1# Revision history for pandoc
3## pandoc 2.14.0.2 (2021-06-13)
5 * Fix MediaBag regressions (#7345). iIn the 2.14 release `--extract-media`
6 stopped working as before; there could be mismatches between the
7 paths in the rendered document and the extracted media.
8 This patch makes several changes that restore the earlier behavior
9 (while keeping the same API). The `mediaPath` in 2.14 was always
10 constructed from the SHA1 hash of the media contents. Now, we
11 preserve the original path unless it's an absolute path or contains
12 `..` segments (in that case we use a path based on the SHA1 hash of
15 In Text.Pandoc.MediaBag, `mediaDirectory` and `mediaItems` now use the
16 `mediaPath`, rather than the mediabag key, for the first component of the
17 tuple. This makes more sense, I think, and fits with the documentation of
18 these functions; eventually, though, we should rework the API so that
19 `mediaItems` returns both the keys and the MediaItems.
21 In Text.Pandoc.Class.IO, rewriting of source paths in `extractMedia` has
24 In Text.Pandoc.Class.PandocMonad, `fillMediaBag` has been modified so that
25 it doesn't modify image paths (that was part of the problem in #7345).
27 We now do path normalization (e.g. `\` separators on Windows) in
32 + Text.Pandoc.PDF: Fix regression in 2.14 for generation of PDFs with
34 + Only print relevant part of environment on `--verbose`. Since
35 `--verbose` output might be put in an issue, we want to avoid
36 spilling out secrets in environment variables.
38 * Markdown reader: fix pipe table regression in 2.11.4 (#7343).
39 Previously pipe tables with empty headers (that is, a header
40 line with all empty cells) would be rendered as headerless
41 tables. This broke in 2.11.4. The fix here is to produce an
42 AST with an empty table head when a pipe table has all empty
45 * LaTeX reader: don't allow optional `*` on symbol control sequences
46 (#7340). Generally we allow optional starred variants of LaTeX commands
47 (since many allow them, and if we don't accept these explicitly,
48 ignoring the star usually gives acceptable results). But we
49 don't want to do this for `\(*\)` and similar cases.
51 * Docx reader: handle absolute URIs in Relationship Target (#7374).
53 * Docx writer: fix handling of empty table headers (Albert Krewinkel,
54 #7369). A table header which does not contain any cells is now treated as
57 * LaTeX writer: Fix regression in table header position (#7347).
58 In recent versions the table headers were no longer bottom-aligned
59 (if more than one line). This patch fixes that by using minipages
60 for table headers in non-simple tables.
64 + Do not use simple class for fenced-divs (Jan Tojnar, amends #7242.)
65 + Do not throw away attributes when `Ext_attributes` is enabled.
66 `Ext_attributes` covers at least the following:
67 `Ext_fenced_code_attributes`, `Ext_header_attributes`,
68 `Ext_inline_code_attributes`, `Ext_link_attributes`.
72 + Allow `pipe_tables` to be disabled for commonmark formats
73 (`commonmark_x`, `gfm`) (#7375).
74 + Re-use functions from Text.Pandoc.Markdown.Inline (Jan Tojnar).
76 * DocBook writer: Remove non-existent admonitions (Jan Tojnar).
77 `attention`, `error` and `hint` are reStructuredText specific.
79 * HTML writer: Don't omit width attribute on div (#7342).
81 * Text.Pandoc.MIME, `extensionFromMimeType`: add a few special cases.
82 When we do a reverse lookup in the MIME table, we just get the
83 last match, so when the same mime type is associated with several
84 different extensions, we sometimes got weird results, e.g. `.vs`
85 for `text/plain`. These special cases help us get the most standard
86 extensions for mime types like `text/plain`.
88 * Lua utils: fix handling of table headers in `from_simple_table` (Albert
89 Krewinkel, #7369). Passing an empty list of header cells now results
90 in an empty table header.
92 * Text.Pandoc.Citeproc:
94 + Avoid duplicate classes and attributes on references div.
95 + Fix regression in citeproc processing (#7376). If inline
96 references are used (in the metadata `references` field), we
97 should still only include in the bibliography items that are
98 actually cited (unless `nocite` is used).
100 * Require citeproc 0.4.0.1. This fixes a bug which led to doubled
101 "et al." in some (rare) circumstances.
105 + Mention GladTeX for EPUB export (Sebastian Humenda).
106 This updates the manual and the web site about the GladTeX usage.
107 + More details and a useful link for YAML syntax.
109 * CONTRIBUTING.md: update modules overview (Albert Krewinkel).
111 * using-the-pandoc-api.md: switch from String to Text (Albert Krewinkel).
114## pandoc 2.14.0.1 (2021-06-01)
116 * Commonmark reader: Fix regression in 2.14 with YAML metdata block parsing,
117 which could cause the document body to be omitted after metadata (#7339).
119 * HTML reader: fix column width regression in 2.14 (#7334).
120 Column widths specified with a style attribute were off by a factor of 100.
122 * Markdown reader: in `rebasePaths`, check for both Windows and Posix
123 absolute paths. Previously Windows pandoc was treating
124 `/foo/bar.jpg` as non-absolute.
126 * Text.Pandoc.Logging: In rendering `LoadedResource`, use relative paths.
128 * Docx writer: fix regression on captions (#7328). The "Table Caption"
129 style was no longer getting applied. (It was overwritten by "Compact.")
131 * Use commonmark-extensions 0.2.1.2
133## pandoc 2.14 (2021-05-28)
135 * Change reader types, allowing better tracking of source positions
136 [API change]. Previously, when multiple file arguments were provided,
137 pandoc simply concatenated them and passed the contents to the readers,
138 which took a Text argument. As a result, the readers had no way of knowing
139 which file was the source of any particular bit of text. This meant that
140 we couldn't report accurate source positions on errors or include accurate
141 source positions as attributes in the AST. More seriously, it meant that
142 we couldn't resolve resource paths relative to the files containing them
143 (see e.g. #5501, #6632, #6384, #3752).
145 * Add `rebase_relative_paths` extension (#3752). When enabled, this
146 extension rewrites relative image and link paths by prepending
147 the (relative) directory of the containing file. This
148 behavior is useful when your input sources are split
149 into multiple files, across several directories, with files
150 referring to images stored in the same directory. The
151 extension can be enabled for all markdown and commonmark-based formats.
153 * Add Text.Pandoc.Sources (exported module), with a `Sources` type and a
154 `ToSources` class. A `Sources` wraps a list of `(SourcePos, Text)` pairs
155 [API change]. A parsec `Stream` instance is provided for `Sources`. The
156 module also exports versions of parsec's `satisfy` and other Char parsers
157 that track source positions accurately from a `Sources` stream (or any
158 instance of the new `UpdateSourcePos` class).
160 * Text.Pandoc.Parsing
162 + Export the modified Char parsers defined in Text.Pandoc.Sources
163 instead of the ones parsec provides. Modified parsers to use a
164 `Sources` as stream [API change].
165 + Improve include file functions [API change]. Remove old
166 `insertIncludedFileF`. Give `insertIncludedFile` a more general type,
167 allowing it to be used where `insertIncludedFileF` was.
168 + Add parameter to the `citeKey` parser from
169 Text.Pandoc.Parsing, which controls whether the `@{..}`
170 syntax is allowed [API change].
172 * Text.Pandoc.Error: Modified the constructor `PandocParsecError` to take a
173 `Sources` rather than a `Text` as first argument, so parse error locations
174 can be accurately reported.
176 * Fix source position reporting for YAML bibliographies (#7273).
178 * Issue error message when reader or writer format is malformed
179 (#7231). Previously we exited with an error status but (due to a bug)
182 * Smarter smart quotes (#7216, #2103). Treat a leading `"` with no
183 closing `"` as a left curly quote. This supports the practice, in
184 fiction, of continuing paragraphs quoting the same speaker without an
185 end quote. It also helps with quotes that break over lines in line blocks.
189 + Use MetaInlines not MetaBlocks for multimarkdown metadata fields.
190 This gives better results in converting to e.g. pandoc markdown.
191 + Implement curly-brace syntax for Markdown citation keys (#6026).
192 The change provides a way to use citation keys that contain
193 special characters not usable with the standard citation key syntax.
194 Example: `@{foo_bar{x}'}` for the key `foo_bar{x}`. It also allows
195 separating citation keys from immediately following text, e.g. `@{foo}A`.
199 + Seek include files in the directory of the file
200 containing the include directive, as RST requires (#6632).
201 + Use `insertIncludedFile` from Text.Pandoc.Parsing
202 instead of reproducing much of its code.
204 * Org reader: Resolve org includes relative to the directory containing the
205 file containing the INCLUDE directive (#5501).
207 * ODT reader: Treat tabs as spaces (#7185, niszet).
211 + Add handling of vml image objects (#7257, mbrackeantidot).
212 + Support new table features (Emily Bourke, #6316): column
213 spans, row spans, multiple header rows, table description
214 (parsed as a simple caption), captions, column widths.
218 + Improved siunitx support (#6658, #6620).
219 + Better support for `\xspace` (#7299).
220 + Improve parsing of `\def` macros. We previously set "verbatim mode"
221 even for parsing the initial `\def`; this caused problems
222 for `\def` nested inside another `\def`.
223 + Implement `\newif`.
225 * ConTeXt writer: improve ordered lists (#5016, Denis Maier).
226 Change ordered list from itemize to enumerate. Add new
227 itemgroup for ordered lists. Remove manual insertion of
228 width attributes. Use tabular figures in ordered list
233 + Don't fail on unmatched closing "script" tag (Albert Krenkel, #7282).
234 + Keep h1 tags as normal headers (#2293, Albert Krewinkel).
235 The tags `<title>` and `<h1 class="title">` often contain the same
236 information, so the latter was dropped from the document. However, as
237 this can lead to loss of information, the heading is now always
238 retained. Use `--shift-heading-level-by=-1` to turn the `<h1>`
239 into the document title, or a filter to restore the previous behavior.
240 + Handle relative lengths (e.g. `2*`) in HTML column widths (#4063).
241 See <https://www.w3.org/TR/html4/types.html#h-6.6>.
243 * DocBook/JATS readers:
245 + Fix mathml regression caused by the switch in XML libraries (#7173).
246 + Fix "phrase" in DocBook: take classes from "role" not "class" (#7195).
248 * DocBook reader: ensure that first and last names are separated (#6541).
250 * Jira reader (Albert Krewinkel, #7218):
252 + Support "smart" links: `[alias|https://example.com|smart-card]` syntax.
253 + Allow spaces and most unicode characters in attachment links.
254 + No longer require a newline character after `{noformat}`.
255 + Only allow URI path segment characters in bare links.
256 + The `file:` schema is no longer allowed in bare links; these
259 * Plain writer: handle superscript unicode minus (#7276).
263 + Better handling of line breaks in simple tables (#7272).
264 Now we also handle the case where they're embedded in other elements,
266 + For beamer output, support `exampleblock` and `alertblock` (#7278).
267 A block will be rendered as an `exampleblock` if the heading
268 has class `example` and an `alertblock` if it has class `alert`.
269 + Separate successive quote chars with thin space (#6958,
270 Albert Krewinkel). Successive quote characters are separated with
271 a thin space to improve readability and to prevent unwanted ligatures.
272 Detection of these quotes sometimes had failed if the second quote
273 was nested in a span element.
274 + Separate successive quote chars with thin space (#6958, Albert
277 * EPUB Writer: Fix belongs-to-collection XML id choice (#7267, nuew).
278 The epub writer previously used the same XML id for both the book
279 identifier and the epub collection. This causes an error on epubcheck.
281 * BibTeX/BibLaTeX writer: Handle `annote` field (#7266).
283 * ZimWiki writer: allow links and emphasis in headers (#6605,
288 + Support blank lines in line blocks (#6564, Albert Krewinkel,
289 thanks to @denismaier).
290 + Use span identifiers as reference anchors (#7246, Albert Krewinkel).
294 + Keep attributes from code nested below `pre` tag (#7221,
295 Albert Krewinkel). If a code block is defined with `<pre><code
296 class="language-x">…</code></pre>`, where the `<pre>` element has no
297 attributes, then the attributes from the `<code>` element are used
298 instead. Any leading `language-` prefix is dropped in the code's
299 `class` attribute are dropped to improve syntax highlighting.
300 + Ensure headings only have valid attribs in HTML4 (#5944, Albert
302 + Parse `<header>` as a Div (Albert Krewinkel).
306 + Inline latex envs need newlines (#7252, tecosaur).
307 As specified in https://orgmode.org/manual/LaTeX-fragments.html, an
308 inline \begin{}...\end{} LaTeX block must start on a new line.
309 + Use LaTeX style maths deliminators (#7196, tecosaur).
311 * JATS writer (Albert Krewinkel):
313 + Use either styled-content or named-content for spans (#7211).
314 If the element has a content-type attribute, or at least one class,
315 then that value is used as `content-type` and the span is put inside
316 a `<named-content>` element. Otherwise a `<styled-content>` element
318 + Reduce unnecessary use of `<p>` elements for wrapping (#7227).
319 The `<p>` element is used for wrapping in cases were the contents
320 would otherwise not be allowed in a certain context. Unnecessary
321 wrapping is avoided, especially around quotes (`<disp-quote>` elements).
322 + Convert spans to `<named-content>` elements (#7211). Spans with
323 attributes are converted to `<named-content>` elements instead of
324 being wrapped with `<milestone-start/>` and `<milestone-end>`
325 elements. Milestone elements are not allowed in documents using the
326 articleauthoring tag set, so this change ensures the creation of valid
328 + Add footnote number as label in backmatter (#7210). Footnotes in the
329 backmatter are given the footnote's number as a label. The
330 articleauthoring output is unaffected from this change, as footnotes
331 are placed inline there.
332 + Escape disallows chars in identifiers. XML identifiers must start
333 with an underscore or letter, and can contain only a limited set
334 of punctuation characters. Any IDs not adhering to these rules are
335 rewritten by writing the offending characters as `Uxxxx`,
336 where `xxxx` is the character's hex code.
338 * Jira writer: use `{color}` when span has a color attribute
339 (Albert Krewinkel, tarleb/jira-wiki-markup#10).
343 + Autoset table width if no column has an explicit width (Albert
345 + Extract Table handling into separate module (Albert Krewinkel).
346 + Support colspans and rowspans in tables (Albert Krewinkel, #6315).
347 + Support multirow table headers (Albert Krewinkel).
348 + Improve integration of settings from reference.docx (#1209).
349 This change allows users to create a reference.docx that
350 sets `w:proofState` for spelling or grammar to `dirty`,
351 so that spell/grammar checking will be triggered on the
353 + Copy over more settings from reference.docx (#7240). From settings.xml
354 in the reference-doc, we now include: `zoom`, `embedSystemFonts`,
355 `doNotTrackMoves`, `defaultTabStop`, `drawingGridHorizontalSpacing`,
356 `drawingGridVerticalSpacing`, `displayHorizontalDrawingGridEvery`,
357 `displayVerticalDrawingGridEvery`, `characterSpacingControl`,
358 `savePreviewPicture`, `mathPr`, `themeFontLang`, `decimalSymbol`,
359 `listSeparator`, `autoHyphenation`, `compat`.
360 + Set zoom to 100% by default in settings.xml.
361 + Align math options more with current Word defaults (e.g. Cambria Math
363 + Remove `rsid`s from default settings.xml. Word will add these
364 when revisions are made.
366 * Ms writer: Handle tables with multiple paragraphs (#7288).
367 Previously they overflowed the table cell width. We now set line lengths
368 per-cell and restore them after the table has been written.
372 + Use cleaner braceless syntax for code blocks with a
373 single class (#7242, Jan Tojnar).
374 + Add quotes properly in markdown YAML metadata fields (#7245).
375 This fixes a bug, which caused the writer to look at the *last*
376 rather than the *first* character in determining whether quotes
377 were needed. So we got spurious quotes in some cases and
378 didn't get necessary quotes in others.
379 + Use `@{..}` syntax for citations when needed.
380 + Use fewer unneeded escapes for `#` (see #6259).
381 + Improve escaping of `@`. We need to escape literal `@` before
382 `{` because of the new citation syntax.
384 * Commonmark writer: Use backslash escapes for `<` and `|`...
385 instead of entities (#7208).
387 * Powerpoint writer: allow `monofont` to be specified in metadata
392 + Use non-starred names for xcolor color names (#6109).
393 This should make svgnames and x11names work properly.
394 + Fix bad vertical spacing after bibliography (#7234, badumont).
395 + List of figures before list of tables (#7235, Julien Dutant).
396 + Move CSL macro definitions before header-includes so they can be
398 + Improve treatment of CSL `entry-spacing` (#7296).
399 Previously with the default template settings (`indent` variable
400 not set), we would get interparagraph spaces separating bib
401 entries even with `entry-spacing="0"`. On the other hand,
402 setting `entry-spacing="2"` gave ridiculously large spacing.
403 This change makes the spacing caused by `entry-spacing` a multiple
404 of `\parskip` by default, which gives aesthetically reasonable
405 output. Those who want a larger or smaller unit (e.g. because
406 they use `indent` which sets `\parskip` to 0) may
407 `\setlength{\cslentryspacingunit}{10pt}` in header-includes
408 to override the defaults.
409 + Move title, author, date up to top of preamble (#7295).
410 This allows header-includes to use them, and puts them
411 in a position where you can see them immediately.
412 + Define commands for zero width non-joiner character
413 (#6639, Albert Krewinkel). The zero-width non-joiner character
414 is used to avoid ligatures (e.g. in German).
418 + Define `enumerate` itemgroup (#5016, Denis Maier).
419 + List of figures before list of tables (#7235, Julien Dutant).
421 * reveal.js template:
423 + Support `toc-title` (#7171, Florian Kohrt).
424 + Use `hash: true` by default rather than `history: true` (#6968).
426 * HTML-based slide shows: add support for `institute` (#7289, Thomas
429 * Text.Pandoc.Extensions: Add constructor `Ext_rebase_relative_paths` to
430 `Extensions` [API change].
432 * Text.Pandoc.XML.Light: add Eq, Ord instances for Content,
433 Element, Attr, CDataKind [API change].
435 * Text.Pandoc.MediaBag:
437 + Change type to use a `Text` key instead of `[FilePath]`.
438 We normalize the path and use `/` separators for consistency.
439 + Export `MediaItem` type [API change].
440 + Change `MediaBag` type to a map from Text to MediaItem [API change].
441 + `lookupMedia` now returns a `MediaItem` [API change].
442 + Change `insertMedia` so it sets the `mediaPath` to a filename based on
443 the SHA1 hash of the contents. This will be used when contents
446 * Text.Pandoc.Class.PandocMonad:
448 + Remove `fetchMediaResource` [API change]. Use `fetchItem` to get
449 resources in `fillMediaBag`.
450 + Add informational message in `downloadOrRead` indicating what path
451 local resources have been loaded from.
453 * Text.Pandoc.Logging:
455 + Remove single quotes around paths in messages.
456 + Add LoadedResource constructor to LogMessage [API change].
457 This is for INFO-level messages telling where image data has been
458 loaded from. (This can vary because of the resource path.)
460 * Text.Pandoc.Asciify: simplify code and export `toAsciiText` [API change].
461 Instead of encoding a giant (and incomplete) map, we now
462 just use unicode-transforms to normalize the text to
463 a canonical decomposition, and manipulate the result.
465 * App: allow tabs expansion even if file-scope is used (Albert Krewinkel,
466 #6709). Tabs in plain-text inputs are now handled correctly, even if
467 the `--file-scope` flag is used.
469 * Add new internal module Text.Pandoc.Writers.GridTable (Albert Krewinkel).
471 * Text.Pandoc.Highlighting: Change type of `languagesByExtension`, adding
472 a parameter for a `SyntaxMap` [API change] (Jan Tojnar, #7241).
473 Languages defined using `--syntax-definition` were not recognized by
474 `languagesByExtension`. This patch corrects that, allowing the writers
475 to see all custom definitions. The LaTeX writer still uses the default
476 syntax map, but that's okay in that context, since
477 `--syntax-definition` won't create new listings styles.
479 * Text.Pandoc.Citeproc:
481 + Ensure that CSL-related attributes are passed on to a Div with id
482 'refs'. Otherwise things like `entry-spacing` won't work when
484 + Use metadata's `lang` for the lang parameter of citeproc, overriding
486 + Recognize locators spelled with a capital letter (#7323).
487 + Add a comma and a space in front of the suffix if it doesn't start
488 with space or punctuation (#7324).
489 + Don't detect math elements as locators (#7321).
491 * Remove Text.Pandoc.BCP47 module [API change]. Use types and functions
492 from UnicodeCollation.Lang instead. This is a richer implementation
495 * Text.Pandoc.Shared:
497 + Fix regression in grid tables for wide characters (#7214).
498 In the translation from String to Text, a char-width-sensitive
499 `splitAt'` was dropped. This commit reinstates it and uses it to make
500 `splitTextByInstances` char-width sensitive.
501 + Add `getLang` (formerly in the now-removed BCP47) [API change].
503 * Text.Pandoc.SelfContained: use `application/octet-stream`
504 for unknown mime types instead of halting with an error (#7202).
506 * Lua filters: respect Inlines/Blocks filter functions in `pandoc.walk_*`
509 * Add text as build-depend for trypandoc (#7193, Roman Beránek).
511 * Bump upper-bounds for network-uri, time, attoparsec.
515 * Use texmath 0.12.3.
517 * Use jira-wiki-markup 1.3.5 (Albert Krewinkel).
519 * Require latest skylighting (fixes a bug in XML syntax highlighting).
521 * Use latest xml-conduit.
523 * Use latest commonmark, commonmark-extensions, commonmark-pandoc.
525 * Use haddock-library-1.10.0 (Albert Krewinkel).
527 * Allow compilation with base 4.15 (Albert Krewinkel).
531 + Add information about `lang` and bibliography sorting.
532 + Add info about YAML escape sequences, link to spec (#7152,
534 + Note that `institute` variable works for HTML-based slides.
535 + Update documentation on citation syntax.
536 + Add citation example for locators and suffixes (Tristan Stenner)
538 * Updated and fixed typos in documentation (Charanjit Singh,
539 Anti-Distinctlyminty, Tatiana Porras, obcat).
541 * Add instructions for installing pandoc-types before compiling filter.
543 * INSTALL: add note that parallel installations should be avoided
546 * Remove `biblatex-nussbaum.md` test. It is basically the same
547 as `biblaetx-quotes.md`.
549 * Command tests: fail if a file contains no tests---and fix a
550 test that failed in that way!
552 * Use smaller images in tests, reducing the size of the source tarball by 8 MB.
555## pandoc 2.13 (2021-03-21)
557 * Support `yaml_metadata_block` extension for `commonmark`, `gfm` (#6537).
558 This supported is a bit more limited than with pandoc's
559 `markdown`. The YAML block must be the first thing in the input,
560 and the leaf notes are parsed in isolation from the rest of
561 the document. So, for example, you can't use reference
562 links if the references are defined later in the document.
564 * Fix fallback to default partials when custom templates are used.
565 If the directory containing a template does not contain the partial,
566 it should be sought in the default templates, but this was not
567 working properly (#7164).
569 * Handle `nocite` better with `--biblatex` and `--natbib` (#4585).
570 Previously the nocite metadata field was ignored with these formats.
571 Now it populates a `nocite-ids` template variable and causes a
572 `\nocite` command to be issued.
574 * Text.Pandoc.Citeproc: apply `fixLinks` correctly (#7130). This is code
575 that incorporates a prefix like `https://doi.org/` into a following link
578 * Text.Pandoc.Shared:
580 + Remove `backslashEscapes`, `escapeStringUsing` [API change]. Replace
581 these inefficient association list lookups with more efficient escaping
582 functions in the writers that used them (for a 10-25% performance boost
583 in org, haddock, rtf, texinfo writers).
584 + Remove `ToString`, `ToText` typeclasses [API change]. These were needed
585 for the transition from String to Text, but they are no longer used and
586 may clash with other things.
587 + Simplify `compactDL`.
589 * Text.Pandoc.Parsing:
591 + Change type of `readWithM` so that it is no longer polymorphic
592 [API change]. The `ToText` class has been removed, and now that we've
593 completed the transition to Text we no longer need this to operate
595 + Remove `F` type synonym [API change]. Muse and Org were defining their
598 * Text.Pandoc.Readers.Metadata:
600 + Export `yamlMetaBlock` [API change].
601 + Make `yamlBsToMeta`, `yamlBsToRefs` polymorphic on the parser state
604 * Markdown reader: Fix regression with `tex_math_backslash` (#7155).
606 * MediaWiki reader: Allow block-level content in notes (ref) (#7145).
608 * Jira reader (Albert Krewinkel):
610 + Fixed parsing of autolinks (i.e., of bare URLs in the text).
611 Previously an autolink would take up the rest of a line, as spaces
612 were allowed characters in these items.
613 + Emoji character sequences no longer cause parsing failures. This was
614 due to missing backtracking when emoji parsing fails.
615 + Mark divs created from panels with class "panel".
617 * RST reader: fix logic for ending comments (#7134). Previously comments
618 sometimes got extended too far.
620 * DocBook writer: include Header attributes as XML attributes on
621 section (Erik Rask). Attributes with key names that are not allowed
622 as XML attributes are dropped, as are attributes with invalid values
623 and `xml:id` (DocBook 5) and `id` (DocBook 4).
627 + Make `nsid` in `abstractNum` deterministic. Previously we assigned
628 a random number, but we don't need random values, so now we just
629 assign a value based on the list marker.
630 + Use integral values for `w:tblW` (#7141).
632 * Jira writer (Albert Krewinkel):
634 + Block quotes are only rendered as `bq.` if they do not contain a
636 + Jira writer: improve div/panel handling. Include div attributes in
637 panels, always render divs with class `panel` as panels, and
638 avoid nesting of panels.
640 * HTML writer: Add warnings on duplicate attribute values.
641 This prevents emitting invalid HTML. Ultimately it would be good to
642 prevent this in the types themselves, but this is better for now.
644 * Org writer: Prevent unintended creation of ordered list items (#7132,
645 Albert Krewinkel). Adjust line wrapping if default wrapping would cause
646 a line to be read as an ordered list item.
648 * JATS templates: support 'equal-contrib' attrib for authors (Albert
649 Krewinkel). Authors who contributed equally to a paper may be marked
650 with `equal-contrib`.
652 * reveal.js template: replace JS comment with HTML (#7154, Florian Kohrt).
654 * Text.Pandoc.Logging: Add `DuplicateAttribute` constructor to `LogMessage`.
657 * Use `-j4` for linux release build. This speeds up the build dramatically
660 * cabal.project: remove ghcoptions. Move flags to top level, so they can
661 be set differently on the command line.
663 * Require latest texmath, skylighting, citeproc, jira-wiki-markup.
664 (The latest skylighting fixes a bad bug with Haskell syntax highlighting.)
665 Narrow version bounds for texmath, skylighting, and citeproc, since
666 the test output depend on them.
668 * Use doclayout 0.3.0.2. This significantly reduces the time and memory
669 needed to compile pandoc.
671 * Use `foldl'` instead of `foldl` everywhere.
673 * Update bounds for random (#7156, Alexey Kuleshevich).
675 * Remove uses of some partial functions.
677 * Don't bake in a larger stack size for the executable.
681 + Use `getExecutablePath` from base, avoiding the dependency on
683 + Factor out `setupEnvironment` in Helpers, to avoid code duplication.
684 + Fix finding of data files by setting teh `pandoc_datadir` environment
685 variable when we shell out to pandoc. This avoids the need to use
686 `--data-dir` for the tests, which caused problems finding `pandoc.lua`
687 when compiling without the `embed_data_files` flag (#7163).
689 * Benchmark improvements:
691 + Build `+RTS -A8m -RTS` into default ghc-options for benchmark.
692 This is necessary to get accurate benchmark results; otherwise we
693 are largely measuring garbage collecting, some not related to the
695 + Allow specifying BASELINE file in 'make bench' for comparison
696 (otherwise the latest benchmark is chosen by default).
697 + Force `readFile` in benchmarks early (Bodigrim).
699 * CONTRIBUTING: suggest using a `cabal.project.local` file (#7153,
702 * Add ghcid-test to Makefile. This loads the test suite in ghcid.
705## pandoc 2.12 (2021-03-08)
707 * `--resource-path` now accumulates if specified multiple
708 times (#6152). Resource paths specified later on the command line are
709 prepended to those specified earlier. Thus,
710 `--resource-path foo --resource-path bar:baz` is equivalent to
711 `--resource-path bar:bas:foo`. (The previous behavior was
712 for the last `--resource-path` to replace all the rest.)
713 `resource-path` in defaults files behaves the same way: it
714 will be prepended to the resource path set by earlier
715 command line options or defaults files. This change
716 facilitates the use of multiple defaults files: each can
717 specify a directory containing resources it refers to
718 without clobbering the resource paths set by the others.
720 * Allow defaults files to refer to the home directory, the
721 user data directory, and the directory containing the defaults file
722 itself (#5871, #5982, #5977). In fields that expect file paths
723 (and only in these fields),
725 + `${VARIABLE}` will expand to the value of the environment variable
726 `VARIABLE` (and in particular `${HOME}` will expand to the path
727 of the home directory). A warning will be raised for undefined
729 + `${USERDATA}` will expand to the path of the user data
730 directory in force when the defaults file is being processed.
731 + `${.}` will expand to the directory containing the defaults file.
732 (This allows default files to be placed in a directory containing
733 resources they make use of.)
735 * When downloading content from URL arguments, be sensitive to
736 the character encoding (#5600). We can properly handle UTF-8 and latin1
737 (ISO-8859-1); for others we raise an error. Fall back to latin1 if
738 no charset is given in the mime type and UTF-8 decoding fails.
740 * Allow abbreviations that don't end in a period to be
741 specified using `--abbreviations` (#7124).
743 * Add new unexported module Text.Pandoc.XML.Light, as well
744 as Text.Pandoc.XML.Light.Types, Text.Pantoc.XML.Light.Proc,
745 Text.Pandoc.XML.Light.Output. (Closes #6001, #6565, #7091).
747 This module exports definitions of `Element` and `Content`
748 that are isomorphic to xml-light's, but with Text
749 instead of String. This allows us to keep most of the code in existing
750 readers that use xml-light, but avoid lots of unnecessary allocation.
752 We also add versions of the functions from xml-light's
753 Text.XML.Light.Output and Text.XML.Light.Proc that operate on our
754 modified XML types, and functions that convert xml-light types to our
755 types (since some of our dependencies, like texmath, use xml-light).
757 We export functions that use xml-conduit's parser to produce an
758 `Element` or `[Content]`. This allows existing pandoc code to use
759 a better parser without much modification.
761 The new parser is used in all places where xml-light's parser was
762 previously used. Benchmarks show a significant performance improvement
763 in parsing XML-based formats (with docbook, opml, jats, and docx
764 almost twice as fast, odt and fb2 more than twice as fast).
766 In addition, the new parser gives us better error reporting than
767 xml-light. We report XML errors, when possible, using the new
768 `PandocXMLError` constructor in `PandocError`.
770 These changes revealed the need for some changes in the tests. The
771 docbook-reader.docbook test lacked definitions for the entities it used;
772 these have been added. And the docx golden tests have been updated,
773 because the new parser does not preserve the order of attributes.
777 + Avoid expensive tree normalization step, as it is not necessary
778 with the new XML parser.
779 + Support `informalfigure` (#7079) (Nils Carlson).
783 + Use Map instead of list for Namespaces. This gives a speedup of
784 about 5-10%. With this and the XML parsing changes, the docx reader
785 is now about twice as fast as in the previous release.
789 + Small performance tweaks.
790 + Also, remove exported class `NamedTag(..)` [API change]. This was just
791 intended to smooth over the transition from String to Text and is no
793 + As a result, the functions `isInlineTag` and `isBlockTag`
794 are no longer polymorphic; they apply to a `Tag Text` [API change].
795 + Do a lookahead to find the right parser to use. This takes
796 benchmarks from 34ms to 23ms, with less allocation.
797 + Fix bad handling of empty `src` attribute in `iframe` (#7099).
798 If `src` is empty, we simply skip the `iframe`.
799 If `src` is invalid or cannot be fetched, we issue a warning
800 nd skip instead of failing with an error.
804 + Avoid tree normalization, which is no longer necessary given the
809 + Don't export `tokenize`, `untokenize` [API change]. These are internal
810 implementation details, which were only exported for testing.
811 They don't belong in the public API.
812 + Improved efficiency of the parser. With these changes the reader
813 is almost twice as fast as in the last release in our benchmarks.
814 + Code cleanup, removing some unnecessary things.
815 + Rewrite `withRaw` so it doesn't rely on fragile assumptions
816 about token positions (which break when macros are expanded)
817 (#7092). This requires the addition of `sEnableWithRaw` and
818 `sRawTokens` in `LaTeXState`, and a new combinator `disablingWithRaw`
819 to disable collecting of raw tokens in certain contexts.
820 Add `parseFromToks` to Text.Pandoc.Readers.LaTeX.Parsing.
821 Fix parsing of single character tokens so it doesn't mess
822 up the new raw token collecting. These changes slightly increase
823 allocations and have a small performance impact.
824 + Handle some bibtex/biblatex-specific commands that used to be
825 dealt with in pandoc-citeproc (#7049).
826 + Optimize `satisfyTok`, avoiding unnecessary macro expansion steps.
827 Benchmarks after this change show 2/3 of the run time and 2/3 of the
828 allocation of the Feb. 10 benchmarks.
829 + Removed `sExpanded` in state. This isn't actually needed and checking
830 it doesn't change anything.
831 + Improve `braced'`. Remove the parameter, have it parse the
832 opening brace, and make it more efficient.
833 + Factor out pieces of the LaTeX reader to make the module smaller.
834 This reduces memory demands when compiling. Created
835 Text.Pandoc.Readers.{LaTeX,Math,Citation,Table,Macro,Inline}.
836 Changed Text.Pandoc.Readers.LaTeX.SIunitx to export a command map
837 instead of individual commands.
838 + Handle table cells containing `&` in `\verb` (#7129).
840 * Make Text.Pandoc.Readers.LaTeX.Types an unexported module [API change].
844 + Improved handling of mmd link attributes in references (#7080).
845 Previously they only worked for links that had titles.
846 + Improved efficiency of the parser (benchmarks show a 15% speedup).
850 + Avoid tree normalization, which is no longer necessary with the
855 + Finer-grained errors on parse failure (#7091).
856 + Give more information if the zip container can't be unpacked.
860 + Support `task_lists` extension (Albert Krewinkel, #6336).
861 + Fix bug in org-ref citation parsing (Albert Krewinkel, #7101).
862 The org-ref syntax allows to list multiple citations separated by
863 comma. Previously commas were accepted as part of the citation id,
864 so all citation lists were parsed as one single citation.
868 + Use `getTimestamp` instead of `getCurrentTime` to fetch timestamp.
869 Setting `SOURCE_DATE_EPOCH` will allow reproducible builds.
870 + RST reader: fix handling of header in CSV tables (#7064).
871 The interpretation of this line is not affected by the delim option.
875 + Modified the Doc parser to skip leading blank lines. This fixes
876 parsing of documents which start with multiple blank lines (Albert
878 + Prevent URLs within link aliases to be treated as autolinks (Albert
883 + Remove formerly exported functions that are no longer used in the
884 code base: `splitByIndices`, `splitStringByIndicies`, `substitute`,
885 and `underlineSpan` (which had been deprecated in April 2020)
887 + Export `handleTaskListItem` (Albert Krewinkel) [API change].
888 + Change `defaultUserDataDirs` to `defaultUserDataDir` [API
889 change]. We determine what is the default user data directory
890 by seeing whether the XDG directory and/or legacy
895 + BibTeX writer: use doclayout and doctemplate. This change allows
896 bibtex/biblatex output to wrap as other formats do,
897 depending on the settings of `--wrap` and `--columns` (#7068).
901 + Output `[]` if no references in input, instead of raising a
902 PandocAppError as before.
906 + Use `getTimestamp` instead of `getCurrentTime` for timestamp.
907 Setting `SOURCE_DATE_EPOCH` will allow reproducible builds.
911 + Use `getTimestamp` instead of `getCurrentTime` for timestamp.
912 Setting `SOURCE_DATE_EPOCH` will allow reproducible builds (#7093).
913 This does not suffice to fully enable reproducible in EPUB, since
914 a unique id is still being generated for each build.
915 + Support `belongs-to-collection` metadata (#7063) (Nick Berendsen).
919 + Escape special chars in reference elements (Albert Krewinkel).
920 Prevents the generation of invalid markup if a citation element
921 contains an ampersand or another character with a special meaning
926 + Use Span identifiers as anchors (Albert Krewinkel).
927 + Use `{noformat}` instead of `{code}` for unknown languages (Albert
928 Krewinkel). Code blocks which are not marked as a language supported
929 by Jira are rendered as preformatted text via `{noformat}` blocks.
933 + Adjust hypertargets to beginnings of paragraphs (#7078).
934 Use `\vadjust pre` so that the hypertarget takes you to the beginning
935 of the paragraph rather than one line down.
936 This makes a particular difference for links to citations using
937 `--citeproc` and `link-citations: true`.
938 + Change BCP47 lang tag from `jp` to `ja` (Mauro Bieg, #7047).
939 + Use function instead of map for accent lookup (should be
941 + Split the module to make it easier to compile on low-memory
942 systems: added Text.Pandoc.Writers.LaTeX.{Util,Citation,Lang}.
946 + Handle math right before digit. We insert an HTML comment to
947 avoid a `$` right before a digit, which pandoc will not recognize
949 + Split the module to make it easier to compile on low-memory
950 systems: added Text.Pandoc.Writers.Markdown.{Types,Inline}.
954 + Use `getTimestamp` instead of `getCurrentTime` for timestamp.
955 Setting `SOURCE_DATE_EPOCH` will allow reproducible builds.
956 + Update default ODT style (Lorenzo). Previously, the "First paragraph"
957 style inherited from "Standard" but not from "Text body." Now
958 it is adjusted to inherit from "Text body", to avoid some ugly
959 spacing issues. It may be necessary to update a custom `reference.odt`
960 in light of this change.
964 + Support `task_lists` extension (Albert Krewinkel, #6336).
968 + Use `getTimestamp` instead of `getCurrentTime` for timestamp.
969 Setting `SOURCE_DATE_EPOCH` will allow reproducible builds.
971 * JATS templates: tag `author.name` as `string-name` (Albert Krewinkel).
972 The partitioning the components of a name into surname, given names,
973 etc. is not always possible or not available. Using `author.name`
974 allows to give the full name as a fallback to be used when
975 `author.surname` is not available.
977 * Add default templates for bibtex and biblatex, so that
978 the variables `header-include`, `include-before`, `include-after`
979 (or alternatively the command line options
980 `--include-in-header`, `--include-before-body`, `--include-after-body`)
985 + Update to iftex package (#7073) (Andrew Dunning)
986 + Wrap url colours in braces (#7121) (Loïc Grobol).
988 * revealjs template: Add 'center' option for vertical slide centering.
991 * Text.Pandoc.XML: Improve efficiency of `fromEntities`.
995 + Add exported function `getCharset` [API change].
997 * Text.Pandoc.UTF8: change IO functions to return Text, not String
998 [API change]. This affects `readFile`, `getContents`, `writeFileWith`,
999 `writeFile`, `putStrWith`, `putStr`, `putStrLnWith`, `putStrLn`.
1000 `hPutStrWith`, `hPutStr`, `hPutStrLnWith`, `hPutStrLn`, `hGetContents`.
1001 This avoids the need to uselessly create a linked list of characters
1002 when emiting output.
1006 + Add `parseOptionsFromArgs` [API change, new exported function].
1007 + Add fields for CSL options to `Opt` [API change]:
1008 `optCSL`, `optbibliography`, `optCitationAbbreviations`.
1010 * Text.Pandoc.Citeproc.BibTeX
1012 + `Text.Pandoc.Citeproc.writeBibTeXString` now returns
1013 `Doc Text` instead of `Text` (#7068).
1014 + Correctly handle `pages` (= `page` in CSL) (#7067).
1015 + Correctly handle BibLaTeX `langid` (= `language` in CSL, #7067).
1016 + In BibTeX output, protect foreign titles since there's no language
1018 + Clean up BibTeX parsing (#7049). Previously there was a messy code
1019 path that gave strange results in some cases, not passing through raw
1020 tex but trying to extract a string content. This was an artefact of
1021 trying to handle some special bibtex-specific commands in the BibTeX
1022 reader. Now we just handle these in the LaTeX reader and simplify
1023 parsing in the BibTeX reader. This does mean that more raw tex will
1024 be passed through (and currently this is not sensitive to the
1025 `raw_tex` extension; this should be fixed).
1027 * Text.Pandoc.Citeproc.MetaValue
1029 + Correctly parse "raw" date value in markdown references metadata.
1030 (See jgm/citeproc#53.)
1032 * Text.Pandoc.Citeproc
1034 + Use https URLs for links (Salim B, #7122).
1038 + Add `getTimestamp` [API change]. This attempts to read the
1039 `SOURCE_DATE_EPOCH` environment variable and parse a UTC time
1040 from it (treating it as a unix date stamp, see
1041 https://reproducible-builds.org/specs/source-date-epoch/). If the
1042 variable is not set or can't be parsed as a unix date stamp, then the
1043 function returns the current date.
1047 + Add `PandocUnsupportedCharsetError` constructor for
1048 `PandocError` [API change].
1049 + Export `renderError` [API change].
1050 + Refactor `handleError` to use `renderError`. This allows us render
1051 error messages without exiting.
1053 * Text.Pandoc.Extensions
1055 + `Ext_task_lists` is now supported by org (and turned
1056 on by default) (Albert Krewinkel, #6336).
1057 + Remove `Ext_fenced_code_attributes` from allowed commonmark attributes
1058 (#7097). This attribute was listed as allowed, but it didn't actually
1059 do anything. Use `attributes` for code attributes and more.
1063 + Always load built-in Lua scripts from default data-dir (Albert
1064 Krewinkel). The Lua modules `pandoc` and `pandoc.List` are now always
1065 loaded from the system's default data directory. Loading from a
1066 different directory by overriding the default path, e.g. via
1067 `--data-dir`, is no longer supported to avoid unexpected behavior
1068 and to address security concerns.
1069 + Add module "pandoc.path" (Albert Krewinkel, #6001, #6565).
1070 The module allows to work with file paths in a convenient and
1071 platform-independent manner.
1072 + Use strict evaluation when retrieving AST value from the stack
1073 (Albert Krewinkel, #6674).
1077 + Disable `smart` extension when building PDF via LaTeX.
1078 This is to prevent accidental creation of ligatures like
1079 `` ?` `` and `` !` `` (especially in languages with quotations like
1080 German), and similar ligature issues. (See jgm/citeproc#54.)
1084 + Fix parsing of unquoted values (#7112). Previously we didn't allow
1085 unescaped quotes in unquoted values, but they are allowed
1090 + Use a more robust method for testing the executable. Many
1091 of our tests require running the pandoc executable. This
1092 is problematic for a few different reasons. First,
1093 cabal-install will sometimes run the test suite after
1094 building the library but before building the executable,
1095 which means the executable isn't in place for the tests.
1096 One can work around that by first building, then building
1097 and running the tests, but that's fragile. Second, we
1098 have to find the executable. So far, we've done that using
1099 a function `findPandoc` that attempts to locate it
1100 relative to the test executable (which can be located
1101 using findExecutablePath). But the logic here is delicate
1102 and work with every combination of options. To solve both
1103 problems, we add an `--emulate` option to the
1104 `test-pandoc` executable. When `--emulate` occurs as the
1105 first argument passed to `test-pandoc`, the program simply
1106 emulates the regular pandoc executable, using the rest of
1107 the arguments (after `--emulate`). Thus, `test-pandoc
1108 --emulate -f markdown -t latex` is just like `pandoc -f
1109 markdown -t latex`. Since all the work is done by library
1110 functions, implementing this emulation just takes a couple
1111 lines of code and should be entirely reliable. With this
1112 change, we can test the pandoc executable by running the
1113 test program itself (locatable using `findExecutablePath`)
1114 with the `--emulate` option. This removes the need for the
1115 fragile `findPandoc` step, and it means we can run our
1116 integration tests even when we're just building the
1117 library, not the executable. [Note: part of this change
1118 involved simplifying some complex handling to set
1119 environment variables for dynamic library paths. I have
1120 tested a build with `--enable-dynamic-executable`, and it
1121 works, but further testing may be needed.]
1122 + Print accurate location if a test fails (Albert
1123 Krewinkel). Ensures that tasty-hunit reports the location
1124 of the failing test instead of the location of the helper
1127 * Documentation: Update URLs and use `https` where possible (#7122,
1130 * Add `doc/libraries.md`, a description of libraries that support pandoc.
1134 + MANUAL: block-level formatting is not allowed in line blocks (#7107).
1135 + Clarify `tex_math_dollars` extension. Note that no blank lines
1136 are allowed between the delimiters in display math.
1137 + Add MANUAL section on reproducible builds.
1138 + Document no template fallback for absolute path (#7077, Nixon
1140 + Improve docs for cite-method.
1141 + Update README and man page.
1143 * Makefile: in `make bench`, create CSV files for comparison and compare
1144 against previous benchmark run. Add timestamp to CSV filenames.
1146 * cabal.project: don't explicitly set -trypandoc.
1147 If we do, this can't be overridden on the cabal command line.
1149 * doc/lua-filters.md: improve documentation for
1150 `pandoc.mediabag.insert`, `pandoc.mediabag.fetch`,
1151 `directory`, `normalize` (Albert Krewinkel).
1153 * Allow base64-bytestring-1.2.* (Dmitrii Kovanikov)
1155 * Require jira-wiki-markup 1.3.3 (Albert Krewinkel)
1157 * Require citeproc 0.3.0.8, which correctly titlecases when titles
1158 contain non-ASCII characters.
1160 * Use skylighting 0.10.4. This version of skylighting uses xml-conduit
1161 rather than hxt. This speeds up parsing of XML syntax definitions
1162 fourfold, and removes four packages from pandoc's dependency graph:
1163 hxt-charproperties, hxt-unicode, hxt-regex-xmlschema, hxt.
1165 * Add script `tools/parseTimings.pl` to help pin down which
1166 modules take the most time and memory to compile.
1168 * Avoid unnecessary use of NoImplicitPrelude pragma (#7089) (Albert
1173 + Use the lighter-weight tasty-bench instead of criterion.
1174 + Run writer benchmarks for binary formats too.
1175 + Alphabetize benchmarks.
1176 + Don't run benchmarks for bibliography formats
1177 (yet; we need a special input for them).
1178 + Show allocation data
1179 + Clean up benchmark code.
1180 + Allow specifying patterns using `-p blah'.
1182 * trypandoc: add 2 second timeout.
1184 * Use `-split-sections` in creating linux release binary.
1185 This reduces executable size significantly (by about 30%).
1187 * Remove `weigh-pandoc`. It's not really useful any more, now that our
1188 regular benchmarks include data on allocation.
1190 * Improve linux package build process and add script to
1191 automate building an arm64 binary package.
1194## pandoc 2.11.4 (2021-01-22)
1196 * Add `biblatex`, `bibtex` as output formats (closes #7040).
1198 * Recognize more extensions as markdown by default (#7034):
1199 `mkdn`, `mkd`, `mdwn`, `mdown`, `Rmd`.
1201 * Implement defaults file inheritance (#6924, David Martschenko).
1202 Allow defaults files to inherit options from other defaults files by
1203 specifying them with the following syntax:
1204 `defaults: [list of defaults files or single defaults file]`.
1206 * Fix infinite HTTP requests when writing epubs from URL source (#7013).
1207 Due to a bug in code added to avoid overwriting the cover image
1208 if it had the form `fileX.YYY`, pandoc made an endless sequence
1209 of HTTP requests when writing epub with input from a URL.
1213 + Allow multiple pipe chars in todo sequences (Albert Krewinkel, #7014).
1214 Additional pipe chars, used to separate "action" state from "no further
1215 action" states, are ignored. E.g., for the following sequence, both
1216 `DONE` and `FINISHED` are states with no further action required:
1217 `#+TODO: UNFINISHED | DONE | FINISHED`.
1218 + Restructure output of captioned code blocks (Albert Krewinkel, #6977).
1219 The Div wrapper of code blocks with captions now has the class
1220 "captioned-content". The caption itself is added as a Plain block
1221 inside a Div of class "caption". This makes it easier to write filters
1222 which match on captioned code blocks. Existing filters will need to be
1224 + Mark verbatim code with class `verbatim` (Dimitri Sabadie, #6998).
1228 + Handle `filecontents` environment (#7003).
1229 + Put contents of unknown environments in a Div when `raw_tex` is not
1230 enabled (#6997). (When `raw_tex` is enabled, the whole environment is
1231 parsed as a raw block.) The class name is the name of the environment.
1232 Previously, we just included the contents without the surrounding Div,
1233 but having a record of the environment's boundaries and name can be
1238 + Allow space around storng/emph delimiters (#6993).
1240 * New module Text.Pandoc.Writers.BibTeX, exporting
1241 writeBibTeX and writeBibLaTeX. [API change]
1245 + Revert table line height increase in 2.11.3 (#6996).
1246 In 2.11.3 we started adding `\addlinespace`, which produced less dense
1247 tables. This wasn't an intentional change; I misunderstood a comment in
1248 the discussion leading up to the change. This commit restores the earlier
1249 default table appearance. Note that if you want a less dense table, you
1250 can use something like `\def\arraystretch{1.5}` in your header.
1254 + Adjust internal links to identifiers defined in raw HTML sections
1255 after splitting into chapters (#7000).
1256 + Recognize `Format "html4"`, `Format "html5"` as raw HTML.
1257 + Adjust internal links to images, links, and tables after splitting into
1258 chapters. Previously we only did this for Div and Span and Header
1259 elements (see #7000).
1263 + Don't justify text inside table cells.
1267 + Use `<element-citation>` if `element_citations`
1268 extension is enabled (Albert Krewinkel).
1269 + Fix citations (Albert Krewinkel, #7018). By default
1270 we use formatted citations.
1271 + Ensure that `<disp-quote>` is always wrapped in `<p>` (#7041).
1275 + Cleaned up raw formats. We now react appropriately
1276 to `gfm`, `commonmark`, and `commonmark_x` as raw formats.
1280 + Fix bug with dropped content from inside spans with a class in
1285 + Handle table header using styles (#7008). Instead of hard-coding
1286 the border and header cell vertical alignment, we now let this
1287 be determined by the Table style, making use of Word's
1288 "conditional formatting" for the table's first row. For
1289 headerless tables, we use the tblLook element to tell Word
1290 not to apply conditional first-row formatting.
1292 * Commonmark writer:
1294 + Implement start number on ordered lists (#7009). Previously they always
1295 started at 1, but according to the spec the start number is respected.
1299 + Fix implicit_figure at end of footnotes (#7006).
1301 * ConTeXt template: Remove `\setupthinrules` from default template.
1302 The width parameter this used is not actually supported,
1303 and the command didn't do anything.
1305 * Text.Pandoc.Extensions:
1307 + Add `Ext_element_citations` constructor (Albert Krewinkel).
1309 * Text.Pandoc.Citeproc.BibTeX: New unexported function
1310 `writeBibtexString`.
1312 * Text.Pandoc.Citeproc:
1314 + Use finer grained imports (Albert Krewinkel).
1315 + Factor out and export `getStyle` [API change].
1316 + Export `getReferences` [API change, #7106].
1317 + Factor out getLang.
1319 * Text.Pandoc.Parsing: modify `gridTableWith'` for headerless tables.
1320 If the table lacks a header, the header row should be an empty
1321 list. Previously we got a list of empty cells, which caused
1322 an empty header to be emitted instead of no header. In LaTeX/PDF
1323 output that meant we got a double top line with space between.
1325 * ImageSize: use `viewBox` for SVG if no length, width attributes (#7045).
1326 This change allows pandoc to extract size information from more SVGs.
1328 * Add simple default.nix.
1330 * Use commonmark 0.1.1.3.
1332 * Use citeproc 0.3.0.5.
1334 * Update default CSL to use latest chicago-author-date.csl.
1336 * CONTRIBUTING.md: add note on GNU xargs.
1340 + Update description of `-L`/`--lua-filter`.
1341 + Document use of citations in note styles (#6828).
1343## pandoc 2.11.3.2 (2020-12-29)
1345 * HTML reader: use renderTags' from Text.Pandoc.Shared (Albert Krewinkel).
1346 A side effect of this change is that empty `<col>` elements are written
1347 as self-closing tags in raw HTML blocks.
1349 * Asciidoc writer: Add support for writing nested tables (#6972, timo-a).
1350 Asciidoc supports one level of nesting. If deeper tables are to be
1351 written, they are omitted and a warning is issued.
1353 * Docx writer: fix nested tables with captions (#6983).
1354 Previously we got unreadable content, because docx seems
1355 to want a `<w:p>` element (even an empty one) at the end of
1358 * Powerpoint writer: allow arbitrary OOXML in raw inline elements
1359 (Albert Krewinkel). The raw text is now included verbatim in the
1360 output. Previously is was parsed into XML elements, which prevented
1361 the inclusion of partial XML snippets.
1363 * LaTeX writer: support colspans and rowspans in tables (#6950,
1364 Albert Krewinkel). Note that the multirow package is needed for
1365 rowspans. It is included in the latex template under a variable,
1366 so that it won't be used unless needed for a table.
1368 * HTML writer: don't include p tags in CSL bibliography entries
1369 (#6966). Fixes a regression in 2.11.3.
1371 * Add `meta-description` variable to HTML templates (#6982). This
1372 is populated by the writer by stringifying the `description`
1373 field of metadata (Jerry Sky). The `description` meta tag will
1374 make the generated HTML documents more complete and SEO-friendly.
1376 * Citeproc: fix handling of empty URL variables (`DOI`, etc.).
1377 The `linkifyVariables` function was changing these to links
1378 which then got treated as non-empty by citeproc, leading
1379 to wrong results (e.g. ignoring nonempty URL when empty DOI is present).
1380 See jgm/citeproc#41.
1382 * Use citeproc 0.3.0.3. Fixes an issue in author-only citations when
1383 both an author and translator are present, and an issue with
1384 citation group delimiters.
1386 * Require texmath 0.12.1. This improves siunitx support in math,
1387 fixes bugs with `\*mod` family operators and arrays, and avoids
1388 italicizing symbols and operator names in docx output.
1390 * Ensure that the perl interpreter used for filters with `.pl`
1393 * MANUAL: note that textarea content is never parsed as Markdown
1397## pandoc 2.11.3.1 (2020-12-18)
1399 * Added some missing files to extra-source-files and data
1400 files, so they are included in the sdist tarball. Closes #6961.
1401 Cleaned up some extraneous data and test files, and added
1402 a CI check to ensure that the test and data files included
1403 in the sdist match what is in the git repository.
1405 * Use citeproc 0.3.0.1, which avoids removing nonbreaking
1406 space at the end of the `initialize-with` attribute. (Some
1407 journals require nonbreaking space after initials, and this
1408 makes that possible.)
1410## pandoc 2.11.3 (2020-12-17)
1412 * With `--bibliography` (or `bibliography` in metadata), a
1413 URL may now be provided, and pandoc will fetch the resource.
1414 In addition, if a file path is provided and it is not
1415 found relative to the working directory, the resource
1416 path will be searched (#6940).
1418 * Add `sourcepos` extension for `commonmark`, `gfm`, `commonmark_x`
1419 (#4565). With the `sourcepos` extension set set, `data-pos`
1420 attributes are added to the AST by the commonmark reader. No other
1421 readers are affected. The `data-pos` attributes are put on elements
1422 that accept attributes; for other elements, an enlosing Div or Span
1423 is added to hold the attributes.
1425 * Change extensions for `commonmark_x`: replace `auto_identifiers`
1426 with `gfm_auto_identifiers` (#6863). `commonmark_x` never actually
1427 supported `auto_identifiers` (it didn't do anything), because the
1428 underlying library implements gfm-style identifiers only. Attempts
1429 to add the `auto_identifiers` extension to `commonmark` will now
1434 + Split module into several submodules (Albert Krewinkel). Reducing
1435 module size should reduce memory use during compilation.
1436 + Support advanced table features (Albert Krewinkel):
1437 block level content in captions, row and colspans,
1438 body headers, row head columns, footers, attributes.
1439 + Disable round-trip testing for tables. Information for cell
1440 alignment in a column is not preserved during round-trips (Albert
1442 + Allow finer grained options for tag omission (Albert Krewinkel).
1443 + Simplify list attribute handling (Albert Krewinkel).
1444 + Pay attention to `lang` attributes on body element (#6938).
1445 These (as well as `lang` attributes on the html element) should update
1447 + Retain attribute prefixes and avoid duplicates (#6938).
1448 Previously we stripped attribute prefixes, reading `xml:lang` as
1449 `lang` for example. This resulted in two duplicate `lang`
1450 attributes when `xml:lang` and `lang` were both used. This commit
1451 causes the prefixes to be retained, and also avoids invald
1452 duplicate attributes.
1454 * Commonmark reader:
1456 + Refactor `specFor`.
1457 + Set input name to `""` to avoid clutter in sourcepos output.
1461 + Parse `#+LANGUAGE` into `lang` metadata field (#6845, Albert
1463 + Preserve targets of spurious links (#6916, Albert
1464 Krewinkel). Links with (internal) targets that the reader doesn't
1465 know about are converted into emphasized text. Information on the
1466 link target is now preserved by wrapping the text in a Span of class
1467 `spurious-link`, with an attribute `target` set to the link's
1468 original target. This allows to recover and fix broken or unknown
1473 + Table text width support (#6791, Nils Carlson).
1474 Table width in relation to text width is not natively supported
1475 by docbook but is by the docbook `fo` stylesheets through an XML
1476 processing instruction, `<?dbfo table-width="50%"?>`.
1480 + Improve parsing of command options (#6869, #6873).
1481 In cases where we run into trouble parsing inlines til the
1482 closing `]`, e.g. quotes, we return a plain string with the
1483 option contents. Previously we mistakenly included the brackets
1485 + Preserve center environment (#6852, Igor Pashev).
1486 The contents of the `center` environment are put in a `Div`
1487 with class `center`.
1488 + Don't parse `\rule` with width 0 as horizontal rule. These are
1489 sometimes used as spacers in LaTeX.
1490 + Don't apply theorem default styling to a figure inside (#6925).
1491 If we put an image in italics, then when rendering to Markdown
1492 we no longer get an implicit figure.
1496 + Handle unknown interwiki links better (#6932).
1497 DokuWiki lets the user define his own Interwiki links. Previously
1498 pandoc reacted to these by emitting a google search link, which is
1499 not helpful. Instead, we now just emit the full URL including the
1500 wikilink prefix, e.g. `faquk>FAQ-mathml`. This at least gives users
1501 the ability to modify the links using filters.
1505 + Properly handle boolean values in writing YAML metadata (#6388).
1506 + Ensure that a new csl-block begins on a new line (#6921).
1507 This just looks better and doesn't affect the semantics.
1511 + Better image handling (#6948). An image alone in its paragraph
1512 (but not a figure) is now rendered as an independent image, with an
1513 `alt` attribute if a description is supplied. An inline image that
1514 is not alone in its paragraph will be rendered, as before, using a
1515 substitution. Such an image cannot have a "center", "left", or
1516 "right" alignment, so the classes `align-center`, `align-left`, or
1517 `align-right` are ignored. However, `align-top`, `align-middle`,
1518 `align-bottom` will generate a corresponding `align` attribute.
1522 + Keep raw openxml strings verbatim (#6933, Albert Krewinkel).
1523 + Use Content instead of Element. This allows us to inject
1524 raw OpenXML into the document without reparsing it into an
1525 Element, which is necessary if you want to inject an open
1527 + Fix bullets/lists indentation, so that the first level is slightly
1528 indented to the right instead of right on the margin (cholonam).
1529 + Support bold and italic in "complex script" (#6911).
1530 Previously bold and italics didn't work properly in LTR
1531 text. This commit causes the w:bCs and w:iCs attributes
1532 to be used, in addition to w:b and w:i, for bold and
1533 italics respectively.
1537 + Fix image bounding box for custom widths/heighta (Mauro Bieg, #6936).
1541 + Improve table spacing (#6842, #6860).
1542 Remove the `\strut` that was added at the end of minipage
1543 environments in cells. Replace `\tabularnewline` with
1545 + Improve calculation of column spacing (#6883).
1546 + Extract table handling into separate module (Albert Krewinkel).
1547 + Fix bug with nested `csl-` display Spans (#6921).
1548 + Improve longtable output (#6883). Don't create minipages for
1549 regular paragraphs. Put width and alignment information in the
1550 longtable column descriptors.
1552 * OpenDocument writer:
1554 + Support for table width as a percentage of text width
1555 (#6792, Nils Carson).
1556 + Implement Div and Span ident support (#6755, Nils Carson).
1557 Spans and Divs containing an ident in the Attr will become bookmarks
1558 or sections with idents in OpenDocument format.
1559 + Add two extensions, `xrefs_name` and `xrefs_number` (#6774, Nils
1560 Carlson). Links to headings, figures and tables inside the
1561 document are substituted with cross-references that will use the
1562 name or caption of the referenced item for `xrefs_name` or the
1563 number for `xrefs_number`. For the `xrefs_number` to be useful
1564 heading numbers must be enabled in the generated document and
1565 table and figure captions must be enabled using for example the
1566 `native_numbering` extension. In order for numbers and reference
1567 text to be updated the generated document must be refreshed.
1571 + Support advanced table features (Albert Krewinkel).
1572 + Support author affiliations (#6687, Albert Krewinkel).
1576 + Use correct id attribute consistently (Jan Tojnar).
1577 DocBook5 should always use `xml:id` instead of `id`.
1578 + Handle admonition titles better (Jan Tojnar).
1579 Docbook reader produces a `Div` with `title` class for `<title>`
1580 element within an “admonition” element. Markdown writer then turns
1581 this into a fenced div with `title` class attribute. Since fenced
1582 divs are block elements, their content is recognized as a
1583 paragraph by the Markdown reader. This is an issue for Docbook
1584 writer because it would produce an invalid DocBook document from
1585 such AST – the `<title>` element can only contain “inline”
1586 elements. Handle this special case separately by unwrapping
1587 the paragraph before creating the `<title>` element.
1588 + Add XML namespaces to top-level elements (#6923, Jan Tojnar).
1589 Previously, we only added `xmlns` attributes to chapter
1590 elements, even when running with `--top-level-division=section`.
1591 These namespaces are now added to part and section elements too,
1592 when they are the selected top-level divisions.
1593 We do not need to add namespaces to documents produced with
1594 `--standalone` flag, since those will already have xmlns attribute on
1595 the root element in the template.
1599 + Fix handling of nested `csl-` display spans (#6921).
1600 Previously inner Spans used to represent CSL display attributes were
1601 not rendered as div tags as intended.
1605 + Include title page in landmarks (#6919).
1606 Note that the toc is also included if `--toc` is specified.
1607 + Add frontmatter type on body element for nav.xhtml (#6918).
1609 * EPUB templates: use preserveAspectRatio="xMidYMid" for cover image (#6895,
1610 Shin Sang-jae). This change affects both the epub2 and the epub3
1611 templates. It avoids distortion of the cover image by requiring that the
1612 aspect ratio be preserved.
1616 + Include `csquotes` package if `csquotes` variable set.
1617 + Put back `amssymb`. We need it for checkboxes in todo lists,
1618 and maybe for other things. In this location it seems compatible
1619 with the cases that prompted #6469 and PR #6762.
1620 + Disable language-specific shorthands in babel (#6817, #6887).
1621 Babel defines "shorthands" for some languages, and these can
1622 produce unexpected results. For example, in Spanish, `1.22`
1623 gets rendered as `122`, and `et~al.` as `etal`.
1624 One would think that babel's `shorthands=off` option (which
1625 we were using) would disable these, but it doesn't. So we
1626 remove `shorthands=off` and add some code that redefines
1627 the shorthands macro. Eventually this will be fixed in babel,
1628 I hope, and we can revert to something simpler.
1630 * JATS template: allow array of persistent institute ids in `pid`
1633 * Text.Pandoc.Parsing: minor code and efficiency improvements.
1635 * Text.Pandoc.Extension:
1637 + Add `Ext_sourcepos` constructor for `Extension` [API change].
1638 + Add `Ext_xrefs_name` and `Ext_xrefs_number` constructors for
1639 `Extension` (Nils Carson) [API change].
1641 * Text.Pandoc.Citeproc:
1643 + Fix truncation of `[Citation]` list in `Cite` inside footnotes (#6890).
1644 This affected author-in-text citations in footnotes. It didn't cause
1645 problems for the printed output, but for filters that expected the
1646 citation id and other information.
1647 + Allow the use of both inline and external references (#6951),
1648 as with pandoc-citeproc. References defined in the document's
1649 metadata take priority over references with the same id defined in
1650 an external bibliography.
1651 + Use `fetchItem` to get external bibliography (#6940).
1652 + Ensure that BCP47 lang codes can be used. We ignore the variants
1653 and just use the base lang code and country code when passing off
1655 + Citeproc BibTeX parser: revert change in `getRawField`
1656 which was made (for reasons forgotten) when transferring
1657 this code from pandoc-citeproc. The change led to `--` in
1658 URLs being interpreted as en-dashes, which is unwanted (#6874).
1660 * Text.Pandoc.ImageSize:
1662 + Default to DPI 72 if the format specifies DPI of 0 (#6880).
1663 This shouldn't happen, in general, but it can happen with
1664 JPEGs that don't conform to the spec. Having a DPI of 0
1665 will blow up size calculations (division by 0).
1666 + ImageSize: use JuicyPixels to determine size for png, jpeg, and
1667 gif, instead of doing our own binary parsing (#6936). This
1668 gives more reliable results.
1672 + Remove `foldOrElse` (internal module) (Albert Krewinkel).
1674 * Use skylighting 0.10.2 (#6625).
1676 * Use citeproc 0.3. This fixes issues with references with
1677 duplicate ids (jgm/citeproc#36).
1679 * Use doctemplates 0.9. This fixes issues with boolean
1680 metadata values in the Markdown writer (#6388)
1681 and in `meta-json` (#6650). It also fixes
1682 issues with nested for loops in templates.
1684 * Add translations zh-Hans.yaml and zh-Hant.yaml (#6904, #6909,
1685 Kolen Cheung, taotieren).
1687 * Add tests: True to cabal.project.
1688 This fixes some CI failures for cabal.
1690 * Normalize test/tables/*.native (#6888, Kolen Cheung).
1692 * Move executable to `app` directory to avoid problems with cabal repl.
1694 * CONTRIBUTING: add section "How can I help?" (#6892, Albert Krewinkel).
1695 Also adds a paragraph aimed at highlighting the importance of feature
1698 * MANUAL: Document that --number-sections works in `ms` (#6935).
1701## pandoc 2.11.2 (2020-11-19)
1703 * Default to using ATX (`##`-style) headings for Markdown output
1704 (#6662, Aner Lucero). Previously we used Setext (underlined) headings
1705 by default for levels 1--2.
1707 * Add option `--markdown-headings=atx|setext`, and deprecate
1708 `--atx-headers` (#6662, Aner Lucero).
1710 * Support `markdown-headings` in defaults files.
1712 * Fix corner case in YAML metadata parsing (#6823). Previously YAML
1713 metadata would sometimes not get recognized if a field ended with a
1714 newline followed by spaces.
1716 * `--self-contained`: increase coverage (#6854).
1717 Previously we only self-contained attributes for certain tag names
1718 (`img`, `embed`, `video`, `input`, `audio`, `source`, `track`,
1719 `section`). Now we self-contain any occurrence of `src`,
1720 `data-src`, `poster`, or `data-background-image`, on any tag; and
1721 also `href` on `link` tags.
1725 + Fix detection of locators following in-text citations.
1726 Prevously, if we had `@foo [p. 33; @bar]`, the `p. 33` would be
1727 incorrectly parsed as a prefix of `@bar` rather than a suffix of
1729 + Improve period suppression algorithm for citations in notes
1730 in note citation styles (#6835).
1731 + Don't increment `stateNoteNumber` for example list references.
1732 This helps with #6836 (a bug in which example list references
1733 disturb calculation of citation note number and affect when
1734 `ibid` is triggered).
1738 + Move `getNextNumber` from Readers.LaTeX to Readers.LaTeX.Parsing.
1739 + Fix negative numbers in siunitx commands. A change in pandoc 2.11
1740 broke negative numbers, e.g. `\SI{-33}{\celcius}` or `\num{-3}`.
1741 This fixes the regression.
1743 * DocBook reader: drop period in formalpara title
1744 and put it in a div with class `formalpara-title`, so that
1745 people can reformat with filters (#6562).
1747 * Man reader: improve handling of `.IP` (#6858). We now better handle
1748 `.IP` when it is used with non-bullet, non-numbered lists, creating a
1749 definition list. We also skip blank lines like groff itself.
1751 * Bibtex reader: fall back on `en-US` if locale for LANG not found.
1752 This reproduces earlier pandoc-citeproc behavior (jgm/citeproc#26).
1756 + Wrap all tables (Albert Krewinkel).
1757 All `<table>` elements are put inside `<table-wrap>` elements, as the
1758 former are not valid as immediate child elements of `<body>`.
1759 + Move Table handling to separate module (Albert Krewinkel).
1760 Adds two new unexported modules:
1761 Text.Pandoc.Writers.JATS.Types, Text.Pandoc.Writers.JATS.Table.
1765 + Replace org #+KEYWORDS with #+keywords (TEC).
1766 As of ~2 years ago, lower case keywords became the standard (though
1767 they are handled case insensitive, as always).
1768 + Update org supported languages and identifiers according to the
1769 current list contained in
1770 <https://orgmode.org/worg/org-contrib/babel/languages/index.html>
1773 * Only use `filterIpynbOutput` if input format is ipynb (#6841).
1774 Before this change content could go missing from divs with class
1775 `output`, even when non-ipynb was being converted.
1777 * When checking reader/writer name, check base name now that we permit
1778 extensions on formats other than markdown.
1780 * Text.Pandoc.PDF: Fix `changePathSeparators` for Windows (#6173).
1781 Previously a path beginning with a drive, like `C:\foo\bar`, was
1782 translated to `C:\/foo/bar`, which caused problems.
1783 With this fix, the backslashes are removed.
1785 * Text.Pandoc.Logging:
1787 + Add constructor `ATXHeadingInLHS` to `LogMessage` [API change].
1788 + Add constructor `EnvironmentVariableUndefined` to
1789 `LogMessage` [API change].
1791 * Fix error that is given when people specify `doc` output (#6834,
1794 * LaTeX template: add a `\break` after parbox in `CSLRightInline`.
1795 This should fix spacing problems between entries with numeric styles.
1796 Also fix number of params on `CSLReferences`.
1798 * reveal.js template: Put quotes around `controlsLayout`,
1799 `controlsBackArrows`, and `display`, since these require strings.
1800 Add `showSlideNumber`, `hashOneBasedIndex`, `pause`.
1802 * Use citeproc 0.2. This fixes a bug with title case around parentheses.
1804 * pandoc.cabal: remove 'static' flag.
1805 This isn't really necessary and can be misleading (e.g. on macOS,
1806 where a fully static build isn't possible). cabal's new option
1807 `--enable-executable-static` does the same. On stack you can add
1808 something like this to the options for your executable in package.yaml:
1810 ld-options: -static -pthread
1812 * Remove obsolete bibutils flag setting in `linux/make_artifacts.sh`.
1816 + Correct `link-citation` -> `link-citations`.
1817 + Add a sentence about `pagetitle` for HTML (#6843, Alex Toldaiev).
1819 * INSTALL.md: Remove references to `pandoc-citeproc` (#6857).
1821 * CONTRIBUTING: describe hlint and how it's used (#6840, Albert
1824## pandoc 2.11.1.1 (2020-11-07)
1826 * Citeproc: improve punctuation in in-text note citations (#6813).
1827 Previously in-text note citations inside a footnote would sometimes have
1828 the final period stripped, even if it was needed (e.g. on the end of
1831 * Use citeproc 0.1.1.1. This improves the decision about when
1832 to use `ibid` in cases where citations are used inside
1835 * Support `nocase` spans for `csljson` output.
1837 * Require latest commonmark, commonmark-extensions.
1838 This fixes a bug with `autolink_bare_uris` and commonmark.
1840 * LaTeX reader: better handling of `\\` inside math in table cells (#6811).
1842 * DokuWiki writer: translate language names for code elements
1843 and improve whitespace (#6807).
1845 * MediaWiki writer: use `syntaxhighlight` tag instead of deprecated
1846 `source` for highlighted code (#6810). Also support `startFrom`
1847 attribute and `numberLines`.
1849 * Lint code in PRs and when committing to master (#6790,
1852 * doc/filters.md: describe technical details of filter invocations (#6815,
1856## pandoc 2.11.1 (2020-11-03)
1858 * DocBook Reader: fix duplicate bibliography bug (#6773, Nils Carlson).
1862 + Parse contents of iframes (#6770).
1863 + Parse inline svg as image unless `raw_html` is set in the reader (in
1864 which case the svg is passed through as raw HTML) (#6770).
1868 + Fix bug parsing macro arguments (#6796). If `\cL` is
1869 defined as `\mathcal{L}`, and `\til` as `\tilde{#1}`, then
1870 `\til\cL` should expand to `\tilde{\mathcal{L}}`, but
1871 pandoc was expanding it to `\tilde\mathcal{L}`. This is
1872 fixed by parsing the arguments in "verbatim mode" when the
1873 macro expands arguments at the point of use.
1874 + Properly support optional (cite) argument for
1875 `\blockquote` from `csquotes` (#6802).
1877 * LaTeX writer: Improved calculation of table column widths.
1878 We now have LaTeX do the calculation, using `\tabcolsep`.
1879 So we should now have accurate relative column widths no
1880 matter what the text width. The default template has been modified to load
1881 the calc package if tables are used.
1883 * HTML writer: Fix duplicate "class" attribute for table
1886 * Text.Pandoc.Filter: allow shorter YAML representation of Citeproc
1887 (Albert Krewinkel). The map-based YAML representation of filters expects
1888 `type` and `path` fields. The path field had to be present for all filter
1889 types, but is not used for citeproc filters. The field can now be omitted
1890 when type is "citeproc", as described in the MANUAL.
1892 * Text.Pandoc.Error: Add `PandocBibliographyError` constructor
1893 for `PandocError` [API change]. This ensures that bibliography parsing
1894 errors generate messages that include the bibliography file name --
1895 otherwise it can be quite mysterious where it is coming from.
1897 * Citeproc: properly handle `csl` field with `data:` URI (#6783).
1898 This is used with the JATS writer, so this fixes a regression
1899 in pandoc 2.11 with JATS output and citeproc.
1901 * Allow `citation-abbreviations` in defaults file.
1903 * JATS templates: ensure `jats_publishing` output is valid
1906 * LaTeX template: Fix `CSLRightInline`, so that it does not
1907 run over the right margin.
1909 * HTML template: default CSS tweaks (Mauro Bieg and John
1912 - Fix margin before codeblock
1913 - Add `monobackgroundcolor` variable, making the background color
1914 and padding of code optional.
1915 - Ensure that backgrounds from highlighting styles take precedence over
1917 - Remove list markers from TOC
1918 - Add margin-bottom where needed
1919 - Remove italics from blockquote styling
1920 - Change borders and spacing in tables to be more consistent with other
1923 - Set font-size for print media to 12pt.
1924 - Reduce interline space.
1925 - Reduce interparagraph space.
1926 - Reduce line width.
1927 - Remove the special `line-height: 1` for table cells.
1928 - Remove the special line-height for pre.
1929 - Ensure that there is a bit more space before a heading
1931 - Slightly reduced space after title header.
1932 - Add CSS example to MANUAL
1934 * man template: Change comment that triggers `tbl` from
1935 `.\"t` to `'\" t`, as specified in groff_man(7) (#6803).
1937 * Use latest commonmark, commonmark-extensions.
1938 This fixes a bug with nested blocks in footnotes with the
1939 `footnote` extension to `commonmark`. See jgm/commonmark-hs#63.
1941 * Citeproc: use comma for in-text citations inside footnotes.
1942 When an author-in-text citation like `@foo` occurs in a footnote,
1943 we now render it with: `AUTHOR NAME + COMMA + SPACE + REST`.
1944 Previously we rendered: `AUTHOR NAME + SPACE + "(" + REST + ")"`.
1945 This gives better results. Note that normal citations are still
1946 rendered in parentheses.
1948 * Use latest citeproc:
1950 + citeproc no longer capitalizes notes, so we do it
1951 in pandoc when appropriate.
1954 * Clarify manual on `--track-changes` (#6801).
1956 * Add `doc/jats.md` to document pandoc's handling of JATS
1957 (#6794, Albert Krewinkel).
1959 * Fix code example in lua-filters.md (#6795).
1961## pandoc 2.11.0.4 (2020-10-21)
1963 * Commonmark writer: fix regression with fenced divs (#6768).
1964 Starting with 2.10.1, fenced divs no longer render with
1965 HTML div tags in commonmark output. This is a regression
1966 due to our transition from cmark-gfm. This commit fixes it.
1968 * Use released version of citeproc. (This fixes a mis-step
1969 in the 2.11.0.3 release, which is now deprecated.)
1971 * Use latest sylighting, with support for `groovy`.
1973 * Document that --html-q-tags requires the smart extension on the reader
1976## pandoc 2.11.0.3 (2020-10-20)
1978 * Use latest citeproc (closes #6765). This fixes a problem with
1979 author-in-text citations for references including both an author
1980 and an editor. Previously, both were included in the text, but only
1981 the author should be.
1983 * With `--citeproc`, ensure that the final period is removed when
1984 citations that occur in notes in note-based styles get put in
1985 parentheses. See jgm/citeproc#20.
1987 * Normalize rewritten image paths with `--extract-media` (#6761).
1988 This change will avoid mixed paths like this one when
1989 `--extract-media` is used with a Word file:
1990 ``. Instead we'll get
1991 ``.
1993 * Modify `--version` output. Use space more efficiently and report the
1994 citeproc and ipynb versions, along with skylighting, texmath, and
1995 pandoc-types. Drop the word "default" before "user data directory."
1997 * DocBook reader: bibliomisc and anchor support (#6754, Nils Carlson).
1998 Also ensure that bibliodiv without a title no longer results in
2001 * ConTeXt template: adds `\setupinterlinespace` to fonts larger
2002 than normal (#6763, Denis Maier).
2004 * LaTeX template: Do not load amssymb if not needed (#6469, Angelo Peronio).
2005 See <https://tex.stackexchange.com/a/549938>.
2007 * Relax upper bound on hslua, allow hslua-1.3.* (Albert Krewinkel).
2011 + Improve explanation of "indent" variable (#6767, Cyrus Yip).
2012 + Remove org from list of input formats supporting raw_tex (#6753,
2016## pandoc 2.11.0.2 (2020-10-15)
2018 * Fix handling of `xdata` in bibtex/biblatex bibliographies (#6752).
2020 * Fix some small typos in the API documentation (#6751, Michael Hoffmann).
2022 * Require citeproc 0.1.0.2. This fixes a regression from pandoc-citeproc
2023 involving spacing between components of a reference in certain styles
2026 * Fix typos in comments, doc strings, error messages, and tests
2027 (Albert Krewinkel, #6738).
2029## pandoc 2.11.0.1 (2020-10-13)
2031 * LaTeX reader: support more acronym commands (#6746):
2032 `\acl`, `\aclp`, and capitalized versions of already
2035 * Commonmark reader: add `pipe_table` extension *after* defaults (#6739).
2036 Otherwise we get bad results for non-table, non-paragraph
2037 lines containing pipe characters.
2039 * Markdown writer: Fix autolinks rendering for gfm (#6740).
2040 Previously, autolinks rendered as raw HTML, due to the
2041 `class="uri"` added by pandoc's markdown reader.
2045 + Escape option values in lstlistings environment (#6742).
2046 + Fix handling of `lang` `pt-BR` (#2953). For polyglossia we now
2047 use `\setmainlanguage[variant=brazilian]{portuguese}` and for babel
2048 `\usepackage[shorthands=off,main=brazilian]{babel}`.
2050 * Depend on latest citeproc (0.1.0.1).
2052 + This fixes the citation number issue with ieee.csl and other
2053 styles that do not explicitly sort bibliographies (#6741). (Pandoc
2054 was numbering them by their order in the bibliography file,
2055 rather than the order cited, as required by the CSL spec.)
2056 + Fixes groupin/collapsing with citation items with prefixes.
2058 * default.latex: fix `CSLReference` macro definition.
2060 * Fix MANUAL.txt CSL JSON conversion examples.
2062 * Fix spelling errors in changelog, MANUAL.txt, `doc/org.md` (#6738).
2065## pandoc 2.11 (2020-10-11)
2067 * Add `--citeproc` (`-C`) option to trigger built-in citation processing.
2068 It is no longer necessary to use the external `pandoc-citeproc`
2069 filter. `--citeproc` behaves like a filter and can be positioned
2070 relative to other filters as they appear on the command line.
2072 The new built-in citation processing uses the citeproc library,
2073 a reimplementation of CSL citation processing that fixes many
2074 of the shortcomings of pandoc-citeproc. In general, citation
2075 processing should work much the same as it did with pandoc-citeproc,
2076 but with greater fidelity to CSL styles and better performance.
2077 (The tests from the pandoc-citeproc package have been carried
2078 over to pandoc.) The following differences should be noted:
2080 - At this point, only some of the writers (HTML, ms, LaTeX) properly
2081 interpret CSL display styles. You should get decent output in
2082 all formats (at least as good as with pandoc-citeproc), but
2083 indentation and block-alignment may not be right.
2085 - pandoc-citeproc searches the `~/.csl` directory for `.csl`
2086 styles. Pandoc instead searches the `csl` subdirectory of
2087 the pandoc user data directory (e.g., `~/.pandoc/csl` or
2088 `~/.local/share/pandoc/csl`). Users who already keep
2089 CSL styles in `~/.csl` may find it convenient to add a
2092 - Some of the bibliography formats supported by pandoc-citeproc (via
2093 hs-bibutils) are no longer supported: Copac, EndNote,
2094 ISI, MEDLINE, MODS, and RIS. If you use one of these formats,
2095 you may use the `bibutils` utility to convert to BibLaTeX.
2096 We now support only BibTeX, BibLaTeX, CSL JSON,
2097 and pandoc's YAML/Markdown analogue of CSL JSON.
2099 - pandoc-citeproc would always retrieve the independent parent
2100 of a dependent style by doing an HTTP request. pandoc will
2101 now first seek the independent parent locally (in the resource
2102 path or in the `csl` subdirectory of the pandoc user data
2103 directory) before resorting to HTTP. In addition, you may
2104 omit the `.csl` extension, e.g. `--csl zoology`.
2106 - Using the `--bibliography` option (or including
2107 `bibliography` in YAML metadata) no longer triggers
2108 citation processing implicitly: one must always use the
2109 `--citeproc` option if citation processing is wanted.
2111 * Add `csljson` as and input and output format. This allows pandoc
2112 to convert between `csljson` and other bibliography formats
2113 (e.g. `-f csljson -t markdown -s` or `-f bibtex -t csljson`),
2114 and to generate formatted versions of CSL JSON bibliographies
2115 (e.g., `pandoc -f csljson --citeproc pl.json -o pl.pdf`).
2117 * Added `bibtex`, `biblatex` as input formats. This allows pandoc
2118 to convert between BibLaTeX and BibTeX and other bibliography formats,
2119 and to generated formatted versions of BibTeX/BibLaTeX bibliographies
2120 (e.g., `pandoc -f biblatex --citeproc pl.bib -o pl.pdf`).
2122 * Raise informative errors when YAML metadata parsing fails (#6730).
2123 Previously the command would succeed, returning empty metadata,
2124 with no errors or warnings.
2126 * Sort languages in `--list-highlight-languages` output (#6718,
2127 Albert Krewinkel). Languages were previously sorted by their
2128 long name, which leads to unexpected results).
2130 * Add CSS to default HTML template (#6601, Mauro Bieg). This
2131 greatly improves the default typography in pandoc's HTML
2132 output. The CSS is sensitive to a number of variables
2133 (e.g. `mainfont`, `fontsize`, `linestretch`): see the manual for
2134 details. To restore the earlier, more spartan output, you can
2135 disable this with `-M document-css=false`.
2137 * Support `--toc-depth` option for ODT writer (#6696, niszet).
2139 * Fix issues with Windows UNC paths with some options (#5127).
2141 * Remove `fenced_code_blocks` and `backtick_code_blocks` from allowed
2142 `commonmark` and `gfm` extensions. These shouldn't really be counted
2143 as extensions, because they can't be disabled in commonmark.
2144 Adjust markdown writer to check for the commonmark variant in addition
2147 * Add these extensions to `gfm` and `commonmark`:
2148 `fenced_code_blocks`, `backtick_code_blocks`, `fenced_code_attributes`.
2149 These can't really be disabled in the reader, but they need to be enabled
2150 in the writer or we just get indented code.
2152 * Make sure proper set of extensions is recognized for `commonmark_x`.
2154 * Allow `gfm_auto_identifiers`, `ascii_identifiers` extensions for `docx`.
2158 + Add `Maybe FilePath` parameter to `yamlToMeta` [API change].
2159 + Export `yamlToRefs` [API change], a version of `yamlToMeta`
2160 specialized to references.
2161 + Set `citationNoteNum` accurately in citations.
2162 + Revise abbreviation support. Don't insert a nonbreaking space after a
2163 potential abbreviation if it comes right before a note or citation.
2164 This causes problems for citeproc's moving of note citations.
2168 + Support missing siunitx commands (#6658).
2169 + Support `squared`, `cubed`, `tothe` in siunitx (#6657).
2170 + Improved uncertainty handling in slunitx.
2171 + Factored out siunitx stuff into separate unexposed module.
2172 + Fix improper empty cell filtering (#6689, Christian Despres).
2173 + Fix parsing of "show name" in `\newtheorem` (#6734).
2174 Previously we were just treating it as a string and
2175 ignoring accents and formatting.
2176 + Prevent wrong nesting of `\multirow` and `\multicolumn` table
2177 cells (#6603, Laurent P. René de Cotret).
2178 + Table cell parser not consuming spaces correctly (#6596,
2179 Laurent P. René de Cotret).
2180 + Change `SIRange` to `SIrange` (#6617, Emerson Harkin).
2181 + Allow blank lines inside `\author` (#6324).
2185 + Don't squelch space at end of emphasis and other inline elements;
2186 instead, move it outside the element (#6719).
2187 + Implement table cell alignment (#6698, Nils Carlson).
2188 + Implement column span support for tables (#6492, Nils Carlson).
2189 + Update list of block level tags (#6610).
2194 + Don't squelch space at end of emphasis and other inline elements;
2195 instead, move it outside the element (#6719).
2199 + Apply `.. class::` directly to following Header rather than creating
2200 a surrounding Div (#6699).
2204 + Allow empty dates in comments and tracked changes (#6726, Diego
2209 + Be less aggressive about using quotes for YAML values,
2210 allowing e.g. a quotation mark or bracket as long as it's not at the
2211 beginning of the line.
2212 + Use double quotes for YAML metadata (#6727).
2213 + Sort YAML metadata keys in Markdown output case-insensitive.
2218 + Support asciidoctor's block figures (#6538, argent0).
2222 + Fix spacing issue with list in definition list.
2223 When a list occurs at the beginning of a definition list definition,
2224 it can start on the same line as the label, which looks bad.
2225 Fix that by starting such lists with an `\item[]`.
2229 + Support intermediate table headers (#5314, Albert Krewinkel).
2230 + Support attributes on all table elements (Albert Krewinkel).
2231 + Render table footers if present (#6314, Albert Krewinkel).
2232 + Fix addition of `doc-biblioentry` role.
2233 + Support colspans and rowspans in HTML tables (#6312, Albert Krewinkel).
2237 + Support internal document links (#5541, Leonard Rosenthol).
2238 + Changed default link state to invisible (#6676, Leonard Rosenthol).
2242 + Better handle list items whose contents are lists (#5948, Michael
2243 Hoffmann). If the first element of a bulleted or ordered list is
2244 another list, then that first item would previously disappear if the
2245 target format is docx.
2246 + Separate adjacent tables (#4315). Word combines adjacent tables,
2247 so to prevent this we insert an empty paragraph between two
2252 + Don't force blank line after headings (#6554).
2254 * OpenDocument writer:
2256 + Implement table cell alignment (#6700 Nils Carson, Mauro Bieg).
2257 + New table cell support with row and column spans (#6682, Nils Carson).
2258 + Syntax highlighting for inline code (#6711, niszet).
2260 * Add Text.Pandoc.Citeproc module, exporting `processCitations`
2261 [API change]. This depends on several other, unexported
2262 modules under Text.Pandoc.Citeproc.
2264 * Add module Text.Pandoc.Writers.CslJson, exporting `writeCslJson`.
2267 * Add module Text.Pandoc.Readers.CslJson, exporting `readCslJson`.
2270 * Add module Text.Pandoc.Readers.BibTeX, exporting `readBibTeX` and
2271 `readBibLaTeX`. [API change]
2273 * Text.Pandoc.Filter: Add `CiteprocFilter` constructor to Filter.
2274 [API change] This runs the processCitations transformation.
2275 We need to treat it like a filter so it can be placed
2276 in the sequence of filter runs (after some, before others).
2277 In FromYAML, this is parsed from `citeproc` or `{type: citeproc}`,
2278 so this special filter may be specified either way in a defaults file
2279 (or by `citeproc: true`, though this gives no control of positioning
2280 relative to other filters).
2282 * Add new exported module Text.Pandoc.Writers.AnnotatedTable [API change]
2283 (#6655, Christian Despres). This module (which should generally
2284 be imported qualified to avoid name conflicts) provides a
2285 `Table` type that mirrors the structure of a pandoc `Table`,
2286 but with added inferred information so that the writers do not have to
2287 lay out tables themselves. The `toTable` and `fromTable` functions convert
2288 between an annotated `Table` and a regular pandoc `Table`. In addition to
2289 producing a `Table` with coherent and well-formed annotations, the
2290 `toTable` function also normalizes its input table like the table
2291 builder does. Tests ensure that `toTable` normalizes tables exactly
2292 like the table builder, and that its annotations are coherent.
2294 * Text.Pandoc.Logging:
2296 + Remove unused `CouldNotParseYamlMetadata` constructor for `LogMessage`
2298 + Add `CiteprocWarning` constructor to `LogMessage` [API change].
2300 * Text.Pandoc.Readers.Metadata: export `yamlBsToRefs` [API change].
2301 These allow specifying an id filter so we parse only references
2302 that are used in the document.
2304 * Text.Pandoc.Parsing:
2306 + Export ParseError [API change].
2307 + Add `stateInNote` and `stateNoteNumber` to `ParserState`
2308 [API change]. These are used to populate note numbers for citation
2311 * Fix apparent typos in sample.lua (#6729, William Lupton).
2312 Also make the writer less aggressive in escaping quotes.
2314 * Text.Pandoc.Options:
2316 + `defaultMathJaxURL`: use `tex-chtml-full` instead of `tex-mml-chtml`
2317 (#6599, Kolen Cheung). This drops the MathML support (which we
2318 don't need for HTML math rendering) and includes the full JavaScript,
2319 which makes it possible to use `--self-contained` (though there may
2320 still be issues if the required math fonts aren't available). This
2321 change should also reduce latency in pages with lots of formulas.
2322 + Add `/tex-chtml-full.js` to `defaultMathJaxURL` (#6593) Previously we
2323 added this in processing command line options, but not in processing
2324 defaults files, which was inconsistent.
2326 * epub.css: Fix cover page selectors and add note explaining their use
2329 * Add data files needed for Text.Pandoc.Citeproc: these include
2330 `default.csl` in the data directory and a `citeproc` directory that
2331 is only used at compile-time for biblatex localizations. Note that we've
2332 added `file-embed` as a mandatory rather than a conditional
2333 dependency, because of the biblatex localization files.
2337 + Add SimpleTable for backwards compatibility (#6575, Albert Krewinkel).
2338 A new type `SimpleTable` is made available to Lua filters. It is
2339 similar to the `Table` type in pandoc versions before 2.10;
2340 conversion functions from and to the new Table type are provided.
2341 Old filters using tables now require minimal changes and can use, e.g.,
2343 if PANDOC_VERSION > {2,10,1} then
2344 pandoc.Table = pandoc.SimpleTable
2349 function Table (tbl)
2350 tbl = pandoc.utils.to_simple_table(tbl)
2352 return pandoc.utils.from_simple_table(tbl)
2355 to work with the current pandoc version.
2357 + Make `attr` argument optional in `Table` constructor (Albert Krewinkel).
2358 This changes the Lua API. It is highly unlikely for this change to affect
2359 existing filters, since the documentation for the new Table constructor
2360 (and type) was incomplete and partly wrong before. The Lua API is now
2361 more consistent, as all constructors for elements with attributes now
2362 take attributes as the last parameter.
2366 * Add a dedicated Citations section which consolidates the information
2367 the manual used to contain about citation processing, and incorporates
2368 some information formerly found in the pandoc-citeproc man page.
2369 + Add note about lualatex using `selnolig`.
2370 + Remove duplicate `seriespage` (#6568, Blake Eryx).
2371 + Remove lists of support extensions for markdown variants (#6604).
2372 Instead, offer the advice to use `--list-extensions=FORMAT`.
2373 + Fix position of attributes in header (Albert Krewinkel).
2374 + Delete obsolete section on compact and loose lists (#6684).
2377 * doc/lua-filters.md:
2379 + Add info on how to debug Lua filters (#6732, Ian Max Andolina).
2380 + Document Underline type and constructor (Albert Krewinkel).
2381 + Document `body` field (Albert Krewinkel).
2382 + Add missing header attribute
2383 + Add missing Link.title field (Albert Krewinkel).
2384 + Make the setting-the-date example conditional (the-solipsist).
2385 This makes the example a bit more realistic.
2386 + Remove outdated link table example.
2390 + Add section on tables (Albert Krewinkel).
2391 + Add section on handling of unknown directives (Albert Krewinkel).
2393 * CONTRIBUTING.md: fix typo (#6584, Dmitry Volodin).
2395 * Use golden test framework for command tests. This means that
2396 `--accept` can be used to update expected output.
2398 * Use the `smart` extension when generating pandoc's man page (#6613).
2400 * Release-candidate: don't build windows i386.
2401 So far we haven't been able to figure out how to get
2402 stack to use a 32-bit ghc.
2404 * Use `null` instead of deprecated `Builder.isNull`.
2408 + Fix macospkg target to fetch target from S3 artifacts.
2409 + Fix pandoc-templates target to include all partials.
2411 * Remove duplicated dependency in pandoc.cabal (#6591, Felix Yan).
2413 * Sort build depends in pandoc.cabal alphabetically (#6691,
2416 * Add .travis.yml for macos release candidate build (#6622).
2417 We need to build the release candidate on Travis rather
2418 than GitHub actions, because GH has macos 10.15, and
2419 binaries compiled on that OS will not work with 10.13.
2420 This build is only triggered on `rc/*` branches.
2422 * Remove instructions for building pandoc-citeproc from CI and
2423 release binary build instructions. We will no longer distribute
2426 * Fix math rendering in trypandoc (this broke after
2427 commit d8ad766d17603784b86fc5c2e1b22864125d04d1).
2429 * Use latest versions of skylighting, commonmark (#6589),
2430 comonmark-extensions, commonmark-pandoc, texmath.
2432 * Relax version bounds for hslua, hslua-module-text, bytestring.
2434 * Use released pandoc-types 1.22. This changes the JSON
2435 encoding slightly for the new table types introduced
2436 in 1.21, so they're more consistent with the rest.
2437 Developers of libraries for pandoc filters will want
2440 * Fix hlint suggestions, update hlint.yaml (#6680, Christian Despres).
2442 * Code cleanup (#6678, Joseph C. Sible).
2444 * Add haddocks to functions in Text.Pandoc.Writers.Shared (Albert
2447 * Remove duplicate `tshow` definition.
2449 * Linux release candidate build: use ghc-musl container. This simplifies
2450 our build process (over using a customized alpine container).
2453## pandoc 2.10.1 (2020-07-23)
2455 * Add `commonmark_x` output format. This is `commonmark` with a number
2456 of useful pandoc extensions enabled.
2458 * Many more extensions now work with `commonmark` and `gfm`.
2460 * Add generic `attributes` extension. This allows attributes to
2461 be added to any block or inline element in a uniform way. Since
2462 the Pandoc AST doesn't include attributes on each element type,
2463 the attributes will sometimes be added by creating a surrounding
2464 Div or Span container. Currently this extension is only
2465 compatible with the `commonmark` and `gfm` readers.
2467 To add an attribute to a block-level element, e.g. a paragraph,
2468 put it before the block:
2471 This is a paragraph.
2473 Multiple attributes may be used and will be combined:
2476 {.blue .warning key="val"}
2477 This is a paragraph.
2479 To add an attribute to an inline-level element, put it
2480 immediately after the element:
2482 *emphasized text*{.special}
2485 * Support `--number-sections` for docx output (#1413).
2489 + Support `\SIRange` reader (#6418, Emerson Harkin).
2490 + Support table col-span and row-span (#6311, Laurent P. René de Cotret).
2491 Supports `\multirow` and `\multicolumn`.
2492 + Support amsthm: `\newtheorem`, `\theoremstyle`, and theorem and
2493 proof environments, including labels and references. The only thing
2494 that is unsupported is the second optional argument, which causes
2495 numbering to be reset after the specified series is incremented.
2496 + Moved some code to T.P.LaTeX.Parsing. We need to reduce the size
2497 of the LaTeX reader to ease compilation on resource-limited systems.
2501 + Fix csv tables with multiline cells (#6549).
2502 + Fix spurious newlines in some attributes from directives.
2503 + Avoid extra newline in included code blocks.
2505 * Commonmark reader:
2507 + Switch from cmark-gfm to commonmark-hs for commonmark and gfm
2508 parsing. This avoids depending on a C library
2509 and allows us to support more pandoc extensions for
2510 `commonmark` and `gfm`.
2514 + Parse releaseinfo as metadata (#6542).
2518 + Only use `bCs/iCs` on runs with `rtl` or `cs` property (#6514, Nikolay
2520 + Code cleanup/refactoring (Nikolay Yakimov).
2522 * Org reader (Albert Krewinkel):
2524 + Respect export setting which disables entities
2525 MathML-like entities, e.g., `\alpha`, can be disabled with the
2526 `#+OPTION: e:nil` export setting (Albert Krewinkel).
2527 + Respect export setting disabling footnotes. Footnotes can be removed
2528 from the final document with the `#+OPTION: f:nil` export setting.
2529 + Respect tables-excluding export setting. Tables can be removed from
2530 the final document with the `#+OPTION: |:nil` export setting.
2534 + Move `asciify` out of `escapeString`. Otherwise `unsmartify`
2535 doesn't catch quotes that have already been turned to entities.
2536 + Add `writeCommonmark` (new exported function, API change).
2537 + Use unicode super/subscript characters when possible if the
2538 `superscript` or `subscript` extension or `raw_html` aren't available.
2539 + Render caption as following paragraph when `table_caption` extension
2541 + Use numerical labels for reference links that are longer
2542 than 999 characters or contain square brackets, for conformity
2543 with commonmark (#6560).
2545 * Commonmark writer:
2547 + Instead of using cmark-gfm, use `writeCommonmark` from the
2548 Markdown writer. This function calls the markdown writer
2549 with appropriate extensions and a few small modifications
2550 (e.g. not requiring backslashes before spaces inside
2551 super/subscripts). With this change `comonmark` and
2552 `gfm` output can be used with a wider selection of
2555 * Jira writer: keep image caption as alt attribute (#6529, Albert
2560 + Improve alt-text/caption handling for HTML5 (#6491, Albert Krewinkel).
2561 Screen readers read an image's `alt` attribute and the figure caption,
2562 both of which come from the same source in pandoc. The figure caption is
2563 hidden from screen readers with the `aria-hidden` attribute. This
2564 improves accessibility. For HTML4, where `aria-hidden` is not allowed,
2565 pandoc still uses an empty `alt` attribute to avoid duplicate contents.
2569 + Fix code highlighting with blank lines. Previously blank lines
2570 were simply omitted from highlighted code.
2571 + Escape starting periods in ms writer code blocks (#6505, Michael
2572 Hoffmann). If a line of ms code block output starts with a period (.),
2573 it should be prepended by `\&` so that it is not interpreted as a roff
2576 * Text.Pandoc.Extensions:
2578 + Add `raw_markdown` extension (which only affects `ipynb`
2580 + Trim down `githubMarkdownExtensions`.
2581 Previously it included all of the following, which make
2582 sense for the legacy `markdown_github` but not for `gfm`,
2583 since they are part of base commonmark and thus
2584 can't be turned off in `gfm`:
2586 - `Ext_all_symbols_escapable`
2587 - `Ext_backtick_code_blocks`
2588 - `Ext_fenced_code_blocks`
2589 - `Ext_space_in_atx_header`
2590 - `Ext_intraword_underscores`
2591 - `Ext_lists_without_preceding_blankline`
2592 - `Ext_shortcut_reference_links`
2594 These have been removed from `githubMarkdownExtensions`, though
2595 they're still turned on for legacy `markdown_github`.
2596 + Add `Ext_attributes` constructor for `Extension` [API change].
2598 * LaTeX template: use selnolig to selectively suppress ligatures with
2601 * Benchmark bytestring readers (Nikolay Yakimov).
2605 + Update using-the-pandoc-api.md (favonia).
2606 + Fix Typos in lua-filters.md (tajmone).
2607 + Rewrite Raw HTML/TeX section in MANUAL.txt to avoid duplicate
2608 headings for the extensions.
2609 + Fix typo in MANUAL.txt (Benjamin Wuethrich).
2610 + Remove duplicate 'titlepage' in MANUAL.txt (Blake Eryx).
2611 + CONTRIBUTING.md: Advertise the official nightlies in GitHub actions.
2612 Replaces #6500, thanks to @ickc.
2616## pandoc 2.10 (2020-06-29)
2618 * Use pandoc-types 1.21. This adds two things:
2620 + A native Underline constructor for Inline (#6277, Vaibhav Sagar).
2621 + More expressive types for tables (#1024, Christian Despres).
2622 Tables can now take attributes; and rowspans and colspans,
2623 column headers, multiple row headers, table head and foot
2624 can all be represented. (Note, however, that reader and
2625 writer support for these features is still lacking, so
2626 most users won't see any differences in table conversion
2627 yet. These changes just lay the foundation for further
2630 * Support new Underline element in readers and writers (#6277,
2633 * Support new Table type (Christian Despres).
2634 The Builder.simpleTable now only adds a row to the TableHead
2635 when the given header row is not null. This uncovered an
2636 inconsistency in the readers: some would unconditionally
2637 emit a header filled with empty cells, even if the header
2638 was not present. Now every reader has the conditional
2639 behaviour. Only the XWiki writer depended on the header row
2640 being always present; it now pads its head as necessary.
2642 * Add an option to disable certificate validation (#6156, Cédric Couralet,
2643 Cécile Chemin, Juliette Fourcot). This commit adds the option
2644 `--no-check-certificate`, which disables certificate checking when
2645 resources are fetched by HTTP.
2647 * Unify defaults and markdown metadata parsers (#6328, Nikolay
2648 Yakimov). Clean up code in Text.Pandoc.Readers.Metadata and
2649 properly handle errors in `yamlToMeta`. This fixes parsing
2650 of Boolean fields in metadata withinin defaults files and reduces
2655 + Implement `<procedure>` (#6442, Mathieu Boespflug).
2656 + Implement `<phrase>` (#6438, Mathieu Boespflug).
2657 + Treat envar and systemitem like code (#6435, Mathieu Boespflug).
2658 + Implement `<replaceable>` (#6437, Mathieu Boespflug)
2659 + Map `<simplesect>` to unnumbered section (#6436, Mathieu Boespflug).
2663 + Handle "label" element in section title (#6288).
2664 + Parse abstract element into metadata field of same name
2665 (#6480, Albert Krewinkel).
2667 * Jira reader (Albert Krewinkel):
2669 + Resolve multiple parsing problems, including issues with empty
2670 table cells, faulty recognition of closing emphasis characters, and
2671 parsing of image attributes (#6212, #6219, #6220).
2672 + Two consecutive markup chars are now parsed verbatim (#6343);
2673 styled text must not be empty.
2674 + Newlines are no longer allowed within styled text (#6325).
2675 + Links to anchors are now parsed as links (#6407).
2676 + Retain image attributes (#6234). Jira images attributes as in
2677 `!image.jpg|align=right!` are retained as key-value pairs. Thumbnail
2678 images, such as `!example.gif|thumbnail!`, are marked by a
2679 `thumbnail` class in their attributes.
2680 + Use Underline for inserted text (#6237). Previously, the span was
2681 marked with the non-standard class `inserted`.
2682 + Improve icon conversion for `(/)`, `(x)`, `(!)`, `(?)`
2683 `(+)`, `(-)`, `(off)`, `(*)`. (#6236, #6264).
2684 + Support citations, attachment links, and user links (#6231, #6238,
2686 + Resolve parsing issues of blockquote, color (#6233, #6235).
2690 + Parse attributes into table attributes.
2691 + Support `<bdo>` (#5794, Tristan de Cacqueray).
2692 + Add `summary` to list of block-level HTML tags (#6385).
2693 This improves support for summary/details inside Markdown.
2694 NOTE: you need to include a blank line before the closing
2695 `</details>`, if you want the last part of the content to
2696 be parsed as a paragraph.
2697 + Fix parsing unclosed th elements in a table (#6247).
2699 * Commonmark reader: Implement `implicit_figures` extension (#6350).
2703 + Fix inline code in lists (#6284, Nikolay Yakimov).
2704 Previously inline code containing list markers was sometimes
2706 + Don't require blank line after grid table (#6481).
2707 This allows grid tables to be enclosed in fenced divs with no
2708 intervening blank lines.
2712 + Don't parse beyond `\end{document}` (#6380).
2713 This required some internal changes to `\subfile` handling.
2714 + Better handling of `\lettrine`. SmallCaps instead of Span
2715 for the part after the initial capital. Ensure that both
2716 arguments are parsed, so that in Markdown both
2717 are treated as raw LateX. (Closes #6258.)
2719 * Org reader (Albert Krewinkel):
2721 + Recognize images with uppercase extensions (#6472).
2722 + Keep unknown keyword lines as raw org. The lines of unknown
2723 keywords, like `#+SOMEWORD: value` are no longer read as metadata,
2724 but kept as raw `org` blocks. This ensures that more information
2725 is retained when round-tripping org-mode files; additionally,
2726 this change makes it possible to support non-standard org
2727 extensions via filters.
2728 + Unify keyword handling. Handling of export settings and other
2729 keywords (like `#+LINK`) has been combined and unified.
2730 + Support `LATEX_HEADER_EXTRA` and `HTML_HEAD_EXTRA`
2731 settings. These export settings are treated like their non-extra
2732 counterparts, i.e., the values are added to the `header-includes`
2734 + Allow multiple `#+SUBTITLE` export settings. The values of all
2735 lines are read as inlines and collected in the `subtitle`
2737 + Read `#+INSTITUTE` values as text with markup. The value is
2738 stored in the `institute` metadata field and used in the
2739 default beamer presentation template.
2740 + The behavior of the `#+AUTHOR` and `#+KEYWORD` export
2741 settings has changed: Org now allows multiple such lines
2742 and adds a space between the contents of each line. Pandoc
2743 now always parses these settings as meta inlines; setting
2744 values are no longer treated as comma-separated lists.
2745 Note that a Lua filter can be used to restore the previous
2747 + Read description lines as inlines (#6485). `#+DESCRIPTION` lines
2748 are now treated as text with markup. If multiple such
2749 lines are given, then all lines are read and separated by soft
2751 + Honor tex export option (#4070). The `tex` export option can be set
2752 with `#+OPTION: tex:nil` and allows three settings:
2753 `t` (the default) causes LaTeX fragments to be parsed as TeX or added
2754 as raw TeX. `nil` removes all LaTeX fragments from the document.
2755 `verbatim` treats LaTeX as text.
2760 + Pass arbitrary attributes through in code blocks (#6465).
2761 Exceptions: name (which becomes the id), class (which becomes the
2762 classes), and number-lines (which is treated specially to fit
2763 with pandoc highlighting).
2764 + Handle `date::` directive (#6276).
2766 * Textile reader: support `pre.` for code blocks (#6454).
2770 + Implement `raw_markdown` extension (#5408). Specifying
2771 `-f ipynb+raw_markdown` will cause Markdown cells
2772 to be represented as raw Markdown blocks, instead of being
2773 parsed. This is not what you want when going from `ipynb`
2774 to other formats, but it may be useful when going from `ipynb`
2775 to Markdown or to `ipynb`, to avoid semantically insignificant
2776 changes in the contents of the Markdown cells that might
2777 otherwise be introduced.
2778 + Handle application/pdf output as image (#6430).
2779 + Properly handle image/svg+xml as an image (#6430).
2783 + Distinguish between docx parsing and docx container unpacking errors.
2787 + Fix `gfm_auto_identifiers` so that `-` is not replaced by `_` (#6335).
2791 + Add nested syntax highlighting (#6256, Vlad Hanciuta).
2792 Nested syntaxes are specified like this:
2798 The preformatted code block parser has been extended to check if the
2799 first attribute of the block is not a `key=value` pair, and in that
2800 case it will be considered as a class.
2803 * Jira writer (Albert Krewinkel):
2805 + Always escape braces (#6478). Braces are now always escaped, even
2806 within words or when surrounded by whitespace. Jira and
2807 Confluence treat braces specially.
2808 + Convert Underline to inserted text (`+inserted+`).
2809 + Add image attributes (#6234). Image attributes are added
2810 to the output as image parameters. If the image has a
2811 class "thumbnail", then a thumbnail image is generated;
2812 all other attributes are discarded in this case.
2816 + Ensure that `-M csquotes` works even in fragment mode (#6265).
2817 + Escape `^` specially for listings (#6460).
2818 + Create hypertarget for links with identifier (#6360).
2819 + Distinguish between single and double quotes when using enquote
2820 package (#6457, dbecher-ito).
2821 + Add support for customizable alignment of columns in beamer (#6331,
2823 + Add support for customizable alignment of columns in beamer
2824 (#4805, #4150, andrebauer).
2828 + Use CSS in favor of `<br>` for display math (#6372)
2829 Some CSS to ensure that display math is
2830 displayed centered and on a new line is now included
2831 in the default HTML-based templates; this may be
2832 overridden if the user wants a different behavior.
2836 + Clean-up Div handling (Albert Krewinkel).
2840 + Enable column and row bands for tables (#6371).
2841 This change will not have any effect with the default style.
2842 However, it enables users to use a style (via a reference.docx)
2843 that turns on row and/or column bands.
2845 * OpenDocument (and ODT) writer:
2847 + Add custom-style "Abstract" in metadata abstract.
2848 This ensures that the abstract is rendered with style Abstract.
2849 + Enable custom-style attribute on a Div.
2850 This allows you to apply a custom style to contained paragraphs.
2854 + Add id of figure to enclosed image.
2855 + Add personname element to docbook author (#6244).
2859 + Properly handle cover-image containing spaces (#6391).
2863 + Ensure consistent padding for pipe tables (#6240).
2864 + Avoid unnecessary escapes before intraword `_` when
2865 `intraword_underscores` extension is enabled (#6296).
2869 + Properly handle images with same alt text (#6194).
2870 Previously we created duplicate references for these in rendering RST.
2874 + Add blank line after Div (#6308).
2878 + Support Haddock tables (Joe Hermaszewski).
2879 See this PR on Haddock for details on the table format:
2880 <https://github.com/haskell/haddock/pull/718>.
2882 * PowerPoint writer (Jesse Rosenthal):
2884 + Write math input verbatim in speaker notes (#6301).
2885 OMML in speaker notes would lead to corrupt PowerPoint output. We now
2886 output the OMML verbatim as LaTeX in the speaker notes.
2888 * LaTeX template: Make polyglossia package options list-aware
2889 (#6444, Frederik Elwert).
2891 * Reveal.js template:
2893 + Update template for reveal.js 4.0.0 (#6390, Salim B).
2894 + Update template with newly available options (#6347, Jake Zimmerman).
2895 + Use CDN version of revealjs v4 by default (#6408).
2897 * opendocument template: Add abstract and subtitle to opendocument
2900 * reference.odt: clean up styles. Add Abstract.
2901 Change Author, Date to centered paragraphs with no character
2904 * epub.css: wrap overlong lines in highlighted code blocks (#6242).
2905 This fixes a problem in iBooks v2.4 with our earlier
2906 horizontally scrolling code blocks. The problem seems to
2907 be a bug in iBooks, not pandoc, but since iBooks is a major
2908 target we're changing pandoc's default behavior so that
2909 pandoc-produced epubs work on that platform.
2913 + Use `--enable-local-file-access` in invoking `wkhtmltopdf` (#6474).
2914 `wkhtmltopdf` changed in recent versions to require this for
2915 access to local files. This fixes PDF via HTML5 with `--css`.
2916 + Send verbose output to stderr, not stdout (#6483).
2918 * Text.Pandoc.MIME: Fix MIME type for TrueType fonts in EPUBs
2919 (#6464, Michael Reed).
2921 * Text.Pandoc.Shared:
2923 + `makeSections`: omit number attribute when unnumbered class
2924 is present (#6339). Previously the attribute was included but given
2925 an empty value, and this caused the table of contents creation
2926 functions in Text.Pandoc.Writers.Shared to think these items had
2927 numbers, which meant that they were included in the TOC even if the
2928 `unlisted` class was used.
2929 + Deprecate `underlineSpan` in Shared in favor of
2930 `Text.Pandoc.Builder.underline` (Vaibhav Sagar).
2931 + `renderTags'`: use self-closing tag for col element (#6295).
2933 * Text.Pandoc.UUID: Fix `getRandomUUID`, which previously would
2934 return the same value twice in a row. Make `getRandomUUID`
2935 polymorphic in PandocMonad. Remove `getUUID` (#6228, Joseph C. Sible).
2937 * Text.Pandoc.Class: Generalize `PandocIO` functions to `MonadIO`.
2939 * Fixed Katex standalone script (#6399, Lucas Escot).
2940 Global macros are now persistent when using the HTML Writer with the
2943 * Lua subsystem (Albert Krewinkel):
2945 + Use new type PandocLua for all pandoc Lua operations (API change).
2946 The new type `PandocLua` is an instance of the `PandocMonad` typeclass
2947 and can thus be used in a way similar to `PandocIO`.
2948 + Use PandocError for exceptions in Lua subsystem (API change).
2949 The PandocError type is used throughout the Lua subsystem. All Lua
2950 functions throw an exception of this type if an error occurs. The
2951 `LuaException` type is removed and no longer exported from
2952 `Text.Pandoc.Lua`. In its place, a new constructor `PandocLuaError`
2953 is added to PandocError.
2955 * Lua filters: improve error messages for failing filters (#6332,
2956 Albert Krewinkel). Print the Lua error properly instead of
2957 displaying their `show` string.
2959 * Use latest skylighting. This fixes a bug with lua multiline
2960 comments (and may improve handling of other syntaxes as well).
2961 IT also adds `aria-hidden="true"` to the empty a elements, which
2962 helps people who use screen readers.
2964 * Use latest texmath.
2966 * Require latest doctemplates 0.8.2.
2967 This adds support for template pipes `first`, `rest`,
2968 `last`, `allbutlast`.
2970 * Revert 0e48a02 and dependency on base-noprelude, which hasn't
2971 been updated for ghc 8.10 (see #6187).
2973 * Dependency adjustments:
2975 + Allow haddock-library 1.9.x.
2976 + Allow hslua 1.1 (#6243, Felix Yan).
2977 + Allow base64-bytestring 1.1.
2978 + Use latest jira-wiki-markup.
2979 + Allow http-client 0.7.
2980 + Allow tasty 1.3.x.
2981 + Allow aeson 1.5 (#6400, Felix Yan).
2982 + Remove unused dependency `vector` (#6462, Laurent P. René de Cotret).
2983 + Bump QuickCheck upper bound.
2985 * Significant code cleanup and simplification (Joseph C. Sible, #6223,
2986 #6209, #6225, #6229, #6226, #6340).
2988 * Remove unnecessary hlint ignores (#6341, Joseph C. Sible).
2990 * Remove obsolete RelaxedPolyRec extension (#6487, Nikolay Yakimov).
2992 * trypandoc improvements (Mike Tzou):
2994 + Add standalone option to the command text (#6210).
2995 + Update third party libraries.
2999 + Clarify template partial naming (#6476, Mauro Bieg).
3000 + Describe `jira` as "Jira/Confluence wiki markup" (#6351, Albert
3001 Krewinkel). In the past, Jira's wiki markup was also used by -- and
3002 could be imported into -- Atlassian Confluence.
3003 + Add link to print-css.rocks (#6272, Mauro Bieg).
3004 + Clarify pipe table column width adjustment (#6254).
3005 + Fix ATX header syntax.
3006 + Fix misleading note about image size conversions (#6353).
3007 + Update links to reveal.js documentation (#6386, Salim B).
3008 + Separate adjacent verbatim code blocks (#6307, tom-audm).
3012 + Document behavior of `smart` extension (#4387, Albert Krewinkel).
3013 + Describe all supported export options in detail.
3017 + Fix description of BulletList Lua type (Levi Gruspe).
3018 + Use pandoc.system module in TikZ example (Albert
3019 Krewinkel). Showcase temporary directory handling with
3020 `with_temporary_directory` and `with_working_directory`.
3022 * INSTALL.md: fix FreeBSD port link (#6422, Mo).
3023 The FreeBSD port was renamed from pandoc to hs-pandoc in 2010.
3024 The old pandoc port is still at version 1.5.1.1
3026 * Propagate `(DY)LD_LIBRARY_PATH` in tests (#6376, Lila).
3028 * Bump `cabal-version` to 2.2 (#6377).
3030 * Make it possible to compile using Stack on NixOS (#6439, Mathieu
3033 * CI action to check for commit message length (Nikolay Yakimov, #6398).
3036## pandoc 2.9.2.1 (2020-03-23)
3038 * Markdown reader: Fix table alignment when heading begins with t (#6153).
3039 Due to a typo (`t` instead of `\t`) we were center aligning column
3040 headings that begin with a lowercase `t`!
3042 * Text.Pandoc.Readers.Roff:
3044 + Fix parsing of `\.` in man/ms readers (#6175).
3045 Previously due to a typo it was being parsed as `` ` ``.
3046 + Fix parsing of `\'` in man/ms readers (#6175). It was being parsed
3049 * Jira reader (Albert Krewinkel):
3051 + Fix parsing of tables without preceding blankline (#6198).
3052 A bug was fixed which caused faulty parsing if a table was not
3053 preceded by a newline and the first table cell had no space
3054 after the initial `|` characters.
3055 + Fix parsing of strikeout, emphasis (#6196). A bug was fixed which
3056 caused non-emphasized text containing digits and/or non-special
3057 symbols (like dots) to sometimes be parsed incorrectly.
3058 + Support colored inline text, indented lists (#6183, #6184).
3062 + Fix definition lists so indent even when paragraph indent is
3063 set to 0 (as is the default). Also ensure indent for display math
3064 that falls back to TeX.
3065 + Use `.QS/.QE` instead of `.RS/.RE` for block quotes.
3067 * EPUB writer: fix regression on detection of front/back/bodymatter
3068 (#6170). This bug caused sections with epub:type `dedication` to be
3069 misplaced in bodymatter instead of frontmatter as specified
3070 in the manual. The same problem would affect other epub:types.
3071 The pattern matching needed to be changed with the use of
3074 * AsciiDoc writer: remove redundant `otherwise` guard in
3075 `inlineToAsciiDoc` (#6146, Ryan Scott).
3077 * Text.Pandoc.Class:
3079 + Fix missing import when data files are not embedded (Albert Krewinkel).
3080 + Subdivide Text.Pandoc.Class into small unexported modules
3081 and ensure that all functions have Haddock documentation (#6106,
3083 + Finer grained imports of Text.Pandoc.Class submodules (#6203, Albert
3086 * Text.Pandoc.XML: Add to list of HTML5 attributes:
3087 `allow`, `autocapitalize`, `decoding`, `enterkeyhint`,
3088 `imagesizes`, `imagesrcset`, `loading`.
3090 * Use implicit Prelude (#6187, Albert Krewinkel).
3091 The previous behavior was introduced as a fix for #4464. It seems that
3092 this change alone did not fix the issue, and `stack ghci` and `cabal
3093 repl` only work with GHC 8.4.1 or newer, as no custom Prelude is loaded
3094 for these versions. Given this, it seems cleaner to revert to the
3097 * Always use custom prelude (#6187, Albert Krewinkel).
3098 Previously, the custom prelude was used only with older GHC versions, as
3099 a workaround for problems with ghci. The ghci problems are resolved by
3100 replacing package `base` with `base-noprelude`, allowing for consistent
3101 use of the custom prelude across all GHC versions.
3103 * Remove outdated checks for no longer supported base versions
3106 * PDF via wkhtmltopdf: put user-specified options last (#6171).
3107 Certain options (e.g. `cover`) need to come after flags on
3110 * Text.Pandoc.App: set resource path at the beginning so it can affect
3111 things like include-in-header (#5982).
3113 * Change macOS release candidate CI process so that notarized
3114 packages can be produced (#6169).
3116 * Make MANUAL more explicit about nonbreaking space handling by
3117 `all_symbols_escapable` (#6154, Fabien Schurter).
3119 * trypandoc (Mike Tzou):
3121 + Add checkbox for standalone option (#6189).
3122 + Use strict mode for JavaScript code (#6188).
3123 + Fetch resources over https (#6188).
3124 + Remove unnecessary attributes on style, script elements (#6188).
3126 * Use details tag to make GitHub releases changelog collapsible.
3128 * Update filter code in doc/filters.md so it works with latest pandoc
3131 * linux/Dockerfile: upgrade to alpine 3.11 (#6180, Albert Krewinkel).
3132 This is used to build the static linux binaries.
3135## pandoc 2.9.2 (2020-02-15)
3137 * Add `csv` as an input format (#6100). The CSV table is converted into a
3138 pandoc simple table. A new module Text.Pandoc.Readers.CSV
3139 exports `readCSV` [API change].
3141 * Introduce new format variants for JATS writer (#6014, Albert Krewinkel):
3143 - `jats_archiving` for the "Archiving and Interchange Tag Set",
3144 - `jats_publishing` for the "Journal Publishing Tag Set", and
3145 - `jats_articleauthoring` for the "Article Authoring Tag Set."
3147 The `jats` output format is now an alias for `jats_archiving`.
3148 The module Text.Pandoc.Writers.JATS now exports
3149 `writeJatsArchiving`, `writeJatsPublishing`, and
3150 `writeJatsArticleAuthoring`, as well as the legacy
3151 `writeJATS` [API change].
3153 * `--defaults`: Support `bibliography` and `csl` fields.
3154 Move `addMeta` from Text.Pandoc.App.CommandLineOptions to
3155 Text.Pandoc.App.Opt (internal change).
3157 * Add timing info for filters in `--verbose` mode (#6112).
3158 When verbose mode is specified (verbosity == INFO), print a
3159 notice when running a filter and when a filter completes (including
3164 + Allow `&` in LaTeX citation keys (#6110).
3165 + Improve caption and label parsing.
3166 + Don't emit empty Span elements for labels.
3167 + Put tables with labels in a surrounding Div.
3168 + Resolve `\ref` to table numbers (#6137).
3169 + Skip comments in more places where this is needed (#6114).
3170 + Allow beamer overlays for all commands in all raw tex (#6043).
3171 This affects parsing of raw tex in LaTeX and in Markdown and
3173 + Improve parsing of raw environments (#6034). If parsing fails
3174 in a raw environment (e.g. due to special characters like unescaped
3175 `_`), try again as a verbatim environment, which is less sensitive to
3176 special characters. This allows us to capture special environments
3177 that change catcodes as raw tex when `-f latex+raw_tex` is used.
3181 + Add highlight directive (#6140, Lucas Escot).
3185 + Prevent triple `[[[` which confuses MediaWiki (#6119).
3189 + Don't parse `data-id` as `id` attribute. And similarly don't
3190 parse any `data-X` as `X` when `X` is a valid HTML attribute.
3194 + Simplify parsing of sub- and superscripts (#6127, Albert Krewinkel).
3195 Speeds up parsing of single-word, markup-less sub- and superscripts.
3199 + Group biblatex citations even with prefix and suffix (#5849, Ethan
3200 Riley). Previously biblatex citations were only grouped if there
3201 was no prefix. This patch allows them to be grouped in subgroups split
3202 by prefixes and suffixes, which allows better citation sorting.
3203 + Fix regression in handling of columns in beamer slides (#6033).
3204 Columns in title slides were causing problems with
3206 + Fix duplicate frame classes in LaTeX/Beamer output (#6107).
3210 + Fix duplicate attributes on headings (#6062), regression from 2.7.x.
3211 + Fix `--number-offset` with HTML TOC. Eventually it would be worth
3212 adding a parameter to `makeSections` so this could be done at that
3213 level; then it would also affect other writers that construct
3215 + reveal.js: restore old behavior for 2D nesting (#6032).
3216 The fix to #6030 actually changed behavior, so that the
3217 2D nesting occurred at slide level N-1 and N, instead of
3218 at the top-level section. This commit restores the v2.7.3 behavior.
3219 If there are more than 2 levels, the top level is horizontal
3220 and the rest are collapsed to vertical.
3221 + reveal.js: ensure that pauses work even in title slides (#5819).
3225 + Fix regression: spurious dots in markdown_mmd metadata output (#6133).
3229 + Fix regression with Compact style on tight lists (#6072).
3230 Starting in 2.8, the docx writer no longer distinguishes between tight
3231 and loose lists, since the Compact style is omitted. This is a
3232 side-effect of the fix to #5670, as explained in the changelog. This
3233 patch fixes the problem by extending the exception currently offered to
3234 Plain blocks inside tables to Plain blocks inside list items.
3238 + Fix output of table headers (Albert Krewinkel, #6035).
3240 * Add Text.Pandoc.Image with unexported svgToPng.
3242 * Text.Pandoc.XML: Export `html5Attributes`, `html4Attributes`,
3243 `rdfaAttributes` (formerly unexported in Text.Pandoc.Writers.HTML).
3246 * Text.Pandoc.Shared: Export a new function `findM` (#6125,
3249 * Text.Pandoc.Logging: Add `RunningFilter`, `FilterCompleted`
3250 constructors to LogMessage [API change].
3252 * Text.Pandoc.CSV: fix bug in CSV parser; previously an extra blank record
3253 would sometimes be inserted at the end.
3255 * LaTeX template: add space option to xeCJK with PassOptionsToPackage
3256 (#6002). Otherwise we can get a clash with documentclasses that
3257 already load the package.
3261 + Allow filtering of element lists (#6038, Albert Krewinkel). Lists of
3262 Inline and Block elements can now be filtered via `Inlines` and
3263 `Blocks` functions, respectively. This is helpful if a filter
3264 conversion depends on the order of elements rather than a single
3265 element. For example, the following filter can be used to remove all
3266 spaces before a citation:
3268 function isSpaceBeforeCite (spc, cite)
3269 return spc and spc.t == 'Space'
3270 and cite and cite.t == 'Cite'
3273 function Inlines (inlines)
3274 for i = #inlines-1,1,-1 do
3275 if isSpaceBeforeCite(inlines[i], inlines[i+1]) then
3282 + Add methods `insert`, `remove`, and `sort` to pandoc.List
3283 (Albert Krewinkel). Example of use:
3285 local numbers = pandoc.List {2, 3, 1}
3286 numbers:sort() -- numbers is now {1, 2, 3}
3287 + Make `pandoc.List` a callable constructor (Albert Krewinkel).
3288 It is now possible to construct a new List via
3289 `pandoc.List()` instead of `pandoc.List:new()`.
3290 + Add tests for pandoc.List module (Albert Krewinkel).
3292 * Text.Pandoc.App.CommandLineOptions: Change `setVariable` to use `Text`
3293 instead of `String`. This avoids some unnecessary unpacking.
3295 * Use versioned directory for windows release zipfile.
3296 Also remove old `make-windows-installer.bat`, superseded by GitHub
3297 actions workflow, and modify `pandoc.wxs` for new paths.
3299 * Extensive code cleanup (#6141, #6128, #6129, #6130, #6123,
3300 #6105, 6102, #6117, #6124, #6115, #6116, #6111, Joseph C. Sible).
3302 * Fix hlint warnings (Albert Krewinkel).
3304 * Use latest doclayout, doctemplates (#6031). The new version of
3305 doclayout fixes a memory leak that affected `--include-in-header` with
3306 large files (and possibly other cases involving extremely long lines).
3308 * Use latest texmath.
3310 * Use latest skylighting and fix test suite (#6086).
3312 * sample.lua: Fix typo in descriptive comments (#6136, Caleb Maclennan).
3313 Fix typo in error message (#6135).
3315 * Add Docker and GH Actions instructions/links to INSTALL.md.
3317 * Update filter documentation (#6065). Improve cabal v2 instructions.
3318 Remove example using pandoc API directly (we have other
3319 docs for that and it was outdated).
3323 + Cross-link constructors and types (Albert Krewinkel).
3324 Thanks to @bpj for the idea.
3325 + Sort pandoc.List methods alphabetically (Albert Krewinkel).
3326 + Unify, fix anchors and internal links (#6061, Albert Krewinkel).
3327 Links and anchors now follow consistent conventions, like
3328 lowercase-only anchor names. This breaks some links to specific
3329 sections in the document, but will make it much easier to link
3330 documentation in the future.
3331 + Clarify filter function execution order (#6059, Albert Krewinkel).
3333 * In docs, update URLs and use `https:` wherever possible (#6090,
3336## pandoc 2.9.1.1 (2020-01-05)
3340 + Fix parsing bug affected indented code after raw HTML (#6009, #5360).
3344 + Fix regression in beamer slide structure with certain slide levels
3346 + Allow framebreaks for beamer's TOC (Heiko Schlittermann, #6012)
3347 + Properly handle unnumbered headings level 4+ (#6018).
3348 Previously the `\paragraph` command was used instead of
3349 `\paragraph*` for unnumbered level 4 headings.
3353 + Fix revealjs slide structure regression with certain slide levels
3355 + Add newlines to make slide show output more readable.
3359 + Remove extra spaces from table cells (Albert Krewinkel, #6024).
3361 * JATS template: Update JATS dtd (Arfon Smith, #6020). Use the archiving
3362 and interchange DTD rather than the more restrictive journal publishing
3363 DTD (which doesn't permit ext-link as a valid child).
3365 * Text.Pandoc.PDF: Fix `runTeXProgram` so that the input source is always
3366 overwritten (#6027). Previously it wasn't overridden if the file already
3367 existed, which led to bad results on subsequent runs when
3368 `pdf-engine-opt=-output-directory=` was used to specify an explicit temp
3371 * Text.Pandoc.BCP47: Change `getLang` to handle block-level contents
3372 (#6008). Some readers (e.g. RST) will populate the `lang` metadata field
3373 with block-level content. `getLang` has been modified to handle this.
3374 Previously in these cases the LaTeX writer would not properly set the
3375 "main language" of the document.
3377 * Fix `test/tables.org` (Albert Krewinkel).
3379 * Use HTTPS in copyright message (Felix Yan, #6010)
3382## pandoc 2.9.1 (2019-12-23)
3384 * Add Jira reader (Albert Krewinkel, #5556).
3386 * Jira writer: use jira-wiki-markup renderer (Albert Krewinkel,
3387 #5926). The following improvements are included in this change:
3389 + non-jira raw blocks are fully discarded instead of showing
3391 + table cells can contain multiple blocks;
3392 + unnecessary blank lines are removed from the output;
3393 + markup chars within words are properly surrounded by
3395 + preserving soft linebreaks via `--wrap=preserve` is
3398 Note that backslashes are rendered as HTML entities, as there
3399 appears no alternative to produce a plain backslash if it is
3400 followed by markup. This may cause problems when used with
3401 confluence, where rendering seems to fail in this case.
3403 * Fix regression with `--number-sections`. Starting with 2.8,
3404 `--number-sections` also had the effect of `--section-divs`,
3405 even if `--section-divs` was not specified.
3407 * Improved table of contents generation in markdown, RTF,
3408 commonmark, better handling cases where section headings are
3411 * Ensure that later default file values for `variable` replace
3412 earlier ones (5988).
3414 * HTML reader: Add `nav` to list of block-level tags.
3416 * Org reader (Albert Krewinkel):
3418 + Wrap named table in Div, using name as id (#5984).
3419 Tables which are given a name via `#+NAME:` or `#+LABEL:`
3420 are wrapped in an additional Div, with the name set as the
3422 + Report parsing errors properly.
3423 + Fix parsing problem for colons in headline (#5993).
3425 * Text.Pandoc.PDF: Ensure UTF8 when printing source in
3426 `--verbose` mode, avoiding an error on platforms that
3427 default to something other than UTF-8 (#5997).
3429 * Text.Pandoc.Templates: Strip directory before trying to find
3430 partial in data files (#5987).
3432 * Text.Pandoc.Shared: Improve `makeSections` so we don't get
3433 doubled "number" attributes in EPUB output (or anywhere
3436 * Added tests for `--toc` and `--section-divs`.
3438 * Text.Pandoc.MIME: Added glsl MIME type for WebGL maps (#6000,
3441 * MANUAL: A bit clearer explanation for `--base-header-level`.
3442 We now say exactly how to translate between the deprecated
3443 `--base-header-level` and `--shift-heading-level-by`.
3447 + Remove spurious dot in title (#5996, Mauro Bieg).
3448 + Replace metadata example with image centering (#6004,
3449 Albert Krewinkel). Thanks to @efx for proposing this filter.
3452## pandoc 2.9 (2019-12-11)
3454 * Text.Pandoc.Templates [API change]
3456 + Add Monad wrappers `WithDefaultPartials` and `WithPartials`.
3457 Wrapping these around an instance of `PandocMonad` gives
3458 us different instances of `TemplateMonad`, with different
3459 search behavior in retrieving partials.
3460 To compile a template and limit partial search to pandoc's
3461 data files, use `runWithDefaultPartials (compileTemplate ...)`.
3462 To compile a template and allow partials to be found locally
3463 (either on the file system or via HTTP, in the event that
3464 the main template has an absolute URL), ue
3465 `runWithPartials (compileTemplate ...)`.
3466 + Export `getTemplate`, which seeks a template locally,
3467 or via HTTP if the template has an absolute URL, falling
3468 back to the data files if not found.
3469 + Export `compileDefaultTemplate` -- does `getDefaultTemplate`
3470 and compiles the result, raising an error on failure.
3472 * Text.Pandoc.Class [API change]
3474 + Remove `TemplateMonad` instances for `PandocIO` and `PandocPure`.
3475 These were too limiting and caused a bug whereby a local
3476 partial could be used even when the default template was requested.
3477 We now rely on instances provided in the Templates module.
3479 * Text.Pandoc.App.OutputSettings: Simplify template retrieval code.
3481 * ConTeXt template: Adjust to title formatting (#5949, Denis Maier).
3482 Add `\setupinterlinespace` to `title`, `subtitle`, `date` and `author`
3483 elements: otherwise longer titles that run over multiple lines will look
3484 squashed as `\tfd` etc. won't adapt the line spacing to the font size.
3486 * reveal.js template: Add title-slide-attributes variable (#5981,
3489 * More informative JSON parse error (#5973).
3491 * Use external emojis package (forked from pandoc). Removed emoji data
3492 in Text.Pandoc.Emoji.
3494 * Fix regression in `makeSections` (#5965).
3495 Previously `hierarchicalize` (the ancestor of `makeSections`) would put
3496 header attributes on the containing Div. In 2.8 this behavior changed,
3497 which broke some tools depending on pandoc. Here we roll back this change,
3498 so that attributes again migrate from the header to the containing Div when
3499 `makeSections` is run. Note that attributes are retained on the header as
3500 well (unlike before) -- with the exception of the `id` attribute, which of
3501 course cannot be duplicated.
3503 * Fix `--toc-depth` regression in 2.8 (#5967).
3505 * Use doctemplates 0.8. Rename template 'filters' as 'pipes'
3506 to avoid confusion with the other notion of filter used by pandoc.
3508 * Fix README.md so that relative links from manual become absolute.
3509 Previously they'd be broken links when viewed on GitHub or Hackage.
3510 So we add the base URL for the pandoc manual.
3512 * Document display math syntax in manual.
3514## pandoc 2.8.1 (2019-12-05)
3516 * Add `ascii_identifiers` as a supported extension for `markdown`.
3517 This fixes a regression in 2.8.
3519 * Fix regression with behavior of `--variable` (#5962).
3520 Previously `-Vfoo=1 -Vfoo=2` would produce a list value for foo;
3521 with 2.8 it produced just `2`. This commit restores the earlier
3524 * Roll back part of of `--shift-heading-level-by` change (#5957).
3525 With positive heading shifts, starting in 2.8 this option
3526 caused metadata titles to be removed and changed to regular
3527 headings. This behavior is incompatible with the old
3528 behavior of `--base-header-level` and breaks old workflows,
3529 so we have rolled back the change. Note that there is now an
3530 asymmetry in positive and negative heading level shifts:
3531 With positive shifts, the metadata title stays the same and
3532 does not get changed to a heading in the body, while with
3533 negative shifts, a heading can be converted into the
3536 * Text.Pandoc.Shared: Fix `makeSections` so it doesn't turn
3537 column Divs into sections.
3539 * HTML writer: add task-list class to ul if all elements are
3540 task list items. This will allow styling unordered task
3541 lists in a way that omits the bullet.
3543 * HTML-based templates: Add CSS to suppress bullet on unordered task lists.
3545 * ConTeXt template: Fix `\startcslreferences` and use ConTeXt syntax
3546 conventions (#5945, Denis Maier). The old version had a too large
3547 a skip at the beginning of the reference list.
3549 * LaTeX template: keep the `\author{}` command even if author is not
3550 specified (#5961, Yihui Xie). Avoids a LaTeX warning.
3552 * Generate Emoji module with TH.
3554 + Add Text.Pandoc.Emoji.TH.
3555 + Replace long literal list in Text.Pandoc.Emoji with one-liner
3556 generating it from `emoji.json` using TH.
3557 + Add Makefile target to download `emoji.json`.
3558 + Remove `tools/emoji.hs`.
3560 * Increase GC allocation space for compilation in cabal.project.
3562 * Clean up manual on PDF generation backend options (#5940).
3564 * Update release checklist to include code signing step and update
3565 Windows release-candidate builds (#5950).
3568## pandoc 2.8.0.1 (2019-11-26)
3570 * List `pdf` in `--list-output-formats`.
3571 * EPUB writer: Fix regression with `--css` (#5937). In 2.8 `--css`
3572 would not have an effect on EPUB output.
3573 * RST writer: Use grid tables for one-column tables, since
3574 simple tables clash with heading syntax in this case (#5936).
3575 * Add unexported module Text.Pandoc.Readers.Metadata (see #5914).
3576 * Use doctemplates 0.7.2, which adds the `nowrap` filter to
3578 * Update default man template using `nowrap` for .TH heading (#5929).
3579 * HTML templates: Add support for `toc-title` variable (#5930,
3581 * Remove `grffile` (LaTeX package) requirement in MANUAL.txt
3582 (#5927, Ian Max Andolina).
3583 * Use skylighting 0.8.3.
3585## pandoc 2.8 (2019-11-22)
3587 * Improvements in templates system (from doctemplates):
3589 + Pandoc templates now support a number of new features that
3590 have been added in doctemplates: notably, `elseif`, `it`,
3591 partials, filters, and syntax to control nesting and reflowing of
3592 text. These changes make pandoc more suitable out of the
3593 box for generating plain-text documents from data in YAML
3594 metadata. It can create enumerated lists and even tabular
3596 + We now used templates parameterized on doclayout Doc types.
3597 The main impact of this change is better reflowing of
3598 content interpolated into templates. Previously,
3599 interpolated variables were rendered independently and
3600 interpolated as strings, which could lead to overly long
3601 lines. Now the templates interpolated as Doc values which
3602 may include breaking spaces, and reflowing occurs after
3603 template interpolation rather than before.
3604 + Remove code from the LaTeX, Docbook, and JATS writers that
3605 looked in the template for strings to determine whether it
3606 is a book or an article, or whether csquotes is used. This
3607 was always kludgy and unreliable.
3608 + Change template code to use new API for doctemplates.
3610 * Add `--defaults`/`-d` option. This adds the ability to specify
3611 a collection of default values for options in a YAML file. For
3612 example, one might define a set of defaults for letters,
3613 and then do `pandoc -d letter myletter.md -o myletter.pdf`.
3614 See the documentation of this feature in MANUAL.txt.
3616 * Raise error on unsupported extensions (#4338).
3618 * The `--list-extensions[=FORMAT]` option now lists only
3619 extensions that affect the given FORMAT.
3621 * Add `-L` option as shortcut for `--lua-filter`.
3623 * Add `--shift-heading-level-by` option and deprecate
3624 `--base-heading-level` (#5615). The new option does
3625 everything the old one does, but also allows negative shifts.
3626 It also promotes the document metadata (if not null) to a
3627 level-1 heading with a +1 shift, and demotes an initial
3628 level-1 heading to document metadata with a -1 shift. This
3629 supports converting documents that use an initial level-1
3630 heading for the document title.
3632 * Allow `--metadata-file` to be used repeatedly to include
3633 multiple metadata files (Owen McGrath, #5702). Values in
3634 files specified first will be overridden by those in later
3637 * `--ascii` now uses numerical hex character references (#5718).
3639 * Allow PDF output to stdout (#5751). PDF output now behaves like other
3640 binary formats: it will not be output to the terminal, but can be
3641 sent to stdout using either `-o -` or a pipe. The intermediate format
3642 will be determined based on the setting of `--pdf-engine`.
3644 * Make some writers sensitive to 'unlisted' class on headings (#1762).
3645 If this is present on a heading with the 'unnumbered' class,
3646 the heading won't appear in the TOC. This class has no
3647 effect if 'unnumbered' is not also specified. This affects HTML-based
3648 writers (including slide shows and EPUB), LateX (including beamer), RTF,
3649 and PowerPoint. Other writers do not yet support `unlisted`.
3651 * Fix `gfm_auto_identifiers` behavior with emojis (#5813). Note that
3652 we also now use emoji names for emojis when `ascii_identifiers`
3655 * When `--ipynb-output` is used with the default "best" format, strip
3656 ANSI escape codes for non-ipynb output (#5633). These cause problems
3657 in many formats, including LaTeX.
3659 * Don't look for template files remotely for remote input (#5579).
3660 Previously pandoc would look for the template at a remote URL when a
3661 URL was used for the input file, instead of taking it from the
3664 * Allow combining `-Vheader-includes` and `--include-in-header` (#5904).
3665 Previously `header-includes` set as a variable would be
3666 clobbered by material included using `--include-in-header`.
3668 * Change merge behavior for metadata. Previously, if a document
3669 contained two YAML metadata blocks that set the same field, the
3670 conflict would be resolved in favor of the first. Now it is resolved
3671 in favor of the second (due to a change in pandoc-types).
3672 This makes the behavior more uniform with other things in pandoc
3673 (such as reference links and `--metadata-file`).
3675 * Don't add a newline to fragment output if there's already one.
3677 * Change exit codes and document in MANUAL.txt:
3679 + `PandocAppError` was 1, is now 4
3680 + `PandocOptionError` was 2, is now 6
3681 + `PandocMakePDFError` was 65, is now 66
3683 * Switch to new pandoc-types and use Text instead of String [API change].
3684 (Christian Despres, #5884).
3688 + Better handling of `<q>` with cite attribute (#5798, Ole Martin Ruud).
3689 If a `<q>` tag has a `cite` attribute, we interpret it as a Quoted
3690 element with an inner Span.
3691 + Add support for HTML `<samp>` element (#5792, Amogh Rathore).
3692 The `<samp>` element is parsed as Code with class `sample`.
3693 + Add support for HTML `<var>` element (#5799, Amogh Rathore).
3694 The `<var>` element is parsed as Code with class `variable`.
3695 + Add support for `<mark>` elements (Florian B, #5797). Parse
3696 `<mark>` elements from HTML as Spans with class `mark`.
3697 + Add support for `<kbd>` elements, parsing them as Span with class
3698 `kbd` (Daniele D'Orazio, #5796).
3699 + Add support for `<dfn>`, parsing this as a Span with class `dfn`
3700 (#5882, Florian Beeres).
3704 + Headers: don't parse content over newline boundary (#5714).
3705 + Handle inline code more eagerly within lists (Brian Leung, #5627).
3706 + Removed some needless lookaheads.
3707 + Don't parse footnote body unless extension enabled.
3708 + Fix small super/subscript issue (#5878). Superscripts and subscripts
3709 cannot contain spaces, but newlines were previously allowed
3710 (unintentionally). This led to bad interactions in some cases
3711 with footnotes. With this change newlines are also not allowed inside
3713 + Use `take1WhileP` for `str`, table row. This yields a small but
3714 measurable performance improvement.
3719 + Fix parsing of optional arguments that contain braced text (#5740).
3720 + Don't try to parse includes if `raw_tex` is set (#5673).
3721 When the `raw_tex` extension is set, we just carry through
3722 `\usepackage`, `\input`, etc. verbatim as raw LaTeX.
3723 + Properly handle optional arguments for macros (#5682).
3724 + Fix `\\` in `\parbox` inside a table cell (#5711).
3725 + Improve `withRaw` so it can handle cases where the token string is
3726 modified by a parser (e.g. accent when it only takes part of a Word
3727 token) (#5686). This fixes a bug that caused the ends of
3728 certain documents to be dropped.
3729 + Handle `\passthrough` macro used by latex writer (#5659).
3730 + Support tex `\tt` command (#5654).
3731 + Search for image with list of extensions like latex does, if an
3732 extension is not provided (#4933).
3733 + Handle `\looseness` command values better (#4439).
3734 + Add `mbox` and `hbox` handling (Vasily Alferov, #5586).
3735 When `+raw_tex` is enabled, these are passed through literally.
3736 Otherwise, they are handled in a way that emulates LaTeX's behavior.
3737 + Properly handle `\providecommand` and `\provideenvironment` (#5635).
3738 They are now ignored if the corresponding command or environment
3740 + Support epigraph command in LaTeX Reader (oquechy, #3523).
3741 + Ensure that expanded macros in raw LaTeX end with a space
3742 if the original did (#4442).
3743 + Treat `ly` environment from lilypond as verbatim (Urs Liska, #5671).
3744 + Add `tikzcd` to list of special environments (Eigil Rischel).
3745 This allows it to be processed by filters, in the same way that
3746 one can do for `tikzpicture`.
3750 + Better support for `while`.
3751 + More improvements in parsing conditionals.
3752 + Fix problem parsing comments before macro.
3753 + Improve handling of groups.
3754 + Better parsing of groups (#5410). We now allow groups
3755 where the closing `\\}` isn't at the beginning of a line.
3759 + Keep `name` property in `imgAttr` (Brian Leung, #5619).
3760 + Fixed parsing of indented blocks (#5753). We were requiring
3761 consistent indentation, but this isn't required by RST.
3762 + Use title, not admonition-title, for admonition title.
3763 This puts RST reader into alignment with docbook reader.
3764 + Don't strip final underscore from absolute URI (#5763).
3765 + Avoid spurious warning when resolving links to internal anchors
3766 ending with `_` (#5763).
3770 + Accept `ATTR_LATEX` in block attributes (Albert Krewinkel, #5648).
3771 Attributes for LaTeX output are accepted as valid block attributes;
3772 however, their values are ignored.
3773 + Modify handling of example blocks (Brian Leung, #5717).
3774 + Allow the `-i` switch to ignore leading spaces (Brian Leung).
3775 + Handle awkwardly-aligned code blocks within lists (Brian Leung).
3776 Code blocks in Org lists must have their `#+BEGIN_` aligned in a
3777 reasonable way, but their other components can be
3778 positioned otherwise.
3779 + Fix parsing of empty comment lines (#5856, Albert Krewinkel).
3780 Comment lines in Org-mode can be completely empty.
3782 * Muse reader (Alexander Krotov):
3784 + Add RTL support (#5551).
3785 + Do not allow closing asterisks to be followed by `*`.
3786 + Do not split series of asterisks into symbols and emphasis (#5821).
3787 + Do not terminate emphasis on `*` not followed by space.
3791 + Parse markup inside monospace ('') (#5916, Alexander Krotov).
3795 + Move style-parsing-specific code to a new unexported module,
3796 Text.Pandoc.Readers.Docx.Parse.Styles (Nikolay Yakimov).
3797 + Move StyleMap to docx writer (Nikolay Yakimov).
3798 + Only use LTR when it is overriding BiDi setting (#5723, Jesse
3799 Rosenthal). The left-to-right direction setting in docx is used
3800 in the spec only for overriding an explicit right-to-left setting.
3801 We only process it when it happens in a paragraph set with BiDi.
3802 This is especially important for docs exported from Google Docs,
3803 which explicitly (and unnecessarily) set `rtl=0` for every paragraph.
3804 + Fix list number resumption for sublists (#4324).
3805 The first list item of a sublist should not resume numbering
3806 from the number of the last sublist item of the same level,
3807 if that sublist was a sublist of a different list item.
3811 + Richer parse for admonitions (Michael Peyton Jones, #1234).
3812 Instead of parsing admonitions as blockquotes, we now parse
3813 them as Divs with an appropriate class. We also handle titles
3814 for admonitions as a nested Div with the "title" class.
3815 + Fix nesting of chapters and sections (#5864, Florian Klink,
3816 Félix Baylac-Jacqué).
3817 + Fix bug with entities in mathphrase element (#5885).
3821 + Skip optional `{{table}}` template (#5757).
3825 + Fix dollar-math parsing to ensure that space is left between a
3826 control sequence and a following letter (#5836).
3827 + In `untokenize`, ensure space between control sequence
3828 and following letter (#5836).
3829 + Don't omit macro definitions defined in the preamble.
3830 These were formerly omitted (though they still affected macro
3831 resolution if `latex_macros` was set). Now they are included in
3833 + Parse macro definitions as raw LaTeX when `latex_macros` is
3834 disabled. (When `latex_macros` is enabled, we omit them, since
3835 pandoc is applying the macros itself.)
3836 + Fix a hang/memory leak in certain circumstances (#5845).
3837 + Text.Pandoc.Readers.LaTeX.Parsing: add `[Tok]` parameter to
3838 `rawLaTeXParser`. This allows us to repeat retokenizing
3839 unnecessarily in e.g. `rawLaTeXBlock`.
3840 + Add KOMA-Script metadata commands (#5910, Andrew Dunning).
3841 Add all titling commands to existing definition for `\dedication`.
3842 + Parse `\micro` siunitx unit command (#5921, Jose Luis Duran).
3846 + Ensure proper nesting when we have long ordered list markers (#5705).
3847 + Make `plain` output plainer (#5741). Previously we used the following
3848 Project Gutenberg conventions for plain output: extra space before and
3849 after level 1 and 2 headings, all-caps for strong emphasis,
3850 underscores surrounding regular emphasis. Now these
3851 conventions are used only when the `gutenberg` extension is
3852 enabled. By default, Strong and Emph are rendered without
3853 special formatting, and headings are rendered without
3854 special formatting, and with only one blank line following.
3855 To restore the former behavior, use `-t plain+gutenberg`.
3856 + Prefer using raw_attribute when enabled (#4311).
3857 The `raw_attribute` will be used to mark raw bits, even HTML
3858 and LaTeX, and even when `raw_html` and `raw_tex` are
3859 enabled, as they are by default. To get the old behavior,
3860 disable `raw_attribute` in the writer.
3861 + Prefer `pipe_tables` to raw HTML even when we must
3862 lose width information (#2608, #4497).
3863 + Improve escaping in list items (#5918).
3867 + Don't include `+` in code blocks for regular asciidoc.
3868 This is asciidoctor-specific.
3869 + Handle admonitions (#5690).
3873 + Add thin space when needed in LaTeX quote ligatures (#5684).
3874 + Use `\hspace{0pt}` for 0-width space U+200B (#5756).
3875 + Use `cslreferences` environment for csl bibliographies.
3876 This allows bibliographies to receive special formatting.
3877 The template now contains definition of this environment (enabled
3878 only when CSL is used). It also defines a `\cslhangindent` length.
3879 This is set to 2em by default when the bibliography style specifies
3880 a hanging indent. To override the length, you can use e.g.
3881 `\setlength{\cslhangindent}{7em}` in header-includes.
3882 See jgm/pandoc-citeproc#410.
3883 + Strip off `{}` around locator for biblatex/natbib output (#5722).
3884 + Fix line breaks at start of paragraph (#3324).
3885 Previously we just omitted these. Now we render them
3886 using `\hfill\break` instead of `\\`. This is a revision
3887 of a PR by @sabine (#5591) who should be credited with the idea.
3888 + We no longer look in the template or header-includes to see if a
3889 book or article documentclass is used, or to see whether the
3890 `csquotes` package is used. To use `csquotes` for LaTeX, set
3891 `csquotes` in your variables or metadata. To specify a book
3892 style, use the `documentclass` variable or
3893 `--top-level-division`.
3894 + Fix horizontal rule (#5801). We change to use 0.5pt rather than
3895 `\linethickness`, which apparently only ever worked "by accident"
3896 and no longer works with recent updates to texlive.
3900 + Add option to include source files in ConTeXt PDFs (Tristan
3901 Stenner, #5578). The metadata field or variable
3902 (`includesource`) can be set to attach the source documents
3903 to the resulting PDF.
3904 + Customizable type of PDF/A for the ConTeXt writer (Karl
3905 Pettersson, #5608). The `pdfa` variable may now be set in metadata.
3906 Also updated color profile settings in accordance with ConTeXt wiki,
3907 and made ICC profile and output intent for PDF/A customizable
3908 using `pdfaiccprofile` and `pdfaintent`.
3909 + Unit tests: adjust code property to avoid an irrelevant
3910 failure involving inline code with two consecutive newlines.
3911 + Set `csl-hanging-indent` variable if needed.
3912 + Use special environment for CSL references.
3913 + Use braces, not start/stop, for inline language tags.
3914 This prevents unwanted gobbling of spaces.
3918 + Use numeric character references with `--ascii` (#5718).
3919 Previously we used named character references with html5 output.
3920 But these aren't valid XML, and we aim to produce html5 that is
3921 also valid XHTML (polyglot markup). (This is also needed for epub3.)
3922 + Ensure that line numbers in code blocks get id-prefix (#5650).
3923 + Ensure TeX formulas are rendered correctly (Philip Pesca,
3924 #5658). The web service passed in to `--webtex` may render
3925 formulas using inline or display style by default.
3926 Prefixing formulas with the appropriate command ensures
3927 they are rendered correctly.
3928 + Render inline formulas correctly with `--webtex` (Philip
3929 Pesca, #5655). We add `\textstyle` to the beginning of the
3930 formula to ensure it will be rendered in inline style.
3931 + Pass through `aria-` attributes to HTML5 (#5642).
3932 + Render a Quoted element with an inner Span with
3933 `cite` attribute using a `<q>` tag (#5798, Ole Martin Ruud).
3934 + Render a Span with class `mark` using the `<mark>` element
3935 (Florian Beeres, #5797).
3936 + Render Span with class `dfn` using `<dfn>` element
3937 (Florian Beeres, #5882).
3938 + Render Span with class `kbd` using `<kbd>` element (Daniele
3940 + Render Code with class `variable` using `<var>` element
3941 (Amogh Rathore, #5799).
3942 + Render Code with class `sample` using `<samp>` element
3943 (Amogh Rathore, #5799).
3947 + Improve splitting into chapters (#5761), using `makeSection`.
3948 + Avoid issuing warning multiple times when title not set (see #5760).
3949 + Use svg tag wrapper for cover image (#5638). In addition, the
3950 code generating the image has been moved to the template, to make it
3951 more customizable. NOTE: Those who use custom EPUB
3952 templates will need to adjust their templates, adding the
3953 code to generate the cover image. (Previously this was just
3954 inserted into 'body'.)
3955 + Improve toChapters, making it work better if there are Divs
3957 + Add support for EPUB2 covers (blmage, #3992).
3958 + Do not override existing "fileN" medias when writing to EPUB format
3960 + Ensure that `lang` variable is set on all chapters (so that it
3961 will add an `xml:lang` attribute on the `html` element).
3965 + Removed remnants of `admonition-title`.
3966 + Fix handling of `:align:` on figures and images (#4420).
3967 When the image has the `align-right` (etc.) class, we now use
3968 an `:align:` attribute.
3969 + Improve spacing for tables with no width information (#5899).
3970 If a simple table would be too wide, we use a grid table.
3971 + Fix backslash escaping after strings (Albert Krewinkel, #5906).
3972 The check whether a complex inline element following a string must
3973 be escaped, now depends on the last character of the string instead
3975 + Ensure there's a blank line before tables (#5898).
3979 + Handle mixed lists without HTML fallback (#5107).
3983 + Fix multiline table (Zihang Chen, #5683).
3987 + Add RTL support (Alexander Krotov, #5551).
3991 + Suppress non-absolute link URLs (#5770). Absolute URLs are
3992 still printed in parentheses following the link text, but relative
3993 URLs are suppressed (just as internal links starting with '#'
3995 + Improved definition list term output. Now we boldface code but
3996 not other things. This matches the most common style in man pages
3997 (particularly option lists).
4001 + Use `.LP` instead of `.PP` for line block (#5588).
4002 + Use boldface for definition terms in DefinitionLists.
4006 + Do not emit empty `<fn-group>` (Mauro Bieg, #5595).
4007 + Update template to v1.1dtd (#5632, Arfon Smith).
4008 + Update `data/jats.csl` to avoid commas between editor
4009 name-part elements. (#5629)
4010 + Add `abstract` to template (Mauro Bieg).
4014 + Don't strip hash from internal links (#5922).
4018 + Remove escapeStringForJira for code blocks (Jan-Otto Kröpke).
4019 + Remove extraneous newline after single-line block quotes
4020 (#5858, Albert Krewinkel).
4021 + Improve escaping of special characters, using backslash escapes
4022 instead of HTML entities (#5858, Albert Krewinkel).
4024 * OpenDocument writer:
4026 + Avoid duplicate attributes (#4634). We use the innermost
4027 attribute in nested cases.
4028 + If `native_numbering` extension is set, use native OpenDocument
4029 enumeration for figures and tables (Nils Carlson).
4030 + Place caption before table (#5681, Dmitry Pogodin).
4034 + Add a test for MathML formulas in ODT documents (blmage).
4035 + Improve the parsing of frames in ODT documents (blmage).
4039 + Make handling of styles more robust in localized versions
4040 of Word (Nikolay Yakimov, #5523, #5052, #5074). We now use
4041 style names, not ids, for assigning semantic meaning, since
4042 the ids can change depending on the locale. Style name
4043 comparisons are case-insensitive, since those are
4044 case-insensitive in Word. Since docx style names can have
4045 spaces in them, and pandoc-markdown classes can't, anywhere
4046 when style name is used as a class name, spaces are
4047 replaced with ASCII dashes `-`. Code styles, i.e. "Source
4048 Code" and "Verbatim Char" now honor style inheritance. Docx
4049 Reader now honours "Compact" style (used in
4050 Pandoc-generated docx). The side-effect is that "Compact"
4051 style no longer shows up in docx+styles output. Styles
4052 inherited from "Compact" will still show up.
4053 + Re-use Readers.Docx.Parse for StyleMap (#5766, Nikolay Yakimov).
4054 + Internal improvements and code simplification (Nikolay Yakimov).
4055 + Preserve built-in styles in DOCX with custom style (Ben Steinberg,
4056 #5670). This change prevents custom styles on divs and spans
4057 from overriding styles on certain elements inside them, like
4058 headings, blockquotes, and links. On those elements, the
4059 "native" style is required for the element to display correctly.
4060 This change also allows nesting of custom styles; in order to do so,
4061 it removes the default "Compact" style applied to Plain blocks,
4062 except when inside a table.
4063 + Add `proofState` to list of elements carried over from
4064 settings.xml in the reference.docx (Krystof Beuermann, #5703).
4065 + Change order of `ilvl` and `numId` in `document.xml` (Agustín
4066 Martín Barbero, #5645). Also, make list para properties go first.
4067 This reordering of properties shouldn't be necessary but
4068 it seems Word Online does not understand the docx correctly otherwise.
4070 * PowerPoint writer:
4072 + Code formatting is now context dependent (Jeroen de Haas, #5573).
4073 This commit alters the way in which the Powerpoint writer treats
4074 inline code and code blocks. Inline code is now formatted at
4075 the same size as the surrounding text. Code blocks are now given
4076 a margin and font size according to their level.
4077 Furthermore this commit allows changing the font with which code is
4078 formatted via the `monofont` option.
4079 + Start numbering at appropriate numbers (Jesse Rosenthal, #5709).
4080 Starting numbers for ordered lists were previously ignored. Now
4081 we specify the number if it is something other than 1.
4083 * Text.Pandoc.Parsing:
4085 + Add `manyChar`, `many1Char`, `manyTillChar`, `many1TillChar`,
4086 `many1Till`, `manyUntil`, `mantyUntilChar`: these are like their
4087 unsuffixed counterparts but pack some or all of their output
4088 (Christian Despres, #5884).
4089 + Add `stateAllowLineBreaks` to `ParserState` [API change].
4090 + Fix inline parsing in grid table cells (#5708).
4091 + Change type of `setLastStrPos` so it takes a `Maybe SourcePos`
4092 rather than a `SourcePos` [API change].
4093 + Make `parseFromString'` and `gridTableWith` and
4094 `gridTableWith'` polymorphic in the parser state,
4095 constraining it with `HasLastStrPosition` [API change].
4096 + `parseFromString'`: reset `stateLastStrPos` to `Nothing` before parse.
4097 + Rename takeWhileP -> take1WhileP and clean it up.
4098 (It doesn't match the empty sequence.)
4102 + For PDFs via HTML, ensure temp file is deleted even if the pdf
4103 program is not found (#5720).
4104 + Better detection of a Cygwin environment (#5451).
4105 + Don't assume tex log file is UTF8-encoded (#5872).
4106 Fall back to latin1 if it can't be read as UTF-8.
4108 * Text.Pandoc.Extensions:
4110 + Export new function `getAllExtensions`, which returns the
4111 extensions that affect a given format (whether enabled by default
4112 or not) [API change].
4113 + Change type of `parseFormatSpec` from
4114 `Either ParseError (String, Extensions -> Extensions)`
4115 to `Either ParseError (String, [Extension], [Extension])`
4117 + Add `Ext_gutenberg` constructor to `Extension` [API change].
4118 + Add `Ext_native_numbering` constructor to `Extension` [API change]
4121 * Text.Pandoc.Readers, Text.Pandoc.Writers:
4123 + Change type of `getReader` and `getWriter` so they return
4124 a value in the PandocMonad instance rather than an Either
4125 [API change]. Exceptions for unknown formats and unsupported
4126 extensions are now raised by these functions.
4130 + Change `optMetadataFile` type from `Maybe FilePath` to `[FilePath]`
4131 (Owen McGrath, #5702) [API change].
4133 * Text.Pandoc.Logging:
4135 + Add `CouldNotDeduceFormat` constructor to `LogMessage` [API change].
4136 Issue this warning when we're falling back to markdown or
4137 html because we don't recognize the extension of the input or
4139 + Clarify warning for missing title (#5760).
4140 + Add `UnusualConversion` constructor to `LogMessage` [API change]
4141 (Mauro Bieg, #5736). Emit warning on `-f latex -o out.pdf`.
4145 + Improve function documentation (Albert Krewkinkel).
4146 + Traverse nested blocks and inlines in correct order (Albert
4147 Krewinkel, #5667). Traversal methods are updated to use the new
4148 Walk module so that sequences with nested Inline (or Block) elements
4149 are traversed in the order in which they appear in the linearized
4151 + New unexported module `Text.Pandoc.Lua.Walk` (Albert Krewinkel).
4152 Lua filters must be able to traverse sequences of AST
4153 elements and to replace elements by splicing sequences back
4154 in their place. Special `Walkable` instances can be used for
4155 this; those are provided in a new module
4156 `Text.Pandoc.Lua.Walk`.
4157 + `Attr` values can now be given as normal Lua tables (Albert
4158 Krewinkel, #5744). This can be used as a convenient alternative
4159 to constructing `Attr` values with `pandoc.Attr`. Identifiers
4160 are taken from the `id` field, classes must be given as
4161 space separated words in the `class` field. All remaining fields
4162 are included as attributes. With this change, the following lines
4163 now create equal elements:
4165 pandoc.Span('test', {id = 'test', class = 'a b', check = 1})
4166 pandoc.Span('test', pandoc.Attr('test', {'a','b'}, {check = 1}))
4168 This also works when using the *attr* setter:
4170 local span = pandoc.Span 'text'
4171 span.attr = {id = 'test', class = 'a b', check = 1}
4173 Furthermore, the *attributes* field of AST elements can now be a
4174 plain key-value table even when using the `attributes` accessor:
4176 local span = pandoc.Span 'test'
4177 span.attributes = {check = 1} -- works as expected now
4179 + Export `make_sections`, remove `hierarchicalize`. Lua filters that
4180 use `hierarchicalize` will need to be rewritten to use
4182 + Add a `clone()` method to all AST elements (Albert Krewinkel, #5568).
4183 + Fix Lua function names in pandoc.system (niszet). Change
4184 `get_current_directory` to `get_working_directory` and
4185 `with_temp_directory` to `with_temporary_directory`, to
4186 conform to the manual.
4188 * Text.Pandoc.Error:
4190 + Add constructors `PandocUnknownReaderError`,
4191 `PandocUnknownWriterError`, `PandocUnsupportedExtensionError`.
4193 + Better message for `PandocShouldNeverHappenError`.
4194 + Better message for `PandocTemplateError`.
4196 * Text.Pandoc.Emoji:
4198 + Update emoji list (#5666). Done using new `tools/emojis.hs`,
4199 which uses the list from the gem GitHub uses. Future updates
4200 can be done with this tool.
4204 + Pass value of `--dpi` to `rsvg-convert` when converting SVG to PDF
4205 in the process of creating a PDF (#5721).
4207 * Text.Pandoc.Shared:
4209 + Replace `Element` and `makeHierarchical` with `makeSections`.
4210 Now that we have Divs, we can use them to represent the
4211 structure of sections, and we don't need a special Element type.
4212 `makeSections` reorganizes a block list, adding Divs with class
4213 `section` around sections, and adding numbering if needed.
4214 This change also fixes some longstanding issues recognizing section
4215 structure when the document contains Divs (#3057, see also #997).
4216 + Remove `Element` type [API change]
4217 + Remove `makeHierarchicalize` [API change]
4218 + Add `makeSections` [API change]
4219 + Export `deLink` [API change]
4220 + Make `filterIpynbOutput` strip ANSI escapes from code in output
4221 for non-ipynb formats, when the default "best" option is used with
4222 `--ipynb-output` (#5633).
4223 + Fix `camelCaseToHyphenated` so it handles `ABCDef` better.
4224 + Improve `isTight` (#5857). If a list has an empty item,
4225 this should not count against its being a tight list.
4226 + Export `htmlSpanLikeElements` [API change] (Daniele D'Orazio, #5796).
4227 This is a mapping of HTML span-like elements that are internally
4228 represented as a Span with a single class.
4229 + Change the implementation of `htmlSpanLikeElements` to retain
4230 classes and attributes (#5882, Florian Beeres).
4232 * Text.Pandoc.Slides: recognize content in Divs when determining
4235 * Text.Pandoc.SelfContained:
4237 + Omit content-type on type attribute for `<style>` (#5725).
4238 It doesn't seem to be valid for HTML5, and as a result Chrome
4239 ignores the style element.
4241 * Text.Pandoc.Pretty has been removed [API change].
4242 We now use the new external doclayout module instead.
4244 * Text.Pandoc.Writers.Shared:
4246 + Remove `metaToJSON`, `metaToJSON'` [API change].
4247 + Modify `addVariablesToContext`, `defField`, `setField`, `getField`,
4248 `resetField` to work with Context rather than JSON values. [API change]
4249 + Export new function `endsWithPlain` [API change].
4250 + Change `gridTables` so it does better at keeping the widths of
4251 columns (#4320) and does better at figuring out column widths
4252 when no widths are given (#5899).
4254 * Text.Pandoc.Options
4256 + Change type of `writerTemplate` to `Maybe Template` instead
4258 + Change To/FromJSON instances for `HTMLMathMethod`, `CiteMethod`,
4259 `ObfuscationMethod`, `TrackChanges`, `WrapOption`,
4260 `TopLevelDivision`, `ReferenceLocation`, `HTMLSlideVariant` (#5790).
4261 In each case we use lowercase (or hyphenated lowercase) for
4262 constructors to line up more closely with command-line option
4263 values. This is a breaking change for those who manually decode or
4264 encode JSON for these data types (e.g. for `ReaderOptions` or
4267 * Text.Pandoc.Filters:
4269 + Add `FromYAML` instance for `Filter`.
4270 + `applyFilters`: Add and apply filters in order (not reversed)
4271 This changes `applyFilters` from Text.Pandoc.Filter so
4272 that it does a left fold rather than a right fold, applying
4273 the filters in the order listed.
4277 + Change `toEntities` to emit numerical hex character references
4280 * Text.Pandoc.Highlighting:
4282 + Add additional listings languages (Wandmalfarbe).
4284 * Text.Pandoc.MediaBag:
4286 + Some of the types using Strings were switched to use FilePath instead
4287 (Christian Despres, #5884).
4289 * Text.Pandoc.Templates:
4291 + Don't import/export `varListToJSON` [API change]. It is removed in
4292 doctemplates >= 0.3.
4293 + Change type of `renderTemplate'` [API change]. Return value is
4294 now Text rather than being polymorphic. This makes room for
4295 upcoming removal of the `TemplateTarget` class from doctemplates.
4297 * Text.Pandoc.App.Opt [API changes]:
4299 + More convenient To/FromJSON instances. Make the field names
4300 like `strip-empty-paragraphs` rather than `optStripEmptyParagraphs`.
4301 Anyone who is using JSON serialization of Opt will need to adjust
4303 + Change `optHighlightStyle` to a `Maybe String` instead of
4304 `Maybe Style`. Do the parsing/loading of themes later, after
4306 + Remove `optBaseHeaderLevel` from `Opt`. We now just use
4307 `optShiftHeadingLevelBy`, to avoid redundancy.
4308 + Change `optShiftHeadingLevel` to `optShiftHeadingLevelBy` to
4310 + Custom FromJSON instance for `LineEnding`, so either `CRLF`
4311 or `crlf` will work.
4312 + Change `optVariables` from `[(String, String)]` to `Context Text`.
4313 + Change `optMetadata` to `Meta`, to allow structured values.
4314 The current behavior of the `--metadata` option stays the same.
4315 + Rename `optReader`, `optWriter` as `optFrom`, `optTo`.
4316 + Add `FromYAML` instances to `Opt` and to all subsidiary types.
4317 + Rename `optMetadataFile` to `optMetadataFiles`.
4318 + Rename `optPDFEngineArgs` to `optPDFEngineOpts`.
4319 + Rename `optWrapText` to `optWrap`.
4320 + Add `IpynbOutput` enumerated type: use this instead of
4321 a string for `optIpynbOutput`.
4322 + Change optInputFiles to a `Maybe [FilePath]` (#5888) [API change].
4323 `Nothing` means: nothing specified.
4324 `Just []` means: an empty list specified (e.g. in defaults).
4325 + List fields in Opt so they aren't reversed (#5881) [API change].
4326 Previously `optIncludeInHeader`, etc. were in reverse order.
4327 + The `sourcefile` variable is now always a list. It used to be
4328 sometimes a string, sometimes a list (when there was more than one).
4332 + default.latex: added code for `cslreferences` environment,
4333 to be used for pandoc-citeproc references. A `csl-hanging-indent`
4334 variable (set automatically if there is a `hanging-ident`
4335 class on the references Div) controls whether contents of this
4336 environment receive a hanging indent.
4337 + default.latex: Add `space` as default option for xeCJK, so that
4338 spaces between words are preserved (#5855, jeongminkim-islab).
4339 This is necessary for Korean.
4340 + default.latex: Remove include of `grffile` (#5848).
4341 This package used to be needed for proper handling of image filenames
4342 containing periods (in addition to the period before the extension).
4343 It no longer works with the latest LaTeX kernel and graphicx,
4344 so we have removed it. Future versions of graphicx will handle
4345 these filenames without the need for `grffile`.
4346 + default.context: add a saner default for page numbers.
4347 Previously they appeared centered at the top of the page;
4348 now we put them centered at the bottom, unless the `pagenumbering`
4350 + default.context: define a start-stop-pair `cslreferences` to
4351 allow for hanging indents in the bibliography (#5875, Denis Maier).
4352 + default.ms: update defaults. Use Palatino font, use slightly
4353 wider interparagraph space, don't indent paragraphs,
4354 and put page numbers on the bottom. This brings ms output
4355 closer to default LaTeX output.
4356 + default.revealjs: add navigationMode (Mauro Bieg, #5657).
4357 + default.muse: handle multiple authors better.
4358 + docbook4, docbook5 templates: add indentation to body.
4359 + HTML-based templates: use `styles.html` partial to avoid
4361 + HTML-based templates: change indentation of styles in template.
4363 * reference.docx (#5820):
4365 + Change Block Text (block quote) style so that the same font
4366 is used as in the body text, and the block text is indented
4368 + All headings now have a uniform color.
4369 + Level-1 headings no longer set `w:themeShade="B5"`.
4370 + Level-2 headings are now 14 point rather than 16 point.
4371 + Level-3 headings are now 12 point rather than 14 point.
4372 + Level-4 headings are italic rather than bold.
4374 * epub.css: Add CSS for hanging-indent div to support pandoc-citeproc's
4375 new hanging indents.
4379 + Repeat ghc-options in all stanzas.
4380 + Remove conditionals for ghc < 8 (Albert Krewinkel, #5693).
4381 + Compile with `-Wcpp-undef` and `-fhide-source-paths` when possible
4383 + Add cabal.project to extra-source-files (Albert Krewinkel).
4384 + Add dependency on skylighting-core (#5729). Even though it shouldn't
4385 be needed, some tools require it.
4386 + Require latest pandoc-types, texmath, skylighting, haddock-library.
4387 + Ensure TemplateHaskell is added to other-extensions when it is
4388 used (Vanessa McHale, #5728).
4389 + Remove `derive_json_via_th` flag; always use TH. This cuts down
4390 on code duplication and reduces the chance for errors. See #4083.
4396 + Remove references to obsolete flag (#5694).
4398 * Benchmarks: fix failure on ipynb.
4400 * Use MathJax 3 (zorkow).
4402 * KaTeX math: respect `classoption=fleqn` variable,
4403 bump KaTeX version to 0.11.1 (#5815, Mauro Bieg).
4405 * Fix redundant constraint compiler warnings (Pete Ryland, #5625).
4407 * Use throwError instead of fail when appropriate.
4409 * Use Prelude.fail to avoid ambiguity with fail from GHC.Base.
4411 * Add `diff-zip.sh` to tools (John MacFarlane, Agustín Martín Barbero).
4412 This is intended to make it easier to test differences in zip
4413 containers, like epub, docx, or odt.
4415 * Add `.gitattributes` (#5747). This ensures that the golden files
4416 in `test/fb2/reader/` don't have newlines converted. This should fix
4417 a test failure on GitHub CI with Windows.
4419 * Reorder options in `--help`.
4421 * Revise code for HsYAML-0.2.0.0 (@vijayphoenix, #5704).
4423 * Remove blank line in code example in Haddocks (Leif Metcalf, #5679).
4425 * Fix trypandoc with `getReader`/`getWriter` changes.
4427 * Allow building pandoc with GHC 8.8.
4429 * linux tarball: add architecture `-amd64` to filename.
4430 Now it will be: `pandoc-VERSION-linux-ARCH.tar.gz`.
4434 + Add section for exit codes.
4435 + Document some pptx limitations in slide show section:
4436 No incremental display (#5689). No pause with `. . .` (#5701).
4437 + reveal.js flags (Mauro Bieg, #5653).
4438 + Document addition of `data-` prefix to unknown attributes in HTML5.
4439 + Link to YAML spec (Mauro Bieg, #5687).
4440 + Fix capitalization of "Linux" (#5859, Marcus Stollsteimer).
4441 + Use a table for exit codes.
4442 + Put all template variable docs into one section.
4443 + Use ATX headers consistently.
4444 + Add fuller documentation of templates (#5779), including
4445 new template syntax, partials, etc.
4446 + Add documentation for the variable `hyperrefoptions` (Wandmalfarbe).
4447 + Clarify when macro definitions are passed as raw latex.
4448 In Markdown input, they are always passed through. In LaTeX, only if
4449 `latex_macros` is disabled.
4450 + Clarify that `--dpi` provides a default and doesn't override
4451 dpi values specified in the images themselves (#5721).
4452 + Document how to use custom writers with `--standalone` (#5866).
4453 + Clarify `--preserve-tabs` default.
4457 + Fix instructions for libicu.
4458 + Add Void Linux instructions (Volodymyr Kozachnyskyi).
4462 + Add information on tests (Agustín Martín Barbero, #5652).
4463 + Add information about command test naming to CONTRIBUTING (Florian B).
4465 * Fix typos in changelog and comments (#5896, Brian Wignall).
4467 * doc/lua-filters.md:
4469 + Fix mistakes in mediabag module docs (#5851, Albert Krewinkel).
4470 + Improve metadata replacement example in lua-filters doc (#5851).
4471 + Mention which Lua version is shipped with pandoc (Albert Krewinkel,
4474## pandoc 2.7.3 (2019-06-11)
4476 * Add `jira` (Atlassian's Jira wiki markup) as output format (#2497,
4479 * Add `tex_math_dollars` to `multimarkdownExtensions` (#5512).
4480 This form is now supported in multimarkdown,
4481 in addition to `tex_math_double_backslash`.
4483 * Fix `--self-contained` so it works when output format has extensions.
4484 Previously if you used `--self-contained` with `html-smart` or
4485 `html+smart`, it wouldn't work.
4487 * Add template variable `curdir` with working directory
4488 from which pandoc is run (#5464).
4490 * Markdown reader: don't create implicit reference for empty header
4493 * Muse reader: allow images inside link descriptions (Alexander Krotov).
4495 * HTML reader: epub related fixes.
4497 + With epub extensions, check for `epub:type` in addition to `type`.
4498 + Fix problem with noteref parsing which caused block-level
4499 content to be eaten with the noteref.
4500 + Rename `pAnyTag` to `pAny`.
4501 + Refactor note resolution.
4502 + Trim definition list terms (Alexander Krotov).
4506 + Add braces when resolving `\DeclareMathOperator`
4507 (#5441). These seem to be needed for xelatex but not pdflatex.
4508 + Allow newlines in `\mintinline`.
4509 + Pass through unknown listings language as class (#5540).
4510 Previously if the language was not in the list of languages supported
4511 by listings, it would not be added as a class, so highlighting
4512 would not be triggered.
4513 + `rawLaTeXInline`: Include trailing `{}`s in raw latex commands (#5439).
4514 This change affects the markdown reader and other readers that allow raw
4515 LaTeX. Previously, trailing `{}` would be included for unknown
4516 commands, but not for known commands. However, they are sometimes used
4517 to avoid a trailing space after the command. The chances that a `{}`
4518 after a LaTeX command is not part of the command are very small.
4520 * MediaWiki reader: handle multiple attributes in table row (#5471,
4523 * Docx reader: Add support for `w:rtl` (#5545). Elements with this
4524 property are put into Span inlines with `dir="rtl"`.
4526 * DocBook reader: Issue `IgnoredElement` warnings.
4528 * Org reader (Albert Krewinkel):
4530 + Fix planning elements in headers level 3 and higher
4531 (#5494). Planning info is now always placed before
4532 the subtree contents. Previously, the planning info was placed after
4533 the content if the header's subtree was converted to a list, which
4534 happens with headers of level 3 and higher per default.
4535 + Omit, but warn about unknown export options. Unknown export
4536 options are properly ignored and omitted from the output.
4537 + Prefer plain symbols over math symbols (#5483).
4538 Symbols like `\alpha` are output plain and unemphasized, not as math.
4539 + Recognize emphasis after TODO/DONE keyword (#5484).
4543 + Skip unknown elements rather than throwing errors (#5560).
4544 Sometimes custom elements are used (e.g. `id` element
4545 inside `author`); previously the reader would halt with an error.
4546 Now it skips the element and issues an `IgnoredElement` warning.
4547 + Parse notes (#5493, Alexander Krotov).
4548 + Internal improvements (Alexander Krotov).
4550 * OpenDocument writer: Roll back automatic figure/table numbering
4551 (#5474). This was added in pandoc 2.7.2, but it makes it impossible
4552 to use pandoc-crossref. So this has been rolled back for now,
4553 until we find a good solution to make this behavior optional
4554 (or a creative way to let pandoc-crossref and this feature
4557 * New module Text.Pandoc.Writers.Jira, exporting `writeJira` [API
4558 change] (Albert Krewinkel).
4562 + Don't include 'landmarks' if there aren't any.
4563 Previously we could get an empty ol element, which caused
4564 validation errors with epubcheck.
4565 + Ensure unique ids for styleesheets in content.opf (#5463).
4566 + Make stylesheet link compatible with kindlegen (#5466,
4567 Eric Schrijver). Pandoc omitted `type="text/css"` from both
4568 `<style>` and `<rel="stylesheet">` elements in all templates, which
4569 is valid according to the spec. However, Amazon’s kindlegen software
4570 relies on this attribute on `<link>` elements when detecting
4571 stylesheets to include.
4575 + Output video and audio elements depending on file
4576 extension of the image path (Mauro Bieg).
4577 + Emit empty alt tag in figures (#5518, Mauro Bieg).
4578 The same text is already in the <figcaption> and
4579 screen-readers would read it twice, see #4737.
4580 + Don't add variation selector if it's already there.
4581 This fixes round-trip failures.
4582 + Prevent gratuitous emojification on iOS (#5469).
4583 iOS chooses to render a number of Unicode entities, including '↩', as
4584 big colorful emoji. This can be defeated by appending Unicode
4585 VARIATION SELECTOR-15'/'VARIATION SELECTOR-16'. So we now append this
4586 character when escaping strings, for both '↩' and '↔'. If other
4587 characters prove problematic, they can simply be added to
4588 `needsVariationSelector`. + Add `class="heading"` to level 7+ Headers
4589 rendered as `<p>` elements (#5457).
4591 * RST writer: treat Span with no attributes as transparent (#5446).
4592 Previously an Emph inside a Span was being treated as
4593 nested markup and ignored. With this patch, the Span
4598 + Include inline code attributes with `--listings` (#5420).
4599 + Don't produce columns environment unless beamer (#5485).
4600 + Fix footnote in image caption. Regression: the fix for #4683 broke
4602 + Don't highlight code in headings (#5574). This causes
4604 + Use `\mbox` to get proper behavior inside `\sout` (#5529).
4606 * EPUB writer: Fix document section assignments (#5546).
4607 For example, introduction should go in bodymatter, not frontmatter, and
4608 epigraph, conclusion, and afterward should go in bodymatter, not
4609 backmatter. For the full list of assignments, see the manual.
4613 + Add backslashes to avoid unwanted interpretation of
4614 definition list terms as other kinds of block (#554).
4615 + Ensure the code fence is long enough (#5519).
4616 Previously too few backticks were used when the code block
4617 contained an indented line of backticks. (Ditto tildes.)
4618 + Handle labels with integer names (Jesse Rosenthal, #5495).
4619 Previously if labels had integer names, it could produce a conflict
4620 with auto-labeled reference links. Now we test for a conflict and
4621 find the next available integer. This involves adding a new state
4622 variable `stPrevRefs` to keep track of refs used in other document
4623 parts when using `--reference-location=block|section`
4625 * Textile writer: fix closing tag for math output (Albert Krewinkel).
4626 Opening and closing tag for math output match now.
4628 * Org writer: always indent src blocks content by 2 spaces (#5440, Albert
4629 Krewinkel). Emacs always uses two spaces when indenting the content of
4630 src blocks, e.g., when exiting a `C-c '` edit-buffer. Pandoc used to
4631 indent contents by the space-equivalent of one tab, but now always uses
4636 + Use `` `+...+` `` form for inline code.
4637 The old `` `a__b__c` `` yields emphasis inside code in asciidoc.
4638 To get a pure literal code span, use `` `+a__b__c+` ``.
4639 + Use proper smart quotes with asciidoctor (#5487).
4640 Asciidoctor has a different format for smart quotes.
4641 + Use doubled ## when necessary for spans (#5566).
4642 + Ensure correct nesting of strong/emph (#5565): strong
4643 must be the outer element.
4647 + Wrap elements with p when needed (#5570). The JATS spec restricts
4648 what elements can go inside `fn` and `list-item`. So we wrap other
4649 elements inside `<p specific-use="wrapper">` when needed.
4650 + Properly handle footnotes (#5511) according to "best practice."
4651 (Group them at the end in `<fn-group>` and use `<xref>` elements
4653 + Fix citations with PMID so they validate (#5481). This includes
4654 an update to data/jats.csl.
4655 + Ensure validity of `<pub-date>` by parsing the date and
4656 extracting year, month, and day, as expected. Also add an
4657 iso-8601-date attribute automatically.
4658 + Don't use `<break>` element for LineBreak. It is only
4659 allowed in a few special contexts, and not in `<p>` elements.
4660 + Don't make `<string-name>` a child of `<string>`, which is illegal.
4664 + Do not wrap note references into `<sup>` and brackets
4665 (Alexander Krotov). Existing FB2 readers, such as FBReader, already
4666 display links with type="note" as a superscript.
4667 + Use genre metadata field (#5478).
4669 * Muse writer: do not escape empty line after `<br>` (Alexander Krotov).
4671 * Add unicode code point in "Missing character" warning (#5538).
4672 If the character isn't in the console font, the message is pretty useless,
4673 so we show the code point for anything non-ASCII.
4675 * Lua: add Version type to simplify comparisons (Albert Krewinkel).
4676 Version specifiers like `PANDOC_VERSION` and `PANDOC_API_VERSION` are
4677 turned into `Version` objects. The objects simplify version-appropriate
4678 comparisons while maintaining backward-compatibility.
4679 A function `pandoc.types.Version` is added as part of the newly
4680 introduced module `pandoc.types`, allowing users to create version
4683 * pandoc lua module (Albert Krewinkel):
4685 - Fix deletion of nonexistent attributes (#5569).
4686 - Better tests for Attr and AttributeList.
4688 * pandoc.mediabag lua module (Albert Krewinkel):
4690 + Add function `delete` for deleting a single item.
4691 + Add function `empty` for removing all entries.
4692 + Add function `items` for iterating over mediabag.
4694 * Text.Pandoc.Class: Fix handling of `file:` URL scheme in
4695 `downloadOrRead` (#5517, Mauro Bieg). Previously `file:/`
4696 URLs were handled wrongly and pandoc attempted to make HTTP
4697 requests, which failed.
4699 * Text.Pandoc.MIME: add `mediaCategory` [API change] (Mauro Bieg).
4701 * Text.Pandoc.Shared:
4703 + Add `onlySimpleTableCells` [API change]
4704 (Mauro Bieg) and use this to consolidate simple-table detection
4705 (#5524). This fixes an inconsistency in the HTML reader, which did not
4706 treat tables with `<p>` inside cells as simple.
4707 + `metaToJSON`: treat digits starting with 0 as a string, not a number
4708 (#5479). This fixes a regression in YAML metadata in pandoc 2.7.2.
4710 * Text.Pandoc.Logging: Add `IgnoredElement` constructor for
4711 `LogMessage`. `SkippedConetnt` doesn't work for some XML-based
4712 readers which don't have access to parsec source positions.
4714 * Text.Pandoc.Asciify: Add Turkish undotted-i (#5433, Mauro Bieg).
4716 * Improve output of Lua tests (#5499, Albert Krewinkel).
4717 This makes use of tasty-lua, a package to write tests in Lua
4718 and integrate the results into Tasty output. Test output becomes
4719 more informative: individual tests and test groups become visible
4720 in test output. Failures are reported with helpful error messages.
4722 * Lua: add `pandoc.system` module (#5468, Albert Krewinkel).
4723 The `system` Lua module provides utility functions to interact with the
4724 operating- and file system. E.g.
4725 `print(pandoc.system.get_current_directory())`
4728 pandoc.system.with_temporary_directory('tikz', function (dir)
4729 -- write and compile a TikZ file with pdflatex
4733 * LaTeX template: Add pdflang to hypersetup if lang is set (#5443).
4735 * beamer template: Fix using Beamer with geometry (#5505, Daniel Maslowski).
4736 Beamer already loads geometry, so we need to use the `\geometry`
4737 command to set geometry options.
4739 * EPUB2/3 templates: Move inline styles to default epub.css (#5466).
4740 NOTE: Those who use a custom CSS stylesheet with EPUB should
4743 code{ white-space: pre-wrap; }
4744 span.smallcaps{ font-variant: small-caps; }
4745 span.underline{ text-decoration: underline; }
4746 q { quotes: "“" "”" "‘" "’"; }
4747 div.column{ display: inline-block; vertical-align: top; width: 50%; }
4749 * reveal.js template:
4751 + Updates for revealjs 3.8.0 (#5435, ebiim).
4752 + Remove reference to head.min.js (#5448, Winnie Hellmann).
4753 NOTE: users will need to update reveal.js to at least 3.8.0
4754 for their presentations generated with this version of pandoc
4759 + Replace `</>` with literal `/` (#5451).
4760 We use forward-slash for a directory separator in tmpDir,
4761 even on Windows (because that's what tex likes). So we
4762 should not put a backslash between the tmpDir and the
4763 filename on Windows. This is harmless enough in normal
4764 Windows setups, but it breaks on Cygwin.
4765 Thanks to @cc2x for noticing and diagnosing the problem.
4766 + Allow use of `-output-directory` in `--pdf-engine-opt` (#5462).
4767 This is currently possible with `mklatex` and `-outdir`, but
4768 was not yet possible with xelatex and `-output-directory`.
4769 + For PDF via ms/pdfroff, make TOC appear at beginning and in
4770 PDF bookmarks (#5553). Previously the TOC appeared at the end
4771 of the document, and was not bookmarked. To keep it at the end,
4772 add `--pdf-engine-opt=--no-toc-relocation` to your command line.
4774 * Fix broken links in documents (#5473, Shim Myeongseob).
4775 Also, use absolute links to pandoc.org when possible, so that
4776 the links can be followed by people viewing these documents
4779 * Improved sample lua tikz filter in lua-filters docs (#5445,
4780 Matthew Doty). There are three changes:
4782 + It only processes elements which begin with `\begin{tikzpicture}`
4783 + It uses pdf2svg instead of imagemagick to preserve fidelity
4784 + The images produced have transparent backgrounds
4788 + Add note about `title-meta`, `author-meta`, `date-meta` (#5486).
4789 + Fix typo (#5489, Christian Krause).
4791 * add test/tables.xwiki to git and pandoc.cabal (#5498, Mauro Bieg).
4793 * Disable missing-home-modules warning in `stack.yaml`.
4794 Otherwise `stack ghci` fails.
4797## pandoc 2.7.2 (2019-04-05)
4799 * Add XWiki writer (#1800, Derek Chen-Becker).
4800 Add `Text.Pandoc.Writers.XWiki`, exporting `writeXWiki` [API change].
4802 * Dokuwiki Reader: parse single curly brace (#5416, Mauro Bieg).
4804 * Vimwiki reader: improve handling of internal links (#5414).
4805 We no longer append `.html` to link targets, and we add a title
4806 `wikilink`. This mirrors behavior of other wiki readers. Generally
4807 the `.html` extension is not wanted. It may be important for output
4808 to HTML in certain circumstances, but it can always be added using a
4809 filter that matches on links with title `wikilink`.
4811 If your workflow requires the current behavior, here is a lua filter
4812 that will add the `.html` extension:
4816 if el.title == 'wikilink' then
4817 el.target = el.target .. ".html"
4824 + Use format `ipynb` for raw cell where no format given.
4825 + Avoid introducing spurious `.0` on integers in metadata.
4827 * Markdown reader: fenced div takes priority over setext header.
4829 * HTML reader: read `data-foo` attribute into `foo` (#5392).
4830 The HTML writer adds the `data-` prefix for HTML5 for nonstandard
4831 attributes. But the attributes are represented in the AST without
4832 the `data-` prefix, so we should strip this when reading HTML.
4834 * LaTeX reader: Improve autolink detection (#5340).
4836 * PowerPoint writer (Jesse Rosenthal):
4838 + Expand builtin reference doc to model all layouts.
4839 The previous built-in reference doc had only title and content
4840 layouts. Add in a section-header slide and a two-content slide, so
4841 users can more easily modify it to build their own templates.
4842 + Always open up in slide view. When editing a
4843 template/reference-doc, the user might be in Master view, but when
4844 producing a slide show, it is assumed that slide view will be
4846 + Remove `handoutsMasterList` from template presentation.xml
4847 + Fix numerous errors in templating (#5402). Previously, some templates
4848 produced by Office 365 (MacOS) would not render with `--reference-doc`
4849 correctly. We now apply correct shapes for content, and build
4850 shape trees correctly.
4851 + Make default placeholder type for template lookup.
4852 + Apply speaker notes to metadata slide if applicable.
4853 + Test for speaker notes after breaking header.
4854 + Correctly handle notes after section-title header.
4855 Previously, if notes came after a section-title header (ie, a level-1
4856 header in a slide-level=2 presentation), they would go on the next
4857 slide. This keeps them on the slide with the header.
4858 + Internal improvements.
4862 + Use format `ipynb` for raw cell where no format given.
4863 According to nbformat docs, this is supposed to render in every
4864 format. We don't do that, but we at least preserve it as a raw
4865 block in markdown, so you can round-trip.
4866 + Consolidate adjacent raw blocks. Sometimes pandoc creates two
4867 HTML blocks, e.g. one for the open tag and one for a close tag.
4868 If these aren't consolidated, only one will show up in output cell.
4869 + Fixed carry-over of nbformat from metadata.
4870 + Preserve `nbformat_minor` if it's given. This helps with round-tripping.
4874 + Avoid inadvertently creating ?` or !` ligatures (#5407).
4875 These are upside down ? and !, resp.
4876 + Fix footnotes in table caption and cells (#5367). This fixes a
4877 bug wherein footnotes appeared in the wrong order, and with
4878 duplicate numbers, when in table captions and cells.
4879 We now use regular `\footnote` commands, even in the table
4880 caption and the minipages containing cells. Apparently
4881 longtable knows how to handle this.
4883 * HTML writer: Don't add data- prefix to RDFa attributes (#5403).
4885 * JATS writer: Ensure that plain strings go inside `<pub-id>` tag (#5397).
4889 + Better rendering of numbers (#5398). If the number is integral,
4890 we render it as an integral not a float.
4891 + Proper rendering of empty map in YAML metadata (#5398). Should
4892 be `{}`, not empty string.
4893 + Properly escape attributes in Markdown writer (#5369).
4894 + Be sure implicit figures work in list contexts (#5368).
4895 Previously they would sometimes not work: e.g., when they
4896 occurred in final paragraphs in lists that were originally
4897 parsed as Plain and converted later using PlainToPara.
4899 * Docx writer: Use `w:br` without attributes for line breaks (#5377).
4900 We previously added the attribute `type="textWrapping"`, but
4901 this causes problems on Word Online.
4903 * LaTeX template (Andrew Dunning):
4905 + Ensure correct heading/table order (#5365). Improve workaround
4906 (#1658) for tables following headings. The new solution works whether
4907 or not the `indent` variable is enabled.
4908 + Remove `subparagraph` variable. The default is now to use run-in
4909 style for level 4 and 5 headings (`\paragraph` and `\subparagraph`).
4910 To get the previous default behavior (where these were formatted as
4911 blocks, like `\subsubsection`), set the `block-headings` variable.
4912 + Add pandoc to PDF metadata (#5388).
4913 + Group graphics-related code (#5389).
4914 + Move `\setstretch` after front matter (#5179). Ensures that
4915 `\maketitle`, `\tableofcontents`, and so forth are not affected by
4916 changes to line spacing.
4918 * Update data/jats.csl to avoid commas between name-part elements
4921 * Add support for golang (`go`) with `--listings` (#5427).
4923 * Text.Pandoc.Shared - improve `metaToJSON` behavior with numbers.
4924 We now do a better job marshalling numbers from MetaString
4925 or MetaInlines into JSON Number.
4927 * Text.Pandoc.Writers.Shared: `metaValueToJSON`: use Number Values for
4928 integers. Pandoc's MetaValue doesn't have a distinguished number type,
4929 so numbers are put in MetaStrings. If the MetaString consists
4930 entirely of digits, we convert it to a Number. We should probably
4931 consider adding a MetaNumber constructor to MetaValue, for better
4932 round-tripping with JSON etc. This change aids round-tripping in
4933 ipynb metadata fields, like `toc_depth`.
4935 * Text.Pandoc.Class: `fetchItem`: don't treat UNC paths as
4936 protocol-relative URLs (#5127). These are paths beginning `//?/UNC/...`.
4938 * Text.Pandoc.ImageSize: Improve `pdfSize` so it handles
4939 a wider range of PDFs (#4322, with help from Richard Davis).
4941 * Text.Pandoc.Pretty: avoid stack overflow by using strict sum (#5401).
4943 * Fix harmless error in file-scope code (#5422).
4947 + Improve 'header' and 'heading' usage (#5423, Andrew
4948 Dunning). The term 'header' was being used where 'heading'
4949 is more appropriate.
4950 + Add paragraph on options affecting markdown in ipynb.
4952 * stack.yaml - remove -Wmissing-home-modules
4953 This seems to cause problems with stack ghci. Remove RTS options.
4955 * Add ghc-options to cabal.project.
4957 * appveyor.yml - use ghc 8.6.4. Fixes segfault issues on Windows (#5037).
4959 * linux build process: Remove clone of pandoc-citeproc (#5366).
4960 It wasn't being used; cabal.project specifies the version to use.
4962## pandoc 2.7.1 (2019-03-14)
4964 * Add tectonic as an option for --pdf-engine (#5345, Cormac Relf).
4965 Runs tectonic on STDIN instead of a temporary .tex file, so that it
4966 looks in the working directory for `\include` and `\input` like the rest
4967 of the engines. Allows overriding the output directory
4968 args with `--pdf-engine-opt=--outdir --pdf-engine-opt="$DIR"`.
4970 * Allow `-o/--output` to be used with `--print-default-data-file`,
4971 `--print-highlighting-style`, `--print-default-template`.
4972 Note that `-o` must occur BEFORE the `--print*` command on the command
4973 line (this is documented, #5357).
4977 + Support `\underline`, `\ul`, `\uline` (#5359, Paul
4978 Tilley). These are parsed as a Span with class `underline`, as
4980 + Ensure that `\Footcite` and `\Footcites` get put in a note.
4984 + Remove sensitivity to `raw_html`, `raw_tex` extensions.
4985 We now include every output format. Pruning is handled by
4987 + Better handling of cell metadata. We now include even complex cell
4988 metadata in the Div's attributes (as JSON, in complex cases, or
4989 as plain strings in simple cases).
4993 + Recurse into native divs for output cell data (#5354).
4994 + Render cell metadata fields from div attributes.
4996 * Docx writer: avoid extra copy of abstractNum and num elements
4997 in numbering.xml. This caused pandoc-produced docx files to
4998 be uneditable using Word Online (#5358).
5000 * Markdown writer: improve handling of raw blocks/inline.
5001 We now emit raw content using `raw_attribute` when no more
5002 direct method is available. Use of `raw_attribute` can be
5003 forced by disabling `raw_html` and `raw_tex`.
5005 * LaTeX writer: Add classes for frontmatter support (#5353, Andrew Dunning)
5006 and remove frontmatter from `scrreprt`.
5010 + Improve readability (#5363, Andrew Dunning).
5011 + Robust section numbering removal (#5351, Andrew Dunning).
5012 Ensures that section numbering does not reappear with custom
5013 section levels. See <https://tex.stackexchange.com/questions/473653/>.
5014 + Better handling of front/main/backmatter (#5348).
5015 In pandoc 2.7 we assumed that every class with chapters would accept
5016 `\frontmatter`, `\mainmatter`, and `\backmatter`. This is not so (e.g.
5017 report does not). So pandoc 2.7 breaks on report class by including an
5018 unsupported command. Instead of the `book-class` variable, we use two
5019 variables, `has-chapters` and `has-frontmatter`, and set these
5020 intelligently in the writer.
5022 * Text.Pandoc.Shared: Improve `filterIpynbOutput`. Ensure that
5023 images are prioritized over text. `best` should include everything
5026 * Tests.Old: specify `--data-dir=../data` to ensure tests can find
5027 data files even if they haven't been installed. Remove old
5028 `pandoc_datadir` environment variable, which hasn't done anything for a
5031 * MANUAL.txt: Add recommendation to use `raw_attribute` with ipynb (#5354).
5033 * Use cmark-gfm-hs 0.1.8 (note that 0.1.7 is buggy).
5035 * Use latest pandoc-citeproc, texmath.
5038## pandoc 2.7 (2019-03-03)
5040 * Use XDG data directory for user data directory (#3582). Instead of
5041 `$HOME/.pandoc`, the default user data directory is now
5042 `$XDG_DATA_HOME/pandoc`, where `XDG_DATA_HOME` defaults to
5043 `$HOME/.local/share` but can be overridden by setting the
5044 environment variable. If this directory is missing, then
5045 `$HOME/.pandoc` is searched instead, for backwards compatibility.
5046 However, we recommend moving local pandoc data files from
5047 `$HOME/.pandoc` to `$HOME/.local/share/pandoc`. On Windows the
5048 default user data directory remains the same.
5050 * Slide show formats behavior change: content under headers
5051 less than slide level is no longer ignored, but included in
5052 the title slide (for HTML slide shows) or in a slide after
5053 the title slide (for beamer). This change makes possible
5054 2D reveal.js slideshows with content in the top slide on
5055 each stack (#4317, #5237).
5057 * Add command line option `--ipynb-output=all|none|best` (#5339).
5058 Output cells in ipynb notebooks often contain several different
5059 versions of an output, with different MIME types, e.g. an HTML
5060 table and a plain-text fallback. Specifying `--ipynb-output=best`
5061 (the default) ensures that the best version for the output format
5062 is used. `all` includes all versions, and `none` suppresses them
5063 all, leaving output cells empty.
5065 * `asciidoctor` is now an output format separate from `asciidoc`,
5066 to accommodate some minor implementation-specific differences
5067 (currently just in the treatment of display math).
5069 * Add `latexmk` as an option for `--pdf-engine` (#3195).
5070 Note that you can use `--pdf-engine-opt=-outdir=bar` to specify
5071 a persistent temp directory.
5075 + Improve tight/loose list handling (#5285). Previously the
5076 algorithm allowed list items with a mix of Para and Plain, which
5078 + Add newline when parsing blocks in YAML (#5271). Otherwise last
5079 block gets parsed as a Plain rather than a Para. This is a
5080 regression in pandoc 2.x. This patch restores pandoc 1.19
5082 + Make `yamlToMeta` respect extensions (#5272, Mauro Bieg).
5083 This adds a `ReaderOptions` parameter to `yamlToMeta` [API change].
5084 + Fix bug parsing fenced code blocks (#5304). Previously parsing
5085 would break if the code block contained a string of backticks of
5086 sufficient length followed by something other than end of line.
5088 * LaTeX reader: don't let `\egroup` match `{`. `braced` now actually
5089 requires nested braces. Otherwise some legitimate command and
5090 environment definitions can break.
5092 * Docx reader (Jesse Rosenthal):
5094 + Rename `getDocumentPath` as `getDocumentXmlPath`.
5095 + Use field notation for setting `ReaderEnv`.
5096 + Figure out `document.xml` path once at the beginning of parsing, and
5097 add it to the environment, so we can avoid repeated lookups.
5098 + Dynamically determine main document xml path (#5277).
5099 The desktop Word program places the main document file in
5100 `word/document.xml`, but the online word places it in
5101 `word/document2.xml`. This file path is actually stated in the root
5102 `_rels/.rels` file, in the `Relationship` element with an
5103 `http://../officedocument` type.
5104 + Fix paths in archive to prevent Windows failure (#5277).
5105 Some paths in archives are absolute (have an opening slash) which, for
5106 reasons unknown, produces a failure in the test suite on MS
5107 Windows. This fixes that by removing the leading slash if it exists.
5108 + Add comments to aid code readability.
5109 + Trim space inside the last inline (#5273).
5110 + Unwrap sdt elements in footnotes and comments (#5302).
5112 * Muse reader (Alexander Krotov):
5114 + Test that block level markup does not break `<verbatim>`.
5115 + Add secondary note support.
5117 * ipynb reader: handle images referring to attachments. Previously
5118 we didn't strip off the attachment: prefix, so even though the
5119 attachment was available in the mediabag, pandoc couldn't find it.
5123 + Fix parsing of figures (#5321). This ensures that a figure
5124 containing a single image is parsed as a pandoc "implicit
5125 figure" (i.e., a Para with a single Image whose title attribute
5126 begins with `fig:`). More complex figures will still be parsed
5128 + Support `fig-group` block element (#5317).
5129 + Handle citations with multiple references (#5310). The `rid`
5130 attribute can have a space-separated list of ids.
5132 * AsciiDoc Writer: Add `writeAsciiDoctor` [API change, Tarik Graba].
5133 Handle display math appropriately for Asciidoctor.
5135 * JATS writer: wrap figure caption in `<p>` to fix validation (#5290,
5140 + Implement WAI-ARIA roles for (end)notes, citations, and
5141 bibliography (#4213). Note that doc-biblioref is only used when
5142 link-citations produces links, since it belongs on links.
5143 + Include content (including speaker notes) in title slides
5148 + Ensure final newline.
5149 + Only include metadata under `jupyter` field.
5150 + Don't create attachments for images with absolute URIs,
5151 including data: URIs (#5303).
5152 + Keep plain text fallbacks in output even if a richer format is
5153 included (#5293). We don't know what output format will be needed.
5154 See the `--ipynb-output` command line option for a way to control
5155 what formats are included in the output.
5157 * Markdown writer: use `markdown="1"` when appropriate for Divs:
5158 when `native_divs` and `markdown_in_html_blocks` are disabled
5159 but `raw_html` and `markdown_attribute` are enabled.
5163 + Use right fold for `escapeString`. This is more elegant than
5164 the explicit recursive code we were using.
5165 + Avoid `{}` after control sequences when escaping.
5166 `\ldots{}.` doesn't behave as well as `\ldots.` with the latex
5167 ellipsis package. This patch causes pandoc to avoid emitting
5168 the `{}` when it is not necessary. Now `\ldots` and other
5169 control sequences used in escaping will be followed by either
5170 a `{}`, a space, or nothing, depending on context.
5171 + For beamer, include contents under headers superordinate to
5172 slidelevel (#4317). Currently we keep the fancy title slide, and
5173 add a new slide with the same title and whatever content was
5176 * Powerpoint writer (Jesse Rosenthal): support underlines.
5177 Use span with single class "underline" as in docx writer.
5179 * Muse writer: escape secondary notes (Alexander Krotov).
5181 * FB2 writer: add section identifiers support (#5229, John KetzerX).
5183 * Make `--fail-if-warnings` work for PDF output (#5343).
5185 * Lua filters (Albert Krewinkel):
5187 + Load module `pandoc` before calling `init.lua` (#5287). The file
5188 `init.lua` in pandoc's data directory is run as part of pandoc's
5189 Lua initialization process. Previously, the `pandoc` module was
5190 loaded in `init.lua`, and the structure for marshaling was
5191 set up after. This allowed simple patching of element
5192 marshaling, but made using `init.lua` more difficult. Now, all
5193 required modules are now loaded before calling `init.lua`. The
5194 file can be used entirely for user customization. Patching
5195 marshaling functions, while discouraged, is still possible via
5197 + All Lua modules bundled with pandoc, i.e., `pandoc.List`,
5198 `pandoc.mediabag`, `pandoc.utils`, and `text` are re-exported from the
5199 `pandoc` module. They are assigned to the fields `List`, `mediabag`,
5200 `utils`, and `text`, respectively.
5202 * Text.Pandoc.Lua (Albert Krewinkel):
5204 + Split `StackInstances` into smaller Marshaling modules.
5205 + Get `CommonState` from Lua global. This allows more control over
5206 the common state from within Lua scripts.
5210 + Support the `subject` metadata variable (#5289, Pascal Wagler).
5211 + Add `\frontmatter`, `\mainmatter`, `\backmatter`
5212 for book classes (#5306).
5214 * epub3 template: Add titlepage class to section (#5269).
5216 * HTML5 template: Add ARIA role `doc-toc` for table of contents (#4213).
5218 * Make `--metadata-file` use selected extensions (#5279, #5272, Mauro Bieg).
5220 * Text.Pandoc.Shared:
5222 + Remove `withTempDir` [API change].
5223 + Add new exported function `defaultUserDataDirs` [API change].
5224 + Add `filterIpynbOutput` [API change].
5225 + `compactify`: Avoid lists with a mix of Plain and Para elements
5228 * Text.Pandoc.Translations: reorder alphabetically and remove `Author`
5229 (#5334, Mauro Bieg).
5231 * Text.Pandoc.Extensions:
5233 + More carefully groom ipynb default extensions.
5234 + Add `all_symbols_escapable` to `githubMarkdownExtensions`.
5238 + Use system temp directory when possible (#1192). Previously we
5239 created temp dirs in the working directory, partly (a) because
5240 there were problems using the system temp directory on Windows,
5241 when their pathnames included tildes, and partly (b) because
5242 programs like `epstopdf.pl` would not be allowed to write to
5243 directories outside the working directory in restricted mode. We
5244 now (a) use the system temp dir except when the path includes
5245 tildes, and (b) set `TEXMFOUTPUT` when creating the PDF, so that
5246 subsidiary programs can use the system temp directory. This
5247 addresses problems that occurred when pandoc was used in a
5248 synced directory (such as Dropbox).
5249 + Change types of subsidiary functions to PandocIO, to allow
5250 warnings to be threaded through (#5343).
5252 * Text.Pandoc.MIME: add WebP (#5267, Mauro Bieg).
5254 * Tests: avoid calling `findPandoc` multiple times.
5256 * Old tests: remove need for temp files by using `pipeProcess`.
5258 * Added simple ipynb reader/writer tests (#5274).
5260 * Rearrange `--help` output in a more rational way, with common
5261 options at the beginning and options grouped by function (#5336).
5263 * trypandoc: Add JATS and other missing formats (Arfon Smith, #5291).
5265 * Add missing copyright notices and remove license boilerplate (#4592,
5268 * Use latest basement/foundation on 32bit windows.
5270 * Use latest skylighting (#5328). Custom syntax definitions no
5271 longer try to load `language.dtd`.
5273 * Require texmath 0.11.2.1
5275 * Use latest pandoc-citeproc (0.16.1.1).
5279 + Clarify variable substitution indentation in templates (#5338,
5280 Agustín Martín Barbero).
5281 + Reorder custom-styles section (#5324, Mauro Bieg).
5283## pandoc 2.6 (2019-01-30)
5285 * Support ipynb (Jupyter notebook) as input and output format.
5287 + Add `ipynb` as input and output format (extension `.ipynb`).
5288 + Added Text.Pandoc.Readers.Ipynb [API change].
5289 + Added Text.Pandoc.Writers.Ipynb [API change].
5290 + Add `PandocIpynbDecodingError` constructor to Text.Pandoc.Error.Error
5292 + Depend on ipynb library.
5293 + Note: there is no template for ipynb.
5295 * Add DokuWiki reader (#1792, Alexander Krotov). This adds
5296 Text.Pandoc.Readers.DokuWiki [API change], and adds `dokuwiki`
5299 * Implement task lists (#3051, Mauro Bieg). Added `task_lists`
5300 extension. Task lists are supported from markdown and gfm input.
5301 They should work, to some degree, in all output formats, though in
5302 most formats you'll get a bullet list with a unicode character for
5303 the box. In HTML, you get checkboxes and in LaTeX/PDF output, a
5304 box is used as the list marker. API changes:
5306 + Added constructor `Ext_task_lists` to `Extension`.
5307 + Added `taskListItemFromAscii` and `taskListItemToAscii` to
5310 * Allow some command line options to take URL in addition to FILE.
5311 `--include-in-header`, `--include-before-body`, `--include-after-body`.
5315 + Handle empty `start` attribute (see #5162).
5316 + Treat `textarea` as a verbatim environment (#5241) and preserve
5321 + Change treatment of `number-lines` directive (Brian Leung,
5322 #5207). Directives of this type without numeric inputs should
5323 not have a `startFrom` attribute; with a blank value, the
5324 writers can produce extra whitespace.
5325 + Removed superfluous `sourceCode` class on code blocks (#5047).
5326 + Handle `sourcecode` directive as synonynm for `code` (#5204).
5330 + Remove `sourceCode` class for literate Haskell code blocks
5331 (#5047). Reverse order of `literate` and `haskell` classes on
5332 code blocks when parsing literate Haskell, so `haskell` is
5334 + Treat `<textarea>` as a verbatim environment (#5241).
5338 + Handle `minlevel` option differently (#5190, Brian Leung).
5339 When `minlevel` exceeds the original minimum level observed in
5340 the file to be included, every heading should be shifted
5342 + Allow for case of `:minlevel == 0` (#5190).
5343 + Fix treatment of links to images (#5191, Albert Krewinkel).
5344 Links with descriptions which are pointing to images are no
5345 longer parsed as inline images, but as links.
5346 + Add support for #+SELECT_TAGS (Brian Leung).
5347 + Separate filtering logic from conversion function (Brian Leung).
5349 * TWiki reader: Fix performance issue with underscores (#3921).
5351 * MediaWiki reader: use `_` instead of `-` in auto-identifiers (#4731).
5352 We may not still be exactly matching mediawiki's algorithm.
5356 + Remove `sourceCode` class for literate Haskell code blocks
5357 (#5047). Reverse order of `literate` and `haskell` classes on
5358 code blocks when parsing literate Haskell, so `haskell` is
5360 + Support `\DeclareMathOperator` (#5149).
5361 + Support `\inputminted` (#5103).
5362 + Support `\endinput` (#5233).
5363 + Allow includes with dots like `cc_by_4.0`. Previously the `.0`
5364 was interpreted as a file extension, leading pandoc not to add
5365 `.tex` (and thus not to find the file). The new behavior matches
5370 + Use `mapLeft` from Shared instead of defining own.
5372 * Docx reader (Jesse Rosenthal):
5374 + Handle level overrides (#5134).
5378 + Support custom properties (#3024, #5252, Agustín Martín Barbero).
5379 Also supports additional core properties: `subject`, `lang`,
5380 `category`, `description`.
5381 + Make Level into a real type, instead of an alias for a tuple
5384 * ICML writer (Mauro Bieg):
5386 + Support custom-styles (#5137, see #2106).
5387 + Support unnumbered headers (#5140).
5389 * Texinfo writer: Use header identifier for anchor if present (#4731).
5390 Previously we were overwriting an existing identifier with a new one.
5392 * Org writer: Preserve line-numbering for example and code blocks
5395 * Man/Ms writers: Don't escape `-` as `\-`. The `\-` gets rendered
5396 in HTML and PDF as a unicode minus sign.
5398 * Ms writer: Ensure we have a newline after .EN in display math (#5251).
5400 * RST writer: Don't wrap simple table header lines (#5128).
5402 * Asciidoc writer: Shorter delimiters for tables, blockquotes
5403 (#4364). This matches asciidoctor reference docs.
5405 * Dokuwiki writer: Remove automatic `:` prefix before internal image
5406 links (#5183, Damien Clochard). This prevented users from making
5407 relative image links.
5409 * Zimwiki writer: remove automatic colon prefix before internal
5410 images (#5183, Damien Clochard).
5412 * MediaWiki writer: fix caption, use 'thumb' instead of 'frame'
5413 (#5105). Captions used to have the word 'caption' prepended; this
5414 has been removed. Also, 'thumb' is used instead of 'frame' to
5415 allow images to be resized.
5419 + Ensure that we don't get > 2 levels of section nesting,
5420 even with slide level > 2 (#5168).
5421 + If slide level == N but there is no N-level header, make
5422 sure the next header with level > N gets treated as a slide
5423 and put in a section, rather than remaining loose (#5168).
5427 + Make `plain` RawBlocks pass through in `plain` output.
5428 + Include needed whitespace after HTML figure (#5121).
5429 We use HTML for a figure in markdown dialects that can't
5430 represent it natively.
5432 * Commonmark writer:
5434 + Fix handling of SoftBreak with `hard_line_breaks` (#5195).
5435 + Implement `--toc` (`writerTableOfContents`)
5436 in commonmark/gfm writers (#5172).
5440 + Ensure that picture transforms are done on metadata too.
5441 + Small fixes to `nav.xhtml`:
5442 Add 'landmarks' id attribute to the landmarks nav.
5443 Replace old default CSS removing numbers from ol.toc li
5444 with new rules that match `nav#toc ol, nav#landmarks ol`.
5445 We keep the `toc` class on `ol` for backwards compatibility.
5449 + Make raw content marked `beamer` pass through in `beamer`
5450 output (pandoc/lua-filters#40).
5451 + Beamer: avoid duplicated `fragile` property in some cases
5453 + Add `#` special characters for listings (#4939).
5454 This character needs special handling in `\lstinline`.
5456 * RTF writer: use `toTableOfContents` from Shared to replace
5457 old duplicated code.
5461 + Support custom properties. Also supports additional core
5462 properties: `subject`, `category`, `description` (#5252,
5463 Agustín Martín Barbero).
5464 + Use `toTableOfContents` from Shared to replace old duplicated code.
5466 * ODT writer (Augustín Martín Barbero):
5468 + Fix typo in custom properties (#2839).
5469 + Improve standard properties, including the following core
5470 properties: `generator` (Pandoc/VERSION), `description`, `subject`,
5471 `keywords`, `initial-creator` (from authors), `creation-date`
5472 (actual creation date) (#5252).
5476 + Allow '-' in filenames for custom lua writers (#5187).
5477 + sample.lua: add `SingleQuoted`, `DoubleQuoted` (#5104).
5478 + sample.lua: Add a missing `>` (MichaWiedenmann).
5480 * reveal.js template: Add `zoomKey` config (#4249).
5482 * HTML5 template: Remove unnecessary type="text/css" on style and
5483 link for HTML5 (#5146).
5485 * LaTeX template (Andrew Dunning, except where noted):
5487 + Prevent fontspec from scaling `mainfont` to match the default
5488 font, Latin Modern. A main font set to 12pt could
5489 previously appear between 11pt to 13pt depending on its design.
5490 To return to the earlier rendering, use `-V
5491 mainfontoptions="Scale=MatchLowercase"` (#5212, #5218).
5492 + Display monospaced fonts without TeX ligatures when using
5493 `--pdf-engine=lualatex`. It now matches the behaviour of other
5494 engines (#5212, #5218).
5495 + Remove the deprecated `romanfont` variable. The functionality of
5496 `mainfont` is identical (#5218).
5497 + Render `\subtitle` with the standard document classes.
5498 Previously, `subtitle` only appeared when using the KOMA-Script
5499 classes or Beamer (#5213, #5244).
5500 + Use Babel instead of Polyglossia for LuaLaTeX. This avoids
5501 several language selection problems, notably with retaining
5502 French spacing conventions when switching to a verbatim
5503 environment or another language; and in printing Greek text
5504 without hyphenation (#5193).
5505 + Use the [`xurl`](https://ctan.org/pkg/xurl) package if
5506 available, improving the appearance of URLs by allowing them to
5507 break at additional points (#5193).
5508 + Use [`bookmark`](https://ctan.org/pkg/bookmark) if available
5509 to correct heading levels in PDF bookmarks: see the [KOMA-Script
5510 3.26 release notes](https://komascript.de/release3.26) (#5193).
5511 + Require the [`xcolor`](https://ctan.org/pkg/xcolor) package to
5512 avoid a possible error when using additional packages alongside
5513 footnotes in tables (#5193, closes #4861).
5514 + Remove obsolete `fixltx2e` package, which has no functionality
5515 with TeX Live 2015 or later (#5193).
5516 + Allow multiple `fontfamilies.options` (#5193, closes #5194).
5517 + Restrict `institute` variable to Beamer (#5219).
5518 + Use [`footnotehyper`](https://ctan.org/pkg/footnotehyper)
5519 package if available to make footnotes in tables compatible
5520 with `hyperref` (#5234).
5521 + Number parts and chapters in book classes only if the
5522 `numbersections` variable is set, for consistency with other
5523 output formats. To return to the previous behaviour, use
5524 `-V numbersections -V secnumdepth=0` (#5235).
5525 + Reindent file (#5193).
5526 + Use built-in parskip handling with KOMA-Script classes (#5143, Enno).
5527 + Set default listings language for lua, assembler (#5227, John
5528 MacFarlane). Otherwise we get an error when trying to compile code
5529 with lua or assembler code. To change the default dialect
5530 (currently 5.3 for lua and x86masm for assembler), you can use
5531 `--include-in-header` to inject something like
5532 `\lstset{defaultdialect=[5.2]Lua}`.
5534 * Text.Pandoc.Readers: Changed types of `readJSON`; it now runs
5535 in an instance of PandocMonad, like the other readers and
5536 writers. [API change]
5538 * Text.Pandoc.Writers: Changed types of `writeJSON`; it now runs
5539 in an instance of PandocMonad, like the other readers and
5540 writers. [API change]
5542 * Text.Pandoc.Error: Added `PandocUTF8DecodingError` constructor
5543 for `PandocError`. [API change]
5545 * Text.Pandoc.Writers.Shared - add `toTableOfContents`. [API change]
5546 This is refactored out from the Markdown writer.
5547 It can be used in other writers to create a table of contents.
5549 * Improve error messages for UTF-8 decoding errors. Now we give
5550 the filename and byte offset (#4765).
5552 * Text.Pandoc.XML: Strip out illegal XML characters in
5553 `escapeXMLString` (#5119).
5555 * Text.Pandoc.Process: update `pipeProcess` (Albert Krewinkel).
5556 The implementation of `pipeProcess` was rewritten to fix sporadic
5557 failures caused by prematurely closed pipes.
5559 * Use `safeRead` instead of `read` everywhere in the code
5560 (John MacFarlane, Mauro Bieg, #5162, #5180).
5562 * Text.Pandoc.SelfContained: Decompress `.svgz` when
5563 converting to `data:` URI (#5163, Alexander Krotov).
5565 * Text.Pandoc.Parsing: Remove unused `HasHeaderMap`
5566 (#5175, Alexander Krotov). [API change]
5568 * Normalize Windows paths to account for change in ghc 8.6 (#5127).
5569 When pandoc is compiled with ghc 8.6, Windows paths are treated
5570 differently, and paths beginning `\\server` no longer work.
5571 This commit rewrites such patsh to `\\?\UNC\server` which works.
5572 The change operates at the level of argument parsing, so it
5573 only affects the command line program.
5575 * Simplify/fix reading of `--metadata` values on command line
5576 to avoid problems relating to hvr/HsYAML#7 (#5177).
5578 * data/pandoc.lua: auto-fix nested constructor arguments (Albert
5579 Krewinkel). Incorrect types to pandoc element constructors are
5580 automatically converted to the correct types when possible. This was
5581 already done for most constructors, but conversions are now also done
5582 for nested types (like lists of lists).
5584 * Removed custom Setup.hs, use build-type: simple. The only thing we
5585 gained from the custom build was automatic installation of the man
5586 page when using 'cabal install'. But custom builds cause problems,
5587 e.g., with cross-compilation. Installation of the man page is
5588 better handled by packagers. Note to packagers (e.g. Debian): it
5589 may be necessary to add a step installing the man page with the
5592 * Allow latest http-client, tasty, zip-archive, Glob.
5594 * Require skylighting >= 0.7.5, adding support for sml, J,
5597 * Tests: Cleaned up `findPandoc` in `Tests.Helpers`, so it
5598 works well with cabal v2.
5602 + Use button for installer links (John MacFarlane, Mauro Bieg,
5604 + Fix links and bump required stack version (max).
5605 + Improve installation notes on associated software (Andrew
5606 Dunning). Includes explanation of how to install related tools
5607 with package managers (since the method of installing
5608 `rsvg-convert` is not obvious).
5610 * doc/org.md: improve documentation of org features (Albert
5613 * doc/lua-filters.md: use 3rd level headers for module fields.
5617 + Clarify automatic identifiers (#5201). We remove
5618 non-alphanumerics. This includes, e.g., emojis.
5619 + Fix example for Div with id (Geoffrey Ely).
5620 + Update list of LaTeX packages used.
5621 + Make it clear that `hard_line_breaks` works in `gfm` (see #5195).
5622 + Mention `raw_attribute` in documentation for `raw_html` and
5623 `raw_tex` (#5240, thanks to @eiro).
5624 + Clarify that `$sep$` must come right before `$endfor$` in templates
5626 + Document metadata support for docx, odt, pptx writers (Agustín
5628 + Reorganize template variables (#5249, Andrew Dunning).
5629 Add additional headings to categorize variables, and
5630 alphabetize when there is large number; add more examples.
5631 + Document `date-meta` template variable (#5260, Tristan Stenner).
5633 * trypandoc: Fix CSS and viewport.
5635## pandoc 2.5 (2018-11-27)
5637 * Text.Pandoc.App: split into several unexported submodules (Albert
5638 Krewinkel): Text.Pandoc.App.FormatHeuristics,
5639 Text.Pandoc.App.Opt, Text.Pandoc.App.CommandLineOptions,
5640 Text.Pandoc.App.OutputSettings. This is motivated partly by the
5641 desire to reduce recompilations when something is modified,
5642 since App previously depended on virtually every other module.
5644 * Text.Pandoc.Extensions
5646 + Semantically, `gfm_auto_identifiers` is now a modifier of
5647 `auto_identifiers`; for identifiers to be set, `auto_identifiers` must
5648 be turned on, and then the type of identifier produced depends on
5649 `gfm_auto_identifiers` and `ascii_identifiers` are set. Accordingly,
5650 `auto_identifiers` is now added to `githubMarkdownExtensions` (#5057).
5651 + Remove `ascii_identifiers` from `githubMarkdownExtensions`.
5652 GitHub doesn't seem to strip non-ascii characters any more.
5654 * Text.Pandoc.Lua.Module.Utils (Albert Krewinkel)
5656 + Test AST object equality via Haskell (#5092). Equality of Lua
5657 objects representing pandoc AST elements is tested by unmarshalling
5658 the objects and comparing the result in Haskell.
5659 A new function `equals` which performs this test has been added to the
5660 `pandoc.utils` module.
5661 + Improve stringify. Meta value strings (MetaString)
5662 and booleans (MetaBool) are now converted to the literal string and the
5663 lowercase boolean name, respectively. Previously, all values of these
5664 types were converted to the empty string.
5666 * Text.Pandoc.Parsing: Remove Functor and Applicative constraints where Monad
5667 already exists (Alexander Krotov).
5669 * Text.Pandoc.Pretty: Don't render BreakingSpace at end of line
5670 or beginning of line (#5050).
5672 * Text.Pandoc.Readers.Markdown
5674 + Fix parsing of citations, quotes, and underline emphasis
5675 after symbols. Starting with pandoc 2.4, citations, quoted inlines,
5676 and underline emphasis were no longer recognized after certain
5677 symbols, like parentheses (#5099, #5053).
5678 + In pandoc 2.4, a soft break after an abbreviation would be
5679 relocated before it to allow for insertion of a nonbreaking
5680 space after the abbreviation. This behavior is here reverted.
5681 A soft break after an abbreviation will remain, and no nonbreaking
5682 space will be added. Those who care about this issue should take care not
5683 to end lines with an abbreviation, or to insert nonbreaking spaces
5686 * Text.Pandoc.Readers.FB2: Do not throw error for unknown elements in
5687 `<body>` (Alexander Krotov). Some libraries include custom elements
5690 * Text.Pandoc.Readers.HTML
5692 + Allow `tfoot` before body rows (#5079).
5693 + Parse `<small>` as a Span with class "small" (#5080).
5694 + Allow thead containing a row with `td` rather than `th` (#5014).
5696 * Text.Pandoc.Readers.LaTeX
5698 + Cleaned up handling of dimension arguments. Allow decimal points,
5700 + Don't allow arguments for verbatim, etc.
5701 + Allow space before bracketed options.
5702 + Allow optional arguments after `\\` in tables.
5703 + Improve parsing of `\tiny`, `\scriptsize`, etc. Parse as raw,
5704 but know that these font changing commands take no arguments.
5706 * Text.Pandoc.Readers.Muse
5708 + Trim whitespace before parsing grid table cells (Alexander Krotov).
5709 + Add grid tables support (Alexander Krotov).
5711 * Text.Pandoc.Shared
5713 + For bibliography match Div with id `refs`, not class `references`.
5714 This was a mismatch between pandoc's docx, epub, latex, and markdown
5715 writers and the behavior of pandoc-citeproc, which actually looks for a
5716 div with id `refs` rather than one with class `references`.
5717 + Exactly match GitHub's identifier generating algorithm (#5057).
5718 + Add parameter for `Extensions` to `uniqueIdent` and
5719 `inlineListToIdentifier` (#5057). [API change]
5720 This allows these functions to be sensitive to the settings of
5721 `Ext_gfm_auto_identifiers` and `Ext_ascii_identifiers`, and allows us to
5722 use `uniqueIdent` in the CommonMark reader, replacing custom code. It
5723 also means that `gfm_auto_identifiers` can now be used in all formats.
5725 * Text.Pandoc.Writers.AsciiDoc
5727 + Use `.`+ as list markers to support nested ordered lists (#5087).
5728 + Support list number styles (#5089).
5729 + Render Spans using `[#id .class]#contents#` (#5080).
5731 * Text.Pandoc.Writers.CommonMark
5733 + Respect `--ascii` (#5043, quasicomputational).
5734 + Make sure `--ascii` affects quotes, super/subscript.
5736 * Text.Pandoc.Writers.Docx
5738 + Fix bookmarks to headers with long titles (#5091).
5739 Word has a 40 character limit for bookmark names. In addition, bookmarks
5740 must begin with a letter. Since pandoc's auto-generated identifiers may
5741 not respect these constraints, some internal links did not work. With
5742 this change, pandoc uses a bookmark name based on the SHA1 hash of the
5743 identifier when the identifier isn't a legal bookmark name.
5744 + Add bookmarks to code blocks (Nikolay Yakimov).
5745 + Add bookmarks to images (Nikolay Yakimov).
5746 + Refactor common bookmark creation code into a function (Nikolay Yakimov).
5748 * Text.Pandoc.Writers.EPUB: Handle calibre metadata (#5098).
5751 <meta name="calibre:series" content="Classics on War and Politics"/>
5753 are now included from an epub XML metadata file. You can also include
5754 this information in your YAML metadata, like so:
5757 series: Classics on War and Policitics
5759 In addition, ibooks-specific metadata can now be included via an XML
5760 file. (Previously, it could only be included via YAML metadata, see
5763 * Text.Pandoc.Writers.HTML: Use plain `"` instead of `"` outside of
5766 * Text.Pandoc.Writers.ICML: Consolidate adjacent strings, inc. spaces.
5767 This avoids splitting up the output unnecessarily into separate elements.
5769 * Text.Pandoc.Writers.LaTeX: Don't emit `[<+->]` unless beamer output, even
5770 if `writerIncremental` is True (#5072).
5772 * Text.Pandoc.Writers.Muse (Alexander Krotov).
5774 + Output tables as grid tables if they have multi-line cells.
5775 + Indent simple tables only on the top level.
5776 + Output tables with one column as grid tables.
5777 + Add support for `--reference-location`.
5778 + Internal improvements.
5780 * Text.Pandoc.Writers.OpenDocument: Fix list indentation (Nils Carlson,
5781 #5095). This was a regression in pandoc 2.4.
5783 * Text.Pandoc.Writers.RTF: Fix warnings for skipped raw inlines.
5785 * Text.Pandoc.Writers.Texinfo: Add blank line before `@menu` section (#5055).
5787 * Text.Pandoc.XML: in `toHtml5Entities`, prefer shorter entities
5788 when there are several choices for a particular character.
5790 * data/abbreviations
5792 + Add additional abbreviations (Andrew Dunning)
5793 Many of these borrowed from the Chicago Manual of Style 10.42,
5794 'Scholarly abbreviations'.
5798 + Asciidoc template: add :lang: to title header is lang is set in
5801 * pandoc.cabal: Add cabal flag `derive_json_via_th` (Albert Krewinkel)
5802 Disabling the flag will cause derivation of ToJSON and FromJSON
5803 instances via GHC Generics instead of Template Haskell. The flag is
5804 enabled by default, as deriving via Generics can be slow (see #4083).
5808 + Tweaked drop-down lists.
5809 + Put link to site in footer.
5810 + Preselect output format.
5811 + Update on change of in or out format.
5812 + Add man input format.
5816 + Fix outdated description of latex_macros extension.
5817 + Clarified placement of bibliography.
5818 + Added "A note on security."
5819 + Fix note on curly brace syntx for locators.
5820 + Document new explicit syntax for citeproc locators.
5821 + Remove confusing cross-links for some extensions.
5822 + Don't put pandoc in code ticks in heading.
5823 + Document that `--ascii` works for gfm and commonmark too.
5824 + Add `man` to `--from` options.
5826 * doc/customizing-pandoc.md: various improvements (Mauro Bieg).
5829## pandoc 2.4 (2018-11-03)
5833 * New input format `man` (Yan Pashkovsky, John MacFarlane).
5837 * `--ascii` is now implemented in the writers, not in Text.Pandoc.App,
5838 via the new `writerPreferAscii` field in `WriterOptions`.
5839 Now the `write*` functions for Docbook, HTML, ICML, JATS, LaTeX,
5840 Ms, Markdown, and OPML are sensitive to `writerPreferAscii`.
5841 Previously the to-ascii translation was done in Text.Pandoc.App,
5842 and thus not available to those using the writer functions
5845 * `--ascii` now works with Markdown output. HTML5 character reference
5848 * `--ascii` now works with LaTeX output. 100% ASCII output can't be
5849 guaranteed, but the writer will use commands like `\"{a}` and `\l`
5850 whenever possible, to avoid emitting a non-ASCII character.
5852 * For HTML5 output, `--ascii` now uses HTML5 character reference
5853 entities rather than numerical entities.
5855 * Improved detection of format based on extension (in Text.Pandoc.App).
5856 We now ensure that if someone tries to convert a file for a
5857 format that has a pandoc writer but not a reader, it won't just
5858 default to markdown.
5860 * Add viz. to abbreviations file (#5007, Nick Fleisher).
5862 * AsciiDoc writer: always use single-line section headers,
5863 instead of the old underline style (#5038). Previously the single-line
5864 style would be used if `--atx-headers` was specified, but
5865 now it is always used.
5867 * RST writer: Use simple tables when possible (#4750).
5869 * CommonMark (and gfm) writer: Add plain text fallbacks. (#4528,
5870 quasicomputational). Previously, the writer would unconditionally
5871 emit HTML output for subscripts, superscripts, strikeouts (if the
5872 strikeout extension is disabled) and small caps, even with
5873 `raw_html` disabled. Now there are plain-text (and, where
5874 possible, fancy Unicode) fallbacks for all of these corresponding
5875 (mostly) to the Markdown fallbacks, and the HTML output is only
5876 used when `raw_html` is enabled.
5878 * Powerpoint writer: support raw openxml (Jesse Rosenthal, #4976).
5879 This allows raw openxml blocks and inlines to be used in the pptx
5880 writer. Caveats: (1) It's up to the user to write
5881 well-formed openxml. The chances for corruption, especially with
5882 such a brittle format as pptx, is high. (2) Because of
5883 the tricky way that blocks map onto shapes, if you are using
5884 a raw block, it should be the only block on a slide
5885 (otherwise other text might end up overlapping it). (3) The
5886 pptx ooxml namespace abbreviations are different from the
5887 docx ooxml namespaces. Again, it's up to the user to get it
5888 right. Unzipped document and ooxml specification should be
5891 * With `--katex` in HTML formats, do not use the autorenderer (#4946).
5892 We no longer surround formulas with `\(..\)` or `\[..\]`. Instead,
5893 we tell katex to convert the contents of span elements with
5894 class "math". Since math has already been identified, this
5895 avoids wasted time parsing for LaTeX delimiters. Note, however,
5896 that this may yield unexpected results if you have span elements
5897 with class "math" that don't contain LaTeX math.
5898 Also, use latest version of KaTeX by default (0.9.0).
5900 * The man writer now produces ASCII-only output, using groff escapes,
5905 + Add title, author and date to metadata; any remaining
5906 metadata fields are added as `meta:user-defined` tags.
5907 + Implement table caption numbering (#4949, Nils Carlson).
5908 Captioned tables are numbered and labeled with format "Table 1:
5909 caption", where "Table" is replaced by a translation, depending
5910 on the value of `lang` in metadata. Uncaptioned tables are not
5912 + OpenDocument writer: Implement figure numbering in captions (#4944,
5913 Nils Carlson). Figure captions are now numbered 1, 2, 3, ...
5914 The format in the caption is "Figure 1: caption" and so on
5915 (where "Figure" is replaced by a translation, depending on the
5916 value of `lang` in the metadata). Captioned figures are numbered
5917 consecutively and uncaptioned figures are not enumerated. This
5918 is necessary in order for LibreOffice to generate an
5919 Illustration Index (Table of Figures) for included figures.
5921 * RST reader: Pass through fields in unknown directives as div attributes
5922 (#4715). Support `class` and `name` attributes for all directives.
5924 * Org reader: Add partial support for `#+EXCLUDE_TAGS` option. (#4284,
5925 Brian Leung). Headers with the corresponding tags should not
5926 appear in the output.
5928 * Log warnings about missing title attributes now include a
5929 suggestion about how to fix the problem (#4909).
5931 * Lua filter changes (Albert Krewinkel):
5933 + Report traceback when an error occurs. A proper Lua traceback is
5934 added if either loading of a file or execution of a filter
5935 function fails. This should be of help to authors of Lua filters
5936 who need to debug their code.
5938 + Allow access to pandoc state (#5015). Lua filters and custom
5939 writers now have read-only access to most fields of pandoc's
5940 internal state via the global variable `PANDOC_STATE`.
5942 + Push ListAttributes via constructor (Albert Krewinkel).
5943 This ensures that ListAttributes, as present in OrderedList elements,
5944 have additional accessors (viz. `start`, `style`, and `delimiter`).
5946 + Rename ReaderOptions fields, use snake_case. Snake case is used
5947 in most variable names, using camelCase for these fields was an
5948 oversight. A metatable is added to ensure that the old field
5949 names remain functional.
5951 + Iterate over AST element fields when using `pairs`. This makes
5952 it possible to iterate over all ield names of an AST element by
5953 using a generic `for` loop with pairs`:
5955 for field_name, field_content in pairs(element) do
5959 Raw table fields of AST elements should be considered an
5960 implementation detail and might change in the future. Accessing
5961 element properties should always happen through the fields
5962 listed in the Lua filter docs.
5964 Note that the iterator currently excludes the `t`/`tag` field.
5966 + Ensure that MetaList elements behave like Lists. Methods usable
5967 on Lists can also be used on MetaList objects.
5969 + Fix MetaList constructor (Albert Krewinkel). Passing a MetaList
5970 object to the constructor `pandoc.MetaList` now returns the
5971 passed list as a MetaList. This is consistent with the
5972 constructor behavior when passed an (untagged) list.
5974 * Custom writers: Custom writers have access to the global variable
5975 `PANDOC_DOCUMENT`(Albert Krewinkel, #4957). The variable contains
5976 a userdata wrapper around the full pandoc AST and exposes two
5977 fields, `meta` and `blocks`. The field content is only
5978 marshaled on-demand, performance of scripts not accessing the
5979 fields remains unaffected.
5983 * Text.Pandoc.Options: add `writerPreferAscii` to `WriterOptions`.
5985 * Text.Pandoc.Shared:
5987 + Export `splitSentences`. This was previously duplicated in the Man and
5989 + Add `ToString` typeclass (Alexander Krotov).
5991 * New exported module Text.Pandoc.Filter (Albert Krewinkel).
5993 * Text.Pandoc.Parsing
5995 + Generalize `gridTableWith` to any `Char` Stream (Alexander Krotov).
5996 + Generalize `readWithM` from `[Char]` to any `Char` Stream
5997 that is a `ToString` instance (Alexander Krotov).
5999 * New exposed module Text.Pandoc.Filter (Albert Krewinkel).
6001 * Text.Pandoc.XML: add `toHtml5Entities`.
6003 * New exported module Text.Pandoc.Readers.Man (Yan Pashkovsky, John
6006 * Text.Pandoc.Writers.Shared
6008 + Add exported functions `toSuperscript` and
6009 `toSubscript` (quasicomputational, #4528).
6010 + Remove exported functions `metaValueToInlines`,
6011 `metaValueToString`. Add new exported functions
6012 `lookupMetaBool`, `lookupMetaBlocks`, `lookupMetaInlines`,
6013 `lookupMetaString`. Use these whenever possible for uniformity
6014 in writers (Mauro Bieg, #4907). (Note that
6015 removed function `metaValueToInlines` was in previous
6017 + Add `metaValueToString`.
6021 + Expose more useful internals (Albert Krewinkel):
6023 - `runFilterFile` to run a Lua filter from file;
6024 - data type `Global` and its constructors; and
6025 - `setGlobals` to add globals to a Lua environment.
6027 This module also contains `Pushable` and `Peekable` instances
6028 required to get pandoc's data types to and from Lua. Low-level
6029 Lua operation remain hidden in Text.Pandoc.Lua.
6031 + Rename `runPandocLua` to `runLua` (Albert Krewinkel).
6033 + Remove `runLuaFilter`, merging this into Text.Pandoc.Filter.Lua's
6034 `apply` (Albert Krewinkel).
6036### Bug fixes and under-the-hood improvements
6038 * Text.Pandoc.Parsing
6040 + Make `uri` accept any stream with Char tokens (Alexander Krotov).
6041 + Rewrite `uri` without `withRaw` (Alexander Krotov).
6042 + Generalize `parseFromString` and `parseFromString'` to any
6043 streams with Char token (Alexander Krotov)
6044 + Rewrite `nonspaceChar` using `noneOf` (Alexander Krotov)
6046 * Text.Pandoc.Shared: Reimplement `mapLeft` using `Bifunctor.first`
6049 * Text.Pandoc.Pretty: Simplify `Text.Pandoc.Pretty.offset`
6054 + Work around HXT limitation for --syntax-definition with windows
6056 + Always preserve tabs for man format. We need it for tables.
6057 + Split command line parsing code into a separate unexported
6058 module, Text.Pandoc.App.CommandLineOptions (Albert Krewinkel).
6060 * Text.Pandoc.Readers.Roff: new unexported module for tokenizing
6063 * New unexported module Text.Pandoc.RoffChar, provided character
6064 escape tables for roff formats.
6066 * Text.Pandoc.Readers.HTML: Fix `htmlTag` and `isInlineTag` to
6067 accept processing instructions (#3123, regression since 2.0).
6069 * Text.Pandoc.Readers.JATS: Use `foldl'` instead of `maximum` to
6070 account for empty lists (Alexander Krotov).
6072 * Text.Pandoc.Readers.RST: Don't allow single-dash separator in
6073 headerless table (#4382).
6075 * Text.Pandoc.Readers.Org: Parse empty argument array in inline src
6076 blocks (Brian Leung).
6078 * Text.Pandoc.Readers.Vimwiki:
6080 + Get rid of `F`, `runF` and `stateMeta'` in favor of `stateMeta`
6082 + Parse `Text` without converting to `[Char]` (Alexander Krotov).
6084 * Text.Pandoc.Readers.Creole: Parse `Text` without converting to
6085 `[Char]` (Alexander Krotov).
6087 * Text.Pandoc.Readers.LaTeX
6089 + Allow space at end of math after `\` (#5010).
6090 + Add support for `nolinkurl` command (#4992, Brian Leung).
6091 + Simplified type on `doMacros'`.
6092 + Tokenize before pulling tokens, rather than after (#4408). This
6093 has some performance penalty but is more reliable.
6094 + Make macroDef polymorphic and allow in inline context.
6095 Otherwise we can't parse something like `\lowercase{\def\x{Foo}}`.
6096 I have actually seen tex like this in the wild.
6097 + Improved parsing of `\def`, `\let`. We now correctly parse:
6107 + Improve parsing of `\def` argspec.
6108 + Skip `\PackageError` commands (see #4408).
6109 + Fix bugs omitting raw tex (#4527). The default is `-raw_tex`,
6110 so no raw tex should result unless we explicitly say `+raw_tex`.
6111 Previously some raw commands did make it through.
6112 + Moved `isArgTok` to Text.Pandoc.Readers.LaTeX.Parsing.
6113 + Moved `babelLangToBCP`, `polyglossiaLangToBCP` to new module,
6114 Text.Pandoc.Readers.LaTeX.Lang (unexported).
6115 + Simplified accent code using unicode-transforms.
6116 New dependency on unicode-transforms package for normalization.
6117 + Allow verbatim blocks ending with blank lines (#4624).
6118 + Support `breq` math environments: `dmath`, `dgroup`, `darray`.
6119 This collects some of the general-purpose code from the LaTeX reader,
6120 with the aim of making the module smaller.
6122 * Text.Pandoc.Readers.Markdown
6124 + Fix awkward soft break movements before abbreviations (#4635).
6125 + Add updateStrPos in a couple places where needed.
6127 * Text.Pandoc.Readers.Docx: Trigger bold/italic with bCs, iCs
6128 (#4947). These are variants for "complex scripts" like Arabic
6129 and are now treated just like b, i (bold, italic).
6131 * Text.Pandoc.Readers.Muse (Alexander Krotov)
6133 + Try to parse lists before trying to parse table.
6134 This ensures that tables inside lists are parsed correctly.
6135 + Forbid whitespace after opening and before closing markup
6137 + Parse page breaks.
6138 + Simplify `museToPandocTable` to get rid of partial functions.
6139 + Allow footnotes to start with empty line.
6140 + Make sure that the whole text is parsed.
6141 + Allow empty headers. Previously empty headers caused parser to
6142 terminate without parsing the rest of the document.
6143 + Allow examples to be indented with tabs.
6144 + Remove indentation from examples indicated by `{{{` and `}}}`.
6145 + Fix parsing of empty cells.
6146 + Various changes to internals.
6147 + Rewrite some parsers in applicative style.
6148 + Avoid tagsoup dependency.
6149 + Allow table caption to contain `+`.
6151 * Text.Pandoc.Writers.LaTeX
6153 + Add newline if math ends in a comment (#4880). This prevents the
6154 closing delimiter from being swalled up in the comment.
6155 + With `--listings`, don't pass through org-babel attributes (#4889).
6156 + With `--biblatex`, use `\autocite` when possible (#4960).
6157 `\autocites{a1}{a2}{a3}` will not collapse the entries. So, if
6158 we don't have prefixes and suffixes, we use instead
6159 `\autocite{a1,a2,a3}`.
6160 + Fix description lists contining highlighted code (#4662).
6162 * Text.Pandoc.Writers.Man
6164 + Don't wrap `.SH` and `.SS` lines (#5019).
6165 + Avoid unnecessary `.RS`/`.RE` pair in definition lists with
6166 one paragraph definitions.
6167 + Moved common groff functions to Text.Pandoc.Writers.Groff.
6168 * Fix strong/code combination on man (should be `\f[CB]` not
6169 `\f[BC]`, see #4973).
6170 + Man writer: use `\f[R]` instead of `\f[]` to reset font
6171 (Alexander Krotov, #4973).
6172 + Move `splitSentences` to Text.Pandoc.Shared.
6174 * Text.Pandoc.Writers.Docx
6176 + Add framework for custom properties (#3034). So far, we don't
6177 actually write any custom properties, but we have the
6178 infrastructure to add this.
6180 + Handle tables in table cells (#4953). Although this is not
6181 documented in the spec, some versions of Word require a `w:p`
6182 element inside every table cell. Thus, we add one when the
6183 contents of a cell do not already include one (e.g. when a table
6184 cell contains a table).
6186 * Text.Pandoc.Writers.AsciiDoc: Prevent illegal nestings.
6187 Adjust header levels so that n+1 level headers are only
6188 found under n level headers, and the top level is 1.
6190 * Text.Pandoc.Writers.OpenDocument: Improve bullet/numbering
6191 alignment (#4385). This change eliminates the large gap we used
6192 to have between bullet and text, and also ensures that numbers
6193 in numbered lists will be right-aligned.
6195 * Text.Pandoc.Writers.ZimWiki
6197 + Number ordered list items sequentially, rather than always
6199 + Remove extra indentation on lists (#4963).
6201 * Text.Pandoc.Writers.EPUB: Use metadata field `css` instead of
6202 `stylesheet` (Mauro Bieg, #4990).
6204 * Text.Pandoc.Writers.Markdown: Ensure blank between raw block and
6205 normal content (#4629). Otherwise a raw block can prevent a
6206 paragraph from being recognized as such.
6208 * Text.Pandoc.Writers.Ms
6210 + Removed old `escapeBar`. We don't need this now that we use
6212 + Moved common code to Text.Pandoc.Writers.Roff and to
6213 Text.Pandoc.RoffChar.
6214 + Move `splitSentences` to Text.Pandoc.Shared (to avoid duplication
6215 with the man writer).
6217 * Text.Pandoc.Writers.Muse (Alexander Krotov).
6219 + Add support for grid tables.
6220 + Fix Muse writer style.
6221 + Use `length` instead of `realLength` to calculate definition
6222 indentation. Muse parsers don't take character width into
6223 account when calculating indentation.
6224 + Do not insert newline before lists.
6225 + Use lightweight markup after `</em>` tag.
6227 * New unexported module Text.Pandoc.Writers.Roff, providing functions
6228 useful for all roff format writers (man, ms).
6232 + Move globals handling to separate module Text.Pandoc.Lua.Global
6235 + Lua filter internals: push Shared.Element as userdata (Albert
6236 Krewinkel). Hierarchical Elements were pushed to Lua as plain
6237 tables. This is simple, but has the disadvantage that marshaling
6238 is eager: all child elements will be marshaled as part of the
6239 object. Using a Lua userdata object instead allows lazy access
6240 to fields, causing content marshaling just (but also each time)
6241 when a field is accessed. Filters which do not traverse the full
6242 element contents tree become faster as a result.
6244### Default template changes
6248 + Add variable `hyperrefoptions` (#4925, Mathias Walter).
6249 + Add variable `romanfont`, `romanfontoptions` (#4665, OvidiusCicero).
6251 * AsciiDoc template: use single-line style for title.
6253 * revealjs template: Fix typo in the socket.io javascript plugin (#5006,
6256 * Text.Pandoc.Lua.Util: add missing docstring to `defineHowTo`
6259 * data/pandoc.lua: add datatype ListAttributes (Albert Krewinkel)
6261 * data/sample.lua: replace custom pipe function with pandoc.utils.pipe
6264### Documentation improvements
6268 + Add chromeos install instructions (#4958) (Evan Pratten).
6269 + Add note about TinyTeX.
6273 + Change `groff` -> `roff`.
6274 + Implement `--ascii` for Markdown writer.
6275 + Clarify LaTeX image dimensions output (Mauro Bieg).
6277 * doc/customizing-pandoc.md: added skeleton (Mauro Bieg, #3288).
6279 * doc/getting-started.md: Added title to test1.md to avoid warning.
6281 * doc/lua-filters.md: merge type references into main document,
6282 fix description of Code.text (Albert Krewinkel).
6284### Build infrastructure improvements
6288 + Makefile: added quick-cabal, full-cabal targets.
6289 + Make .msi download targets insensitive to order of appveyor builds.
6291 * Update benchmarks for ghc 8.6.1.
6295 + Enable more compiler warnings (Albert Krewinkel).
6296 + Make base lower bound 4.8.
6297 + Bump upper bound for QuickCheck.
6298 + Bump upper bound for binary.
6299 + Updated version bounds for containers and haddock-library (#4974).
6300 + Added docx/docPropos/custom.xml to cabal data-files.
6301 + Require skylighting 0.7.4 (#4920).
6302 + New dependency on unicode-transforms package for normalization.
6304 * Improved .travis.yml testing and test with GHC 8.6.1 (Albert Krewinkel).
6306 * Added `tools/changelog-helper.sh`.
6308 * Added test/grofftest.sh for testing the man reader on real man pages.
6310## pandoc 2.3.1 (2018-09-28)
6314 + Parse RST inlines containing newlines (#4912, Francesco Occhipinti).
6315 This eliminates a regression introduced after pandoc 2.1.1, which
6316 caused inline constructions containing newlines not to be recognized.
6317 + Fix bug with internal link targets (#4919). They were gobbling up
6318 indented content underneath.
6320 * Markdown reader: distinguish autolinks in the AST. With this change,
6321 autolinks are parsed as Links with the `uri` class. (The same is true
6322 for bare links, if the `autolink_bare_uris` extension is enabled.)
6323 Email autolinks are parsed as Links with the `email` class. This
6324 allows the distinction to be represented in the AST.
6328 + Force inline code blocks to honor export options (Brian Leung).
6329 + Parse empty argument array in inline src blocks (Brian Leung).
6331 * Muse reader (Alexander Krotov):
6333 + Added additional tests.
6334 + Do not allow code markup to be followed by digit.
6335 + Remove heading level limit.
6336 + Simplify `<literal>` tag parsers
6337 + Parse Text instead of String. Benchmark shows 7% improvement.
6338 + Get rid of HTML parser dependency.
6339 + Various code improvements.
6341 * ConTeXt writer: change `\` to `/` in Windows image paths (#4918).
6342 We do this in the LaTeX writer, and it avoids problems.
6343 Note that `/` works as a LaTeX path separator on Windows.
6347 + Add support for multiprenote and multipostnote arguments
6348 with `--biblatex` (Brian Leung, #4930). The multiprenotes occur before
6349 the first prefix of a multicite, and the multipostnotes follow the
6351 + Fix a use of `last` that might take empty list. If you ran with
6352 `--biblatex` and have an empty document (metadata but no blocks),
6353 pandoc would previously raise an error because of the use of
6354 `last` on an empty list.
6356 * RTF writer: Fix build failure with ghc-8.6.1 caused by missing
6357 MonadFail instance (Jonas Scholl).
6359 * ODT Writer: Improve table header row style handling (Nils Carlson).
6360 This changes the way styles for cells in the header row
6361 and normal rows are handled in ODT tables.
6362 Previously a new (but identical) style was generated for
6363 every table, specifying the style of the cells within the table.
6364 After this change there are two style definitions for table cells,
6365 one for the cells in the header row, one for all other cells.
6366 This doesn't change the actual styles, but makes post-processing
6367 changes to the table styles much simpler as it is no longer
6368 necessary to introduce new styles for header rows and there are
6369 now only two styles where there was previously one per table.
6373 + Don't add `uri` class to presumed autolinks. Formerly the `uri`
6374 class was added to autolinks by the HTML writer, but it had to
6375 guess what was an autolink and could not distinguish
6376 `[http://example.com](http://example.com)` from
6377 `<http://example.com>`. It also incorrectly recognized
6378 `[pandoc](pandoc)` as an autolink. Now the HTML writer
6379 simply passes through the `uri` attribute if it is present,
6380 but does not add anything.
6381 + Avoid adding extra section nestings for revealjs.
6382 Previously revealjs title slides at level (slidelevel - 1)
6383 were nested under an extra section element, even when
6384 the section contained no additional (vertical) content.
6385 That caused problems for some transition effects.
6386 + Omit unknown attributes in EPUB2 output. For example,
6387 `epub:type` attributes should not be passed through,
6388 or the epub produced will not validate.
6390 * JATS writer: remove 'role' attribute on 'bold' and 'sc' elements (#4937).
6391 The JATS spec does not allow these.
6393 * Textile writer: don't represent `uri` class explicitly
6394 for autolinks (#4913).
6396 * Lua filters (Albert Krewinkel):
6398 + Cleanup filter execution code.
6399 + Better error on test failure.
6401 * HTML, Muse reader tests: reduce time taken by round-trip test.
6403 * Added cabal.project.
6405 * MANUAL: `epub:type` is only useful for epub3 (Maura Bieg).
6407 * Use hslua v1.0.0 (Albert Krewinkel).
6409 * Fix `translations/ru` to use modern Russian orthography
6412 * Build Windows binary using ghc 8.6.1 and cabal new-build. This
6413 fixes issues with segfaults in the 32-bit Windows binaries (#4283).
6415## pandoc 2.3 (2018-09-16)
6417 * Add `--metadata-file` option (Mauro Bieg, #1960), which allows
6418 users to specify metadata in a YAML file, regardless of the
6419 input format (#1960).
6421 * Text.Pandoc.Writers.Shared: export `isDisplayMath` (API change).
6423 * Text.Pandoc.Readers.Markdown: export `yamlToMeta` (API change,
6426 * Text.Pandoc.Readers.LaTeX.Types:
6428 + New type `ArgSpec` (API change).
6429 + Second parameter of `Macro` constructor is now `[ArgSpec]` instead of
6434 + Use `tex` instead of `latex` for raw tex-ish content. We can't
6435 always tell if it's LaTeX, ConTeXt, or plain TeX. Better just to
6436 use `tex` always. Note that if `context` or `latex` specifically
6437 is desired, you can still force that in a markdown document by
6438 using the raw attribute. Note that this change may affect some
6439 filters, if they assume that raw tex parsed by the Markdown reader
6440 will be RawBlock (Format `latex`). In most cases it should be
6441 trivial to modify the filters to accept `tex` as well.
6442 + Refactor and reorganize YAML code (Mauro Bieg).
6443 + Make `example_lists` work for interrupted lists
6444 without `startnum` (#4908).
6448 + Parse `<script type="math/tex` tags as math (#4877).
6449 These are used by MathJax in some configurations.
6450 + Extract spaces inside links instead of trimming them
6451 (Alexander Krotov, #4845).
6452 + Added round-trip tests (Alexander Krotov).
6453 + Make parsing sensitive to the `raw_tex` extension (#1126). This now
6454 allows raw LaTeX environments, `\ref`, and `\eqref` to be parsed
6455 (which is helpful for translation HTML documents using MathJaX).
6457 * Org reader (Albert Krewinkel):
6459 + Respect export option `p` for planning info. Inclusion of planning info
6460 (`*DEADLINE*`, `*SCHEDULED*`, and `*CLOSED*`) can be controlled via the
6461 `p` export option: setting the option to `t` will add all planning
6462 information in a *Plain* block below the respective headline.
6463 + Org reader internals: don't rely on RecordWildCards and ViewPatterns
6465 + Strip planning info from output. Planning info is parsed, but not
6466 included in the output (as is the default with Emacs Org-mode, #4867).
6470 + Handle parameter patterns for `\def` (#4768, #4771).
6471 For example: `\def\foo#1[#2]{#1 and #2}`.
6472 + Allow `%` characters in URLs. This affects `\href` and `\url` (#4832).
6473 + Fixed parsing of `\texorpdfstring`. We were returning the wrong
6474 argument as the content.
6475 + Support `blockcquote`, `foreignblockquote`, `foreigncblockquote`,
6476 `hyphenblockquote`, `hyphencblockquote`, `enquote*`, `foreignquote`,
6477 `hyphenquote` from `csquotes` (#4848). Note that `foreignquote`
6478 will be parsed as a regular Quoted inline (not using the quotes
6479 appropriate to the foreign language).
6480 + Support more text-mode accents (#4652). Add support for `\|`, `\b`, `\G`,
6481 `\h`, `\d`, `\f`, `\r`, `\t`, `\U`, `\i`, `\j`, `\newtie`,
6482 `\textcircled`. Also fall back to combining characters when composed
6483 characters are not available.
6484 + Resolve `\ref` for figure numbers.
6485 + Support `mintinline` (#4365, Marc Schreiber).
6486 + Fix siunitx unit commands so they are only recognized in
6487 siunitx contexts (#4842). For example, `\l` outside of an
6488 siunitx context should be l-slash, not l (for liter).
6489 + Fix double `unnumbered` class (#4838). The `unnumbered` class
6490 was being included twice for starred sections.
6492 * RST reader: Don't skip link definitions after comments (#4860).
6494 * Muse reader (Alexander Krotov):
6496 + Close the `</quote>` in indented tag test.
6497 There is already a separate test for unclosed `</quote>`.
6498 + Autonumber sections in the correct order. Parsing now stops
6499 at each section header to ensure the header is registered before
6500 parsing of the next section starts.
6501 + Move duplicate code into `headingStart` function.
6502 + Allow newline after opening `*` or `**`.
6503 + Don't allow digits after closing marker in lightweight markup
6504 This change makes reader more compatible with Emacs Muse.
6505 + Parse `<verse>` tag in one pass instead of using
6506 `parseFromString`. This change makes it possible to have
6507 verbatim `</verse>` tag inside verse.
6509 * ODT reader: deal gracefully with missing `<office:font-face-decls/>`
6510 (#4336). This allows pandoc to parse ODT document produced by KDE's
6513 * Muse writer (Alexander Krotov):
6515 + Output headers without asterisks if not on the top level.
6516 + Never wrap definition list terms.
6517 + Set `envInsideBlock = True` when rendering notes.
6518 + Use `""` instead of `[]` for empty String.
6519 + Check for whitespace in the beginning and end of Str's.
6520 + Escape `-`, `;` and `>` in the beginning of strings.
6521 + Escape list markers in the beginning of notes.
6522 + Normalize inline list before testing if tags should be used.
6523 + Use tags instead of lightweight markup for empty strings.
6524 + Use lightweight markup when possible.
6525 + Escape empty strings. This guarantees that `conditionalEscapeString`
6526 never returns empty string.
6527 + Wrap conditionalEscapeString result into `Muse` type.
6528 This removes the need to pass `envInsideLinkDescription` to it.
6529 + Separate `shouldEscapeString` function.
6530 + Simplify inline list rendering.
6531 + Replace newlines in strings with spaces.
6535 + Add MetaString case for abstract, subtitle (#4900, Mauro Bieg).
6536 + Properly handle display math in spans (#4826). This isn't a
6537 complete solution, since other nestings of display math may still
6538 cause problems, but it should work for what is by far the most
6543 + Always output `<dt>` element, even if it is empty (#4883,
6545 + Don't prefix `epub:` attributes with `data-`.
6547 * Org writer: Don't escape literal `_`, `^` (#4882). Org doesn't
6548 recognize these escapes.
6550 * ODT writer: Fix percentage image scaling (#4881, Nils Carlson).
6551 Image scaling was broken when a width was set to a percentage.
6553 * EPUB writer: set `epub:type` on body element in each chapter,
6554 depending on the `epub:type` of the first section (#4823). This
6555 only affects epub3. See
6556 http://www.idpf.org/epub/profiles/edu/structure/#h.l0bzsloklt10
6558 * FB2 writer: put `coverpage` element between title and date rather than in
6559 `document-info` element (#4854).
6561 * Markdown writer: Escape `~` if strikeout extension enabled (#4840).
6563 * Haddock writer: Use proper format for latex math in haddock (#4571, Joe
6564 Hermaszewski). Inline math in `\(..\)`, display math in `\[..\]`,
6565 tex is now used. Previously we'd "fake it with unicode" and fall
6566 back to tex when that didn't work. But newer haddock versions
6571 + Ensure that title element is always present, even if empty (#4839).
6572 + Put author tags in the template, rather than adding them in
6575 * LaTeX writer/template: be sensitive to `filecolor` variable (#4822).
6576 `linkcolor` only affects internal links, and `urlcolor` only
6577 affects linked URLs. For external links, the option to use is
6580 * ConTeXt writer: output raw `tex` blocks as well as `context` (#969).
6584 + Use `.. raw:: latex` for `tex` content.
6585 + Use `.. container` for generic Divs, instead of raw HTML.
6586 + Render Divs with admonition classes as admonitions (#4833).
6587 Also omit Div with class `admonition-title`. These are generated
6588 by the RST reader and should be omitted on round-trip.
6590 * Text.Pandoc.PDF: fix message printed when rsvg-convert is not available
6591 (#4855, Antonio Terceiro).
6593 * HTML5 template: add the `title-block-header` identifier to the
6594 `header` element, to make it easier to style precisely (#4767,
6597 * OpenDocument template: Remove unnecessary indenting of TOC title (#4798,
6598 José de Mattos Neto).
6600 * latex template: Add support for $toc-title$ to LaTeX (and PDF)
6601 (#4853, Wandmalfarbe).
6603 * TEI template: improve `publicationStmt`. Add support for
6604 `publisher`, `address`, `pubPlace`, and `date` variables.
6606 * beamer template: Support "toc-title" (#4835, Cyril Roelandt).
6608 * Text.Pandoc.Extensions: Fix haddock on `Ext_footnotes` (Chris Martin).
6610 * Lua: cleanup Lua utils, remove unused functions (Albert Krewinkel).
6614 + Clarify that `--biblatex/--natbib` don't work directly for PDF
6616 + Document `epub:type` attribute (Mauro Bieg, #4901)
6617 + Clarify when `--resource-path` has an effect.
6618 + More detail on customization in syntax highlighting section.
6619 + Document encoding issue with `--listings` (#4871, Damien Clochard).
6620 + Remove docs on removed `--katex-stylesheet` (Mauro Bieg, #4862).
6621 + Use https for context wiki links (#4910).
6625 + Link to lua-filters repository (#4874).
6626 + Fix mistake in REPL instructions for stack. (#4849, Brian Leung).
6628 * lua-filters.md: add links to filters, and to lua-filters repository
6633 + Indicate that cabal >= 2.0 is needed.
6634 + Added chocolatey installation method (#4844, Miodrag Milić).
6636 * Travis: exclude round-trip tests, except for nightly test which can fail.
6638 * Use latest texmath, pandoc-citeproc.
6640 * Use a patched version of foundation until
6641 https://github.com/haskell-foundation/foundation/pull/503
6644 * Clean up appveyor build and Windows package creation.
6645 We now use 64-bit stack and ghc 8.4.3, lts-12 for the 64-bit build. The
6646 WiX-based msi is now 64-bit for 64-bit builds (fixing #4795).
6648 * Remove obsolete RELEASE-CHECKLIST.md.
6650 * Added additional compiler warnings in Makefile and CI builds.
6652## pandoc 2.2.3.2 (2018-08-07)
6654 * Markdown reader: Properly handle boolean values in YAML metadata (#4819).
6655 This fixes a regression in 2.2.3, which cause boolean values to
6656 be parsed as MetaInlines instead of MetaBool.
6658 We here record another undocumented (but desirable) change in 2.2.3:
6659 numerical metadata fields are now parsed as MetaInlines rather than
6662## pandoc 2.2.3.1 (2018-08-06)
6664 * Markdown reader: Fix parsing of embedded mappings in YAML metadata
6665 (#4817). This fixes a regression in 2.2.3 which caused embedded
6666 mappings (e.g. mappings in sequences) not to work in YAML metadata.
6668## pandoc 2.2.3 (2018-08-05)
6670 * RST reader: improve parsing of inline interpreted text roles (#4811).
6672 + Use a Span with class "title-reference" for the default
6673 title-reference role.
6674 + Use `B.text` to split up contents into `Space`s, `SoftBreak`s,
6675 and `Str`s for `title-reference`.
6676 + Use Code with class "interpreted-text" instead of Span and Str for
6677 unknown roles. (The RST writer has also been modified to round-trip
6679 + Disallow blank lines in interpreted text.
6680 + Backslash-escape now works in interpreted text.
6681 + Backticks followed by alphanumerics no longer end interpreted text.
6682 + Remove support for nested inlines (Francesco Occhipinti).
6683 RST does not allow nested emphasis, links, or other inline
6684 constructs. This fixes several bugs (#4581, #4561, #4792).
6686 * Org reader: fix parsers relying on `parseFromString` (#4784, Albert
6687 Krewinkel). Emphasis was not parsed when it followed directly after
6688 some block types (e.g., lists).
6690 * Markdown reader: Allow unquoted numbers and booleans as YAML mapping
6691 keys. Previously in 2.2.2 you could not do
6703 With this change, both forms work.
6705 * DocBook reader: metadata handling improvements.
6706 Now we properly parse title and subtitle elements that are direct
6707 children of book and article (as well as children of bookinfo,
6708 articleinfo, or info). We also now use the `subtitle` metadata
6709 field for subtitles, rather than tacking the subtitle on to the
6714 + Allow images to be directly nested within links (#4810, Francesco
6716 + Use `titleblock` instead of `title` variable for title block (#4803,
6717 Francesco Occhipinti). `titleblock` contains a properly formatted
6718 title and subtitle (using top-level headers). `title` and
6719 `subtitle` variables are still available and just contain the
6720 title and subtitle text. Note that this change will require an
6721 update to custom rst templates.
6722 + Render Code with class "interpreted-text" as interpreted text role.
6724 * MediaWiki writer: Avoid extra blank line in tables with empty cells
6725 (#4794). Note that the old output is semantically identical, but the
6726 new output looks better.
6728 * Lua Utils module: add function `blocks_to_inlines` (#4799, Albert
6729 Krewinkel). Exposes a function converting which flattenes a list of
6730 blocks into a list of inlines. An example use case would be the
6731 conversion of Note elements into other inlines.
6733 * RST template: use `titleblock` instead of `title`. Users of
6734 custom RST templates will want to update this.
6736 * LaTeX template: Moved some beamer code in default.latex template.
6737 This change allows beamer themes to change the template and font (as
6738 Metropolis does) (#4450).
6740 * Better error message on `-t pdf -o out.pdf` (#1155, Mauro Bieg).
6742 * Added test case for #4669 to repository.
6744 * INSTALL.md: Fix broken link for cabal-install (#4806, ChanHoHo).
6748 + Add beamer info for slide backgrounds (#4802, John Muccigrosso).
6749 + Clarify when `csquotes` is used in LaTeX writer (#4514).
6750 + Add `commonmark` to list of output formats where `raw_tex` has an
6754## pandoc 2.2.2.1 (2018-07-19)
6756 * Fix regression finding templates in user data directory (#4777).
6757 Under version 2.2.1 and prior pandoc found latex templates in the
6758 templates directory under the data directory, but this broke in 2.2.2.
6760 * Fix for bug in parsing `\input` in `rawLaTeXBlock` and
6761 `rawLaTeXInline` (#4781). (This primarily affects the markdown
6762 reader, and other readers that accept raw tex.)
6763 Starting in 2.2.2, everything after an `\input` (or `\include`)
6764 in a markdown file would be parsed as raw LaTeX.
6768 + Clarify `gfm` vs `markdown_github` (#4783, Mauro Bieg).
6769 * Use `keywords` instead of `tags` in YAML metadata example (#4779).
6770 Unlike `tags`, `keywords` is used in some of the writers
6771 and default templates.
6773 * Add missing `rollingLinks` option to revealjs template (#4778,
6776## pandoc 2.2.2 (2018-07-16)
6778 * Use HsYAML instead of yaml for translations and YAML metadata (#4747).
6779 yaml wraps a C library; HsYAML is pure Haskell. Advances #4535.
6781 Note: HsYAML implements YAML 1.2, in which the valid true
6782 values are `true`, `True`, `TRUE`. This means a change in
6783 the semantics of YAML metadata that could affect users:
6784 `y`, `yes`, and `on` no longer count as true values.
6786 * Fix regression: make `--pdf-engine` work with full paths (#4681, Mauro
6789 * CommonMark reader: Handle `ascii_identifiers` extension (#4742,
6790 Anders Waldenborg). Non-ascii characters were not stripped from
6791 identifiers even if the `ascii_identifiers` extension was
6792 enabled (which is is by default for gfm).
6794 * TikiWiki reader: Improve list parsing (#4722, Mauro Bieg).
6795 Remove trailing Space from list items. Parse lists that have no space
6800 + Treat `lilypond` as a verbatim environment (#4725).
6801 + Parse figure label into Image id (#4700, Mauro Bieg).
6802 + Beamer: Allow "noframenumbering" option (#4696, Raymond Ehlers).
6803 + Allow spaces around `\graphicspath` arguments (#4698).
6804 + Handle includes without surrounding blanklines (#4553).
6805 In addition, `\input` can now be used in an inline context,
6806 e.g. to provide part of a paragraph, as it can in LaTeX.
6807 + In `rawLaTeXBlock`, handle macros that resolve to a
6808 `\begin` or `\end` (#4667).
6809 + In `rawLaTeXBlock`, don't expand macros in macro definitions (#4653).
6810 Note that this only affected LaTeX in markdown.
6811 + Tighten up reading of beamer overlay specifications (#4669).
6812 Ideally we'd turn these on only when reading beamer, but currently
6813 beamer is not distinguished from latex as an input format.
6814 This commit also activates parsing of overlay specifications
6815 after commands in general (e.g. `\item`), since they can occur
6816 in many contexts in beamer.
6817 + Parse more siunitx unit commands (#4296, #4773).
6818 + Be more forgiving in key/value option parsing (#4761).
6822 + Allow empty code spans, e.g. `` ` ` ``.
6823 + Emojis are now wrapped in Spans with class `emoji` and
6824 attribute `data-emoji` (Anders Waldenborg, #4743).
6825 This allows the writer to handle them in a special way
6826 (e.g. using a special font, or just rendering the
6829 * Muse reader (Alexander Krotov, except where indicated):
6831 + Get rid of non-exhaustive pattern match warning (Mauro Bieg).
6832 + Add support for floating images.
6833 + Add support for images with specified width.
6834 + Parse image URLs without "guard" and "takeExtension".
6835 + Split link and image parsing into separate functions.
6836 + Parse links starting with "URL:" explicitly instead of trying to strip
6837 "URL:" prefix after parsing.
6839 * Texinfo writer: Use `@sup` and `@sub` instead of custom macros (#4728,
6842 * Markdown writer: Preserve `implicit_figures` with attributes, even if
6843 `implicit_attributes` is not set, by rendering in raw HTML (#4677).
6845 * Markdown and commonmark/github writers now respect the `emoji`
6846 extension. So, `-f markdown+emoji -t markdown+emoji` now leaves
6847 `:smile:` as `:smile:` rather than converting it to a smile
6850 * Docx writer: Be sensitive to `toc` in YAML metadata (#4645).
6852 * ODT/OpenDocument writer: Make internal links work (#4358).
6853 This adds proper bookmarks to the headers with non-null IDs.
6855 * EPUB writer: Properly escape pagetitle. Previously we weren't
6856 escaping `&` and other XML characters in the pagetitle, so a title
6857 containing a `&` would be invalid.
6859 * AsciiDoc Writer: Eescape square brackets at start of line (#4545,
6864 + Don't treat 'example' as a syntax name (#4748).
6865 This fixes conversions from org with example blocks.
6866 + Support `--number-sections` via the `section-numbering`
6867 directive in standalone output.
6869 * reveal.js writer and template: reuse mathjax URL
6870 provided by the argument to `--mathjax` or the normal pandoc default,
6871 rather than a hard-coded one in the template (#4701).
6875 + Properly handle footnotes in table captions (#4683).
6876 Refactored code from figure captions to use in both places.
6877 + In beamer output, fix single digit column percentage (#4690, Mauro
6880 * FB2 writer (Alexander Krotov):
6882 + Convert Plain to Para in annotation (#2424).
6883 + Fix order of items in title-info (#2424).
6885 * Custom writer: fix error message on script failure (Albert Krewinkel).
6886 Error messages produced by Lua were not displayed by Pandoc.
6888 * Text.Pandoc.Emoji now exports `emojiToInline`, which returns a Span
6889 inline containing the emoji character and some attributes with metadata
6890 (class `emoji`, attribute `data-emoji` with emoji name). (API change,
6891 Anders Waldenborg, #4743).
6895 + Revert fix for #4484 (only compress images on last run, #4755).
6896 This will mean some increase in the time it takes to
6897 produce an image-heavy PDF with xelatex, but it will
6898 make tables of contents correct, which is more important.
6899 + Fix logic error in `runTeXProgram`. We were running the tex program
6900 one more time than requested. This should speed up PDF production.
6902 * Allow `--template` to take a URL as argument.
6904 * Text.Pandoc.Highlighting: Add missing re-export of `breezeDark`
6905 highlighting style (#4687, Adrian Sieber, API change).
6907 * Clarify macOS install in INSTALL.md (#4661). Make the binary package
6908 installer the recommended method, and note that on some older versions of
6909 macOS, homebrew installs from source and takes a lot of disk space
6914 + Clarify EPUB linked media (#4756, Mauro Bieg)
6915 + Update manual for "true" YAML values. Now that we're using HsYAML and
6916 YAML 1.2, the valid true values are `true`, `True`, `TRUE`. NOTE!
6917 `y`, `yes`, `on` no longer count as true values.
6918 + Document `-F` as alias for `--filter` (thanks to Gandalf Saxe).
6919 + Update manual on how math is rendered in LaTeX.
6920 + Add proxy description (#4131, Mauro Bieg).
6921 + Clarify that `--toc` requires `--standalone` (#4703).
6922 + Update citation styles link (#4699, wiefling).
6924 * In API docs, clarify how `Ext_east_asian_line_breaks` extension works
6925 (kaizshang91). Note that it will not take effect when readers/writers
6926 are called as libraries (#4674).
6928 * Improved translations/fr (#4766, lux-lth).
6930 * Removed inadvertently added `.orig` files from repository (#4648).
6932 * Remove `network-uri` flag and use 'Network.Socket'.
6933 This removes a compiler warning. There is no need for the old
6934 `network-uri` flag, since network 2.6 was released in 2014.
6936 * Add stack.lts10.yaml, stack.lts11.yaml. use lts-12 in stack.yaml.
6938 * Bump upper bounds for dependent packages.
6940 * Exclude foundation 0.0.21 for ghc 7.10. Otherwise cabal gets
6941 confused because of the way ghc 7.10 is excluded in foundation's
6942 cabal file. This can be removed when haskell-foundation/foundation#500
6945 * Require cabal-version >= 2.0. This is needed for haddock-library.
6950## pandoc 2.2.1 (2018-05-10)
6952 * Restored and undeprecated gladtex for HTML math (#4607).
6954 + Added `GladTeX` constructor to `Text.Pandoc.Options.HTMLMathMethod`
6955 [API change, reverts removal in v2.2]
6956 + Restored and undeprecated `--gladtex` option, removed in v2.2.
6958 * LaTeX reader: Handle `$` in `/text{..}` inside math (#4576).
6960 * Org reader: Fix image filename recognition (Albert Krewinkel).
6961 Use a function from the filepath library to check whether a string is a
6962 valid file name. The custom validity checker that was used before gave
6963 wrong results (e.g. for absolute file paths on Windows,
6964 kawabata/ox-pandoc#52).
6966 * FB2 reader: Replace some errors with warnings (Alexander Krotov).
6970 + Strip links from headers when creating TOC (#4340).
6971 Otherwise the TOC entries will not link to the sections.
6972 + Fix regression with tex math environments in HTML + MathJax (#4639).
6974 * Muse writer (Alexander Krotov): Add support for left-align and
6975 right-align classes (#4542).
6977 * Docx writer: Support underline (#4633).
6979 * Text.Pandoc.Parsing: Lookahead for non-whitespace after
6980 `singleQuoteStart` and `doubleQuoteStart` (#4637).
6982 * `test-pandoc-utils.lua`: more robust testing on both windows
6983 and \*nix. Previously the pipe tests were only run if
6984 `\bin/false` and `/bin/sed` were present, which they aren't
6985 in default MacOS and Windows systems. Fixed by using `tr`
6986 and `false`, which should always be in the path on a \*nix
6987 system, and `find` and `echo` for Windows.
6989 * Text.Pandoc.Shared: add `uriPathToPath`.
6990 This adjusts the path from a file: URI in a way that is sensitive
6991 to Windows/Linux differences. Thus, on Windows,
6992 `/c:/foo` gets interpreted as `c:/foo`, but on Linux,
6993 `/c:/foo` gets interpreted as `/c:/foo`. See #4613.
6995 * Use `uriPathToPath` with file: URIs (#4613).
6997 * Revert piping HTML to pdf-engine (Mauro Bieg, #4413). Use a temp
7000 * Text.Pandoc.Class: Catch IO errors when writing media files
7001 and issue a warning, rather than an error (Francesco Occhipinti, #4559).
7003 * Don't lowercase custom writer filename (Alexander Krotov, #4610).
7005 * MANUAL (Mauro Bieg):
7007 + Clarify truthiness in template variables (#2281).
7008 + Clarify pipe table width calculation (#4520).
7010 * ConTeXt template: New Greek fallback typeface (Pablo Rodríguez, #4405).
7011 CMU Serif gives better typographic results than the previous
7012 Greek fallback DejaVu Serif.
7014 * Make HTML template polyglot (#4606, OvidiusCicero), by making
7015 `<link rel="stylesheet" href="$css$">` self-closing.
7017 * Use texmath 0.11, allowing better translation of non-ASCII
7018 characters in math (#4642).
7021## pandoc 2.2 (2018-04-27)
7023 * New input format: `fb2` (FictionBook2) (Alexander Krotov).
7025 * Make `--ascii` work for all XML formats (ICML, OPML, JATS,...),
7026 and for `ms` and `man`.
7028 * Remove deprecated `--latexmathml`, `--gladtex`, `--mimetex`, `--jsmath`, `-m`,
7029 `--asciimathml` options.
7031 * New module Text.Pandoc.Readers.FB2, exporting readFB2 (Alexander
7032 Krotov, API change).
7036 + Allow empty key-value attributes, like `title=""` (#2944).
7037 + Handle table w/o following blank line in fenced div (#4560).
7038 + Remove "fallback" for `doubleQuote` parser. Previously the
7039 parser tried to be efficient -- if no end double quote was found,
7040 it would just return the contents. But this could backfire in a
7041 case `**this should "be bold**`, since the fallback would return
7042 the content `"be bold**` and the closing boldface delimiter
7043 would never be encountered.
7044 + Improve computation of the relative width of the last column in a
7045 multiline table, so we can round-trip tables without constantly
7046 shrinking the last column.
7050 + Fix images with space in file path (#4344).
7054 + Properly resolve section numbers with `\ref` and chapters (#4529).
7055 + Parse sloppypar environment (#4517, Marc Schreiber).
7056 + Improve handling of raw LaTeX (for markdown etc.) (#4589, #4594).
7057 Previously there were some bugs in how macros were handled.
7058 + Support `\MakeUppercase`, `\MakeLowercase`, `\uppercase`,
7059 `\lowercase`, and also `\MakeTextUppercase` and
7060 `\MakeTextLowercase` from textcase (#4959).
7064 + Fixed tables with no body rows (#4513).
7065 Previously these raised an exception.
7069 + Improve table parsing (#4508). This fixes detection of table
7070 attributes and also handles `!` characters in cells.
7074 + Properly handle title in `section` element (#4526).
7075 Previously we just got `section_title` for `section` (though `sect1`,
7076 `sect2`, etc. were handled properly).
7077 + Read tex math as output by asciidoctor (#4569, Joe Hermaszewski).
7081 + Combine adjacent CodeBlocks with the same attributes into
7082 a single CodeBlock. This prevents a multiline codeblock in
7083 Word from being read as different paragraphs.
7087 + Allow < 3 spaces indent under directives (#4579).
7088 + Fix anonymous redirects with backticks (#4598).
7090 * Muse reader (Alexander Krotov):
7092 + Add support for Text::Amuse multiline headings.
7093 + Add `<math>` tag support.
7094 + Add support for `<biblio>` and `<play>` tags.
7095 + Allow links to have empty descriptions.
7096 + Require block `<literal>` tags to be on separate lines.
7097 + Allow `-` in anchors.
7098 + Allow verse to be indented.
7099 + Allow nested footnotes.
7100 + Internal improvements.
7102 * Muse writer (Alexander Krotov):
7104 + Escape `>` only at the beginning of a line.
7105 + Escape `]` in image title.
7106 + Escape `]` brackets in URLs as `%5D`.
7107 + Only escape brackets when necessary.
7108 + Escape ordered list markers.
7109 + Do not escape list markers unless preceded by space.
7110 + Escape strings starting with space.
7111 + Escape semicolons and markers after line break.
7112 + Escape `;` to avoid accidental comments.
7113 + Don't break headers, line blocks and tables with line breaks.
7114 + Correctly output empty headings.
7115 + Escape horizontal rule only if at the beginning of the line.
7116 + Escape definition list terms starting with list markers.
7117 + Place header IDs before header.
7118 + Improve span writing.
7119 + Do not join Spans in normalization.
7120 + Don't align ordered list items.
7121 + Remove key-value pairs from attributes before normalization.
7122 + Enable `--wrap=preserve` for all tests by default.
7123 + Reduced `<verbatim>` tags in output.
7128 + Use more consistent indentation (#4563). Previously we
7129 used an odd mix of 3- and 4-space indentation. Now we use 3-space
7130 indentation, except for ordered lists, where indentation must
7131 depend on the width of the list marker.
7132 + Flatten nested inlines (#4368, Francesco Occhipinti).
7133 Nested inlines are not valid RST syntax, so we flatten them following
7134 some readability criteria discussed in #4368.
7138 + Ensure that `pagetitle` is always set, even when structured titles
7139 are used. This prevents spurious warnings about empty title
7142 * FB2 writer (Alexander Krotov):
7144 + Output links inline instead of producing notes. Previously all links
7145 were turned into footnotes with unclickable URLs inside.
7146 + Allow emphasis and notes in titles.
7147 + Don't intersperse paragraph with empty lines.
7148 + Convert metadata value `abstract` to book annotation.
7149 + Use `<empty-line />` for `HorizontalRule` rather than `LineBreak`.
7150 FB2 does not have a way to represent line breaks inside paragraphs;
7151 previously we used `<empty-line />` elements, but these are not allowed
7154 * Powerpoint writer (Jesse Rosenthal):
7156 + Handle Quoted Inlines (#4532).
7157 + Simplify code with `ParseXml`.
7158 + Allow fallback options when looking for placeholder type.
7159 + Check reference-doc for all layouts.
7160 + Simplify speaker notes logic.
7161 + Change notes state to a simpler per-slide value.
7162 + Remove `Maybe` from `SpeakerNotes` in `Slide`. `mempty`
7163 means no speaker notes.
7164 + Add tests for improved speaker notes.
7165 + Handle speaker notes earlier in the conversion process.
7166 + Keep notes with related blocks (#4477). Some blocks automatically
7167 split slides (imgs, tables, `column` divs). We assume that any
7168 speaker notes immediately following these are connected to these
7169 elements, and keep them with the related blocks, splitting after them.
7170 + Remove `docProps/thumbnail.jpeg` in data dir (Jesse Rosenthal, #4588).
7171 It contained a nonfree ICC color calibration profile and is not needed
7172 for production of a powerpoint document.
7176 + Include a blank line at the end of the row in a single-row multiline
7177 table, to prevent it from being interpreted as a simple table (#4578).
7179 * CommonMark writer:
7181 + Correctly ignore LaTeX raw blocks when `raw_tex` is not
7182 enabled (#4527, quasicomputational).
7186 + Add `epub:type="footnotes"` to notes section in EPUB3 (#4489).
7190 + In beamer, don't use format specifier for default ordered lists
7191 (#4556). This gives better results for styles that put ordered list
7192 markers in boxes or circles.
7193 + Update `\lstinline` delimiters (#4369, Tim Parenti).
7197 + Use `\f[R]` rather than `\f[]` to reset font (#4552).
7198 + Use `\f[BI]` and `\f[CB]` in headers, instead of `\f[I]` and `\f[C]`,
7199 since the header font is automatically bold (#4552).
7200 + Use `\f[CB]` rather than `\f[BC]` for monospace bold (#4552).
7201 + Create pdf anchor for a Div with an identifier (#4515).
7202 + Escape `/` character in anchor ids (#4515).
7203 + Improve escaping for anchor ids: we now use _uNNN_ instead of uNNN
7208 + Don't escape U+2019 as `'` (#4550).
7210 * Text.Pandoc.Options:
7212 + Removed `JsMath`, `LaTeXMathML`, and `GladTeX` constructors from
7213 `Text.Pandoc.Options.HTMLMathMethod` [API change].
7215 * Text.Pandoc.Class:
7217 + `writeMedia`: unescape URI-escaping in file path. This avoids
7218 writing things like `file%20one.png` to the file system.
7220 * Text.Pandoc.Parsing:
7222 + Fix `romanNumeral` parser (#4480). We previously accepted 'DDC'
7224 + `uri`: don't treat `*` characters at end as part of URI (#4561).
7228 + Use the alias `application/eps` for EPS (#2067).
7229 This will ensure that we retain the eps extension after reading the
7230 image into a mediabag and writing it again.
7234 + Use `withTempDir` in `html2pdf`.
7235 + With `xelatex`, don't compress images til the last run (#4484).
7236 This saves time for image-heavy documents.
7237 + Don't try to convert EPS files (#2067). `pdflatex` converts them
7238 itself, and JuicyPixels can't do it.
7239 + For `pdflatex`, use a temp directory in the working directory.
7240 Otherwise we can have problems with the EPS conversion pdflatex
7241 tries to do, which can't operate on a file above the working
7242 directory without `--shell-escape`.
7244 * Changes to tests to accommodate changes in pandoc-types.
7245 In <https://github.com/jgm/pandoc-types/pull/36> we changed
7246 the table builder to pad cells. This commit changes tests
7247 (and two readers) to accord with this behavior.
7249 * Set default extensions for `beamer` same as `latex`.
7253 + Add `beameroption` variable (#4359, Étienne Bersac).
7254 + Use `pgfpages` package; this is needed for notes on second
7255 screen in beamer (Étienne Bersac).
7256 + Add `background-image` variable (#4601, John Muccigrosso).
7258 * reveal.js template: Add `background-image` variable (#4600,
7261 * ms template: Fix date. Previously `.ND` was used, but this only
7262 works if you have a title page, which we don't. Thanks to @teoric.
7264 * Removed pragmas for unused extensions (#4506, Anabra).
7266 * Fix bash completion for `--print-default-data-file` (#4549).
7267 Previously this looked in the filesystem, even if pandoc
7268 was compiled with `embed_data_files` (and sometimes it looked
7269 in a nonexistent build directory). Now the bash completion
7270 script just includes a hard-coded list of data file names.
7274 + Clarify template vs metadata variables (#4501, Mauro Bieg).
7275 + Fix raw content example (#4479, Mauro Bieg).
7276 + Specify that you use html for raw output in epub.
7277 + Add examples for raw docx blocks (#4472, Tristan Stenner).
7278 The documentation states that the target format name should match
7279 the output format, which isn't the case for `docx`/`openxml` and
7281 + Don't say that `empty_paragraphs` affects markdown output (#4540).
7282 + Consolidate input/output format documentation (#4577, Mauro Bieg).
7284 * New README template. Take in/out formats from manual.
7286 * Fix example in lua-filters docs (#4459, HeirOfNorton).
7288 * Use the `-threaded` GHC flag when building benchmarks (#4587,
7289 Francesco Occhipinti).
7291 * Bump temporary upper bound to 1.4.
7293 * Use pandoc-citeproc 0.14.3.1.
7295 * Use texmath-0.10.1.2 (fixes escapes in math in ms, #4597).
7297 * Removed old lib directory. This was used for something long ago,
7298 but plays no role now.
7300 * Removed unneeded data file `LaTeXMathML.js`.
7302 * Create 64- and 32-bit versions of Windows binary packages.
7304## pandoc 2.1.3 (2018-03-18)
7306 * Docx reader (Jesse Rosenthal):
7308 + Add tests for nested smart tags.
7309 + Parse nested smart tags.
7310 + Make unwrapSDT into a general `unwrap` function that can unwrap both
7311 nested SDT tags and smartTags. This makes the SmartTags constructor in
7312 the Docx type unnecessary, so we remove it (#4446).
7313 + Remove unused `docxWarnings` (Alexander Krotov).
7315 * RST reader: Allow unicode bullet characters (#4454).
7317 * Haddock reader: Better table handling, using haddock-library's
7318 new table support, if compiled against a version that
7319 includes it. Note that tables with col/rowspans will not
7320 translate well into Pandoc.
7322 * Muse reader (Alexander Krotov):
7324 + Various internal improvements.
7325 + Require closing tag to have the same indentation as opening.
7326 + Do not reparse blocks inside unclosed block tag (#4425).
7327 + Parse `<class>` tag (supported by Emacs Muse).
7328 + Do not produce empty Str element for unindented verse lines.
7332 + Add support to parse unit string of `\SI` command (closes #4296,
7335 * Haddock writer: In the writer, we now render tables always as
7336 grid tables, since Haddock supports these.
7338 * DokuWiki writer: rewrite backSlashLineBreaks (#4445, Mauro Bieg).
7340 * Docx writer: Fixed formatting of `DefaultStyle` ordered lists in
7341 docx writer. We want decimal for the top level, not lower roman.
7345 + Strip whitespace at beginning and ending of inline containers
7346 (#4327, Francesco Occhipinti).
7347 + Filter out empty inline containers (#4434). There is nothing in
7348 RST that corresponds to e.g. `Emph []`, so we just ignore elements
7349 like this (Francesco Occhipinti).
7351 * Muse writer (Alexander Krotov):
7353 + Support spans with anchors.
7354 + Replace smallcaps with emphasis before normalization.
7355 + Output smallcaps as emphasis.
7356 + Expand Cite before list normalization.
7357 + Write empty inline lists as `<verbatim></verbatim>`.
7358 + Remove empty Str from the beginning of inline lists during normalization.
7359 + Escape "-" to avoid creating bullet lists.
7360 + Fix math expansion for more than one expression per paragraph.
7361 + Expand math before inline list normalization.
7363 * Dokuwiki writer: fix LineBreaks in Tables (#4313, Mauro Bieg).
7367 + Asciify pdf anchors, since unicode anchors don't work (#4436).
7368 Internal links should be converted automatically, so this shouldn't
7369 affect users directly.
7370 + Don't escape hyphens as `\-`; that's for a minus sign (#4467).
7372 * Beamer writer: put hyperlink after `\begin{frame}` and not in the title
7373 (#4307). If it's in the title, then we get a titlebar on slides with
7374 the `plain` attribute, when the id is non-null. This fixes a regression
7377 * EPUB writer: Remove notes from TOC in nav.xhtml (#4453, Mauro Bieg).
7379 * JATS writer: Remove extraneous, significant whitespace (#4335,
7382 * html2pdf: inject base tag with current working directory (#4413, Mauro
7383 Bieg). This helps ensure that linked resources are included.
7385 * Add Semigroup instances for everything for which we defined a
7386 Monoid instance previously (API change):
7388 + Text.Pandoc.Class.FileTree.
7389 + Text.Pandoc.Translations.Translations.
7390 + Text.Pandoc.Extensions.Extensions.
7391 + Text.Pandoc.Readers.Odt.StyleReader.Styles.
7392 + Text.Pandoc.Pretty.Doc.
7393 + Text.Pandoc.MediaBag.MediaBag.
7395 * Add custom Prelude to give clean code for Monoid and Semigroup
7396 that works with ghc 7.10-8.4. The custom Prelude (`prelude/Prelude`)
7397 is used for ghc versions < 8.4. `NoImplicitPrelude` is used
7398 in all source files, and Prelude is explicitly imported
7399 (this is necessary for ghci to work properly with the custom prelude).
7401 * Text.Pandoc.Writers.Shared (Francesco Occhipinti):
7403 + Export `stripLeadingTrailingSpace`.
7404 + Don't wrap lines in grid tables when `--wrap=none` (#4320).
7405 + `gridTable`: Don't wrap lines in tables when `--wrap=none`. Instead,
7406 expand cells, even if it results in cells that don't respect relative
7407 widths or surpass page column width. This change affects RST,
7408 Markdown, and Haddock writers.
7410 * Raise error if someone tries to print docx, odt, etc. template (#4441).
7412 * LaTeX template: Provide `bidi` package's option using
7413 `\PassOptionsToPackage` (#4357, Václav Haisman). This avoid a
7414 clash when `polyglossia` loads it first and then it is loaded again
7417 * ConTeXt template: Added `pdfa` variable to generate PDF/A (#4294, Henri
7418 Menke). Instructions on how to install the ICC profiles on ConTeXt
7419 standalone can be found in the wiki:
7420 <http://wiki.contextgarden.net/PDFX#ICC_profiles>.
7421 If the ICC profiles are not available the log will contain error
7424 * Use latest pandoc-types, skylighting
7426 * Use latest pandoc-citeproc in binary package.
7428 * Bump upper bound for time, criterion, haddock-library, exceptions,
7429 http-types, aeson, haddock-library.
7431 * Bump upper bound tasty-quickcheck 0.10 (#4429, Felix Yan).
7433 * pandoc.cabal: fix up other-extensions and language fields.
7434 Language is now consistently `Haskell2010`, and other-extensions
7435 is consistently `NoImplicitPrelude`. Everything else to be specified
7436 in the module header as needed.
7438 * Removed `old-locale` flag and Text.Pandoc.Compat.Time.
7439 This is no longer necessary since we no longer support ghc 7.8.
7441 * Make `weigh-pandoc` into a benchmark program.
7442 Remove `weigh-pandoc` flag. `weigh-pandoc` is now built (and run)
7443 automatically when you build (and run) benchmarks.
7445 * MANUAL: add instructions for background images reveal.js (#4325, John
7448 * appveyor: use VS 2013 environment instead of VS 2015 for Windows builds.
7450## pandoc 2.1.2 (2018-03-02)
7454 + Fix parsing bug with nested fenced divs (#4281). Previously we allowed
7455 "nonindent spaces" before the opening and closing `:::`, but this
7456 interfered with list parsing, so now we require the fences to be flush with
7457 the margin of the containing block.
7459 * Commonmark reader:
7461 + `raw_html` is now on by default. It can be disabled explicitly
7462 using `-f commonmark-raw_html`.
7464 * Org reader (Albert Krewinkel):
7466 + Move citation tests to separate module.
7467 + Allow changing emphasis syntax (#4378). The characters allowed before
7468 and after emphasis can be configured via `#+pandoc-emphasis-pre` and
7469 `#+pandoc-emphasis-post`, respectively. This allows to change which
7470 strings are recognized as emphasized text on a per-document or even
7471 per-paragraph basis. Example:
7473 #+pandoc-emphasis-pre: "-\t ('\"{"
7474 #+pandoc-emphasis-post: "-\t\n .,:!?;'\")}["
7478 + Fixed comments inside citations (#4374).
7479 + Fix regression in package options including underscore (#4424).
7480 + Make `--trace` work.
7481 + Fixed parsing of `tabular*` environment (#4279).
7485 + Fix regression in parsing of headers with trailing space (#4280).
7487 * Muse reader (Alexander Krotov):
7489 + Enable `<literal>` tags even if amuse extension is enabled.
7490 Amusewiki disables <literal> tags for security reasons.
7491 If user wants similar behavior in pandoc, RawBlocks and RawInlines
7492 can be removed or replaced with filters.
7493 + Remove space prefix from `<literal>` tag contents.
7494 + Do not consume whitespace while looking for closing end tag.
7495 + Convert alphabetical list markers to decimal in round-trip test.
7496 Alphabetical lists are an addition of Text::Amuse.
7497 They are not present in Emacs Muse and can be ambiguous
7498 when list starts with "i.", "c." etc.
7499 + Allow `<quote>` and other tags to be indented.
7500 + Allow single colon in definition list term.
7501 + Fix parsing of verse in lists.
7502 + Improved parsing efficiency. Avoid `parseFromString`.
7503 Lists are parsed in linear instead of exponential time now.
7504 + Replace ParserState with MuseState.
7505 + Prioritize lists with roman numerals over alphabetical lists.
7506 This is to make sure "i." starts a roman numbered list,
7507 instead of a list with letter "i" (followed by "j", "k", ...").
7508 + Fix directive parsing.
7509 + Parse definition lists with multiple descriptions.
7510 + Parse next list item before parsing more item contents.
7511 + Fixed a bug: headers did not terminate lists.
7512 + Move indentation parsing from `definitionListItem` to `definitionList`.
7513 + Paragraph indentation does not indicate nested quote.
7514 Muse allows indentation to indicate quotation or alignment,
7515 but only on the top level, not within a <quote> or list.
7516 + Require that block tags are on separate lines.
7517 Text::Amuse already explicitly requires it anyway.
7518 + Fix matching of closing inline tags.
7519 + Various internal changes.
7520 + Fix parsing of nested definition lists.
7521 + Require only one space for nested definition list indentation.
7522 + Do not remove trailing whitespace from `<code>`.
7523 + Fix parsing of trailing whitespace. Newline after whitespace now
7524 results in softbreak instead of space.
7526 * Docx reader (Jesse Rosenthal, except where noted):
7528 + Handle nested sdt tags (#4415).
7529 + Don't look up dependant run styles if `+styles` is enabled.
7530 + Move pandoc inline styling inside custom-style span.
7531 + Read custom styles (#1843). This will read all paragraph and
7532 character classes as divs and spans, respectively. Dependent styles
7533 will still be resolved, but will be wrapped with appropriate style
7534 tags. It is controlled by the `+styles` extension (`-f docx+styles`).
7535 This can be used in conjunction with the `custom-style` feature in the
7536 docx writer for a pandoc-docx editing workflow. Users can convert from
7537 an input docx, reading the custom-styles, and then use that same input
7538 docx file as a reference-doc for producing an output docx file. Styles
7539 will be maintained across the conversion, even if pandoc doesn't
7541 + Small change to Fields hyperlink parser. Previously, unquoted string
7542 required a space at the end of the line (and consumed it). Now we
7543 either take a space (and don't consume it), or end of input.
7544 + Pick table width from the longest row or header (Francesco Occhipinti,
7547 * Muse writer (Alexander Krotov):
7549 + Change verse markup: `> ` instead of `<verse>` tag.
7550 + Remove empty strings during inline normalization.
7551 + Don't indent nested definition lists.
7552 + Use unicode quotes for quoted text.
7553 + Write image width specified in percent in Text::Amuse mode.
7554 + Don't wrap displayMath into `<verse>`.
7555 + Escape nonbreaking space (`~~`).
7556 + Join code with different attributes during normalization.
7557 + Indent lists inside Div.
7558 + Support definitions with multiple descriptions.
7560 * Powerpoint writer (Jesse Rosenthal):
7562 + Use table styles This will use the default table style in the
7563 reference-doc file. As a result they will be easier when using
7564 in a template, and match the color scheme.
7565 + Remove empty slides. Because of the way that slides were split, these
7566 could be accidentally produced by comments after images. When animations
7567 are added, there will be a way to add an empty slide with either
7568 incremental lists or pauses.
7569 + Implement syntax highlighting. Note that background colors can't
7570 be implemented in PowerPoint, so highlighting styles that require
7571 these will be incomplete.
7572 + New test framework for pptx. We now compare the output of the
7573 Powerpoint writer with files that we know to (a) not be corrupt,
7574 and (b) to show the desired output behavior (details below).
7575 + Add `notesMaster` to `presentation.xml` if necessary.
7576 + Ignore links and (end)notes in speaker notes.
7577 + Output speaker notes.
7578 + Read speaker note templates conditionally. If there are speaker
7579 notes in the presentation, we read in the notesMasters templates
7580 from the reference pptx file.
7581 + Fix deletion track changes (#4303, Jesse Rosenthal).
7583 * Markdown writer: properly escape @ to avoid capture as citation
7588 + Put hypertarget inside figure environment (#4388).
7589 This works around a problem with the endfloat package and
7590 makes pandoc's output compatible with it.
7591 + Fix image height with percentage (#4389). This previously caused
7592 the image to be resized to a percentage of textwidth, rather than
7595 * ConTeXt writer (Henri Menke):
7597 + New section syntax and support `--section-divs` (#2609).
7598 `\section[my-header]{My Header}` ->
7599 `\section[title={My Header},reference={my-header}]`.
7600 The ConTeXt writer now supports the `--section-divs` option to
7601 write sections in the fenced style, with `\startsection` and
7603 + xtables: correct wrong usage of caption (Henri Menke).
7607 + Fix image resizing with multiple images (#3930, Andrew Pritchard).
7608 + Use new golden framework (Jesse Rosenthal).
7609 + Make more deterministic to facilitate testing (Jesse Rosenthal).
7610 - `getUniqueId` now calls to the state to get an incremented digit,
7611 instead of calling to P.uniqueHash.
7612 - we always start the PRNG in mkNumbering/mkAbstractNum with the same
7613 seed (1848), so our randoms should be the same each time.
7614 + Fix ids in comment writing (Jesse Rosenthal). Comments from
7615 `--track-changes=all` were producing corrupt docx, because the
7616 writer was trying to get id from the `(ID,_,_)` field of
7617 the attributes, and ignoring the "id" entry in the key-value pairs. We
7620 * Ms writer: Added papersize variable.
7624 + Use `height` instead of `depth` for images (#4331).
7625 + Ensure that id prefix is always used.
7626 + Don't emit `role` attribute; that was a leftover from the
7628 + Use 'xml:id', not 'id' attribute (#4371).
7632 + Do not output implicit heading IDs (#4363, Alexander
7633 Krotov). Convert to `asciidoc-auto_identifiers` for old behaviour.
7637 + Remove `blockToRST'` moving its logic into `fixBlocks`
7638 (Francesco Occhipinti).
7639 + Insert comment between lists and quotes (#4248, Francesco Occchipinti).
7641 * RST template: remove definition of 'math' role as raw.
7642 This used to be needed prior to v 0.8 of docutils, but
7643 now math support is built-in.
7645 * Slides: Use divs to set incremental/non-incremental (#4381,
7646 Jesse Rosenthal). The old method (list inside blockquote) still
7647 works, but we are encouraging the use of divs with class
7648 `incremental` or `nonincremental`.
7650 * Text.Pandoc.ImageSize:
7652 + Make image size detection for PDFs more robust (#4322).
7653 + Determine image size for PDFs (#4322).
7654 + EMF Image size support (#4375, Andrew Pritchard).
7656 * Text.Pandoc.Extensions:
7658 + Add `Ext_styles` (Jesse Rosenthal, API change). This will be used in
7659 the docx reader (defaulting to off) to read pargraph and character
7660 styles not understood by pandoc (as divs and spans, respectively).
7661 + Made `Ext_raw_html` default for `commonmark` format.
7663 * Text.Pandoc.Parsing:
7665 + Export `manyUntil` (Alexander Krotov, API change).
7666 + Export improved `sepBy1` (Alexander Krotov).
7667 + Export list marker parsers: `upperRoman`, `lowerRoman`,
7668 `decimal`, `lowerAlpha`, `upperAlpha` (Alexander Krotov, API change).
7670 * Tests/Lua: fix tests on windows (Albert Krewinkel).
7672 * Lua: register script name in global variable (#4393). The name of the Lua
7673 script which is executed is made available in the global Lua variable
7674 `PANDOC_SCRIPT_FILE`, both for Lua filters and custom writers.
7676 * Tests: Abstract powerpoint tests out to OOXML tests (Jesse Rosenthal).
7677 There is very little pptx-specific in these tests, so we abstract out
7678 the basic testing function so it can be used for docx as well. This
7679 should allow us to catch some errors in the docx writer that slipped
7680 by the roundtrip testing.
7682 * Lua filters: store constructors in registry (Albert Krewinkel). Lua
7683 functions used to construct AST element values are stored in the Lua
7684 registry for quicker access. Getting a value from the registry is much
7685 faster than getting a global value (partly to idiosyncrasies of hslua);
7686 this change results in a considerable performance boost.
7690 + `doc/org.md` Add draft of Org-mode documentation (Albert Krewinkel).
7691 + `doc/lua-filters.md`: document global vars set for filters
7693 + INSTALL.md: mention Stack version. (#4343, Adam Brandizzi).
7694 + MANUAL: add documentation on custom styles (Jesse Rosenthal).
7695 + MANUAL.txt: Document incremental and nonincremental divs (Jesse
7696 Rosenthal). Blockquoted lists are still described, but fenced divs are
7697 presented in preference.
7698 + MANUAL.txt: document header and footer variables (newmana).
7699 + MANUAL.txt: self-contained implies standalone (#4304, Daniel Lublin).
7700 + CONTRIBUTING.md: label was renamed. (#4310, Alexander Brandizzi).
7702 * Require tagsoup 0.14.3 (#4282), fixing HTML tokenization bug.
7704 * Use latest texmath.
7706 * Use latest pandoc-citeproc.
7708 * Allow exceptions 0.9.
7710 * Require aeson-pretty 0.8.5 (#4394).
7712 * Bump blaze-markup, blaze-html lower bounds to 0.8, 0.9 (#4334).
7714 * Update tagsoup to 0.14.6 (Alexander Krotov, #4282).
7716 * Removed ghc-prof-options. As of cabal 1.24, sensible defaults are used.
7718 * Update default.nix to current nixpkgs-unstable for hslua-0.9.5 (#4348,
7721## pandoc 2.1.1 (2018-01-18)
7725 + Don't coalesce adjacent raw LaTeX blocks if they are separated by a
7726 blank line. See lierdakil/pandoc-crossref#160.
7727 + Improved `inlinesInBalancedBrackets` (#4272, jgm/pandoc-citeproc#315).
7728 The change both improves performance and fixes a regression whereby
7729 normal citations inside inline notes and figure captions were not
7734 + Better handling for headers with an anchor (#4240). Instead of creating a
7735 Div containing the header, we put the id directly on the header.
7736 This way header promotion will work properly.
7737 + Add aligned environment when needed in math (#4254). `rst2latex.py`
7738 uses an `align*` environment for math in `.. math::` blocks, so this
7739 math may contain line breaks. If it does, we put the math in an
7740 `aligned` environment to simulate `rst2latex.py`'s behavior.
7744 + Fix col width parsing for percentages < 10% (#4262, n3fariox).
7748 + Advance source position at end of stream.
7749 + Pass through macro defs in `rawLaTeXBlock` even if the `latex_macros`
7750 extension is set (#4246). This reverts to earlier behavior and is
7751 probably safer on the whole, since some macros only modify things in
7752 included packages, which pandoc's macro expansion can't modify.
7753 + Fixed pos calculation in tokenizing escaped space.
7754 + Allow macro definitions inside macros (#4253). Previously we went into
7755 an infinite loop with
7757 \newcommand{\noop}[1]{#1}
7758 \noop{\newcommand{\foo}[1]{#1}}
7761 + Fix inconsistent column widths (#4238). This fixes a bug whereby column
7762 widths for the body were different from widths for the header in some
7765 * Docx reader (Jesse Rosenthal):
7767 + Parse hyperlinks in `instrText` tags (#3389, #4266). This was a form of
7768 hyperlink found in older versions of word. The changes introduced for
7769 this, though, create a framework for parsing further fields in MS Word
7770 (see the spec, ECMA-376-1:2016, §17.16.5, for more on these fields).
7771 We introduce a new module, `Text.Pandoc.Readers.Docx.Fields` which
7772 contains a simple parsec parser. At the moment, only simple hyperlink
7773 fields are accepted, but that can be extended in the future.
7775 * Muse reader (Alexander Krotov):
7777 + Parse `~~` as non-breaking space in Text::Amuse mode.
7778 + Refactor list parsing.
7780 * Powerpoint writer (Jesse Rosenthal):
7782 + Change reference to `notesSlide` to `endNotesSlide`.
7783 + Move image sizing into `picProps`.
7784 + Improve table placement.
7785 + Make our own `_rels/.rels` file.
7786 + Import reference-doc images properly.
7787 + Move `Presentation.hs` out of `PandocMonad`.
7788 + Refactor into separate modules. T.P.W.Powerpoint.Presentation
7789 defines the Presentation datatype and goes Pandoc->Presentation;
7790 T.P.W.Pandoc.Output goes Presentation->Archive.
7791 Text.Pandoc.Writers.Powerpoint a thin wrapper around the two modules.
7792 + Avoid overlapping blocks in column output.
7793 + Position images correctly in two-column layout.
7794 + Make content shape retrieval environment-aware.
7795 + Improve image handling. We now determine image and caption placement
7796 by getting the dimensions of the content box in a given layout.
7797 This allows for images to be correctly sized and positioned in a
7798 different template. Note that images without captions and headers are
7799 no longer full-screened. We can't do this dependably in different
7800 layouts, because we don't know where the header is (it could be to
7801 the side of the content, for example).
7802 + Read presentation size from reference file. Our presentation size is
7803 now dependent on the reference/template file we use.
7804 + Handle (sub)headers above slidelevel correctly. Above the slidelevel,
7805 subheaders will be printed in bold and given a bit of extra space
7806 before them. Note that at the moment, no distinction is made between
7807 levels of headers above the slide header, though that can be changed.
7808 + Check for required files. Since we now import from reference/dist
7809 file by glob, we need to make sure that we're getting the files we
7810 need to make a non-corrupt Powerpoint. This performs that check.
7811 + Improve templating using `--reference-doc`. Templating should work
7812 much more reliably now.
7813 + Include Notes slide in TOC.
7814 + Set notes slide header to slide-level.
7815 + Add table of contents. This is triggered by the `--toc` flag. Note
7816 that in a long slide deck this risks overrunning the text box. The user
7817 can address this by setting `--toc-depth=1`.
7818 + Set notes slide number correctly.
7819 + Clean up adding metadata slide. We want to count the slide numbers
7820 correctly if it's in there.
7821 + Add anchor links. For anchor-type links (`[foo](#bar)`) we produce
7822 an anchor link. In powerpoint these are links to slides, so we keep
7823 track of a map relating anchors to the slides they occur on.
7824 + Make the slide number available to the blocks. For anchors,
7825 block-processing functions need to know what slide number
7826 they're in. We make the `envCurSlideId` available to blocks.
7827 + Move `curSlideId` to environment.
7828 + Allow setting `toc-title` in metadata.
7829 + Link notes to endnotes slide.
7833 + Fix cell width calculation (#4265). Previously we could get
7834 ever-lengthening cell widths when a table was run repeatedly through
7835 `pandoc -f markdown -t markdown`.
7839 + Escape `&` in lstinline (Robert Schütz).
7843 + Use xtables instead of Tables (#4223, Henri Menke).
7844 Default to xtables for context output. Natural Tables are used
7845 if the new `ntb` extension is set.
7849 + Fixed footnote backlinks with `--id-prefix` (#4235).
7851 * `Text.Pandoc.Extensions`: Added `Ext_ntb` constructor (API change,
7854 * `Text.Pandoc.ImageSize`: add derived `Eq` instance to `Dimension`
7855 (Jesse Rosenthal, API change).
7857 * Lua filters (Albert Krewinkel):
7859 + Make `PANDOC_READER_OPTIONS` available.
7860 The options which were used to read the document are made available to
7861 Lua filters via the `PANDOC_READER_OPTIONS` global.
7862 + Add lua module `pandoc.utils.run_json_filter`, which runs a JSON filter
7863 on a Pandoc document.
7864 + Refactor filter-handling code into `Text.Pandoc.Filter.JSON`,
7865 `Text.Pandoc.Filter.Lua`, and `Text.Pandoc.Filter.Path`.
7866 + Improve error messages. Provide more context about the task
7867 which caused an error.
7869 * data/pandoc.lua (Albert Krewinkel):
7871 + Accept singleton inline as a list. Every constructor which accepts a
7872 list of inlines now also accepts a single inline element for
7874 + Accept single block as singleton list. Every constructor which accepts
7875 a list of blocks now also accepts a single block element for
7876 convenience. Furthermore, strings are accepted as shorthand for
7877 `{pandoc.Str "text"}` in constructors.
7878 + Add attr, listAttributes accessors. Elements with
7879 attributes got an additional `attr` accessor. Attributes were
7880 accessible only via the `identifier`, `classes`, and `attributes`,
7881 which was in conflict with the documentation, which indirectly states
7882 that such elements have the an `attr` property.
7883 + Drop `_VERSION`. Having a `_VERSION` became superfluous, as this
7884 module is closely tied to the pandoc version, which is available via
7886 + Fix access to Attr components. Accessing an Attr value (e.g.,
7887 ` Attr().classes`) was broken; the more common case of accessing it via
7888 an Inline or Block element was unaffected by this.
7890 * Move `metaValueToInlines` to from Docx writer to
7891 `Text.Pandoc.Writers.Shared`, so it can be used by other writers
7896 + Clarify otherlangs in LaTeX (#4072).
7897 + Clarify `latex_macros` extension.
7898 + Recommend use of `raw_attribute` extension in header includes (#4253).
7900 * Allow latest QuickCheck, tasty, criterion.
7902 * Remove custom prelude and ghc 7.8 support.
7904 * Reduce compiler noise (exact paths for compiled modules).
7906## pandoc 2.1 (2018-01-07)
7908 * Allow filters and lua filters to be interspersed (#4196). Previously
7909 we ran all lua filters before JSON filters. Now we run filters in
7910 the order they are presented on the command line, whether lua or JSON.
7911 There are two incompatible API changes: The type of `applyFilters`
7912 has changed, and `applyLuaFilters` has been removed. `Filter` is
7915 * Use latest skylighting and omit the `missingIncludes` check, fixing
7916 a major performance regression in earlier releases of the 2.x series
7917 (#4226). Behavior change: If you use a custom syntax definition that
7918 refers to a syntax you haven't loaded, pandoc will now complain when
7919 it is highlighting the text, rather than doing a check at the start.
7920 This change dramatically speeds up invocations of pandoc on short
7923 * Text.Pandoc.Class: make `FileTree` opaque (don't export
7924 `FileTree` constructor). This forces users to interact with it using
7925 `insertInFileTree` and `getFileInfo`, which normalize file names.
7929 + Rewrite `inlinesInBalancedBrackets`. The rewrite is much more
7930 direct, avoiding `parseFromString`. And it performs significantly
7931 better; unfortunately, parsing time still increases exponentially
7933 + Avoid parsing raw tex unless `\` + letter seen. This seems to
7934 help with the performance problem, #4216.
7936 * LaTeX reader: Simplified a check for raw tex command.
7938 * Muse reader (Alexander Krotov):
7940 + Enable round trip test (#4107).
7941 + Automatically translate `#cover` into `#cover-image`.
7942 Amusewiki uses #cover directive to specify cover image.
7944 * Docx reader (Jesse Rosenthal):
7946 + Allow for insertion/deletion of paragraphs (#3927).
7947 If the paragraph has a deleted or inserted paragraph break (depending
7948 on the track-changes setting) we hold onto it until the next
7949 paragraph. This takes care of accept and reject. For this we introduce
7950 a new state which holds the ils from the previous para if necessary.
7951 For `--track-changes=all`, we add an empty span with class
7952 `paragraph-insertion`/`paragraph-deletion` at the end of the paragraph
7953 prior to the break to be inserted or deleted.
7954 + Remove unused anchors (#3679). Docx produces a lot of anchors with
7955 nothing pointing to them---we now remove these to produce cleaner
7956 output. Note that this has to occur at the end of the process
7957 because it has to follow link/anchor rewriting.
7958 + Read multiple children of `w:sdtContents`.
7959 + Combine adjacent anchors. There isn't any reason to have numerous
7960 anchors in the same place, since we can't maintain docx's
7961 non-nesting overlapping. So we reduce to a single anchor.
7964 * Muse writer (Alexander Krotov): don't escape URIs from AST
7968 + Removed redundant subtitle in title (Sebastian Talmon).
7969 + `firstRow` table definition compatibility for Word 2016 (Sebastian
7970 Talmon). Word 2016 seems to use a default value of "1" for table
7971 headers, if there is no firstRow definition (although a default
7972 value of 0 is documented), so all tables get the first Row formatted
7973 as header. Setting the parameter to 0 if the table has no header
7974 row fixes this for Word 2016
7975 + Fix custom styles with spaces in the name (#3290).
7977 * Powerpoint writer (Jesse Rosenthal):
7979 + Ignore Notes div for parity with other slide outputs.
7980 + Set default slidelevel correctly. We had previously defaulted to
7981 slideLevel 2. Now we use the correct behavior of defaulting to the
7982 highest level header followed by content. We change an expected test
7983 result to match this behavior.
7984 + Split blocks correctly for linked images.
7985 + Combine adjacent runs.
7986 + Make inline code inherit code size. Previously (a) the code size
7987 wasn't set when we force size, and (b) the properties was set from
7988 the default, instead of inheriting.
7989 + Simplify `replaceNamedChildren` function.
7990 + Allow linked images. The following markdown:
7991 `[](http://www.example.com)`
7992 will now produce a linked image in the resulting PowerPoint file.
7993 + Fix error with empty table cell. We require an empty `<a:p>` tag,
7994 even if the cell contains no paragraphs---otherwise PowerPoint
7995 complains of corruption.
7996 + Implement two-column slides. This uses the columns/column div
7997 format described in the pandoc manual. At the moment, only two
7998 columns (half the screen each) are allowed. Custom widths are not
8002 * OpenDocument/ODT writers: improved rendering of formulas (#4170, oltolm).
8004 * Lua filters (Albert Krewinkel):
8006 + `data/pandoc.lua`: drop 'pandoc-api-version' from Pandoc objects
8007 + The current pandoc-types version is made available to Lua programs in
8008 the global `PANDOC_API_VERSION`. It contains the version as a list of
8010 + The pandoc version available as a global `PANDOC_VERSION` (a list
8012 + `data/pandoc.lua`: make `Attr` an `AstElement`.
8013 + `data/pandoc.lua`: make all types subtypes of `AstElement`.
8014 `Pandoc`, `Meta`, and `Citation` were just plain functions and did
8015 not set a metatable on the returned value, which made it difficult
8016 to amend objects of these types with new behavior. They are now
8017 subtypes of AstElement, meaning that all their objects can gain
8018 new features when a method is added to the behavior object
8019 (e.g., `pandoc.Pandoc.behavior`).
8020 + `data/pandoc.lua`: split type and behavior tables. Clearly distinguish
8021 between a type and the behavioral properties of an instance of that
8022 type. The behavior of a type (and all its subtypes) can now be
8023 amended by adding methods to that types `behavior` object, without
8024 exposing the type objects internals. E.g.:
8026 pandoc.Inline.behavior.frob = function () print'42' end
8027 local str = pandoc.Str'hello'
8028 str.frob() -- outputs '42'
8030 + `data/pandoc.lua`: fix Element inheritance. Extending all elements
8031 of a given type (e.g., all inline elements) was difficult, as the
8032 table used to lookup unknown methods would be reset every time a
8033 new element of that type was created, preventing recursive property
8034 lookup. This is was changed in that all methods and attributes of
8035 supertypes are now available to their subtypes.
8036 + `data/pandoc.lua`: fix attribute names of Citation (#4222). The
8037 fields were named like the Haskell fields, not like the documented,
8038 shorter version. The names are changed to match the documentation
8039 and Citations are given a shared metatable to enable simple
8041 + `data/pandoc.lua`: drop function `pandoc.global_filter`.
8042 + Bump `hslua` version to 0.9.5. This version fixes a bug that made it
8043 difficult to handle failures while getting lists or a Map from Lua.
8044 A bug in pandoc, which made it necessary to always pass a tag when
8045 using MetaList or MetaBlock, is fixed as a result. Using the pandoc
8046 module's constructor functions for these values is now optional
8047 (if still recommended).
8048 + Stop exporting `pushPandocModule` (API change). The introduction
8049 of `runPandocLua` renders direct use of this function obsolete.
8050 + Update generation of module docs for lua filters.
8051 + `Lua.Module.Utils`: make stringify work on `MetaValues` (John
8052 MacFarlane). I'm sure this was intended in the first place,
8053 but currently only `Meta` is supported.
8055 * Improve benchmarks.
8057 + Set the default extensions properly.
8058 + Improve benchmark argument parsing. You can now say
8059 `make bench BENCHARGS="markdown latex reader"` and both the
8060 markdown and latex readers will be benchmarked.
8062 * MANUAL.txt simplify and add more structure (Mauro Bieg).
8064 * Generate README.md from template and MANUAL.txt.
8065 `make README.md` will generate the README.md after changes
8066 to MANUAL.txt have been made.
8068 * Update copyright notices to include 2018 (Albert Krewinkel).
8070## pandoc 2.0.6 (2017-12-28)
8072 * Added `jats` as an input format.
8074 + Add Text.Pandoc.Readers.JATS, exporting `readJATS` (API
8075 change) (Hamish Mackenzie).
8076 + Improved citation handling in JATS reader. JATS citations
8077 are now converted to pandoc citations, and JATS ref-lists
8078 are converted into a `references` field in metadata, suitable
8079 for use with pandoc-citeproc. Thus a JATS article with embedded
8080 bibliographic information can be processed with pandoc and
8081 pandoc-citeproc to produce a formatted bibliography.
8083 * Allow `--list-extensions` to take an optional FORMAT argument.
8084 This lists the extensions set by default for the selected FORMAT.
8085 The extensions are now alphabetized, and the `+` or `-`
8086 indicating the default setting comes before, rather than after,
8091 + Preserve original whitespace between blocks.
8092 + Recognize `\placeformula` as context.
8093 + Be pickier about table captions. A caption starts with a `:` which
8094 can't be followed by punctuation. Otherwise we can falsely interpret
8095 the start of a fenced div, or even a table header line like
8096 `:--:|:--:`, as a caption.
8097 + Always use four space rule for example lists. It would be awkward
8098 to indent example list contents to the first non-space character after
8099 the label, since example list labels are often long. Thanks to
8100 Bernhard Fisseni for the suggestion.
8101 + Improve raw tex parsing. Note that the Markdown reader is also
8102 affected by the `latex_macros` extension changes described below
8103 under the LaTeX reader.
8107 + `latex_macros` extension changes (#4179). Don't pass through macro
8108 definitions themselves when `latex_macros` is set. The macros
8109 have already been applied. If `latex_macros` is enabled, then
8110 `rawLaTeXBlock` in Text.Pandoc.Readers.LaTeX will succeed in parsing
8111 a macro definition, and will update pandoc's internal macro map
8112 accordingly, but the empty string will be returned.
8113 + Export `tokenize`, `untokenize` (API change).
8114 + Use `applyMacros` in `rawLaTeXBlock`, `rawLaTeXInline`.
8115 + Refactored `inlineCommand`.
8116 + Fix bug in tokenizer. Material following `^^` was
8117 dropped if it wasn't a character escape. This only affected
8118 invalid LaTeX, so we didn't see it in the wild, but it appeared
8119 in a QuickCheck test failure.
8120 + Fix regression in LateX tokenization (#4159). This mainly affects the
8121 Markdown reader when parsing raw LaTeX with escaped spaces.
8122 + Add tests of LaTeX tokenizer.
8123 + Support `\foreignlanguage` from babel.
8124 + Be more tolerant of `&` character (#4208). This allows us to parse
8125 unknown tabular environments as raw LaTeX.
8127 * Muse reader (Alexander Krotov):
8129 + Parse anchors immediately after headings as IDs.
8130 + Require that note references does not start with 0.
8131 + Parse empty comments correctly.
8133 * Org reader (Albert Krewinkel):
8135 + Fix asterisks-related parsing error (#4180).
8136 + Support minlevel option for includes (#4154). The level of headers
8137 in included files can be shifted to a higher level by specifying a
8138 minimum header level via the `:minlevel` parameter. E.g.
8139 `#+include: "tour.org" :minlevel 1` will shift the headers in
8140 tour.org such that the topmost headers become level 1 headers.
8141 + Break-up org reader test file into multiple modules.
8145 + Enable raw HTML and other extensions by default for notes
8146 (#4164). This fixes a regression in 2.0. Note that extensions can
8147 now be individually disabled, e.g. `-f opml-smart-raw_html`.
8151 + Allow empty list items (#4193).
8152 + More accurate parsing of references (#4156). Previously we erroneously
8153 included the enclosing backticks in a reference ID (#4156). This
8154 change also disables interpretation of syntax inside references, as
8155 in docutils. So, there is no emphasis in `` `my *link*`_ ``.
8159 + Continue lists after interruption (#4025, Jesse Rosenthal).
8160 Docx expects that lists will continue where they left off after an
8161 interruption and introduces a new id if a list is starting again. So
8162 we keep track of the state of lists and use them to define a "start"
8163 attribute, if necessary.
8164 + Add tests for structured document tags unwrapping (Jesse Rosenthal).
8165 + Preprocess Document body to unwrap `w:sdt` elements (Jesse Rosenthal,
8170 + Don't linkify table of contents.
8174 + Fix anchors for headers (#4188). We were missing an `_`.
8176 * PowerPoint writer (Jesse Rosenthal):
8178 + Treat lists inside BlockQuotes as lists. We don't yet produce
8179 incremental lists in PowerPoint, but we should at least treat lists
8180 inside BlockQuotes as lists, for compatibility with other slide formats.
8181 + Add ability to force size. This replaces the more specific
8182 `blockQuote runProp`, which only affected the size of blockquotes. We
8183 can use this for notes, etc.
8184 + Implement notes. This currently prints all notes on a final slide.
8185 Note that at the moment, there is a danger of text overflowing the
8186 note slide, since there is no logic for adding further slides.
8187 + Implement basic definition list functionality to PowerPoint writer.
8188 + Don't look for default template file for Powerpoint (#4181).
8189 + Add pptx to isTextFormat list. This is used to check standalone
8190 and not writing to the terminal.
8191 + Obey slide level option (Jesse Rosenthal).
8196 + Ensure that `distArchive` is the one that comes with pandoc
8197 (#4182). Previously a `reference.docx` in `~/.pandoc` (or the user data
8198 dir) would be used instead, and this could cause problems because a
8199 user-modified docx sometimes lacks vital sections that we count
8200 on the `distArchive` to supply.
8204 + Do not wrap "-" to avoid accidental bullet lists (Alexander Krotov).
8205 + Don't allow fn refs to wrap to beginning of line (#4171, with help from
8206 Alexander Krotov). Otherwise they can be interpreted as footnote
8209 * Muse writer (Alexander Krotov):
8211 + Don't wrap note references to the next line (#4172).
8215 + Use br elements in line blocks instead of relying on CSS
8216 (#4162). HTML-based templates have had the custom CSS for
8217 `div.line-block` removed. Those maintaining custom templates will want
8218 to remove this too. We still enclose line blocks in a div with class
8223 + Use `\renewcommand` for `\textlatin` with babel (#4161).
8224 This avoids a clash with a deprecated `\textlatin` command defined
8226 + Allow fragile=singleslide attribute in beamer slides (#4169).
8227 + Use `\endhead` after `\toprule` in headerless tables (#4207).
8231 + Add cover image specified by `cover-image` meta (Alexander Krotov,
8234 * JATS writer (Hamish Mackenzie):
8236 + Support writing `<fig>` and `<table-wrap>` elements
8237 with `<title>` and `<caption>` inside them by using Divs with class set
8238 to one of `fig`, `table-wrap` or `caption` (Hamish Mackenzie). The
8239 title is included as a Heading so the constraint on where Heading can
8240 occur is also relaxed.
8241 + Leave out empty alt attributes on links.
8242 + Deduplicate image mime type code.
8243 + Make `<p>` optional in `<td>` and `<th>` (#4178).
8244 + Self closing tags for empty xref (#4187).
8245 + Improve support for code language.
8249 + Use init file to setup Lua interpreter (Albert Krewinkel).
8250 The same init file (`data/init`) that is used to setup the Lua
8251 interpreter for Lua filters is also used to setup the interpreter of
8253 + Define instances for newtype wrapper (Albert Krewinkel). The custom
8254 writer used its own `ToLuaStack` instance definitions, which made
8255 it difficult to share code with Lua filters, as this could result
8256 in conflicting instances. A `Stringify` wrapper is introduced to
8258 + Added tests for custom writer.
8259 + Fixed definition lists and tables in `data/sample.lua`.
8261 * Fixed regression: when target is PDF, writer extensions were being
8262 ignored. So, for example, `pandoc -t latex-smart -o file.pdf`
8263 did not work properly.
8265 * Lua modules (Albert Krewinkel):
8267 + Add `pandoc.utils` module, to hold utility functions.
8268 + Create a Haskell module Text.Pandoc.Lua.Module.Pandoc to
8269 define the `pandoc` lua module.
8270 + Make a Haskell module for each Lua module. Move definitions for the
8271 `pandoc.mediabag` modules to a separate Haskell module.
8272 + Move `sha1` from the main `pandoc` module to `pandoc.utils`.
8273 + Add function `pandoc.utils.hierarchicalize` (convert list of
8274 Pandoc blocks into (hierarchical) list of Elements).
8275 + Add function `pandoc.utils.normalize_date` (parses a date and
8276 converts it (if possible) to "YYYY-MM-DD" format).
8277 + Add function `pandoc.utils.to_roman_numeral` (allows conversion
8278 of numbers below 4000 into roman numerals).
8279 + Add function `pandoc.utils.stringify` (converts any AST element
8280 to a string with formatting removed).
8281 + `data/init.lua`: load `pandoc.utils` by default
8282 + Turn pipe, read into full Haskell functions. The `pipe` and `read`
8283 utility functions are converted from hybrid lua/haskell functions
8284 into full Haskell functions. This avoids the need for intermediate
8285 `_pipe`/`_read` helper functions, which have dropped.
8286 + pandoc.lua: re-add missing MetaMap function. This was a bug
8287 introduced in version 2.0.4.
8289 * Text.Pandoc.Class: Add `insertInFileTree` [API change]. This gives
8290 a pure way to insert an ersatz file into a `FileTree`. In addition, we
8291 normalize paths both on insertion and on lookup.
8293 * Text.Pandoc.Shared: export `blocksToInlines'` (API change, Maura Bieg).
8295 * Text.Pandoc.MIME: Add opus to MIME type table as audio/ogg (#4198).
8297 * Text.Pandoc.Extensions: Alphabetical order constructors for
8298 `Extension`. This makes them appear in order in `--list-extensions`.
8300 * Allow lenient decoding of latex error logs, which are not always
8301 properly UTF8-encoded (#4200).
8303 * Update latex template to work with recent versions of beamer.
8304 The old template produced numbered sections with some recent
8305 versions of beamer. Thanks to Thomas Hodgson.
8307 * Updated reference.docx (#4175). Instead of just "Hello, world", the
8308 document now contains exemplars of most of the styles that have an
8309 effect on pandoc documents. This makes it easier to see the effect
8312 * Removed `default.theme` data file (#4096). It is no longer needed now
8313 that we have `--print-highlight-style`.
8315 * Added `stack.lts9.yaml` for building with lts 9 and ghc 8.0.2.
8316 We still need this for the alpine static linux build, since
8317 we don't have ghc 8.2.2 for that yet.
8319 * Removed `stack.pkg.yaml`. We only really need `stack.yaml`; we
8320 can put flag settings for pandoc-citeproc there.
8322 * Makefile: Add 'trypandoc' and 'pandoc-templates' targets to
8323 make releases easier.
8327 + Add note on what formats have `+smart` by default.
8328 + Use native syntax for custom-style (#4174, Mauro Bieg).
8329 + Introduce dedicated Extensions section, since some extensions
8330 affect formats other than markdown (Mauro Bieg, #4204).
8331 + Clarify default html output for `--section-divs` (Richard Edwards).
8333 * filters.md: say that Text.Pandoc.JSON comes form pandoc-types.
8334 Closes jgm/pandoc-website#16.
8336 * epub.md: Delete removed `-S` option from command (#4151, Georger Araújo).
8338## pandoc 2.0.5 (2017-12-12)
8340 * Fix a bug in 2.0.4, whereby pandoc could not read the theme files
8341 generated with `--print-highlight-style` (#4133). Improve JSON
8342 serialization of styles.
8344 * Fix CSS issues involving line numbers (#4128).
8345 Highlighted code blocks are now enclosed in a div with class `sourceCode`.
8346 Highlighting CSS no longer sets a generic color for pre and code; we only
8347 set these for class `sourceCode`.
8349 * `--pdf-engine-opt`: fix bug where option order was reversed (#4137).
8351 * Add PowerPoint (pptx) writer (Jesse Rosenthal).
8352 It works following the standard Pandoc conventions for making other
8353 sorts of slides. Caveats:
8355 + Syntax highlighting is not yet implemented. (This is difficult
8356 because there are no character classes in Powerpoint.)
8357 + Footnotes and Definition lists are not yet implemented. (Notes will
8358 usually take the form of a final slide.
8359 + Image placement and auto-resizing has a few glitches.
8360 + Reference powerpoint files don't work dependably from the command
8361 line. This will be implemented, but at the moment users are advised
8362 to change themes from within Powerpoint.
8364 * Create shared Text.Pandoc.Writers.OOXML module (Jesse Rosenthal).
8365 This is for functions used by both Powerpoint and Docx writers.
8367 * Add default pptx data for Powerpoint writer (Jesse Rosenthal).
8369 * Add `empty_paragraphs` extension.
8371 + Deprecate `--strip-empty-paragraphs` option. Instead we now
8372 use an `empty_paragraphs` extension that can be enabled on
8373 the reader or writer. By default, disabled.
8374 + Add `Ext_empty_paragraphs` constructor to `Extension`.
8375 + Revert "Docx reader: don't strip out empty paragraphs."
8376 This reverts commit d6c58eb836f033a48955796de4d9ffb3b30e297b.
8377 + Implement `empty_paragraphs` extension in docx reader and writer,
8378 opendocument writer, html reader and writer.
8379 + Add tests for `empty_paragraphs` extension.
8383 + Don't parse native div as table caption (#4119).
8384 + Improved computation of column widths in pipe tables.
8385 Pipe tables with lines longer than the text width (as set
8386 by `--columns`) are now scaled to text width, with the relative
8387 widths of columns determined by the ratios between the
8388 header lines. Previously we computed column widths using
8389 the ratio of header line lengths to column width, so that
8390 tables with narrow header lines were extremely thin, which
8391 was very rarely the desired result.
8393 * LaTeX reader: fix `\` before newline (#4134). This should be a space,
8394 as long as it's not followed by a blank line. This has been fixed at the
8397 * Muse reader (Alexander Krotov):
8399 + Add test for `#disable-tables` directive in Emacs mode.
8400 + Don't allow emphasis to be preceded by letter.
8401 + Add underline support in Emacs Muse mode..
8402 + Support multiline directives in Amusewiki mode
8404 * Man writer: omit internal links (#4136). That is, just print the link
8405 text without the URL.
8407 * Markdown reader: accept processing instructions as raw HTML (#4125).
8409 * Lua filters (Albert Krewinkel):
8411 + Use script to initialize the interpreter. The file `init.lua` is
8412 used to initialize the Lua interpreter which is used in Lua filters.
8413 This gives users the option to require libraries which they want to
8414 use in all of their filters, and to extend default modules.
8415 + Fix package loading for Lua 5.1. The list of package searchers is
8416 named `package.loaders` in Lua 5.1 and LuaJIT, and `package.searchers`
8417 in Lua 5.2 and later.
8418 + Refactor lua module handling. The integration with Lua's package/module
8419 system is improved: A pandoc-specific package searcher is prepended to
8420 the searchers in `package.searchers`. The modules `pandoc` and
8421 `pandoc.mediabag` can now be loaded via `require`.
8422 + Bump lower bound of hslua. The release hslua 0.9.3 contains a new
8423 function which makes using Haskell functions as package loaders much
8426 * reveal.js template: add title-slide identifier to title slide (#4120).
8427 This allows it to be styled more easily.
8429 * LaTeX template: Added support for `pagestyle` variable (#4135,
8432 * Add `-threaded` to ghc-options for executable (#4130, fixes a build
8436## pandoc 2.0.4 (2017-12-03)
8438 * Add `--print-highlight-style` option. This generates a JSON version
8439 of a highlighting style, which can be saved as a `.theme` file, modified,
8440 and used with `--highlight-style` (#4106, #4096).
8442 * Add `--strip-empty-paragraphs` option. This works for any input format.
8443 It is primarily intended for use with docx and odt documents where
8444 empty paragraphs have been used for inter-paragraph spaces.
8446 * Support `--webtex` for `gfm` output.
8448 * Recognize `.muse` file extension.
8450 * Support beamer `\alert` in LaTeX reader. Closes #4091.
8452 * Docx reader: don't strip out empty paragraphs (#2252).
8453 Users who have a conversion pipeline from docx may want to consider adding
8454 `--strip-empty-paragraphs` to the command line.
8456 * Org reader (Albert Krewinkel): Allow empty list items (#4090).
8458 * Muse reader (Alexander Krotov):
8460 + Parse markup in definition list terms.
8461 + Allow definition to end with EOF.
8462 + Make code blocks round trip.
8463 + Drop common space prefix from list items.
8464 + Add partial round trip test.
8465 + Don't interpret XML entities.
8467 + Parse `~~` as non-breaking space in Emacs mode.
8468 + Correctly remove indentation from notes. Exactly one space is
8469 required and considered to be part of the marker.
8470 + Allow list items to be empty.
8471 + Add ordered list test.
8472 + Add more multiline definition tests.
8473 + Don't allow blockquotes within lists.
8474 + Fix reading of multiline definitions.
8475 + Add inline `<literal>` support.
8476 + Concatenate inlines of the same type
8478 * Docx writer: allow empty paragraphs (#2252).
8480 * CommonMark/gfm writer:
8482 + Use raw html for native divs/spans (#4113). This allows a pandoc
8483 markdown native div or span to be rendered in gfm using raw html tags.
8484 + Implement `raw_html` and `raw_tex` extensions. Note that `raw_html`
8485 is enabled by default for `gfm`, while `raw_tex` is disabled by default.
8487 * Muse writer (Alexander Krotov):
8489 + Test that inline math conversion result is normalized.
8490 Without normalization this test produced
8491 `<em>a</em><em>b</em><em>c</em>`.
8492 + Improve inline list normalization and move to writer.
8493 + Escape hash symbol.
8494 + Escape `----` to avoid accidental horizontal rules.
8495 + Escape only `</code>` inside code tag.
8496 + Additional `<verbatim>` is not needed as `<code>` is verbatim already.
8500 + Allow specifying just width or height for image size.
8501 Previously both needed to be specified (unless the image was
8502 being resized to be smaller than its original size).
8503 If height but not width is specified, we now set width to
8504 textwidth. If width but not height is specified, we now set
8505 height to textheight. Since we have `keepaspectratio`, this
8506 yields the desired result.
8507 + Escape `~` and `_` in code with `--listings` (#4111).
8509 * HTML writer: export `tagWithAttributes`. This is a helper allowing
8510 other writers to create single HTML tags.
8512 * Let papersizes `a0`, `a1`, `a2`, ... be case-insensitive by
8513 converting the case as needed in LaTeX and ConTeXt writers.
8515 * Change `fixDisplayMath` from `Text.Pandoc.Writers.Shared`
8516 so that it no longer produces empty `Para`'s as an artifact.
8518 * `Text.Pandoc.Shared.blocksToInlines`: rewrote using builder.
8519 This gives us automatic normalization, so we don't get
8520 for example two consecutive Spaces.
8522 * Include default CSS for 'underline' class in HTML-based templates.
8524 * revealjs template: add `tex2jax` configuration for the
8525 math plugin. With the next release of reveal.js, this will
8526 fix the problem of `$`s outside of math contexts being
8527 interpreted as math delimiters (#4027).
8529 * `pandoc.lua` module for use in lua filters (Albert Krewinkel):
8531 + Add basic lua List module (#4099, #4081). The List module is
8532 automatically loaded, but not assigned to a global variable. It can be
8533 included in filters by calling `List = require 'List'`. Lists of blocks,
8534 lists of inlines, and lists of classes are now given `List` as a metatable,
8535 making working with them more convenient. E.g., it is now possible to
8536 concatenate lists of inlines using Lua's concatenation operator `..`
8537 (requires at least one of the operants to have `List` as a metatable):
8539 function Emph (emph)
8540 local s = {pandoc.Space(), pandoc.Str 'emphasized'}
8541 return pandoc.Span(emph.content .. s)
8544 The `List` metatable is assigned to the tables which get passed to
8545 the constructors `MetaBlocks`, `MetaInline`, and `MetaList`. This
8546 enables the use of the resulting objects as lists.
8547 + `Lua/StackInstances`: push Pandoc and Meta via constructor.
8548 Pandoc and Meta elements are now pushed by calling the respective
8549 constructor functions of the pandoc Lua module. This makes serialization
8550 consistent with the way blocks and inlines are pushed to lua and allows
8551 to use List methods with the `blocks` value.
8552 + Add documentation for pandoc.List in `lua-filters.md`.
8554 * Use latest tagsoup. This fixes a bug in parsing HTML tags with
8555 `&` (but not a valid entity) following them (#4094, #4088).
8557 * Use skylighting 0.4.4.1, fixing the color of unmarked code text
8558 when `numberLines` is used (#4103).
8560 * Make `normalizeDate` more forgiving (Mauro Bieg, #4101), not
8561 requiring a leading 0 on single-digit days.
8563 * Fix `--help` output for `--highlight-style` to include `FILE` (Mauro
8566 * Clearer deprecation warning for `--latexmathml, --asciimathml, -m`.
8567 Previously we only mentioned `--latexmathml`, even if `-m` was
8570 * Changelog: fix description of lua filters in 2.0 release
8571 (Albert Krewinkel). Lua filters were initially run *after* conventional
8572 (JSON) filters. However, this was changed later to make it easier to deal
8573 with files in the mediabag. The changelog is updated to describe that
8574 feature of the 2.0 release correctly.
8576 * Change Generic JSON instances to TemplateHaskell (Jasper Van der Jeugt,
8577 #4085). This reduces compile time and memory usage significantly.
8579 * `lua-filters.md`: Added tikz filter example.
8581 * Create alternative zip file for macOS binaries.
8583 * Create alternative zip file for Windows binaries.
8585 * Update INSTALL.md since we now provide zips for binaries.
8587 * Relax `http-types` dependency (Justus Sagemüller, #4084).
8589 * Add `epub.md`, `getting-started.md` to docs. These used to live in
8592 * Add `packages` target to Makefile.
8594 * Bump bounds for binary, http-types, tasty-hunit
8596## pandoc 2.0.3 (2017-11-20)
8598 * Lua filters: preload text module (Albert Krewinkel, #4077).
8599 The `text` module is preloaded in lua. The module contains some UTF-8
8600 aware string functions, implemented in Haskell. The module is loaded on
8603 text = require 'text'
8605 s.text = text.upper(s.text)
8609 * Allow table-like access to attributes in lua filters (Albert Krewinkel,
8610 #4071). Attribute lists are represented as associative lists in Lua. Pure
8611 associative lists are awkward to work with. A metatable is attached to
8612 attribute lists, allowing to access and use the associative list as if
8613 the attributes were stored in as normal key-value pair in table.
8614 Note that this changes the way `pairs` works on attribute lists. Instead
8615 of producing integer keys and two-element tables, the resulting iterator
8616 function now returns the key and value of those pairs. Use `ipairs` to
8617 get the old behavior. Warning: the new iteration mechanism only works if
8618 pandoc has been compiled with Lua 5.2 or later (current default: 5.3).
8620 * Text.Pandoc.Parsing.uri: allow `&` and `=` as word characters (#4068).
8621 This fixes a bug where pandoc would stop parsing a URI with an
8622 empty attribute: for example, `&a=&b=` wolud stop at `a`.
8623 (The uri parser tries to guess which punctuation characters
8624 are part of the URI and which might be punctuation after it.)
8626 * Introduce `HasSyntaxExtensions` typeclass (Alexander Krotov, #4074).
8628 + Added new `HasSyntaxExtensions` typeclass for `ReaderOptions` and
8630 + Reimplemented `isEnabled` function from `Options.hs` to accept both
8631 `ReaderOptions` and `WriterOptions`.
8632 + Replaced `enabled` from `CommonMark.hs` with new `isEnabled`.
8634 * Add `amuse` extension (Alexander Krotov) to enable Amuse wiki
8635 behavior for `muse`. New `Ext_amuse` constructor for
8636 `Extension`. Note: this is switched on by default; for
8637 Emacs behavior, use `muse-amuse`.
8639 * Muse reader (Alexander Krotov):
8641 + Count only one space as part of list item marker.
8642 + Produce SoftBreaks on newlines. Now wrapping can be preserved
8643 with `--wrap=preserve`.
8644 + Add Text::Amuse footnote extensions. Footnote end is indicated by
8645 indentation, so footnotes can be placed anywhere in the text,
8646 not just at the end of it.
8647 + Accept Emacs Muse definition lists when `-amuse`.
8648 Emacs Muse does not require indentation.
8652 + Ensure we don't produce level 0 headers (#4076), even for chapter
8653 sections in epubs. This causes problems because writers aren't set
8655 + Allow spaces after `\(` and before `\)` with `tex_math_single_backslash`.
8656 Previously `\( \frac{1}{a} < \frac{1}{b} \)` was not parsed as math in
8657 `markdown` or `html` `+tex_math_single_backslash`.
8659 * MANUAL: clarify that math extensions work with HTML.
8660 Clarify that `tex_math_dollars` and `tex_math_single_backslash`
8661 will work with HTML as well as Markdown.
8663 * Creole reader: Fix performance issue for longer lists (Sascha Wilde,
8666 * RST reader: better support for 'container' directive (#4066).
8667 Create a div, incorporate name attribute and classes.
8671 + Support column specs like `*{2}{r}` (#4056). This is equivalent to
8672 `rr`. We now expand it like a macro.
8673 + Allow optional args for parbox (#4056).
8674 + Allow optional arguments on `\footnote` (#4062).
8676 * EPUB writer: Fixed path for cover image (#4069). It was previously
8677 `media/media/imagename`, and should have been `media/imagename`.
8679 * Markdown writer: fix bug with doubled footnotes in grid tables
8682 * LaTeX template: include natbib/biblatex after polyglossia (#4073).
8683 Otherwise we seem to get an error; biblatex wants polyglossia
8684 language to be defined.
8686 * Added examples to lua filters documentation.
8689## pandoc 2.0.2 (2017-11-12)
8691 * Deprecated ancient HTML math methods: `--latexmathml`, `--gladtex`,
8692 `--mimetex`, `--jsmath`.
8694 * Fixed URIs in `data/jats.csl`. They were being rendered twice,
8695 leading to invalid XML in default JATS output with pandoc-citeproc.
8697 * `lua-filters.md`: use real-world man page filter as example.
8699 * Add lua filter functions `walk_inline` and `walk_block`
8700 in the pandoc module, to apply filters inside particular
8701 inline and block elements.
8703 * Refactored some code from `Text.Pandoc.Lua.PandocModule`
8704 into new internal module `Text.Pandoc.Lua.Filter`.
8708 + Allow fenced code blocks to be indented 1-3 spaces (#4011).
8709 This brings our handling of them into alignment with CommonMark's.
8710 + Fix YAML metadata with "chomp" (`|-`). Previously if a
8711 YAML block under `|-` contained a blank line, pandoc would
8712 not parse it as metadata.
8714 * Removed `etc.` from abbreviations file. Often `etc.` ends a
8715 sentence, and we want the period to be treated as a
8716 sentence-ending period.
8718 * Fix regression with `--metadata` (#4054). Values specified with
8719 `--metadata` should replace a metadata value set in the document
8720 itself, rather than creating a list including a new value.
8724 + Fix EPUB OCF structure. #3720 had been improperly implemented.
8725 + Fix modified paths for raw HTML tags (src, poster, etc.)
8726 (#4050, #4055). This had not been updated for the new EPUB
8727 container layout, with a separate text/ subdirectory.
8728 + Fix image paths with empty `--epub-subdirectory`.
8730 * Miscellaneous code cleanup (Alexander Krotov).
8732 * Use pandoc-types 1.17.3, which adds `Walkable` instances
8733 for `[Block] Block` and `[Inline] Inline`.
8735 * Remove obsolete `stack.full.yaml` (#4052).
8737 * Change to using pandoc-citeproc 0.12.1 in binary packages.
8739 * Consolidate math output method documentation (#4049, Mauro Bieg).
8741 * `MANUAL.txt`: fix header level of "Extension: emoji" (Albert Krewinkel).
8743 * Use lua filter to generate man page from `MANUAL.txt`, replacing old
8744 Haskell filters. This is easier and faster.
8746 * Improved `INSTALL.md`.
8748 * Update commands to extract deb archive on Linux (#4043, Salim B).
8751## pandoc 2.0.1.1 (2017-11-04)
8753 * Improved fix to #3989 (parsing of HTML tags containing
8754 `>` in an attribute or comment). The previous fix (in 2.0.1) only
8755 worked in certain cases.
8757 * FB2 writer (Alexander Krotov):
8759 + Add `unrecognised` genre to `<title-info>`
8760 (Alexander Krotov). XML schema requires at least one genre.
8761 + Remove `<annotation>` from `<body>`.
8763 * CommonMark writer: fix strikethrough for `gfm` (#4038).
8765 * Use texmath 0.10, which adds support for a wider range of
8766 symbols and fixes default column alignments in MathML
8769 * Highlighting fixes, using skylighting 0.4.3.2:
8772 + Support `lineAnchors` (or `line-anchors`) in HTML code blocks.
8773 + Ensure that code lines don't get duplicate identifiers (#4031).
8774 The line identifiers are built using the code block's identifier
8775 as a prefix. If the code block has null identifier, we use
8778 * Added a few abbreviations to `data/abbreviations`,
8779 and sorted the list (#3984, Wandmalfarbe).
8781 * Improved support for columns in HTML writer (#4028).
8783 + Remove `width` attribute from the `div`.
8784 + Remove space between `<div class="column">` elements,
8785 since this prevents columns whose widths sum to 100%
8786 (the space takes up space).
8787 + Move as much as possible of the CSS to the template.
8788 + Ensure that all the HTML-based templates (including epub)
8789 contain the CSS for columns.
8790 + Columns default to 50% width unless they are given a width
8791 attribute. So if you want two equal-width columns, you
8792 can use a div with class `column` and no `width` attribute.
8794 * SelfContained: use `base64` for css links with media attribute (#4026).
8795 This fixes `--self-contained` with S5.
8797 * Improve `pandoc-template-mode.el` (Vaclav Haisman).
8799 * INSTALL.md: MacOS instructions needed xar -f (adam234).
8803 + Clarify that --setext-headers doesn't affect gfm output (#4035).
8804 + Clarify what is needed to open and close a div in `fenced_divs`
8805 (#4039, Tristano Ajmone).
8806 + Removed reference to `default.beamer` in docs (#4024).
8807 Also added mention of other templates affecting PDF output
8808 with different settings.
8811## pandoc 2.0.1 (2017-10-31)
8813 * Fixed regression in parsing of HTML comments in markdown and other
8814 non-HTML formats (`Text.Pandoc.Readers.HTML.htmlTag`) (#4019).
8815 The parser stopped at the first `>` character, even if it wasn't
8816 the end of the comment.
8818 * Creole reader (Sascha Wilde):
8820 + Fix some minor typos and formatting.
8821 + Add additional test on nowiki-block after para.
8822 + Fix lists with trailing white space.
8824 * LaTeX reader: handle `%` comment right after command.
8825 For example, `\emph%`.
8827 * Markdown reader: make sure fenced div closers work in lists.
8828 Previously the following failed:
8835 and you needed a blank line before the closing `:::`.
8837 * Make `fenced_divs` affect the Markdown writer. If `fenced_divs` is
8838 enabled, Divs will be rendered as fenced divs.
8840 * LaTeX/Beamer writer: support "blocks" inside columns and other Divs
8843 * HTML Writer: consistently use dashed class-names (Mauro Bieg, #3556).
8844 Note: this change may require some changes in CSS rules.
8845 `footnoteRef` has become `footnote-ref`, `titleslide` has
8846 become `title-slide`, and `footnoteBack` has become `footnote-back`.
8848 * JATS writer: Properly pass through author metadata (#4020).
8850 * FB2 writer (Alexander Krotov):
8852 + Write blocks outside of `<p>` in definitions.
8853 + Make bullet lists consistent with ordered lists, repeating
8854 the marker for the outer list rather than indenting sublists,
8855 since indentation does not work in readers.
8856 + Add new style FB2 tests.
8858 * `Text.Pandoc.ImageSize`: Add `Millimeter` constructor to `Dimension`
8859 (#4012) [API change]. Now sizes given in 'mm' are no longer converted
8862 * Revise documentation of small caps syntax (Andrew Dunning, #4013).
8864 * Fix broken reference links in manual (Andrew Dunning, #4014)
8866 * Fixed example of slide columns structure in changelog (#4015).
8867 Also documented this feature in MANUAL.txt.
8870## pandoc 2.0.0.1 (2017-10-30)
8874 + Fixed filepaths for nonstandard epub-subdirectory values.
8875 + Ensure that epub2 is recognized as a non-text format,
8876 so that a template is used.
8877 + Don't include "prefix" attribute for ibooks for epub2.
8878 It doesn't validate.
8879 + Fix stylesheet paths; previously we had an incorrect
8880 stylesheet path for the cover page and nav page.
8884 + Insert space when needed in macro expansion (#4007).
8885 Sometimes we need to insert a space after a control sequence
8886 to prevent it merging with a following letter.
8887 + Allow unbraced arguments for macros (#4007).
8888 + Allow body of macro definition to be unbraced (#4007).
8890 * Linux package build: ensure that pandoc-citeproc is statically linked.
8892 * trypandoc: add native, ms.
8895## pandoc 2.0 (2017-10-29)
8899 * New output format `ms` (groff ms). Complete support, including
8900 tables, math, syntax highlighting, and PDF bookmarks. The writer uses
8901 texmath's new eqn writer to convert math to eqn format, so a ms file
8902 produced with this writer should be processed with `groff -ms -e` if
8905 * New output format `jats` (Journal Article Tag Suite). This is an XML
8906 format used in archiving and publishing articles. Note that a
8907 URI-encoded CSL stylesheet (`data/jats.csl`) is added automatically
8908 unless a stylesheet is specified using `--css`.
8910 * New output format `gfm` (GitHub-flavored CommonMark) (#3841).
8911 This uses bindings to GitHub's fork of cmark, so it should parse
8912 gfm exactly as GitHub does (excepting certain postprocessing
8913 steps, involving notifications, emojis, etc.). `markdown_github`
8914 has been deprecated in favor of `gfm`.
8916 * New output format `muse` (Emacs Muse) (Alexander Krotov, #3489).
8918 * New input format `gfm` (GitHub-flavored CommonMark) (#3841).
8919 This uses bindings to GitHub's fork of cmark. `markdown_github`
8920 has been deprecated in favor of `gfm`.
8922 * New input format `muse` (Emacs Muse) reader (Alexander Krotov, #3620).
8924 * New input format `tikiwiki` (TikiWiki markup) (rlpowell, #3800).
8926 * New input format `vimwiki` (Vimwiki markup) (Yuchen Pei, #3705).
8927 Note that there is a new data file, `data/vimwiki.css`, which can
8928 be used to display the HTML produced by this reader and
8929 pandoc's HTML writer in the style of vimwiki's own HTML
8932 * New input format `creole` (Creole 1.0) (#3994, Sascha Wilde).
8934 * New syntax for Divs, with `fenced_divs` extension enabled by
8935 default (#168). This gives an attractive, plain-text way to create
8936 containers for block-level content.
8938 * Added new syntax for including raw content in any output format,
8939 enabled by the `raw_attribute` extension (which is on by default
8940 for `markdown` and `multimarkdown`). The syntax is the same as
8941 for fenced code blocks or code inlines, only with `{=FORMAT}` for
8942 attributes, where `FORMAT` is the name of the output format
8943 (e.g., `ms`, `html`).
8945 * Implement multicolumn support for slide formats (#1710).
8946 The structure expected is:
8948 :::::::::::::: {.columns}
8949 ::: {.column width="40%"}
8952 ::: {.column width="60%"}
8957 Support has been added for beamer and all HTML slide formats.
8959 * Allows line comments in templates, beginning with `$--` (#3806).
8960 (Requires doctemplates 0.2.1.)
8962 * Add `--eol=crlf|lf|native` flag and writer option to control line endings
8963 (Stefan Dresselhaus, #3663, #2097).
8965 * Add `--log` option to save log messages in JSON format to a file (#3392).
8967 * Add `--request-header` option, to set request headers when pandoc
8968 makes HTTP requests to fetch external resources. For example:
8969 `--request-header User-Agent:blah`.
8971 * Added lua filters (Albert Krewinkel, #3514). The new `--lua-filter`
8972 option works like `--filter` but takes pathnames of special lua filters
8973 and uses the lua interpreter baked into pandoc, so that no external
8974 interpreter is needed. Note that lua filters are all applied after
8975 regular filters, regardless of their position on the command line.
8976 For documentation of lua filters, see `doc/lua-filters.md`.
8978 * Set `PANDOC_READER_OPTIONS` in environment where filters are run.
8979 This contains a JSON representation of `ReaderOptions`, so filters
8982 * Support creation of pdf via groff `ms` and pdfroff.
8983 `pandoc -t ms -o output.pdf input.txt`.
8985 * Support for PDF generation via HTML and `weasyprint` or `prince`
8986 (Mauro Bieg, #3909). `pandoc -t html5 -o output.pdf --pdf-engine=prince`.
8988 * Added `--epub-subdirectory` option (#3720). This specifies the
8989 subdirectory in the OCF container that holds the EPUB specific content.
8990 We now put all EPUB related content in an `EPUB/` subdirectory by default
8991 (later this will be configurable).
8996 com.apple.ibooks.display-options.xml
8998 EPUB/ <<--configurable-->>
8999 fonts/ <<--static-->>
9001 media/ <<--static-->>
9004 styles/ <<--static-->>
9008 text/ <<--static-->>
9012 * Added `--resource-path=SEARCHPATH` command line option (#852).
9013 SEARCHPATH is separated by the usual character, depending on OS
9014 (: on unix, ; on windows). Default resource path is just working
9015 directory. However, the working directory must be explicitly
9016 specified if the `--resource-path` option is used.
9018 * Added --abbreviations=FILE option for custom abbreviations file
9019 (#256). Dfault abbreviations file (`data/abbreviations`) contains
9020 a list of strings that will be recognized by pandoc's
9021 Markdown parser as abbreviations. (A nonbreaking space will
9022 be inserted after the period, preventing a sentence space in
9023 formats like LaTeX.) Users can override the default by putting a file
9024 abbreviations in their user data directory (`~/.pandoc` on *nix).
9026 * Allow a theme file as argument to `--highlight-style`.
9027 Also include a sample, `default.theme`, in `data/`.
9029 * Allow `--syntax-definition` option for dynamic loading of syntax
9030 highlighting definitions (#3334).
9032 * Lists in `markdown` by default now use the CommonMark variable
9033 nesting rules (#3511). The indentation required for a block-level
9034 item to be included in a list item is no longer fixed, but is
9035 determined by the first line of the list item. To be included in
9036 the list item, a block must be indented to the level of the first
9037 non-space content after the list marker. Exception: if are 5 or more
9038 spaces after the list marker, then the content is interpreted as an
9039 indented code block, and continuation paragraphs must be indented
9040 two spaces beyond the end of the list marker. See the CommonMark
9041 spec for more details and examples.
9043 Documents that adhere to the four-space rule should, in most cases,
9044 be parsed the same way by the new rules. Here are some examples
9045 of texts that will be parsed differently:
9050 will be parsed as a list item with a sublist; under the four-space
9051 rule, it would be a list with two items.
9057 Here we have an indented code block under the list item, even though it
9058 is only indented six spaces from the margin, because it is four spaces
9059 past the point where a continuation paragraph could begin. With the
9060 four-space rule, this would be a regular paragraph rather than a code
9067 Here the code block will start with two spaces, whereas under
9068 the four-space rule, it would start with `code`. With the four-space
9069 rule, indented code under a list item always must be indented eight
9070 spaces from the margin, while the new rules require only that it
9071 be indented four spaces from the beginning of the first non-space
9072 text after the list marker (here, `a`).
9074 This change was motivated by a slew of bug reports from people
9075 who expected lists to work differently (#3125, #2367, #2575, #2210,
9076 #1990, #1137, #744, #172, #137, #128) and by the growing prevalance
9077 of CommonMark (now used by GitHub, for example). Those who
9078 prefer the old behavior can use `-f markdown+four_space_rule`.
9080 * Added `four_space_rule` extension. This triggers the old pandoc
9081 parsing rule for content nested under list items (the "four space
9084 * Added `spaced_reference_links` extension (#2602). It allows whitespace
9085 between the two parts of a reference link: e.g.
9091 This was previously enabled by default; now it is forbidden by default.
9093 * Add `space_in_atx_header` extension (#3512). This is enabled by default
9094 in pandoc and GitHub markdown but not the other flavors.
9095 This requirse a space between the opening #'s and the header
9096 text in ATX headers (as CommonMark does but many other implementations
9097 do not). This is desirable to avoid falsely capturing things ilke
9105 * Add `sourcefile` and `outputfile` template variables (Roland Hieber,
9108 * Allow ibooks-specific metadata in epubs (#2693). You can now have
9109 the following fields in your YAML metadata, and it will be treated
9110 appropriately in the generated EPUB:
9115 specified-fonts: false
9116 ipad-orientation-lock: portrait-only
9117 iphone-orientation-lock: landscape-only
9119 scroll-axis: vertical
9125 * Reader functions no longer presuppose that CRs have been
9126 stripped from the input. (They strip CRs themselves, before
9127 parsing, to simplify the parsers.)
9129 * Added support for translations (localization) (#3559).
9130 Currently this only affects the LaTeX reader, for things
9131 like `\figurename`. Translation data files for 46 languages
9132 can be found in `data/translations`.
9134 * Make `--ascii` work with DocBook output too.
9136 * Rename `--latex-engine` to `--pdf-engine`,
9137 and `--latex-engine-opt` to `--pdf-engine-opt`.
9139 * Removed `--parse-raw` and `readerParseRaw`. These were confusing.
9140 Now we rely on the `+raw_tex` or `+raw_html` extension with latex or html
9141 input. Thus, instead of `--parse-raw -f latex` we use `-f latex+raw_tex`,
9142 and instead of `--parse-raw -f html` we use `-f html+raw_html`.
9144 * With `--filter` R filters are now recognized, even if they are
9145 not executable (#3940, #3941, Andrie de Vries).
9147 * Support SVG in PDF output, converting with `rsvg2pdf` (#1793).
9149 * Make epub an alias for epub3, not epub2.
9151 * Removed `--epub-stylesheet`; use `--css` instead (#3472, #847).
9152 Multiple stylesheets may be used. Stylesheets will be taken both from
9153 `--css` and from the `stylesheet` metadata field (which can contain
9154 either a file path or a list of them).
9156 * `--mathml` and MathML in HTMLMathMethod no longer take an argument.
9157 The argument was for a bridge JavaScript that used to be necessary
9158 in 2004. We have removed the script already.
9160 * `--katex` improvements. The latest version is used, and the
9161 autoload script is loaded by default.
9163 * Change MathJax CDN default since old one is shutting down (#3544).
9164 Note: The new URL requires a version number, which we'll have
9165 to update manually in subsequent pandoc releases in order to
9166 take advantage of mathjax improvements.
9168 * `--self-contained`: don't incorporate elements with `data-external="1"`
9169 (#2656). You can leave an external link as it is by adding the attribute
9170 data-external="1" to the element. Pandoc will then not try to
9171 incorporate its content when `--self-contained` is used. This is
9172 similar to a feature already supported by the EPUB writer.
9174 * Allow `--extract-media` to work with non-binary input formats
9175 (#1583, #2289). If `--extract-media` is supplied with a non-binary
9176 input format, pandoc will attempt to extract the contents of all
9177 linked images, whether in local files, data: uris, or external uris.
9178 They will be named based on the sha1 hash of the contents.
9180 * Make `papersize: a4` work regardless of the case of `a4`.
9181 It is converted to `a4` in LaTeX and `A4` in ConTeXt.
9183 * Make `east_asian_line_breaks` affect all readers/writers (#3703).
9185 * Underlined elements are now treated consistently by readers
9186 (#2270, hftf); they are always put in a Span with class `underline`.
9187 This allows the user to treat them differently from other emphasis,
9188 using a filter. Docx, Org, Textile, Txt2Tags, and HTML readers
9191 * Improved behavior of `auto_identifiers` when there are explicit ids
9192 (#1745). Previously only autogenerated ids were added to the list
9193 of header identifiers in state, so explicit ids weren't taken
9194 into account when generating unique identifiers. Duplicated
9195 identifiers could result. This simple fix ensures that explicitly given
9196 identifiers are also taken into account.
9198 * Use `table-of-contents` for contents of toc, make `toc` a boolean
9199 (#2872). Changed markdown, rtf, and HTML-based templates accordingly.
9200 This allows you to set `toc: true` in the metadata; this
9201 previously produced strange results in some output formats.
9202 For backwards compatibility, `toc` is still set to the
9203 toc contents. But it is recommended that you update templates
9204 to use `table-of-contents` for the toc contents and `toc`
9207 * Change behavior with binary format output to stdout.
9208 Previously, for binary formats, output to stdout was disabled
9209 unless we could detect that the output was being piped (and not
9210 sent to the terminal). Unfortunately, such detection is not
9211 possible on Windows, leaving windows users no way to pipe binary
9212 output. So we have changed the behavior in the following way:
9214 + Output to stdout is allowed when it can be determined that
9215 the output is being piped (on non-Windows platforms).
9216 + If the `-o` option is not used, binary output is never sent
9217 to stdout by default; instead, an error is raised.
9218 + If `-o -` is used, binary output is sent to stdout, regardless
9219 of whether it is being piped. This works on Windows too.
9221 * Better error behavior: uses of `error` have been replaced by
9222 raising of `PandocError`, which can be trapped and handled by the
9225 * Removed `hard_line_breaks` extension from `markdown_github` (#3594).
9226 GitHub has two Markdown modes, one for long-form documents like READMEs
9227 and one for short things like issue comments. In issue comments, a line
9228 break is treated as a hard line break. In README, wikis, etc., it is
9229 treated as a space as in regular Markdown. Since pandoc is more likely to
9230 be used to convert long-form documents from GitHub Markdown,
9231 `-hard_line_breaks` is a better default.
9233 * Include `backtick_code_blocks` extension in `mardkown_mmd` (#3637).
9235 * Escape `MetaString` values (as added with `-M/--metadata` flag) (#3792).
9236 Previously they would be transmitted to the template without any
9237 escaping. Note that `--M title='*foo*'` yields a different result from
9243 In the latter case, we have emphasis; in the former case, just
9244 a string with literal asterisks (which will be escaped
9245 in formats, like Markdown, that require it).
9247 * Allow `em`, `cm`, `in` for image height/width in HTML, LaTeX (#3450).
9249 * HTML writer: Insert `data-` in front of unsupported attributes. Thus,
9250 a span with attribute `foo` gets written to HTML5 with `data-foo`, so
9251 it is valid HTML5. HTML4 is not affected. This will allow us to use
9252 custom attributes in pandoc without producing invalid HTML. (With help
9253 from Wandmalfarbe, #3817.)
9255 * Plain writer: improved super/subscript rendering. We now
9256 handle more non-digit characters for which there are
9257 sub/superscripted unicode characters. When unicode
9258 sub/superscripted characters are not available, we use
9259 `_(..)` or `^(..)` (#3518).
9261 * Docbook, JATS, TEI writers: print INFO message when omitting interior
9262 header (#3750). This only applies to section headers inside list items,
9263 e.g., which were otherwise silently omitted.
9265 * Change to `--reference-links` in Markdown writer (#3701). With
9266 `--reference-location` of `section` or `block`, pandoc will now repeat
9267 references that have been used in earlier sections. The Markdown
9268 reader has also been modified, so that *exactly* repeated references
9269 do not generate a warning, only references with the same label but
9270 different targets. The idea is that, with references after every block,
9271 one might want to repeat references sometimes.
9273 * ODT/OpenDocument writer:
9275 + Support `lang` attribute (#1667).
9276 + Added support for `--toc` (#2836). Thanks to @anayrat.
9280 + `lang` meta, see #1667 (Mauro Bieg, #3515).
9281 + Change `FigureWithCaption` to `CaptionedFigure` (iandol, #3658).
9282 + Use `Table` rather than `Table Normal` for table style (#3275).
9283 `Table Normal` is the default table style and can't be modified.
9284 + Pass through comments (#2994). We assume that comments are defined as
9285 parsed by the docx reader:
9287 I want <span class="comment-start" id="0" author="Jesse Rosenthal"
9288 date="2016-05-09T16:13:00Z">I left a comment.</span>some text to
9289 have a comment <span class="comment-end" id="0"></span>on it.
9291 We assume also that the id attributes are unique and properly
9292 matched between comment-start and comment-end.
9293 + Bookmark improvements. Bookmark start/end now surrounds content rather
9294 than preceding it. Bookmarks generated for Div with id
9295 (jgm/pandoc-citeproc#205).
9296 + Add `keywords` metadata to docx document properties (Ian).
9298 * RST writer: support unknown interpreted text roles by
9299 parsing them as `Span` with `role` attributes (#3407). This
9300 way they can be manipulated in the AST.
9304 + Line block: Use class instead of style attribute (#1623). We now
9305 issue `<div class="line-block">` and include a default definition
9306 for `line-block` in the default templates, instead of hard-coding a
9308 + Add class `footnoteBack` to footnote back references (Timm Albers).
9309 This allows for easier CSS styling.
9310 + Render SmallCaps as span with smallcaps class (#1592), rather than
9311 using a style attribute directly. This gives the user more flexibility
9312 in styling small caps in CSS.
9313 + With reveal.js we use `data-src` instead of `src` for images for
9315 + Special-case `.stretch` class for images in reveal.js (#1291).
9316 Now in reveal.js, an image with class `stretch` in a paragraph
9317 by itself will stretch to fill the whole screen, with no
9318 caption or figure environment.
9320 * Added warnings for non-rendered blocks to writers.
9322 * Writers now raise an error on template failure.
9324 * When creating a PDF via LaTeX, warn if the font is missing some
9327 * Remove initial check for PDF-creating program (#3819).
9328 Instead, just try running it and raise the exception if it
9329 isn't found at that point. This improves things for users of Cygwin
9330 on Windows, where the executable won't be found by `findExecutable`
9331 unless `.exe` is added. The same exception is raised as before, but
9334 * Readers issue warning for duplicate header identifiers (#1745).
9335 Autogenerated header identifiers are given suffixes so as not to clash
9336 with previously used header identifiers. But they may still coincide with
9337 an explicit identifier that is given for a header later in the document,
9338 or with an identifier on a div, span, link, or image. We now issue
9339 a warning in this case, so users can supply an explicit identifier.
9341 * CommonMark reader now supports `emoji`, `hard_line_breaks`, `smart`,
9342 and `raw_html` extensions.
9346 + Don't allow backslash + newline to affect block structure (#3730).
9347 Note that as a result of this change, the following, which formerly
9348 produced a header with two lines separated by a line break, will
9349 now produce a header followed by a paragraph:
9354 This may affect some existing documents that relied on
9355 this undocumented and unintended behavior. This change makes pandoc
9356 more consistent with other Markdown implementations, and with itself
9357 (since the two-space version of a line break doesn't work inside ATX
9358 headers, and neither version works inside Setext headers).
9361 * Org reader (Albert Krewinkel, unless noted):
9363 + Support `table.el` tables (#3314).
9364 + Support macros (#3401).
9365 + Support the `#+INCLUDE:` file inclusion mechanism (#3510).
9366 Recognized include types are `example`, `export`, `src`, and
9367 normal org file inclusion. Advanced features like line numbers
9368 and level selection are not implemented yet.
9369 + Interpret more meta value as inlines. The values of the following
9370 meta variables are now interpreted using org-markup instead of
9371 treating them as pure strings: `keywords` (comma-separated list of
9372 inlines), `subtitle` (inline values), `nocite` (inline values, can
9374 + Support `\n` export option (#3940). This turns all newlines in the
9375 text into hard linebreaks.
9379 + Improved admonition support (#223). We no longer add an
9380 `admonition` class, we just use the class for the type of admonition,
9381 `note` for example. We put the word corresponding to the label in
9382 a paragraph inside a `Div` at the beginning of the admonition with
9383 class `admonition-title`. This is about as close as we can get to
9385 + Initial support of `.. table` directive. This allows adding captions
9387 + Support `.. line-block` directive. This is deprecated but may still
9388 be in older documents.
9389 + Support scale and align attributes of images (#2662).
9390 + Implemented implicit internal header links (#3475).
9391 + Support RST-style citations (#853). The citations appear at the end
9392 of the document as a definition list in a special div with id
9393 `citations`. Citations link to the definitions.
9394 + Recurse into bodies of unknown directives (#3432).
9395 In most cases it's better to preserve the content than
9396 to emit it. This isn't guaranteed to have good results;
9397 it will fail spectacularly for unknown raw or verbatim directives.
9398 + Handle chained link definitions (#262). For example,
9401 .. _goodbye: example.com
9403 Here both `hello` and `goodbye` should link to `example.com`.
9404 + Support anchors (#262). E.g.
9412 This is supported by putting "paragraph" in a `Div` with id `hello`.
9413 + Support `:widths:` attribute for table directive.
9414 + Implement csv-table directive (#3533). Most attributes are supported,
9415 including `:file:` and `:url:`.
9416 + Support unknown interpreted text roles by parsing them as Span
9417 with "role" attributes (#3407). This way they can be manipulated in
9420 * HTML reader: parse a span with class `smallcaps` as `SmallCaps`.
9424 + Implemented `\graphicspath` (#736).
9425 + Properly handle column prefixes/suffixes. For example, in
9426 `\begin{tabular}{>{$}l<{$}>{$}l<{$} >{$}l<{$}}`
9427 each cell will be interpreted as if it has a `$`
9428 before its content and a `$` after (math mode).
9429 + Handle komascript `\dedication` (#1845). It now adds a
9430 `dedication` field to metadata. It is up to the user to supply
9431 a template that uses this variable.
9432 + Support all `\textXX` commands, where XX = `rm`, `tt`, `up`, `md`,
9433 `sf`, `bf` (#3488). Spans with a class are used when there is
9435 + Expand `\newenvironment` macros (#987).
9436 + Add support for LaTeX subfiles package (Marc Schreiber, #3530).
9437 + Better support for subfigure package (#3577).
9438 A figure with two subfigures turns into two pandoc
9439 figures; the subcaptions are used and the main caption
9440 ignored, unless there are no subcaptions.
9441 + Add support for `\vdots` (Marc Schreiber, #3607).
9442 + Add basic support for hyphenat package (Marc Schreiber, #3603).
9443 + Add basic `\textcolor` support (Marc Schreiber).
9444 + Add support for `tabularx` environment (Marc Schreiber, #3632).
9445 + Better handling of comments inside math environments (#3113).
9446 This solves a problem with commented out `\end{eqnarray}` inside
9447 an eqnarray (among other things).
9448 + Parse tikzpicture as raw verbatim environment if `raw_tex` extension
9449 is selected (#3692). Otherwise skip with a warning. This is better
9450 than trying to parse it as text!
9451 + Add `\colorbox` support (Marc Schreiber).
9452 + Set identifiers on Spans used for `\label`.
9453 + Have `\setmainlanguage` set `lang` in metadata.
9454 + Support etoolbox's `\ifstrequal`.
9455 + Support `plainbreak`, `fancybreak` et al from the memoir class
9456 (bucklereed, #3833).
9457 + Support `\let`. Also, fix regular macros so they're expanded at the
9458 point of use, and NOT also the point of definition. `\let` macros,
9459 by contrast, are expanded at the point of definition. Added an
9460 `ExpansionPoint` field to `Macro` to track this difference.
9461 + Support simple `\def` macros. Note that we still don't support
9462 macros with fancy parameter delimiters, like `\def\foo#1..#2{...}`.
9463 + Support `\chaptername`, `\partname`, `\abstractname`, etc. (#3559,
9465 + Put content of `\ref`, `\label`, `\eqref` commands into `Span` with
9466 attributes, so they can be handled in filters (Marc Schreiber, #3639)
9467 + Add Support for `glossaries` and `acronym` package (Marc Schreiber,
9468 #3589). Acronyms are not resolved by the reader, but acronym and
9469 glossary information is put into attributes on Spans so that they
9470 can be processed in filters.
9471 + Use `Link` instead of `Span` for `\ref`. This makes more sense
9472 semantically and avoids unnecessary `Span [Link]` nestings when
9473 references are resolved.
9474 + Rudimentary support for `\hyperlink`.
9475 + Support `\textquoteleft|right`, `\textquotedblleft|right` (#3849).
9476 + Support `\lq`, `\rq`.
9477 + Implement `\newtoggle`, `\iftoggle`, `\toggletrue|false` from etoolbox
9479 + Support `\RN` and `\Rn`, from biblatex (bucklereed, #3854).
9480 + Improved support for `\hyperlink`, `\hypertarget` (#2549).
9481 + Support `\k` ogonek accent.
9482 + Improve handling of accents. Handle ogonek, and fall back correctly
9483 with forms like `\"{}`.
9484 + Better support for ogonek accents.
9485 + Support for `\faCheck` and `\faClose` (Marc Schreiber, #3727).
9486 + Support for `xspace` (Marc Schreiber, #3797).
9487 + Support `\setmainlanguage` or `\setdefaultlanguage` (polyglossia)
9489 + Better handling of `\part` in LaTeX (#1905). Now we parse chapters as
9490 level 0 headers, and parts as level -1 headers. After parsing, we
9491 check for the lowest header level, and if it's less than 1 we bump
9492 everything up so that 1 is the lowest header level. So `\part` will
9493 always produce a header; no command-line options are needed.
9494 + Add block version of `\textcolor` (Marc Schreiber).
9495 + `\textcolor` works as inline and block command (Marc Schreiber).
9496 + `\textcolor` will be parse as span at the beginning of a paragraph
9498 + Read polyglossia/babel `\text(LANG){...}` (bucklereed)
9499 + Improved handling of include files in LaTeX reader (#3971).
9500 Previously `\include` wouldn't work if the included file
9501 contained, e.g., a begin without a matching end.
9502 + Support `\expandafter` (#3983).
9503 + Handle `\DeclareRobustCommand` (#3983). Currently it's just treated
9504 as a synonym for `\newcommand`.
9505 + Handle `\lettrine` (Mauro Bieg).
9507 * Math improvements due to updates in texmath:
9509 + Improved handling of accents and upper/lower delimiters.
9510 + Support for output in GNU eqn format (used with *roff).
9511 + Allow `\boldsymbol` + a token without braces, and similarly
9512 with other styling commands.
9513 + Improve parsing of `\mathop` to allow multi-character operator names.
9514 + Add thin space after math operators when "faking it with
9517 * `walk` is now used instead of `bottomUp` in the `ToJSONFilter`
9518 instance for `a -> [a]` (pandoc-types). Note that behavior
9519 will be slightly different, since `bottomUp`'s treatment of
9520 a function `[a] -> [a]` is to apply it to each sublist of a
9521 list, while walk applies it only to maximal sublists.
9522 Usually the latter behavior is what is wanted, and the
9523 former can be simulated when needed. But there may be
9524 existing filters that need to be rewritten in light of the
9525 new behavior. Performance should be improved.
9527 * There are some changes to syntax highlighting due to revisions
9528 in the `skylighting` library:
9530 + Support for `powershell` has been added, and many syntax
9531 definitions have been updated.
9532 + Background colors have been added to the `kate` style.
9533 + The way highlighted code blocks are formatted in HTML has
9534 been changed (David Baynard), in ways that may require
9535 changes in hard-coded CSS affecting highlighting.
9536 (If you haven't included hard-coded highlighting CSS in
9537 your template, you needn't change anything.)
9542 * New module `Text.Pandoc.Class` (Jesse Rosenthal, John MacFarlane).
9543 This contains definitions of the `PandocMonad` typeclass, the
9544 `PandocIO` and `PandocPure` monads, and associated functions.
9546 * Changed types of all writers and readers.
9548 + We now use `Text` instead of `String` in the interface (#3731).
9549 (We have not yet changed the internals of most readers to work
9550 with `Text`, but making this change in the API now opens up a
9551 path to doing that.)
9552 + The result is now of form `m a` with constraint `PandocMonad m`.
9553 Readers and writers can be combined to form monadic values which
9554 can be run using either `runIO` or `runPure`. If `runIO` is used,
9555 then both readers and writers will be able to do IO when needed
9556 (for include files, for example); if `runPure` is used,
9557 then the functions are pure and will not touch IO.
9558 + Where previously you used
9559 `writeRST def (readMarkdown def "[foo](url)")`, now you
9561 `runPure $ readMarkdown def (pack "[foo](url)") >>= writeRST def`.
9563 * New module `Text.Pandoc.Readers` (Albert Krewinkel). This
9564 contains reader helper functions formerly defined in the
9565 top-level `Text.Pandoc` module.
9567 + Changed `StringReader` -> `TextReader`.
9568 + `getReader` now returns a pair of a reader and
9569 `Extensions`, instead of building the extensions into the
9570 reader (#3659). The calling code must explicitly set
9571 `readerExtensions` using the `Extensions` returned. The
9572 point of the change is to make it possible for the calling
9573 code to determine what extensions are being used.
9575 * New module `Text.Pandoc.Writers` (Albert Krewinkel).
9576 This contains writer helper functions formerly defined in the
9577 top-level `Text.Pandoc` module.
9579 + Changed `StringWriter` -> `TextWriter`.
9580 + `getWriter` now returns a pair of a reader and
9581 `Extensions`, instead of building the extensions into the
9582 reader (#3659). The calling code must explicitly set
9583 `readerExtensions` using the `Extensions` returned. The
9584 point of the change is to make it possible for the calling
9585 code to determine what extensions are being used.
9587 * New module `Text.Pandoc.Lua`, exporting `runLuaFilter` (Albert Krewinkel,
9590 * New module `Text.Pandoc.App`. This abstracts out the functionality
9591 of the command line program (`convertWithOpts`), so it can be reproduced
9592 e.g. in a desktop or web application. Instead of exiting, we throw errors
9593 (#3548), which are caught (leading to exit) in pandoc.hs, but allow other
9594 users of `Text.Pandoc.App` to recover. `pandoc.hs` is now a 2-liner.
9595 The module also exports some utility functions for parsing options
9596 and running filters.
9598 * New module `Text.Pandoc.Logging` (exported module) (#3392).
9599 This now contains the `Verbosity` definition previously in
9600 `Text.Pandoc.Options`, as well as a new `LogMessage` datatype that will
9601 eventually be used instead of raw strings for warnings. This will enable
9602 us, among other things, to provide machine-readable warnings if desired.
9603 Include ToJSON instance and showLogMessage. This gives us the possibility
9604 of both machine-readable and human-readable output for log messages.
9606 * New module `Text.Pandoc.BCP47`, with `getLang`, `Lang(..)`, `parseBCP47`.
9608 * New module `Text.Pandoc.Translations`, exporting `Term`,
9609 `Translations`, `readTranslations`.
9611 * New module `Text.Pandoc.Readers.LaTeX.Types', exporting `Macro`, `Tok`,
9612 `TokType`, `Line`, `Column`.
9614 * `Text.Pandoc.Error`: added many new constructors for `PandocError`.
9616 * Expose some previously private modules (#3260). These are often
9617 helpful to people writing their own reader or writer modules:
9619 + `Text.Pandoc.Writers.Shared`
9620 + `Text.Pandoc.Parsing`
9621 + `Text.Pandoc.Asciify`
9622 + `Text.Pandoc.Emoji`
9623 + `Text.Pandoc.ImageSize`
9624 + `Text.Pandoc.Highlighting`
9626 * New module `Text.Pandoc.Extensions` (Albert Krewinkel):
9627 Extension parsing and processing functions were defined in the top-level
9628 `Text.Pandoc` module. These functions are moved to the Extensions
9629 submodule as to enable reuse in other submodules.
9631 * Add `Ext_raw_attribute` constructor for `Extension`.
9633 * Add `Ext_fenced_divs` constructor for `Extension'.
9635 * Add `Ext_four_space_rule` constructor in `Extension`.
9637 * Add `Ext_gfm_auto_identifiers` constructor for `Extension`.
9639 * Add `Monoid` instance for `Extensions`.
9641 * Add `Text.Pandoc.Writers.Ms`, exporting `writeMs`.
9643 * Add `Text.Pandoc.Writers.JATS`, exporting `writeJATS`.
9645 * Add `Text.Pandoc.Writers.Muse`, exporting `writeMuse`.
9647 * Add `Text.Pandoc.Readers.Muse`, exporting `readMuse`.
9649 * Add `Text.Pandoc.Readers.TikiWiki`, exporting `readTikiWiki`.
9651 * Add `Text.Pandoc.Readers.Vimwiki`, exporting `readVimwiki`.
9653 * Add `Text.Pandoc.Readers.Creole`, exporting `readCreole`.
9655 * Export `setVerbosity` from `Text.Pandoc`.
9657 * `Text.Pandoc.Pretty`: Add `Eq` instance for `Doc`.
9659 * `Text.Pandoc.XML`: `toEntities`: changed type to `Text -> Text`.
9661 * `Text.Pandoc.UTF8`:
9663 + Export `fromText`, `fromTextLazy`, `toText`, `toTextLazy`.
9664 Define `toString`, `toStringLazy` in terms of them.
9665 + Add new functions parameterized on `Newline`: `writeFileWith`,
9666 `putStrWith`, `putStrLnWith`, `hPutStrWith`, `hPutStrLnWith`.
9668 * `Text.Pandoc.MediaBag`: removed `extractMediaBag`.
9670 * `Text.Pandoc.Highlighting`:
9672 + `highlighting` now returns an Either rather than Maybe.
9673 This allows us to display error information returned by the skylighting
9674 library. Display a warning if the highlighting library throws an error.
9675 + Add parameter for `SyntaxMap` to `highlight`.
9677 * `Text.Pandoc.Writers.Math`:
9679 + Export `defaultMathJaxURL`, `defaultKaTeXURL`. This will ensure that
9680 we only need to update these in one place.
9682 * `Text.Pandoc.SelfContained`:
9684 + Removed `WriterOptions` parameter from `makeSelfContained`.
9685 + Put `makeSelfContained` in PandocMonad instead of IO. This removes
9686 the need to pass MediaBag around and improves exceptions. It also
9687 opens up the possibility of using makeSelfContained purely.
9688 + Export `makeDataURI`.
9690 * `Text.Pandoc.ImageSize`:
9692 + Export `lengthToDim`, new function `scaleDimension`.
9693 + Export `inEm` from ImageSize (#3450).
9694 + Change `showFl` and `show` instance for `Dimension` so
9695 extra decimal places are omitted.
9696 + Added `Em` as a constructor of `Dimension`.
9697 + Add `WriterOptions` parameter to `imageSize` signature (Mauro Bieg).
9699 * `Text.Pandoc.Templates`:
9701 + Change type of `renderTemplate'`. Now it runs in `PandocMonad`
9702 and raises a proper `PandocTemplateError` if there are problems, rather
9703 than failing with uncatchable `error`.
9704 + Change signature of `getDefaultTemplate`. Now it runs in any instance
9705 of `PandocMonad`, and returns a `String` rather than an `Either` value.
9706 And it no longer takes a `datadir` parameter, since this can be
9707 retrieved from `CommonState`.
9709 * `Text.Pandoc.Options`:
9711 + Added `writerEpubSubdirectory` to `WriterOptions` (#3720).
9712 The EPUB writer now takes its EPUB subdirectory from this option.
9713 + In `WriterOptions`, rename `writerLaTeXEngine` to `writerPdfEngine`
9714 and `writerLaTeXArgs` to `writerPdfArgs` (Mauro Bieg, #3909).
9715 + Add `writerSyntaxMap` to `WriterOptions`.
9716 + Removed `writerEpubStylesheet` from `WriterOptions`.
9717 + Remove `writerUserDataDir` from `WriterOptions`. It is now carried
9718 in `CommonState` in `PandocMonad` instances. (And thus it can be used
9720 + Changed `writerEpubMetadata` to a `Maybe String`.
9721 + Removed `readerApplyMacros` from `ReaderOptions`. Now we just check
9722 the `latex_macros` reader extension.
9723 + FromJSON/ToJSON instances for `ReaderOptions`.
9724 + In `HTMLMathMethod`, the `KaTeX` contsructor now takes only
9725 one string (for the KaTeX base URL), rather than two.
9726 + Removed `writerSourceURL` from `WriterOptions`. We now use
9727 `stSourceURL` in `CommonState`, which is set by `setInputFiles`.
9729 * `Text.Pandoc.Shared`:
9731 + `tabFilter` now takes a `Text`, not `String`.
9732 + `openURL`: Changed type from an Either. Now it will just raise
9733 an exception to be trapped later.
9734 + Remove `normalizeSpaces` (#1530).
9735 + Remove `warn`. (Use `report` from `Text.Pandoc.Class` instead.)
9736 + Export a new function `crFilter`.
9737 + Add `eastAsianLineBreakFilter` (previously in Markdown reader).
9738 + Provide custom `isURI` that rejects unknown schemes.
9739 (Albert Krewinkel, #2713). We also export the set of known
9740 `schemes`. The new function replaces the function of the same name
9741 from `Network.URI`, as the latter did not check whether a scheme is
9742 well-known. All official IANA schemes (as of 2017-05-22) are
9743 included in the set of known schemes. The four non-official schemes
9744 `doi`, `isbn`, `javascript`, and `pmid` are kept.
9746 + Remove `readDataFile`, `readDefaultDataFile`, `getReferenceDocx`,
9747 `getReferenceODT`. These now live in `Text.Pandoc.Class`,
9748 where they are defined in terms of `PandocMonad`
9749 primitives and have different signatures.
9750 + Remove `openURL`. Use `openURL` from `Text.Pandoc.Class` instead.
9751 + Add `underlineSpan`.
9753 * `Text.Pandoc.Readers.HTML`: export new `NamedTag` class.
9755 * `Text.Pandoc.Readers.Markdown`: remove `readDocxWithWarnings`.
9756 With the new API one can simply use `getLog` after running
9759 * `Text.Pandoc.Readers.LaTeX`: Changed types for `rawLaTeXInline`
9760 and `rawLaTeXBlock`. (Both now return a `String`, and they are
9761 polymorphic in state.)
9764### Bug fixes and under-the-hood improvements
9766 * TEI writer: Added identifiers on `<div>` elements.
9768 * DokuWiki reader: Better handling for code block in list item (#3824).
9770 * Custom writer: Remove old preprocesesor conditionals (Albert Krewinkel).
9772 * ZimWiki writer: Removed internal formatting from note and table cells,
9773 because ZimWiki does not support it (Alex Ivkin, #3446).
9777 + Updated list of syntax highlighting languages (#3461).
9778 Now `r` gets you `<source>` rather than `<code>` (among others).
9779 + Add display attribute on `<math>` tags (#3452). This allows display
9780 math to be rendered properly.
9781 + Remove newline before `</ref>` (#2652).
9782 + Don't softbreak lines inside list items (#3531).
9786 + Reduce to two spaces after bullets (#3417, Albert Krewinkel).
9787 + Add unit tests (Alexander Krotov).
9788 + Stop using raw HTML to wrap divs (Albert Krewinkel, #3771).
9789 + Do not strip `#` from Org anchor links (Alexander Krotov).
9791 * CommonMark writer:
9793 + Avoid excess blank lines at end of output.
9794 + Prefer pipe tables to HTML tables even if it means losing relative
9795 column width information (#3734).
9796 + Support table, strikethrough extensions, when enabled (as with gfm).
9797 Note that we bypass the commonmark writer from cmark and construct our
9798 own pipe tables, with better results.
9799 + Properly support `--wrap=none`.
9800 + Use smallcaps class for `SmallCaps` (#1592).
9801 + Omit "fig:" prefix in image titles. This is used internally to
9802 indicate internal figures.
9806 + Properly handle table captions.
9807 + Don't wrap lines in in definition list terms. Wrapping is not allowed.
9808 + Implemented `+/-smart` and improved escaping with `+smart`.
9809 + Add empty comments when needed to avoid including a blockquote
9810 in the indented content of a preceding block (#3675).
9811 + Improve grid table output, fix bug with empty rows (#3516).
9812 Uses the new `gridTable` in Writers.Shared, which is here
9813 improved to better handle 0-width cells.
9814 + Remove space at beginning/end of RST code span (#3496). Otherwise
9815 we get invalid RST. There seems to be no way to escape the space.
9816 + Add header anchors when header has non-standard id (#3937).
9817 + Correctly handle inline code containing backticks, using a `:literal:`
9819 + Don't backslash-escape word-internal punctuation (#3978).
9823 + Don't include variables in metadata blocks. Previously variables set
9824 on the command line were included in e.g. YAML metadata, contrary to
9825 documentation and intentions.
9826 + Improved escaping with `+smart`.
9827 + Fixed grid tables embedded in grid tables (#2834).
9828 + Use span with class 'smallcaps' for SmallCaps, instead of a style
9829 attribute as before (#1592).
9830 + Escape initial `%` in a paragraph if the `pandoc_title_blocks`
9831 extension is enabled (#3454). Otherwise in a document starting with
9832 a literal `%` the first line is wrongly interpreted as a title.
9833 + Fixed false ordered lists in YAML metadata (#3492, #1685). Now we
9834 properly escape things that would otherwise start ordered lists,
9840 + Better handling of tables with empty columns (#3337). We now
9841 calculate the number of columns based on the longest row (or the
9842 length of aligns or widths).
9843 + Escape unordered list markers at beginning of paragraph (#3497), to
9844 avoid false interpretation as a list.
9845 + Escape `|` appropriately.
9846 + Ensure space before list at top level (#3487).
9847 + Avoid spurious blanklines at end of document after tables and list,
9849 + Fixed bugs in simple/multiline list output (#3384).
9850 Previously we got overlong lists with `--wrap=none`. This is fixed.
9851 Previously a multiline list could become a simple list (and would
9852 always become one with `--wrap=none`).
9853 + Don't emit a simple table if `simple_tables` disabled (#3529).
9854 + Case-insensitive reference links (David A Roberts, #3616).
9855 Ensure that we do not generate reference links whose labels differ only
9856 by case. Also allow implicit reference links when the link
9857 text and label are identical up to case.
9858 + Put space before reference link definitions (Mauro Bieg, #3630).
9859 + Better escaping for links (David A. Roberts, #3619). Previously the
9860 Markdown writer would sometimes create links where there were none
9861 in the source. This is now avoided by selectively escaping bracket
9862 characters when they occur in a place where a link might be created.
9863 + Added missing `\n` (David A. Roberts, #3647).
9864 + Fixed duplicated reference links with `--reference-links`
9865 and `--reference-location=section` (#3674). Also ensure that there
9866 are no empty link references `[]`.
9867 + Avoid inline surround-marking with empty content (#3715).
9868 E.g. we don't want `<strong></strong>` to become `****`.
9869 Similarly for emphasis, super/subscript, strikeout.
9870 + Don't allow soft break in header (#3736).
9871 + Make sure `plain`, `markdown_github`, etc. work for raw.
9872 Previously only `markdown` worked. Note: currently a raw block labeled
9873 `markdown_github` will be printed for any `markdown` format.
9874 + Ensure that `+` and `-` are escaped properly so they don't cause
9875 spurious lists (#3773). Previously they were only
9876 if succeeded by a space, not if they were at end of line.
9877 + Use pipe tables if `raw_html` disabled and `pipe_tables` enabled,
9878 even if the table has relative width information (#3734).
9879 + Markdown writer: don't crash on `Str ""`.
9880 + Make `Span` with null attribute transparent. That is, we don't use
9881 brackets or `<span>` tags to mark spans when there are no attributes;
9882 we simply output the contents.
9883 + Escape pipe characters when `pipe_tables` enabled (#3887).
9884 + Better escaping of `<` and `>`. If `all_symbols_escapable` is set,
9885 we backslash escape these. Otherwise we use entities as before.
9886 + When writing plain, don't use ` ` to separate list and indented
9887 code. There's no need for it in this context, since this isn't to be
9888 interpreted using Markdown rules.
9889 + Preserve classes in JS obfuscated links (Timm Albers, #2989).
9890 HTML links containing classes originally now preserve them when using
9891 javascript email obfuscation.
9892 + Render `SmallCaps` as a native span when `native_spans` are enabled.
9893 + Always write attributes with `bracketed_spans` (d-dorazio).
9897 + Fix handling of nested font commands (#3568). Previously pandoc emitted
9898 incorrect markup for bold + italic, for example, or bold + code.
9899 + Avoid error for definition lists with no definitions (#3832).
9903 + Fix internal links with `writerIdentifierPrefix opt`
9904 (#3397, Mauro Bieg).
9908 + Don't include bookmarks on headers unless non-null id (#3476).
9909 + Support 9 levels of headers (#1642).
9910 + Allow 9 list levels (#3519).
9911 + Don't take `distArchive` from datadir (#3322). The docx writer takes
9912 components from the distribution's version of `reference.docx` when it
9913 can't find them in a user's custom `reference.docx`. Previously, we
9914 allowed a `reference.docx` in the data directory (e.g. `~/.pandoc`)
9915 to be used as the distribution's reference.docx. This led to a
9916 bizarre situation where pandoc would produce a good docx using
9917 `--template ~/.pandoc/ref.docx`, but if `ref.docx` were moved to
9918 `~/.pandoc/reference.docx`, it would then produce a corrupted docx.
9919 + Fixed handling of soft hyphen (0173) (#3691).
9920 + Better handling of keywords (#3719).
9921 + Cleaner code for handling dir and style attributes for `Div`.
9922 + Use `Set` for dynamic styles to avoid duplicates.
9923 + Removed redundant element from data/docx/word/numbering.xml.
9924 The elements we need are generated when the document is
9925 compiled; this didn't do anything.
9926 + Activate `evenAndOddHeaders` from reference docx (#3901,
9927 Augustín Martín Barbero).
9929 * ODT/OpenDocument writer:
9931 + Calculate aspect ratio for percentage-sized images (Mauro Bieg, #3239).
9932 + Use more widely available bullet characters (#1400). The old
9933 characters weren't available in some font sets. These seem to work
9934 well on Windows and Linux versions of LibreOffice.
9935 + Wider labels for lists (#2421). This avoids overly narrow labels for
9936 ordered lists with `()` delimiters. However, arguably it creates
9937 overly wide labels for bullets. Also, lists now start flush with
9938 the margin, rather than indented.
9939 + Fixed dropped elements in some ordered lists (#2434).
9943 + Don't render `RawBlock` as code.
9944 + Don't fail with an error on interior headers (e.g. in list) (#3750).
9945 Instead, omit them with an INFO message.
9946 + Add support for "lang" metadata (Alexander Krotov, #3625).
9947 + Format `LineBlock` as poem (Alexander Krotov). Previously writer
9948 produced one paragraph with `<empty-line/>` elements, which are not
9949 allowed inside `<p>` according to FB2 schema.
9950 + Replace `concatMap` with `cMap` (Alexander Krotov).
9951 + Write FB2 lists without nesting blocks inside `<p>` (Alexander
9956 + Make sure `html4`, `html5` formats work for raw blocks/inlines.
9957 + Render raw inline environments when `--mathjax` used (#3816).
9958 We previously did this only with raw blocks, on the assumption
9959 that math environments would always be raw blocks. This has changed
9960 since we now parse them as inline environments.
9961 + Ensure we don't get two style attributes for width and height.
9962 + Report when not rendering raw inline/block.
9963 + Issue warning if no title specified and template used (#3473).
9964 + Info message if `lang` is unspecified (#3486).
9965 + Removed unused parameter in `dimensionsToAttributeList`.
9966 + Avoid two class attributes when adding `uri` class (#3716).
9967 + Fix internal links with `writerIdentifierPrefix opt` (#3397, Mauro
9969 + Use revealjs's math plugin for mathjax (#3743). This is a thin
9970 wrapper around mathjax that makes math look better on revealjs.
9971 + Slidy: use h1 for all slides, even if they were originally
9972 level 2 headers (#3566). Otherwise the built-in table of contents
9977 + Don't render LaTeX images with data: URIs (#3636). Note that
9978 `--extract-media` can be used when the input contains data: URIs.
9979 + Make highlighted code blocks work in footnotes (Timm Albers).
9980 + Don't use figure inside table cell (#3836).
9981 + Use proper code for list enumerators (#3891). This should fix problems
9982 with lists that don't use arabic numerals.
9983 + Always add hypertarget when there's a non-empty identifier (#2719).
9984 Previously the hypertargets were only added when there was actually
9985 a link to that identifier.
9986 + Use `%` after hypertarget before code block.
9987 + Add `\leavevmode` before hypertarget at start of paragraph (#2704,
9988 fixes formatting problems in beamer citations).
9989 + Don't use `lstinline` in `\item[..]` (#645). If you do, the contents
9990 of item disappear or are misplaced. Use `\texttt` instead.
9991 + Fix problem with escaping in `lstinline` (#1629). Previously the
9992 LaTeX writer created invalid LaTeX when `--listings` was specified and
9993 a code span occurred inside emphasis or another construction.
9994 + Fix error with line breaks after empty content (#2874). LaTeX
9995 requires something before a line break, so we insert a `~` if no
9996 printable content has yet been emitted.
9998 + Fixed detection of otherlangs (#3770). We weren't recursing into
10000 + Handle language in inline code with `--listings` (#3422).
10001 + Write euro symbol directly in LaTeX (Andrew Dunning, #3801).
10002 The textcomp package allows pdfLaTeX to parse `€` directly, making the
10003 `\euro` command unneeded.
10004 + Fixed footnotes in table captions (#2378). Note that if the table has
10005 a first page header and a continuation page header, the notes will
10006 appear only on the first occurrence of the header.
10007 + In `writeBeamer` output, allow hyperlinks to frames (#3220).
10008 Previously you could link to a header above or below slide level but
10009 not *to* slide level. This commit changes that. Hypertargets are
10010 inserted inside frame titles; technically the reference is to just
10011 after the title, but in normal use (where slides are viewed full
10012 screen in a slide show), this does not matter.
10013 + Remove `\strut` at beginning of table cells (#3436). This fixes a
10014 problem with alignment of lists in table cells. The `\strut` at the
10015 end seems to be enough to avoid the too-close spacing that motivated
10016 addition of the strut in #1573.
10017 + Add partial siunitx Support (Marc Schreiber, #3588).
10021 + Refactored to use BCP47 module.
10022 + Remove unnecessary `$` (Alexander Krotov, #3482).
10023 + Use header identifiers for chapters (#3968).
10027 + `title_page.xhtml` is now put in `text/`.
10028 + Don't strip formatting in TOC (#1611).
10032 + Fix bug for certain links in table cells (#3667).
10033 + Allow 'pre' code in list item (#3916).
10037 + Added warnings for ignored material (#3392).
10038 + Better sanity checks to avoid parsing unintended things as
10039 raw HTML in the Markdown reader (#3257).
10040 + Revise treatment of `li` with `id` attribute (#3596). Previously we
10041 always added an empty div before the list item, but this created
10042 problems with spacing in tight lists. Now we do this: If the list
10043 item contents begin with a `Plain` block, we modify the `Plain`
10044 block by adding a `Span` around its contents. Otherwise, we add a
10045 `Div` around the contents of the list item (instead of adding an
10046 empty `Div` to the beginning, as before).
10047 + Add `details` tag to list of block tags (#3694).
10048 + Removed `button` from block tag list (#3717). It is already in the
10049 `eitherBlockOrInlineTag` list, and should be both places.
10050 + Use `Set`s instead of lists for block tag lookup.
10051 + Rewrote to use `Text` throughout. Effect on memory usage is modest
10053 + Use the lang value of `<html>` to set the lang meta value (bucklereed,
10055 + Ensure that paragraphs are closed properly when the parent block
10056 element closes, even without `</p>` (#3794).
10057 + Parse `<figure>` and `<figcaption>` (Mauro Bieg, #3813).
10058 + Parse `<main>` like `<div role=main>` (bucklereed, #3791).
10059 `<main>` closes `<p>` and behaves like a block element generally
10060 + Support column alignments (#1881). These can be set either
10061 with a `width` attribute or with `text-width` in a `style` attribute.
10062 + Modified state type to be an instance of `HasLogMessages`, so
10063 `registerHeader` can issue warnings.
10064 + `</td>` or `</th>` should close any open block tag (#3991).
10065 + `<td>` should close an open `<th>` or `<td>`.
10066 + `htmlTag` improvements (#3989). We previously failed on cases
10067 where an attribute contained a `>` character. This patch fixes the
10068 bug, which especially affects raw HTML in Markdown.
10072 + Newline is not indentation (Alexander Krotov).
10074 * MediaWiki reader:
10076 + Allow extra hyphens after `|-` in tables (#2649).
10077 + Allow blank line after table start (#2649).
10078 + Fixed more table issues (#2649).
10079 + Ensure that list starts begin at left margin (#2606). Including when
10080 they're in tables or other list items.
10081 + Make smart double quotes depend on `smart` extension (#3585).
10082 + Don't do curly quotes inside `<tt>` contexts (#3585). Even if `+smart`.
10083 + Modified state type to be an instance of `HasLogMessages`, so
10084 `registerHeader` can issue warnings.
10086 * TWiki reader (Alexander Krotov):
10088 + Remove unnecessary `$` (#3597).
10089 + Simplify `linkText` (#3605).
10093 + Minor refactoring, avoiding explicit MediaBag handling.
10094 This all works behind the scenes in CommonState plumbing.
10098 + Don't drop smartTag contents (#2242).
10099 + Handle local namespace declarations (#3365). Previously we didn't
10100 recognize math, for example, when the xmlns declaration occurred on
10101 the element and not the root.
10102 + More efficient trimSps (#1530). Replacing `trimLineBreaks`. This
10103 does the work of `normalizeSpaces` as well, so we avoid the need for
10104 that function here.
10105 + Avoid 0-level headers (Jesse Rosenthal, #3830). We used to parse
10106 paragraphs styled with "HeadingN" as "nth-level header." But if a
10107 document has a custom style named "Heading0", this will produce a
10108 0-level header, which shouldn't exist. We only parse this style
10109 if N>0. Otherwise we treat it as a normal style name, and
10110 follow its dependencies, if any.
10111 + Add tests for avoiding zero-level header (Jesse Rosenthal).
10115 + Replaced `collectRights` with Rights from `Data.Either`.
10116 + Remove dead code (Albert Krewinkel).
10118 * Org reader (Albert Krewinkel, unless noted).
10120 + Don't allow tables inside list items (John MacFarlane, #3499).
10121 + Disallow tables on list marker lines (#3499).
10122 + Convert markup at beginning of footnotes (John MacFarlane, #3576).
10123 + Allow emphasized text to be followed by `[` (#3577).
10124 + Handle line numbering switch for src blocks.
10125 The line-numbering switch that can be given to source blocks (`-n` with
10126 an start number as an optional parameter) is parsed and translated to a
10127 class/key-value combination used by highlighting and other readers and
10129 + Stop adding rundoc prefix to src params. Source block parameter names
10130 are no longer prefixed with `rundoc`. This was intended to simplify
10131 working with the rundoc project, a babel runner. However, the rundoc
10132 project is unmaintained, and adding those markers is not the reader's
10133 job anyway. The original language that is specified for a source
10134 element is now retained as the `data-org-language` attribute and only
10135 added if it differs from the translated language.
10136 + Allow multi-word arguments to src block params (#3477). The reader now
10137 correctly parses src block parameter list even if parameter arguments
10138 contain multiple words.
10139 + Avoid creating `nullMeta` by applying `setMeta` directly
10140 (Alexander Krotov).
10141 + Replace `sequence . map` with `mapM`.
10142 + Fix smart parsing behavior. Parsing of smart quotes and special
10143 characters can either be enabled via the `smart` language extension or
10144 the `'` and `-` export options. Smart parsing is active if either the
10145 extension or export option is enabled. Only smart parsing of special
10146 characters (like ellipses and en and em dashes) is enabled by default,
10147 while smart quotes are disabled. Previously, all smart parsing was
10148 disabled unless the language extension was enabled.
10149 + Subject full doc tree to headline transformations (Albert Krewinkel,
10150 #3695). Emacs parses org documents into a tree structure, which is
10151 then post-processed during exporting. The reader is changed to do the
10152 same, turning the document into a single tree of headlines starting
10154 + Fix cite parsing behaviour (Herwig Stuetz). Until now, `org-ref`
10155 cite keys included special characters also at the end. This caused
10156 problems when citations occur right before colons or at the end of
10157 a sentence. With this change, all non alphanumeric characters at
10158 the end of a cite key are ignored. This also adds `,` to the list
10159 of special characters that are legal in cite keys to better mirror
10160 the behaviour of org-export.
10161 + Fix module names in haddock comments. Copy-pasting had lead to
10162 haddock module descriptions containing the wrong module names.
10163 + Recognize babel result blocks with attributes (#3706). Babel
10164 result blocks can have block attributes like captions and names.
10165 Result blocks with attributes were not recognized and were parsed
10166 as normal blocks without attributes.
10167 + Include tags in headlines. The Emacs default is to include tags in the
10168 headline when exporting. Instead of just empty spans, which contain the
10169 tag name as attribute, tags are rendered as small caps and wrapped in
10170 those spans. Non-breaking spaces serve as separators for multiple tags.
10171 + Respect export option for tags (#3713). Tags are appended to
10172 headlines by default, but will be omitted when the `tags` export option
10174 + Use `tag-name` attribute instead of `data-tag-name`.
10175 + Use `org-language` attribute rather than `data-org-language`.
10176 + Modified state type to be an instance of `HasLogMessages`, so
10177 `registerHeader` can issue warnings.
10178 + End footnotes after two blank lines. Footnotes can not only be
10179 terminated by the start of a new footnote or a header, but also by two
10180 consecutive blank lines.
10181 + Update emphasis border chars (#3933). The org reader was updated to
10182 match current org-mode behavior: the set of characters which are
10183 acceptable to occur as the first or last character in an org emphasis
10184 have been changed and now allows all non-whitespace chars at the
10185 inner border of emphasized text (see `org-emphasis-regexp-components`).
10189 + Fixed small bug in list parsing (#3432). Previously the parser didn't
10190 handle properly this case:
10196 + Handle multiline cells in simple tables (#1166).
10197 + Parse list table directive (Keiichiro Shikano, #3432).
10198 + Make use of `anyLineNewline` (Alexander Krotov, #3686).
10199 + Use `anyLineNewline` in `rawListItem` (Alexander Krotov, #3702).
10200 + Reorganize block parsers for ~20% faster parsing.
10201 + Fixed `..include::` directive (#3880).
10202 + Handle blank lines correctly in line blocks (Alexander Krotov, #3881).
10203 Previously pandoc would sometimes combine two line blocks separated
10204 by blanks, and ignore trailing blank lines within the line block.
10205 + Fix indirect hyperlink targets (#512).
10209 + Allow attributes in reference links to start on next line (#3674).
10210 + Parse YAML metadata in a context that sees footnotes defined in
10211 the body of the document (#1279).
10212 + When splitting pipe table cells, skip tex math (#3481).
10213 You might have a `|` character inside math. (Or for that matter
10214 something that the parser might mistake for raw HTML.)
10215 + Treat span with class `smallcaps` as SmallCaps.
10216 This allows users to specify small caps in Markdown this way:
10217 `[my text]{.smallcaps}` (#1592).
10218 + Fixed internal header links (#2397).
10219 This patch also adds `shortcut_reference_links` to the list
10221 + Treat certain environments as inline
10222 when they occur without space surrounding them (#3309, #2171).
10223 E.g. equation, math. This avoids incorrect vertical space
10225 + Optimized `nonindentSpaces`. Makes the benchmark go from 40 to 36 ms.
10226 + Allow latex macro definitions indented 1-3 spaces.
10227 Previously they only worked if nonindented.
10228 + Improved parsing of indented raw HTML blocks (#1841).
10229 Previously we inadvertently interpreted indented HTML as
10230 code blocks. This was a regression. We now seek to determine the
10231 indentation level of the contents of an HTML block, and (optionally)
10232 skip that much indentation. As a side effect, indentation may be
10233 stripped off of raw HTML blocks, if `markdown_in_html_blocks` is
10234 used. This is better than having things interpreted as indented
10236 + Fixed smart quotes after emphasis (#2228). E.g. in `*foo*'s 'foo'`.
10237 + Warn for notes defined but not used (#1718).
10238 + Use `anyLineNewline` (Alexander Krotov).
10239 + Interpret YAML metadata as Inlines when possible (#3755). If
10240 the metadata field is all on one line, we try to interpret it as
10241 Inlines, and only try parsing as Blocks if that fails. If it
10242 extends over one line (including possibly the `|` or `>` character
10243 signaling an indented block), then we parse as Blocks. This was
10244 motivated by some German users finding that `date: '22. Juin 2017'`
10245 got parsed as an ordered list.
10246 + Fixed spurious parsing as citation as reference def (#3840).
10247 We now disallow reference keys starting with `@` if the
10248 `citations` extension is enabled.
10249 + Parse `-@roe` as suppress-author citation (pandoc-citeproc#237).
10250 Previously only `[-@roe]` (with brackets) was recognized as
10251 suppress-author, and `-@roe` was treated the same as `@roe`.
10252 + Fixed parsing of fenced code after list when there is no intervening
10253 blank line (#3733).
10254 + Allow raw latex commands starting with `\start` (#3558). Previously
10255 these weren't allowed because they were interpreted as starting
10256 ConTeXt environments, even without a corresponding `\stop`...
10257 + Added `inlines`, `inlines1`.
10258 + Require nonempty alt text for `implicit_figures` (#2844).
10259 A figure with an empty caption doesn't make sense.
10260 + Removed texmath macro material; now all this is handled
10261 in the LaTeX reader functions.
10262 + Fixed bug with indented code following raw LaTeX (#3947).
10266 + Rewrote LaTeX reader with proper tokenization (#1390,
10267 #2118, #3236, #3779, #934, #982). This rewrite is primarily
10268 motivated by the need to get macros working properly. A side benefit
10269 is that the reader is significantly faster. We now tokenize the
10270 input text, then parse the token stream. Macros modify the token
10271 stream, so they should now be effective in any context, including
10272 math. Thus, we no longer need the clunky macro processing
10273 capacities of texmath.
10274 + Parse `\,` to `\8198` (six-per-em space) (Henri Werth).
10275 + Allow `\newcommand\foo{blah}` without braces.
10276 + Support `\lstinputlisting` (#2116).
10277 + Issue warnings when skipping unknown latex commands (#3392).
10278 + Include contents of `\parbox`.
10279 + Allow `\hspace` and `\vspace` to count as raw block or inline.
10280 Previously we would refuse to parse anything as raw inline if
10281 it was in the `blockCommands` list. Now we allow exceptions
10282 if they're listed under ignoreInlines in inlineCommands.
10283 This should make it easier e.g. to include an `\hspace`
10284 between two side-by-side raw LaTeX tables.
10285 + Don't drop contents of `\hypertarget`.
10286 + Handle spaces before `\cite` arguments.
10287 + Allow newpage, clearpage, pagebreak in inline contexts as well as
10288 block contexts (#3494).
10289 + Treat `{{xxx}}` the same as `{xxx}` (#2115).
10290 + Use `pMacroDefinition` in macro (for more direct parsing).
10291 Note that this means that `macro` will now parse one
10292 macro at a time, rather than parsing a whole group together.
10293 + Fixed failures on `\ref{}`, `\label{}` with `+raw_tex`. Now these
10294 commands are parsed as raw if `+raw_tex`; otherwise, their argument
10295 is parsed as a bracketed string.
10296 + Don't crash on empty `enumerate` environment (#3707).
10297 + Handle escaped `&` inside table cell (#3708).
10298 + Handle block structure inside table cells (#3709). `minipage` is no
10300 + Handle some width specifiers on table columns (#3709). Currently
10301 we only handle the form `0.9\linewidth`. Anything else would have
10302 to be converted to a percentage, using some kind arbitrary assumptions
10304 + Make sure `\write18` is parsed as raw LaTeX. The change is in the
10305 LaTeX reader's treatment of raw commands, but it also affects the
10307 + Fixed regression with starred environment names (#3803).
10308 + Handle optional args in raw `\titleformat` (#3804).
10309 + Improved heuristic for raw block/inline. An unknown command at the
10310 beginning of the line that could be either block or inline is
10311 treated as block if we have a sequence of block commands followed by
10312 a newline or a `\startXXX` command (which might start a raw ConTeXt
10314 + Don't remove macro definitions from the output, even if
10315 `Ext_latex_macros` is set, so that macros will be applied.
10316 Since they're only applied to math in Markdown, removing the macros
10317 can have bad effects. Even for math macros, keeping them should be
10319 + Removed `macro`. It is no longer necessary, since the
10320 `rawLaTeXBlock` parser will parse macro definitions. This also avoids
10321 the need for a separate `latexMacro` parser in the Markdown reader.
10322 + Use `label` instead of `data-label` for label in caption (#3639).
10323 + Fixed space after `\figurename` etc.
10324 + Resolve references to section numbers.
10325 + Fix `\let\a=0` case, with single character token.
10326 + Allow `@` as a letter in control sequences. `@` is commonly used
10327 in macros using `\makeatletter`. Ideally we'd make the tokenizer
10328 sensitive to `\makeatletter` and `\makeatother`, but until then this
10329 seems a good change.
10330 + Track header numbers and correlate with labels.
10331 + Allow `]` inside group in option brackets (#3857).
10332 + lstinline with braces can be used (verb cannot be used with braces)
10333 (Marc Schreiber, #3535).
10334 + Fix keyval function: pandoc did not parse options in braces correctly
10335 (Marc Schreiber, #3642).
10336 + When parsing raw LaTeX commands, include trailing space (#1773).
10337 Otherwise things like `\noindent foo` break and turn into
10338 `\noindentfoo`. Affects `-f latex+raw_tex` and `-f markdown` (and other
10339 formats that allow `raw_tex`).
10340 + Don't treat "..." as Quoted (#3958). This caused quotes to be omitted in
10341 `\texttt` contexts.
10342 + Add tests for existing `\includegraphics` behaviour (Ben Firshman).
10343 + Allow space before `=` in bracketd options (Ben Firshman).
10344 + Be more forgiving in parsing command options. This was needed, for
10345 example, to make some minted options work.
10346 + Strip off quotes in `\include` filenames.
10348 * Added `Text.Pandoc.CSV`, simple (unexported) CSV parser.
10350 * `Text.Pandoc.PDF`:
10352 + Got `--resource-path` working with PDF output (#852).
10353 + Fetch images when generating PDF via context (#3380).
10354 To do this, we create the temp directory as a subdirectory
10355 of the working directory. Since context mk IV by default looks
10356 for images in the parent directory, this works.
10357 + Use `report` instead of `warn`, make it sensitive to verbosity settings.
10358 + Use `fillMediaBag` and `extractMedia` to extract media to temp dir.
10359 This reduces code duplication.
10360 + `html2pdf`: use stdin instead of intermediate HTML file
10361 + Removed useless `TEXINPUTS` stuff for `context2pdf`. mkiv context
10362 doesn't use `TEXINPUTS`.
10364 * `Text.Pandoc.Pretty`:
10366 + Simplified definition of `realLength`.
10367 + Don't error for blocks of size < 1. Instead, resize to 1 (see #1785).
10369 * `Text.Pandoc.MIME`:
10371 + Use `application/javascript` (not `application/x-javascript`).
10372 + Added `emf` to mimeTypes with type `application/x-msmetafile` (#1713).
10374 * `Text.Pandoc.ImageSize`:
10376 + Improve SVG image size code (Marc Schreiber, #3580).
10377 + Make `imageSize` recognize basic SVG dimensions (Mauro Bieg, #3462).
10379 * Use `Control.Monad.State.Strict` throughout. This gives 20-30% speedup
10380 and reduction of memory usage in most of the writers.
10382 * Use `foldrWithKey` instead of deprecated `foldWithKey`.
10384 * `Text.Pandoc.SelfContained`:
10386 + Fixed problem with embedded fonts (#3629).
10387 + Refactored getData from `getDataURI` in `SelfContained`.
10388 + Don't use data URIs for script or style (#3423). Instead, just use
10389 script or style tags with the content inside. The old method with
10390 data URIs prevents certain optimizations outside pandoc. Exception:
10391 data URIs are still used when a script contains `</script>` or a
10392 style contains `</`.
10393 + SelfContained: Handle URL inside material retrieved from a URL
10394 (#3629). This can happen e.g. with an @import of a google web font.
10395 (What is imported is some CSS which contains an url reference
10396 to the font itself.) Also, allow unescaped pipe (|) in URL.
10397 + Load resources from `data-src` (needed for lazy loading in
10398 reveal.js slide shows).
10399 + Handle `data-background-image` attribute on section (#3979).
10401 * `Text.Pandoc.Parsing`:
10403 + Added `indentWith` (Alexander Krotov, #3687).
10404 + Added `stateCitations` to `ParserState`.
10405 + Removed `stateChapters` from `ParserState`.
10406 + In `ParserState`, make `stateNotes'` a Map, add `stateNoteRefs`.
10407 + Added `gobbleSpaces` and `gobbleAtMostSpaces`.
10408 + Adjusted type of `insertIncludedFile` so it can be used with token
10410 + Replace old texmath macro stuff from Parsing. Use Macro from
10411 Text.Pandoc.Readers.LaTeX.Types instead.
10412 + Export `insertIncludedFile`.
10413 + Added `HasLogMessages`, `logMessage`, `reportLogMessages` (#3447).
10414 + Replace partial with total function (Albert Krewinkel).
10415 + Introduce `HasIncludeFiles` type class (Albert Krewinkel). The
10416 `insertIncludeFile` function is generalized to work with all parser
10417 states which are instances of that class.
10418 + Add `insertIncludedFilesF` which returns F blocks (Albert Krewinkel).
10419 The `insertIncludeFiles` function was generalized and renamed
10420 to `insertIncludedFiles'`; the specialized versions are based on that.
10421 + `many1Till`: Check for the end condition before parsing (Herwig
10422 Stuetz). By not checking for the end condition before the first
10423 parse, the parser was applied too often, consuming too much of the
10424 input. This only affects `many1Till p end` where `p` matches on a
10426 + Provide `parseFromString` (#3690). This is a version of
10427 `parseFromString` specialied to ParserState, which resets
10428 `stateLastStrPos` at the end. This is almost always what we want.
10429 This fixes a bug where `_hi_` wasn't treated as emphasis in the
10430 following, because pandoc got confused about the position of the
10431 last word: `- [o] _hi_`.
10432 + Added `takeP`, `takeWhileP` for efficient parsing of `[Char]`.
10433 + Fix `blanklines` documentation (Alexander Krotov, #3843).
10434 + Give less misleading line information with `parseWithString`.
10435 Previously positions would be reported past the end of the chunk.
10436 We now reset the source position within the chunk and report
10437 positions "in chunk."
10438 + Add `anyLineNewline` (Alexander Krotov).
10439 + Provide shared F monad functions for Markdown and Org readers
10440 (Albert Krewinkel). The `F` monads used for delayed evaluation
10441 of certain values in the Markdown and Org readers are based on a
10442 shared data type capturing the common pattern of both `F` types.
10443 + Add `returnF` (Alexander Krotov).
10444 + Avoid parsing `Notes:**` as a bare URI (#3570). This avoids parsing
10445 bare URIs that start with a scheme + colon + `*`, `_`, or `]`.
10446 + Added `readerAbbreviations` to `ParserState`. Markdown reader
10447 now consults this to determine what is an abbreviation.
10448 + Combine grid table parsers (Albert Krewinkel, #3638). The grid table
10449 parsers for markdown and rst was combined into one single
10450 parser `gridTable`, slightly changing parsing behavior of both
10451 parsers: (1) The markdown parser now compactifies block content
10452 cell-wise: pure text blocks in cells are now treated as paragraphs
10453 only if the cell contains multiple paragraphs, and as plain blocks
10454 otherwise. Before, this was true only for single-column tables. (2)
10455 The rst parser now accepts newlines and multiple blocks in header
10457 + Generalize tableWith, gridTableWith (Albert Krewinkel).
10458 The parsing functions `tableWith` and `gridTableWith` are generalized
10459 to work with more parsers. The parser state only has to be an
10460 instance of the `HasOptions` class instead of requiring a concrete
10461 type. Block parsers are required to return blocks wrapped into a
10462 monad, as this makes it possible to use parsers returning results
10463 wrapped in `Future`s.
10465 * `Text.Pandoc.Shared`:
10467 + Simplify `toRomanNumeral` using guards (Alexander Krotov, #3445)
10468 + `stringify`: handle Quoted better (#3958). Previously we were losing
10469 the quotation marks in Quoted elements.
10471 * `Text.Pandoc.Writers.Shared`:
10473 + Export `metaToJSON'`, `addVariablesToJSON` (#3439).
10474 This allows us to add the variables AFTER using the metadata
10475 to generate a YAML header (in the Markdown writer).
10476 + Added `unsmartify` (previously in RST writer).
10477 Undo literal double curly quotes. Previously we left these.
10478 + Generalize type of `metaToJSON` so it can take a Text. Previously a
10479 String was needed as argument; now any ToJSON instance will do.
10480 + Added `gridTable` (previously in Markdown writer).
10481 + `gridTable`: Refactored to use widths in chars.
10482 + `gridTable`: remove unnecessary extra space in cells.
10483 + Fixed `addVariablesToJSON`. It was previously not allowing multiple
10484 values to become lists.
10485 + Pipe tables: impose minimum cell size (see #3526).
10488### Default template changes
10490 * HTML templates (including EPUB and HTML slide show templates):
10492 + Make default.html5 polyglot markup conformant (John Luke Bentley,
10493 #3473). Polyglot markup is HTML5 that is also valid XHTML. See
10494 <https://www.w3.org/TR/html-polyglot>. With this change, pandoc's
10495 html5 writer creates HTML that is both valid HTML5 and valid XHTML.
10496 + Regularized CSS in html/epub/html slide templates (#3485).
10497 All templates now include `code{white-space: pre-wrap}`
10498 and CSS for `q` if `--html-q-tags` is used. Previously some templates
10499 had `pre` and others `pre-wrap`; the `q` styles were only sometimes
10501 + CSS for `.smallcaps`, (Mauro Bieg, #1592)
10502 + `default.revealjs`: make `history` default to true.
10503 + `default.revealjs`: use lazy loading (#2283).
10504 + `default.revealjs`: add `mathjax` variable and some conditional code
10505 to use the MathJaX plugin.
10506 + `default.slidy` uses `https` instead of `http` (ickc, #3848).
10507 + `default.dzslides`: Load Google Font using HTTPS by default
10510 * DocBook5 template: Use `lang` and `subtitle` variables (Jens Getreu,
10513 * LaTeX/Beamer template:
10515 + Combine LaTeX/Beamer templates (Andrew Dunning, #3878).
10516 `default.beamer` has been removed; beamer now uses the
10517 `default.latex` template. Beamer-specific parts are conditional
10518 on the `beamer` variable set by the writer. Note that
10519 `pandoc -D beamer` will return this (combined) template.
10520 + Use `xcolor` for `colorlinks` option (Andrew Dunning, #3877).
10521 Beamer loads `xcolor` rather than `color`, and thus the
10522 `dvipsnames` option doesn't take effect. This also provides a wider
10523 range of colour selections with the `svgnames` option.
10524 + Use starred versions of `xcolor` names (Andrew Dunning).
10525 Prevents changes to documents defined using the `dvipsnames` list (e.g.
10526 `Blue` gives a different result with svgnames enabled).
10527 + Load `polyglossia` after header-includes (#3898). It needs to be
10528 loaded as late as possible.
10529 + Use `unicode-math` (Vaclav Haisman). Use `mathspec` with only
10530 XeLaTeX on request.
10531 + Don't load `fontspec` before `unicode-math` (over there).
10532 The `unicode-math` package loads `fontspec` so explicit loading of
10533 `fontspec` before `unicode-math` is not necessary.
10534 + Use `unicode-math` by default in default.latex template. mathspec will
10535 be used in xelatex if the `mathspec` variable is set; otherwise
10536 unicode-math will be used (Václav Haisman).
10537 + Use `dvipsnames` options when `colorlinks` specified (otherwise
10538 we get an error for `maroon`) (Thomas Hodgson).
10539 + Added beamer `titlegraphic` and `logo` variables (Thomas Hodgson).
10540 + Fix typo in fix for notes in tables (#2378, zeeMonkeez).
10541 + Fix `hyperref` options clash (Andrew Dunning, #3847) Avoids an options
10542 clash when loading a package (e.g. `tufte-latex`) that uses
10543 `hyperref` settings different from those in the template.
10544 + Add `natbiboptions` variable (#3768).
10545 + Fix links inside captions in LaTeX output with links-as-notes
10546 (Václav Haisman, #3651). Declare our redefined `\href` robust.
10547 + Load `parskip` before `hyperref` (Václav Haisman, #3654).
10548 + Allow setting Japanese fonts when using LuaLaTeX (Václav Haisman,
10549 #3873). by using the `luatexja-fontspec` and `luatexja-preset`
10550 packages. Use existing `CJKmainfont` and `CJKoptions` template
10551 variables. Add `luatexjafontspecoptions` for `luatexja-fontspec`
10552 and `luatexjapresetoptions` for `luatexja-preset`.
10553 + Added `aspectratio` variable to beamer template (Václav Haisman,
10555 + Modified template.latex to fix XeLaTex being used with tables
10556 (lwolfsonkin, #3661). Reordered `lang` variable handling to
10557 immediately before `bidi`.
10559 * ConTeXt template: Improved font handling: `simplefonts` is now
10560 obsolete in ConTeXt (Pablo Rodríguez).
10563### Documentation improvements
10567 + Add URL for Prince HTML > PDF engine (Ian, #3919).
10568 + Document that content above slide-level will be omitted in
10569 slide shows. See #3460, #2265.
10570 + Explain `--webtex` SVG url (Mauro Bieg, #3471)
10571 + Small clarification in YAML metadata section.
10572 + Document that html4 is technically XHTML 1.0 transitional.
10573 + Remove refs to highlighting-kate (#3672).
10574 + Document ibooks specific epub metadata.
10575 + Clarify that mathml is used for ODT math.
10576 + Mention limitations of Literate Haskell Support (#3410,
10578 + Add documentation of limitations of grid tables (Stephen
10580 + Clarify that meta-json contains transformed values (Jakob Voß,
10581 #3491) Make clear that template variable `meta-json` does not
10582 contain plain text values or JSON output format but field values
10583 transformed to the selected output format.
10587 + Clarify that templates are dual-licensed.
10588 + Clarify that pandoc-types is BSD3 licensed.
10589 + List new files not written by jgm (Albert Krewinkel).
10590 + Update dates in copyright notices (Albert Krewinkel). This follows
10591 the suggestions given by the FSF for GPL licensed software.
10592 <https://www.gnu.org/prep/maintain/html_node/Copyright-Notices.html>
10596 + Improved instructions for tests with patterns.
10597 + Put RPM-based distros on separate point (Mauro Bieg, #3449)
10601 + Fixed typos (Wandmalfarbe, #3479).
10602 + Add "ask on pandoc-discuss" (Mauro Bieg).
10604 * Add lua filter documentation in `doc/lua-filters.md`. Note that the
10605 end of this document is autogenerated from `data/pandoc.lua`
10606 using `make doc/lua-filters.md`, which uses `tools/ldoc.ltp`
10607 (Albert Krewinkel).
10609 * Add `doc/filters.md`. This is the old scripting tutorial from
10612 * Add `doc/using-the-pandoc-api.md` (#3289). This gives an introduction
10613 to using pandoc as a Haskell library.
10616### Build infrastructure improvements
10618 * Removed `data/templates` submodule. Templates are now a subtree
10619 in `data/templates`. This removes the need to do `git submodule
10622 * Renamed `tests` -> `test`.
10624 * Remove `https` flag. Always build with HTTPS support.
10626 * Use `file-embed` instead of `hsb2hs` to embed data files when
10627 `embed_data_files` flag is set. `file-embed` gives us better dependency
10628 tracking: if a data file changes, ghc/stack/cabal know to recompile
10629 the Data module. This also removes `hsb2hs` as a build dependency.
10631 * Add `custom-setup` stanza to pandoc, lowercase field names.
10633 * Add `static` Cabal flag.
10635 * Name change OSX -> MacOS. Add a -MacOS suffix to mac package rather
10636 than -OSX. Changed local names from osx to macos.
10638 * make_macos_package.sh - Use strip to reduce executable size.
10640 * Revised binary linux package. Now a completely static executable
10641 is created, using Docker and alpine. We create both a deb and a
10642 tarball. The old `deb` directory has been replaced with a `linux`
10643 directory. Running `make` in the `linux` directory should
10644 perform the build, putting the binary packages in `artifacts/`.
10646 * `linux/control.in`: add `Replaces:`, so existing pandoc-citeproc and
10647 pandoc-data packages will be uninstalled; this package provides
10648 both (#3822). Add latex packages as 'suggested', update
10651 * Remove cpphs build requirement -- it is no longer needed.
10653 * Replaced `{deb,macos,windows}/stack.yaml` with `stack.pkg.yaml`.
10655 * Name change OSX -> macOS (ickc, #3869).
10657 * Fix casing of Linux, UNIX, and Windows (ickc).
10659 * `.travis.yml`: create a source dist and do cabal build and test there.
10660 That way we catch errors due to files missing from the data
10661 section of pandoc.cabal.
10665 + Split `make haddock` from `make full`.
10666 + Add BRANCH variable for winpkg.
10667 + Add `lint` target.
10668 + Improve `make full`. Disable optimizations.
10669 Build everything, inc. trypandoc and benchmarks. Use parallel build.
10670 + Allow `make test` to take `TESTARGS`.
10672 * Added new command tests (`Tests.Command`), using small text files
10673 in `test/command/`. Any files added in this directory will be treated
10674 as shell tests (see smart.md for an example). This makes it very easy
10675 to add regression tests etc.
10677 * Test fixes so we can find data files. In old tests & command tests,
10678 we now set the environment variable `pandoc_datadir`. In lua tests,
10679 we set the datadir explicitly.
10681 * Refactored `compareOutput` in docx writer test.
10683 * Consolidated some common functions in `Tests.Helper`.
10685 * Small change to unbalanced bracket test to speed up test suite.
10687 * Speed up Native writer quickcheck tests.
10689 * Use tasty for tests rather than test-framework.
10691 * Add simple Emacs mode to help with Pandoc templates editing.
10692 (Václav Haisman, #3889). `tools/pandoc-template-mode.el`
10695## pandoc 1.19.2.4 (2017-09-10)
10697 * Add dependencies on texmath and skylighting to the executable.
10698 This is needed for dependency version numbers to be available,
10701## pandoc 1.19.2.3 (2017-09-09)
10703 * Add CPP to Setup.hs so it works with Cabal >= 2 and < 2.
10705## pandoc 1.19.2.2 (2017-09-08)
10707 * Fix build with GHC 8.2.1 (#3876, Peter Simons). Setup.hs does not
10708 compile with Cabal 2.x, so we require an earlier version via
10709 setup-depends. The following packages need newer versions with
10710 GHC 8.2.1 and had their constraints relaxed accordingly:
10711 executable-path, process, syb, and time.
10713## pandoc 1.19.2.1 (2017-01-31)
10715 * Require skylighting >= 0.1.1.4.
10716 * Adjust test output for skylighting version.
10717 * Relax upper bounds on blaze-html and blaze-markup.
10719## pandoc 1.19.2 (2017-01-29)
10721 * Use skylighting library instead of highlighting-kate for syntax
10722 highlighting. Skylighting is faster and more accurate (#3363).
10723 Later we'll be able to add features like warning messages, dynamic
10724 loading of xml syntax definitions, and dynamic loading of themes.
10726 * Added a new highlight style, `breezeDark`.
10728 * Text.Pandoc.Highlighting: Update list of `listings` languages (#3374).
10729 This allows more languages to be used when using the `--listings`
10732 * OpenDocument writer:
10734 + Small refactoring. Removed separate 'parent' parameter in paraStyle.
10735 + Don't profilerate text styles unnecessarily (#3371).
10736 This change makes the writer create only as many temporary
10737 text styles as are absolutely necessary. It also consolidates
10738 adjacent nodes with the same style.
10740 * Org reader (Albert Krewinkel):
10742 + Allow short hand for single-line raw blocks (Albert Krewinkel,
10743 #3366). Single-line raw blocks can be given via `#+FORMAT: raw line`,
10744 where `FORMAT` must be one of `latex`, `beamer`, `html`, or `texinfo`.
10745 + Accept org-ref citations followed by commas (Albert Krewinkel).
10746 Bugfix for an issue which, whenever the citation was immediately
10747 followed by a comma, prevented correct parsing of org-ref citations.
10748 + Ensure emphasis markup can be nested. Nested emphasis markup (e.g.
10749 `/*strong and emphasized*/`) was interpreted incorrectly in that the
10750 inner markup was not recognized.
10751 + Remove pipe char irking the haddock coverage tool (Albert Krewinkel).
10753 * Docx reader: Empty header should be list of lists (Jesse Rosenthal).
10754 In the past, the docx reader wrote an empty header as an empty list. It
10755 should have the same width as a row (and be filled with empty cells).
10757 * MediaWiki reader:
10759 + Improved handling of display math (#3362). Sometimes display math is
10760 indented with more than one colon. Previously we handled these cases
10761 badly, generating definition lists and missing the math.
10762 + Fix quotation mark parsing (#3336, tgkokk). Change MediaWiki reader's
10763 behavior when the smart option is parsed to match other readers'
10768 + Fixed `-f markdown_github-hard_line_breaks+escaped_line_breaks`
10769 (#3341). Previously this did not properly enable escaped line breaks.
10770 + Disallow space between inline code and attributes (#3326, #3323,
10773 * DocBook5 writer: make id attribute xml:id, fixes #3329 (#3330, Mauro Bieg).
10775 * Added some test cases for ODT reader (#3306, #3308, Hubert Plociniczak).
10777 * LaTeX writer: allow tables with empty cells to count as "plain."
10778 This addresses a problem of too-wide tables when empty cells
10779 are used. Thanks to Joost Kremers for reporting the issue.
10781 * Org writer: prefix footnote numbers with `fn:` (Albert Krewinkel).
10782 Unprefixed numbers where used by older org-mode versions, but are no
10785 * HTML writer: don't process pars with empty RawInline, (#1040, #3327,
10788 * Markdown writer: Fix display math with `--webtex` (#3298).
10790 * Fix sample.lua so it properly handles raw blocks/inlines (#3358,
10795 + default.latex: Moved geometry after hyperref (Václav Haisman).
10796 Otherwise PDF sizes can be wrong in some circumstances.
10797 + Copied a few changes from default.latex to default.beamer
10799 + default.latex, default.beamer: Changed position of `\VerbatimNotes`
10800 and `fancyvrb`. This fixes hyperlinks on footnotes in documents
10801 that contain verbatim in notes (#3361). (Note: the beamer template
10802 was updated to match the LaTeX template, but at this point verbatim
10803 in notes seems not to work in beamer.)
10804 + default.latex: Allow passing `microtypeoptions` to microtype
10806 + default.latex: Add hyphen option to url package.
10807 + default.docbook5: Fix namespace declarations (Mauro Bieg).
10809 * Moved `make_osx_package.sh` to `osx/` directory.
10811 * Travis continuous integration:
10813 + Fix false positives with dist build.
10814 + Speed improvements (Kolen Cheung, #3304, #3357).
10818 + Clarify that blank space is needed around footnotes (#3352).
10819 + Fixed typo (#3351, Alexey Rogechev).
10820 + Note that `--wrap=auto` does not work in HTML output.
10821 + Default `--columns` width is 72, not 80.
10822 + Fixed broken links (#3316, Kolen Cheung).
10823 + Document usage of `@*` in nocite section (#3333, John Muccigrosso).
10827 + Indent code so it's properly formatted (#3335, Bheesham Persaud).
10828 + Added instructions for extracting binary from OSX, Windows packages.
10830 * CONTRIBUTING.md: Describe labels currently used in issue tracker
10831 (Albert Krewinkel). The labels have changed over time, the list of
10832 labels is updated to reflect the current set of labels used in the
10835 * Rearrange and extend badges in README (Albert Krewinkel, #3354)
10837 * Bumped version bounds for dependencies.
10840## pandoc 1.19.1 (2016-12-10)
10842 * Set `PANDOC_VERSION` environment variable for filters (#2640).
10843 This allows filters to check the pandoc version that produced
10844 the JSON they are receiving.
10846 * Docx reader: Ensure one-row tables don't have header (#3285,
10847 Jesse Rosenthal). Tables in MS Word are set by default to have
10848 special first-row formatting, which pandoc uses to determine whether
10849 or not they have a header. This means that one-row tables will, by
10850 default, have only a header -- which we imagine is not what people
10851 want. This change ensures that a one-row table is not understood to
10852 be a header only. Note that this means that it is impossible to
10853 produce a header-only table from docx, even though it is legal
10854 pandoc. But we believe that in nearly all cases, it will be an
10855 accidental (and unwelcome) result
10859 + Fixed some bad regressions in HTML table parser (#3280).
10860 This regression leads to the introduction of empty rows
10861 in some circumstances.
10862 + Understand `style=width:` as well as `width` in `col` (#3286).
10866 + Print warnings when keys, substitition, notes not found.
10867 Previously the parsers failed and we got raw text. Now we get a
10868 link with an empty URL, or empty inlines in the case of a note or
10871 + Fix hyperlink aliases (#3283).
10873 * Man writer: Ensure that periods are escaped at beginning of line
10876 * LaTeX writer: Fix unnumbered headers when used with `--top-level`
10877 (#3272, Albert Krewinkel). Fix interaction of top-level
10878 divisions `part` or `chapter` with unnumbered headers when
10879 emitting LaTeX. Headers are ensured to be written using
10880 stared commands (like `\subsection*{}`).
10882 * LaTeX template: use comma not semicolon to separate keywords for
10883 `pdfkeywords`. Thanks to Wandmalfarbe.
10885 * Markdown writer: Fixed incorrect word wrapping (#3277).
10886 Previously pandoc would sometimes wrap lines too early due to
10889 * Text.Pandoc.Pretty: Added `afterBreak` [API change]. This makes it
10890 possible to insert escape codes for content that needs escaping at the
10891 beginning of a line.
10893 * Removed old MathMLInHTML.js from 2004, which should no longer
10894 be needed for MathML with modern browsers.
10896 * Fixed tests with dynamic linking (#2709).
10898 * Makefile: Use stack instead of cabal for targets. This is just
10899 a convenience for developers.
10901 * Fixed bash completion of filenames with space (#2749).
10903 * MANUAL: improved documentation on how to create a custom
10906 * Fix minor spelling typos in the manual (#3273, Anthony Geoghegan)
10908## pandoc 1.19 (2016-12-01)
10910 * Changed resolution of filter paths.
10912 + We now first treat the argument of `--filter` as a full (absolute
10913 or relative) path, looking for a program there. If it's found, we
10915 + If not, and if it is a simple program name or a relative path, we
10916 try resolving it relative to `$DATADIR/filters`.
10917 + If this fails, then we treat it as a program name and look in the
10919 + Removed a hardcoded '/' that may have caused problems with
10922 Previously if you did `--filter foo` and you had `foo` in your path and
10923 also an executable `foo` in your working directory, the one in the path
10924 would be used. Now the one in the working directory is used.
10926 In addition, when you do `--filter foo/bar.hs`, pandoc will now find a
10927 filter `$DATADIR/filters/foo/bar.hs` -- assuming there isn't a
10928 `foo/bar.hs` relative to the working directory.
10930 * Allow `file://` URIs as arguments (#3196). Also improved default reader
10931 format detection. Previously with a URI ending in .md or .markdown,
10932 pandoc would assume HTML input. Now it treats these as markdown.
10934 * Allow to overwrite top-level division type heuristics (#3258,
10935 Albert Krewinkel). Pandoc uses heuristics to determine the most
10936 reasonable top-level division type when emitting LaTeX or
10937 Docbook markup. It is now possible to overwrite this implicitly set
10938 top-level division via the `top-level-division` command line parameter.
10940 * Text.Pandoc.Options \[API changes\]:
10942 + Removed `writerStandalone` field in `WriterOptions`, made
10943 `writerTemplate` a `Maybe` value. Previously setting
10944 `writerStandalone = True` did nothing unless a template was provided
10945 in writerTemplate. Now a fragment will be generated if
10946 `writerTemplate` is `Nothing`; otherwise, the specified template
10947 will be used and standalone output generated.
10948 + `Division` has been renamed `TopLevelDivision` (#3197). The
10949 `Section`, `Chapter`, and `Part` constructors were renamed to
10950 `TopLevelSection`, `TopLevelChapter`, and
10951 `TopLevelPart`, respectively. An additional `TopLevelDefault`
10952 constructor was added, which is now also the new default value of
10953 the `writerTopLevelDivision` field in `WriterOptions`.
10955 * Improved error if they give wrong arg to `--top-level-division`.
10957 * Use new module from texmath to lookup MS font codepoints in Docx reader.
10958 Removed unexported module Text.Pandoc.Readers.Docx.Fonts. Its code now
10959 lives in texmath (0.9).
10961 * DocBook reader: Fixed xref lookup (#3243). It previously only worked
10962 when the qnames lacked the docbook namespace URI.
10966 + Improved table parsing (#3027). We now check explicitly for non-1
10967 rowspan or colspan attributes, and fail when we encounter them.
10968 Previously we checked that each row had the same number of cells,
10969 but that could be true even with rowspans/colspans. And there are
10970 cases where it isn't true in tables that we can handle fine -- e.g.
10971 when a tr element is empty. So now we just pad rows with empty cells
10973 + Treat `<math>` as MathML by default unless something else is
10974 explicitly specified in xmlns. Provided it parses as MathML,
10975 of course. Also fixed default which should be to inline math if no
10976 display attribute is used.
10977 + Only treat "a" element as link if it has href (#3226). Otherwise
10980 * Docx reader (Jesse Rosenthal):
10982 + Add a placeholder value for CHART. We wrap `[CHART]` in a
10983 `<span class="chart">`. Note that it maps to inlines because, in
10984 docx, anything in a drawing tag can be part of a larger paragraph.
10985 + Be more specific in parsing images We not only want `w:drawing`,
10986 because that could also include charts. Now we specify
10987 `w:drawing/pic:pic`. This shouldn't change behavior at all, but it's
10988 a first step toward allowing other sorts of drawing data as well.
10989 + Abstract out function to avoid code repetition.
10990 + Update tests for img title and alt (#3204).
10991 + Handle Alt text and titles in images. We use the "description" field
10992 as alt text and the "title" field as title. These can be accessed
10993 through the "Format Picture" dialog in Word.
10994 + Docx reader utils: handle empty namespace in `elemName`. Previously,
10995 if given an empty namespace `(elemName ns "" "foo")` `elemName`
10996 would output a QName with a `Just ""` namespace. This is never what
10997 we want. Now we output a `Nothing`. If someone *does* want a
10998 `Just ""` in the namespace, they can enter the QName
11001 * ODT reader/writer:
11003 + Inline code when text has a special style (Hubert Plociniczak). When
11004 a piece of text has a text `Source_Text` then we assume that this is
11005 a piece of the document that represents a code that needs to
11006 be inlined. Adapted the writer to also reflect that change.
11007 Previously it was just writing a 'preformatted' text using a
11008 non-distinguishable font style. Code blocks are still not recognized
11009 by the ODT reader. That's a separate issue.
11010 + Infer table's caption from the paragraph (#3224,
11011 Hubert Plociniczak). ODT's reader always put empty captions for the
11012 parsed tables. This commit
11014 1. checks paragraphs that follow the table definition
11015 2. treats specially a paragraph with a style named 'Table'
11016 3. does some postprocessing of the paragraphs that combines tables
11017 followed immediately by captions
11019 The ODT writer used the `TableCaption` style for the caption
11020 paragraph. This commit follows the OpenOffice approach which allows
11021 for appending captions to table but uses a built-in style named
11022 `Table` instead of `TableCaption`. Users of a custom `reference.odt`
11023 should change the style's name from `TableCaption` to `Table`.
11025 * ODT reader: Infer tables' header props from rows (#3199,
11026 Hubert Plociniczak). ODT reader simply provided an empty header list
11027 which meant that the contents of the whole table, even if not empty, was
11028 simply ignored. While we still do not infer headers we at least have to
11029 provide default properties of columns.
11033 + Allow reference link labels starting with `@...` if `citations`
11034 extension disabled (#3209). Example: in
11036 \[link text\]\[@a\]
11038 `link text` isn't hyperlinked because `[@a]` is parsed as
11039 a citation. Previously this happened whether or not the `citations`
11040 extension was enabled. Now it happens only if the `citations`
11041 extension is enabled.
11042 + Allow alignments to be specified in Markdown grid tables. For
11045 +-------+---------------+--------------------+
11046 | Right | Left | Centered |
11047 +=========:+:=================+:=============:+
11048 | Bananas | $1.34 | built-in wrapper |
11049 +-------+---------------+--------------------+
11051 + Allow Small Caps elements to be created using bracketed spans (as
11052 they already can be using HTML-syntax spans) (#3191, Kolen Cheung).
11056 + Don't treat `\vspace` and `\hspace` as block commands (#3256).
11057 Fixed an error which came up, for example, with `\vspace` inside
11058 a caption. (Captions expect inlines.)
11059 + Improved table handling. We can now parse all of the tables emitted
11060 by pandoc in our tests. The only thing we don't get yet are
11061 alignments and column widths in more complex tables. See #2669.
11062 + Limited support for minipage.
11063 + Allow for `[]`s inside LaTeX optional args. Fixes cases like:
11064 + Handle BVerbatim from fancyvrb (#3203).
11065 + Handle hungarumlaut (#3201).
11066 + Allow beamer-style `<...>` options in raw LaTeX (also in Markdown)
11067 (#3184). This allows use of things like `\only<2,3>{my content}` in
11068 Markdown that is going to be converted to beamer.
11070 * Use pre-wrap for code in dzslides template (Nicolas Porcel). Otherwise
11071 overly long code will appear on every slide.
11073 * Org reader (Albert Krewinkel):
11075 + Respect column width settings (#3246). Table column properties can
11076 optionally specify a column's width with which it is displayed in
11077 the buffer. Some exporters, notably the ODT exporter in org-mode
11078 v9.0, use these values to calculate relative column widths. The org
11079 reader now implements the same behavior. Note that the org-mode
11080 LaTeX and HTML exporters in Emacs don't support this feature yet,
11081 which should be kept in mind by users who use the column
11083 + Allow HTML attribs on non-figure images (#3222). Images which are
11084 the only element in a paragraph can still be given HTML attributes,
11085 even if the image does not have a caption and is hence not a figure.
11086 The following will add set the `width` attribute of the image to
11089 +ATTR\_HTML: :width 50%
11090 =======================
11092 \[\[file:image.jpg\]\]
11094 + Support `ATTR_HTML` for special blocks (#3182). Special
11095 blocks (i.e. blocks with unrecognized names) can be prefixed with an
11096 `ATTR_HTML` block attribute. The attributes defined in that
11097 meta-directive are added to the `Div` which is used to represent the
11099 + Support the `todo` export option. The `todo` export option allows to
11100 toggle the inclusion of TODO keywords in the output. Setting this to
11101 `nil` causes TODO keywords to be dropped from headlines. The default
11102 is to include the keywords.
11103 + Add support for todo-markers. Headlines can have optional
11104 todo-markers which can be controlled via the `#+TODO`, `#+SEQ_TODO`,
11105 or `#+TYP_TODO` meta directive. Multiple such directives can be
11106 given, each adding a new set of recognized todo-markers. If no
11107 custom todo-markers are defined, the default `TODO` and `DONE`
11108 markers are used. Todo-markers are conceptually separate from
11109 headline text and are hence excluded when autogenerating
11110 headline IDs. The markers are rendered as spans and labelled with
11111 two classes: One class is the markers name, the other signals the
11112 todo-state of the marker (either `todo` or `done`).
11116 + Use `\autocites*` when "suppress-author" citation used.
11117 + Ensure that simple tables have simple cells (#2666). If cells
11118 contain more than a single Plain or Para, then we need to set
11119 nonzero widths and put contents into minipages.
11120 + Remove invalid inlines in sections (#3218, Hubert Plociniczak).
11124 + Fix calculation of column widths for aligned multiline tables
11125 (#1911, Björn Peemöller). This also fixes excessive CPU and memory
11126 usage for tables when `--columns` is set in such a way that cells
11127 must be very tiny. Now cells are guaranteed to be big enough so that
11128 single words don't need to line break, even if this pushes the line
11129 length above the column width.
11130 + Use bracketed form for native spans when `bracketed_spans`
11132 + Fixed inconsistent spacing issue (#3232). Previously a tight bullet
11133 sublist got rendered with a blank line after, while a tight ordered
11134 sublist did not. Now we don't get the blank line in either case.
11135 + Fix escaping of spaces in super/subscript (#3225). Previously two
11136 backslashes were inserted, which gave a literal backslash.
11137 + Adjust widths in Markdown grid tables so that they match
11142 + Give full detail when there are errors converting tex math.
11143 + Handle title text in images (Jesse Rosenthal). We already handled
11144 alt text. This just puts the image "title" into the docx
11146 + Fixed XML markup for empty cells (#3238). Previously the Compact
11147 style wasn't being applied properly to empty cells.
11151 + Updated `renderHtml` import from blaze-html.
11153 * Text.Pandoc.Pretty:
11155 + Fixed some bugs that caused blank lines in tables (#3251). The bugs
11156 caused spurious blank lines in grid tables when we had things like
11157 `blankline $$ blankline`.
11158 + Add exported function `minOffet` \[API change\] (Björn Peemöller).
11159 + Added error message for illegal call to `block` (Björn Peemöller).
11161 * Text.Pandoc.Shared:
11163 + Put `warn` in MonadIO.
11164 + `fetchItem`: Better handling of protocol-relative URL (#2635). If
11165 URL starts with `//` and there is no "base URL" (as there would be
11166 if a URL were used on the command line), then default to http:.
11168 * Export Text.Pandoc.getDefaultExtensions \[API change\] (#3178).
11170 * In --version, trap error in `getAppUserDataDirectory` (#3241). This
11171 fixes a crash with `pandoc --version` on unusual systems with no real
11172 user (e.g. SQL Server 2016).
11174 * Added weigh-pandoc for memory usage diagnostics (#3169).
11176 * Use correct mime types for woff and woff2 (#3228).
11178 * Remove make\_travis\_yml.hs (#3235, Kolen Cheung).
11180 * changelog: Moved an item that was misplaced in the 1.17.2 section to the
11181 1.18 section where it belongs.
11183 * CONTRIBUTING.md: minor change in wording and punctuation (#3252,
11186 * Further revisions to manual for `--version` changes (#3244).
11190## pandoc 1.18 (2016-10-26)
11192 * Added `--list-input-formats`, `--list-output-formats`,
11193 `--list-extensions`, `--list-highlight-languages`, and
11194 `--list-highlight-styles` (#3173). Removed list of highlighting
11195 languages from `--version` output. Removed list of input and output
11196 formats from default `--help` output.
11198 * Added `--reference-location=block|section|document` option
11199 (Jesse Rosenthal). This determines whether Markdown link references
11200 and footnotes are placed at the end of the document, the end of the
11201 section, or the end of the top-level block.
11203 * Added `--top-level-division=section|chapter|part` (Albert Krewinkel).
11204 This determines what a level-1 header corresponds to in LaTeX,
11205 ConTeXt, DocBook, and TEI output. The default is `section`.
11206 The `--chapters` option has been deprecated in favor of
11207 `--top-level-division=chapter`.
11209 * Added `LineBlock` constructor for `Block` (Albert Krewinkel). This
11210 is now used in parsing RST and Markdown line blocks, DocBook
11211 `linegroup`/`line` combinations, and Org-mode `VERSE` blocks.
11212 Previously `Para` blocks with hard linebreaks were used. `LineBlock`s
11213 are handled specially in the following output formats: AsciiDoc
11214 (as `[verse]` blocks), ConTeXt (`\startlines`/`\endlines`),
11215 HTML (`div` with a style), Markdown (line blocks if `line_blocks`
11216 is enabled), Org-mode (`VERSE` blocks), RST (line blocks). In
11217 other output formats, a paragraph with hard linebreaks is emitted.
11219 * Allow binary formats to be written to stdout (but not to tty) (#2677).
11220 Only works on posix, since we use the unix library to check whether
11221 output is to tty. On Windows, pandoc works as before and always requires
11222 an output file parameter for binary formats.
11224 * Changed JSON output format (Jesse Rosenthal). Previously we used
11225 generically generated JSON, but this was subject to change depending
11226 on the version of aeson pandoc was compiled with. To ensure stability,
11227 we switched to using manually written ToJSON and FromJSON
11228 instances, and encoding the API version. **Note:** pandoc filter
11229 libraries will need to be revised to handle the format change.
11230 Here is a summary of the essential changes:
11232 + The toplevel JSON format is now `{"pandoc-api-version" :
11233 [MAJ, MIN, REV], "meta" : META, "blocks": BLOCKS}`
11234 instead of `[{"unMeta": META}, [BLOCKS]]`.
11235 Decoding fails if the major and minor version numbers don't
11237 + Leaf nodes no longer have an empty array for their "c" value.
11238 Thus, for example, a `Space` is encoded as `{"t":"Space"}`
11239 rather than `{"t":"Space","c":[]}` as before.
11241 * Removed `tests/Tests/Arbitrary.hs` and added a `Text.Pandoc.Arbitrary`
11242 module to pandoc-types (Jesse Rosenthal). This makes it easier
11243 to use QuickCheck with pandoc types outside of pandoc itself.
11245 * Add `bracketed_spans` Markdown extension, enabled by default
11246 in pandoc `markdown`. This allows you to create a native span
11247 using this syntax: `[Here is my span]{#id .class key="val"}`.
11249 * Added `angle_brackets_escapable` Markdown extension (#2846).
11250 This is needed because github flavored Markdown has a slightly
11251 different set of escapable symbols than original Markdown;
11252 it includes angle brackets.
11254 * Export `Text.Pandoc.Error` in `Text.Pandoc` [API change].
11256 * Print highlighting-kate version in `--version`.
11258 * `Text.Pandoc.Options`:
11260 + `Extension` has new constructors `Ext_brackted_spans` and
11261 `Ext_angle_brackets_escapable` [API change].
11262 + Added `ReferenceLocation` type [API change] (Jesse Rosenthal).
11263 + Added `writerReferenceLocation` field to `WriterOptions` (Jesse
11266 * `--filter`: we now check `$DATADIR/filters` for filters before
11267 looking in the path (#3127, Jesse Rosenthal, thanks to Jakob
11268 Voß for the idea). Filters placed in this directory need not
11269 be executable; if the extension is `.hs`, `.php`, `.pl`, `.js`,
11270 or `.rb`, pandoc will run the right interpreter.
11272 * For `--webtex`, replace deprecated Google Chart API by CodeCogs as
11273 default (Kolen Cheung).
11275 * Removed `raw_tex` extension from `markdown_mmd` defaults (Kolen Cheung).
11277 * Execute .js filters with node (Jakob Voß).
11281 + Support `bc..` extended code blocks (#3037). Also, remove trailing
11282 newline in code blocks (consistently with Markdown reader).
11283 + Improve table parsing. We now handle cell and row attributes, mostly
11284 by skipping them. However, alignments are now handled properly.
11285 Since in pandoc alignment is per-column, not per-cell, we
11286 try to devine column alignments from cell alignments.
11287 Table captions are also now parsed, and textile indicators
11288 for thead and tfoot no longer cause parse failure. (However,
11289 a row designated as tfoot will just be a regular row in pandoc.)
11290 + Improve definition list parsing. We now allow multiple terms
11291 (which we concatenate with linebreaks). An exponential parsing
11292 bug (#3020) is also fixed.
11293 + Disallow empty URL in explicit link (#3036).
11297 + Use Div instead of BlockQuote for admonitions (#3031).
11298 The Div has class `admonition` and (if relevant) one of the
11299 following: `attention`, `caution`, `danger`, `error`, `hint`,
11300 `important`, `note`, `tip`, `warning`. **Note:** This will change
11301 the rendering of some RST documents! The word ("Warning", "Attention",
11302 etc.) is no longer added; that must be done with CSS or a filter.
11303 + A Div is now used for `sidebar` as well.
11304 + Skip whitespace before note (Jesse Rosenthal, #3163). RST requires a
11305 space before a footnote marker. We discard those spaces so that footnotes
11306 will be adjacent to the text that comes before it. This is in line with
11307 what rst2latex does.
11308 + Allow empty lines when parsing line blocks (Albert Krewinkel).
11312 + Allow empty lines when parsing line blocks (Albert Krewinkel).
11313 + Allow attributes on autolinks (#3183, Daniele D'Orazio).
11317 + More robust parsing of unknown environments (#3026).
11318 We no longer fail on things like `^` inside options for tikz.
11319 + Be more forgiving of non-standard characters, e.g. `^` outside of math.
11320 Some custom environments give these a meaning, so we should try not to
11321 fall over when we encounter them.
11322 + Drop duplicate `*` in bibtexKeyChars (Albert Krewinkel)
11324 * MediaWiki reader:
11326 + Fix for unquoted attribute values in mediawiki tables (#3053).
11327 Previously an unquoted attribute value in a table row
11328 could cause parsing problems.
11329 + Improved treatment of verbatim constructions (#3055).
11330 Previously these yielded strings of alternating Code and Space
11331 elements; we now incorporate the spaces into the Code. Emphasis
11332 etc. is still possible inside these.
11333 + Properly interpret XML tags in pre environments (#3042). They are meant
11334 to be interpreted as literal text.
11336 * EPUB reader: don't add root path to data: URIs (#3150).
11337 Thanks to @lep for the bug report and patch.
11339 * Org reader (Albert Krewinkel):
11341 + Preserve indentation of verse lines (#3064). Leading spaces in verse
11342 lines are converted to non-breaking spaces, so indentation is preserved.
11343 + Ensure image sources are proper links. Image sources as those in plain
11344 images, image links, or figures, must be proper URIs or relative file
11345 paths to be recognized as images. This restriction is now enforced
11346 for all image sources. This also fixes the reader's usage of uncleaned
11347 image sources, leading to `file:` prefixes not being deleted from
11348 figure images. Thanks to @bsag for noticing this bug.
11349 + Trim verse lines properly (Albert Krewinkel).
11350 + Extract meta parsing code to module. Parsing of meta-data is well
11351 separable from other block parsing tasks. Moving into new module to
11352 get small files and clearly arranged code.
11353 + Read markup only for special meta keys. Most meta-keys should be read
11354 as normal string values, only a few are interpreted as marked-up text.
11355 + Allow multiple, comma-separated authors. Multiple authors can be
11356 specified in the `#+AUTHOR` meta line if they are given as a
11357 comma-separated list.
11358 + Give precedence to later meta lines. The last meta-line of any given
11359 type is the significant line. Previously the value of the first line
11360 was kept, even if more lines of the same type were encountered.
11361 + Read LaTeX_header as header-includes. LaTeX-specific header commands
11362 can be defined in `#+LaTeX_header` lines. They are parsed as
11363 format-specific inlines to ensure that they will only show up in LaTeX
11365 + Set documentclass meta from LaTeX_class.
11366 + Set classoption meta from LaTeX_class_options.
11367 + Read HTML_head as header-includes. HTML-specific head content can be
11368 defined in `#+HTML_head` lines. They are parsed as format-specific
11369 inlines to ensure that they will only show up in HTML output.
11370 + Respect `author` export option. The `author` option controls whether
11371 the author should be included in the final markup. Setting
11372 `#+OPTIONS: author:nil` will drop the author from the final meta-data
11374 + Respect `email` export option. The `email` option controls whether the
11375 email meta-field should be included in the final markup. Setting
11376 `#+OPTIONS: email:nil` will drop the email field from the final
11378 + Respect `creator` export option. The `creator` option controls whether
11379 the creator meta-field should be included in the final markup. Setting
11380 `#+OPTIONS: creator:nil` will drop the creator field from the final
11381 meta-data output. Org-mode recognizes the special value `comment` for
11382 this field, causing the creator to be included in a comment. This is
11383 difficult to translate to Pandoc internals and is hence interpreted the
11384 same as other truish values (i.e. the meta field is kept if it's
11386 + Respect unnumbered header property (#3095). Sections the `unnumbered`
11387 property should, as the name implies, be excluded from the automatic
11388 numbering of section provided by some output formats. The Pandoc
11389 convention for this is to add an "unnumbered" class to the header. The
11390 reader treats properties as key-value pairs per default, so a special
11391 case is added to translate the above property to a class instead.
11392 + Allow figure with empty caption (Albert Krewinkel, #3161).
11393 A `#+CAPTION` attribute before an image is enough to turn an image into
11394 a figure. This wasn't the case because the `parseFromString` function,
11395 which processes the caption value, would fail on empty values. Adding
11396 a newline character to the caption value fixes this.
11400 + Use XML convenience functions (Jesse Rosenthal).
11401 The functions `isElem` and `elemName` (defined in Docx/Util.hs) make
11402 the code a lot cleaner than the original XML.Light functions, but they
11403 had been used inconsistently. This puts them in wherever applicable.
11404 + Handle anchor spans with content in headers. Previously, we would only
11405 be able to figure out internal links to a header in a docx if the
11406 anchor span was empty. We change that to read the inlines out of the
11407 first anchor span in a header.
11408 + Let headers use existing id. Previously we always generated an id for
11409 headers (since they wouldn't bring one from Docx). Now we let it use an
11410 existing one if possible. This should allow us to recurs through anchor
11412 + Use all anchor spans for header ids. Previously we only used the first
11413 anchor span to affect header ids. This allows us to use all the anchor
11414 spans in a header, whether they're nested or not (#3088).
11415 + Test for nested anchor spans in header. This ensures that anchor spans
11416 in header with content (or with other anchor spans inside) will resolve
11417 to links to a header id properly.
11419 * ODT reader (Hubert Plociniczak)
11421 + Include list's starting value. Previously the starting value of
11422 the lists' items has been hardcoded to 1. In reality ODT's list
11423 style definition can provide a new starting value in one of its
11425 + Infer caption from the text following the image.
11426 Frame can contain other frames with the text boxes.
11427 + Add `fig:` to title for Image with a caption (as expected
11428 by pandoc's writers).
11429 + Basic support for images in ODT documents.
11430 + Don't duplicate text for anchors (#3143). When creating an anchor
11431 element we were adding its representation as well as the original
11432 content, leading to text duplication.
11436 + Include an anchor element when a div or span has an id (#3102).
11437 Note that DocBook does not have a class attribute, but at least this
11438 provides an anchor for internal links.
11442 + Don't use * for unnumbered paragraph, subparagraph. The starred
11443 variants don't exist. This helps with part of #3058...it gets rid of
11444 the spurious `*`s. But we still have numbers on the 4th and 5th level
11446 + Properly escape backticks in verbatim (#3121, Jesse Rosenthal).
11447 Otherwise they can cause unintended ligatures like `` ?` ``.
11448 + Handle NARRAOW NO-BREAK SPACE into LaTeX (Vaclav Zeman) as `\,`.
11449 + Don't include `[htbp]` placement for figures (#3103, Václav Haisman).
11450 This allows figure placement defaults to be changed by the user
11453 * TEI writer: remove heuristic to detect book template (Albert Krewinkel).
11454 TEI doesn't have `<book>` elements but only generic `<divN>` division
11455 elements. Checking the template for a trailing `</book>` is nonsensical.
11457 * MediaWiki writer: transform filename with underscores in images (#3052).
11458 `foo bar.jpg` becomes `foo_bar.jpg`. This was already done
11459 for internal links, but it also needs to happen for images.
11461 * ICML writer: replace partial function (!!) in table handling (#3175,
11464 * Man writer: allow section numbers that are not a single digit (#3089).
11466 * AsciiDoc writer: avoid unnecessary use of "unconstrained" emphasis
11467 (#3068). In AsciiDoc, you must use a special form of emphasis
11468 (double `__`) for intraword emphasis. Pandoc was previously using
11469 this more than necessary.
11471 * EPUB writer: use stringify instead of plain writer for metadata
11472 (#3066). This means that underscores won't be used for emphasis,
11473 or CAPS for bold. The metadata fields will just have unadorned
11478 + Implement user-defined styles (Jesse Rosenthal). Divs and Spans
11479 with a `custom-style` key in the attributes will apply the corresponding
11480 key to the contained blocks or inlines.
11481 + Add ReaderT env to the docx writer (Jesse Rosenthal).
11482 + Clean up and streamline RTL behavior (Jesse Rosenthal, #3140).
11483 You can set `dir: rtl` in YAML metadata, or use `-M dir=rtl`
11484 on the command line. For finer-grained control, you can set
11485 the `dir` attribute in Div or Span elements.
11487 * Org writer (Albert Krewinkel):
11489 + Remove blank line after figure caption. Org-mode only treats an image
11490 as a figure if it is directly preceded by a caption.
11491 + Ensure blank line after figure. An Org-mode figure should be surrounded
11492 by blank lines. The figure would be recognized regardless, but images
11493 in the following line would unintentionally be treated as figures as
11495 + Ensure link targets are paths or URLs. Org-mode treats links as
11496 document internal searches unless the link target looks like a URL or
11497 file path, either relative or absolute. This change ensures that this
11498 is always the case.
11499 + Translate language identifiers. Pandoc and Org-mode use different
11500 programming language identifiers. An additional translation between
11501 those identifiers is added to avoid unexpected behavior. This fixes a
11502 problem where language specific source code would sometimes be output
11504 + Drop space before footnote markers (Albert Krewinkel, #3162).
11505 The writer no longer adds an extra space before footnote markers.
11509 + Don't emit HTML for tables unless `raw_html` extension is set (#3154).
11510 Emit `[TABLE]` if no suitable table formats are enabled and raw HTML
11512 + Check for the `raw_html` extension before emitting a raw HTML block.
11513 + Abstract out note/ref function (Jesse Rosenthal).
11514 + Add ReaderT monad for environment variables (Jesse Rosenthal).
11516 * HTML, EPUB, slidy, revealjs templates: Use `<p>` instead of `<h1>` for
11517 subtitle, author, date (#3119). Note that, as a result of this change,
11518 authors may need to update CSS.
11520 * revealjs template: Added `notes-server` option
11521 (jgm/pandoc-templates#212, Yoan Blanc).
11525 + Restore whitespace between paragraphs. This was
11526 a regression in the last release (jgm/pandoc-templates#207).
11527 + Added `themeoptions` variable (Carsten Gips).
11528 + Added `beamerarticle` variable. This causes the `beamerarticle`
11529 package to be loaded in beamer, to produce an article from beamer
11530 slides. (Carsten Gips)
11531 + Added support for `fontfamilies` structured variable
11533 + Added hypersetup options (Jake Zimmerman).
11537 + Added dummy definition for `\institute`.
11538 This isn't a standard command, and we want to avoid a crash when
11539 `institute` is used with the default template.
11540 + Define default figure placement (Václav Haisman), since pandoc
11541 no longer includes `[htbp]` for figures. Users with custom templates
11542 will want to add this. See #3103.
11543 + Use footnote package to fix notes in tables (jgm/pandoc-templates#208,
11546 * Moved template compiling/rendering code to a separate library.
11547 `doctemplates`. This allows the pandoc templating system to be
11548 used independently.
11550 * Text.Pandoc.Error: Fix out of index error in `handleError`
11551 (Matthew Pickering). The fix is to not try to show the exact line when
11552 it would cause an out-of-bounds error as a result of included files.
11554 * Text.Pandoc.Shared: Add `linesToBlock` function (Albert Krewinkel).
11556 * Text.Pandoc.Parsing.emailAddress: tighten up parsing of email
11557 addresses. Technically `**@user` is a valid email address, but if we
11558 allow things like this, we get bad results in markdown flavors
11559 that autolink raw email addresses (see #2940). So we exclude a few
11560 valid email addresses in order to avoid these more common bad cases.
11562 * Text.Pandoc.PDF: Don't crash with nonexistent image (#3100). Instead,
11563 emit the alt text, emphasized. This accords with what the ODT writer
11564 currently does. The user will still get a warning about a nonexistent
11567 * Fix example in API documentation (#3176, Thomas Weißschuh).
11569 * Tell where to get tarball in INSTALL (#3062).
11571 * Rename README to MANUAL.txt and add GitHub-friendly README.md
11572 (Albert Krewinkel, Kolen Cheung).
11574 * Replace COPYING with Markdown version COPYING.md from GNU (Kolen Cheung).
11578 + Put note on structured vars in separate paragraph (#2148, Albert
11579 Krewinkel). Make it clearer that structured author variables require a
11581 + Note that `--katex` works best with `html5` (#3077).
11582 + Fix the LaTeX and EPUB links in manual (Morton Fox).
11583 + Document `biblio-title` variable.
11585 * Improve spacing of footnotes in `--help` output (Waldir Pimenta).
11587 * Update KaTeX to v0.6.0 (Kolen Cheung).
11589 * Allow latest dependencies.
11591 * Use texmath 0.8.6.6 (#3040).
11593 * Allow http-client 0.4.30, which is the version in stackage lts.
11594 Previously we required 0.5.
11595 Remove CPP conditionals for earlier versions.
11597 * Remove support for GHC < 7.8 (Jesse Rosenthal).
11599 + Remove Compat.Monoid.
11600 + Remove an inline monad compatibility macro.
11601 + Remove Text.Pandoc.Compat.Except.
11602 + Remove directory compat.
11603 + Change constraint on mtl.
11604 + Remove unnecessary CPP condition in UTF8.
11605 + Bump base lower bound to 4.7.
11606 + Remove 7.6 build from .travis.yaml.
11607 + Bump supported ghc version in CONTRIBUTING.md.
11608 + Add note about GHC version support to INSTALL.
11609 + Remove GHC 7.6 from list of tested versions (Albert Krewinkel).
11610 + Remove TagSoup compat.
11611 + Add EOL note to time compat module. Because time 1.4 is a boot library
11612 for GHC 7.8, we will support the compatibility module as long as we
11613 support 7.8. But we should be clear about when we will no longer need
11615 + Remove blaze-html CPP conditional.
11616 + Remove unnecessary CPP in custom Prelude.
11618## pandoc 1.17.2 (2016-07-17)
11620 * Added Zim Wiki writer, template and tests. `zimwiki` is now
11621 a valid output format. (Alex Ivkin)
11623 * Changed email-obfuscation default to no obfuscation (#2988).
11624 + `writerEmailObfuscation` in `defaultWriterOptions` is now
11626 + the default for the command-line `--email-obfuscation` option is
11629 * Docbook writer: Declare xlink namespace in Docbook5 output (Ivo Clarysse).
11633 + Support arbitrary raw inlines (Albert Krewinkel).
11634 Org mode allows arbitrary raw inlines ("export snippets" in Emacs
11635 parlance) to be included as `@@format:raw foreign format text@@`.
11636 + Improve Div handling (Albert Krewinkel). Div blocks handling is
11637 changed to make the output look more like idiomatic org mode:
11638 - Div-wrapped content is output as-is if the div's attribute is the
11640 - Div containers with an id but neither classes nor key-value pairs
11641 are unwrapped and the id is added as an anchor.
11642 - Divs with classes associated with greater block elements are
11643 wrapped in a `#+BEGIN`...`#+END` block.
11644 - The old behavior for Divs with more complex attributes is kept.
11646 * HTML writer: Better support for raw LaTeX environments (#2758).
11647 Previously we just passed all raw TeX through when MathJax
11648 was used for HTML math. This passed through too much.
11649 With this patch, only raw LaTeX environments that MathJax
11650 can handle get passed through.
11651 This patch also causes raw LaTeX environments to be treated
11652 as math, when possible, with MathML and WebTeX output.
11654 * Markdown writer: use raw HTML for simple, pipe tables with linebreaks
11655 (#2993). Markdown line breaks involve a newline, and simple and pipe
11656 tables can't contain one.
11658 * Make --webtex work with the Markdown writer (#1177).
11659 This is a convenient option for people using
11660 websites whose Markdown flavors don't provide for math.
11664 + Set paragraph to FirstPara after display math (Jesse Rosenthal).
11665 We treat display math like block quotes, and apply FirstParagraph style
11666 to paragraphs that follow them. These can be styled as the user
11667 wishes. (But, when the user is using indentation, this allows for
11668 paragraphs to continue after display math without indentation.)
11669 + Use actual creation time as doc prop (Jesse Rosenthal).
11670 Previously, we had used the user-supplied date, if available, for Word's
11671 document creation metadata. This could lead to weird results, as in
11672 cases where the user post-dates a document (so the modification might be
11673 prior to the creation). Here we use the actual computer time to set the
11678 + Don't URI-escape image source (#2825). Usually this is a local file,
11679 and replacing spaces with `%20` ruins things.
11680 + Allow 'standout' as a beamer frame option (#3007).
11681 `## Slide title {.standout}`.
11683 * RST reader: Fixed links with no explicit link text. The link
11684 `` `<foo>`_ `` should have `foo` as both its link text and its URL.
11685 See RST spec at <http://docutils.sourceforge.net/docs/ref/rst/restructuredtext.html#embedded-uris-and-aliases>
11686 Closes Debian #828167 -- reported by Christian Heller.
11690 + Fixed attributes (#2984). Attributes can't be followed by
11691 a space. So, `_(class)emph_` but `_(noclass) emph_`.
11692 + Fixed exponential parsing bug (#3020).
11693 + Fix overly aggressive interpretation as images (#2998).
11694 Spaces are not allowed in the image URL in textile.
11698 + Fix `\cite` so it is a NormalCitation not AuthorInText.
11699 + Strip off double quotes around image source if present (#2825).
11700 Avoids interpreting these as part of the literal filename.
11704 + Add semicolon to list of special chars (Albert Krewinkel)
11705 Semicolons are used as special characters in citations syntax. This
11706 ensures the correct parsing of Pandoc-style citations: `[prefix; @key;
11707 suffix]`. Previously, parsing would have failed unless there was a space
11708 or other special character as the last <prefix> character.
11709 + Add support for "Berkeley-style" cites (Albert Krewinkel, #1978).
11710 A specification for an official Org-mode citation syntax was drafted by
11711 Richard Lawrence and enhanced with the help of others on the orgmode
11712 mailing list. Basic support for this citation style is added to the
11714 + Support arbitrary raw inlines (Albert Krewinkel).
11715 Org mode allows arbitrary raw inlines ("export snippets" in Emacs
11716 parlance) to be included as `@@format:raw foreign format text@@`.
11717 + Remove partial functions (Albert Krewinkel, #2991).
11718 Partial functions like `head` lead to avoidable errors and should be
11719 avoided. They are replaced with total functions.
11720 + Support figure labels (Albert Krewinkel, #2496, #2999).
11721 Figure labels given as `#+LABEL: thelabel` are used as the ID of the
11722 respective image. This allows e.g. the LaTeX to add proper `\label`
11724 + Improve tag and properties type safety (Albert Krewinkel).
11725 Specific newtype definitions are used to replace stringly typing of tags
11726 and properties. Type safety is increased while readability is improved.
11727 + Parse as headlines, convert to blocks (Albert Krewinkel).
11728 Emacs org-mode is based on outline-mode, which treats documents as trees
11729 with headlines are nodes. The reader is refactored to parse into a
11730 similar tree structure. This simplifies transformations acting on
11731 document (sub-)trees.
11732 * Refactor comment tree handling (Albert Krewinkel).
11733 Comment trees were handled after parsing, as pattern matching on lists
11734 is easier than matching on sequences. The new method of reading
11735 documents as trees allows for more elegant subtree removal.
11736 * Support archived trees export options (Albert Krewinkel).
11737 Handling of archived trees can be modified using the `arch` option.
11738 Archived trees are either dropped, exported completely, or collapsed to
11739 include just the header when the `arch` option is nil, non-nil, or
11740 `headline`, respectively.
11741 * Put export setting parser into module (Albert Krewinkel).
11742 Export option parsing is distinct enough from general block parsing to
11743 justify putting it into a separate module.
11744 * Support headline levels export setting (Albert Krewinkel).
11745 The depths of headlines can be modified using the `H` option. Deeper
11746 headlines will be converted to lists.
11747 * Replace ugly code with view pattern (Albert Krewinkel).
11748 Some less-than-smart code required a pragma switching of overlapping
11749 pattern warnings in order to compile seamlessly. Using view patterns
11750 makes the code easier to read and also doesn't require overlapping
11751 pattern checks to be disabled.
11752 * Fix parsing of verbatim inlines (Albert Krewinkel, #3016).
11753 Org rules for allowed characters before or after markup chars were not
11754 checked for verbatim text. This resultet in wrong parsing outcomes of
11755 if the verbatim text contained e.g. space enclosed markup characters as
11756 part of the text (`=is_substr = True=`). Forcing the parser to update
11757 the positions of allowed/forbidden markup border characters fixes this.
11759 * LaTeX template: fix for obscure hyperref/xelatex issue.
11760 Here's a minimal case:
11762 \documentclass[]{article}
11763 \usepackage{hyperref}
11768 Without this change, this fails on the second invocation of xelatex.
11769 This affects inputs this like `# %á` with pdf output via xelatex.
11771 * trypandoc: call results 'html' instead of 'result'.
11772 This is for better compatibility with babelmark2.
11774 * Document MultiMarkdown as input/output format (Albert Krewinkel, #2973).
11775 MultiMarkdown was only mentioned as a supported Markdown dialect but not
11776 as a possible input or output format. A brief mention is added
11777 everywhere the other supported markdown dialects are mentioned.
11779 * Document Org mode as a format containing raw HTML (Albert Krewinkel)
11780 Raw HTML is kept when the output format is Emacs Org mode.
11782 * Implement `RawInline` and `RawBlock` in sample lua custom writer (#2985).
11784 * Text.Pandoc.Shared:
11786 + Introduce blocksToInlines function (Jesse Rosenthal).
11787 This is a lossy function for converting `[Block] -> [Inline]`. Its main
11788 use, at the moment, is for docx comments, which can contain arbitrary
11789 blocks (except for footnotes), but which will be converted to spans.
11790 This is, at the moment, pretty useless for everything but the basic
11791 `Para` and `Plain` comments. It can be improved, but the docx reader
11792 should probably emit a warning if the comment contains more than this.
11793 + Add BlockQuote to blocksToInlines (Jesse Rosenthal).
11794 + Add further formats for `normalizeDate` (Jesse Rosenthal).
11795 We want to avoid illegal dates -- in particular years with greater than
11796 four digits. We attempt to parse series of digits first as `%Y%m%d`, then
11797 `%Y%m`, and finally `%Y`.
11798 + `normalizeDate` should reject illegal years (Jesse Rosenthal).
11799 We only allow years between 1601 and 9999, inclusive. The ISO 8601
11800 actually says that years are supposed to start with 1583, but MS Word
11801 only allows 1601-9999. This should stop corrupted word files if the date
11802 is out of that range, or is parsed incorrectly.
11803 + Improve year sanity check in normalizeDate (Jesse Rosenthal).
11804 Previously we parsed a list of dates, took the first one, and then
11805 tested its year range. That meant that if the first one failed, we
11806 returned nothing, regardless of what the others did. Now we test for
11807 sanity before running `msum` over the list of Maybe values. Anything
11808 failing the test will be Nothing, so will not be a candidate.
11812 + Add simple comment functionality. (Jesse Rosenthal).
11813 This adds simple track-changes comment parsing to the docx reader. It is
11814 turned on with `--track-changes=all`. All comments are converted to
11815 inlines, which can list some information. In the future a warning will be
11816 added for comments with formatting that seems like it will be excessively
11817 denatured. Note that comments can extend across blocks. For that reason
11818 there are two spans: `comment-start` and `comment-end`. `comment-start`
11819 will contain the comment. `comment-end` will always be empty. The two
11820 will be associated by a numeric id.
11821 + Enable warnings in top-level reader (Jesse Rosenthal).
11822 Previously we had only allowed for warnings in the parser. Now we allow
11823 for them in the `Docx.hs` as well. The warnings are simply concatenated.
11824 + Add warning for advanced comment formatting. (Jesse Rosenthal).
11825 We can't guarantee we'll convert every comment correctly, though we'll
11826 do the best we can. This warns if the comment includes something other
11827 than Para or Plain.
11828 + Add tests for warnings. (Jesse Rosenthal).
11829 + Add tests for comments (Jesse Rosenthal).
11830 We test for comments, using all track-changes options. Note that we
11831 should only output comments if `--track-changes=all`. We also test for
11832 emitting warnings if there is complicated formatting.
11834 * README: update to include track-changes comments. (Jesse Rosenthal)
11836 * Improved Windows installer - don't ignore properties set on command-line.
11837 See #2708. Needs testing to see if this resolves the issue.
11840 * Process markdown extensions on command line in L->R order (#2995).
11841 Previously they were processed, very unintuitively, in R->L
11842 order, so that `markdown-tex_math_dollars+tex_math_dollars`
11843 had `tex_math_dollars` disabled.
11845 * Added `secnumdepth` variable to LaTeX template (#2920).
11847 * Include table of contents in README.html in Windows package.
11849 * Writers: treat SoftBreak as space for stripping (Jesse Rosenthal)
11850 In Writers.Shared, we strip leading and trailing spaces for display
11851 math. Since SoftBreak's are treated as spaces, we should strip those
11854 * beamer, latex templates: pass biblatexoptions directly in package load.
11855 This allows runtime optinos to be used. Fixes jgm/pandoc-citeproc#201
11857 * CPP workaround for deprecation of `parseUrl` in http-client.
11859 * Removed some redundant class constraints.
11861 * make_oxs_package.sh - use OSX env variable.
11863 * Added `winpkg` target to Makefile. This downloads the windows package
11864 from appveyor and signs it using the key.
11866 * Document Org mode as a format containing raw TeX (Albert Krewinkel).
11867 Raw TeX is kept verbatim when the output format is Emacs Org mode.
11869 * Support math with haddock-library >= 1.4.
11871 * Removed `-rtsopts` from library stanza. It has no effect, and Hackage
11872 wouldn't accept the package.
11874 * Update library dependency versions.
11876## pandoc 1.17.1 (2016-06-04)
11878 * New output format: `docbook5` (Ivo Clarysse).
11880 * `Text.Pandoc.Options`: Add `writerDocBook5` to `WriterOptions`
11885 + Add :PROPERTIES: drawer support (Albert Krewinkel, #1962).
11886 This allows header attributes to be added to org documents in the form
11887 of `:PROPERTIES:` drawers. All available attributes are stored as
11888 key/value pairs. This reflects the way the org reader handles
11889 `:PROPERTIES:` blocks.
11890 + Add drawer capability (Carlos Sosa). For the implementation of the
11891 Drawer element in the Org Writer, we make use of a generic Block
11892 container with attributes. The presence of a `drawer` class defines
11893 that the `Div` constructor is a drawer. The first class defines the
11894 drawer name to use. The key-value list in the attributes defines
11895 the keys to add inside the Drawer. Lastly, the list of Block elements
11896 contains miscellaneous blocks elements to add inside of the Drawer.
11897 + Use `CUSTOM_ID` in properties (Albert Krewinkel). The `ID` property is
11898 reserved for internal use by Org-mode and should not be used.
11899 The `CUSTOM_ID` property is to be used instead, it is converted to the
11900 `ID` property for certain export format.
11904 + Ignore `--incremental` unless output format is beamer (#2843).
11905 + Fix polyglossia to babel env mapping (Mauro Bieg, #2728).
11906 Allow for optional argument in square brackets.
11907 + Recognize `la-x-classic` as Classical Latin (Andrew Dunning).
11908 This allows one to access the hyphenation patterns in CTAN's
11910 + Add missing languages from hyph-utf8 (Andrew Dunning).
11911 + Improve use of `\strut` with `\minipage` inside tables
11912 (Jose Luis Duran). This improves spacing in multiline
11914 + Use `{}` around options containing special chars (#2892).
11915 + Avoid lazy `foldl`.
11916 + Don't escape underscore in labels (#2921). Previously they were
11918 + brazilian -> brazil for polyglossia (#2953).
11920 * HTML writer: Ensure mathjax link is added when math appears in footnote
11921 (#2881). Previously if a document only had math in a footnote, the
11922 MathJax link would not be added.
11924 * EPUB writer: set `navpage` variable on nav page.
11925 This allows templates to treat it differently.
11929 + Use docbook5 if `writerDocbook5` is set (Ivo Clarysse).
11930 + Properly handle `ulink`/`link` (Ivo Clarysse).
11934 + Unescape URIs in spine (#2924).
11935 + EPUB reader: normalise link id (Mauro Bieg).
11939 + Parse `moveTo` and `moveFrom` (Jesse Rosenthal).
11940 `moveTo` and `moveFrom` are track-changes tags that are used when a
11941 block of text is moved in the document. We now recognize these tags and
11942 treat them the same as `insert` and `delete`, respectively. So,
11943 `--track-changes=accept` will show the moved version, while
11944 `--track-changes=reject` will show the original version.
11945 + Tests for track-changes moving (Jesse Rosenthal).
11947 * ODT, EPUB, Docx readers: throw `PandocError` on unzip failure
11948 (Jesse Rosenthal) Previously, `readDocx`, `readEPUB`, and `readOdt`
11949 would error out if zip-archive failed. We change the archive extraction
11950 step from `toArchive` to `toArchiveOrFail`, which returns an Either value.
11952 * Markdown, HTML readers: be more forgiving about unescaped `&` in
11953 HTML (#2410). We are now more forgiving about parsing invalid HTML with
11954 unescaped `&` as raw HTML. (Previously any unescaped `&`
11955 would cause pandoc not to recognize the string as raw HTML.)
11959 + Fix pandoc title blocks with lines ending in 2 spaces (#2799).
11960 + Added `-s` to markdown-reader-more test.
11962 * HTML reader: fixed bug in `pClose`. This caused exponential parsing
11963 behavior in documnets with unclosed tags in `dl`, `dd`, `dt`.
11965 * MediaWiki reader: Allow spaces before `!` in MediaWiki table header
11968 * RST reader: Support `:class:` option for code block in RST reader
11971 * Org reader (all Albert Krewinkel, except where noted otherwise):
11973 + Stop padding short table rows.
11974 Emacs Org-mode doesn't add any padding to table rows. The first
11975 row (header or first body row) is used to determine the column count,
11976 no other magic is performed.
11977 + Refactor rows-to-table conversion. This refactors
11978 the codes conversing a list table lines to an org table ADT.
11979 The old code was simplified and is now slightly less ugly.
11980 + Fix handling of empty table cells, rows (Albert Krewinkel, #2616).
11981 This fixes Org mode parsing of some corner cases regarding empty cells
11982 and rows. Empty cells weren't parsed correctly, e.g. `|||` should be
11983 two empty cells, but would be parsed as a single cell containing a pipe
11984 character. Empty rows where parsed as alignment rows and dropped from
11986 + Fix spacing after LaTeX-style symbols.
11987 The org-reader was dropping space after unescaped LaTeX-style symbol
11988 commands: `\ForAll \Auml` resulted in `∀Ä` but should give `∀ Ä`
11989 instead. This seems to be because the LaTeX-reader treats the
11990 command-terminating space as part of the command. Dropping the trailing
11991 space from the symbol-command fixes this issue.
11992 + Print empty table rows. Empty table rows should not
11993 be dropped from the output, so row-height is always set to be at least 1.
11994 + Move parser state into separate module.
11995 The org reader code has become large and confusing. Extracting smaller
11996 parts into submodules should help to clean things up.
11997 + Add support for sub/superscript export options.
11998 Org-mode allows to specify export settings via `#+OPTIONS` lines.
11999 Disabling simple sub- and superscripts is one of these export options,
12000 this options is now supported.
12001 + Support special strings export option Parsing of special strings
12002 (like `...` as ellipsis or `--` as en dash) can be toggled using the `-`
12004 + Support emphasized text export option. Parsing of emphasized text can
12005 be toggled using the `*` option. This influences parsing of text marked
12006 as emphasized, strong, strikeout, and underline. Parsing of inline math,
12007 code, and verbatim text is not affected by this option.
12008 + Support smart quotes export option. Reading of smart quotes can be
12009 toggled using the `'` option.
12010 + Parse but ignore export options. All known export options are parsed
12012 + Refactor block attribute handling. A parser state attribute was used
12013 to keep track of block attributes defined in meta-lines. Global state
12014 is undesirable, so block attributes are no longer saved as part of the
12015 parser state. Old functions and the respective part of the parser state
12017 + Use custom `anyLine`. Additional state changes need to be made after
12018 a newline is parsed, otherwise markup may not be recognized correctly.
12019 This fixes a bug where markup after certain block-types would not be
12021 + Add support for `ATTR_HTML` attributes (#1906).
12022 Arbitrary key-value pairs can be added to some block types using a
12023 `#+ATTR_HTML` line before the block. Emacs Org-mode only includes these
12024 when exporting to HTML, but since we cannot make this distinction here,
12025 the attributes are always added. The functionality is now supported
12027 + Add `:PROPERTIES:` drawer support (#1877).
12028 Headers can have optional `:PROPERTIES:` drawers associated with them.
12029 These drawers contain key/value pairs like the header's `id`. The
12030 reader adds all listed pairs to the header's attributes; `id` and
12031 `class` attributes are handled specially to match the way `Attr` are
12032 defined. This also changes behavior of how drawers of unknown type
12033 are handled. Instead of including all unknown drawers, those are not
12034 read/exported, thereby matching current Emacs behavior.
12035 + Use `CUSTOM_ID` in properties. See above on Org writer changes.
12036 + Respect drawer export setting. The `d` export option can be used
12037 to control which drawers are exported and which are discarded.
12038 Basic support for this option is added here.
12039 + Ignore leading space in org code blocks (Emanuel Evans, #2862).
12040 Also fix up tab handling for leading whitespace in code blocks.
12041 + Support new syntax for export blocks. Org-mode version 9
12042 uses a new syntax for export blocks. Instead of `#+BEGIN_<FORMAT>`,
12043 where `<FORMAT>` is the format of the block's content, the new
12044 format uses `#+BEGIN_export <FORMAT>` instead. Both types are
12046 + Refactor `BEGIN...END` block parsing.
12047 + Fix handling of whitespace in blocks, allowing content to be indented
12048 less then the block header.
12049 + Support org-ref style citations. The *org-ref* package is an
12050 org-mode extension commonly used to manage citations in org
12051 documents. Basic support for the `cite:citeKey` and
12052 `[[cite:citeKey][prefix text::suffix text]]` syntax is added.
12053 + Split code into separate modules, making for cleaner code and
12056 * Added `docbook5` template.
12058 * `--mathjax` improvements:
12060 + Use new CommonHTML output for MathJax (updated default MathJax URL,
12062 + Change default mathjax setup to use `TeX-AMS_CHTML` configuration.
12063 This is designed for cases where the input is always TeX and maximal
12064 conformity with TeX is desired. It seems to be smaller and load faster
12065 than what we used before. See #2858.
12066 + Load the full MathJax config to maximize loading speed (KolenCheung).
12068 * Bumped upper version bounds to allow use of latest packages
12069 and compilation with ghc 8.
12071 * Require texmath 0.8.6.2. Closes several texmath-related bugs (#2775,
12072 #2310, #2310, #2824). This fixes behavior of roots, e.g.
12073 `\sqrt[3]{x}`, and issues with sub/superscript positioning
12074 and matrix column alignment in docx.
12078 + Clarified documentation of `implicit_header_references` (#2904).
12079 + Improved documentation of `--columns` option.
12081 * Added appveyor setup, with artefacts (Jan Schulz).
12083 * stack.yaml versions: Use proper flags used for texmath, pandoc-citeproc.
12085 * LaTeX template: support for custom font families (vladipus).
12086 Needed for correct polyglossia operation with Cyrillic fonts and perhaps
12087 can find some other usages. Example usage in YAML metadata:
12090 - name: \cyrillicfont
12091 font: Liberation Serif
12092 - name: \cyrillicfonttt
12093 options: Scale=MatchLowercase
12096 * Create unsigned msi as build artifact in appveyor build.
12098 * On travis, test with ghc 8.0.1; drop testing for ghc 7.4.1.
12102## pandoc 1.17.0.3 (2016-03-24)
12104 * LaTeX writer: Fixed position of label in figures (#2813).
12105 Previously the label wasn't in the right place, and `\ref`
12106 wouldn't work properly.
12107 * Added .tei test files to pandoc.cabal so they'll be included
12108 in tarball (#2811).
12109 * Updated copyright dates.
12111## pandoc 1.17.0.2 (2016-03-23)
12113 * Fixed serious regression in `htmlInBalanced`, which caused
12114 newlines to be omitted in some raw HTML blocks in Markdown
12117## pandoc 1.17.0.1 (2016-03-21)
12119 * File scope is no longer used when there are no input files (i.e.,
12120 when input comes from stdin). Previously file scope was triggered
12121 when the `json` reader was specified and input came from `stdin`,
12122 and this caused no output to be produced. (Fix due to Jesse Rosenthal;
12123 thanks to Fedor Sheremetyev for calling the bug to our attention.)
12124 * Improved documentation of templates (#2797).
12126## pandoc 1.17 (2016-03-20)
12128 * Added `--file-scope` option (Jesse Rosenthal).
12129 By default pandoc operates on multiple files by first concatenating
12130 them (around extra line breaks) and then processing the joined file. So
12131 it only parses a multi-file document at the document scope. This has the
12132 benefit that footnotes and links can be in different files, but for
12133 some purposes it is useful to parse the individual files first
12134 and then combine their outputs (e.g. when the files use footnotes
12135 or links with the same labels). The `--file-scope` option causes
12136 pandoc to parse the files first, and then combine the parsed output,
12137 instead of combining before parsing. `--file-scope` is selected
12138 automatically for binary input files (which cannot be concatenated)
12139 and for pandoc json.
12141 * Add TEI Writer (Chris Forster) and `tei` output format.
12143 * Added a general `ByteStringReader` with warnings, used by the docx
12144 reader (API change, Jesse Rosenthal).
12146 * Add `readDocxWithWarnings` (API change, Jesse Rosenthal).
12148 * Changed type of `Shared.uniqueIdent` argument from
12149 `[String]` to `Set String`. This avoids performance problems in documents
12150 with many identically named headers (API change, #2671).
12152 * Removed `tex_math_single_backslash` from `markdown_github` options
12155 * Make language extensions as well as full language names
12156 trigger syntax highlighting. For example, `py` will now work as
12157 well as `python` (jgm/highlighting-kate#83).
12159 * Added `institute` variable to latex, beamer templates (Fraser
12160 Tweedale, Josef Svenningsson).
12162 * Docx reader (Jesse Rosenthal):
12164 + Handle alternate content. Some word functions (especially graphics)
12165 give various choices for content so there can be backwards compatibility.
12166 + Don't turn numbered headers into lists.
12167 + Docx Reader: Add state to the parser, for warnings
12168 + Update feature checklist in source code.
12169 + Get rid of `Modifiable` typeclass.
12170 + Add tests for adjacent hyperlinks.
12171 + Add a "Link" modifier to `Reducible`. We want to make sure that
12172 links have their spaces removed, and are appropriately smushed
12177 + Fixed behavior of base tag (#2777).
12178 If the base path does not end with slash, the last component
12179 will be replaced. E.g. base = `http://example.com/foo`
12180 combines with `bar.html` to give `http://example.com/bar.html`.
12181 If the href begins with a slash, the whole path of the base
12182 is replaced. E.g. base = `http://example.com/foo/` combines
12183 with `/bar.html` to give `http://example.com/bar.html`.
12184 + Rewrote `htmlInBalanced`. This version avoids an exponential
12185 performance problem with `<script>` tags, and it should be faster
12186 in general (#2730).
12187 + Properly handle an empty cell in a simple table (#2718).
12188 + Handle multiple `<meta>` tags with same name. Put them in a list
12189 in the metadata so they are all preserved, rather than (as before)
12190 throwing out all but one..
12194 + Improved pipe table parsing (#2765).
12195 + Allow `+` separators in pipe table cells. We already allowed
12196 them in the header, but not in the body rows, for some reason.
12197 This gives compatibility with org-mode tables.
12198 + Don't cross line boundary parsing pipe table row.
12199 Previously an Emph element could be parsed across the newline
12200 at the end of the pipe table row.
12201 + Use `htmlInBalanced` for `rawVerbatimBlock`, for better
12202 performance (#2730).
12203 + Fixed bug with smart quotes around tex math.
12207 + Handle interior `$` characters in math (#2743). For example,
12209 + `inlineCommand` now gobbles an empty `{}` after any command (#2687).
12210 This gives better results when people write e.g. `\TeX{}` in Markdown.
12211 + Properly handle LaTeX "math" environment as inline math (#2171).
12213 * Textile reader: Support `>`, `<`, `=`, `<>` text alignment attributes.
12216 * Org reader (Albert Krewinkel):
12218 + Prefix even empty figure names with "fig:" (#2643). The
12219 convention used by pandoc for figures is to mark them by prefixing
12220 the name with `fig:`. The org reader failed to do this if a figure
12222 + Refactor link-target processing (#2684).
12224 * ConTeXt writer: Fix whitespace at line beginning in line blocks (#2744).
12225 Thanks to @c-foster.
12227 * HTML writer: Don't include alignment attribute for default table columns.
12228 Previously these were given "left" alignment. Better to leave off
12229 alignment attributes altogether (#2694).
12231 * Markdown writer: Use hyphens for YAML metadata block bottom line, for
12232 better compatibility with other Markdown flavors (Henrik Tramberend).
12236 + Use image identifier to create a label and hypertarget for
12237 figures (Mauro Bieg).
12238 + Avoid double toprule in headerless table with caption (#2742).
12239 + Clean up options parser (Jesse Rosenthal).
12240 + Treat `memoir` template with `article` option as article, instead
12241 of treating all `memoir` templates as books.
12242 + Allow more flexible table alignment (Henrik Tramberend, #2665).
12243 New default is not to include `[c]` option (which is the default
12244 anyway if no positioning is specified). Now LaTeX emplates can
12245 control the overall table alignment in a document by setting the
12246 longtable length variables `LTleft` and `LTright`. For example,
12247 `\setlength\LTleft\parindent\setlength\LTright\fill`
12248 will create left-aligned tables that respect paragraph indentation.
12250 * Docx writer: Handle image alt text (#2754, Mauro Bieg).
12252 * Org writer - pass through RawInline with format "org".
12254 * DokuWiki writer: use `$$` for display math.
12256 * Custom writer: Pass attributes parameter to CaptionedImage (#2697).
12258 * Make protocol-relative URIs work again (#2737).
12260 * make_osx_package.sh: Use env variable for developer id certs.
12262 * Raise `tagsoup` lower bound to 0.13.7 to fix entity-related
12265 * Allow `zip-archive` 0.3.
12267 * Allow `aeson` 0.11.
12269## pandoc 1.16.0.2 (2016-01-12)
12271 * Depend on deepseq rather than deepseq-generics (fpco/stackage#1096).
12273 * Fixed regression in latex smart quote parsing (#2645).
12274 In cases where a match was not found for a quote, everything
12275 from the open quote to the end of the paragraph was being dropped.
12277## pandoc 1.16.0.1 (2016-01-10)
12279 * Fixed regression with `--latex-engine` (#2618). In 1.16
12280 `--latex-engine` raises an error if a full path is given.
12282 * Org reader: Fix function dropping subtrees tagged `:noexport`
12283 (Albert Krewinkel, #2628):
12285 * Markdown reader: renormalize table column widths if they exceed 100%
12288 * Textile reader: don't allow block HTML tags in inline contexts.
12289 The reader previously did allow this, following redcloth,
12290 which happily parses
12292 Html blocks can be <div>inlined</div> as well.
12296 <p>Html blocks can be <div>inlined</div> as well.</p>
12298 This is invalid HTML. The above sample now produces;
12300 <p>Html blocks can be</p>
12306 * Improved default template lookup for custom lua scripts (#2625).
12307 Previously, if you tried to do `pandoc -s -t /path/to/lua/script.lua`,
12308 pandoc would look for the template in
12309 `~/.pandoc/templates/default./path/to/lua/script.lua`.
12310 With this change it will look in the more reasonable
12311 `~/.pandoc/templates/default.script.lua`. This makes it possible to
12312 store default templates for custom writers.
12314 * RST, Markdown writers: Fixed rendering of grid tables with blank rows
12317 * LaTeX writer: restore old treatment of Span (#2624). A Span is
12318 now rendered with surrounding `{}`, as it was before 1.16.
12320 * Entity handling fixes: improved handling of entities like
12321 `⟨` that require a trailing semicolon. Allow uppercase
12322 `x` in numerical hexadecimal character references, working
12323 around a tagsoup bug.
12325 * `stack.yaml` - use lts-4.0, but with older aeson to avoid excessive
12326 memory use on compile. With aeson 0.10 we were getting an out of
12327 memory error on a 2GB Ubuntu 64-bit VM.
12329 * Improved deb package creation script. Made `DPKGVER` work.
12330 Renamed `COMMIT` to `TREE`. You should now be able to do
12331 `TREE=1.16.0.1 DPKGVER=2 make deb`.
12334## pandoc 1.16 (2016-01-02)
12336 * Added `Attr` field to `Link` and `Image` (Mauro Bieg, #261, API change).
12338 + Added syntax for link and image attributes to pandoc's Markdown.
12339 + Updated readers and writers to use link and image attributes
12341 + Support image attributes in Docx, Textile, RST readers.
12343 * Renamed link attribute extensions. The old `link_attributes` is
12344 now `mmd_link_attributes`, and `link_attributes` now enables the
12345 new pandoc-style link and image attributes (API change).
12346 Note: this change could break some existing workflows.
12348 * Implemented `SoftBreak` and new `--wrap` option (#1701, API change).
12349 Added threefold wrapping option.
12351 + Command line option: deprecated `--no-wrap`, added
12352 `--wrap=[auto|none|preserve]`
12353 + Added `WrapOption`, exported from `Text.Pandoc.Options`
12354 + Changed type of `writerWrapText` in `WriterOptions` from
12355 `Bool` to `WrapOption`.
12356 + Modified `Text.Pandoc.Shared` functions to allow `SoftBreak`.
12357 + Supported `SoftBreak` in readers and writers.
12359 * Text.Pandoc.Options: Added `writerDpi` to `WriterOptions` (API
12360 change, Mauro Bieg).
12362 * Added `--dpi` command-line option (Mauro Bieg).
12364 * Rationalized behavior of `--no-tex-ligatures` and `--smart` (#2541).
12365 This change makes `--no-tex-ligatures` affect the LaTeX reader
12366 as well as the LaTeX and ConTeXt writers. If it is used,
12367 the LaTeX reader will parse characters `` ` ``, `'`, and `-`
12368 literally, rather than parsing ligatures for quotation marks
12369 and dashes. And the LaTeX writer will print unicode quotation
12370 mark and dash characters literally, rather than converting
12371 them to the standard ASCII ligatures. Note that `--smart` has
12372 no effect on the LaTeX reader. `--smart` is still the default
12373 for all input formats when LaTeX or ConTeXt is the output format,
12374 *unless* `--no-tex-ligatures` is used.
12376 Some examples to illustrate the logic:
12379 % echo "'hi'" | pandoc -t latex
12381 % echo "'hi'" | pandoc -t latex --no-tex-ligatures
12383 % echo "'hi'" | pandoc -t latex --no-tex-ligatures --smart
12385 % echo "'hi'" | pandoc -f latex --no-tex-ligatures
12387 % echo "'hi'" | pandoc -f latex
12391 * Removed deprecated options `--offline` and `--html5`.
12393 * Fixed language code for Czech (`cs` not `cz`) (#2597).
12395 * Implemented `east_asian_line_breaks` extension (#2586).
12396 In `Text.Pandoc.Options`, added `Ext_east_asian_line_breaks` constructor
12397 to `Extension` (API change). This extension is like
12398 `ignore_line_breaks`, but smarter -- it only ignores line breaks
12399 between two East Asian wide characters. This makes it better suited
12400 for writing with a mix of East Asian and non-East Asian scripts.
12402 * Added support for PDF creation via `wkhtmltopdf`.
12403 To use this: `pandoc -t html5 -o result.pdf` (and add `--mathjax`
12404 if you have math.) Margins can be set using the variables
12405 `margin-top`, `margin-bottom`, `margin-left`, `margin-right`.
12406 Other styling can be done through CSS.
12408 * Fixed cite key parsing regression (jgm/pandoc-citeproc#201).
12409 We were capturing final colons as in `[@foo: bar]`; the citation id
12410 was being parsed as `@foo:`.
12414 + Fixed image syntax for local files (#2589).
12415 + Changed type of `writeICML` (Mauro Bieg).
12416 API change: It is now `WriterOptions -> Pandoc -> IO String`.
12417 Also handle new image attributes.
12418 + Intersperse line breaks instead of appending them to
12419 every `ParagraphStyleRange` (Mauro Bieg, #2501).
12420 + Add `Cite` style to citations (Mauro Bieg).
12421 + Added figure handling (#2590, Mauro Bieg).
12422 + Better handling of math. Instead of just printing the raw tex,
12423 we now try to fake it with unicode characters.
12425 * HTML writer: Include `example` class for example lists (#2524).
12427 * ODT/OpenDocument writer: improved image attributes (Mauro Bieg).
12429 + Support for percentage widths/heights
12430 + Use `Attr` instead of title to get dimensions from ODT walker
12431 to `writeOpenDocument`.
12435 + Support anchors in spans and divs with id elements
12436 (jgm/pandoc-citeproc#143).
12437 + Fixed code blocks (#1861).
12439 * Haddock writer: omit formatting inside links, which isn't supported
12440 by Haddock (#2515).
12442 * MediaWiki writer: Fixed spacing issues in table cells.
12444 + Start cell on new line unless it's a single Para or Plain
12446 + For single Para or Plain, insert a space after the `|` to
12447 avoid problems when the text begins with a character like
12450 * Beamer writer: mark frame as fragile when it contains verbatim (#1613).
12454 + Add support for GAP highlighting using listings (Raniere Silva).
12455 + Consider `header-includes` content as well as templates
12456 when determining whether to use csquotes (Andreas Lööw).
12457 + Create defaults for geometry using `margin-left` etc.
12458 If `geometry` has no value, but `margin-left`, `margin-right`,
12459 `margin-top`, and/or `-margin-bottom` are given, a default value
12460 for `geometry` is created from these. Note that these variables
12461 already affect PDF production via HTML5 with `wkhtmltopdf`.
12463 * ConTeXt writer: set default layout based on `margin-left`, etc.
12464 This sets up `\setuplayout` based on the variables `margin-left`,
12465 `margin-right`, `margin-bottom`, and `margin-top`, if no layout
12468 * Docx writer: better handling of PDF images. Previously we tried
12469 to get the image size from the image even if an explicit size was
12470 specified. Since we still can't get image size for PDFs, this made
12471 it impossible to use PDF images in docx. Now we don't try to get
12472 the image size when a size is already explicitly specified.
12474 * Markdown writer: use raw HTML for link/image attributes when
12475 the `link_attributes` extension is unset and `raw_html` is set (#2554).
12477 * MediaWiki reader: interpret markup inside `<tt>`, `<code>` (#2607).
12481 + Improved smart quote parsing (#2555). This fixes redering of
12483 + Use curly quotes for unmatched ` (#2555).
12484 + Allow blank space between braced arguments of commands (#2592).
12488 + Improved pipe table relative widths. Previously pipe table
12489 columns got relative widths (based on the header underscore lines)
12490 when the source of one of the rows was greater in width than the
12491 column width. This gave bad results in some cases where much of
12492 the width of the row was due to nonprinting material (e.g. link
12493 URLs). Now pandoc only looks at printable width (the width of a
12494 plain string version of the source), which should give better results.
12495 Thanks to John Muccigrosso for bringing up the issue.
12496 + Fixed parsing bug with macros. Previously macro definitions in
12497 indented code blocks were being parsed as macro definitions, not code.
12499 * Textile reader: skip over attribute in image source (#2515).
12500 We don't have a place yet for styles or sizes on images, but
12501 we can skip the attributes rather than incorrectly taking them
12502 to be part of the filename.
12504 * Docx reader: Handle dummy list items (Jesse Rosenthal).
12505 These come up when people create a list item and then delete the
12506 bullet. It doesn't refer to any real list item, and we used to ignore
12509 * CommonMark reader/writer rewritten to use latest `cmark`.
12511 * Fixed Emoji character definitions (#2523). There were many bugs in the
12514 * `Text.Pandoc.CSS`:
12516 + Added `pickStylesToKVs` function to extract multiple properties at
12517 once (API change, Mauro Bieg).
12518 + Parse CSS that doesn't contain the optional semicolon (Mauro Bieg).
12520 * `trypandoc`: sort drop-down lists.
12524 + Made `\euro` conditional on presence of character.
12525 for xelatex and lualatex, as it is for pdflatex (Andrew Dunning).
12526 + Moved `header-includes` before setting of title (Thomas Hodgson),
12527 to match the LaTeX template (jgm/pandoc-templates#168).
12528 + Added `section-titles` variable (defaults to true)
12529 to enable/suppress section title pages in beamer
12530 slide shows (Thomas Hodgson).
12531 + Moved beamer themes after fonts, so that themes can
12532 change fonts. (Previously the fonts set were being
12533 clobbered by lmodern.sty.) (Thomas Hodgson).
12535 * Beamer/LaTeX template changes (Thomas Hodgson):
12536 + Added `thanks` variable
12537 + Use `parskip.sty` when `indent` isn't set (fall back to using
12538 `setlength` as before if `parskip.sty` isn't available).
12539 + Use `biblio-style` with biblatex.
12540 + Added `biblatexoptions` variable.
12542 * LaTeX template changes:
12544 + Added `paper` after `$papersize$` variable in latex template.
12545 Thus you can say `papersize: a4` and the latex will contain
12546 `a4paper`. This change may break some existing workflows; if
12547 you currently specify `a4paper`, you'll get `a4paperpaper` which
12548 is meaningless. However, the change seems worth it, as it will
12549 make the `papersize` variable work uniformly across ConTeXt, LaTeX,
12550 and html->pdf via wkhtmltopdf.
12551 + Only pass options to color package if `colorlinks` is set
12553 + Make definition of `\euro` conditional in xelatex/lualatex,
12554 as it is already for pdflatex (Andrew Dunning).
12555 + Removed setting of `subject` in PDF metadata.
12556 This used to be set to the subtitle, but really the subtitle
12557 need not give the subject. Also, `subtitle` can contain formatting,
12558 so we'd need, at least, a plain text version for this.
12559 + Moved `header-includes` before setting of `\title`, `\author`,
12560 etc. This allows these macros to be redefined.
12561 + Use `\subtitle` command for `subtitle`, instead of tacking it
12562 on to the title as before. We give a no-op fallback definition if it
12563 is not defined. This change should produce much better results
12564 in classes that support `\subtitle`. With the default article
12565 class, which does not define `\subtitle`, subtitles will no
12566 longer be printed unless the user defines `\subtitle` and
12567 redefines `\maketitle`.
12568 + Moved redefinitions of `\paragraph` and `\subparagraph` to
12569 before header-includes.
12571 * Context template:
12573 + Use `simplefonts` for font loading (Paolo Rodríguez). This is
12574 needed for things to work on ConTeXt stable from TeXLive 2015.
12575 + Revert use of `\setuphead` in title block (Andrew Dunning,
12578 * Update LaTeX/ConTeXt link colour usage (Andrew Dunning).
12580 * Fixed man template so disabling hyphenation actually works.
12581 The command needs to come after .TH.
12583 * Added 'navigation' variable to beamer template (#2543).
12584 Valid values are `empty` (the default), `horizontal`, `vertical`,
12585 and `frame`. Note that this changes the default behavior from
12586 `horizontal` to `empty`. Closes #2543.
12588 * Added `toc` to HTML slide format templates (Andrew Dunning),
12589 so that `--toc` creates a contents slide.
12591 * Added `stack.full.yaml` to build `pandoc-citeproc` as well.
12593 * Allow pipe tables with no body rows (#2556).
12594 Previously this raised a runtime error.
12596 * Shared: Improved `fetchItem` so that `C:/Blah/Blah.jpg` isn't treated
12597 as URL. The Haskell URI parsing routines will accept "C:" as a
12598 scheme, so we rule that out manually. This helps with
12599 `--self-contained` and absolute Windows paths.
12601 * Define a `meta-json` variable for all writers (#2019). This contains
12602 a JSON version of all the metadata, in the format selected for the
12603 writer. So, for example, to get just the YAML metadata, you can run
12604 pandoc with the following custom template: `$meta-json$`. The intent
12605 is to make it easier for static site generators and other tools to get
12608 * Document limitations of --self-contained (#2553).
12610 * Improved Citations section of README (#2551). Added information
12611 about `link-citations` and a link to the pandoc-citeproc man page.
12613 * `ImageSize`: use `safeRead` instead of `readMaybe`, which isn't
12616 * Allow .adoc file extension for AsciiDoc (Andrew Dunning).
12618 * Improved implicit pandoc-citeproc inclusion.
12619 The filter pandoc-citeproc is automatically used when
12620 `--bibliography` is specified on the command line, unless
12621 `--natbib` or `--biblatex` is used. However, previously this
12622 only worked if `--bibliography` was spelled out in full, and not
12623 if `--biblio` was used.
12625 * reveal.js: Interpret pauses correctly for all headers (#2530).
12626 Previously, when using headers below the slide level, pauses are left
12627 uninterpreted into pauses. In my opinion, unexpected behavior but
12628 intentional looking at the code.
12630 * Remove redundant `center` variable for reveal.js (Andrew Dunning).
12632 * Parsing: Add `extractIdClass`, modified type of `KeyTable` (Mauro
12635 * ImageSize: Added functions for converting between image dimensions
12638 * Use lts-3.18 in stack.yaml. This avoids Windows build
12639 issues with the HTTP library.
12641 * Bump version bounds for dependencies.
12643## pandoc 1.15.2.1 (2015-11-16)
12645 * Added two missing test files, and `stack.yaml`, to
12646 `extra-source-files` so they're included in the source tarball.
12648 * reveal.js template: Fixed parallaxBackground options.
12649 `parallaxBackgroundHorizontal` and `parallaxBackgroundVertical`
12650 need integer values, not strings. (Vaughn Iverson)
12652## pandoc 1.15.2 (2015-11-15)
12654 * `pandoc my.md -t context -o my.pdf` will now create a PDF using
12655 ConTeXt rather than LaTeX (#2463).
12657 * Fixed omitted `url(...)` in CSS data-uri with `--self-contained` (#2489).
12659 * Added `emoji` Markdown extension, enabled by default in `markdown_github`
12660 (#2523). Added `Ext_emoji` to `Extension` in `Text.Pandoc.Options`
12663 * `Text.Pandoc.Readers.HTML.parseTags`: Fixed over-eager raw HTML inline
12664 parsing (#2469). Tightened up the inline HTML parser so it disallows
12667 * Derive `Generic` instances for the types in `Text.Pandoc.Options`.
12671 + Fix paragraph/list interaction (Albert Krewinkel, #2464).
12672 Paragraphs can be followed by lists, even if there is no blank line
12673 between the two blocks. However, this should only be true if the
12674 paragraph is not within a list, were the preceding block should be
12675 parsed as a plain instead of paragraph (to allow for compact lists).
12676 Thanks to @rgaiacs for bringing this up.
12677 + Allow toggling header args (Albert Krewinkel, #2269).
12678 Org-mode allows to skip the argument of a code block header argument if
12679 it's toggling a value. Argument-less headers are now recognized,
12680 avoiding weird parsing errors.
12681 + Fix markup parsing in headers (Albert Krewinkel, #2504).
12682 Markup as the very first item in a header wasn't recognized. This was
12683 caused by an incorrect parser state: positions at which inline markup
12684 can start need to be marked explicitly by changing the parser state.
12685 This wasn't done for headers. The proper function to update the state
12686 is now called at the beginning of the header parser, fixing this issue.
12687 + Fix emphasis rules for smart parsing (Albert Krewinkel, #2513).
12688 Smart quotes, ellipses, and dashes should behave like normal quotes,
12689 single dashes, and dots with respect to text markup parsing.
12690 + Require whitespace around definition list markers (#2518).
12691 This rule was not checked before, resulting in bugs with footnotes
12692 and some link types.
12696 + Pipe tables with long lines now get relative cell widths (#2471).
12697 If a pipe table contains a line longer than the column width (as set by
12698 `--columns` or 80 by default), relative widths are computed based on the
12699 widths of the separator lines relative to the column width. This should
12700 solve persistent problems with long pipe tables in LaTeX/PDF output, and
12701 give more flexibility for determining relative column widths in other
12702 formats, too. For narrower pipe tables, column widths of 0 are used,
12703 telling pandoc not to specify widths explicitly in output formats that
12705 + Improved parser for `mmd_title_block`. We now allow blank metadata
12706 fields. These were explicitly disallowed before.
12707 + Citation keys can now contain `://`, so URLs and DOIs can be used
12708 as citation keys (jgm/pandoc-citeproc#166).
12710 * Beamer template: fix incompatibility of section slides with natbib.
12711 Natbib (and presumably biblatex) bibliography commands create
12712 their own section. Since these are in frame environments,
12713 we have an incompatibility with the `\AtBeginSection` macro
12714 which creates a special frame when a new section occurs.
12715 (We can't have a frame inside another frame.) This change disables
12716 `\AtBeginSection` inside bibliography slides. Thinks to Yihui Xie for
12717 bringing the problem to my attention. This supersedes #145. See
12720 * Textile reader: don't do smart punctuation unless explicitly asked
12721 (#2480). Note that although smart punctuation is part of the textile
12722 spec, it's not always wanted when converting from textile
12723 to, say, Markdown. So it seems better to make this an option.
12725 * LaTeX reader: Handle `comment` environment (Arata Mizuki).
12726 The `comment` environment is handled in a similar way to the
12727 `verbatim` environment, except that its content is discarded.
12729 * Docx reader: Follow relationships correctly in foot/endnotes (#2258,
12730 Jesse Rosenthal). This fixes a problem with links in notes.
12732 * LaTeX and ConTeXt writers: support `lang` attribute on divs and spans
12733 (mb21). For LaTeX, also collect `lang` and `dir` attributes on spans and
12734 divs to set the `lang`, `otherlangs` and `dir` variables if they aren’t set
12739 + Use proper command for `\textarabic` (mb21).
12740 + Added `de-CH-1901`, fixed `el-polyton` in `toPloyglossia` (Nick Bart).
12741 + Use `\hypertarget` and `\hyperlink` for links. This works correctly
12742 to link to Div or Span elements. We now don't bother defining `\label`
12743 for Div or Span elements. Closes jgm/pandoc-citeproc#174.
12744 + Avoid footnotes in list of figures (#1506).
12745 + Properly handle footnotes in captions (#1506).
12746 + Add `\protect` to `\hyperlink` (#2490). Thanks to Hadrien Mary.
12747 + Set `colorlinks` if `linkcolor`, `urlcolor`, `citecolor`, or
12748 `toccolor` is set (#2508).
12750 * Textile writer: support start number in ordered lists (#2465).
12752 * OpenDocument writer: Allow customization of opendocument
12753 automatic styles. Automatic styles can now be inserted in the
12754 template, which now provides the enclosing `<office:automatic-styles>`
12757 * Docx writer: insert space between footnote reference and note (#2527).
12758 This matches Word's default behavior.
12760 * EPUB writer: don't download linked media when `data-external` attribute
12761 set (#2473). By default pandoc downloads all linked media and includes it
12762 in the EPUB container. This can be disabled by setting `data-external` on
12763 the tags linking to media that should not be downloaded. Example:
12765 <audio controls="1">
12766 <source src="http://www.sixbarsjail.it/tmp/bach_toccata.mp3"
12767 type="audio/mpeg"></source>
12770 * HTML writer: use width on whole table if col widths sum to < 100%.
12771 Otherwise some browsers display the table with the columns
12772 separated far apart.
12774 * AsciiDoc template: Fix `author` and `date`; add `keywords`,
12775 `abstract` (Andrew Dunning).
12777 * HTML-based templates (Andrew Dunning):
12779 + Use en dash instead of hyphen between title prefix and title.
12780 + Add `keywords` to metadata.
12781 + Add `lang`, `dir`, `quotes` where missing.
12782 + Always make author and date display conditional.
12783 + Updated dzslides template from source.
12785 * Man template: make "generated by" comment conditional.
12787 * LaTeX, Beamer templates:
12789 + Add `babel-otherlangs` for language divs/spans; `babel-newcommands`,
12790 filled by commands that make babel understand the polyglossia-style
12791 language directives (mb21, #137).
12792 + Improved formatting of conditionals; `$for$` is always provided to allow
12793 multiple options (Andrew Dunning, #141).
12794 + Use `Ligatures=TeX` rather than `Mapping=tex-text` with `fontspec`
12795 to improve support for LuaTeX (Andrew Dunning, #135).
12796 + Revise `hyperref` usage (Andrew Dunning, #139, #141):
12797 - use same options for all LaTeX engines;
12798 - add `subtitle` and `keywords` to PDF metadata;
12799 - do not override `hyperref` link coloring without user input, effectively making
12800 the `hidelinks` option the default (removed as a separate variable);
12801 - link colors can be enabled (using a slightly darker version of the old
12802 defaults) using a new `colorlinks` variable, automatically used by
12803 the LaTeX writer when custom colors are specified;
12804 - `pdfborder={0 0 0}` is automatically set by `hyperref` with
12805 `colorlinks`, and is only applied if `colorlinks` is disabled.
12807 * ConTeXt template (Andrew Dunning):
12809 + New variables for controlling styles: `linkstyle`, `linkcolor`,
12810 `linkcontrastcolor`, `layout`, `pagenumbering`, `whitespace`, `indenting`,
12811 `interlinespace`, `headertext`, `footertext`, `mainfont`, `sansfont`,
12812 `monofont`, `mathfont`, `fontsize`.
12813 + Default template no longer supports MkII.
12814 + Improve writing of title block (suppressing numbering of first page).
12815 + Add `title` `subtitle`, `author`, `date`, `keywords` to PDF metadata.
12816 + Support `subtitle`, `abstract`.
12817 + Support list of figures (`lof`), list of tables (`lot`).
12818 + Disable link styling by default.
12819 + Define styles for all section types.
12820 + Enable microtype.
12821 + Improved formatting of conditionals.
12823 * Beamer template: added code to prevent slide breaks inside paragraphs
12824 (#2422, thanks to Nick Bart). This will matter, in practice, only when
12825 `allowframebreaks` is used. It is especially helpful for bibliography
12828 * OpenDocument template: Add `<office:automatic-styles>` tag around
12829 automatic styles. The writer now longer provides this (see #2520).
12831 * Restored Text.Pandoc.Compat.Monoid.
12833 * Do not export (<>) from custom Prelude. The Prelude now matches
12834 base 4.8 Prelude's API.
12836 * Don't use custom prelude with ghc 7.10. Use the custom prelude
12837 only for earlier versions. This change makes `stack ghci` and
12838 `cabal repl` work (#2503), at least with ghc 7.10.
12840 * Changed § to % in operators from Odt.Arrows.Utils (#2457).
12841 This prevents problems building haddocks with "C" locale.
12843 * Change default for old-locale flag to False.
12845 * Use stack in deb, osx, and Windows package generators.
12847 * Added Vagrantfile for building deb in vm.
12848 This should help in automating binary package creation. 'make package'
12849 will make the package. 'make package COMMIT=blah' will make the package
12854 + Consistent capitalization for pandoc and Markdown.
12855 + Fixed `auto_identifiers` examples (Benoit Schweblin).
12856 + Improved documentation of template variables (Andrew Dunning).
12858## pandoc 1.15.1 (2015-10-15)
12860 * `pandocVersion` is now defined in `Text.Pandoc.Shared`
12861 and reexported from `Text.Pandoc` (Alex Vong). This allows
12862 writers to access it. (Alex Vong) (API change)
12864 * For `markdown_mmd`, add: `implicit_figures`, `superscripts`,
12865 `subscripts` (#2401).
12867 * Added `odt` as input format (MarLinn). Added new module
12868 `Text.Pandoc.Reader.ODT` (API change). Fully implemented features:
12869 Paragraphs, Headers, Basic styling, Unordered lists, Ordered lists,
12870 External Links, Internal Links, Footnotes, Endnotes, Blockquotes.
12871 Partly implemented features: Citations, Tables.
12875 + Add basic tests for each header style (Ophir Lifshitz).
12876 + Add implicit header ref tests for headers with spaces (Ophir Lifshitz).
12877 + Skip spaces in headers (Ophir Lifshitz).
12878 + Handle 'id' and 'class' in parsing key/value attributes (#2396).
12879 `# Header {id="myid" class="foo bar"}`
12880 is now equivalent to `# Header {#myid .foo .bar}`.
12881 + Use '=' instead of '#' for atx-style headers in markdown+lhs.
12882 (Kristof Bastiaensen)
12883 + Pipe tables: allow indented columns. Previously the left-hand column
12884 could not start with 4 or more spaces indent. This was inconvenient
12885 for right-aligned left columns. Note that the first (header column)
12886 must still have 3 or fewer spaces indentation, or the table will be
12887 treated as an indented code block.
12888 + Fix regression: allow HTML comments containing `--`.
12889 Technically this isn't allowed in an HTML comment, but
12890 we've always allowed it, and so do most other implementations.
12891 It is handy if e.g. you want to put command line arguments
12896 + Don't eat excess whitespace after macros with only optional
12898 + Support longtable (#2411).
12899 + Implement `\Cite` (#2335).
12900 + Support abstract environment. The abstract populates an
12901 `abstract` metadata field.
12902 + Properly handle booktabs lines. Lines aren't part of the
12903 pandoc table model, so we just ignore them (#2307).
12907 + Handle type attribute on ol, e.g. `<ol type="i">` (#2313).
12908 + Updated for new automatic header attributes.
12909 + Add auto identifiers if not present on headers. This makes
12910 TOC linking work properly.
12911 + Detect `font-variant` with `pickStyleAttrProps` (Ophir Lifshitz).
12912 + Test `<ol>` type, class, and inline list-style(-type) CSS
12914 + Better handling of "section" elements (#2438). Previously
12915 `<section>` tags were just parsed as raw HTML blocks. With
12916 this change, section elements are parsed as Div elements with
12917 the class "section".
12919 * MediaWiki reader: handle unquoted table attributes (#2355).
12923 + Added proper support for DocBook `xref` elements (Frerich Raabe).
12924 Added `dbContent` field to reader state, so we can lookup
12926 + Handle `informalexample` (#2319).
12930 + Create special punctuation test (Ophir Lifshitz).
12931 + Parse soft, no-break hyphen elements (Ophir Lifshitz).
12932 + Updated headers test (Ophir Lifshitz). Replaced `styles.xml`
12933 in `headers.docx` with pandoc's current `styles.xml`, which
12934 contains styles for Heading 1 through 6. Added Heading 4
12935 through 7 to the test document. Note that Heading 7 is not
12936 parsed as a Heading because there is no Heading 7 style.
12938 * RST reader: better handling of indirect roles.
12939 Previously the parser failed on this kind of case
12941 .. role:: indirect(code)
12943 .. role:: py(indirect)
12948 Now it correctly recognizes `:py:` as a code role.
12952 + Add auto identifiers if not present on headers
12953 (#2354, Juliusz Gonera).
12954 + Allow verse blocks to contain empty lines (#2402,
12957 * EPUB reader: stop mangling external URLs (#2284).
12961 + Don't insert `\ ` when complex expression in matched pairs.
12962 E.g. `` [:sup:`3`] `` is okay; you don't need `` [:sup:`3`\ ] ``.
12963 + Ensure that `\ ` is inserted when needed before Cite and Span
12964 elements that begin with a "complex" element (jgm/pandoc-citeproc#157).
12965 + Normalize headers only in "standalone" mode (#2394).
12967 * Haddock writer: escape `*` and `^` (G. Bataille).
12971 + In TOC, add links to headers (#829).
12972 + Use unicode super/subscripts for digits in plain output
12973 (when the `superscripts` and `subscripts` extensions are
12978 + Moved invalid character stripping to `formattedString`.
12979 This avoids an inefficient generic traversal (#2356).
12980 + Use user data directory for `reference.docx` archive.
12981 This allows the test suite to work without installing pandoc first.
12982 It also brings the docx writer in line with the odt writer.
12983 + Tests: docx writer tests now use `../data` for data directory.
12984 This allows tests to be run without installing first.
12985 + Tests: Use real jpg (not empty) for docx tests to avoid warning.
12989 + Fixed detection of 'chapters' from template.
12990 If a documentclass isn't specified in metadata, but the
12991 template has a hardwired bookish documentclass, act as if
12992 `--chapters` was used. This was the default in earlier
12993 versions, but it has been broken for a little while.
12994 + Correctly recognize book documentclass in metadata (#2395).
12995 + Set language-related variables automatically, depending
12996 on the value of the `lang` field, which is now always
12997 assumed to be in BCP47 format (mb21, #1614, #2437).
12998 + Add `\protect` to `\hyperdef` in inline context. This way we
12999 don't get an error when this is used as a moveable argument (#2136).
13000 + Support all frame attributes in Beamer.
13001 + Percent-encode more special characters in URLs (#1640, #2377).
13002 The special characters are '<','>','|','"','{','}','[',']','^', '`'.
13006 + Update KaTeX JS and CSS versions (Emily Eisenberg).
13007 + For dzslides, add `role="note"` for speaker notes (#1693).
13008 + Percent-encode more special characters in URLs (#1640, #2377).
13009 The special characters are '<','>','|','"','{','}','[',']','^', '`'.
13010 + Render Div with class `section` as `<section>` in HTML5.
13014 + In TOC, replace literal `<br/>` with space (#2105).
13015 + With `--webtex`, include image file rather than `data:` URI (#2363).
13017 * Native writer: format Div properly, with blocks separated.
13019 * Support bidirectional text output with XeLaTeX, ConTeXt and HTML
13024 + Add missing Header 6 style (steel blue) (Ophir Lifshitz).
13025 + Correct `outlineLvl` for Header styles (Ophir Lifshitz).
13029 + Beamer: Add `innertheme`, `outertheme` variables
13030 (Guilhem Bonnefille, #121). Add space after colon in figure caption.
13031 Integrate recent font and language updates from LaTeX template;
13032 allow use of `mainfont` variable for changing the slide text
13033 in XeTeX and LuaTeX (Andrew Dunning, #131).
13034 + LaTeX: Add `mainfontoptions`, `sansfontoptions`,
13035 `monofontoptions`, `mathfontoptions`, `fontfamilyoptions`
13036 (Andrew Dunning, #122). Support handling of bidirectional
13037 text (mb21, #120). Improve reliability of superscripts/subscripts
13038 under XeTeX and prevent letters and numbers from appearing on a
13039 different baseline by removing use of the `realscripts` package
13040 (via `xltxtra`). To restore use of OpenType characters for these
13041 features under XeTeX or LuaTeX, add `\usepackage{realscripts}` to
13042 `header-includes` (Andrew Dunning, #130). Remove redundant
13043 reference to `xunicode` (Andrew Dunning, #130). Add `fontenc`,
13044 `indent`, `subparagraph` variables (Andrew Dunning).
13045 Allow use of `hidelinks` variable for `hyperref` package (Hugo Roy,
13046 #113). Prevent package clash with `tufte-latex` and other classes that
13047 include `hyperref` or `color` (Xavier Olive, #115).
13048 + ConTeXt: Support handling of bidirectional text (mb21, #120).
13049 + LaTeX and ConTeXt: Use more specific language variables.
13050 Instead of directly using `lang`, we now use `babel-lang` and
13051 `polyglossia-lang` and `context-lang`. These variables are set by
13052 the writers to the necessary values, based on the `lang` variable
13053 (which now always takes a value in BCP47 format). (mb21, #114, #129).
13054 + HTML: Support handling of bidirectional text (mb21, #120).
13055 Move HTML5 shiv after CSS and fix URL (Andrew Dunning).
13056 Add dir attribute in html5 (Andrew Dunning).
13057 + reveal.js: Add `controls`, `progress` variables (Grégoire Pineau, #127).
13058 Add `width`, `height` variables (Anrew Dunning). Update template
13059 from 3.1 source (Andrew Dunning). All configuration options are now
13060 available as variables, but are only be included if set (reveal.js
13061 uses defaults otherwise).
13062 + man: Added comment stating that the page is autogenerated by pandoc,
13063 giving version. Added `adjusting` and `hyphenate` variables
13066 * epub.css: added selectors for nested emphasis (Pablo Rodriguez).
13068 * MediaBag: ensure that `/` is always used as path separator.
13070 * `sample.lua`: define `CaptionedImage`, add newline at end (#2393).
13072 * Added `--bash-completion` option. This generates a bash completion
13073 script. To use: `eval "$(pandoc --bash-completion)"`.
13075 * Text.Pandoc.Error: Define Typeable and Exception instances
13076 for PandocError (#2386).
13078 * Text.Pandoc.Parsing: `toKey`: strip off outer brackets.
13079 This makes keys with extra space at the beginning and end
13086 will now be a link to bar (it wasn't before).
13088 * Text.Pandoc: disable `auto_identifiers` for epub.
13089 The epub writer inserts its own auto identifiers;
13090 this is more complex due to splitting into "chapter" files.
13092 * Renamed Text.Pandoc.Compat.Locale -> Text.Pandoc.Compat.Time.
13093 It now reexports Data.Time.
13095 * Use custom Prelude to avoid compiler warnings.
13097 + The (non-exported) prelude is in prelude/Prelude.hs.
13098 + It exports Monoid and Applicative, like base 4.8 prelude,
13099 but works with older base versions.
13100 + It exports (<>) for mappend.
13101 + It hides 'catch' on older base versions.
13103 * Added a `stack.ymal` and stack install instructions to INSTALL.
13105 * Clarified what is "out of scope" in README and CONTRIBUTING.md.
13107 * Added note to CONTRIBUTING.md about ghc versions and travis.
13109 * Clarify docs on block quotes. The space after `>` is optional (#2346).
13111 * Removed obsolete reference to default.csl (#2372).
13113 * List all styles in manual for `--reference-docx` (Chris Black)
13115 * Don't capitalize header links in man page.
13117 * Added section on repl to CONTRIBUTING.md.
13119 * README: Added space after backslash in image example (#2329).
13121 * Document details of citation locator terms (Nick Bart).
13123 * Fixed some internal links in README (#2309).
13125 * Improve CSL documentation, variables documentations,
13126 links, and cross-references in README. (Andrew Dunning)
13128 * Fix build failure with `--flags=-https` (Sergei Trofimovich).
13130 * Use `newManager` instead of `withManager` in recent `http-client`.
13131 This avoids a deprecation warning.
13133 * Allow building with latest versions of http-types,
13134 HUnit, criterion, syb, aeson.
13136 * Updated benchmark program for new criterion API.
13138 * Setup.hs: rewrite so as not to use process, directory, filepath.
13139 Using anything outside base is dangerous, since older
13140 versions of ghc may link against two different versions.
13142 * Added appveyor (Windows continuous integration) builds.
13144 * New `.travis.yml`. Autgenerated using `make_travis_yml.hs`.
13145 This script has been modified in a few ways, e.g. to add `GHCOPTS`.
13146 `make .travis.yml` regenerates it based on the tested-with
13147 field of the cabal file.
13149## pandoc 1.15.0.6 (2015-07-15)
13151 * `--self-contained`: Fixed overaggressive CSS minimization
13152 (#2301, 2286). Previously `--self-contained` wiped out all
13153 spaces in CSS, including semantically significant spaces.
13154 This was a regression from 1.14.x.
13156 * Markdown reader: don't allow bare URI links or autolinks in link
13157 label (#2300). Added test cases.
13159 * `Text.Pandoc.Parsing`, `uri`: Improved bare autolink detection (#2299).
13160 Previously we disallowed `-` at the end of an autolink,
13161 and disallowed the combination `=-`. This commit liberalizes the
13162 rules for allowing punctuation in a bare URI, and adds test cases.
13163 One potential drawback is that you can no longer put a bare
13164 URI in em dashes like this:
13165 `this uri---http://example.com---is an example.`
13166 But in this respect we now match github's treatment of bare URIs.
13168 * HTML writer: support speaker notes in dzslides.
13169 With this change `<div class="notes">` and also `<div class="notes"
13170 role="note">` will be output if `-t dzslides` is used. So we can
13171 have speaker notes in dzslides too. Thanks to maybegeek.
13173 * Updated dzslides template.
13175 * Improved documentation of options to print system default files (#2298).
13176 `--print-default-data-file` and `--print-default-template`.
13178 * DokuWiki writer: use `$..$` for Math instead of `<math>..</math>`
13179 (Tiziano Müller). MathJax seems currently to be the only maintained
13180 math rendering extension for DokuWiki.
13182 * `Text.Pandoc.Shared`: Changed `hierarchicalize` so it treats references
13183 div as top-level header (#2294). This fixes a bug with `--section-divs`,
13184 where the final references section added by pandoc-citeproc, enclosed in
13185 its own div, got nested in the div for the section previous to it.
13187 * Allow vector 0.11.
13189 * Require cmark > 0.4.
13191## pandoc 1.15.0.5 (2015-07-10)
13193 * HTML writer: Fixed email javascript obfuscation with `mailto:`
13194 URLs (#2280). This fixes a potential security issue. Because
13195 single quotes weren't being escaped in the link portion, a
13196 specially crafted email address could allow javascript code injection.
13198 * RST reader: allow inline formatting in definition list field
13199 names (Lars-Dominik Braun).
13201 * PDF: Make sure `--latex-engine-opt` goes before the filename
13202 on the command line. LaTeX needs the argument to come after
13203 the options (#1779).
13205 * CommonMark writer: fixed tags used for super/subscript.
13207 * ConTeXt template: activate hanging indent for definition lists
13210 * Make cabal require `hsb2hs` >= 0.3.1 if `embed_data_files` specified.
13211 This is done by adding `hookedPrograms` in `Setup.hs`, which allows us
13212 to include `hsb2hs` in Build-Tools in cabal.
13214 * Improved Windows installer (thanks to nkalvi).
13216 + When per-machine installation is chosen, the system path
13217 is updated instead of the user's.
13218 + An appropriate default is used for per-machine installation
13220 + Admin privileges are no longer required for a per-user install
13222 * Travis: unpack sdist for build to catch packaging bugs.
13224 * Improved documentation on where user templates go (#2272).
13226## pandoc 1.15.0.4 (2015-07-03)
13228 * Added pandoc.1 man page to the repository. It is no longer
13229 built as part of the cabal build process. (This proved too
13230 fragile.) pandoc.1 can be regenerated (`make man/pandoc.1`)
13231 when `README` is changed.
13233 * Copying of the man page now respects `--destdir` (#2262).
13235 * Improved error messages for filters. User is now informed if
13236 the filter requires an interpreter that isn't found in the path,
13237 or if the filter returns an error status.
13239## pandoc 1.15.0.3 (2015-07-02)
13241 * Ensure target directory is created when installing man page.
13243## pandoc 1.15.0.2 (2015-07-02)
13245 * Added files needed for building man page to Extra-Source-Files.
13247## pandoc 1.15.0.1 (2015-07-01)
13249 * Man page is now built and installed as part of the cabal build
13250 process. Removed Makefile target for man page.
13252## pandoc 1.15 (2015-07-01)
13254 * Man page changes:
13256 + Removed `--man1`, `--man5` options (breaking change).
13257 + Removed `Text.Pandoc.ManPages` module (breaking API change).
13258 + Makefile target for `man/man1/pandoc.1`. This uses pandoc to
13259 create the man page from README using a custom template and filters.
13260 + Added `man/` directory with template and filters needed to build
13262 + We no longer have two man pages: `pandoc.1` and `pandoc_markdown.5`.
13263 Now there is just pandoc.1, which has all the content from README.
13264 This change was needed because of the extensive cross-references
13265 between parts of the README.
13266 + Removed old `data/pandoc.1.template` and
13267 `data/pandoc_markdown.5.template`.
13269 * OpenDocument writer: Do not add a carriage return after a hard
13270 line break (Michael Chladek).
13274 + use `\goto` for internal links.
13275 + Added a `%` at end for `\reference` to avoid spurious space.
13277 * Ignore sandbox on 'make quick'
13279## pandoc 1.14.1 (2015-06-30)
13281 * Added `--man1` and `--man5` options to pandoc, allowing
13282 pandoc to generate its own man pages. Man pages are no longer
13283 automatically generated in the build process (the process for
13284 this was too complex and prone to failure, #2190). The
13285 `make-pandoc-man-pages` executable has been removed. The
13286 `man/` directory has been removed, and man page templates
13287 have been moved to `data/`. NOTE TO PACKAGERS: You will no
13288 longer find pandoc's man pages in `man/`, but you can
13289 generate them using `pandoc --man1 > pandoc.1` and `pandoc
13290 --man5 > pandoc_markdown.5`.
13292 * Added new unexported module: `Text.Pandoc.ManPages`.
13294 * `README` now acts like a data file (even though it isn't in
13295 `data/`). So, for example, `pandoc --print-default-data-file README`
13296 will produce the README.) This change was required for the `--man1`
13297 and `--man5` options, since the man pages are produced from the
13298 README, but it may be useful for other purposes as well.
13300 * Allow `reference.docx` and `reference.odt` to be used with
13301 `--print-default-data-file` and to shadow defaults if placed in
13302 the user data directory. Note that as of 1.14, we no longer
13303 include these files as data files; instead, we include their
13304 components. This change causes pandoc to behave as if it has
13305 these data files; they are constructed on demand when needed
13306 using `getDefaultReferenceDocx` and `getDefaultReferenceODT`.
13308 * Fixed regression in CSS parsing with `--self-contained` (#2224).
13309 Pandoc 1.14.0.x used css-text to parse the CSS, but its parser
13310 silently drops big sections of CSS. This commit replaces the
13311 use of css-text with a small but principled CSS preprocessor,
13312 which removes whitespace and comments and replaces `url()` with
13313 base 64 data when possible.
13315 * Use `https://` instead of `//` for MathJax and KaTeX CDN URLs (#1920).
13316 This will allow math to work when pages are being viewed locally.
13318 * `Text.Pandoc.Options`: Export `plainExtensions`.
13319 These are the extensions used in `plain` output.
13321 * LaTeX reader: Don't parse `_` and `^` as sub/superscript outside of
13322 math mode; treat them as regular inline text. Normally these will
13323 cause an error in LaTeX, but there are contexts (e.g. `alltt`
13324 environments) where they are allowed.
13326 * HTML reader: allow `<body>` to close `<head>`.
13328 * DocBook reader: support `mediaobject`s and `figures` (#2184, mb21).
13330 * RST reader: Fix reference names with special characters
13331 (Lars-Dominik Braun).
13333 * Textile writer: escape `+` and `-` as entities (#2225).
13335 * DokuWiki writer: Use proper `<code>` tags for code blocks (#2213).
13337 * Plain writer: don't use symbols for super/subscript (#2237).
13338 Simplified code by using `plainExtensions`.
13340 * InDesign writer: Properly escape URLs containing more than one
13341 colon character (gohai).
13343 * Docx writer: Make sure we use dist version of `reference.docx`
13344 (and not the user's version) for certain settings. Taking some
13345 settings values from a user-supplied reference.docx can lead to
13346 corruption. This fixes a regression from the last release (#2249).
13348 * `Text.Pandoc.Shared`: exports `getDefaultReferenceDocx` and
13349 `getDefaultReferenceODT` (API change). These functions have been
13350 removed from the Docx and ODT writers.
13352 * LaTeX template (Xavier Olive):
13353 + Added `CJKmainfont` and `CJKoptions` variables.
13354 + Allow dvipsnames (e.g. `MidnightBlue`) for colors (Xavier Olive).
13356 * Epub templates: use `author.role`, not `author.type`.
13358 * Bump cmark version to >= 0.3.4.
13360 * Improved Windows installer (#2205, thanks to nkalvi).
13361 Users can now select a per-user or systemwide install, and can set
13362 the installation path. At the end of installation, the install location
13363 is given. The install location is also now given in the list of
13364 installed programs in Control Panel. Cleaner WiX syntax is used for
13367 * Added `download_stats` target to Makefile.
13369## pandoc 1.14.0.4 (2015-06-02)
13371 * Added missing commonmark template.
13373 * Improved try pandoc (moved button, show raw command).
13375## pandoc 1.14.0.3 (2015-06-01)
13377 * Allow compilation with syb 0.5.*.
13379 * Custom writer: fixed some compiler warnings for ghc < 7.10.
13381## pandoc 1.14.0.2 (2015-05-31)
13383 * Allow building with hslua 0.4.
13385## pandoc 1.14.0.1 (2015-05-28)
13387 * Fixed problem with building of `reference.docx` and `reference.odt`
13388 when the `embed_data_files` flag is used. Instead of having a phase
13389 of the build where `reference.docx` and `reference.odt` are created
13390 from their constituent data files, we now construct these archives
13391 from their constituents when a `docx` or `odt` is built. The
13392 constituent files have been moved from `extra-source-files` to
13393 `data-files`, and `reference.docx` and `reference.odt` have been
13394 removed. Users can create their own `reference.docx` or
13395 `reference.odt` by using pandoc to create a simple `docx` or `odt`.
13396 `make-reference-files.hs` has been removed, simplifying the build
13399 * Don't include generated man pages in extra-source-files (#2189).
13401 * Bumped upper bound for aeson.
13403 * ConTeXt writer: create internal link anchors for Div elements with
13404 identifiers. (This is needed for linked citations to work.)
13406## pandoc 1.14 (2015-05-27)
13410 * Added `commonmark` as input and output format.
13412 * Added `--verbose` flag for debugging output in PDF production (#1840,
13415 * Allow wildcards in `--epub-embed-font` arguments (#1939).
13417 * Added `--latex-engine-opt` option (#969, #1779, Sumit Sahrawat).
13419 * Added `shortcut_reference_links` extension (Konstantin Zudov, #1977).
13420 This is enabled by default for those markdown flavors that
13421 support reading shortcut reference links, namely: `markdown`,
13422 `markdown_strict`, `markdown_github`, `markdown_php`.
13423 If the extension is enabled, the reader parses shortcut reference
13424 links like `[foo]`, and the writer creates such links unless doing
13425 so would cause problems. Users of markdown flavors that support
13426 shortcut reference links should not notice a difference in reading
13427 markdown, but the markdown pandoc produces may differ.
13428 If shortcut links are not desired, the extension can be disabled
13431### Behavior changes
13433 * `--toc` is now supported for `docx` output (#458, Nikolay Yakimov).
13434 A "dirty" TOC is created at the beginning of document.
13435 It can be regenerated after the document has been opened.
13437 * An implicit `--filter pandoc-citeproc` is now triggered only when the
13438 `--bibliography` option is used, and not when the `bibliography`
13439 field in metadata is specified (#1849).
13443 + Reference links with `implicit_header_references` are no longer
13444 case-sensitive (#1606).
13445 + Definition lists no longer require indentation for first line (#2087).
13446 Previously the body of the definition (after the `:` or `~` marker)
13447 needed to be in column 4. This commit relaxes that requirement,
13448 to better match the behavior of PHP Markdown Extra. So, now
13449 this is a valid definition list:
13453 + Resolve a potentially ambiguity with table captions:
13463 Is "bar" a definition, or the caption for the table? We'll count
13464 it as a caption for the table.
13465 + Disallow headerless pipe tables (#1996), to conform to GFM and PHP
13466 Markdown Extra. Note: If you have been using headerless pipe tables,
13467 this change may cause existing tables to break.
13468 + Allow pipe tables with header but no body (#2017).
13469 + Allow a digit as first character of a citation key (Matthias Troffaes).
13470 See https://github.com/jgm/pandoc-citeproc/issues/97
13474 + Don't limit includes to `.tex` extension (#1882).
13475 If the extension is not `.tex`, it must be given explicitly in
13476 the `\input` or `\include`.
13480 + Allow numbering in the style file. This allows inherited styles
13481 with numbering (lists) (Jesse Rosenthal).
13485 + Support smart punctuation (Craig Bosma).
13486 + Drop trees with a :noexport: tag (Albert Krewinkel). Trees having a
13487 `:noexport:` tag set are not exported. This mirrors org-mode.
13488 + Put header tags into empty spans (Albert Krewinkel, #2160).
13489 Org mode allows headers to be tagged: `* Headline :TAG1:TAG2`.
13490 Instead of being interpreted as part of the headline, the tags are now
13491 put into the attributes of empty spans. Spans without textual content
13492 won't be visible by default, but they are detectable by filters. They
13493 can also be styled using CSS when written as HTML.
13494 + Generalize code block result parsing (Albert Krewinkel).
13495 Previously, only code blocks were recognized as result blocks;
13496 now, any kind of block can be the result.
13498 * Append newline to the LineBreak in Dokuwiki, HTML, EPUB,
13499 LaTeX, MediaWiki, OpenDocument, Texinfo writers (#1924, Tim Lin).
13503 + Add "inline" or "display" class to math spans (#1914).
13504 This allows inline and display math to be styled differently.
13505 + Include raw latex blocks if `--mathjax` specified (#1938).
13506 + Require highlighting-kate >= 0.5.14 (#1903).
13507 This ensures that all code blocks will be wrapped in a `div`
13508 with class `sourceCode`. Also, the default highlighting CSS
13509 now adds `div.sourceCode { x-overflow: auto; }`, which means
13510 that code blocks (even with line numbers) will acquire a scroll
13511 bar on screens too small to display them (e.g. mobile phones).
13512 See also jgm/highlighting-kate#65.
13516 + Use a declaration for tight lists (Jose Luis Duran, Joseph
13517 Harriott). Previously, pandoc hard-coded some commands to make
13518 tight lists in LaTeX. Now we use a custom command instead,
13519 allowing the styling to be changed in a macro in the header.
13520 (Note: existing templates may need to be modified to include
13521 the definition of this macro. See the current template.)
13522 + Beamer output: if the header introducing a slide has the
13523 class `fragile`, add the `[fragile]` option to the slide (#2119).
13525 * MediaWiki writer:
13527 + Use `File:` instead of the deprecated `Image:` for images and
13528 other media files (Greg Rundlett).
13532 + Render a `Div (id,_,_) [Para _]` element as a `para` element
13533 with an `id` attribute. This makes links to citations work in
13534 DocBook with pandoc-citeproc.
13538 + Normalize headings to sequential levels (Nikolay Yakimov).
13539 This is pretty much required by docutils.
13540 + Treat headings in block quotes, etc as rubrics (Nikolay Yakimov).
13541 + Better handling of raw latex inline (#1961). We use
13542 `` :raw-latex:`...` `` and add a definition for this role to
13547 + Remove `linear=no` from cover `itemref` (#1609).
13548 + Don't use `sup` element for epub footnotes (#1995).
13549 Instead, just use an a element with class `footnoteRef`.
13550 This allows more styling options, and provides better results
13551 in some readers (e.g. iBooks, where anything inside the a
13552 tag breaks popup footnotes).
13553 + Take TOC title from `toc-title` metadata field.
13557 + Implemented `FirstParagraph` style (Jesse Rosenthal).
13558 Following the ODT writer, we add the `FirstParagraph` style to the
13559 first text paragraph following an image, blockquote, table, heading,
13560 or beginning of document. This allows it to be styled differently.
13561 The default is for it to be the same as `Normal`.
13562 + Added `BodyText` style (Jesse Rosenthal).
13563 We apply a `BodyText` style to all unstyled paragraphs. This is,
13564 essentially, the same as `Normal`, except that since not everything
13565 inherits from `BodyText` (the metadata won't, for example, or
13566 the headers or footnote numbers), we can change the text in the body
13567 without having to make exceptions for everything. If we do want to
13568 change *everything*, we can still do it through `Normal`.
13569 + Altered `Blockquote` style slightly (Jesse Rosenthal).
13570 Since `BlockQuote` derives from `BodyText`, we just want to specify
13571 by default that it won't indent, regardless of what `BodyText` does.
13572 Note that this will not produce any visible difference in the default
13574 + Take TOC title from `toc-title` metadata field (Nikolay Yakimov).
13575 + Added a style to figure images (Nikolay Yakimov).
13576 Figures with empty captions use style `Figure`.
13577 Figures with nonempty captions use style `Figure with Caption`, which
13578 is based on `Figure`, and additionally has `keepNext` set.
13582 + Added figure captions (Nikolay Yakimov). The following styles are
13584 `Figure` -- for figure with empty caption),
13585 `FigureWithCaption` (based on `Figure`) -- for figure with caption,
13586 `FigureCaption` (based on `Caption`) -- for figure captions.
13587 Also, `TableCaption` (based on `Caption`) is used for table captions.
13591 * New `Text.Pandoc.Error` module with `PandocError` type
13592 (Matthew Pickering).
13594 * All readers now return `Either PandocError Pandoc` instead of `Pandoc`
13595 (Matthew Pickering). This allows better handling of errors.
13597 * Added `Text.Pandoc.Writers.CommonMark`, exporting `writeCommonMark`.
13599 * Added `Text.Pandoc.Readers.CommonMark`, exporting `readCommonMark`.
13601 * Derive `Data` and `Typeable` instances for `MediaBag`, `Extension`,
13602 `ReaderOptions`, `EPUBVersion`, `CiteMethod`, `ObfuscationMethod`,
13603 `HTMLSlideVariant`, `TrackChanges`, `WriterOptions` (Shabbaz
13606 * New `Ext_shortcut_reference_links` constructor for `Extension`
13607 (Konstantin Zudov).
13613 + Allow smart `'` after inline math (#1909, Nikolay Yakimov).
13614 + Check for tex macros after indented code (#1973).
13615 + Rewrote `charsInBalancedBrackets` for efficiency.
13616 + Make sure a closing `</div>` doesn't get included in a
13617 definition list item (#2127).
13618 + Don't parse bracketed text as citation if it might be a link,
13619 image, or footnote (Nikolay Yakimov).
13620 + Require space after key in mmd title block (#2026, Nikolay
13621 Yakimov). Require space after key-value delimiter colon in mmd title
13623 + Require nonempty value in mmd title block (Nikolay Yakimov).
13624 + Disable all metadata block extensions when parsing
13625 metadata field values (#2026, Nikolay Yakimov). Otherwise we
13626 could get a mmd title block inside YAML metadata, for example.
13630 + Improve self-closing tag detection in `htmlInBalanced` (#2146).
13631 + Handle tables with `<th>` in body rows (#1859, mb21).
13632 + Fixed `htmlTag` (#1820). If the tag parses as a comment, we check
13633 to see if the input starts with `<!--`. If not, it's bogus comment
13634 mode and we fail `htmlTag`.
13635 + Handle `base` tag; if it has an `href` value, this is added to
13636 all relative URLs in links and images.
13640 + Look inside "info" elements for section titles (#1931).
13644 + Parse images in deprecated vml format (Jesse Rosenthal).
13645 + Allow sub/superscript verbatims (Jesse Rosenthal).
13646 Verbatim usually shuts off all other run styles, but we don't want it
13647 to shut off sub/superscript.
13651 + Handle `tabular*` environment (#1850).
13652 Note that the table width is not actually parsed or taken into
13653 account, but pandoc no longer chokes on it.
13654 + Ignore options in `\lstinline` rather than raising error (#1997).
13655 + Add some test cases for simple tables (Mathias Schenner).
13656 + Handle valign argument in tables (Mathias Schenner) (currently
13657 we just ignore this).
13658 + Allow non-empty colsep in tables (Mathias Schenner).
13659 The `tabular` environment allows non-empty column separators
13660 with the "@{...}" syntax. Previously, pandoc would fail to
13661 parse tables if a non-empty colsep was present. With this
13662 commit, these separators are still ignored, but the table gets
13663 parsed. A test case is included.
13664 + Recognize `\newpage` as a block command.
13665 + Allow block content in `\title{}` (#2001).
13666 + Check for block-level newcommand aliases in blockCommand (Nikolay
13668 + Guard against paragraph starting with inline macro (Nikolay Yakimov).
13669 + Properly gobble spaces after `\\` (#2007).
13673 + Handle newlines in table cells, and empty cells (#1919).
13677 + Allow image links with non-image targets (Hans-Peter Deifel).
13678 This matches behavior of Org-Mode for links like
13679 `[[http://example.com][https://www.haskell.org/static/img/logo.png]]`.
13683 + Don't print empty id attributes (thanks to Steve Horne).
13687 + Fixed list-style-type for numbered example lists.
13688 Should be "decimal," not "example" (#1902).
13689 + Do not omit missing `alt` attribute on `img` tag (#1131,
13691 + Allow multiple colgroups in table (#2122).
13692 + In revealjs, ensure that lists in speaker notes don't add "fragment"
13693 classes, which can cause additional keypresses to be needed to
13694 advance a slide (#1394).
13698 + Don't escape `$` in URL (#1913).
13699 + Don't use listings in headers (Matthew Pickering, #1963).
13700 + Recognize book documentclass if set in metadata (#1971).
13701 This sets `--chapters` implicitly if the documentclass in metadata
13702 is a book documentclass. Previously this was done only if a book
13703 documentclass was set in a variable.
13704 + Add a `\label` in `\hyperdef` for Div, Span (or links don't work).
13705 + Make `mainlang` work when `lang` is in metadata (#2174).
13709 + Fix wrapping by using breakable spaces (Tim Lin).
13713 + Fixed toc depth in RST writer. Previously the depth was being
13714 rendered as a floating point number with a decimal point.
13718 + Improved escaping (#2086). `<` should not be escaped as `\<`, for
13719 compatibility with original Markdown. We now escape `<` and `>`
13720 with entities. Also, we now backslash-escape square brackets.
13721 + Avoid introducing spurious list items through wrapping (#1946).
13722 + Don't emit span tags if plain or raw HTML disabled.
13724 * MediaWiki writer:
13726 + Convert spaces to underscores in wikilink URL (#1982), like MediaWiki.
13730 + Insert some needed blank lines (#1860).
13731 + Avoid wrapping after list marker (#1858).
13735 + Properly handle internal links to IDs in spans, divs (#1884).
13736 + Use plain writer for metadata dc: fields (#2121).
13737 This gives better results when we have, e.g. multiple paragraphs.
13738 Note that tags aren't allowed in these fields.
13739 + Properly handle image links without an extension (#1855).
13740 + Improved chapter splitting and internal link rewriting (#1887,
13741 #2162, #2163). This will ensure that internal links work and
13742 that the references section produced by pandoc-citeproc is
13743 in its own chapter.
13744 + Fixed handling of svg images (#2183).
13748 + Better handling of raw blocks and inlines (#1951).
13749 Previously these were always escaped and printed verbatim.
13750 Now they are ignored unless the format is `icml`, in which
13751 case they are passed through unescaped.
13752 + Fixed image URIs in ICML output (gohai).
13756 + Raise error if loadstring returns an error status.
13757 + Raise `PandocLuaException` instead of using 'error'.
13758 Eventually we'll change the return type so that no exception
13759 is involved, but at least this can be trapped.
13760 + Use UTF-8 aware bytestring conversion.
13761 + Set foreign encoding to UTF-8 (Nikolay Yakimov, #2101, #1634).
13762 Also factored out ByteString, since it's only used as an intermediate
13767 + Copy hyphenation settings from reference.docx (Nikolay Yakimov).
13768 + Filter out illegal XML characters (#1992, Matthew Pickering).
13769 + Added `noProof` to docx syntax highlighting `SourceCode` style.
13770 + Added footnotes id -1 and 0 (Jesse Rosenthal).
13771 Word uses, by default, footnotes with id -1 and 0 for separators. If a
13772 user modifies `reference.docx`, they will end up with a `settings.xml`
13773 file that references these footnotes, but no such footnotes in the
13774 document. This will produce a corruption error. Here we add these to the
13775 document and `settings.xml` file, so future modifications won't break
13777 + Handle lists correctly inside table cells (Jesse Rosenthal).
13778 Previously we didn't transform lists inside table cells.
13779 + Set firstRow information in tables (Nikolay Yakimov).
13780 + Don't replace `SourceCode` style in `reference.docx` if it is defined
13781 there (Nikolay Yakimov, #1872). If `--no-highlight` specified, remove
13782 any `SourceCode` and `*Tok` styles in `reference.docx`.
13783 + Attempt to match international style names (#1607, Nikolay Yakimov).
13784 + Set these styles as custom (Nikolay Yakimov): `Author`, `Abstract`,
13785 `Compact`, `Image Caption`, `Table Caption`, `Definition Term`,
13786 `Definition`, `First Paragraph`.
13787 + Rename these styles to correspond with Word `Normal.dotm` (Nikolay
13788 Yakimov): `Block Quote -> Block Text`, `Link -> Hyperlink`,
13789 `Footnote Ref -> Footnote Reference`.
13790 + Added `Caption` style (Nikolay Yakimov).
13791 + Changed these styles' inheritance (Nikolay Yakimov):
13792 `Image Caption <- Caption`, `Table Caption <- Caption`.
13793 + Remove `SourceCode` style from `reference.docx` (#1872).
13794 This is added automatically by the docx writer.
13795 + Added toc heading style to `reference.docx` (Nikolay Yakimov).
13797 * `Text.Pandoc.PDF`
13799 + Don't suggest "Try xelatex" if xelatex already in use (mb21, #1832).
13800 + More comprehensible errors on image conversion (#2067).
13801 EPS can't be supported without shelling out to something like
13802 ImageMagick, but at least we can avoid mysterious error messages.
13804 * `Text.Pandoc.Shared`:
13806 + Make safeRead safe (#1801, Matthew Pickering).
13807 + Added `mapLeft`, `hush` (Matthew Pickering).
13809 * `Text.Pandoc.Pretty`:
13811 + Remove partial function (Matthew Pickering).
13813 * `Text.Pandoc.SelfContained`:
13815 + Add `;charset=utf-8` to script mime type if missing (#1842).
13816 + Improved building of data URIs (#1940). Now base64 is used except
13817 for `text/*` mime types.
13818 + `cssURLs` no longer tries to fetch fragment URLs (#2121).
13819 + Properly handle data URIs in css urls (#2129).
13820 Use a proper CSS parser (adds dependency on `text-css`).
13822 * `Text.Pandoc.UTF8`:
13824 + Better handling of bare CRs in input files (#2132).
13825 Previously we just stripped them out; now we convert
13826 other line ending styles to LF line endings.
13828 * `Text.Pandoc.ImageSize`:
13830 + Fixed some exif header parsing bugs (#1834).
13831 + Make imageSize return an Either, not a Maybe (#1834).
13832 Use `runGetOrFail` (with `binary >= 0.7`) to return `Left` on
13833 parse failure (rather than `error`).
13834 + Improved warnings when image size can't be determined.
13835 + Removed error landmines (Matthew Pickering).
13837 * Added woff2 to MIME types (Alfred Wechselberger).
13839 * pandoc: When a binary input format is used, warn that file
13840 arguments past the first one are being ignored (Matthew Pickering).
13842### Template changes
13846 + Degrade gracefully if `\paragraph` not defined.
13847 + Include `grffile` together with `graphicx` (#2074).
13848 This properly handles filenames containing spaces and dots.
13849 + Redefine `\paragraph`, `\subparagraph`... to behave more
13850 like section headers (#1658).
13851 + Import hyperref before polyglossia to avoid an error with xelatex,
13852 "please load package hyperref before bidi package" (Nick Bart).
13853 + Added `toccolor` variable to control link color in toc (Kaixhin).
13855 * LaTeX, Beamer templates:
13857 + Provide `\tightlist`, which is now used by the LaTeX writer.
13858 + Use polyglossia in beamer (#85).
13859 + Use `bibliography` instead of `biblio-files`
13860 (#1661). Also use `\addbibresource` instead of `\bibliography` for
13862 + Added `setotherlanguages` in polyglossia. This uses an `otherlang`
13863 variable that is derived from a comma-separated list in `lang`;
13864 the last language is `mainlang` and the others are `otherlang`.
13868 + Use `div`, not `p`, for "rights" on title page.
13869 + Added header-includes, include-before, include-after (#1987).
13871 * OpenDocument template:
13873 + Use `text:p` instead of `text:h` for title.
13874 Using `text:h` causes problems with numbering. Closes #2059.
13875 Thanks to @nkalvi for diagnosing this.
13877 * reveal.js template:
13879 + Link to non-minified css, js. The minified versions no longer
13880 ship with the library.
13881 + Correctly include style CSS (#1949).
13882 + New configurable options options: `center`, `maxScale`, `slideNuber`
13883 (Dmitry Smirnov, pandoc-templates#89).
13884 + Moved custom CSS after theme. This allows custom CSS to modify
13885 themes, instead of being replaced by them.
13886 + Allow `center` to be set to false.
13888### Under the hood improvements
13890 * Removed pre-built `reference.docx` and `reference.odt` (Nikolay
13891 Yakimov). Instead the repository now includes the component text files,
13892 and the zipped binaries are built from these using a helper
13893 program, `make-reference-files`. This should make maintenance of
13894 these components easier going forward.
13896 * `Text.Pandoc.Parsing`:
13898 + Added new `<+?>` combinator (Nikolay Yakimov).
13899 + Added `stateHeaderKeys` to `ParserState`.
13901 * `make_deb.sh` fixes:
13903 + Detect architecture.
13904 + Add Installed-Size to debian package control file (#1900).
13905 + Use `fakeroot` to get permissions right.
13906 + Use `mkdir` and `cp` instead of `install`.
13907 + Set permissions of directories to 755.
13908 + Install in `/usr` rather than `/usr/local`.
13909 + Compress man pages.
13910 + Combine copyright files for `pandoc`, `pandoc-citeproc`.
13912 * Added `Text.Pandoc.Compat.Locale` and `old-locale` flag
13913 to assist with transition to `time` 1.5.
13915 * Updated CONTRIBUTING.md with information about issue tags (Matthew
13918 * Updated travis installs to the new sudo-less syntax (Tim Lin).
13920 * Updated dependency version bounds.
13922 * EPUB tests: don't use `joinPath`, which varies across platforms.
13923 Instead, use a forward-slash to join paths, regardless of the
13924 platform. This matches the way `MediaBag` now works.
13926 * Clarify JSON input and output in usage message (Caleb McDaniel).
13928 * Improved INSTALL instructions.
13930 * Always build man pages. Removed make-pandoc-man-pages flag.
13932 * Makefile: removed man target, now that we generate man pages by default.
13936 + Fixed typos (J. Lewis Muir).
13937 + Added documentation on backtick_code_blocks (#2135, Nikolay Yakimov).
13938 + Added note on in-field markup in biblio databases (Nick Bart).
13939 + Fixed misleading example of raw HTML block.
13940 + Various minor formatting and consistency fixes for the program
13941 options (Andreas Lööw).
13942 + Made definition lists for options all "loose" for consistency.
13943 + Added YAML biblio format to table, and note on `pandoc-citeproc`'s
13944 `--bib2json` and `--bib2yaml` options (Nick Bart).
13945 + Removed obsolete reference to `mods2yaml` (Nick Bart).
13946 + Added section on syntax highlighting.
13947 + Documented `toccolor` variable.
13949## pandoc 1.13.2.1 (2015-04-15)
13951 * Updated to build with ghc 7.10.1.
13953 * Bumped package upper bounds for filepath, blaze-html, blaze-markup.
13955## pandoc 1.13.2 (2014-12-20)
13957 * TWiki Reader: add new new twiki reader (API chaneg, Alexander Sulfrian).
13961 + Better handling of paragraph in div (#1591).
13962 Previously text that ended a div would be parsed as Plain
13963 unless there was a blank line before the closing div tag.
13964 + Don't treat a citation as a reference link label (#1763).
13965 + Fix autolinks with following punctuation (#1811).
13966 The price of this is that autolinked bare URIs can no longer
13967 contain `>` characters, but this is not a big issue.
13968 + Fix `Ext_lists_without_preceding_blankline` bug (#1636, Artyom).
13969 + Allow `startnum` to work without `fancy_lists`. Formerly
13970 `pandoc -f markdown-fancy_lists+startnum` did not work properly.
13972 * RST reader (all Daniel Bergey):
13974 + Parse quoted literal blocks (#65). RST quoted literal blocks are
13975 the same as indented literal blocks (which pandoc already supports)
13976 except that the quote character is preserved in each line.
13977 + Parse RST class directives. The class directive accepts one or more
13978 class names, and creates a Div value with those classes. If the
13979 directive has an indented body, the body is parsed as the children of
13980 the Div. If not, the first block following the directive is made a
13981 child of the Div. This differs from the behavior of rst2xml, which
13982 does not create a Div element. Instead, the specified classes are
13983 applied to each child of the directive. However, most Pandoc Block
13984 constructors to not take an Attr argument, so we can't duplicate this
13986 + Warn about skipped directives.
13987 + Literal role now produces Code. Code role should have "code" class.
13988 + Improved support for custom roles
13990 - Added `sourceCode` to classes for `:code:` role, and anything
13991 inheriting from it.
13992 - Add the name of the custom role to classes if the Inline
13993 constructor supports Attr.
13994 - If the custom role directive does not specify a parent role,
13995 inherit from the `:span:` role.
13997 This differs somewhat from the `rst2xml.py` behavior. If a custom
13998 role inherits from another custom role, Pandoc will attach both
13999 roles' names as classes. `rst2xml.py` will only use the class of
14000 the directly invoked role (though in the case of inheriting from a
14001 `:code:` role with a `:language:` defined, it will also provide the
14002 inherited language as a class).
14003 + Warn about ignored fields in role directives.
14007 + Parse label after caption into a span instead of
14008 inserting an additional paragraph of bracketed text (#1747).
14009 + Parse math environments as inline when possible (#1821).
14010 + Better handling of `\noindent` and `\greektext` (#1783).
14011 + Handle `\texorpdfstring` more gracefully.
14012 + Handle `\cref` and `\sep` (Wikiwide).
14013 + Support `\smartcite` and `\Smartcite` from biblatex.
14017 + Retain display type of MathML output (#1719, Matthew Pickering).
14018 + Recognise `<br>` tags inside `<pre>` blocks (#1620, Matthew Pickering).
14019 + Make `embed` tag either block or inline (#1756).
14023 + Handle `keycombo`, `keycap` (#1815).
14024 + Get string content in inner tags for literal elements (#1816).
14025 + Handle `menuchoice` elements better, with a `>` between (#1817).
14026 + Include `id` on section headers (#1818).
14027 + Document/test "type" as implemented (Brian O'Sullivan).
14028 + Add support for calloutlist and callout (Brian O'Sullivan).
14029 We treat a calloutlist as a bulleted list. This works well in practice.
14030 + Add support for `classname` (Bryan O'Sullivan).
14034 + Fix window path for image lookup (Jesse Rosenthal).
14035 Don't use os-sensitive "combine", since we always want the paths in our
14036 zip-archive to use forward-slashes.
14037 + Single-item headers in ordered lists are headers (Jesse Rosenthal).
14038 When users number their headers, Word understands that as a single item
14039 enumerated list. We make the assumption that such a list is, in fact,
14041 + Rewrite rewriteLink to work with new headers (Jesse Rosenthal).
14042 There could be new top-level headers after making lists, so we have to
14043 rewrite links after that.
14044 + Use polyglot header list (Jesse Rosenthal).
14045 We're just keeping a list of header formats that different languages
14046 use as their default styles. At the moment, we have English, German,
14047 Danish, and French. We can continue to add to this.
14048 This is simpler than parsing the styles file, and perhaps less
14049 error-prone, since there seems to be some variations, even within a
14050 language, of how a style file will define headers.
14051 + Remove header class properly in other langs (Jesse Rosenthal).
14052 When we encounter one of the polyglot header styles, we want to remove
14053 that from the par styles after we convert to a header. To do that, we
14054 have to keep track of the style name, and remove it appropriately.
14055 + Account for external link URLs with anchors. Previously, if a URL
14056 had an anchor, the reader would incorrectly identify it as an
14057 internal link and return only the anchor as URL. (Caleb McDaniel)
14058 + Fix for Issue #1692 (i18n styles) (Nikolay Yakimov).
14062 + Added state changing blanklines (Jesse Rosenthal).
14063 This allows us to emphasize at the beginning of a new paragraph (or, in
14064 general, after blank lines).
14065 + Fixed bug with bulleted lists:
14071 was being parsed as a list, even though an unindented `*`
14072 should make a heading. See
14073 <http://orgmode.org/manual/Plain-lists.html#fn-1>.
14074 + Org reader: absolute, relative paths in link (#1741, Albert
14075 Krewinkel). The org reader was too restrictive when parsing links;
14076 some relative links and links to files given as absolute paths
14077 were not recognized correctly.
14078 + Org reader: allow empty links (jgm/gitit#471, Albert Krewinkel).
14079 This is important for use in gitit, which uses empty links
14081 + Respect indent when parsing Org bullet lists (#1650, Timothy
14082 Humphries). Fixes issue with top-level bullet list parsing.
14083 + Fix indent issue for definition lists (Timothy Humphries,
14084 see #1650, #1698, #1680).
14085 + Parse multi-inline terms correctly in definition list (#1649,
14086 Matthew Pickering).
14087 + Fix rules for emphasis recognition (Albert Krewinkel).
14088 Things like `/hello,/` or `/hi'/` were falsy recognized as emphasised
14089 strings. This is wrong, as `,` and `'` are forbidden border chars and
14090 may not occur on the inner border of emphasized text.
14091 + Drop COMMENT document trees (Albert Krewinkel).
14092 Document trees under a header starting with the word `COMMENT` are
14093 comment trees and should not be exported. Those trees are dropped
14095 + Properly handle links to `file:target` (Albert Krewinkel).
14096 Org links like `[[file:target][title]]` were not handled correctly,
14097 parsing the link target verbatim. The org reader is changed such that
14098 the leading `file:` is dropped from the link target (see #756, #1812).
14099 + Parse LaTeX-style MathML entities (#1657, Albert Krewinkel).
14100 Org supports special symbols which can be included using LaTeX syntax,
14101 but are actually MathML entities. Examples for this are
14102 `\nbsp` (non-breaking space), `\Aacute` (the letter A with accent acute)
14103 or `\copy` (the copyright sign ©)
14107 + URI handling improvements. Now we outsource most of the work to
14108 `fetchItem'`. Also, do not include queries in file extensions (#1671).
14112 + Use `\texorpdfstring` for section captions when needed (Vaclav Zeman).
14113 + Handle consecutive linebreaks (#1733).
14114 + Protect graphics in headers (Jesse Rosenthal).
14115 Graphics in `\section`/`\subsection` etc titles need to be `\protect`ed.
14116 + Put `~` before header in list item text (Jesse Rosenthal).
14117 Because of the built-in line skip, LaTeX can't handle a section header
14118 as the first element in a list item.
14119 + Avoid using reserved characters as `\lstinline` delimiters (#1595).
14120 + Better handling of display math in simple tables (#1754).
14121 We convert display math to inline math in simple tables,
14122 since LaTeX can't deal with display math in simple tables.
14123 + Escape spaces in code (#1694, Bjorn Buckwalter).
14125 * MediaWiki writer:
14127 + Fixed links with URL = text. Previously these were rendered as bare
14128 words, even if the URL was not an absolute URL (#1825).
14132 + Don't force all citations into footnotes.
14136 + Add blankline at end of output (#1732, Matthew Pickering).
14140 + Ensure blank line after figure.
14141 + Avoid excess whitespace after last list item (#1777).
14142 + Wrap line blocks with spaces before continuations (#1656).
14143 + Fixed double-rendering of footnotes in RST tables (#1769).
14147 + Better handling of block quotes. This change ensures that
14148 multiple paragraph blockquotes are rendered using native `>`
14149 rather than as HTML (#1738).
14150 + Fix external images (#1739). Preface relative links with ":",
14151 absolute URIs without. (Timothy Humphries)
14155 + Use protocol-relative URL for mathjax.
14156 + Put newline btw img and caption paragraph.
14157 + MathML now outputted with tex annotation (#1635, Matthew Pickering).
14158 + Add support for KaTeX HTML math (#1626, Matthew Pickering).
14159 This adds `KaTeX` to `HTMLMathMethod` (API change).
14160 + Don't double render when `email-obfuscation=none` (#1625, Matthew
14162 + Make header attributes work outside top level (#1711).
14163 Previously they only appeared on top level header elements.
14164 Now they work e.g. in blockquotes.
14168 + Correctly handle images without extensions (#1729).
14169 + Strip querystring in ODT write (#1682, Todd Sifleet).
14173 + Add newline to output.
14177 + Don't add `sourceURL` to absolute URIs (#1669).
14178 + Don't use unsupported `opf:title-type` for epub2.
14179 + Include "landmarks" section in nav document for epub3 (#1757).
14180 + Removed playOrder from navpoint elements in ncx file (#1760).
14181 These aren't required, and they make manual modification of epubs
14183 + Extract title even from structured title.
14184 + Don't include nav node in spine unless `--toc` was requested.
14185 Previously we included it in the spine with `linear="no"`, leading
14186 to odd results in some readers (#1593).
14187 + Fixed absolute URI detection (#1672).
14188 + Correctly resolve relative URIs (#1671).
14189 + Use regular page template for `nav.xhtml`, including doctype (#1759).
14193 + Put docx table captions above tables (#1641, Nikolay Yakimov).
14194 + Get the page width from the reference docx file, and use
14195 it to scale images that are too large to fit (Grégory Bataille).
14196 + Partial fix for #1607 (Nikolay Yakimov). International heading styles
14197 are inferred based on `<w:name val="heading #">` fallback, if there
14198 are no en-US "Heading#" styles
14199 + Look in user data dir for archive `reference.docx`.
14200 + Renumber header and footer relationships to avoid collisions (Jesse
14201 Rosenthal). We previously took the old relationship names of the
14202 headers and footer in secptr. That led to collisions. We now make
14203 a map of available names in the relationships file, and then rename
14208 + Add function toLabel (Mark Szepieniec).
14209 This function can be used to sanitize reference labels so that
14210 they do not contain any of the illegal characters \#[]",{}%()|= .
14211 Currently only Links have their labels sanitized, because they
14212 are the only Elements that use passed labels.
14214 * `Text.Pandoc.Shared`:
14216 + Moved import of `toChunks` outside of CPP conditional (#1590).
14217 + Fix `inDirectory` to reset to the original directory in case
14218 an exception occurs (Freiric Barral).
14222 + LaTeX template: load polyglossia before bibtex (jgm/pandoc-templates#70).
14223 Thanks to bluebirch.
14224 + LaTeX template: Added `\VerbatimFootnotes` if there is verbatim in notes
14226 + LaTeX template: Add shorthands=off to babel options (#1648).
14227 + EPUB, EPUB3 templates: Added `id="cover"` to body of cover page.
14228 This aids styling, making it possible for example to set 0 margins
14229 on the title page (#1758).
14230 + EPUB, EPUB3 templates: Handle structured metadata on titlepage.
14231 Previously we just expected 'title', 'subtitle', 'author', 'date'.
14232 Now we still support those, but also support the format recommended
14233 for epub metadata in the pandoc README:
14240 text: An investigation of metadata
14248 text: doi:10.234234.234/33
14249 publisher: My Press
14250 rights: (c) 2007 John Smith, CC BY-NC
14253 * `Text.Pandoc.Templates.getDefaultTemplate`:
14254 don't fail when called with "fb2" (#1660).
14256 * `Text.Pandoc.Parsing`:
14258 + Fixed `inlineMath` so it handles `\text{..}` containing `$`.
14259 For example: `$x = \text{the $n$th root of $y$}` (#1677).
14260 + Change `parseFromString` to fail if not all input is consumed.
14261 (Matthew Pickering)
14262 + Moved `addWarning` from Markdown reader to `Parsing`, so it can be
14263 used by more readers (API change, Daniel Bergey).
14265 * `Text.Pandoc.Pretty`:
14267 + Improve performance of `realLength` (Matthew Pickering).
14268 + Make CR + BLANKLINE = BLANKLINE. This fixes an extra blank line we
14269 were getting at the end of markdown fragments (as well as rst, org,
14272 * `Text.Pandoc.MIME`:
14274 + Add mime type for WebVTT (Jason Ronallo).
14275 + Changed mime type for `otf` to `application/vnd.ms-opentype` (#1761).
14276 This is needed for epub3 validation.
14278 * `Text.Pandoc.MediaBag`:
14280 + Fix Windows specific path problems (#1597).
14282 * `Text.Pandoc.Shared`:
14284 + Make `collapseFilePath` OS-agnostic (Matthew Pickering).
14286 * Link the test suite using `-threaded`.
14287 This allows the test suite to be run using `+RTS -N`.
14289 * Added `network` dependency under `network-uri` flag in test section.
14291 * Give better error messages when someone tries to convert from
14292 pdf, doc, odt (#1683).
14294 * Added `track` to list of tags treated by `--self-contained` (#1664).
14297## pandoc 1.13.1 (2014-08-30)
14299 * Fixed `--self-contained` with Windows paths (#1558).
14300 Previously `C:\foo.js` was being wrongly interpreted as a URI.
14302 * HTML reader: improved handling of tags that can be block or inline.
14303 Previously a section like this would be enclosed in a paragraph,
14304 with RawInline for the video tags (since video is a tag that can
14305 be either block or inline):
14307 <video controls="controls">
14308 <source src="../videos/test.mp4" type="video/mp4" />
14309 <source src="../videos/test.webm" type="video/webm" />
14311 The videos can not be played back on your system.<br/>
14312 Try viewing on Youtube (requires Internet connection):
14313 <a href="http://youtu.be/etE5urBps_w">Relative Velocity on
14318 This change will cause the video and source tags to be parsed
14319 as RawBlock instead, giving better output.
14320 The general change is this: when we're parsing a "plain" sequence
14321 of inlines, we don't parse anything that COULD be a block-level tag.
14325 + Be sensitive to user styles. Note that "Hyperlink" is
14326 "blacklisted," as we don't want the default underline styling to be
14327 inherited by all links by default (Jesse Rosenthal).
14328 + Read single paragraph in table cell as `Plain` (Jesse Rosenthal).
14329 This makes to docx reader's native output fit with the way the markdown
14330 reader understands its markdown output.
14334 + Header is now parsed only if standalone flag is set (Matthew Pickering).
14335 + The header is now parsed as meta information. The first line is the
14336 `title`, the second is the `author` and third line is the `date`
14337 (Matthew Pickering).
14338 + Corrected formatting of `%%mtime` macro (Matthew Pickering).
14339 + Fixed crash when reading from stdin.
14341 * Textile writer: Extended the range of cases where native textile
14342 tables will be used (as opposed to raw HTML): we now handle any
14343 alignment type, but only for simple tables with no captions.
14345 * EPUB writer: Don't use page-progression-direction in EPUB2, which
14346 doesn't support it. Also, if page-progression-direction not specified
14347 in metadata, don't include the attribute even in EPUB3; not including it
14348 is the same as including it with the value "default", as we did before.
14351 * Org writer: Accept example lines with indentation at the beginning
14356 + Refactor to use Reader monad (Matthew Pickering).
14357 + Avoid using raw HTML in table cells; instead, use `\\`
14358 instead of newlines (Jesse Rosenthal).
14359 + Properly handle HTML table cell alignments, and use spacing
14360 to make the tables look prettier (#1566).
14364 + Bibliography entries get `Bibliography` style (#1559).
14365 + Implement change tracking (Jesse Rosenthal).
14369 + Fixed a bug that caused a table caption to repeat across all pages
14371 + Improved vertical spacing in tables and made it customizable using
14372 standard lengths set by booktab. See
14373 <https://groups.google.com/forum/#!msg/pandoc-discuss/qMu6_5lYy0o/ZAU7lzAIKw0J>
14375 + Added `\strut` to fix spacing in multiline tables (Jose Luis Duran).
14376 + Use `\tabularnewline` instead of `\\` in table cells (Jose Luis Duran).
14377 + Made horizontal rules more flexible (Jose Luis Duran).
14379 * Text.Pandoc.MIME:
14381 + Added `MimeType` (type synonym for `String`) and `getMimeTypeDef`.
14382 Code cleanups (Artyom Kazak).
14386 + LaTeX template: disable microtype protrusion for typewriter font (#1549,
14389 * Improved OSX build procedure.
14391 * Added `network-uri` flag, to deal with split of `network-uri` from
14394 * Fix build dependencies for the `trypandoc` flag, so that they are
14395 ignored if `trypandoc` flag is set to False (Gabor Pali).
14397 * Updated README to remove outdated claim that `--self-contained`
14398 looks in the user data directory for missing files.
14400## pandoc 1.13.0.1 (2014-08-17)
14404 + Fixed regression which bungled list numbering (#1544), causing
14405 all lists to appear as basic ordered lists.
14406 + Include row width in table rows (Christoffer Ackelman, Viktor Kronvall).
14407 Added a property to all table rows where the sum of column widths
14408 is specified in pct (fraction of 5000). This helps persuade Word
14409 to lay out the table with the widths we specify.
14411 * Fixed a bug in Windows 8 which caused pandoc not to find the
14412 `pandoc-citeproc` filter (#1542).
14414 * Docx reader: miscellaneous under-the-hood improvements (Jesse Rosenthal).
14415 Most significantly, the reader now uses Builder, leading to some
14416 performance improvements.
14418 * HTML reader: Parse appropriately styled span as SmallCaps.
14420 * Markdown writer: don't escape `$`, `^`, `~` when `tex_math_dollars`,
14421 `superscript`, and `subscript` extensions, respectively, are
14422 deactivated (#1127).
14424 * Added `trypandoc` flag to build CGI executable used in the online
14427 * Makefile: Added 'quick', 'osxpkg' targets.
14429 * Updated README in templates to indicate templates license.
14430 The templates are dual-licensed, BSD3 and GPL2+.
14432## pandoc 1.13 (15 August 2014)
14436 * Added `docx` as an input format (Jesse Rosenthal). The docx
14437 reader includes conversion of native Word equations to pandoc
14438 LaTeX `Math` elements. Metadata is taken from paragraphs at the
14439 beginning of the document with styles `Author`, `Title`, `Subtitle`,
14440 `Date`, and `Abstract`.
14442 * Added `epub` as an input format (Matthew Pickering). The epub
14443 reader includes conversion of MathML to pandoc LaTeX `Math`
14446 * Added `t2t` (Txt2Tags) as an input format (Matthew Pickering).
14447 Txt2tags is a lightweight markup format described at
14448 <http://txt2tags.org/>.
14450 * Added `dokuwiki` as an output format (Clare Macrae).
14452 * Added `haddock` as an output format.
14454 * Added `--extract-media` option to extract media contained in a zip
14455 container (docx or epub) while adjusting image paths to point to the
14458 * Added a new markdown extension, `compact_definition_lists`, that
14459 restores the syntax for definition lists of pandoc 1.12.x, allowing
14460 tight definition lists with no blank space between items, and
14461 disallowing lazy wrapping. (See below under behavior changes.)
14463 * Added an extension `epub_html_exts` for parsing HTML in EPUBs.
14465 * Added extensions `native_spans` and `native_divs` to activate
14466 parsing of material in HTML span or div tags as Pandoc Span
14467 inlines or Div blocks.
14469 * `--trace` now works with the Markdown, HTML, Haddock, EPUB,
14470 Textile, and MediaWiki readers. This is an option intended
14471 for debugging parsing problems; ordinary users should not need
14474### Behavior changes
14476 * Changed behavior of the `markdown_attribute` extension, to bring
14477 it in line with PHP markdown extra and multimarkdown. Setting
14478 `markdown="1"` on an outer tag affects all contained tags,
14479 recursively, until it is reversed with `markdown="0"` (#1378).
14481 * Revised markdown definition list syntax (#1429). Both the reader
14482 and writer are affected. This change brings pandoc's definition list
14483 syntax into alignment with that used in PHP markdown extra and
14484 multimarkdown (with the exception that pandoc is more flexible about
14485 the definition markers, allowing tildes as well as colons). Lazily
14486 wrapped definitions are now allowed. Blank space is required
14487 between list items. The space before a definition is used to determine
14488 whether it is a paragraph or a "plain" element. **WARNING: This change
14489 may break existing documents!** Either check your documents for
14490 definition lists without blank space between items, or use
14491 `markdown+compact_definition_lists` for the old behavior.
14493 * `.numberLines` now works in fenced code blocks even if no language
14494 is given (#1287, jgm/highlighting-kate#40).
14496 * Improvements to `--filter`:
14498 + Don't search PATH for a filter with an explicit path.
14499 This fixed a bug wherein `--filter ./caps.py` would run `caps.py` from
14500 the system path, even if there was a `caps.py` in the working directory.
14501 + Respect shebang if filter is executable (#1389).
14502 + Don't print misleading error message.
14503 Previously pandoc would say that a filter was not found,
14504 even in a case where the filter had a syntax error.
14508 + Parse `div` and `span` elements even without `--parse-raw`,
14509 provided `native_divs` and `native_spans` extensions are set.
14510 Motivation: these now generate native pandoc Div and Span
14511 elements, not raw HTML.
14512 + Parse EPUB-specific elements if the `epub_html_exts`
14513 extension is enabled. These include `switch`, `footnote`,
14514 `rearnote`, `noteref`.
14518 + Support for inline LaTeX. Inline LaTeX is now accepted and parsed by the
14519 org-mode reader. Both math symbols (like `\tau`) and LaTeX commands (like
14520 `\cite{Coffee}`), can be used without any further escaping (Albert
14523 * Textile reader and writer:
14525 + The `raw_tex` extension is no longer set by default. You can
14526 enable it with `textile+raw_tex`.
14530 + Support `equation`, `informalequation`, `inlineequation` elements with
14531 `mml:math` content. This is converted into LaTeX and put into a Pandoc
14534 * Revised `plain` output, largely following the style of Project
14537 + Emphasis is rendered with `_underscores_`, strong emphasis
14539 + Headings are rendered differently, with space to set them off,
14540 not with setext style underlines. Level 1 headers are ALL CAPS.
14541 + Math is rendered using unicode when possible, but without the
14542 distracting emphasis markers around variables.
14543 + Footnotes use a regular `[n]` style.
14547 + Horizontal rules are now a line across the whole page.
14548 + Prettier pipe tables. Columns are now aligned (#1323).
14549 + Respect the `raw_html` extension. `pandoc -t markdown-raw_html`
14550 no longer emits any raw HTML, including span and div tags
14551 generated by Span and Div elements.
14552 + Use span with style for `SmallCaps` (#1360).
14556 + Autolinks now have class `uri`, and email autolinks have class
14557 `email`, so they can be styled.
14561 + Document formatting is carried over from `reference.docx`.
14562 This includes margins, page size, page orientation, header,
14563 and footer, including images in headers and footers.
14564 + Include abstract (if present) with `Abstract` style (#1451).
14565 + Include subtitle (if present) with `Subtitle` style, rather
14566 than tacking it on to the title (#1451).
14570 + Write empty span elements with an id attribute as org anchors.
14571 For example `Span ("uid",[],[]) []` becomes `<<uid>>`.
14575 + Put table captions above tables, to match the conventional
14576 standard. (Previously they appeared below tables.)
14577 + Use `\(..\)` instead of `$..$` for inline math (#1464).
14578 + Use `\nolinkurl` in email autolinks. This allows them to be styled
14579 using `\urlstyle{tt}`. Thanks to Ulrike Fischer for the solution.
14580 + Use `\textquotesingle` for `'` in inline code. Otherwise we get
14581 curly quotes in the PDF output (#1364).
14582 + Use `\footnote<.>{..}` for notes in beamer, so that footnotes
14583 do not appear before the overlays in which their markers appear
14585 + Don't produce a `\label{..}` for a Div or Span element. Do produce
14586 a `\hyperdef{..}` (#1519).
14590 + If the metadata includes `page-progression-direction` (which can be
14591 `ltr` or `rtl`, the `page-progression-direction` attribute will
14592 be set in the EPUB spine (#1455).
14594 * Custom lua writers:
14596 + Custom writers now work with `--template`.
14597 + Removed HTML header scaffolding from `sample.lua`.
14598 + Made citation information available in lua writers.
14600 * `--normalize` and `Text.Pandoc.Shared.normalize` now consolidate
14601 adjacent `RawBlock`s when possible.
14605 * Added `Text.Pandoc.Readers.Docx`, exporting `readDocx` (Jesse Rosenthal).
14607 * Added `Text.Pandoc.Readers.EPUB`, exporting `readEPUB` (Matthew
14610 * Added `Text.Pandoc.Readers.Txt2Tags`, exporting `readTxt2Tags` (Matthew
14613 * Added `Text.Pandoc.Writers.DokuWiki`, exporting `writeDokuWiki`
14616 * Added `Text.Pandoc.Writers.Haddock`, exporting `writeHaddock`.
14618 * Added `Text.Pandoc.MediaBag`, exporting `MediaBag`, `lookupMedia`,
14619 `insertMedia`, `mediaDirectory`, `extractMediaBag`. The docx and epub
14620 readers return a pair of a `Pandoc` document and a `MediaBag` with
14621 the media resources they contain. This can be extracted using
14622 `--extract-media`. Writers that incorporate media (PDF, Docx,
14623 ODT, EPUB, RTF, or HTML formats with `--self-contained`) will look
14624 for resources in the `MediaBag` generated by the reader, in addition to
14625 the file system or web.
14627 * `Text.Pandoc.Readers.TexMath`: Removed deprecated `readTeXMath`.
14628 Renamed `readTeXMath'` to `texMathToInlines`.
14630 * `Text.Pandoc`: Added `Reader` data type (Matthew Pickering).
14631 `readers` now associates names of readers with `Reader`
14632 structures. This allows inclusion of readers, like the docx
14633 reader, that take binary rather than textual input.
14635 * `Text.Pandoc.Shared`:
14637 + Added `capitalize` (Artyom Kazak), and replaced uses of
14638 `map toUpper` (which give bad results for many languages).
14639 + Added `collapseFilePath`, which removes intermediate `.` and
14640 `..` from a path (Matthew Pickering).
14641 + Added `fetchItem'`, which works like `fetchItem` but searches
14642 a `MediaBag` before looking on the net or file system.
14643 + Added `withTempDir`.
14644 + Added `removeFormatting`.
14645 + Added `extractSpaces` (from HTML reader) and generalized its type
14646 so that it can be used by the docx reader (Matthew Pickering).
14648 + Added `normalizeInlines`, `normalizeBlocks`.
14649 + `normalize` is now `Pandoc -> Pandoc` instead of
14650 `Data a :: a -> a`. Some users may need to change their uses of
14651 `normalize` to the newly exported `normalizeInlines` or
14654 * `Text.Pandoc.Options`:
14656 + Added `writerMediaBag` to `WriterOptions`.
14657 + Removed deprecated and no longer used `readerStrict` in
14658 `ReaderOptions`. This is handled by `readerExtensions` now.
14659 + Added `Ext_compact_definition_lists`.
14660 + Added `Ext_epub_html_exts`.
14661 + Added `Ext_native_divs` and `Ext_native_spans`.
14662 This allows users to turn off the default pandoc behavior of
14663 parsing contents of div and span tags in markdown and HTML
14664 as native pandoc Div blocks and Span inlines.
14666 * `Text.Pandoc.Parsing`:
14668 + Generalized `readWith` to `readWithM` (Matthew Pickering).
14669 + Export `runParserT` and `Stream` (Matthew Pickering).
14670 + Added `HasQuoteContext` type class (Matthew Pickering).
14671 + Generalized types of `mathInline`, `smartPunctuation`, `quoted`,
14672 `singleQuoted`, `doubleQuoted`, `failIfInQuoteContext`,
14673 `applyMacros` (Matthew Pickering).
14674 + Added custom `token` (Matthew Pickering).
14675 + Added `stateInHtmlBlock` to `ParserState`. This is used to keep
14676 track of the ending tag we're waiting for when we're parsing inside
14678 + Added `stateMarkdownAttribute` to `ParserState`. This is used
14679 to keep track of whether the markdown attribute has been set in
14681 + Generalized type of `registerHeader`, using new type classes
14682 `HasReaderOptions`, `HasIdentifierList`, `HasHeaderMap` (Matthew
14683 Pickering). These allow certain common functions to be reused
14684 even in parsers that use custom state (instead of `ParserState`),
14685 such as the MediaWiki reader.
14686 + Moved `inlineMath`, `displayMath` from Markdown reader to Parsing,
14687 and generalized their types (Matthew Pickering).
14689 * `Text.Pandoc.Pretty`:
14692 + Added `blanklines`, which guarantees a certain number of blank lines
14699 + Fixed parsing of indented code in list items. Indented code
14700 at the beginning of a list item must be indented eight spaces
14701 from the margin (or edge of the container), or four spaces
14702 from the list marker, whichever is greater.
14703 + Fixed small bug in HTML parsing with `markdown_attribute`, which
14704 caused incorrect tag nesting for input like
14705 `<aside markdown="1">*hi*</aside>`.
14706 + Fixed regression with intraword underscores (#1121).
14707 + Improved parsing of inline links containing quote characters (#1534).
14708 + Slight rewrite of `enclosure`/`emphOrStrong` code.
14709 + Revamped raw HTML block parsing in markdown (#1330).
14710 We no longer include trailing spaces and newlines in the
14711 raw blocks. We look for closing tags for elements (but without
14712 backtracking). Each block-level tag is its own `RawBlock`;
14713 we no longer try to consolidate them (though `--normalize` will do so).
14714 + Combine consecutive latex environments. This helps when you have
14715 two minipages which can't have blank lines between them (#690, #1196).
14716 + Support smallcaps through span.
14717 `<span style="font-variant:small-caps;">foo</span>` will be
14718 parsed as a `SmallCaps` inline, and will work in all output
14719 formats that support small caps (#1360).
14720 + Prevent spurious line breaks after list items (#1137). When the
14721 `hard_line_breaks` option was specified, pandoc would formerly
14722 produce a spurious line break after a tight list item.
14723 + Fixed table parsing bug (#1333).
14724 + Handle `c++` and `objective-c` as language identifiers in
14725 github-style fenced blocks (#1318).
14726 + Inline math must have nonspace before final `$` (#1313).
14730 + Handle comments at the end of tables. This resolves the issue
14731 illustrated in <http://stackoverflow.com/questions/24009489>.
14732 + Correctly handle table rows with too few cells. LaTeX seems to
14733 treat them as if they have empty cells at the end (#241).
14734 + Handle leading/trailing spaces in `\emph` better.
14735 `\emph{ hi }` gets parsed as `[Space, Emph [Str "hi"], Space]`
14736 so that we don't get things like `* hi *` in markdown output.
14737 Also applies to `textbf` and some other constructions (#1146).
14738 + Don't assume preamble doesn't contain environments (#1338).
14739 + Allow (and discard) optional argument for `\caption` (James Aspnes).
14743 + Fixed major parsing problem with HTML tables. Table cells were
14744 being combined into one cell (#1341).
14745 + Fixed performance issue with malformed HTML tables.
14746 We let a `</table>` tag close an open `<tr>` or `<td>` (#1167).
14747 + Allow space between `<col>` and `</col>`.
14748 + Added `audio` and `source` in `eitherBlockOrInline`.
14749 + Moved `video`, `svg`, `progress`, `script`, `noscript`, `svg` from
14750 `blockTags` to `eitherBlockOrInline`.
14751 + `map` and `object` were mistakenly in both lists; they have been removed
14753 + Ignore `DOCTYPE` and `xml` declarations.
14755 * MediaWiki reader:
14757 + Don't parse backslash escapes inside `<source>` (#1445).
14758 + Tightened up template parsing.
14759 The opening `{{` must be followed by an alphanumeric or `:`.
14760 This prevents the exponential slowdown in #1033.
14761 + Support "Bild" for images.
14765 + Better handle elements inside code environments. Pandoc's document
14766 model does not allow structure inside code blocks, but at least this way
14767 we preserve the text (#1449).
14768 + Support `<?asciidoc-br?>` (#1236).
14772 + Fixed list parsing. Lists can now start without an intervening
14773 blank line (#1513).
14774 + HTML block-level tags that do not start a line are parsed as
14775 inline HTML and do not interrupt paragraphs (as in RedCloth).
14779 + Make tildes create inline code (#1345). Also relabeled `code` and
14780 `verbatim` parsers to accord with the org-mode manual.
14781 + Respect `:exports` header argument in code blocks (Craig Bosma).
14782 + Fixed tight lists with sublists (#1437).
14786 + Avoid excess whitespace in `nav.xhtml`. This should improve
14787 TOC view in iBooks (#1392).
14788 + Fixed regression on cover image.
14789 In 1.12.4 and 1.12.4.2, the cover image would not appear properly,
14790 because the metadata id was not correct. Now we derive the id from the
14791 actual cover image filename, which we preserve rather than using
14793 + Keep newlines between block elements. This allows
14794 easier diff-ability (#1424).
14795 + Use `stringify` instead of custom `plainify`.
14796 + Use `renderTags'` for all tag rendering. This properly handles tags
14797 that should be self-closing. Previously `<hr/>` would appear in EPUB
14798 output as `<hr></hr>` (#1420).
14799 + Better handle HTML media tags.
14800 + Handle multiple dates with OPF `event` attributes. Note: in EPUB3 we
14801 can have only one dc:date, so only the first one is used.
14805 + Correctly handle figures in notes. Notes can't contain figures in
14806 LaTeX, so we fake it to avoid an error (#1053).
14807 + Fixed strikeout + highlighted code (#1294).
14808 Previously strikeout highlighted code caused an error.
14812 + Improved detection of autolinks with URLs containing escapes.
14816 + Improved image embedding: `fetchItem'` is now used to get the
14817 images, and calculated image sizes are indicated in the RTF.
14818 + Avoid extra paragraph tags in metadata (#1421).
14822 + Deactivate "incremental" inside slide speaker notes (#1394).
14823 + Don't include empty items in the table of contents for
14824 slide shows. (These would result from creating a slide
14825 using a horizontal rule.)
14827 * MediaWiki writer:
14829 + Minor renaming of `st` prefixed names.
14833 + Double up emphasis and strong emphasis markers in intraword
14834 contexts, as required by asciidoc (#1441).
14838 + Avoid wrapping that might start a list, blockquote, or header (#1013).
14839 + Use Span instead of (hackish) `SmallCaps` in `plainify`.
14840 + Don't use braced attributes for fenced code (#1416).
14841 If `Ext_fenced_code_attributes` is not set, the first class
14842 attribute will be printed after the opening fence as a bare word.
14843 + Separate adjacent lists of the same kind with an HTML comment (#1458).
14847 + Fixed treatment of data uris for images (#1062).
14851 + Use Compact style for empty table cells (#1353).
14852 Otherwise we get overly tall lines when there are empty
14853 table cells and the other cells are compact.
14854 + Create overrides per-image for `media/` in reference docx.
14855 This should be somewhat more robust and cover more types of images.
14856 + Improved `entryFromArchive` to avoid an unneeded parse.
14857 + Section numbering carries over from reference.docx (#1305).
14858 + Simplified `abstractNumId` numbering. Instead of sequential numbering,
14859 we assign numbers based on the list marker styles.
14861 * `Text.Pandoc.Options`:
14863 + Removed `Ext_fenced_code_attributes` from `markdown_github`
14866 * `Text.Pandoc.ImageSize`:
14868 + Use default instead of failing if image size not found
14869 in exif header (#1358).
14870 + ignore unknown exif header tag rather than crashing.
14871 Some images seem to have tag type of 256, which was causing
14874 * `Text.Pandoc.Shared`:
14876 + `fetchItem`: unescape URI encoding before reading local file (#1427).
14877 + `fetchItem`: strip a fragment like `?#iefix` from the extension before
14878 doing mime lookup, to improve mime type guessing.
14879 + Improved logic of `fetchItem`: absolute URIs are fetched from the net;
14880 other things are treated as relative URIs if `sourceURL` is `Just _`,
14881 otherwise as file paths on the local file system.
14882 + `fetchItem` now properly handles links without a protocol (#1477).
14883 + `fetchItem` now escapes characters not allowed in URIs before trying
14885 + Fixed runtime error with `compactify'DL` on certain lists (#1452).
14887 * `pandoc.hs`: Don't strip path off of `writerSourceURL`: the path is
14888 needed to resolve relative URLs when we fetch resources (#750).
14890 * `Text.Pandoc.Parsing`
14892 + Simplified `dash` and `ellipsis` (#1419).
14893 + Removed `(>>~)` in favor of the equivalent `(<*)` (Matthew Pickering).
14894 + Generalized functions to use `ParsecT` (Matthew Pickering).
14895 + Added `isbn` and `pmid` to list of recognized schemes (Matthew
14898### Template changes
14900 * Added haddock template.
14901 * EPUB3: Added `type` attribute to `link` tags. They are supposed to
14902 be "advisory" in HTML5, but kindlegen seems to require them.
14903 * EPUB3: Put title page in section with `epub:type="titlepage"`.
14904 * LaTeX: Made `\subtitle` work properly (#1327).
14905 * LaTeX/Beamer: remove conditional around date (#1321).
14906 * LaTeX: Added `lot` and `lof` variables, which can be set to
14907 get `\listoftables` and `\listoffigures` (#1407). Note that
14908 these variables can be set at the command line with `-Vlot -Vlof`
14909 or in YAML metadata.
14911### Under the hood improvements
14913 * Rewrote normalize for efficiency (#1385).
14915 * Rewrote Haddock reader to use `haddock-library` (#1346).
14917 + This brings pandoc's rendering of haddock markup in line
14918 with the new haddock.
14919 + Fixed line breaks in `@` code blocks.
14920 + alex and happy are no longer build-depends.
14922 * Added `Text.Pandoc.Compat.Directory` to allow building against
14923 different versions of the `directory` library.
14925 + Added `Text.Pandoc.Compat.Except` to allow building against
14926 different versions of `mtl`.
14928 * Code cleanup in some writers, using Reader monad to avoid
14929 passing options parameter around (Matej Kollar).
14931 * Improved readability in `pandoc.hs`.
14933 * Miscellaneous code cleanups (Artyom Kazak).
14935 * Avoid `import Prelude hiding (catch)` (#1309, thanks to Michael
14938 * Changed `http-conduit` flag to `https`. Depend on `http-client`
14939 and `http-client-tls` instead of `http-conduit`. (Note: pandoc still
14940 depends on `conduit` via `yaml`.)
14942 * Require `highlighting-kate >= 0.5.8.5` (#1271, #1317, Debian #753299).
14943 This change to highlighting-kate means that PHP fragments no longer need
14944 to start with `<?php`. It also fixes a serious bug causing failures with
14947 * Require latest `texmath`. This fixes `\tilde{E}` and allows
14948 `\left` to be used with `]`, `)` etc. (#1319), among many other
14951 * Require latest `zip-archive`. This has fixes for unicode path names.
14953 * Added tests for plain writer.
14955 * `Text.Pandoc.Templates`:
14957 + Fail informatively on template syntax errors.
14958 With the move from parsec to attoparsec, we lost good error
14959 reporting. In fact, since we weren't testing for end of input,
14960 malformed templates would fail silently. Here we revert back to
14961 Parsec for better error messages.
14962 + Use `ordNub` (#1022).
14966 + Made benchmarks compile again (Artyom Kazak).
14967 + Fixed so that the failure of one benchmark does not prevent others
14968 from running (Artyom Kazak).
14969 + Use `nfIO` instead of the `getLength` trick to force full evaluation.
14970 + Changed benchmark to use only the test suite, so that benchmarks
14973 * Windows build script:
14975 + Add `-windows` to file name.
14976 + Use one install command for pandoc, pandoc-citeproc.
14977 + Force install of pandoc-citeproc.
14979 * `make_osx_package`: Call zip file `pandoc-VERSION-osx.zip`.
14980 The zip should not be named `SOMETHING.pkg.zip`, or OSX finder
14981 will extract it into a folder named `SOMETHING.pkg`, which it
14982 will interpret as a defective package (#1308).
14986 + Made headers for all extensions so they have IDs and can be
14987 linked to (Beni Cherniavsky-Paskin).
14988 + Fixed typos (Phillip Alday).
14989 + Fixed documentation of attributes (#1315).
14990 + Clarified documentation on small caps (#1360).
14991 + Better documentation for `fenced_code_attributes` extension
14993 + Documented fact that you can put YAML metadata in a separate file
14997## pandoc 1.12.4.2 (2014-05-14)
14999 * Require highlighting-kate >= 0.5.8. Fixes a performance regression.
15001 * Shared: `addMetaValue` now behaves slightly differently:
15002 if both the new and old values are lists, it concatenates their
15003 contents to form a new list.
15007 + Set `bibliography` in metadata from `\bibliography` or
15008 `\addbibresource` command.
15009 + Don't error on `%foo` with no trailing newline.
15013 + Support code block headers (`#+BEGIN_SRC ...`) (Albert Krewinkel).
15014 + Fix parsing of blank lines within blocks (Albert Krewinkel).
15015 + Support pandoc citation extension (Albert Krewinkel). This can
15016 be turned off by specifying `org-citation` as the input format.
15020 + `citeKey` moved to `Text.Pandoc.Parsing` so it can be used by
15021 other readers (Albert Krewinkel).
15023 * `Text.Pandoc.Parsing`:
15025 + Added `citeKey` (see above).
15026 + Added `HasLastStrPosition` type class and `updateLastStrPos`
15027 and `notAfterString` functions.
15029 * Updated copyright notices (Albert Krewinkel).
15031 * Added default.icml to data files so it installs with the package.
15035 + The binary is now built with options to ensure that it can be
15036 used with OSX 10.6+.
15037 + Moved OSX package materials to osx directory.
15038 + Added OSX package uninstall script, included in the zip container
15039 (thanks to Daniel T. Staal).
15041## pandoc 1.12.4 (2014-05-07)
15043 * Made it possible to run filters that aren't executable
15044 (#1096). Pandoc first tries to find the executable (searching
15045 the path if path isn't given). If it fails, but the file
15046 exists and has a `.py`, `.pl`, `.rb`, `.hs`, or `.php`
15047 extension, pandoc runs the filter using the appropriate
15048 interpreter. This should make it easier to use filters on
15049 Windows, and make it more convenient for everyone.
15051 * Added Emacs org-mode reader (Albert Krewinkel).
15053 * Added InDesign ICML Writer (mb21).
15055 * MediaWiki reader:
15057 + Accept image links in more languages (Jaime Marquínez Ferrándiz).
15058 + Fixed bug in certain nested lists (#1213). If a level 2 list was
15059 followed by a level 1 list, the first item of the level 1 list
15061 + Handle table rows containing just an HTML comment (#1230).
15065 + Give better location information on errors, pointing to line
15066 numbers within included files (#1274).
15067 + LaTeX reader: Better handling of `table` environment (#1204).
15068 Positioning options no longer rendered verbatim.
15069 + Better handling of figure and table with caption (#1204).
15070 + Handle `@{}` and `p{length}` in tabular. The length is not actually
15071 recorded, but at least we get a table (#1180).
15072 + Properly handle `\nocite`. It now adds a `nocite` metadata
15073 field. Citations there will appear in the bibliography but not
15074 in the text (unless you explicitly put a `$nocite$` variable
15079 + Ensure that whole numbers in YAML metadata are rendered without
15080 decimal points. (This became necessary with changes to aeson
15081 and yaml libraries. aeson >= 0.7 and yaml >= 0.8.8.2 are now required.)
15082 + Fixed regression on line breaks in strict mode (#1203).
15083 + Small efficiency improvements.
15084 + Improved parsing of nested `div`s. Formerly a closing `div` tag
15085 would be missed if it came right after other block-level tags.
15086 + Avoid backtracking when closing `</div>` not found.
15087 + Fixed bug in reference link parsing in `markdown_mmd`.
15088 + Fixed a bug in list parsing (#1154). When reading a raw list
15089 item, we now strip off up to 4 spaces.
15090 + Fixed parsing of empty reference link definitions (#1186).
15091 + Made one-column pipe tables work (#1218).
15095 + Better support for attributes. Instead of being ignored, attributes
15096 are now parsed and included in Span inlines. The output will be a bit
15097 different from stock textile: e.g. for `*(foo)hi*`, we'll get
15098 `<em><span class="foo">hi</span></em>` instead of
15099 `<em class="foo">hi</em>`. But at least the data is not lost.
15100 + Improved treatment of HTML spans (%) (#1115).
15101 + Improved link parsing. In particular we now pick up on attributes.
15102 Since pandoc links can't have attributes, we enclose the whole link in
15103 a span if there are attributes (#1008).
15104 + Implemented correct parsing rules for inline markup (#1175, Matthew
15106 + Use Builder (Matthew Pickering).
15110 + Better treatment of `formalpara`. We now emit the title (if present)
15111 as a separate paragraph with boldface text (#1215).
15112 + Set metadata `author` not `authors`.
15113 + Added recognition of `authorgroup` and `releaseinfo` elements (#1214,
15114 Matthew Pickering).
15115 + Converted current meta information parsing in DocBook to a more
15116 extensible version which is aware of the more recent meta
15117 representation (Matthew Pickering).
15121 + Require tagsoup 0.13.1, to fix a bug with parsing of script tags
15123 + Treat processing instructions & declarations as block. Previously
15124 these were treated as inline, and included in paragraph tags in HTML
15125 or DocBook output, which is generally not what is wanted (#1233).
15126 + Updated `closes` with rules from HTML5 spec.
15127 + Use Builder (Matthew Pickering, #1162).
15131 + Remove duplicate `http` in PEP links (Albert Krewinkel).
15132 + Make rst figures true figures (#1168, CasperVector)
15133 + Enhanced Pandoc's support for rST roles (Merijn Verstaaten).
15134 rST parser now supports: all built-in rST roles, new role definition,
15135 role inheritance, though with some limitations.
15136 + Use `author` rather than `authors` in metadata.
15137 + Better handling of directives. We now correctly handle field
15138 lists that are indented more than three spaces. We treat an
15139 `aafig` directive as a code block with attributes, so it can be
15140 processed in a filter (#1212).
15144 + Mark span contents with label if span has an ID (Albert Krewinkel).
15145 + Made `--toc-depth` work well with books in latex/pdf output (#1210).
15146 + Handle line breaks in simple table cells (#1217).
15147 + Workaround for level 4-5 headers in quotes. These previously produced
15148 invalid LaTeX: `\paragraph` or `\subparagraph` in a `quote` environment.
15149 This adds an `mbox{}` in these contexts to work around the problem.
15150 See <http://tex.stackexchange.com/a/169833/22451> (#1221).
15151 + Use `\/` to avoid en-dash ligature instead of `-{}-` (Vaclav Zeman).
15152 This is to fix LuaLaTeX output. The `-{}-` sequence does not avoid the
15153 ligature with LuaLaTeX but `\/` does.
15154 + Fixed string escaping in `hyperref` and `hyperdef` (#1130).
15156 * ConTeXt writer: Improved autolinks (#1270).
15160 + Improve handling of hard line breaks in Docbook writer
15161 (Neil Mayhew). Use a `<literallayout>` for the entire paragraph, not
15162 just for the newline character.
15163 + Don't let line breaks inside footnotes influence the enclosing
15164 paragraph (Neil Mayhew).
15165 + Distinguish tight and loose lists in DocBook output, using
15166 `spacing="compact"` (Neil Mayhew, #1250).
15168 * Docx writer: When needed files are not present in the user's
15169 `reference.docx`, fall back on the versions in the `reference.docx`
15170 in pandoc's data files. This fixes a bug that occurs when a
15171 `reference.docx` saved by LibreOffice is used. (#1185)
15175 + Include extension in epub ids. This fixes a problem with duplicate
15176 extensions for fonts and images with the same base name but different
15177 extensions (#1254).
15178 + Handle files linked in raw `img` tags (#1170).
15179 + Handle media in `audio` source tags (#1170).
15180 Note that we now use a `media` directory rather than `images`.
15181 + Incorporate files linked in `video` tags (#1170). `src` and `poster`
15182 will both be incorporated into `content.opf` and the epub container.
15186 + Add colgroup around col tags (#877). Also affects EPUB writer.
15187 + Fixed bug with unnumbered section headings. Unnumbered section
15188 headings (with class `unnumbered`) were getting numbers.
15189 + Improved detection of image links. Previously image links with
15190 queries were not recognized, causing `<embed>` to be used instead
15193 * Man writer: Ensure that terms in definition lists aren't line wrapped
15198 + Use proper escapes to avoid unwanted lists (#980). Previously we used
15199 0-width spaces, an ugly hack.
15200 + Use longer backtick fences if needed (#1206). If the content contains a
15201 backtick fence and there are attributes, make sure longer fences are
15202 used to delimit the code. Note: This works well in pandoc, but github
15203 markdown is more limited, and will interpret the first string of three
15204 or more backticks as ending the code block.
15206 * RST writer: Avoid stack overflow with certain tables (#1197).
15208 * RTF writer: Fixed table cells containing paragraphs.
15212 + Correctly handle UTF-8 in custom lua scripts (#1189).
15213 + Fix bugs with lua scripts with mixed-case filenames and
15214 paths containing `+` or `-` (#1267). Note that `getWriter`
15215 in `Text.Pandoc` no longer returns a custom writer on input
15218 * AsciiDoc writer: Handle multiblock and empty table cells
15219 (#1245, #1246). Added tests.
15221 * `Text.Pandoc.Options`: Added `readerTrace` to `ReaderOptions`
15223 * `Text.Pandoc.Shared`:
15225 + Added `compactify'DL` (formerly in markdown reader) (Albert Krewinkel).
15226 + Fixed bug in `toRomanNumeral`: numbers ending with '9' would
15227 be rendered as Roman numerals ending with 'IXIV' (#1249). Thanks to
15229 + `openURL`: set proxy with value of http_proxy env variable (#1211).
15230 Note: proxies with non-root paths are not supported, due to
15231 limitations in `http-conduit`.
15233 * `Text.Pandoc.PDF`:
15235 + Ensure that temp directories deleted on Windows (#1192). The PDF is
15236 now read as a strict bytestring, ensuring that process ownership will
15237 be terminated, so the temp directory can be deleted.
15238 + Use `/` as path separators in a few places, even on Windows.
15239 This seems to be necessary for texlive (#1151, thanks to Tim Lin).
15240 + Use `;` for `TEXINPUTS` separator on Windows (#1151).
15241 + Changes to error reporting, to handle non-UTF8 error output.
15243 * `Text.Pandoc.Templates`:
15245 + Removed unneeded datatype context (Merijn Verstraaten).
15247 + YAML objects resolve to "true" in conditionals (#1133).
15248 Note: If `address` is a YAML object and you just have `$address$`
15249 in your template, the word `true` will appear, which may be
15250 unexpected. (Previously nothing would appear.)
15252 * `Text.Pandoc.SelfContained`: Handle `poster` attribute in `video`
15255 * `Text.Pandoc.Parsing`:
15257 + Made `F` an instance of Applicative (#1138).
15258 + Added `stateCaption`.
15259 + Added `HasMacros`, simplified other typeclasses.
15260 Removed `updateHeaderMap`, `setHeaderMap`, `getHeaderMap`,
15261 `updateIdentifierList`, `setIdentifierList`, `getIdentifierList`.
15262 + Changed the smart punctuation parser to return `Inlines`
15263 rather than `Inline` (Matthew Pickering).
15264 + Changed `HasReaderOptions`, `HasHeaderMap`, `HasIdentifierList`
15265 from typeclasses of monads to typeclasses of states. This simplifies
15266 the instance definitions and provides more flexibility. Generalized
15267 type of `getOption` and added a default definition. Removed
15268 `askReaderOption`. Added `extractReaderOption`. Added
15269 `extractHeaderMap` and `updateHeaderMap` in `HasHeaderMap`.
15270 Gave default definitions for `getHeaderMap`, `putHeaderMap`,
15271 `modifyHeaderMap`. Added `extractIdentifierList` and
15272 `updateIdentifierList` in `HasIdentifierList`. Gave defaults
15273 for `getIdentifierList`, `putIdentifierList`, and
15274 `modifyIdentifierList`. The ultimate goal here is to allow different
15275 parsers to use their own, tailored parser states (instead of
15276 `ParserState`) while still using shared functions.
15278 * Template changes:
15280 + LaTeX template: Use `fontenc` package only with `pdflatex` (#1164).
15281 + LaTeX template: Add `linestretch` and `fontfamily` variables.
15282 + LaTeX template: Conditionalize author and date commands.
15283 + Beamer template: Consistent styles for figure and table captions
15285 + LaTeX and beamer template: Adjust widths correctly for oversized
15286 images. Use `\setkeys{Gin}{}` to set appropriate defaults for
15287 `\includegraphics` (Yihui Xie, Garrick Aden-Buie). Load
15288 `upquote` only after `fontenc` (Yihui Xie).
15289 + Beamer template: Added caption package (#1200).
15290 + Beamer template: changes for better unicode handling (KarolS).
15291 + DocBook template: use `authorgroup` if there are authors.
15292 + revealjs template: Move `include-after` to end (certainlyakey).
15293 + revealjs template: Fixed PDF print function (#1220, kevinkenan).
15295 * Bumped version bounds of dependencies.
15297 * Added a `--trace` command line option, for debugging backtracking
15298 bugs. So far this only works with the markdown reader.
15300 * MathMLinHTML: Fixed deprecation warning (#362, gwern, Albert Krewinkel).
15302 * Updated travis script to test with multiple GHC versions.
15304 * Force failure of a Travis build if GHC produces warnings (Albert
15307 * Add `.editorconfig` (Albert Krewinkel).
15308 See <http://editorconfig.org/> for details.
15310 * Give more useful error message if '-t pdf' is specified (#1155).
15312 * Added `Cite`, `SmallCaps` to `Arbitrary` instance (#1269).
15314 * Allow `html4` as a synonym of `html` as a reader (it already works
15319 + Added an explanation of how to use YAML metadata to
15320 force items to appear in the bibliography without citations in
15321 the text (like LaTeX `\nocite`).
15322 + Added note to `--bibtex/--natbib`: not for use in making PDF
15323 (#1194, thanks to nahoj).
15324 + Added explanatory notes about `--natbib` and `--biblatex`.
15325 + Added specification of legal syntax for citation keys.
15326 + Fixed variable defaults documentation (Albert Krewinkel).
15328 * Removed copyright statements for files that have been removed
15329 (Albert Krewinkel).
15331 * Moved some doc files from `data-files` to `extra-source-files` (#1123).
15332 They aren't needed at runtime. We keep README and COPYRIGHT in data
15333 to ensure that they'll be available on all systems on which pandoc
15336 * Use cabal sandboxes in Windows build script.
15338## pandoc 1.12.3.3 (2014-02-03)
15340 * To changes to source; recompiled tarball with latest alex and
15341 happy, so they will work with GHC 7.8.
15343## pandoc 1.12.3.2 (2014-02-03)
15345 * Bumped version bounds for blaze-html, blaze-markup.
15347 * ImageSize: Avoid use of lookAhead, which is not in binary >= 0.6
15350 * Fixed mediawiki ordered list parsing (#1122).
15352 * HTML reader: Fixed bug reading inline math with `$$` (#225).
15354 * Added support for LaTeX style literate Haskell code blocks in rST
15355 (Merijn Verstraaten).
15357## pandoc 1.12.3.1 (2014-01-14)
15359 * Relaxed version constraint on binary, allowing the use of binary 0.5.
15362## pandoc 1.12.3 (2014-01-10)
15364 * The `--bibliography` option now sets the `biblio-files` variable.
15365 So, if you're using `--natbib` or `--biblatex`, you can just use
15366 `--bibliography=foo.bib` instead of `-V bibliofiles=foo`.
15368 * Don't run pandoc-citeproc filter if `--bibliography` is
15369 used together with `--natbib` or `--biblatex` (Florian Eitel).
15371 * Template changes:
15373 + Updated beamer template to include booktabs.
15374 + Added `abstract` variable to LaTeX template.
15375 + Put `header-includes` after `title` in LaTeX template (#908).
15376 + Allow use of `\includegraphics[size]` in beamer.
15377 This just required porting a macro definition from the default
15378 LaTeX template to the default beamer template.
15380 * `reference.docx`: Include `FootnoteText` style.
15381 Otherwise Word ignores the style, even when specified in the `pPr`.
15384 * `reference.odt`: Tidied `styles.xml`.
15386 * Relaxed version bounds for dependencies.
15388 * Added `withSocketsDo` around http conduit code in `openURL`,
15389 so it works on Windows (#1080).
15391 * Added `Cite` function to `sample.lua`.
15395 + Fixed regression in title blocks (#1089).
15396 If author field was empty, date was being ignored.
15397 + Allow backslash-newline hard line breaks in grid and
15398 multiline table cells.
15399 + Citation keys may now start with underscores, and may contain
15400 underscores adjacent to internal punctuation.
15404 + Add support for `Verb` macro (jrnold) (#1090).
15405 + Support babel-style quoting: `` "`..."' ``.
15407 * Properly handle script blocks in strict mode. (That is,
15408 `markdown-markdown_in_html_blocks`.) Previously a spurious
15409 `<p>` tag was being added (#1093).
15411 * Docbook reader: Avoid failure if `tbody` contains no `tr` or `row`
15416 + Factored out function for table cell creation.
15417 + Better treatment of footnotes in tables.
15418 Notes now appear in the regular sequence, rather than in the
15419 table cell. (This was a regression in 1.10.)
15421 * HTML reader: Parse name/content pairs from meta tags as metadata.
15424 * Moved `fixDisplayMath` from Docx writer to `Writer.Shared`.
15426 * OpenDocument writer: Fixed `RawInline`, `RawBlock` so they don't escape.
15428 * ODT writer: Use mathml for proper rendering of formulas.
15429 Note: LibreOffice's support for this seems a bit buggy. But
15430 it should be better than what we had before.
15432 * RST writer: Ensure no blank line after def in definition list (#992).
15434 * Markdown writer: Don't use tilde code blocks with braced attributes in
15435 `markdown_github` output. A consequence of this change is that the
15436 backtick form will be preferred in general if both are enabled. That
15437 is good, as it is much more widespread than the tilde form. (#1084)
15439 * Docx writer: Fixed problem with some modified reference docx files.
15440 Include `word/_rels/settings.xml.rels` if it exists, as well as other
15441 `rels` files besides the ones pandoc generates explicitly.
15445 + With `--toc`, headers no longer link to themselves (#1081).
15446 + Omit footnotes from TOC entries. Otherwise we get doubled
15447 footnotes when headers have notes!
15451 + Avoid duplicate notes when headings contain notes.
15452 This arose because the headings are copied into the metadata
15453 "title" field, and the note gets rendered twice. We strip the
15454 note now before putting the heading in "title".
15455 + Strip out footnotes from toc entries.
15456 + Fixed bug with `--epub-stylesheet`. Now the contents of
15457 `writerEpubStylesheet` (set by `--epub-stylesheet`)
15458 should again work, and take precedence over a stylesheet specified
15461 * `Text.Pandoc.Pretty`: Added `nestle`. API change.
15463 * `Text.Pandoc.MIME`: Added `wmf`, `emf`.
15465 * `Text.Pandoc.Shared`: `fetchItem` now handles image URLs beginning
15468 * `Text.Pandoc.ImageSize`: Parse EXIF format JPEGs. Previously
15469 we could only get size information for JFIF format, which led
15470 to squished images in Word documents. Closes #976.
15472 * Removed old `MarkdownTest_1.0.3` directory (#1104).
15475## pandoc 1.12.2.1 (2013-12-08)
15477 * Markdown reader: Fixed regression in list parser, involving
15478 continuation lines containing raw HTML (or even verbatim raw HTML).
15480## pandoc 1.12.2 (2013-12-07)
15482 * Metadata may now be included in YAML blocks in a markdown document.
15490 text: An investigation of metadata
15498 text: doi:10.234234.234/33
15499 publisher: My Press
15500 rights: (c) 2007 John Smith, CC BY-NC
15501 cover-image: img/mypic.jpg
15502 stylesheet: style.css
15505 Metadata may still be provided using `--epub-metadata`; it will
15506 be merged with the metadata in YAML blocks.
15510 + `meta` tags are now used instead of `opf` attributes for EPUB3.
15511 + Insert "svg" property as needed in opf (EPUB 3).
15512 + Simplify `imageTypeOf` using `getMimeType`.
15513 + Add properties attribute to `cover-image` item for EPUB 3.
15514 + Don't include node for `cover.xhtml` if no cover!
15515 + Ensure that same identifier is used throughout (#1044).
15516 If an identifier is given in metadata, we use that; otherwise
15517 we generate a random uuid.
15518 + Add cover reference to guide element (EPUB 2) (Shaun Attfield).
15519 Fixes an issue with Calibre putting the cover at the end of the book
15520 if the spine has `linear="no"`. Apparently this is best practice
15521 for other converters as well:
15522 <http://www.idpf.org/epub/20/spec/OPF_2.0.1_draft.htm#Section2.6>.
15523 + Allow `stylesheet` in metadata. The value is a path to the stylesheet.
15524 + Allow partial dates: `YYYY`, `YYYY-MM`.
15526 * Markdown writer: Fix rendering of tight sublists (#1050).
15527 Previously a spurious blank line was included after a tight sublist.
15529 * ODT writer: Add `draw:name` attribute to `draw:frame` elements (#1069).
15530 This is reported to be necessary to avoid an error from recent
15531 versions of Libre Office when files contain more than one image
15532 Thanks to wmanley for reporting and diagnosing the problem.
15534 * ConTeXt writer: Don't hardcode figure/table placement and numbering.
15535 Instead, let this be set in the template, using `\setupfloat`.
15536 Thanks to on4aa and Aditya Mahajan for the suggestion (#1067).
15538 * Implemented CSL flipflopping spans in DOCX, LaTeX, and HTML writers.
15540 * Fixed bug with markdown intraword emphasis. Closes #1066.
15542 * Docbook writer: Hierarchicalize block content in metadata.
15543 Previously headers just disappeared from block-level metadata
15544 when it was used in templates. Now we apply the 'hierarchicalize'
15545 transformation. Note that a block headed by a level-2 header will
15546 turn into a `<sect1>` element.
15548 * OpenDocument writer: Skip raw HTML (#1035).
15549 Previously it was erroneously included as verbatim text.
15551 * HTML/EPUB writer, footnotes: Put `<sup>` tag inside `<a>` tags.
15552 This allows better control of formatting, since the `<a>`
15553 tags have a distinguishing class (#1049).
15557 + Use mime type info returned by fetchItem.
15558 + Fixed core metadata (#1046).
15559 Don't create empty date nodes if no date given.
15560 Don't create multiple `dc:creator` nodes; instead separate by
15562 + Fix URL for core-properties in `_rels/.rels` (#1046).
15564 * Plain writer: don't print `<span>` tags.
15568 + Fix definition lists with internal links in terms (#1032).
15569 This fix puts braces around a term that contains an internal
15570 link, to avoid problems with square brackets.
15571 + Properly escape pdftitle, pdfauthor (#1059).
15572 + Use booktabs package for tables (thanks to Jose Luis Duran).
15574 * Updated beamer template. Now references should work properly
15575 (in a slide) when `--biblatex` or `--natbib` is used.
15579 + Parse contents of curly quotes or matched `"` as quotes.
15580 + Support `\textnormal` as span with class `nodecor`.
15581 This is needed for pandoc-citeproc.
15582 + Improved citation parsing. This fixes a run-time error that occurred
15583 with `\citet{}` (empty list of keys). It also ensures that empty keys
15584 don't get produced.
15586 * MediaWiki reader: Add automatic header identifiers.
15590 + Use pandoc `Div` and `Span` for raw `<div>`, `<span>` when
15592 + Recognize `svg` tags as block level content (thanks to MinRK).
15593 + Parse LaTeX math if appropriate options are set.
15597 + Yaml block must start immediately after `---`. If there's a blank
15598 line after `---`, we interpreted it as a horizontal rule.
15599 + Correctly handle empty bullet list items.
15600 + Stop parsing "list lines" when we hit a block tag.
15601 This fixes exponential slowdown in certain input, e.g.
15602 a series of lists followed by `</div>`.
15604 * Slides: Preserve `<div class="references">` in references slide.
15606 * `Text.Pandoc.Writer.Shared`:
15608 + Fixed bug in `tagWithAttrs`. A space was omitted before key-value
15609 attributes, leading to invalid HTML.
15610 + `normalizeDate`: Allow dates with year only (thanks to Shaun Attfield).
15611 + Fixed bug in `openURL` with `data:` URIs. Previously the base-64
15612 encoded bytestring was returned. We now decode it so it's a proper
15615 * DocBook reader: Handle numerical attributes starting with decimal.
15616 Also use `safeRead` instead of `read`.
15618 * `Text.Pandoc.Parsing`:
15620 + Generalized type of `registerHeader`, using new type classes
15621 `HasReadeOptions`, `HasIdentifierList`, `HasHeaderMap`.
15622 These allow certain common functions to be reused
15623 even in parsers that use custom state (instead of `ParserState`),
15624 such as the MediaWiki reader.
15625 + Moved inlineMath, displayMath from Markdown reader to Parsing.
15626 Generalize their types and export them from Parsing. (API change.)
15628 * `Text.Pandoc.Readers.TexMath`: Export `readTeXMath'`, which attends
15629 to display/inline. Deprecate `readTeXMath`, and use `readTeXMath'`
15630 in all the writers. Require `texmath >= 0.6.5.2`.
15632 * `Text.Pandoc.MIME`:
15634 + Add entry for `jfif`.
15635 + In looking up extensions, drop the encoding info.
15636 E.g. for 'image/jpg;base64' we should lookup 'image/jpg'.
15638 * Templates: Changed how array variables are resolved. Previously if
15639 `foo` is an array (which might be because multiple values were set on
15640 the command line), `$foo$` would resolve to the concatenation of the
15641 elements of foo. This is rarely useful behavior. It has been changed
15642 so that the first value is rendered. Of course, you can still iterate
15643 over the values using `$for(foo)$`. This has the result that you can
15644 override earlier settings using `-V` by putting new values later on the
15645 command line, which is useful for many purposes.
15647 * `Text.Pandoc`: Don't default to `pandocExtensions` for all writers.
15649 * Allow "epub2" as synonym for "epub", "html4" for "html".
15651 * Don't look for slidy files in data files with `--self-contained`.
15653 * Allow `https:` command line arguments to be downloaded.
15655 * Fixed `make_osx_package.sh` so data files embedded in `pandoc-citeproc`.
15657## pandoc 1.12.1 (2013-10-20)
15659 * `Text.Pandoc.Definition`: Changed default JSON serialization format.
15660 Instead of `{"Str": "foo"}`, for example, we now have `{"t": "Str",
15661 "c": "foo"}`. This new format is easier to work with outside of Haskell.
15662 Incidentally, "t" stands for "tag", "c" for "contents".
15664 * MediaWiki reader: Trim contents of `<math>` tags, to avoid problems
15665 when converting to markdown (#1027).
15669 + Ensure that preamble doesn't contribute to the text of
15671 + Fixed character escaping in `\url{}`. Previously `\~` wasn't handled
15672 properly, among others.
15673 + Parse `{groups}` as `Span`. This is needed for accurate conversion of
15674 bibtex titles, since we need to know what was protected from
15675 titlecase conversions.
15679 + Specially escape non-ascii characters in labels.
15680 Otherwise we can get compile errors and other bugs when
15681 compiled with pdflatex (#1007). Thanks to begemotv2718 for the fix.
15682 + Add link anchors for code blocks with identifiers (#1025).
15684 * Throughout the code, use `isURI` instead of `isAbsoluteURI`.
15685 It allows fragments identifiers.
15689 + A Div element with class "notes" is treated as speaker
15690 notes. Currently beamer goes to `\note{}`, revealjs to
15691 `<aside class="notes">`, and the notes are simply suppressed in
15692 other formats (#925).
15693 + Fixed `. . .` (pause) on HTML slide formats. Closes #1029.
15694 The old version caused a pause to be inserted before the first
15695 material on a slide. This has been fixed.
15696 + Removed data files for s5, slideous, slidy.
15697 Users of s5 and slideous will have to download the needed
15698 files, as has been documented for some time in the README.
15699 By default, slidy code will be sought on the web, as before.
15701 * HTML writer: Insert command to typeset mathjax only in slideous output
15704 * RST writer: Skip spaces after display math. Otherwise we get indentation
15705 problems, and part of the next paragraph may be rendered as part of the
15708 * OpenDocument writer: Fix formatting of strikeout code (#995),
15709 thanks to wilx. don't use `font-face-decls` variable.
15711 * Fixed test suite so it works with cabal sandboxes.
15713## pandoc 1.12.0.2 (2013-09-20)
15715 * Removed an unused dependency (`stringable`) from pandoc.cabal.
15716 This will help packagers, but users should not need to upgrade.
15718## pandoc 1.12.0.1 (2013-09-20)
15720 * Allow `--metadata` to be repeated for the same key to form a list.
15721 This also has the effect that `--bibliography` can be repeated,
15724 * Handle boolean values in `--metadata`. Note that anything not parseable
15725 as a YAML boolean or string is treated as a literal string.
15726 You can get a string value with "yes", or any of the strings interpretable
15727 as booleans, by quoting it:
15729 -M boolvalue=yes -M stringvalue='"yes"'
15731 * LaTeX writer: Don't print references if `--natbib` or `--biblatex`
15734 * DOCX writer: Add `settings.xml` to the zip container. Fixes a bug
15735 in which docx files could not be read by some versions of Word
15736 and LibreOffice (#990).
15738 * Fixed a regression involving slide shows with bibliographies.
15739 The Div container around references messed up the procedure for carving
15740 a document into slides. So we now remove the surrounding Div in
15743 * More informative error message when a filter is not found in path.
15745 * Depend on pandoc-types 1.12.1. This provide `ToJSONFilter`
15746 instances for `Data a => a -> [a]` and `Data a => a -> IO [a]`.
15748 * Don't use unicode_collation in building OSX package:
15749 it adds something like 50MB of dependencies to the package.
15751 * Declare alex and happy as build-tools (#986).
15753## pandoc 1.12 (2013-09-15)
15757 * Much more flexible metadata, including arbitrary fields and structured
15758 values. Metadata can be specified flexibly in pandoc markdown using
15759 YAML metadata blocks, which may occur anywhere in the document:
15762 title: Here is my title.
15764 This is the abstract.
15768 and *inline markup*
15770 tags: [cat, dog, animal]
15773 Metadata fields automatically populate template variables.
15775 * Added `opml` (OPML) as input and output format. The `_note` attribute,
15776 used in OmniOutliner and supported by multimarkdown, is supported.
15777 We treat the contents as markdown blocks under a section header.
15779 * Added `haddock` (Haddock markup) as input format (David Lazar).
15781 * Added `revealjs` output format, for reveal.js HTML 5 slide shows.
15782 (Thanks to Jamie F. Olson for the initial patch.)
15783 Nested vertical stacks are used for hierarchical structure.
15784 Results for more than one level of nesting may be odd.
15786 * Custom writers can now be written in lua.
15788 pandoc -t data/sample.lua
15790 will load the script sample.lua and use it as a custom writer.
15791 (For a sample, do `pandoc --print-default-data-file sample.lua`.)
15792 Note that pandoc embeds a lua interpreter, so lua need not be
15793 installed separately.
15795 * New `--filter/-F` option to make it easier to run "filters"
15796 (Pandoc AST transformations that operate on JSON serializations).
15797 Filters are always passed the name of the output format, so their
15798 behavior can be tailored to it. The repository
15799 <https://github.com/jgm/pandocfilters> contains
15800 a python module for writing pandoc filters in python, with
15801 a number of examples.
15803 * Added `--metadata/-M` option.
15804 This is like `--variable/-V`, but actually adds to metadata, not
15807 * Added `--print-default-data-file` option, which allows printing
15808 of any of pandoc's data files. (For example,
15809 `pandoc --print-default-data-file reference.odt` will print
15812 * Added syntax for "pauses" in slide shows:
15820 * New markdown extensions:
15822 + `ignore_line_breaks`: causes intra-paragraph line breaks to be ignored,
15823 rather than being treated as hard line breaks or spaces. This is useful
15824 for some East Asian languages, where spaces aren't used between words,
15825 but text is separated into lines for readability.
15826 + `yaml_metadata_block`: Parse YAML metadata blocks. (Default.)
15827 + `ascii_identifiers`: This will force `auto_identifiers` to use ASCII
15828 only. (Default for `markdown_github`.) (#807)
15829 + `lists_without_preceding_blankline`: Allow lists to start without
15830 preceding blank space. (Default for `markdown_github`.) (#972)
15832### Behavior changes
15834 * `--toc-level` no longer implies `--toc`.
15835 Reason: EPUB users who don't want a visible TOC may still want
15836 to set the TOC level for in the book navigation.
15838 * `--help` now prints in and out formats in alphabetical order, and
15839 says something about PDF output (#720).
15841 * `--self-contained` now returns less verbose output (telling you
15842 which URLs it is fetching, but not giving the full header). In
15843 addition, there are better error messages when fetching a URL fails.
15845 * Citation support is no longer baked in to core pandoc. Users who
15846 need citations will need to install and use a separate filter
15847 (`--filter pandoc-citeproc`). This filter will take `bibliography`,
15848 `csl`, and `citation-abbreviations` from the metadata, though it
15849 may still be specified on the command line as before.
15851 * A `Cite` element is now created in parsing markdown whether or not
15852 there is a matching reference.
15854 * The `pandoc-citeproc` script will put the bibliography at the
15855 end of the document, as before. However, it will be put inside a `Div`
15856 element with class "references", allowing users some control
15857 over the styling of references. A final header, if any, will
15858 be included in the `Div`.
15860 * The markdown writer will not print a bibliography if the
15861 `citations` extension is enabled. (If the citations are formatted
15862 as markdown citations, it is redundant to have a bibliography,
15863 since one will be generated automatically.)
15865 * Previously we used to store the directory of the first input file,
15866 even if it was local, and used this as a base directory for finding
15867 images in ODT, EPUB, Docx, and PDF. This has been confusing to many
15868 users. So we now look for images relative to the current
15869 working directory, even if the first file argument is in another
15870 directory. Note that this change may break some existing workflows.
15871 If you have been assuming that relative links will be interpreted
15872 relative to the directory of the first file argument, you'll need
15873 to make that the current directory before running pandoc. (#942)
15875 * Better error reporting in some readers, due to changes in `readWith`:
15876 the line in which the error occurred is printed, with a caret pointing
15879 * All slide formats now support incremental slide view for definition lists.
15881 * Parse `\(..\)` and `\[..\]` as math in MediaWiki reader.
15882 Parse `:<math>...</math>` as display math. These notations are used with
15883 the MathJax MediaWiki extension.
15885 * All writers: template variables are set automatically from metadata
15886 fields. However, variables specified on the command line with
15887 `--variable` will completely shadow metadata fields.
15889 * If `--variable` is used to set many variables with the same name,
15892 * Man writer: The `title`, `section`, `header`, and `footer` can now
15893 all be set individually in metadata. The `description` variable has been
15894 removed. Quotes have been added so that spaces are allowed in the
15895 title. If you have a title that begins
15897 COMMAND(1) footer here | header here
15899 pandoc will still parse it into a title, section, header, and
15900 footer. But you can also specify these elements explicitly (#885).
15904 + Added support for YAML metadata blocks, which can come anywhere
15905 in the document (not just at the beginning). A document can contain
15906 multiple YAML metadata blocks.
15907 + HTML span and div tags are parsed as pandoc Span and Div elements.
15911 + Allow simple tables to be printed as grid tables,
15912 if other table options are disabled. This means you can do
15913 `pandoc -t markdown-pipe_tables-simple_tables-multiline_tables`
15914 and all tables will render as grid tables.
15915 + Support YAML title block (render fields in alphabetical order
15916 to make output predictable).
15920 * `Meta` in `Text.Pandoc.Definition` has been changed to allow
15921 structured metadata. (Note: existing code that pattern-matches
15922 on `Meta` will have to be revised.) Metadata can now contain
15923 indefinitely many fields, with content that can be a string,
15924 a Boolean, a list of `Inline` elements, a list of `Block`
15925 elements, or a map or list of these.
15927 * A new generic block container (`Div`) has been added to `Block`,
15928 and a generic inline container (`Span`) has been added to `Inline`.
15929 These can take attributes. They will render in HTML, Textile,
15930 MediaWiki, Org, RST and Markdown (with `markdown_in_html`
15931 extension) as HTML `<div>` and `<span>` elements; in other formats
15932 they will simply pass through their contents. But they can be
15933 targeted by scripts.
15935 * `Format` is now a newtype, not an alias for String.
15936 Equality comparisons are case-insensitive.
15938 * Added `Text.Pandoc.Walk`, which exports hand-written tree-walking
15939 functions that are much faster than the SYB functions from
15940 `Text.Pandoc.Generic`. These functions are now used where possible
15941 in pandoc's code. (`Tests.Walk` verifies that `walk` and `query`
15942 match the generic traversals `bottomUp` and `queryWith`.)
15944 * Added `Text.Pandoc.JSON`, which provides `ToJSON` and `FromJSON`
15945 instances for the basic pandoc types. They use GHC generics and
15946 should be faster than the old JSON serialization using
15947 `Data.Aeson.Generic`.
15949 * Added `Text.Pandoc.Process`, exporting `pipeProcess`.
15950 This is a souped-up version of `readProcessWithErrorcode` that
15951 uses lazy bytestrings instead of strings and allows setting
15952 environment variables. (Used in `Text.Pandoc.PDF`.)
15954 * New module `Text.Pandoc.Readers.OPML`.
15956 * New module `Text.Pandoc.Writers.OPML`.
15958 * New module `Text.Pandoc.Readers.Haddock` (David Lazar).
15959 This is based on Haddock's own lexer/parser.
15961 * New module `Text.Pandoc.Writers.Custom`.
15963 * In `Text.Pandoc.Shared`, `openURL` and `fetchItem` now return an
15964 Either, for better error handling.
15966 * Made `stringify` polymorphic in `Text.Pandoc.Shared`.
15968 * Removed `stripTags` from `Text.Pandoc.XML`.
15970 * `Text.Pandoc.Templates`:
15972 + Simplified `Template` type to a newtype.
15974 + Changed type of `renderTemplate`: it now takes a JSON context
15975 and a compiled template.
15976 + Export `compileTemplate`.
15977 + Export `renderTemplate'` that takes a string instead of a compiled
15979 + Export `varListToJSON`.
15981 * `Text.Pandoc.PDF` exports `makePDF` instead of `tex2pdf`.
15985 + Made `toJsonFilter` an alias for `toJSONFilter` from `Text.Pandoc.JSON`.
15986 + Removed `ToJsonFilter` typeclass. `ToJSONFilter` from
15987 `Text.Pandoc.JSON` should be used instead. (Compiling against
15988 pandoc-types instead of pandoc will also produce smaller executables.)
15989 * Removed the deprecated `jsonFilter` function.
15990 + Added `readJSON`, `writeJSON` to the API (#817).
15992 * `Text.Pandoc.Options`:
15994 + Added `Ext_lists_without_preceding_blankline`,
15995 `Ext_ascii_identifiers`, `Ext_ignore_line_breaks`,
15996 `Ext_yaml_metadataBlock` to `Extension`.
15997 + Changed `writerSourceDirectory` to `writerSourceURL` and changed the
15998 type to a `Maybe`. `writerSourceURL` is set to 'Just url' when the
15999 first command-line argument is an absolute URL. (So, relative links
16000 will be resolved in relation to the first page.) Otherwise, 'Nothing'.
16001 + All bibliography-related fields have been removed from
16002 `ReaderOptions` and `WriterOptions`: `writerBiblioFiles`,
16003 `readerReferences`, `readerCitationStyle`.
16005 * The `Text.Pandoc.Biblio` module has been removed. Users of the
16006 pandoc library who want citation support will need to use
16007 `Text.CSL.Pandoc` from `pandoc-citeproc`.
16011 * In markdown, don't autolink a bare URI that is followed by `</a>`
16014 * `Text.Pandoc.Shared`
16016 + `openURL` now follows redirects (#701), properly handles `data:`
16017 URIs, and prints diagnostic output to stderr rather than stdout.
16018 + `readDefaultDataFile`: normalize the paths. This fixes bugs in
16019 `--self-contained` on pandoc compiled with `embed_data_files` (#833).
16020 + Fixed `readDefaultDataFile` so it works on Windows.
16021 + Better error messages for `readDefaultDataFile`. Instead of
16022 listing the last path tried, which can confuse people who are
16023 using `--self-contained`, so now we just list the data file name.
16024 + URL-escape pipe characters. Even though these are legal, `Network.URI`
16025 doesn't regard them as legal in URLs. So we escape them first (#535).
16027 * Mathjax in HTML slide shows: include explicit "Typeset" call.
16028 This seems to be needed for some formats (e.g. slideous) and won't
16029 hurt in others (#966).
16031 * `Text.Pandoc.PDF`
16033 + On Windows, create temdir in working directory, since the system
16034 temp directory path may contain tildes, which can cause
16035 problems in LaTeX (#777).
16036 + Put temporary output directory in `TEXINPUTS` (see #917).
16037 + `makePDF` tries to download images that are not found locally,
16038 if the first argument is a URL (#917).
16039 + If compiling with `pdflatex` yields an encoding error, offer
16040 the suggestion to use `--latex-engine=xelatex`.
16042 * Produce automatic header identifiers in parsing textile, RST,
16043 and LaTeX, unless `auto_identifiers` extension is disabled (#967).
16045 * `Text.Pandoc.SelfContained`: Strip off fragment, query of relative URL
16046 before treating as a filename. This fixes `--self-contained` when used
16047 with CSS files that include web fonts using the method described here:
16048 <http://paulirish.com/2009/bulletproof-font-face-implementation-syntax/>
16049 (#739). Handle `src` in `embed`, `audio`, `source`, `input` tags.
16051 * `Text.Pandoc.Parsing`: `uri` parser no longer treats punctuation before
16052 percent-encoding, or a `+` character, as final punctuation.
16054 * `Text.Pandoc.ImageSize`: Handle EPS (#903). This change will make
16055 EPS images properly sized on conversion to Word.
16057 * Slidy: Use slidy.js rather than slidy.js.gz.
16058 Reason: some browsers have trouble with the gzipped js file,
16059 at least on the local file system (#795).
16063 + Properly handle blank line at beginning of input (#882).
16064 + Fixed bug in unmatched reference links. The input
16065 `[*infile*] [*outfile*]` was getting improperly parsed:
16066 "infile" was emphasized, but "*outfile*" was literal (#883).
16067 + Allow internal `+` in citation identifiers (#856).
16068 + Allow `.` or `)` after `#` in ATX headers if no `fancy_lists`.
16069 + Do not generate blank title, author, or date metadata elements.
16070 Leave these out entirely if they aren't present.
16071 + Allow backtick code blocks not to be preceded by blank line (#975).
16075 + Correctly handle entities.
16076 + Improved handling of `<pre>` blocks (#927). Remove internal HTML tags
16077 in code blocks, rather than printing them verbatim. Parse attributes
16078 on `<pre>` tag for code blocks.
16080 * HTML reader: Handle non-simple tables (#893). Column widths are read from
16081 `col` tags if present, otherwise divided equally.
16085 + Support alltt environment (#892).
16086 + Support `\textasciitilde`, `\textasciicircum` (#810).
16087 + Treat `\textsl` as emphasized text reader (#850).
16088 + Skip positional options after `\begin{figure}`.
16089 + Support `\v{}` for hacek (#926).
16090 + Don't add spurious ", " to citation suffixes.
16091 This is added when needed in pandoc-citeproc.
16092 + Allow spaces in alignment spec in tables, e.g. `{ l r c }`.
16093 + Improved support for accented characters (thanks to Scott Morrison).
16094 + Parse label after section command and set id (#951).
16098 + Don't insert paragraphs where docutils doesn't.
16099 `rst2html` doesn't add `<p>` tags to list items (even when they are
16100 separated by blank lines) unless there are multiple paragraphs in the
16101 list. This commit changes the RST reader to conform more closely to
16102 what docutils does (#880).
16103 + Improved metadata. Treat initial field list as metadata when
16104 standalone specified. Previously ALL fields "title", "author",
16105 "date" in field lists were treated as metadata, even if not at
16106 the beginning. Use `subtitle` metadata field for subtitle.
16107 + Fixed 'authors' metadata parsing in reST. Semicolons separate
16112 + Allow space before table rows.
16113 + Fixed regression for `<ref>URL</ref>`.
16114 `<` is no longer allowed in URLs, according to the uri parser
16115 in `Text.Pandoc.Parsing`. Added a test case.
16116 + Correctly handle indented preformatted text without preceding
16117 or following blank line.
16118 + Fixed `|` links inside table cells. Improved attribute parsing.
16119 + Skip attributes on table rows. Previously we just crashed if
16120 rows had attributes, now we ignore them.
16121 + Ignore attributes on headers.
16122 + Allow `Image:` for images (#971).
16123 + Parse an image with caption in a paragraph by itself as a figure.
16127 + Don't use ligatures in escaping inline code.
16128 + Fixed footnote numbers in LaTeX/PDF tables. This fixes a bug
16129 wherein notes were numbered incorrectly in tables (#827).
16130 + Always create labels for sections. Previously the labels were only
16131 created when there were links to the section in the document (#871).
16132 + Stop escaping `|` in LaTeX math.
16133 This caused problems with array environments (#891).
16134 + Change `\` to `/` in paths. `/` works even on Windows in LaTeX.
16135 `\` will cause major problems if unescaped.
16136 + Write id for code block to label attribute in LaTeX when listings
16137 is used (thanks to Florian Eitel).
16138 + Scale LaTeX tables so they don't exceed columnwidth.
16139 + Avoid problem with footnotes in unnumbered headers (#940).
16141 * Beamer writer: when creating beamer slides, add `allowframebreaks` option
16142 to the slide if it is one of the header classes. It is recommended
16143 that your bibliography slide have this attribute:
16145 # References {.allowframebreaks}
16147 This causes multiple slides to be created if necessary, depending
16148 on the length of the bibliography.
16150 * ConTeXt writer: Properly handle tables without captions. The old output
16151 only worked in MkII. This should work in MkIV as well (#837).
16153 * MediaWiki writer: Use native mediawiki tables instead of HTML (#720).
16157 + Fixed `--no-highlight` (Alexander Kondratskiy).
16158 + Don't convert to lowercase in email obfuscation (#839).
16159 + Ensure proper escaping in `<title>` and `<meta>` fields.
16163 + Support `--atx-headers` (Max Rydahl Andersen).
16164 + Don't print empty identifier blocks `([[]])` on headers (Max
16169 + Fixing wrong numbered-list indentation in open document format
16170 (Alexander Kondratskiy) (#369).
16171 + `reference.odt`: Added pandoc as "generator" in `meta.xml`.
16172 + Minor changes for ODF 1.2 conformance (#939). We leave the
16173 nonconforming `contextual-spacing` attribute, which is provided by
16174 LibreOffice itself and seems well supported.
16178 + Fixed rendering of display math in lists.
16179 In 1.11 and 1.11.1, display math in lists rendered as a new list
16180 item. Now it always appears centered, just as outside of lists,
16181 and in proper display math style, no matter how far indented the
16182 containing list item is (#784).
16183 + Use `w:br` with `w:type` `textWrapping` for linebreaks.
16184 Previously we used `w:cr` (#873).
16185 + Use Compact style for Plain block elements, to
16186 differentiate between tight and loose lists (#775).
16187 + Ignore most components of `reference.docx`.
16188 We take the `word/styles.xml`, `docProps/app.xml`,
16189 `word/theme/theme1.xml`, and `word/fontTable.xml` from
16190 `reference.docx`, ignoring everything else. This should help
16191 with the corruption problems caused when different versions of
16192 Word resave the reference.docx and reorganize things.
16193 + Made `--no-highlight` work properly.
16197 + Don't add `dc:creator` tags if present in EPUB metadata.
16198 + Add `id="toc-title"` to `h1` in `nav.xhtml` (#799).
16199 + Don't put blank title page in reading sequence.
16200 Set `linear="no"` if no title block. Addresses #797.
16201 + Download webtex images and include as data URLs.
16202 This allows you to use `--webtex` in creating EPUBs.
16203 Math with `--webtex` is automatically made self-contained.
16204 + In `data/epub.css`, removed highlighting styles (which
16205 are no longer needed, since styles are added by the HTML
16206 writer according to `--highlighting-style`). Simplified
16208 + If resource not found, skip it, as in Docx writer (#916).
16212 + Properly handle characters above the 0000-FFFF range.
16213 Uses surrogate pairs. Thanks to Hiromi Ishii for the patch.
16214 + Fixed regression with RTF table of contents.
16215 + Only autolink absolute URIs. This fixes a regression, #830.
16219 + Only autolink absolute URIs. This fixes a regression, #830.
16220 + Don't wrap attributes in fenced code blocks.
16221 + Write full metadata in MMD style title blocks.
16222 + Put multiple authors on separate lines in pandoc titleblock.
16223 Also, don't wrap long author entries, as new lines get treated
16226 * `Text.Pandoc.Templates`:
16228 + Fixed bug retrieving default template for markdown variants.
16229 + Templates can now contain "record lookups" in variables;
16230 for example, `author.institution` will retrieve the `institution`
16231 field of the `author` variable.
16232 + More consistent behavior of `$for$`. When `foo` is not a list,
16233 `$for(foo)$...$endfor$` should behave like $if(foo)$...$endif$.
16234 So if `foo` resolves to "", no output should be produced.
16235 See pandoc-templates#39.
16237 * Citation processing improvements (now part of pandoc-citeproc):
16239 + Fixed `endWithPunct` The new version correctly sees a sentence
16240 ending in '.)' as ending with punctuation. This fixes a bug which
16241 led such sentences to receive an extra period at the end: '.).'.
16242 Thanks to Steve Petersen for reporting.
16243 + Don't interfere with Notes that aren't citation notes.
16244 This fixes a bug in which notes not generated from citations were
16245 being altered (e.g. first letter capitalized) (#898).
16246 + Only capitalize footnote citations when they have a prefix.
16247 + Changes in suffix parsing. A suffix beginning with a digit gets 'p'
16248 inserted before it before passing to citeproc-hs, so that bare numbers
16249 are treated as page numbers by default. A suffix not beginning with
16250 punctuation has a space added at the beginning (rather than a comma and
16251 space, as was done before for not-author-in-text citations).
16252 The result is that `\citep[23]{item1}` in LaTeX will be interpreted
16253 properly, with '23' treated as a locator of type 'page'.
16254 + Many improvements to citation rendering, due to fixes in citeproc-hs
16255 (thanks to Andrea Rossato).
16256 + Warnings are issued for undefined citations, which are rendered
16258 + Fixed hanging behavior when locale files cannot be found.
16260### Template changes
16262 * DocBook: Use DocBook 4.5 doctype.
16264 * Org: '#+TITLE:' is inserted before the title.
16265 Previously the writer did this.
16267 * LaTeX: Changes to make mathfont work with xelatex.
16268 We need the mathspec library, not just fontspec, for this.
16269 We also need to set options for setmathfont (#734).
16271 * LaTeX: Use `tex-ansi` mapping for `monofont`.
16272 This ensures that straight quotes appear as straight, rather than
16273 being treated as curly. See #889.
16275 * Made `\includegraphics` more flexible in LaTeX template.
16276 Now it can be used with options, if needed. Thanks to Bernhard Weichel.
16278 * LaTeX/Beamer: Added `classoption` variable.
16279 This is intended for class options like `oneside`; it may
16280 be repeated with different options. (Thanks to Oliver Matthews.)
16282 * Beamer: Added `fonttheme` variable. (Thanks to Luis Osa.)
16284 * LaTeX: Added `biblio-style` variable (#920).
16286 * DZSlides: title attribute on title section.
16288 * HTML5: add meta tag to allow scaling by user (Erik Evenson)
16290### Under-the-hood improvements
16292 * Markdown reader:Improved strong/emph parsing, using the strategy of
16293 <https://github.com/jgm/Markdown>. The new parsing algorithm requires
16294 no backtracking, and no keeping track of nesting levels. It will give
16295 different results in some edge cases, but these should not affect normal
16298 * Added `Text.Pandoc.Compat.Monoid`.
16299 This allows pandoc to compile with `base` < 4.5, where `Data.Monoid`
16300 doesn't export `<>`. Thanks to Dirk Ullirch for the patch.
16302 * Added `Text.Pandoc.Compat.TagSoupEntity`.
16303 This allows pandoc to compile with `tagsoup` 0.13.x.
16304 Thanks to Dirk Ullrich for the patch.
16306 * Most of `Text.Pandoc.Readers.TeXMath` has been moved to the
16307 `texmath` module (0.6.4). (This allows `pandoc-citeproc` to
16308 handle simple math in bibliography fields.)
16310 * Added `Text.Pandoc.Writers.Shared` for shared functions used
16311 only in writers. `metaToJSON` is used in writers to create a
16312 JSON object for use in the templates from the pandoc metadata
16313 and variables. `getField`, `setField`, and `defField` are
16314 for working with JSON template contexts.
16316 * Added `Text.Pandoc.Asciify` utility module.
16317 This exports functions to create ASCII-only versions of identifiers.
16319 * `Text.Pandoc.Parsing`
16321 + Generalized state type on `readWith` (API change).
16322 + Specialize readWith to `String` input. (API change).
16323 + In `ParserState`, replace `stateTitle`, `stateAuthors`, `stateDate`
16324 with `stateMeta` and `stateMeta'`.
16326 * `Text.Pandoc.UTF8`: use strict bytestrings in reading. The use of lazy
16327 bytestrings seemed to cause problems using pandoc on 64-bit Windows
16330 * Factored out `registerHeader` from markdown reader, added to
16331 `Text.Pandoc.Parsing`.
16333 * Removed `blaze_html_0_5` flag, require `blaze-html` >= 0.5.
16334 Reason: < 0.5 does not provide a monoid instance for Attribute,
16335 which is now needed by the HTML writer (#803).
16337 * Added `http-conduit` flag, which allows fetching https resources.
16338 It also brings in a large number of dependencies (`http-conduit`
16339 and its dependencies), which is why for now it is an optional flag
16342 * Added CONTRIBUTING.md.
16344 * Improved INSTALL instructions.
16346 * `make-windows-installer.bat`: Removed explicit paths for executables.
16348 * `aeson` is now used instead of `json` for JSON.
16350 * Set default stack size to 16M. This is needed for some large
16351 conversions, esp. if pandoc is compiled with 64-bit ghc.
16353 * Various small documentation improvements.
16354 Thanks to achalddave and drothlis for patches.
16356 * Removed comment that chokes recent versions of CPP (#933).
16358 * Removed support for GHC version < 7.2, since pandoc-types now
16359 requires at least GHC 7.2 for GHC generics.
16362## pandoc 1.11.1 (2013-03-17)
16366 + Fixed regression in which parentheses were lost in link URLs.
16367 Added tests. Closes #786.
16368 + Better handling of unmatched double quotes in `--smart` mode.
16369 These occur frequently in fiction, since it is customary not to
16370 close quotes in dialogue if the speaker does not change between
16371 paragraphs. The unmatched quotes now get turned into literal
16372 left double quotes. (No `Quoted` inline is generated, however.)
16373 Closes #99 (again).
16375 * HTML writer: Fixed numbering mismatch between TOC and sections.
16376 `--number-offset` now affects TOC numbering as well
16377 as section numbering, as it should have all along. Closes #789.
16379 * Markdown writer: Reverted 1.11 change that caused citations to be rendered
16380 as markdown citations, even if `--bibliography` was specified, unless
16381 `citation` extension is disabled. Now, formatted citations are always
16382 printed if `--bibliography` was specified. If you want to reformat
16383 markdown keeping pandoc markdown citations intact, don't use
16384 `--bibliography`. Note that citations parsed from LaTeX documents will
16385 be rendered as pandoc markdown citations when `--bibliography` is not
16388 * ODT writer: Fixed regression leading to corrupt ODTs.
16389 This was due to a change in the `Show` instance for
16390 `Text.Pandoc.Pretty.Doc`. Closes #780.
16392 * Fixed spacing bugs involving code block attributes in
16393 RST reader and Markdown writer. Closes #763.
16395 * Windows package: Various improvements due to Fyodor Sheremetyev.
16397 + Automatically set installation path (Program Files or Local App Data).
16398 + Set system PATH environment variable when installing for all users.
16399 + Pandoc can installed for all users using the following command.
16400 `msiexec /i pandoc-1.11.msi ALLUSERS=1`.
16402 * Bumped QuickCheck version bound.
16404## pandoc 1.11 (2013-03-09)
16406 * Added `--number-offset` option. (See README for description.)
16408 * Added `--default-image-extension` option. (See README for description.)
16410 * `--number-sections` behavior change: headers with class `unnumbered`
16411 will not be numbered.
16413 * `--version` now reports the default data directory.
16415 * `Text.Pandoc.Parsing` is no longer exposed. (API change.)
16417 * `Text.Pandoc.Highlighting` is no longer exposed. (API change.)
16419 * `Text.Pandoc.Shared`: Changed type of `Element`. `Sec` now includes
16420 a field for `Attr` rather than just `String`. (API change.)
16422 * Added `markdown_github` as input format. This was an accidental
16425 * Added `readerDefaultImageExtension` field to `ReaderOptions`. (API
16428 * Added `writerNumberOffset` field in `WriterOptions`. (API change.)
16432 + Fixed captions with longtable. Thanks to Joost Kremers.
16433 + Provide `\Oldincludegraphics` as in LaTeX template (Benjamin Bannier).
16437 + Load microtype after fonts. Microtype needs to know
16438 what fonts are being used. Thanks to dfc for the patch.
16439 + Set `secnumdepth` to 5 if `--number-sections` specified.
16440 This yields behavior equivalent to the other writers, numbering
16441 level 4 and 5 headers too. Closes #753.
16445 + Handle `<colgroup>` tag.
16446 + Preserve all header attributes.
16450 + Parse `\hrule` as `HorizontalRule`. Closes #746.
16451 + Parse starred variants of `\section` etc. as headers with
16452 attribute `unnumbered`.
16453 + Read optional attributes in `lstlisting` and `Verbatim` environments.
16454 We convert these to pandoc standard names, e.g. `numberLines`
16455 for `numbers=left`, `startFrom=100` from `firstnumber=100`.
16456 + Handle language attribute for lstlistings.
16457 + Better support for Verbatim and minted environments. Closes #763.
16461 + `-` in an attribute context = `.unnumbered`. The point of this
16462 is to provide a way to specify unnumbered headers in non-English
16464 + Fixed bug parsing key/value attributes. Parsing failed if you
16465 had an unquoted attribute immediately before the final '}'.
16466 + Make backslash escape work in attributes.
16467 + Fix title block parsing. Now if `mmd_title_blocks` is specified,
16468 pandoc will parse a MMD title block if it sees one, even if
16469 `pandoc_title_blocks` is enabled.
16470 + Refactoring: `litChar` now includes entities, so we don't need
16471 to use `fromEntities` e.g. on titles.
16472 + Allow spaces around borders in pipe tables. Closes #772.
16473 + Allow all punctuation in angle-bracket autolinks. Previously
16474 things like `----` were disallowed, because the uri parser
16475 treated them as trailing punctuation. Closes #768.
16476 + Make `implicit_header_references` work properly when
16477 headers are given explicit identifiers.
16478 + Check for tables before line blocks. Otherwise some pipe
16479 tables get treated as line blocks.
16480 + Allow `&` in emails (for entities).
16481 + Properly handle entities in titles and links. A markdown link
16482 `<http://göogle.com>` should be a link to `http://göogle.com`.
16487 + Handle attributes on headers.
16491 + Add `fig:` as title for images with captions.
16492 This is needed for them to be rendered as figures. Closes #766.
16493 + Never emit an empty paragraph. See #761.
16494 + Handle `\caption` for images in figures. Closes #766.
16495 + Parse `\section*`, etc. as unnumbered sections.
16499 + Support header attributes. The attributes go on
16500 the enclosing `section` or `div` tag if `--section-divs` is specified.
16501 + Fixed a regression (only now noticed) in html+lhs output.
16502 Previously the bird tracks were being omitted.
16506 + Omit lists with no items to avoid LaTeX errors.
16507 + Support line numbering with `--listings`.
16508 If `numberLines` class is present, we add `numbers=left`;
16509 if `startFrom` is present, we add `firstnumber=`. (#763)
16513 + Removed `\placecontent`. This produced a duplicate toc,
16514 in conjunction with `\placelist`.
16515 + Use `\title`, `\subject` etc. for headers with
16516 `unnumbered` class.
16520 + Support header attributes.
16524 + Use grid tables when needed, and if enabled. Closes #740.
16525 + Render citations as pandoc-markdown citations.
16526 Previously citations were rendered as citeproc-formatted citations
16527 by default. Now we render them as pandoc citations, e.g. `[@item1]`,
16528 unless the `citations` extension is disabled.
16529 If you still want formatted citations in your markdown output,
16530 use `pandoc -t markdown-citations`.
16534 + Support `:number-lines:` in code blocks.
16538 + Better treatment of display math. Display math inside a
16539 paragraph is now put in a separate paragraph, so it will render
16540 properly (centered and without extra blank lines around it).
16541 Partially addresses #742.
16542 + Content types and document rels xml files are now created from
16543 scratch, rather than being taken over from `reference.docx`.
16544 This fixes problems that arise when you edit the `reference.docx`
16546 + We also now encode mime types for each individual image rather
16547 than using defaults. This should allow us to handle a wider
16548 range of image types (including PDF). Closes #414.
16549 + Changed style names in `reference docx`.
16550 `FootnoteReference` -> `FootnoteRef`, `Hyperlink` -> `Link`.
16551 The old names got changed by Word when the `reference.docx` was
16552 edited. Closes #414.
16556 + Fix section numbering. Previously the numbering restarted from 1
16557 in each chapter (with `--number-sections`), though the numbers in
16558 the table of contents were correct.
16559 + Headers with "unnumbered" attribute are not numbered. (Nor do they
16560 cause an increment in running numbering.) Section numbers now work
16561 properly, even when there is material before the first numbered section.
16562 + Include HTML TOC, even in epub2. The TOC is included in `<spine>`,
16563 but `linear` is set to `no` unless the `--toc` option is specified.
16564 Include `<guide>` element in OPF. This should allow the TOC to
16565 be usable in Kindles when converted with kindlegen. Closes #773.
16567 * `Text.Pandoc.Parsing`: Optimized `oneOfStringsCI`.
16568 This dramatically reduces the speed penalty that comes from enabling the
16569 `autolink_bare_uris` extension. The penalty is still substantial (in one
16570 test, from 0.33s to 0.44s), but nowhere near what it used to be.
16571 The RST reader is also much faster now, as it autodetects URIs.
16573 * `Text.Pandoc.Shared`: `hierarchicalize` will not number section
16574 with class "unnumbered". Unnumbered sections get `[]` for their
16577 * `Text.Pandoc.Pretty`:
16579 + Fixed `chomp` so it works inside `Prefixed` elements.
16580 + Changed `Show` instance so it is better for debugging.
16582 * `Text.Pandoc.ImageSize`: Added `Pdf` to `ImageType`.
16584 * `Text.Pandoc.UTF8`: Strip off BOM if present. Closes #743.
16586 * Windows installer improvements:
16588 + The installer is now signed with a certificate (thanks to
16589 Fyodor Sheremetyev).
16590 + WiX is used instead of InnoSetup. The installer is now a
16592 + The version number is now auto-detected, and need not be
16593 updated separately.
16595 * OSX installer improvements:
16597 + The package and pandoc executable are now signed with a
16598 certificate (thanks to Fyodor Sheremetyev).
16599 + RTF version of license is used.
16600 + Use full path for sysctl in `InstallationCheck` script (jonahbull).
16603 * Converted COPYING to markdown.
16605 * pandoc.cabal: Require latest versions of highlighting-kate,
16606 texmath, citeproc-hs, zip-archive.
16608## pandoc 1.10.1 (2013-01-23)
16610 * Markdown reader: various optimizations, leading to a
16611 significant performance boost.
16613 * RST reader: Allow anonymous form of inline links:
16614 `` `hello <url>`__ `` Closes #724.
16616 * Mediawiki reader: Don't require newlines after tables.
16617 Thanks to jrunningen for the patch. Closes #733.
16619 * Fixed LaTeX macro parsing. Now LaTeX macro definitions are preserved
16620 when output is LaTeX, and applied when it is another format.
16621 Partially addresses #730.
16623 * Markdown and RST readers: Added parser to `block` that skips blank
16624 lines. This fixes a subtle regression involving grid tables with
16625 empty cells. Also added test for grid table with empty cells.
16628 * RST writer: Use `.. code:: language` for code blocks with language.
16631 * DocBook writer: Fixed output for hard line breaks, adding a newline
16632 between `<literallayout>` tags.
16634 * Markdown writer: Use an autolink when link text matches url.
16635 Previously we also checked for a null title, but this
16636 test fails for links produced by citeproc-hs in bibliographies.
16637 So, if the link has a title, it will be lost on conversion
16638 to an autolink, but that seems okay.
16640 * Markdown writer: Set title, author, date variables as before.
16641 These are no longer used in the default template, since we use
16642 titleblock, but we set them anyway for those who use custom templates.
16644 * LaTeX writer: Avoid extra space at start/end of table cell.
16645 Thanks to Nick Bart for the suggestion of using @{}.
16647 * `Text.Pandoc.Parsing`:
16649 + More efficient version of `anyLine`.
16650 + Type of `macro` has changed; the parser now returns `Blocks`
16651 instead of `Block`.
16653 * Relaxed old-time version bound, allowing 1.0.*.
16655 * Removed obsolete `hsmarkdown` script. Those who need `hsmarkdown`
16656 should create a symlink as described in the README.
16658## pandoc 1.10.0.5 (2013-01-23)
16660 * Markdown reader: Try `lhsCodeBlock` before `rawTeXBlock`. Otherwise
16661 `\begin{code}...\end{code}` isn't handled properly in markdown+lhs.
16662 Thanks to Daniel Miot for noticing the bug and suggesting the fix.
16664 * Markdown reader: Fixed bug with headerless grid tables.
16665 The 1.10 code assumed that each table header cell contains exactly one
16666 block. That failed for headerless tables (0) and also for tables with
16667 multiple blocks in a header cell. The code is fixed and tests provided.
16668 Thanks to Andrew Lee for pointing out the bug.
16670 * Markdown reader: Fixed regressions in fenced code blocks. Closes #722.
16672 + Tilde code fences can again take a bare language string
16673 (`~~~ haskell`), not just curly-bracketed attributes
16674 (`~~~ {.haskell}`).
16675 + Backtick code blocks can take the curly-bracketed attributes.
16676 + Backtick code blocks don't *require* a language.
16677 + Consolidated code for the two kinds of fenced code blocks.
16679 * LaTeX template: Use `\urlstyle{same}` to avoid monospace URLs.
16681 * Markdown writer: Use proportional font for email autolinks with
16682 obfuscation. Closes #714.
16684 * Corrected name of `blank_before_blockquote` in README. Closes #718.
16686 * `Text.Pandoc.Shared`: Fixed bug in `uri` parser.
16687 The bug prevented an autolink at the end of a string (e.g.
16688 at the end of a line block line) from counting as a link. Closes #711.
16690 * Use the `hsb2hs` preprocessor instead of TH for embed_data_files.
16691 This should work on Windows, unlike the TH solution with
16694 * Eliminated use of TH in test suite.
16696 * Added `Text.Pandoc.Data` (non-exported) to hold the association
16697 list of embedded data files, if the `embed_data_files` flag is selected.
16698 This isolates the code that needs special treatment with file-embed or
16701 * Changes to `make-windows-installer.bat`.
16703 + Exit batch file if any of the cabal-dev installs fail.
16704 + There's no longer any need to reinstall `highlighting-kate`.
16705 + Don't start with a `cabal update`; leave that to the user.
16706 + Force reinstall of pandoc.
16708 * Fixed EPUB writer so it builds with blaze-html 0.4.x. Thanks to
16711## pandoc 1.10.0.4 (2013-01-20)
16713 * Fixed bug with escaped % in LaTeX reader. Closes #710.
16715## pandoc 1.10.0.3 (2013-01-20)
16717 * Added further missing fb2 tests to cabal file.
16719## pandoc 1.10.0.2 (2013-01-20)
16721 * Added fb2 tests to cabal file's extra-source-files.
16723## pandoc 1.10.0.1 (2013-01-20)
16725 * Bump version bounds on test-framework packages.
16727## pandoc 1.10 (2013-01-19)
16731 * New input formats: `mediawiki` (MediaWiki markup).
16733 * New output formats: `epub3` (EPUB v3 with MathML),
16734 `fb2` (FictionBook2 ebooks).
16736 * New `--toc-depth` option, specifying how many levels of
16737 headers to include in a table of contents.
16739 * New `--epub-chapter-level` option, specifying the header
16740 level at which to divide EPUBs into separate files.
16741 Note that this normally affects only performance, not the
16742 visual presentation of the EPUB in a reader.
16744 * Removed the `--strict` option. Instead of using `--strict`,
16745 one can now use the format name `markdown_strict` for either input
16746 or output. This gives more fine-grained control that `--strict`
16747 did, allowing one to convert from pandoc's markdown to strict
16748 markdown or vice versa.
16750 * It is now possible to enable or disable specific syntax
16751 extensions by appending them (with `+` or `-`) to the writer
16752 or reader name. For example,
16754 pandoc -f markdown-footnotes+hard_line_breaks
16756 disables footnotes and enables treating newlines as hard
16757 line breaks. The literate Haskell extensions are now implemented
16758 this way as well, using either `+lhs` or `+literate_haskell`.
16759 For a list of extension names, see the README under
16760 "Pandoc's Markdown."
16762 * The following aliases have been introduced for specific
16763 combinations of markdown extensions: `markdown_phpextra`,
16764 `markdown_github`, `markdown_mmd`, `markdown_strict`. These aliases
16765 work just like regular reader and writer names, and can be modified
16766 with extension modifiers as described above. (Note that conversion
16767 from one markdown dialect to another does not work perfectly,
16768 because there are differences in markdown parsers besides
16769 just the extensions, and because pandoc's internal document model is
16770 not rich enough to capture all of the extensions.)
16772 * New `--html-q-tags` option. The previous default was to use `<q>`
16773 tags for smart quotes in HTML5. But `<q>` tags are also valid HTML4.
16774 Moreover, they are not a robust way of typesetting quotes, since
16775 some user agents don't support them, and some CSS resets (e.g.
16776 bootstrap) prevent pandoc's quotes CSS from working properly.
16777 We now just insert literal quote characters by default in both
16778 `html` and `html5` output, but this option is provided for
16779 those who still want `<q>` tags.
16781 * The markdown reader now prints warnings (to stderr) about
16782 duplicate link and note references. Closes #375.
16784 * Markdown syntax extensions:
16786 + Added pipe tables. Thanks to François Gannaz for the initial patch.
16787 These conform to PHP Markdown Extra's pipe table syntax. A subset
16788 of org-mode table syntax is also supported, which means that you can
16789 use org-mode's nice table editor to create tables.
16791 + Added support for RST-style line blocks. These are
16792 useful for verse and addresses.
16794 + Attributes can now be specified for headers, using the same
16795 syntax as in code blocks. (However, currently only the
16796 identifier has any effect in most writers.) For example,
16800 See [the header above](#foo).
16802 + Pandoc will now act as if link references have been defined
16803 for all headers without explicit identifiers.
16804 So, you can do this:
16808 Link to [My header].
16809 Another link to [it][My header].
16815 + Command macros now work everywhere, including non-math.
16816 Environment macros still not supported.
16817 + `\input` now works, as well as `\include`. TEXINPUTS is used.
16818 Pandoc looks recursively into included files for more included files.
16820### Behavior changes
16822 * The Markdown reader no longer puts the text of autolinks in a
16823 `Code` inline. This means that autolinks will no longer appear
16824 in a monospace font.
16826 * The character `/` can now appear in markdown citation keys.
16828 * HTML blocks in strict_markdown are no longer required to begin
16829 at the left margin. Technically this is required, according to
16830 the markdown syntax document, but `Markdown.pl` and other markdown
16831 processors are more liberal.
16833 * The `-V` option has been changed so that if there are duplicate
16834 variables, those specified later on the command line take precedence.
16836 * Tight lists now work in LaTeX and ConTeXt output.
16838 * The LaTeX writer no longer relien on the `enumerate` package.
16839 Instead, it uses standard LaTeX commands to change the list numbering
16842 * The LaTeX writer now uses `longtable` instead of `ctable`. This allows
16843 tables to be split over page boundaries.
16845 * The RST writer now uses a line block to render paragraphs containing
16846 linebreaks (which previously weren't supported at all).
16848 * The markdown writer now applies the `--id-prefix` to footnote IDs.
16851 * The plain writer no longer uses backslash-escaped line breaks
16852 (which are not very "plain").
16854 * `Text.Pandoc.UTF8`: Better error message for invalid UTF8.
16855 Read bytestring and use `Text`'s decodeUtf8 instead of using
16856 `System.IO.hGetContents`. This way you get a message saying
16857 "invalid UTF-8 stream" instead of "invalid byte sequence."
16858 You are also told which byte caused the problem.
16860 * Docx, ODT, and EPUB writers now download images specified by a URL
16861 instead of skipping them or raising an error.
16865 + The default CSS now left-aligns headers by default, instead of
16866 centering. This is more consistent with the rest of the writers.
16867 + A proper multi-level table of contents is now used in `toc.ncx`.
16868 There is no longer a subsidiary table of contents at the beginning
16870 + Code highlighting now works by default.
16871 + Section divs are used by default for better semantic markup.
16872 + The title is used instead of "Title Page" in the table of contents.
16873 Otherwise we have a hard-coded English string, which looks
16874 strange in ebooks written in other languages. Closes #572.
16878 + Put mathjax in span with class "math". Closes #562.
16879 + Put citations in a span with class "citation." In HTML5, also include
16880 a `data-cite` attribute with a space-separated list of citation
16883 * `Text.Pandoc.UTF8`: use universalNewlineMode in reading.
16884 This treats both `\r\n` and `\n` as `\n` on input, no matter
16885 what platform we're running on.
16887 * Citation processing is now done in the Markdown and LaTeX
16888 readers, not in `pandoc.hs`. This makes it easier for library users
16891### Template changes
16893 * HTML: Added css to template to preserve spaces in `<code>` tags.
16894 Thanks to Dirk Laurie.
16896 * Beamer: Remove English-centric strings in section pages.
16897 Section pages used to have "Section" and a number as well as the
16898 section title. Now they just have the title. Similarly for part
16899 and subsection. Closes #566.
16901 * LaTeX, ConTeXt: Added papersize variable.
16903 * LaTeX, Beamer templates: Use longtable instead of ctable.
16905 * LaTeX, Beamer templates: Don't require 'float' package for tables.
16906 We don't actually seem to use the '[H]' option.
16908 * Markdown, plain: Fixed titleblock so it is just a single string.
16909 Previously separate title, author, and date variables were used,
16910 but this didn't allow different kinds of title blocks.
16914 + Rationalized templates. Previously there were three different
16915 templates involved in epub production. There is now just one
16916 template, `default.epub` or `default.epub3`. It can now be
16917 overridden using `--template`, just like other templates.
16918 The titlepage is now folded into the default template.
16919 A `titlepage` variable selects it.
16920 + UTF-8, lang tag, meta tags, title element.
16922 * Added scale-to-width feature to beamer template
16926 * `Text.Pandoc.Definition`: Added `Attr` field to `Header`.
16927 Previously header identifiers were autogenerated by the writers.
16928 Now they are added in the readers (either automatically or explicitly).
16930 * `Text.Pandoc.Builder`:
16932 + `Inlines` and `Blocks` are now synonyms for `Many Inline` and
16933 `Many Block`. `Many` is a newtype wrapper around `Seq`, with
16934 custom Monoid instances for `Many Inline` and `Many Block. This
16935 allows `Many` to be made an instance of `Foldable` and `Traversable`.
16936 + The old `Listable` class has been removed.
16937 + The module now exports `isNull`, `toList`, `fromList`.
16938 + The old `Read` and `Show` instances have been removed; derived
16939 instances are now used.
16940 + Added `headerWith`.
16942 * The readers now take a `ReaderOptions` rather than a `ParserState`
16943 as a parameter. Indeed, not all parsers use the `ParserState` type;
16944 some have a custom state. The motivation for this change was to separate
16945 user-specifiable options from the accounting functions of parser state.
16947 * New module `Text.Pandoc.Options`. This includes the `WriterOptions`
16948 formerly in `Text.Pandoc.Shared`, and its associated
16949 data types. It also includes a new type `ReaderOptions`, which
16950 contains many options formerly in `ParserState`, and its associated
16953 + `ParserState.stateParseRaw` -> `ReaderOptions.readerParseRaw`.
16954 + `ParserState.stateColumns` -> `ReaderOptions.readerColumns`.
16955 + `ParserState.stateTabStop` -> `ReaderOptions.readerTabStop`.
16956 + `ParserState.stateOldDashes` -> `ReaderOptions.readerOldDashes`.
16957 + `ParserState.stateLiterateHaskell` -> `ReaderOptions.readerLiterateHaskell`.
16958 + `ParserState.stateCitations` -> `ReaderOptions.readerReferences`.
16959 + `ParserState.stateApplyMacros` -> `ReaderOptions.readerApplyMacros`.
16960 + `ParserState.stateIndentedCodeClasses` ->
16961 `ReaderOptions.readerIndentedCodeClasses`.
16962 + Added `ReaderOptions.readerCitationStyle`.
16964 * `WriterOptions` now includes `writerEpubVersion`, `writerEpubChapterLevel`,
16965 `writerEpubStylesheet`, `writerEpubFonts`, `writerReferenceODT`,
16966 `writerReferenceDocx`, and `writerTOCDepth`. `writerEPUBMetadata` has
16967 been renamed `writerEpubMetadata` for consistency.
16969 * Changed signatures of `writeODT`, `writeDocx`, `writeEPUB`, since they no
16970 longer stylesheet, fonts, reference files as separate parameters.
16972 * Removed `writerLiterateHaskell` from `WriterOptions`, and
16973 `readerLiterateHaskell` from `ReaderOptions`. LHS is now handled
16974 by an extension (`Ext_literate_haskell`).
16976 * Removed deprecated `writerXeTeX`.
16978 * Removed `writerStrict` from `WriterOptions`. Added `writerExtensions`.
16979 Strict is now handled through extensions.
16981 * `Text.Pandoc.Options` exports `pandocExtensions`, `strictExtensions`,
16982 `phpMarkdownExtraExtensions`, `githubMarkdownExtensions`,
16983 and `multimarkdownExtensions`, as well as the `Extensions` type.
16985 * New `Text.Pandoc.Readers.MediaWiki` module, exporting
16988 * New `Text.Pandoc.Writers.FB2` module, exporting `writeFB2`
16989 (thanks to Sergey Astanin).
16993 + Added `getReader`, `getWriter` to `Text.Pandoc`.
16994 + `writers` is now an association list `(String, Writer)`.
16995 A `Writer` can be a `PureStringWriter`, an `IOStringWriter`, or
16996 an `IOByteStringWriter`. ALL writers are now in the 'writers'
16997 list, including the binary writers and FB2 writer. This allows
16998 code in `pandoc.hs` to be simplified.
16999 + Changed type of `readers`, so all readers are in IO.
17000 Users who want pure readers can still get them form the reader
17001 modules; this just affects the function `getReader` that looks up
17002 a reader based on the format name. The point of this change is to
17003 make it possible to print warnings from the parser.
17005 * `Text.Pandoc.Parsing`:
17007 + `Text.Parsec` now exports all Parsec functions used in pandoc code.
17008 No other module directly imports Parsec. This will make it easier
17009 to change the parsing backend in the future, if we want to.
17010 + `Text.Parsec` is used instead of `Text.ParserCombinators.Parsec`.
17011 + Export the type synonym `Parser`.
17012 + Export `widthsFromIndices`, `NoteTable'`, `KeyTable'`, `Key'`, `toKey'`,
17013 `withQuoteContext`, `singleQuoteStart`, `singleQuoteEnd`,
17014 `doubleQuoteStart`, `doubleQuoteEnd`, `ellipses`, `apostrophe`,
17015 `dash`, `nested`, `F(..)`, `askF`, `asksF`, `runF`, `lineBlockLines`.
17016 + `ParserState` is no longer an instance of `Show`.
17017 + Added `stateSubstitutions` and `stateWarnings` to `ParserState`.
17018 + Generalized type of `withQuoteContext`.
17019 + Added `guardEnabled`, `guardDisabled`, `getOption`.
17020 + Removed `failIfStrict`.
17021 + `lookupKeySrc` and `fromKey` are no longer exported.
17023 * `Data.Default` instances are now provided for `ReaderOptions`,
17024 `WriterOptions`, and `ParserState`. `Text.Pandoc` re-exports `def`.
17025 Now you can use `def` (which is re-exported by `Text.Pandoc`) instead
17026 of `defaultWriterOptions` (which is still defined). Closes #546.
17028 * `Text.Pandoc.Shared`:
17030 + Added `safeRead`.
17031 + Renamed `removedLeadingTrailingSpace` to `trim`,
17032 `removeLeadingSpace` to `triml`, and `removeTrailingSpace` to `trimr`.
17033 + Count `\r` as space in `trim` functions.
17034 + Moved `renderTags'` from HTML reader and `Text.Pandoc.SelfContained`
17036 + Removed `failUnlessLHS`.
17037 + Export `compactify'`, formerly in Markdown reader.
17038 + Export `isTightList`.
17039 + Do not export `findDataFile`.
17040 + `readDataFile` now returns a strict ByteString.
17041 + Export `readDataFileUTF8` which returns a String, like the
17042 old `readDataFile`.
17043 + Export `fetchItem` and `openURL`.
17045 * `Text.Pandoc.ImageSize`: Use strict, not lazy bytestrings.
17046 Removed `readImageSize`.
17048 * `Text.Pandoc.UTF8`: Export `encodePath`, `decodePath`,
17049 `decodeArg`, `toString`, `fromString`, `toStringLazy`,
17052 * `Text.Pandoc.UTF8` is now an exposed module.
17054 * `Text.Pandoc.Biblio`:
17056 + csl parameter now a `String` rather than a `FilePath`.
17057 + Changed type of `processBiblio`. It is no longer in the IO monad.
17058 It now takes a `Maybe Style` argument rather than parameters for CSL
17059 and abbrev filenames. (`pandoc.hs` now calls the functions to parse
17060 the style file and add abbreviations.)
17062 * Markdown reader now exports `readMarkdownWithWarnings`.
17064 * `Text.Pandoc.RTF` now exports `writeRTFWithEmbeddedImages` instead of
17069 * Make `--ascii` work properly with `--self-contained`. Closes #568.
17073 + Fixed link parser to avoid exponential slowdowns. Closes #620.
17074 Previously the parser would hang on input like this:
17076 [[[[[[[[[[[[[[[[[[hi
17078 We fixed this by making the link parser parser characters
17079 between balanced brackets (skipping brackets in inline code spans),
17080 then parsing the result as an inline list. One change is that
17082 [hi *there]* bud](/url)
17084 is now no longer parsed as a link. But in this respect pandoc behaved
17085 differently from most other implementations anyway, so that seems okay.
17087 + Look for raw html/latex blocks before tables.
17088 Otherwise the following gets parsed as a table:
17099 + Added support for `:target:` on `.. image::` blocks
17101 + Field list fixes:
17103 - Fixed field lists items with body beginning after a new line
17105 - Allow any char but ':' in names of field lists in RST reader
17107 - Don't allow line breaks in field names.
17108 - Require whitespace after field list field names.
17109 - Don't create empty definition list for metadata field lists.
17110 Previously a field list consisting only of metadata fields (author,
17111 title, date) would be parsed as an empty DefinitionList, which is
17112 not legal in LaTeX and not needed in any format.
17114 + Don't recognize inline-markup starts inside words.
17115 For example, `2*2 = 4*1` should not contain an emphasized
17116 section. Added test case for "Literal symbols". Closes #569.
17117 + Allow dashes as separator in simple tables. Closes #555.
17118 + Added support for `container`, `compound`, `epigraph`,
17119 `rubric`, `highlights`, `pull-quote`.
17120 + Added support for `.. code::`.
17121 + Made directive labels case-insensitive.
17122 + Removed requirement that directives begin at left margin.
17123 This was (correctly) not in earlier releases; docutils doesn't
17124 make the requirement.
17125 + Added support for `replace::` and `unicode::` substitutions.
17126 + Ignore unknown interpreted roles.
17127 + Renamed image parser to `subst`, since it now handles all
17128 substitution references.
17132 + Allow newlines before pipes in table. Closes #654.
17133 + Fixed bug with list items containing line breaks.
17134 Now pandoc correctly handles hard line breaks inside list items.
17135 Previously they broke list parsing.
17136 + Implemented comment blocks.
17137 + Fixed bug affected words ending in hyphen.
17138 + Properly handle links with surrounding brackets.
17139 Square brackets need to be used when the link isn't surrounded by
17140 spaces or punctuation, or when the URL ending may be ambiguous.
17142 + Removed nullBlock. Better to know about parsing problems than
17143 to skip stuff when we get stuck.
17144 + Allow ID attributes on headers.
17145 + Textile reader: Avoid parsing dashes as strikeout.
17146 Previously the input
17153 would be parsed with strikeouts rather than dashes. This fixes
17154 the problem by requiring that a strikeout delimiting - not be
17155 followed by a -. Closes #631.
17156 + Expanded list of `stringBreakers`.
17157 This fixes a bug on input like "(_hello_)" which should
17158 be a parenthesized emphasized "hello".
17159 The new list is taken from the PHP source of textile 2.4.
17160 + Fixed autolinks. Previously the textile reader and writer
17161 incorrectly implemented RST-style autolinks for URLs and email
17162 addresses. This has been fixed. Now an autolink is done this way:
17163 `"$":http://myurl.com`.
17164 + Fixed footnotes bug in textile. This affected notes occurring
17165 before punctuation, e.g. `foo[1].`. Closes #518.
17169 + Better handling of citation commands.
17170 + Better handling of `\noindent`.
17171 + Added a 'try' in rawLaTeXBlock, so we can handle `\begin` without `{`.
17173 + Made `rawLaTeXInline` try to parse block commands as well. This
17174 is usually what we want, given how `rawLaTeXInline` is used in
17175 the markdown and textile readers. If a block-level LaTeX command
17176 is used in the middle of a paragraph (e.g. `\subtitle` inside a title),
17177 we can treat it as raw inline LaTeX.
17178 + Handle `\slash` command. Closes #605.
17179 + Basic `\enquote` support.
17180 + Fixed parsing of paragraphs beginning with a group. Closes #606.
17181 + Use curly quotes for bare straight quotes.
17182 + Support obeylines environment. Closes #604.
17183 + Guard against "begin", "end" in inlineCommand and
17185 + Better error messages for environments. Now it should tell you that
17186 it was looking for \end{env}, instead of giving "unknown parse error."
17190 + Added HTML 5 tags to list of block-level tags.
17191 + HTML reader: Fixed bug in `htmlBalanced`, which
17192 caused hangs in parsing certain markdown input using
17194 + Parse `<q>` as `Quoted DoubleQuote`.
17195 + Handle nested `<q>` tags properly.
17196 + Modified `htmlTag` for fewer false positives.
17197 A tag must start with `<` followed by `!`,`?`, `/`, or a letter.
17198 This makes it more useful in the wikimedia and markdown parsers.
17200 * DocBook reader: Support title in "figure" element. Closes #650.
17202 * MediaWiki writer:
17204 + Remove newline after `<br/>` in translation of `LineBreak`
17205 There's no particular need for a newline (other than making the
17206 generated MediaWiki source look nice to a human), and in fact
17207 sometimes it is incorrect: in particular, inside an enumeration, list
17208 items cannot have embedded newline characters. (Brent Yorgey)
17209 + Use `<code>` not `<tt>` for Code.
17211 * Man writer: Escape `-` as `\-`.
17212 Unescaped `-`'s become hyphens, while `\-`'s are left as ascii minus
17213 signs. That is preferable for use with command-line options. See
17214 <http://lintian.debian.org/tags/hyphen-used-as-minus-sign.html>. Thanks
17215 to Andrea Bolognani for bringing the issue to our attention.
17219 + Improved line block output. Use nonbreaking spaces for
17220 initial indent (otherwise lost in HTML and LaTeX).
17221 Allow multiple paragraphs in a single line block.
17222 Allow soft breaks w continuations in line blocks.
17223 + Properly handle images with no alt text. Closes #678.
17224 + Fixed bug with links with duplicate text. We now (a) use anonymous
17225 links for links with inline URLs, and (b) use an inline link instead
17226 of a reference link if the reference link would require a label that
17227 has already been used for a different link. Closes #511.
17228 + Fixed hyperlinked images. Closes #611. Use `:target:`
17229 field when you have a simple linked image.
17230 + Don't add `:align: center` to figures.
17232 * Texinfo writer: Fixed internal cross-references.
17233 Now we insert anchors after each header, and use `@ref` instead of `@uref`
17234 for links. Commas are now escaped as `@comma{}` only when needed;
17235 previously all commas were escaped. (This change is needed, in part,
17236 because `@ref` commands must be followed by a real comma or period.) Also
17237 insert a blank line in from of `@verbatim` environments.
17241 + Made --id-prefix work in DocBook as well as HTML.
17243 + Don't include empty captions in figures. Closes #581.
17247 + Use `\hspace*` for nonbreaking space after line break,
17248 since `~` spaces after a line break are just ignored.
17250 + Don't escape `_` in URLs or hyperref identifiers.
17251 + Properly escape strings inside `\url{}`. Closes #576.
17252 + Use `[fragile]` only for slides containing code rendered
17253 using listings. Closes #649.
17254 + Escape `|` as `\vert` in LaTeX math. This avoids a clash with
17255 highlighting-kate's macros, which redefine `|` as a short verbatim
17256 delimiter. Thanks to Björn Peemöller for raising this issue.
17257 + Use minipage rather than parbox for block containers in tables.
17258 This allows verbatim code to be included in grid tables.
17260 + Prevent paragraphs containing only linebreaks or spaces.
17264 + Included `highlighting-css` for code spans, too.
17265 Previously it was only included if used in a code block. Closes #653.
17266 + Improved line breaks with `<dd>` tags. We now put a newline between
17267 `</dd>` and `<dd>` when there are multiple definitions.
17268 + Changed mathjax cdn url so it doesn't use https. (This caused
17269 problems when used with `--self-contained`.) See #609.
17273 + `--number-sections` now works properly.
17274 + Don't strip meta and link elements in epub metadata.
17275 Patch from aberrancy. Closes #589.
17276 + Fixed a couple validation bugs.
17277 + Use ch001, ch002, etc. for chapter filenames. This improves sorting
17278 of chapters in some readers, which apparently sort ch2 after ch10.
17281 * ODT writer: properly set title property (Arlo O'Keeffe).
17285 + Fixed bug with nested lists. Previously a list like
17292 would come out with a bullet instead of "2."
17293 Thanks to Russell Allen for reporting the bug.
17294 + Use `w:cr` in `w:r` instead of `w:br` for linebreaks.
17295 This seems to fix a problem viewing pandoc-generated
17296 docx files in LibreOffice.
17297 + Use integer ids for bookmarks. Closes #626.
17298 + Added nsid to abstractNum elements. This helps when merging
17299 word documents with numbered or bulleted lists. Closes #627.
17300 + Use separate footnotes.xml for notes.
17301 This seems to help LibreOffice convert the file, even though
17302 it was valid docx before. Closes #637.
17303 + Use rIdNN identifiers for r:embed in images.
17304 + Avoid reading image files again when we've already processed them.
17305 + Fixed typo in `referenc.docx` that prevented image captions from
17306 working. Thanks to Huashan Chen.
17308 * `Text.Pandoc.Parsing`:
17310 + Fixed bug in `withRaw`, which didn't correctly handle the case
17311 where nothing is parsed.
17312 + Made `emailAddress` parser more correct. Now it is based on RFC 822,
17313 though it still doesn't implement quoted strings in email addresses.
17314 + Revised URI parser. It now allows many more schemes, allows
17315 uppercase URIs, and better handles trailing punctuation and
17316 trailing slashes in bare URIs. Added many tests.
17317 + Simplified and improved singleQuoteStart. This makes `'s'`, `'l'`,
17318 etc. parse properly. Formerly we had some English-centric heuristics,
17319 but they are no longer needed. Closes #698.
17321 * `Text.Pandoc.Pretty`: Added wide punctuation range to `charWidth`.
17322 This fixes bug with Chinese commas in markdown and reST tables, and
17323 a bug that caused combining characters to be dropped.
17325 * `Text.Pandoc.MIME`: Added MIME types for .wof and .eot. Closes #640.
17327 * `Text.Pandoc.Biblio`:
17329 + Run `mvPunc` and `deNote` on metadata too.
17330 This fixed a bug with notes on titles using footnote styles.
17331 + Fixed bug in fetching CSL files from CSL data directory.
17333 * `pandoc.hs`: Give correct value to `writerSourceDirectory` when a URL
17334 is provided. It should be the URL up to the path.
17336 * Fixed/simplified diff output for tests.
17337 Biblio: Make sure mvPunc and deNote run on metadata too.
17338 This fixed a bug with notes on titles using footnote styles.
17340### Under the hood improvements
17342 * We no longer depend on `utf8-string`. Instead we use functions
17343 defined in `Text.Pandoc.UTF8` that use `Data.Text`'s conversions.
17345 * Use `safeRead` instead of using `reads` directly (various modules).
17347 * "Implicit figures" (images alone in a paragraph) are now handled
17348 differently. The markdown reader gives their titles the prefix `fig:`; the
17349 writers look for this before treating the image as a figure. Though this
17350 is a bit of a hack, it has two advantages: (i) implicit figures can be
17351 limited to the markdown reader, and (ii) they can be deactivated by turning
17352 off the `implicit_figures` extension.
17354 * `catch` from `Control.Exception` is now used instead of the
17355 old Preface `catch`.
17357 * `Text.Pandoc.Shared`: Improved algorithm for `normalizeSpaces`
17358 and `oneOfStrings` (which is now non-backtracking).
17360 * `Text.Pandoc.Biblio`: Remove workaround for `toCapital`.
17361 Now citeproc-hs is fixed upstream, so this is no longer needed.
17364 * Textile reader: Improved speed of `hyphenedWords`.
17365 This speeds up the textile reader by about a factor of 4.
17367 * Use `Text.Pandoc.Builder` in RST reader, for more flexibility,
17368 better performance, and automatic normalization.
17370 * Major rewrite of markdown reader:
17372 + Use `Text.Pandoc.Builder` instead of lists. This also
17373 means that everything is normalized automatically.
17374 + Move to a one-pass parsing strategy, returning values in the reader
17375 monad, which are then run (at the end of parsing) against the final
17378 * In HTML writer, we now use `toHtml` instead of pre-escaping.
17379 We work around the problem that blaze-html unnecessarily escapes `'`
17380 by pre-escaping just the `'` characters, instead of the whole string.
17381 If blaze-html later stops escaping `'` characters, we can simplify
17382 `strToHtml` to `toHtml`. Closes #629.
17384 * Moved code for embedding images in RTFs from `pandoc.hs` to the
17385 RTF writer (which now exports `writeRTFWithEmbeddedImages`).
17387 * Moved citation processing from `pandoc.hs` into the readers.
17388 This makes things more convenient for library users.
17390 * The man pages are now built by an executable `make-pandoc-man-pages`,
17391 which has its own stanza in the cabal file so that dependencies can be
17392 handled by Cabal. Special treatment in `Setup.hs` ensures that this
17393 executable never gets installed; it is only used to create the man pages.
17395 * The cabal file has been modified so that the pandoc library is used
17396 in building the pandoc executable. (This required moving `pandoc.hs`
17397 from `src` to `.`.) This cuts compile time in half.
17399 * The `executable` and `library` flags have been removed.
17401 * `-threaded` has been removed from ghc-options.
17403 * Version bounds of dependencies have been raised, and the
17404 `blaze_html_0_5` flag now defaults to True. Pandoc now compiles on
17407 * We now require base >= 4.2.
17409 * Integrated the benchmark program into cabal. One can now do:
17411 cabal configure --enable-benchmarks && cabal build
17412 cabal bench --benchmark-option='markdown' --benchmark-option='-s 20'
17414 The benchmark now uses README + testsuite, so benchmark results
17415 from older versions aren't comparable.
17417 * Integrated test suite with cabal.
17418 To run tests, configure with `--enable-tests`, then `cabal test`.
17419 You can specify particular tests using `--test-options='-t markdown'`.
17420 No output is shown unless tests fail. The Haskell test modules
17421 have been moved from `src/` to `tests/`.
17423 * Moved all data files and templates to the `data/` subdirectory.
17425 * Added an `embed_data_files` cabal flag. This causes all
17426 data files to be embedded in the binary, so that the binary
17427 is self-sufficient and can be relocated anywhere, copied on
17428 a USB key, etc. The Windows installer now uses this.
17429 (Since we no longer have the option to build the executable
17430 without the library, this is the only way to get a relocatable
17431 binary on Windows.)
17433 * Removed pcre3.dll from windows package.
17434 It isn't needed unless highlighting-kate is compiled with the
17435 `pcre-light` flag. By default, regex-prce-builtin is used.
17438## pandoc 1.9.4.5 (2012-10-21)
17440 * Raised version bounds on network, base64-bytestring, json,
17441 and template-haskell.
17443## pandoc 1.9.4.4 (2012-10-20)
17445 * Removed `tests` flag and made test suite into a proper cabal
17446 test suite, which can now be enabled using `--enable-tests`
17447 and run with `cabal test`.
17449 * Moved man page creation out of `Setup.hs` and into an
17450 executable built by Cabal, but never installed. This
17451 allows dependencies to be specified, and solves a problem
17452 with 1.9.4.3, which could only be installed if `data-default`
17453 had already been installed.
17455 * Updated `lhs-latex.tex` test for latest highlighting-kate
17456 representation of backticks.
17458## pandoc 1.9.4.3 (2012-10-20)
17460 * Removed `-threaded` from default compile flags.
17462 * Modified modules to compile with GHC 7.6 and latest version of time
17465## pandoc 1.9.4.2 (2012-06-29)
17467 * Don't encode/decode file paths if base >= 4.4.
17468 Prior to base 4.4, filepaths and command line arguments were treated
17469 as unencoded lists of bytes, not unicode strings, so we had to work
17470 around that by encoding and decoding them. This commit adds CPP
17471 checks for the base version that intelligibly enable encoding/decoding
17472 when needed. Fixes a bug with multilingual filenames when pandoc was
17473 compiled with ghc 7.4 (#540).
17475 * Don't generate an empty H1 after hrule slide breaks.
17476 We now use a slide-level header with contents `[Str "\0"]` to mark
17477 an hrule break. This avoids creation of an empty H1 in these
17478 contexts. Closes #484.
17480 * Docbook reader: Added support for "bold" emphasis. Thanks to mb21.
17482 * In make_osx_package.sh, ensure citeproc-hs is built with the
17483 embed_data_files flag.
17485 * MediaWiki writer: Avoid extra blank lines after sublists (Gavin Beatty).
17487 * ConTeXt writer: Don't escape `&`, `^`, `<`, `>`, `_`,
17488 simplified escapes for `}` and `{` to `\{` and `\}` (Aditya Mahajan).
17490 * Fixed handling of absolute URLs in CSS imports with `--self-contained`.
17493 * Added webm to mime types. Closes #543.
17495 * Added some missing exports and tests to the cabal file
17496 (Alexander V Vershilov).
17498 * Compile with `-rtsopts` and `-threaded` by default.
17500## pandoc 1.9.4.1 (2012-06-08)
17502 * Markdown reader: Added `cf.` and `cp.` to list of likely abbreviations.
17504 * LaTeX template: Added `linkcolor`, `urlcolor` and `links-as-notes`
17505 variables. Make TOC links black.
17507 * LaTeX template improvements.
17509 + Don't print date unless one is given explicitly in the document.
17510 + Simplified templates.
17511 + Use fontenc [T1] by default, and lmodern.
17512 + Use microtype if available.
17516 + Add comma to beginning of bare suffix, e.g. `@item1 [50]`.
17517 Motivation: `@item1 [50]` should be as close as possible to
17519 + Added workaround for a bug in citeproc-hs 0.3.4 that causes footnotes
17520 beginning with a citation to be empty. Closes #531.
17522 * Fixed documentation on mixed lists. Closes #533.
17524## pandoc 1.9.4 (2012-06-03)
17526 * Simplified `Text.Pandoc.Biblio` and fixed bugs with citations inside
17527 footnotes and captions. We now handle note citations by inserting
17528 footnotes during initial citation processing, and doing a separate
17529 pass later to remove notes inside notes.
17531 * Added 'zenburn' highlight style from highlighting-kate.
17533 * Added Slideous writer. Slideous is an HTML + javascript slide show
17534 format, similar to Slidy, but works with IE 7. (Jonas Smedegaard)
17538 + Ensure we don't have extra blank lines at ends of cells.
17539 This can cause LaTeX errors, as they are interpreted as new paragraphs.
17540 + More consistent interblock spacing.
17541 + Require highlighting-kate >= 0.5.1, for proper highlighted inline
17542 code in LaTeX. Closes #527.
17543 + Ensure that a Verbatim at the end of a footnote is followed by
17544 a newline. (Fixes a regression in the previous version.)
17545 + In default template, use black for internal links and TOC.
17546 Added commented-out code to use footnotes for links, as would
17547 be suitable in print output.
17549 * Beamer writer: When `--incremental` is used, lists inside
17550 a block quote should appear all at once. (This makes Beamer
17551 output consistent with the HTML slide show formats.)
17555 + Escape `%` as `\letterpercent{}` not `\letterpercent `,
17556 to avoid gobbling spaces after the `%` sign.
17557 + Ensure space after `\stopformula`.
17561 + Use `:` form instead of `~` in definition lists, for better
17562 compatibility with other markdown implementations.
17563 + Don't wrap the term, because it breaks definition lists.
17564 + Use a nonzero space to prevent false recognition
17565 of list marker in ordered lists. Closes #516.
17567 * Org writer: Add space before language name. Closes #523.
17569 * Docx writer: Simplified bullet characters so they work properly
17570 with Word 2007. Closes #520.
17572 * LaTeX reader: Support `\centerline`.
17574 * RST reader: handle figures. Closes #522.
17576 * Textile reader: fix for `<notextile>` and `==`. Closes #517.
17579## pandoc 1.9.3 (2012-05-12)
17581 * Added docbook reader (with contributions from Mauro Bieg).
17583 * Fixed bug in `fromEntities`. The previous version would turn
17584 `hi & low you know;` into `hi &`.
17588 + Don't skip nonbreaking spaces.
17589 Previously a paragraph containing just ` ` would be rendered
17590 as an empty paragraph. Thanks to Paul Vorbach for pointing out the bug.
17591 + Support `<col>` and `<caption>` in tables. Closes #486.
17595 + Don't recognize references inside delimited code blocks.
17596 + Allow list items to begin with lists.
17600 + Handle `\bgroup`, `\egroup`, `\begingroup`, `\endgroup`.
17601 + Control sequences can't be followed by a letter.
17602 This fixes a bug where `\begingroup` was parsed as `\begin`
17603 followed by `group`.
17604 + Parse 'dimension' arguments to unknown commands. e.g. `\parindent0pt`
17605 + Make `\label` and `\ref` sensitive to `--parse-raw`.
17606 If `--parse-raw` is selected, these will be parsed as raw latex
17607 inlines, rather than bracketed text.
17608 + Don't crash on unknown block commands (like `\vspace{10pt}`)
17609 inside `\author`; just skip them. Closes #505.
17613 + Implemented literal escapes with `==` and `<notextile>`. Closes #473.
17614 + Added support for LaTeX blocks and inlines (Paul Rivier).
17615 + Better conformance to RedCloth inline parsing (Paul Rivier).
17616 + Parse '+text+' as emphasized (should be underlined, but this
17617 is better than leaving literal plus characters in the output.
17619 * Docx writer: Fixed multi-paragraph list items. Previously they each
17620 got a list marker. Closes #457.
17624 + Added `--no-tex-ligatures` option to avoid replacing
17625 quotation marks and dashes with TeX ligatures.
17626 + Use `fixltx2e` package to provide `\textsubscript`.
17627 + Improve spacing around LaTeX block environments:
17628 quote, verbatim, itemize, description, enumerate.
17630 + Use blue instead of pink for URL links in latex/pdf output.
17632 * ConTeXt writer: Fixed escaping of `%`.
17633 In text, `%` needs to be escaped as `\letterpercent`, not `\%`
17634 Inside URLs, `%` needs to be escaped as `\%`
17635 Thanks to jmarca and adityam for the fix. Closes #492.
17637 * Texinfo writer: Escape special characters in node titles.
17638 This fixes a problem pointed out by Joost Kremers. Pandoc used
17639 to escape an '@' in a chapter title, but not in the corresponding
17640 node title, leading to invalid texinfo.
17642 * Fixed document encoding in texinfo template.
17643 Resolves Debian Bug #667816.
17647 + Don't force delimited code blocks to be flush left.
17648 Fixes bug with delimited code blocks inside lists etc.
17649 + Escape `<` and `$`.
17651 * LaTeX writer: Use `\hyperref[ident]{text}` for internal links.
17652 Previously we used `\href{\#ident}{text}`, which didn't work on
17653 all systems. Thanks to Dirk Laurie.
17655 * RST writer: Don't wrap link references. Closes #487.
17657 * Updated to use latest versions of blaze-html, mtl.
17660## pandoc 1.9.2 (2012-04-05)
17664 + Made `lstlisting` work as a proper verbatim environment.
17665 + Fixed bug parsing LaTeX tables with one column.
17669 + Use `{}` around `ctable` caption, so that formatting can be used.
17670 + Don't require eurosym package unless document has a €.
17672 * LaTeX template: Added variables for `geometry`, `romanfont`,
17673 `sansfont`, `mathfont`, `mainfont` so users can more easily
17678 + Run latex engine at least two times, to ensure
17679 that PDFs will have hyperlinked bookmarks.
17680 + Added PDF metadata (title,author) in LaTeX standalone + PDF output.
17682 * Texinfo writer: retain directories in image paths. (Peter Wang)
17684 * RST writer: Better handling of inline formatting, in accord
17685 with docutils' "inline markup recognition rules" (though we don't
17686 implement the unicode rules fully). Now `hi*there*hi` gets
17687 rendered properly as `hi\ *there*\ hi`, and unnecessary
17688 `\ ` are avoided around `:math:`, `:sub:`, `:sup:`.
17692 + Parse `\ ` as null, not escaped space.
17693 + Allow `` :math:`...` `` even when not followed by blank
17694 or `\`. This does not implement the complex rule docutils follows,
17695 but it should be good enough for most purposes.
17696 + Add support for the rST default-role directive. (Greg Maslov)
17698 * Text.Pandoc.Parsing: Added `stateRstDefaultRole` field to `ParserState`.
17701 * Markdown reader: Properly handle citations nested in other inline
17704 * Markdown writer: don't replace empty alt in image with "image".
17706 * DZSlides: Updated template.html and styles in default template.
17707 Removed bizarre CSS for `q` in dzslides template.
17709 * Avoid repeated `id` attribute in section and header in HTML slides.
17711 * README improvements: new instructions on internal links,
17712 removed misleading note on reST math.
17716 + Fixed Windows installer so that dzslides works.
17717 + Removed stripansi.sh.
17718 + Added .travis.yml for Travis continuous integration support..
17719 + Fixed upper bound for zlib (Sergei Trofimovich).
17720 + Fixed upper bound for test-framework.
17721 + Updated haddocks for haddock-2.10 (Sergei Trofimovich).
17724## pandoc 1.9.1.2 (2012-03-09)
17726 * Added `beamer+lhs` as output format.
17728 * Don't escape `<` in `<style>` tags with `--self-contained`.
17729 This fixes a bug which prevented highlighting from working
17730 when using `--self-contained`.
17732 * PDF: run latex engine three times if `--toc` specified.
17733 This fixes page numbers in the table of contents.
17735 * Docx writer: Added TableNormal style to tables.
17737 * LaTeX math environment fixes. `aligned` is now used instead of
17738 the nonexistent `aligned*`. `multline` instead of the nonexistent
17741 * LaTeX writer: Use `\textasciitilde` for literal `~`.
17743 * HTML writer: Don't escape contents of EQ tags with --gladtex.
17744 This fixes a regression from 1.8.
17746 * Use `<q>` tags for Quoted items for HTML5 output.
17747 The quote style can be changed by modifying the template
17748 or including a css file. A default quote style is included.
17750 * LaTeX reader: Fixed accents (\~{a}, `\c{c}`).
17751 Correctly handle \^{}. Support "minted" as a LaTeX verbatim block.
17753 * Updated LaTeX template for better language support.
17754 Use `polyglossia` instead of `babel` with xetex.
17755 Set `lang` as documentclass option.
17756 `\setmainlanguage` will use the last of a comma-separated
17757 list of languages. Thanks to François Gannaz.
17759 * Fixed default LaTeX template so `\euro` and `€` work. The
17760 `eurosym` package is needed if you are using pdflatex.
17762 * Fixed escaping of period in man writer (thanks to Michael Thompson).
17764 * Fixed list label positions in beamer.
17766 * Set `mainlang` variable in context writer.
17767 This parallels behavior of latex writer. `mainlang` is the last
17768 of a comma-separated list of languages in lang.
17770 * EPUB language metadat: convert e.g. `en_US` from locale to `en-US`.
17772 * Changed `-V` so that you can specify a key without a value.
17773 Such keys get the value `true`.
17775 * Fixed permissions on installed man pages - thanks Magnus Therning.
17777 * Windows installer: require XP or higher. The installer is
17778 now compiled on a Windows 7 machine, which fixes a problem
17779 using citation functions on Windows 7.
17781 * OSX package: Check for 64-bit Intel CPU before installing.
17784## pandoc 1.9.1.1 (2012-02-11)
17786 * Better handling of raw latex environments in markdown. Now
17792 turns into a raw latex block as expected.
17794 * Improvements to LaTeX reader:
17796 + Skip options after block commands.
17797 + Correctly handle `{\\}` in braced.
17798 + Added a needed 'try'.
17799 + Citations: add `, ` to suffix if it doesn't start with space or
17800 punctuation. Otherwise we get no space between the year and the
17801 suffix in author-date styles.
17803 * Added two needed data files for S5. This fixes a problem with
17804 `pandoc -t s5 --self-contained`. Also removed `slides.min.js`,
17805 which was no longer being used.
17807 * Fixed some minor problems in `reference.docx`:
17808 name on "Date" style, `xCs` instead of `xIs`.
17810 * Fixed a problem creating docx files using a reference docx
17811 modified using Word. The problem seems to be that Word
17812 modifies `_rels/.rels`, changing the Type of the Relationship to
17813 `docProps/core.xml`. Pandoc now changes this back to the correct
17814 value if it has been altered, fixing the problem.
17816 * Fixed html5 template so it works properly with highlighting.
17818## pandoc 1.9.1 (2012-02-09)
17822 + Fixed regression in 1.9; properly handle escaped $ in latex math.
17823 + Put LaTeX verse environments in blockquotes.
17827 + Limit nesting of strong/emph. This avoids exponential lookahead
17828 in parasitic cases, like `a**a*a**a*a**a*a**a*a**a*a**a*a**a*a**`.
17829 + Improved attributes syntax (in code blocks/spans): (1)
17830 Attributes can contain line breaks. (2) Values in key-value
17831 attributes can be surrounded by either double or single quotes, or
17832 left unquoted if they contain no spaces.
17834 * Headers no longer wrap in markdown or RST writers.
17836 * Added `stateMaxNestingLevel` to `ParserState`.
17837 We set this to 6, so you can still have `Emph` inside `Emph`,
17838 just not indefinitely.
17840 * More efficient implementation of `nowrap` in `Text.Pandoc.Pretty`.
17842 * `Text.Pandoc.PDF`: Only run latex twice if `\tableofcontents`
17845 * Require highlighting-kate >= 0.5.0.2, texmath >= 0.6.0.2.
17847## pandoc 1.9.0.5 (2012-02-06)
17849 * Changed cabal file so that build-depends for the test program
17850 are not required unless the tests flag is used.
17852 * LaTeX writer: insert `{}` between adjacent hyphens so they don't
17853 form ligatures (dashes) in code spans.
17855## pandoc 1.9.0.4 (2012-02-06)
17857 * Raised version bound on test-framework to avoid problems
17858 compiling tests on GHC 7.4.1.
17860 * LaTeX reader: Use raw LaTeX as fallback inline text for Cites,
17861 so citations don't just disappear unless you process with
17862 citeproc. Ignore `\bibliographystyle`, `\nocite`.
17864 * Simplified tex2pdf; it will always run latex twice to
17865 resolve table of contents and hyperrefs.
17867## pandoc 1.9.0.3 (2012-02-06)
17869 * Require Cabal >= 1.10.
17871 * Tweaked cabal file to meet Cabal 1.10 requirements.
17873## pandoc 1.9.0.2 (2012-02-05)
17875 * Allow build with json 0.4 or 0.5. Otherwise we can't build with
17878## pandoc 1.9 (2012-02-05)
17882 * Added a Microsoft Word `docx` writer. The writer includes support
17883 for highlighted code and for math (which is converted from TeX to OMML,
17884 Office's native math markup language, using texmath's new OMML module).
17885 A new option `--reference-docx` allows the user to customize the
17888 * Added an `asciidoc` writer (<http://www.methods.co.nz/asciidoc/>).
17890 * Better support for slide shows:
17892 + Added a `dzslides` writer. DZSlides is a lightweight HTML5/javascript
17893 slide show format due to Paul Rouget (<http://paulrouget.com/dzslides/>).
17895 + Added a LaTeX `beamer` writer. Beamer is a LaTeX package for creating
17896 slide presentations.
17898 + New, flexible rules for dividing documents into sections and slides
17899 (see the "Structuring the slide show" in the User's Guide). These
17900 are backward-compatible with the old rules, but they allow slide
17901 shows to be organized into sections and subsections containing
17904 + A new `--slide-level` option allows users to override defaults
17905 and select a slide level below the first header level with content.
17907 * A new `--self-contained` option produces HTML output that does not
17908 depend on an internet connection or the presence of any external
17909 files. Linked images, CSS, and javascript is downloaded (or fetched
17910 locally) and encoded in `data:` URIs. This is useful for making portable
17911 `HTML slide shows. The --offline` option has been deprecated and is now
17912 `treated as a synonym or --self-contained`.
17914 * Support for PDF output:
17916 + Removed the old `markdown2pdf`.
17917 + `pandoc` can now create PDFs (assuming you have latex and a set of
17918 appropriate packages installed): just specify an output file with the
17920 + A new option `--latex-engine` allows you to specify `pdflatex`,
17921 `xelatex`, or `lualatex` as the processor.
17923 * Highlighting changes:
17925 + Syntax highlighting is now a standard feature; the `highlighting`
17926 flag is no longer needed when compiling.
17927 + A new `--no-highlight` option allows highlighting to be disabled.
17928 + Highlighting now works in `docx`, `latex`, and `epub`, as well as
17929 `html`, `html5`, `dzslides`, `s5`, and `slidy`.
17930 + A new `--highlight-style` option selects between various highlighting
17933 * Internal links to sections now work in ConTeXt and LaTeX as well as HTML.
17935 * LaTeX `\include` and `\usepackage` commands are now processed,
17936 provided the files are in the working directory.
17938 * EPUB improvements:
17940 + Internal and external links now work in EPUB.
17941 + Raw HTML is allowed.
17942 + New `--epub-embed-font` option.
17943 + Customizable templates for EPUB pages offer more control over
17944 formatting: `epub-page.html`, `epub-coverimage.html`,
17945 `epub-titlepage.html`.
17947 * `--mathml` now works with DocBook.
17949 * Added support for math in RST reader and writer. Inline math uses the
17950 `` :math:`...` `` construct. Display math uses
17954 or if the math is multiline,
17960 These constructions are now supported now by `rst2latex.py`.
17962 * Github syntax for fenced code blocks is supported in pandoc's
17963 markdown. You can now write
17975 * Easier scripting: a new `toJsonFilter` function makes it easier to
17976 write Haskell scripts to manipulate the Pandoc AST.
17977 See [Scripting with pandoc](scripting.html#pandoc-1.9-changes).
17979### Behavior changes
17981 * Fixed parsing of consecutive lists in markdown.
17982 Pandoc previously behaved like Markdown.pl for consecutive
17983 lists of different styles. Thus, the following would be parsed
17984 as a single ordered list, rather than an ordered list followed
17985 by an unordered list:
17993 This change makes pandoc behave more sensibly, parsing this as
17994 two lists. Any change in list type (ordered/unordered) or in
17995 list number style will trigger a new list. Thus, the following
17996 will also be parsed as two lists:
18004 Since we regard this as a bug in Markdown.pl, and not something
18005 anyone would ever rely on, we do not preserve the old behavior
18006 even when `--strict` is selected.
18008 * Dashes work differently with `--smart`: `---` is always em-dash,
18009 and `--` is always en-dash. Pandoc no longer tries to guess when
18010 `-` should be en-dash. *Note:* This may change how existing documents
18011 look when processed with pandoc. A new option, `--old-dashes`,
18012 is provided for legacy documents.
18014 * The markdown writer now uses setext headers for levels 1-2.
18015 The old behavior (ATX headers for all levels) can be restored
18016 using the new `--atx-headers` option.
18018 * Links are now allowed in markdown image captions. They are also
18019 allowed in links, but will appear there as regular text. So,
18021 [link with [link](/url)](/url)
18025 <p><a href="/url">link with link</a></p>
18027 * Improved handling of citations using `citeproc-hs-0.3.4`.
18028 Added `--citation-abbreviations` option.
18030 * Citation keys can no longer end with a punctuation character.
18031 This means that `@item1.` will be parsed as a citation with key
18032 'item1', followed by a period, instead of a citation with key
18033 'item1.', as was the case previously.
18035 * In HTML output, citations are now put in a span with class `citation`.
18037 * The markdown reader now recognizes DocBook block and inline tags.
18038 It was always possible to include raw DocBook tags in a markdown
18039 document, but now pandoc will be able to distinguish block from
18040 inline tags and behave accordingly. Thus, for example,
18046 will not be wrapped in `<para>` tags.
18048 * The LaTeX parser has been completely rewritten; it is now much more
18049 accurate, robust, and extensible. However, there are two important
18050 changes in how it treats unknown LaTeX. (1) Previously, unknown
18051 environments became BlockQuote elements; now, they are treated
18052 as "transparent", so `\begin{unknown}xyz\end{unknown}` is the
18053 same as `xyz`. (2) Previously, arguments of unknown commands
18054 were passed through with their braces; now the braces are stripped
18057 * `--smart` is no longer selected automatically with `man` output.
18059 * The deprecated `--xetex` option has been removed.
18061 * The `--html5`/`-5` option has been deprecated. Use `-t html5`
18062 instead. `html5` and `html5+lhs` are now separate output formats.
18064 * Single quotes are no longer escaped in HTML output. They do not
18065 need to be escaped outside of attributes.
18067 * Pandoc will no longer transform leading newlines in code
18068 blocks to `<br/>` tags.
18070 * The ODT writer now sizes images appropriately, using the image
18071 size and DPI information embedded in the image.
18073 * `--standalone` is once again implicitly for a non-text output format
18074 (ODT, EPUB). You can again do `pandoc test.txt -o test.odt`
18075 and get a standalone ODT file.
18077 * The Docbook writer now uses `<sect1>`, `<sect2>`, etc. instead of
18080 * The HTML writer now uses `<del>` for strikeout.
18082 * In HTML output with `--section-divs`, the classes `section` and
18083 `level[1,2,..6]` are put on the `div` tags so they can be styled.
18084 In HTML 5 output with `--section-divs`, the classes
18085 `level[1,2,...6]` are put on `section` tags.
18087 * EPUB writer changes:
18089 + The `lang` variable now sets the language
18090 in the metadata (if it is not set, we default to the locale).
18091 + EPUB: UTF-8 is used rather than decimal entities.
18093 * Added `titleslide` class to title slide in S5 template.
18095 * In HTML, EPUB, and docx metadata, the date is normalized into
18096 YYYY-MM-DD format if possible. (This is required for validation.)
18098 * Attributes in highlighted code blocks are now preserved in HTML.
18099 The container element will have the classes, id, and key-value attributes
18100 you specified in the delimited code block. Previously these were stripped
18103 * The reference backlink in the HTML writer no longer has a special
18104 `footnoteBacklink` class.
18106 * The HTML template has been split into `html` and `html5` templates.
18108 * Author and date are treated more consistently in HTML templates.
18109 Authors are now `<h2>`, date `<h3>`.
18111 * URLs are hyphenated in the ConTeXt writer (B. Scott Michel).
18113 * In `Text.Pandoc.Builder`, `+++` has been replaced by `<>`.
18117 * Better support for combining characters and East Asian wide characters
18118 in markdown and reST.
18120 * Better handling of single quotes with `--smart`.
18121 Previously `D'oh l'*aide*` would be parsed with left and right single
18122 quotes instead of apostrophes. This kind of error is now fixed.
18124 * Highlighting: Use `reads` instead of `read` for better error handling.
18125 Fixes crash on `startNum="abc"`.
18127 * Added blank comment after directives in rst template.
18129 * Unescape entities in citation `refId`. The `refId`s coming
18130 from citeproc contain XML numeric entities, and these don't match with the
18131 citation keys parsed by pandoc. Solution is to unescape them.
18133 * HTML reader: Fixed bug parsing tables with both thead and tbody.
18137 + Better handling of escapes in link URLs and titles.
18138 + Fixed backslash escapes in reference links.
18139 + Fixed bug in table/hrule parsing, by checking that the top
18140 line of a table is not followed by a blank line. This bug caused
18141 slowdowns on some files with hrules and tables, as pandoc tried to
18142 interpret the hrules as the tops of multiline tables.
18143 + Fixed bug in code block attribute parser. Previously the ID attribute
18144 got lost if it didn't come first. Now attributes can come in any order.
18146 * RST reader: allow footnotes followed by newline without space characters.
18150 + Ignore empty groups {}, { }.
18151 + LaTeX reader: Handle `\@`.
18152 + LaTeX reader: Don't crash on commands like `\itemsep`.
18153 + LaTeX reader: Better handling of letter environments.
18155 * RST writer: Fixed bug involving empty table cells. isSimple was being
18156 calculated in a way that assumed there were no non-empty cells.
18160 + Made `--toc` work even without `--number-sections`.
18161 + Escape # in link URLs.
18162 + Use buffering for footnotes containing code blocks.
18163 + Changed 'descr' to 'description', fixed alignment.
18167 + Escape euro character.
18168 + Don't escape `~` inside `\href{...}`.
18169 + Escape `#` in href URLs.
18170 + Improved detection of book classes. We now check the
18171 `documentclass` variable, and if that is not set, we look through
18172 the template itself. Also, we have added the KOMA classes scrreprt
18173 and scrbook. You can now make a book using
18174 `pandoc -V documentclass:book mybook.txt -o mybook.pdf`
18175 + LHS files now set the "listings" variable, so that the definition
18176 of the `code` environment will be included in the template.
18177 + Links are colored blue by default (this can be changed by
18178 modifying `hyperref` settings in the template).
18179 + Added `lang` variable to LaTeX template.
18183 + Fixed bug in HTML template with html5 and mathml.
18184 + Don't use self-closing img, br, hr tags for HTML5.
18185 + Use `<section>` for footnotes if HTML5.
18186 + Update HTML templates to use Content-Style-Type meta tag.
18187 + Use separate variables for meta-date, meta-author.
18188 This makes footnotes work in author and date fields.
18189 + Use 'vertical-align:middle' in WebTeX math for better alignment.
18191 * S5/slidy writer: Make footnotes appear on separate slide at end.
18193 * MIME: Added 'layout-cache' to getMimeType. This ensures that the
18194 META-INF/manifest.xml for ODT files will have everything it needs, so
18195 that ODT files modified by LibreOffice can be used as `--reference-odt`.
18197 * `Text.Pandoc.Templates`: Return empty string for json template.
18199 * `Text.Pandoc.Biblio`:
18201 + Expand citations recursively inside nested inlines.
18202 + Treat `\160` as space when parsing locator and suffix.
18203 This fixes a bug with "p. 33" when `--smart` is used. Previously
18204 the whole "p. 33" would be included in the suffix, with no locator.
18205 + Put whole author-in-text citation in a Cite. Previously just the
18206 date and other info went in the Cite.
18207 + Don't add comma+space to prefix if it ends in punctuation.
18209 * Updated chicago-author-date.csl. The old version did not work
18210 properly for edited volumes with no author.
18214 + Add date to EPUB titlepage and metadata.
18215 + Added TOC identifier in EPUB page template.
18216 + Don't generate superfluous file `cover-image.jpg`.
18218### Under the hood improvements
18220 * Modified `make_osx_package.sh` to use cabal-dev.
18221 Items are no longer installed as root.
18222 Man pages are zipped and given proper permissions.
18224 * Modified windows installer generator to use cabal-dev.
18226 * Setup: Making man pages now works with cabal-dev (at least on OSX). In
18227 Setup.hs we now invoke 'runghc' in a way that points it to the correct
18228 package databases, instead of always falling back to the default user
18231 * Updated to work with GHC 7.4.1.
18233 * Removed dependency on old-time.
18235 * Removed dependency on dlist.
18237 * New slidy directory for "self-contained."
18239 * TeXMath writer: Use unicode thin spaces for thin spaces.
18241 * Markdown citations: don't strip off initial space in locator.
18245 * Removed `Apostrophe`, `EmDash`, `EnDash`, and `Ellipses`
18246 from the native `Inline` type in pandoc-types. Now we use `Str`
18247 elements with unicode.
18249 * Improvements to `Text.Pandoc.Builder`:
18251 + `Inlines` and `Blocks` are now newtypes (not synonyms for
18253 + Instances are defined for `IsString`, `Show`, `Read`, `Monoid`,
18254 and a new `Listable` class, which allows these to be manipulated
18255 to some extent like lists. Monoid append includes automatic
18257 + `+++` has been replaced by `<>` (mappend).
18259 * Use blaze-html instead of xhtml for HTML generation.
18260 This changes the type of `writeHtml`.
18262 * `Text.Pandoc.Shared`:
18264 + Added `warn` and `err`.
18265 + Removed `unescapeURI`, modified `escapeURI`.
18266 (See under [behavior changes], above.)
18268 * Changes in URI escaping: Previously the readers escaped URIs by
18269 converting unicode characters to octets and then percent encoding.
18270 Now unicode characters are left as they are, and `escapeURI` only
18271 percent-encodes space characters. This gives more readable
18272 URIs, and works well with modern user agents. URIs are no longer unescaped
18273 at all on conversion to `markdown`, `asciidoc`, `rst`, `org`.
18275 * New module `Text.Pandoc.SelfContained`.
18277 * New module `Text.Pandoc.Docx`.
18279 * New module `Text.Pandoc.PDF`.
18281 * Added `writerBeamer` to `WriterOptions`.
18283 * Added `normalizeDate` to `Text.Pandoc.Shared`.
18285 * Added `splitStringWithIndices` in `Text.Pandoc.Shared`.
18286 This is like `splitWithIndices`, but it is sensitive to distinctions
18287 between wide, combining, and regular characters.
18289 * `Text.Pandoc.Pretty`:
18291 + Added `chomp` combinator.
18292 + Added `beforeNonBreak` combinator. This allows you to include
18293 something conditionally on it being before a nonblank.
18294 Used for RST inline math.
18295 + Added `charWidth` function. All characters marked W or F in the unicode
18296 spec EastAsianWidth.txt get width 2.
18297 + Added `realLength`, based on `charWidth`. `realLength` is now
18298 used in calculating offsets.
18300 * New module `Text.Pandoc.Slides`, for common functions for breaking
18301 a document into slides.
18303 * Removed `Text.Pandoc.S5`, which is no longer needed.
18305 * Removed `Text.Pandoc.CharacterReferences`. Moved
18306 `characterReference` to `Text.Pandoc.Parsing`.
18307 `decodeCharacterReferences` is replaced by `fromEntities`
18308 in `Text.Pandoc.XML`.
18310 * Added `Text.Pandoc.ImageSize`. This is intened for use
18311 in `docx` and `odt` writers, so the size and dpi of images
18314 * Removed `writerAscii` in `WriterOptions`.
18316 * Added `writerHighlight` to `WriterOptions`.
18318 * Added `DZSlides` to `HTMLSlideVariant`.
18320 * `writeEPUB` has a new argument for font files to embed.
18322 * Added `stateLastStrPos` to `ParserState`. This lets us keep track
18323 of whether we're parsing the position immediately after a regular
18324 (non-space, non-symbol) string, which is useful for distinguishing
18325 apostrophes from single quote starts.
18327 * `Text.Pandoc.Parsing`:
18329 + `escaped` now returns a `Char`.
18330 + Removed `charsInBalanced'`, added a character parser as
18331 a parameter of `charsInBalanced`. This is needed for
18332 proper handling of escapes, etc.
18335 * Added `toEntities` to `Text.Pandoc.XML`.
18337 * `Text.Pandoc.Readers.LaTeX`:
18339 + Export `handleIncludes`.
18340 + Export `rawLaTeXBlock` instead of `rawLaTeXEnvironment'`.
18342 * Added `ToJsonFilter` class and `toJsonFilter` function to
18343 `Text.Pandoc`, deprecating the old `jsonFilter` function.
18345 * `Text.Pandoc.Highlighting`:
18347 + Removed `highlightHtml`, `defaultHighlightingCss`.
18348 + Export `formatLaTeXInline`, `formatLaTeXBlock`, and `highlight`, plus
18349 key functions from highlighting-kate.
18350 + Changed types of highlighting function. `highlight` returns a
18351 `Maybe`, not an `Either`.
18354## pandoc 1.8.2.1 (2011-08-01)
18356 * Adjusted Arbitrary instance to help avoid timeouts in tests.
18358 * Added `Tests.Writers.Markdown` to cabal file.
18360 * Relaxed version bounds on pandoc-types, test-framework.
18362## pandoc 1.8.2 (2011-07-30)
18364 * Added script to produce OS X package.
18366 * Made `templates` directory a git submodule. This should make it
18367 easier for people to revise their custom templates when the default
18370 * Changed template naming scheme: `FORMAT.template` -> `default.FORMAT`.
18371 **Note:** If you have existing templates in `~/.pandoc/templates`, you
18372 must rename them to conform to the new scheme!
18374 * Default template improvements:
18376 + HTML: Display author and date after title.
18377 + HTML: Made table of contents more customizable. The container
18378 for the TOC is now in the template, so users can insert a header
18379 or other styling. (Thanks to Bruce D'Arcus for the suggestion.)
18380 + HTML, Slidy, S5: Enclose scripts in CDATA tags.
18381 + Slidy, S5: Added `s5-url` and `slidy-url` variables, instead of
18382 hard-coding. If you want to put your slidy files in the slidy
18383 subdirectory, for example, you can do
18384 `pandoc -t slidy -V slidy-url=slidy -s`.
18385 + LaTeX: Use `\and` to separate authors in LaTeX documents (reader
18386 & writer). Closes #279.
18387 + LaTeX: Set `\emergencystretch` to prevent overfull lines.
18388 + LaTeX: Use different `hyperref` options for `xetex`, fixing
18389 problems with unicode bookmarks (thanks to CircleCode).
18390 + LaTeX: Removed `ucs` package, use `utf8` rather than `utf8x`
18391 with `inputenc`. This covers fewer characters but is more
18392 robust with other packages, and `ucs` is unmaintained. Users
18393 who need better unicode support should use xelatex or lualatex.
18395 * If a template specified with `--template` is not found, look for it
18396 in `datadir`. Also, if no extension is provided, supply one based
18397 on the writer. So now you can put your `special.latex` template in
18398 `~/.pandoc/templates`, and use it from any directory via
18399 `pandoc -t latex --template special`.
18401 * Added `nonspaceChar` to `Text.Pandoc.Parsing`.
18403 * Fixed smart quotes bug, now handling `'...hi'` properly.
18407 + Partial support for labeled footnotes.
18408 + Improved accuracy of `simpleReferenceName` parser.
18412 + Substitute correct unicode characters for
18413 characters in the 128..159 range, which are often found even in
18414 HTML that purports to be UTF-8.
18416 * LaTeX reader: Handle `\subtitle` command (a subtitle is added
18417 to the title, after a colon and linebreak). Closes #280.
18419 * Leaner `reference.odt`.
18421 * Added unexported module `Text.Pandoc.MIME` for use in
18424 * ODT writer: Construct `manifest.xml` based on archive contents.
18425 This fixes a bug in ODTs containing images. Recent versions of
18426 LibreOffice would reject these as corrupt, because `manifest.xml`
18427 did not contain a reference to the image files.
18431 + Make verbatim environments flush to avoid spurious
18432 blank lines. Closes #277.
18433 + Use `\texttt` and escapes insntead of `\verb!...!`, which
18434 is too fragile (doesn't work in command arguments).
18435 + Use `\enquote{}` for quotes if the template includes
18436 the `csquotes` package. This provides better support for
18437 local quoting styles. (Thanks to Andreas Wagner for the idea.)
18439 * ConTeXt writer: Make `\starttyping`/`\stoptyping` flush with
18440 margin, preventing spurious blank lines.
18444 + Use non-minimized version of `slidy.css` with `--offline`
18445 option, so users can more easily edit it.
18446 + Also fixed a bug in the CSS that prevented proper centering
18447 of title (now reported and fixed upstream).
18451 + Replaced `s5/default/slides.js.{comment,packed}` with
18452 new compressed `s5/default/slides.min.js`.
18453 + Use `data:` protocol to embed S5 CSS in `<link>` tags,
18454 when `--offline` is specified. Using inline CSS didn't
18455 work with Chrome or Safari. This fixes offline
18456 S5 on those browsers.
18458 * HTML writer: Removed English title on footnote backlinks.
18459 This is incongrous in non-English documents.
18463 + Use CALS tables. (Some older docbook software does not work
18464 well with XHTML tables.) Closes #77.
18465 + Use `programlisting` tags (instead of `screen`) for code blocks.
18469 + Calls latex with `-halt-on-error -interaction nonstopmode` instead
18470 of `-interaction=batchmode`, which essentially just ignored errors,
18471 leading to bad results. Better to know when something is wrong.
18472 + Fixed issues with non-UTF-8 output of `pdflatex`.
18473 + Better error reporting.
18475 * `--mathjax` now takes an optional URL argument. If it is not
18476 provided, pandoc links directly to the (secure) mathjax CDN,
18477 as now recommended (thanks to dsanson).
18479 * Deprecated `--xetex` option in `pandoc`. It is no longer needed,
18480 since the LaTeX writer now produces a file that can be processed by
18481 `latex`, `pdflatex`, `lualatex`, or `xelatex`.
18483 * Introduced `--luatex` option to `markdown2pdf`. This causes `lualatex`
18484 to be used to create the PDF.
18487## pandoc 1.8.1.2 (2011-07-16)
18489 * Added `--epub-cover-image` option.
18491 * Documented `--biblatex` and `--natbib` options.
18493 * Allow `--section-divs` with slidy output. Resolves Issue #296.
18495 * Disallow notes within notes in reST and markdown.
18496 These previously caused infinite looping and stack overflows.
18503 Note references are allowed in reST notes, so this isn't a full
18504 implementation of reST. That can come later. For now we need to
18505 prevent the stack overflows. Partially resolves Issue #297.
18507 * EPUB writer: Allow non-plain math methods.
18509 * Forbid ()s in citation item keys. Resolves Issue #304: problems with
18510 `(@item1; @item2)` because the final paren was being parsed as part of
18513 * Changed URI parser so it doesn't include trailing punctuation.
18514 So, in RST, `http://google.com.` should be parsed as a link followed by a
18515 period. The parser is smart enough to recognize balanced parentheses, as
18516 often occur in wikipedia links: `http://foo.bar/baz_(bam)`.
18518 * Markdown+lhs reader: Require space after inverse bird tracks, so that
18519 HTML tags can be used freely at the left margin of a markdown+lhs document.
18520 Thanks to Conal Elliot for the suggestion.
18522 * Markdown reader: Fixed bug in footnote order (reported by CircleCode).
18525 + Fixed bug in in field lists with multi-line items at the
18527 + Added parentheses to RST `specialChars`, so
18528 `(http://google.com)` will be parsed as a link in parens.
18529 Resolves Issue #291.
18530 + Allow `|` followed by newline in RST line block.
18534 + Gobble option & space after linebreak `\\[10pt]`.
18537 + Make it possible to have colons after links. (qerub)
18538 + Make it possible to have colons after links. (Christoffer Sawicki)
18541 + Skip spaces after `<b>`, `<emph>`, etc.
18542 + Handle tbody, thead in simple tables. Closes #274.
18543 + Implicit `Para`s instead of `Plains` in some contexts.
18545 * OpenDocument writer: Use special `First paragraph` style for
18546 first paragraph after most non-paragraph blocks. This allows users to
18547 specify e.g. that only paragraphs after the first paragraph of a block are
18548 to be indented. Thanks to Andrea Rossato for the patch. Closes #20.
18550 * LaTeX writer: use `deVerb` on table and picture captions.
18551 Otherwise LaTeX complains about `\verb` inside command argument.
18552 Thanks to bbanier for reporting the bug.
18554 * Markdown writer: Insert HTML comment btw list and indented code block.
18555 This prevents the code block from being interpreted as part of the list.
18557 * EPUB writer: Add a meta element specify the cover.
18558 Some EPUB e-readers, such as the Nook, require a meta element inside the
18559 OPF metadata block to ensure the cover image is properly displayed.
18562 * HTML writer: Use embed tag for images with non-image extensions.
18563 (e.g. PDFs). Closes #264.
18565 * LaTeX writer: Improved tables.
18567 + More space between lines, top-align cells.
18568 + Use ctable package, which allows footnotes and
18569 provides additional options.
18570 + Made cell alignments work in multiline tables.
18571 + Closes #271, #272.
18573 * Un-URI-escape image filenames in LaTeX, ConTeXt, RTF, Texinfo.
18574 Also do this when copying image files into EPUBs and ODTs.
18577 * Changed to github issue tracker.
18579 * Added failing emph/strong markdown test case due to Perry Wagle.
18581 * Slidy improvements:
18582 + Updated to use Slidy2.
18583 + Fixed bug, unclosed div tag.
18584 + Added `duration` variable in template.
18585 Setting this activates the timer.
18586 + Use 'titlepage' instead of 'cover' for title div.
18589## pandoc 1.8.1.1 (2011-02-13)
18591 * `markdown2pdf`: Removed some debugging lines accidentally included
18592 in the 1.8.1 release. With those lines, the temp directory is created
18593 in the working directory, and it is not deleted. This fix restores
18594 the original behavior.
18596## pandoc 1.8.1 (2011-02-13)
18598 * Added `--ascii` option. Currently supported only in HTML writer,
18599 which it causes to use numerical entities instead of UTF-8.
18601 * EPUB writer: `--toc` now works to provide a table of contents
18602 at the beginning of each chapter.
18604 * LaTeX writer: Change figure defaults to `htbp`.
18605 This prevents "too many unprocessed floats." Resolves
18608 * `Text.Pandoc.UTF8`: Encode filenames even when using recent
18611 * `markdown2pdf`: Fixed filename encoding issues. With help from Paulo
18612 Tanimoto. Resolves Issue #286.
18614 * HTML writer: Put line breaks in section divs.
18616 * `Text.Pandoc.Shared`: Make `writerSectionDivs` default to False.
18618## pandoc 1.8.0.3 (2011-02-05)
18620 * Fixed Source-repository stanza in cabal file.
18622## pandoc 1.8.0.2 (2011-02-05)
18626 + Stringify alt text instead of converting to HTML.
18627 + Break lines after block elements, not inside tags.
18628 HTML output now closely resembles that of tidy. Resolves Issue #134.
18630 * Markdown reader: Fixed bug in footnote block parser (pointed out
18631 by Jesse Rosenthal). The problem arose when the blank line
18632 at the end of a footnote block contained indenting spaces.
18634 * Shared: Improved 'normalize' function so it normalizes Spaces too.
18635 In normal form, Space elements only occur to separate two non-Space
18636 elements. So, we never have [Space], or [, ..., Space].
18640 + Improved Arbitrary instance.
18641 + Added timeout for test instances.
18645 + Added section on four-space rule for lists. Resolves Issue #283.
18646 + Clarified optional arguments on math options.
18648 * markdown2pdf: Fixed bug with output file extensions.
18649 Previously `markdown2pdf test.txt -o test.en.pdf` would produce
18650 `test.pdf`, not `test.en.pdf`. Thanks to Paolo Tanimoto for the fix.
18652## pandoc 1.8.0.1 (2011-01-31)
18654 * Revised Interact.hs so that it works with the CPP macros
18655 in the UTF8 module.
18657 * Revised Setup.hs so that we don't call MakeManPage.hs unless
18658 the man pages are out of date.
18660## pandoc 1.8 (2011-01-30)
18664 * Support for citations using Andrea Rossato's `citeproc-hs` 0.3.
18665 You can now write, for example,
18667 Water is wet [see @doe99, pp. 33-35; also @smith04, ch. 1].
18669 and, when you process your document using `pandoc`, specifying
18670 a citation style using `--csl` and a bibliography using `--bibliography`,
18671 the citation will be replaced by an appropriately formatted
18672 citation, and a list of works cited will be added to the end
18675 This means that you can switch effortlessly between different citation
18676 and bibliography styles, including footnote, numerical, and author-date
18677 formats. The bibliography can be in any of the following formats: MODS,
18678 BibTeX, BibLaTeX, RIS, EndNote, EndNote XML, ISI, MEDLINE, Copac, or JSON.
18679 See the README for further details.
18681 Citations are supported in the markdown reader, using a special
18682 syntax, and in the LaTeX reader, using natbib or biblatex syntax.
18683 (Thanks to Nathan Gass for the natbib and biblatex support.)
18685 * New `textile` reader and writer. Thanks to Paul Rivier for contributing
18686 the `textile` reader, an almost complete implementation of the textile
18687 syntax used by the ruby [RedCloth library](http://redcloth.org/textile).
18688 Resolves Issue #51.
18690 * New `org` writer, for Emacs Org-mode, contributed by Puneeth Chaganti.
18692 * New `json` reader and writer, for reading and writing a JSON
18693 representation of the native Pandoc AST. These are much faster
18694 than the `native` reader and writer, and should be used for
18695 serializing Pandoc to text. To convert between the JSON representation
18696 and native Pandoc, use `encodeJSON` and `decodeJSON` from
18697 `Text.JSON.Generic`.
18699 * A new `jsonFilter` function in `Text.Pandoc` makes it easy
18700 to write scripts that transform a JSON-encoded pandoc document.
18703 -- removelinks.hs - removes links from document
18705 main = interact $ jsonFilter $ bottomUp removeLink
18706 where removeLink (Link xs _) = Emph xs
18709 To use this to remove links while translating markdown to LaTeX:
18711 pandoc -t json | runghc removelinks.hs | pandoc -f json -t latex
18713 * Attributes are now allowed in inline `Code` elements, for example:
18715 In this code, `ulist ! [theclass "special"] << elts`{.haskell} is...
18717 The attribute syntax is the same as for delimited code blocks.
18718 `Code` inline has an extra argument place for attributes, just like
18719 `CodeBlock`. Inline code will be highlighted in HTML output, if pandoc
18720 is compiled with highlighting support. Resolves Issue #119.
18722 * New `RawBlock` and `RawInline` elements (replacing `RawHtml`,
18723 `HtmlInline`, and `TeX`) provide lots of flexibility in writing
18724 scripts to transform Pandoc documents. Scripts can now change
18725 how each element is rendered in each output format.
18727 * You can now define LaTeX macros in markdown documents, and pandoc
18728 will apply them to TeX math. For example,
18730 \newcommand{\plus}[2]{#1 + #2}
18733 yields `3+4`. Since the macros are applied in the reader, they
18734 will work in every output format, not just LaTeX.
18736 * LaTeX macros can also be used in LaTeX documents (both in math
18737 and in non-math contexts).
18739 * A new `--mathjax` option has been added for displaying
18740 math in HTML using MathJax. Resolves issue #259.
18742 * Footnotes are now supported in the RST reader. (Note, however,
18743 that unlike docutils, pandoc ignores the numeral or symbol used in
18744 the note; footnotes are put in an auto-numbered ordered list.)
18745 Resolves Issue #258.
18747 * A new `--normalize` option causes pandoc to normalize the AST
18748 before writing the document. This means that, for example,
18749 `*hi**there*` will be rendered as `<em>hithere</em>`
18750 instead of `<em>hi</em><em>there</em>`. This is not the default,
18751 because there is a significant performance penalty.
18753 * A new `--chapters` command-line option causes headers
18754 in DocBook, LaTeX, and ConTeXt to start with "chapter" (level one).
18755 Resolves Issue #265.
18757 * In DocBook output, `<chapter>` is now used for top-level
18758 headers if the template contains `<book>`. Resolves Issue #265.
18760 * A new `--listings` option in `pandoc` and `markdown2pdf` causes
18761 the LaTeX writer to use the listings package for code blocks.
18762 (Thanks to Josef Svennigsson for the pandoc patch, and Etienne
18763 Millon for the markdown2pdf patch.)
18765 * `markdown2pdf` now supports `--data-dir`.
18767 * URLs in autolinks now have class "url" so they can be styled.
18769 * Improved prettyprinting in most formats. Lines will be wrapped
18770 more evenly and duplicate blank lines avoided.
18772 * New `--columns` command-line option sets the column width for
18773 line wrapping and relative width calculations for tables.
18775 * Made `--smart` work in HTML, RST, and Textile readers, as well
18778 * Added `--html5` option for HTML5 output.
18780 * Added support for listings package in LaTeX reader
18781 (Puneeth Chaganti).
18783 * Added support for simple tables in the LaTeX reader.
18785 * Added support for simple tables in the HTML reader.
18787 * Significant performance improvements in many readers and writers.
18789### API and program changes
18791 * Moved `Text.Pandoc.Definition` from the `pandoc` package to a new
18792 auxiliary package, `pandoc-types`. This will make it possible for other
18793 programs to supply output in Pandoc format, without depending on the whole
18796 * Added `Attr` field to `Code`.
18798 * Removed `RawHtml`, `HtmlInline`, and `TeX` elements; added generic
18799 `RawBlock` and `RawInline`.
18801 * Moved generic functions to `Text.Pandoc.Generic`. Deprecated
18802 `processWith`, replacing it with two functions, `bottomUp` and `topDown`.
18803 Removed previously deprecated functions `processPandoc` and `queryPandoc`.
18805 * Added `Text.Pandoc.Builder`, for building `Pandoc` structures.
18807 * `Text.Pandoc` now exports association lists `readers` and `writers`.
18809 * Added `Text.Pandoc.Readers.Native`, which exports `readNative`.
18810 `readNative` can now read full pandoc documents, block lists, blocks,
18811 inline lists, or inlines. It will interpret `Str "hi"`
18812 as if it were `Pandoc (Meta [] [] []) [Plain [Str "hi"]]`.
18813 This should make testing easier.
18815 * Removed deprecated `-C/--custom-header` option.
18816 Use `--template` instead.
18818 * `--biblio-file` has been replaced by `--bibliography`.
18819 `--biblio-format` has been removed; pandoc now guesses the format
18820 from the file extension (see README).
18822 * pandoc will treat an argument as a URI only if it has an
18823 `http(s)` scheme. Previously pandoc would treat some
18824 Windows pathnames beginning with `C:/` as URIs.
18826 * The `--sanitize-html` option and the `stateSanitize` field in
18827 `ParserState` have been removed. Sanitization is better done in the
18828 resulting HTML using `xss-sanitize`, which is based on pandoc's
18829 sanitization, but improved.
18831 * pandoc now adds a newline to the end of its output in fragment
18832 mode (= not `--standalone`).
18834 * Added support for `lang` in `html` tag in the HTML template,
18835 so you can do `pandoc -s -V lang=es`, for example.
18837 * `highlightHtml` in `Text.Pandoc.Highlighting` now takes
18838 a boolean argument that selects between "inline" and
18841 * `Text.Pandoc.Writers.RTF` now exports `rtfEmbedImage`.
18842 Images are embedded in RTF output when possible (png, jpeg).
18843 Resolves Issue #275.
18845 * Added `Text.Pandoc.Pretty`. This is better suited for pandoc than the
18846 `pretty` package. Changed all writers that used
18847 `Text.PrettyPrint.HughesPJ` to use `Text.Pandoc.Pretty` instead.
18849 * Rewrote `writeNative` using the new prettyprinting module. It is
18850 now much faster. The output has been made more consistent and compressed.
18851 `writeNative` is also now sensitive to writerStandalone`, and will simply
18852 `print a block list if writerStandalone` is False.
18854 * Removed `Text.Pandoc.Blocks`. `Text.Pandoc.Pretty` allows you to define
18855 blocks and concatenate them, so a separate module is no longer needed.
18857 * `Text.Pandoc.Shared`:
18859 + Added `writerColumns`, `writerChapters`, and `writerHtml5` to
18861 + Added `normalize`.
18862 + Removed unneeded prettyprinting functions:
18863 `wrapped`, `wrapIfNeeded`, `wrappedTeX`, `wrapTeXIfNeeded`, `hang'`,
18864 `BlockWrapper`, `wrappedBlocksToDoc`.
18865 + Made `splitBy` take a test instead of an element.
18866 + Added `findDataFile`, refactored `readDataFile`.
18867 + Added `stringify`. Rewrote `inlineListToIdentifier` using `stringify`.
18868 + Fixed `inlineListToIdentifier` to treat `\160` as ' '.
18870 * `Text.Pandoc.Readers.HTML`:
18872 + Removed `rawHtmlBlock`, `anyHtmlBlockTag`, `anyHtmlInlineTag`,
18873 `anyHtmlTag`, `anyHtmlEndTag`, `htmlEndTag`, `extractTagType`,
18874 `htmlBlockElement`, `htmlComment`
18875 + Added `htmlTag`, `htmlInBalanced`, `isInlineTag`, `isBlockTag`,
18878 * Moved `smartPunctuation` from `Text.Pandoc.Readers.Markdown`
18879 to `Text.Pandoc.Readers.Parsing`, and parameterized it with
18882 * Ellipses are no longer allowed to contain spaces.
18883 Previously we allowed '. . .', ' . . . ', etc. This caused
18884 too many complications, and removed author's flexibility in
18885 combining ellipses with spaces and periods.
18887 * Allow linebreaks in URLs (treat as spaces). Also, a string of
18888 consecutive spaces or tabs is now parsed as a single space. If you have
18889 multiple spaces in your URL, use `%20%20`.
18891 * `Text.Pandoc.Parsing`:
18893 + Removed `refsMatch`.
18894 + Hid `Key` constructor.
18895 + Removed custom `Ord` and `Eq` instances for `Key`.
18896 + Added `toKey` and `fromKey` to convert between `Key` and `[Inline]`.
18897 + Generalized type on `readWith`.
18899 * Small change in calculation of relative widths of table columns.
18900 If the size of the header > the specified column width, use
18901 the header size as 100% for purposes of calculating
18902 relative widths of columns.
18904 * Markdown writer now uses some pandoc-specific features when `--strict`
18905 is not specified: `\` newline is used for a hard linebreak instead of
18906 two spaces then a newline. And delimited code blocks are used when
18907 there are attributes.
18909 * HTML writer: improved gladTeX output by setting ENV appropriately
18910 for display or inline math (Jonathan Daugherty).
18912 * LaTeX writer: Use `\paragraph`, `\subparagraph` for level 4,5 headers.
18916 + `\label{foo}` and `\ref{foo}` now become `{foo}` instead of `(foo)`.
18917 + `\index{}` commands are skipped.
18919 * Added `fontsize` variable to default LaTeX template.
18920 This makes it easy to set the font size using `markdown2pdf`:
18921 `markdown2pdf -V fontsize=12pt input.txt`.
18923 * Fixed problem with strikeout in LaTeX headers when using
18924 hyperref, by adding a command to the default LaTeX template
18925 that disables `\sout` inside pdf strings. Thanks to Joost Kremers
18928 * The `COLUMNS` environment variable no longer has any effect.
18930### Under-the-hood improvements
18932 * Pandoc now compiles with GHC 7. (This alone leads to a
18933 significant performance improvement, 15-20%.)
18935 * Completely rewrote HTML reader using tagsoup as a lexer. The
18936 new reader is faster and more accurate. Unlike the
18937 old reader, it does not get bogged down on some input
18938 (Issues #277, 255). And it handles namespaces in tags
18941 * Replaced `escapeStringAsXML` with a faster version.
18943 * Rewrote `spaceChar` and some other parsers in Text.Pandoc.Parsing
18944 for a significant performance boost.
18946 * Improved performance of all readers by rewriting parsers.
18948 * Simplified Text.Pandoc.CharacterReferences by using
18949 entity lookup functions from TagSoup.
18951 * `Text.Pandoc.UTF8` now uses the unicode-aware IO functions
18952 from `System.IO` if base >= 4.2. This gives support for
18953 windows line endings on windows.
18955 * Remove duplications in documentation by generating the
18956 pandoc man page from README, using `MakeManPage.hs`.
18958 * README now includes a full description of markdown syntax,
18959 including non-pandoc-specific parts. A new `pandoc_markdown`
18960 man page is extracted from this, so you can look up markdown
18961 syntax by doing `man pandoc_markdown`.
18963 * Completely revised test framework (with help from Nathan Gass).
18964 The new test framework is built when the `tests` Cabal flag is set. It
18965 includes the old integration tests, but also some new unit and quickcheck
18966 tests. Test output has been much improved, and you can now specify a glob
18967 pattern after `cabal test` to indicate which tests should be run;
18968 for example `cabal test citations` will run all the citation tests.
18970 * Added a shell script, `stripansi.sh`, for filtering ANSI control
18971 sequences from test output: `cabal test | ./stripansi.sh > test.log`.
18973 * Added `Interact.hs` to make it easier to use ghci while developing.
18974 `Interact.hs` loads `ghci` from the `src` directory, specifying
18975 all the options needed to load pandoc modules (including
18976 specific package dependencies, which it gets by parsing
18977 dist/setup-config).
18979 * Added `Benchmark.hs`, testing all readers + writers using criterion.
18981 * Added `stats.sh`, to make it easier to collect and archive
18982 benchmark and lines-of-code stats.
18984 * Added upper bounds to all cabal dependencies.
18986 * Include man pages in extra-source-files. This allows users to
18987 install pandoc from the tarball without needing to build the man
18992 * Filenames are encoded as UTF8. Resolves Issue #252.
18994 * Handle curly quotes better in `--smart` mode. Previously, curly quotes
18995 were just parsed literally, leading to problems in some output formats.
18996 Now they are parsed as `Quoted` inlines, if `--smart` is specified.
18997 Resolves Issue #270.
18999 * Text.Pandoc.Parsing: Fixed bug in grid table parser.
19000 Spaces at end of line were not being stripped properly,
19001 resulting in unintended LineBreaks.
19005 + Allow HTML comments as inline elements in markdown.
19006 So, `aaa <!-- comment --> bbb` can be a single paragraph.
19007 + Fixed superscripts with links: `^[link](/foo)^` gets
19008 recognized as a superscripted link, not an inline note followed by
19010 + Fixed regression, making markdown reference keys case-insensitive again.
19011 Resolves Issue #272.
19012 + Properly handle abbreviations (like `Mr.`) at the end of a line.
19013 + Better handling of intraword underscores, avoiding exponential
19014 slowdowns in some cases. Resolves Issue #182.
19015 + Fixed bug in alignments in tables with blank rows in the header.
19019 + Field lists now allow spaces in field names, and
19020 block content in field values. (Thanks to Lachlan Musicman
19021 for pointing out the bug.)
19022 + Definition list items are now always `Para` instead of
19023 `Plain`, matching behavior of `rst2xml.py`.
19024 + In image blocks, the description is parsed properly and
19025 used for the alt attribute, not also the title.
19026 + Skip blank lines at beginning of file. Resolves
19031 + Improved parsing of preamble.
19032 Previously you'd get unexpected behavior on a document that
19033 contained `\begin{document}` in, say, a verbatim block.
19034 + Allow spaces between `\begin` or `\end` and `{`.
19035 + Support `\L` and `\l`.
19036 + Skip comments inside paragraphs.
19040 + Escape strings in `\href{..}`.
19041 + In nonsimple tables, put cells in `\parbox`.
19043 * OpenDocument writer: don't print raw TeX.
19047 + Fixed bug in `Image`. URI was getting unescaped twice!
19048 + Avoid printing extra blank lines at the end if there are
19049 no notes or references.
19051 * LaTeX and ConTeXt: Escape `[` and `]` as `{[}` and `{]}`.
19052 This avoids unwanted interpretation as an optional argument.
19054 * ConTeXt writer: Fixed problem with inline code. Previously
19055 `}` would be rendered `\type{}}`. Now we check the string for '}' and '{'.
19056 If it contains neither, use `\type{}`; otherwise use `\mono{}`
19057 with an escaped version of the string.
19059 * `:` now allowed in HTML tags. Resolves Issue #274.
19063## pandoc 1.6 (2010-07-24)
19065+ New EPUB and HTML Slidy writers. (Issue #122)
19067 - [EPUB] is a standard ebook format, used in Apple's iBooks for
19068 the iPad and iPhone, Barnes and Noble's nook reader, the Sony
19069 reader, and many other devices, and by online ebook readers like
19070 [bookworm]. (Amazon's Kindle uses a different format, MobiPocket,
19071 but EPUB books can easily be converted to Kindle format.) Now you
19072 can write your book in markdown and produce an ebook with a single
19073 command! I've put up a short [tutorial here].
19074 - [Slidy], like S5, is a system for producing HTML+javascript slide shows.
19076+ All input is assumed to be UTF-8, no matter what the locale and ghc
19077 version, and all output is UTF-8. This reverts to pre-1.5 behavior.
19078 Also, a BOM, if present, is stripped from the input.
19080+ Markdown now supports grid tables, whose cells can contain
19081 arbitrary block elements. (Issue #43)
19083+ Sequentially numbered example lists in markdown with `@` marker.
19085+ Markdown table captions can begin with a bare colon and no longer need
19086 to include the English word "table." Also, a caption can now occur
19087 either before or after the table. (Issue #227)
19089+ New command-line options:
19091 - `--epub-stylesheet` allows you to specify a CSS file that will
19092 be used to style your ebook.
19093 - `--epub-metadata` allows you to specify metadata for the ebook.
19094 - `--offline` causes the generated HTML slideshow to include all
19095 needed scripts and stylesheets.
19096 - `--webtex` causes TeX math to be converted to images using the
19097 Google Charts API (unless a different URL is specified).
19098 - `--section-divs` causes div tags to be added around each section
19099 in an HTML document. (Issue #230, 239)
19101+ Default behavior of S5 writer in standalone mode has changed:
19102 previously, it would include all needed scripts and stylesheets
19103 in the generated HTML; now, only links are included unless
19104 the `--offline` option is used.
19106+ Default behavior of HTML writer has changed. Between 1.2 and 1.5,
19107 pandoc would enclose sections in div tags with identifiers on the
19108 div tags, so that the sections can be manipulated in javascript.
19109 This caused undesirable interactions with raw HTML div tags. So,
19110 starting with 1.6, the default is to put the identifiers directly
19111 on the header tags, and not to include the divs. The `--section-divs`
19112 option selects the 1.2-1.5 behavior.
19116 - `HTMLMathMethod`: Added `WebTeX`, removed `MimeTeX`.
19117 - `WriterOptions`: Added `writerUserDataDir`, `writerSourceDirectory`,
19118 `writerEPUBMetadata` fields. Removed `writerIncludeBefore`,
19119 `writerIncludeAfter`.
19120 - Added `headerShift` to `Text.Pandoc.Shared`.
19121 - Moved parsing code and `ParserState` from `Text.Pandoc.Shared`
19122 to a new module, `Text.Pandoc.Parsing`.
19123 - Added `stateHasChapters` to `ParserState`.
19124 - Added `HTMLSlideVariant`.
19125 - Made `KeyTable` a map instead of an association list.
19126 - Added accessors for `Meta` fields (`docTitle`, `docAuthors`, `docDate`).
19127 - `Pandoc`, `Meta`, `Inline`, and `Block` have been given `Ord` instances.
19128 - Reference keys now have a type of their own (`Key`), with its
19129 own `Ord` instance for case-insensitive comparison.
19130 - Added `Text.Pandoc.Writers.EPUB`.
19131 - Added `Text.Pandoc.UUID`.
19132 - Removed `Text.Pandoc.ODT`, added `Text.Pandoc.Writers.ODT`.
19133 Removed `saveOpenDocumentAsODT`, added `writeODT`.
19134 - Added `Text.Pandoc.Writers.Native` and `writeNative`.
19135 Removed `prettyPandoc`.
19136 - Added `Text.Pandoc.UTF8` for portable UTF8 string IO.
19137 - Removed `Text.Pandoc.Writers.S5` and the `writeS5` function.
19138 Moved `s5Includes` to a new module, `Text.Pandoc.S5`.
19139 To write S5, you now use `writeHtml` with `writerSlideVariant`
19140 set to `S5Slides` or `SlidySlides`.
19142+ Template changes. If you use custom templates, please update them,
19143 particularly if you use syntax highlighting with pandoc. The old HTML
19144 templates hardcoded highlighting CSS that will no longer work with
19145 the most recent version of highlighting-kate.
19147 - HTML template: avoid empty meta tag if no date.
19148 - HTML template: Use default highlighting CSS from highlighting-kate
19149 instead of hard-coding the CSS into the template.
19150 - HTML template: insert-before text goes before the title, and
19151 immediately after the `<body>` tag, as documented. (Issue #241)
19152 - Added slidy and s5 templates.
19153 - Added amssymb to preamble of latex template. (github Issue 1)
19155+ Removed excess newlines at the end of output. Note: because output
19156 will not contain an extra newline, you may need to make adjustments
19157 if you are inserting pandoc's output into a template.
19159+ In S5 and slidy, horizontal rules now cause a new slide, so you
19160 are no longer limited to one slide per section.
19162+ Improved handling of code in man writer. Inline code is now monospace,
19163 not bold, and code blocks now use .nf (no fill) and .IP (indented para).
19165+ HTML reader parses `<tt>` as Code. (Issue #247)
19167+ html+lhs output now contains bird tracks, even when compiled without
19168 highlighting support. (Issue #242)
19170+ Colons are now no longer allowed in autogenerated XML/HTML identifiers,
19171 since they have a special meaning in XML.
19173+ Code improvements in ODT writer. Remote images are now replaced with
19174 their alt text rather than a broken link.
19176+ LaTeX reader improvements:
19178 - Made latex `\section`, `\chapter` parsers more forgiving of whitespace.
19179 - Parse `\chapter{}` in latex.
19180 - Changed `rawLaTeXInline` to accept `\section`, `\begin`, etc.
19181 - Use new `rawLaTeXInline'` in LaTeX reader, and export `rawLaTeXInline`
19182 for use in markdown reader.
19183 - Fixes bug wherein `\section{foo}` was not recognized as raw TeX
19184 in markdown document.
19186+ LaTeX writer: images are automatically shrunk if they would extend
19187 beyond the page margin.
19189+ Plain, markdown, RST writers now use unicode for smart punctuation.
19191+ Man writer converts math to unicode when possible, as in other writers.
19193+ `markdown2pdf` can now recognize citeproc options.
19195+ Command-line arguments are converted to UTF-8. (Issue #234)
19197+ `Text.Pandoc.TeXMath` has been rewritten to use texmath's parser.
19198 This allows it to handle a wider range of formulas. Also, if a formula
19199 cannot be converted, it is left in raw TeX; formulas are no longer
19200 partially converted.
19202+ Unicode curly quotes are left alone when parsing smart quotes. (Issue #143)
19204+ Cabal file changes:
19206 - Removed parsec < 3 restriction.
19207 - Added 'threaded' flag for architectures where GHC lacks a threaded
19209 - Use 'threaded' only for markdown2pdf; it is not needed for pandoc.
19210 - Require highlighting-kate 0.2.7.
19212+ Use explicit imports from `Data.Generics`. Otherwise we have a
19213 conflict with the 'empty' symbol, introduced in syb >= 0.2. (Issue #237)
19215+ New data files: slidy/slidy.min.js, slidy/slidy.min.css, epub.css.
19217[EPUB]: http://en.wikipedia.org/wiki/EPUB
19218[Slidy]: http://www.w3.org/Talks/Tools/Slidy
19219[bookworm]: http://bookworm.oreilly.com/
19220[tutorial here]: http://johnmacfarlane.net/pandoc/epub.html
19222## pandoc 1.5.1.1 (2010-03-29)
19224+ Fixed header identifiers (uniqueIdent in Shared) so they
19225 work as advertized in README and are guaranteed to be
19226 valid XHTML names. Thanks to Xyne for reporting the bug.
19228## pandoc 1.5.1 (2010-03-23)
19230+ Fixed treatment of unicode characters in URIs.
19231+ Revised Setup.hs so it works with debian's build process.
19232+ Fixed bug in OpenDocument writer that led to invalid
19233 XML for some input.
19235## pandoc 1.5.0.1 (2010-03-21)
19237+ HTML writer: Fixed error in math writer (with MathML option)
19238 that caused an infinite loop for unparsable MathML.
19240## pandoc 1.5 (2010-03-20)
19242+ Moved repository to [github](http://github.com/jgm/pandoc).
19243+ New `--mathml` option, for display of TeX math as MathML.
19244+ New `--data-dir` option, allowing users to specify a data
19245 directory other than `~/.pandoc`. Files placed in this directory
19246 will be used instead of system defaults.
19247+ New `--base-header-level` option. For example, `--base-header-level=2`
19248 changes level 1 headers to level 2, level 2 to level 3, etc.
19249+ New 'plain' output format: plain text without pictures, hyperlinks,
19250 inline formatting, or anything else that looks even vaguely
19252+ Titles and authors in title blocks can now span multiple lines,
19253 as long as the continuation lines begin with a space character.
19254+ When given an absolute URI as a parameter, pandoc will fetch
19255 the content via HTTP.
19256+ The HTML reader has been made much more forgiving. It no
19257 longer requires well-formed xhtml as input.
19258+ `html2markdown` has been removed; it is no longer necessary, given
19259 the last two changes. `pandoc` can be used by itself to convert
19260 web pages to markdown or other formats.
19261+ `hsmarkdown` has also been removed. Use `pandoc --strict` instead.
19262 Or symlink pandoc's executable to `hsmarkdown`; `pandoc` will then
19263 behave like `hsmarkdown` used to.
19264+ An image in a paragraph by itself is now rendered as a figure
19265 in most writers, with the alt text as the caption.
19266+ Incomplete support for reST tables (simple and grid). Thanks to
19267 Eric Kow. Colspans and rowspans not yet supported.
19268+ In mediawiki, links with relative URLs are now formatted as wikilinks.
19269 Also, headers have been promoted: `= head =` is now level 1 instead of
19271+ The markdown reader now handles "inverse bird tracks" when parsing
19272 literate haskell. These are used for haskell example code that
19273 is not part of the literate program.
19274+ The `-B` and `-A` options now imply `-s` and no longer work in
19276+ Headerless tables are now printed properly in all writers.
19277 In addition, tbody, thead, and cols are used in HTML and Docbook
19279+ Improved build system; removed obsolete Makefile.
19280+ In LaTeX writer, `\chapter` is now used instead of `\section`.
19281 when the documentclass is book, report, or memoir.
19282+ Many small bug fixes. See [changelog] for details.
19284## pandoc 1.4 (2010-01-02)
19286+ New template system replaces old headers, giving users much
19287 more control over pandoc's output in `--standalone` mode.
19288 Added `--template` and `--variable` options. The `--print-default-header`
19289 option is now `--print-default-template`. See README under
19290 "Templates" for details.
19291+ The old `--custom-header` option should still work, but it has
19293+ New `--reference-odt` option allows users to customize styles
19295+ Users may now put custom templates, s5 styles, and a reference
19296 ODT in the `~/.pandoc` directory, where they will override system
19297 defaults. See README for details.
19298+ Unicode is now used whenever possible in HTML and XML output. Entities
19299 are used only where necessary (`>`, `<`, `"`, `&`).
19300+ Authors and dates may now include formatting and notes.
19301+ Added `--xetex` option for `pandoc` and `markdown2pdf`.
19302+ Windows installer now includes highlighting support and
19303 `markdown2pdf` and `hsmarkdown` wrappers.
19304+ Pandoc no longer requires Template Haskell, which should make
19306+ Pandoc can now be built on GHC 6.12, as well as earlier versions.
19307+ See README for other small improvements and bug fixes.
19309## pandoc 1.3 (2009-12-10)
19311+ Added `--id-prefix` option to help prevent duplicate
19312 identifiers when you're generating HTML fragments.
19313+ Added `--indented-code-classes` option, which specifies
19314 default highlighting syntax for indented code blocks.
19315+ `--number-sections` now affects HTML output.
19316+ Improved syntax for markdown definition lists.
19317+ Better looking simple tables.
19318+ Markdown tables without headers are now possible.
19319+ New hard line break syntax: backslash followed by newline.
19320+ Improved performance of markdown reader by ~10% by eliminating the
19321 need for a separate parsing pass for notes.
19322+ Improved syntax highlighting for literate Haskell.
19323+ Support for "..code-block" directive in RST reader.
19324+ Windows binary now includes highlighting support.
19325+ Many bug fixes and small improvements. See [changelog]
19328## pandoc 1.2.1 (2009-07-18)
19330+ Improved the efficiency of the markdown reader's
19331 abbreviation parsing (should give a big performance
19332 boost with `--smart`).
19333+ HTML writer now wraps sections in divs with unique
19334 identifiers, for easier manipulation.
19335+ Improved LaTeX reader's coverage of math modes.
19336+ Added a portable Haskell version of markdown2pdf (thanks
19337 to Paolo Tanimoto).
19338+ Made `--strict` compatible with `--standalone` and `--toc.`
19339+ Many other small improvements and bug fixes.
19340 See [changelog] for details.
19342## pandoc 1.2 (2009-03-01)
19344+ Added support for literate Haskell. lhs support is triggered by
19345 '+lhs' suffixes in formats. For example, 'latex+lhs' is literate
19346 Haskell LaTeX. '.lhs' files are treated by default as literate
19348+ Added `--email-obfuscation` option.
19349+ Brought citeproc support up to date for citeproc-hs-0.2.
19350+ Many bugs fixed. See [changelog] for details.
19352## pandoc 1.1 (2008-11-06)
19354+ New `--jsmath` option supporting use of pandoc with [jsMath].
19355+ Classes on HTML table output for better CSS styling.
19356+ Windows installer no longer requires admin privileges.
19357+ Many bugs fixed. See [changelog] for details.
19359## pandoc 1.0 (2008-09-13)
19361+ New writers for MediaWiki, GNU Texinfo (thanks to Peter Wang),
19362 OpenDocument XML (thanks to Andrea Rossato), and ODT (OpenOffice
19364+ New [delimited code blocks](README.html#delimited-code-blocks),
19365 with optional syntax highlighting.
19366+ Reorganized build system: pandoc can now be built using standard
19367 Cabal tools. It can be compiled on Windows without Cygwin.
19368 The tests can also be run without perl or unix tools.
19369+ LaTeXMathML replaces ASCIIMathML for rendering math in HTML.
19370+ Support for "displayed" math.
19371+ Common abbreviations are now handled more intelligently, with
19372 a non-breaking space (and not a sentence-ending space) after
19374+ Code is -Wall clean.
19375+ Many bug fixes and small improvements. See [changelog] for
19378## pandoc 0.46 (2008-01-08)
19380+ Added a `--sanitize-html` option (and a corresponding parameter
19381 in `ParserState` for those using the pandoc libraries in programs).
19382 This option causes pandoc to sanitize HTML (in HTML or Markdown
19383 input) using a whitelist method. Possibly harmful HTML elements
19384 are replaced with HTML comments. This should be useful in the
19385 context of web applications, where pandoc may be used to convert
19386 user input into HTML.
19387+ Made -H, -A, and -B options cumulative: if they are specified
19388 multiple times, multiple files will be included.
19389+ Many bug fixes and small improvements. See [changelog] for full
19392## pandoc 0.45 (2007-12-09)
19394+ Many bug fixes and structural improvements. See [changelog] for
19396+ Improved treatment of math. Math is now rendered using unicode
19397 by default in HTML, RTF, and DocBook output. For more accurate
19398 display of math in HTML, `--gladtex`, `--mimetex`, and `--asciimathml`
19399 options are provided. See the [User's Guide](README.html#math) for
19401+ Removed support for box-style block quotes in markdown.
19402+ More idiomatic ConTeXt output.
19403+ Text wrapping in ConTeXt and LaTeX output.
19404+ Pandoc now correctly handles all standard line endings
19406+ New `--no-wrap` option that disables line wrapping and minimizes
19407 whitespace in HTML output.
19408+ Build process is now compatible with both GHC 6.8 and GHC 6.6.
19409 GHC and GHC_PKG environment variables may be used to specify
19410 which version of the compiler to use, when multiple versions are
19413## pandoc 0.44 (2007-09-03)
19415 [ John MacFarlane ]
19417 * Fixed bug in HTML writer: when --toc was used, anchors were put around
19418 headers, which is invalid XHTML (block content within inline element).
19419 Now the anchors are put inside the header tags. Resolves Issue #23.
19421 * Added xmlns attribute to html element in html writer tests.
19422 This attribute is added by more recent versions of the
19423 xhtml library (>= 3000), and is required for valid XHTML.
19427 * On configure, compile 'Setup.hs' to 'setup' and use 'setup' as the build
19428 command instead of 'runhaskell', which, on some platforms (such as s390,
19429 alpha, m68k), throws the following error:
19431 runhaskell Setup.hs configure --prefix=/usr
19432 ghc-6.6.1: not built for interactive use
19434 This causes a serious FTBFS bug. Closes: #440668.
19436## pandoc 0.43 (2007-09-02)
19438 [ John MacFarlane ]
19440 * The focus of this release is performance. The markdown parser
19441 is about five times faster than in 0.42, based on benchmarks
19442 with the TextMate manual.
19444 * Main.hs: Replaced CRFilter and tabFilter with single function
19445 tabFilter, which operates on the whole string rather than breaking
19446 it into lines, and handles dos-style line-endings as well as tabs.
19448 * Added separate LaTeX reader and native reader tests; removed
19451 * Text.Pandoc.Shared:
19453 + Removed tabsToSpaces and tabsInLine (they were used only in Main.hs.)
19454 + General code cleanup (to elimante warnings when compiling with -Wall.)
19455 + Added 'wrapped' function, which helps wrap text into paragraphs,
19456 using the prettyprinting library.
19457 + Rewrote charsInBalanced and charsInBalanced'.
19458 - Documented restriction: open and close must be distinct characters.
19459 - Rearranged options for greater efficiency.
19460 - Bug fix: Changed inner call to charsInBalanced inside
19461 charsInBalanced' to charsInBalanced'.
19462 + anyLine now requires that the line end with a newline (not eof).
19463 This is a harmless assumption, since we always add newlines to the
19464 end of a block before parsing with anyLine, and it yields a 10% speed
19466 + Removed unnecessary 'try' in anyLine.
19467 + Removed unneeded 'try' from romanNumeral parser.
19468 + Use notFollowedBy instead of notFollowedBy' in charsInBalanced.
19469 + Removed unneeded 'try' in parseFromString.
19470 + Removed unneeded 'try' from stringAnyCase. (Now it behaves
19472 + Changed definition of 'enclosed' in Text.Pandoc.Shared so that
19473 'try' is not automatically applied to the 'end' parser. Added
19474 'try' in calls to 'enclosed' where needed. Slight speed increase.
19478 + Replaced individual wrapping routines in RST, Man, and Markdown
19479 writers with 'wrapped' from Text.Pandoc.Shared.
19480 + Rewrote LaTeX writer to use the prettyprinting library,
19481 so we get word wrapping, etc.
19482 + Modified latex writer tests for new latex writer using prettyprinter.
19483 + Fixed bug in LaTeX writer: autolinks would not cause
19484 `\usepackage{url}` to be put in the document header. Also, changes
19485 to state in enumerated list items would be overwritten.
19486 + In Markdown writer, escape paragraphs that begin with ordered list
19487 markers, so they don't get interpreted as ordered lists.
19489 * Text.Pandoc.Reades.LaTeX:
19491 + Fixed bug in LaTeX reader, which wrongly assumed that the roman
19492 numeral after "enum" in "setcounter" would consist entirely of
19493 "i"s. 'enumiv' is legitimate.
19494 + LaTeX command and environment names can't contain numbers.
19495 + Rearranged order of parsers in inline for slight speed improvement.
19496 + Added '`' to special characters and 'unescapedChar'.
19498 * Text.Pandoc.Readers.RST:
19500 + Removed unneeded try's in RST reader; also minor code cleanup.
19502 + Rearranged parsers in inline (doubled speed).
19504 * Text.Pandoc.Readers.Markdown:
19506 + Skip notes parsing if running in strict mode. (This yields a nice
19507 speed improvement in strict mode.)
19508 + Simplify autolink parsing code, using Network.URI to test for
19509 URIs. Added dependency on network library to debian/control and
19511 + More perspicuous definition of nonindentSpaces.
19512 + Removed unneeded 'try' in 'rawLine'.
19513 + Combined linebreak and whitespace into a new whitespace parser, to
19514 avoid unnecessary reparsing of space characters.
19515 + Removed unnecessary 'try' in 'codeBlock', 'ellipses', 'noteMarker',
19516 'multilineRow', 'dashedLine', 'rawHtmlBlocks'.
19517 + Use lookAhead in parsers for setext headers and definition lists
19518 to see if the next line begins appropriately; if not, don't waste
19519 any more time parsing.
19520 + Don't require blank lines after code block. (It's sufficient to
19521 end code block with a nonindented line.)
19522 + Changed definition of 'emph': italics with '_' must not
19523 be followed by an alphanumeric character. This is to help
19524 prevent interpretation of e.g. `[LC_TYPE]: my_type` as
19525 `[LC<em>TYPE]:my</em>type`.
19526 + Improved Markdown.pl-compatibility in referenceLink: the two parts
19527 of a reference-style link may be separated by one space, but not
19528 more... [a] [link], [not] [a link].
19529 + Fixed markdown inline code parsing so it better accords with
19530 Markdown.pl: the marker for the end of the code section is a clump
19531 of the same number of `'s with which the section began, followed
19532 by a non-` character. So, for example,
19533 ` h ``` i ` -> `<code>h ``` i</code>`.
19534 + Split 'title' into 'linkTitle' and 'referenceTitle', since the
19535 rules are slightly different.
19536 + Rewrote 'para' for greater efficiency.
19537 + Rewrote link parsers for greater efficiency.
19538 + Removed redundant 'referenceLink' in definition of inline (it's
19539 already in 'link').
19540 + Refactored escapeChar so it doesn't need 'try'.
19541 + Refactored hrule for performance in Markdown reader.
19542 + More intelligent rearranging of 'inline' so that most frequently
19543 used parsers are tried first.
19544 + Removed tabchar parser, as whitespace handles tabs anyway.
19546 * Text.Pandoc.CharacterReferences:
19549 + Removed unnecessary 'try's for a speed improvement.
19550 + Removed unnecessary '&' and ';' from the entity table.
19554 + Makefile: Get VERSION from cabal file, not Main.hs.
19555 + Modified MacPorts Portfile:
19556 - Depend on haddock
19557 - Build and install libraries and library documentation in
19558 addition to pandoc executable
19559 - Added template item for md5 sum in Portfile.in.
19560 - Incorporated changes from MacPorts repository (r28278).
19561 + FreeBSD port: Don't try to generate distinfo in Makefile.
19562 It can be made using 'make makesum' in FreeBSD.
19563 + Make both freebsd and macports targets depend on tarball.
19565 * Website and documentation:
19567 + Updated INSTALL instructions.
19568 + Added pandocwiki demo to website.
19569 + Removed local references to Portfile, since pandoc is now in the
19570 MacPorts repository.
19572## pandoc 0.42 (2007-08-26)
19574 [ John MacFarlane ]
19576 * Main.hs: Use utf8 conversion on the extra files loaded with
19577 the -H, -C, -B, and -A options. This fixes problems with unicode
19578 characters in these files.
19580 * Exposed Text.Pandoc.ASCIIMathML, since it is imported in
19581 Text.Pandoc.Readers.HTML and without it we get a linking error when
19586 + Added new rule for enhanced markdown ordered lists: if the list
19587 marker is a capital letter followed by a period (including a
19588 single-letter capital roman numeral), then it must be followed by
19589 at least two spaces. The point of this is to avoid accidentally
19590 treating people's initials as list markers: a paragraph might begin,
19591 "B. Russell was an English philosopher," and this shouldn't be
19592 treated as a list. Documented change in README.
19593 + Blocks that start with "p. " and a digit are no longer treated
19594 as ordered lists (it's a page number).
19595 + Added a needed 'try' to listItem.
19596 + Removed check for a following setext header in endline.
19597 A full test is too inefficient (doubles benchmark time), and the
19598 substitute we had before is not 100% accurate.
19599 + Don't use Code elements for autolinks if --strict specified.
19601 * LaTeX writer: When a footnote ends with a Verbatim environment, the
19602 close } of the footnote cannot occur on the same line or an error occurs.
19603 Fixed this by adding a newline before the closing } of every footnote.
19606 + Removed incorrect "{}" around style information in HTML tables.
19607 Column widths now work properly in HTML.
19608 + If --strict option is specified (and --toc is not), don't include
19609 identifiers in headers, for better Markdown compatibility.
19613 + Separated $(web_dest) and website targets.
19614 + In website, index.txt is now constructed from template index.txt.in.
19615 + Added freebsd target to Markefile. This creates the freebsd Makefile
19616 from Makefile.in, and creates distinfo. Removed Makefile and distinfo
19617 from the repository.
19618 + Added macport target to Makefile. Portfile is built from template
19620 + Removed OSX package targets. (Too many difficulties involving
19621 dependencies on dynamic libraries.)
19622 + More complete INSTALL instructions for all architectures.
19625 + Added a programming demo, pandocwiki.
19629 * Do not forget to close pandoc's ITP. Closes: #391666
19631## pandoc 0.41 (2007-08-19)
19633 [ John MacFarlane ]
19635 * Fixed bugs in HTML reader:
19636 + Skip material at end *only if* `</html>` is present (previously,
19637 only part of the document would be parsed if an error was
19638 found; now a proper error message is given).
19639 + Added new constant eitherBlockOrInline with elements that may
19640 count either as block-level or as inline. Modified isInline and
19641 isBlock to take this into account.
19642 + Modified rawHtmlBlock to accept any tag (even an inline tag):
19643 this is innocuous, because rawHtmlBlock is tried only if a regular
19644 inline element can't be parsed.
19645 + Added a necessary 'try' in definition of 'para'.
19647 * Fixed bug in markdown ordered list parsing. The problem was that
19648 anyOrderedListStart did not check for a space following the
19649 ordered list marker. So in 'A.B. 2007' the parser would be
19650 expecting a list item, but would not find one, causing an error.
19651 Fixed a similar bug in the RST reader. Resolves Issue #22.
19653 * Refactored RST and Markdown readers using parseFromString.
19655 * LaTeX reader will now skip anything after \end{document}.
19657 * Fixed blockquote output in markdown writer: previously, block
19658 quotes in indented contexts would be indented only in the first
19661 * Added note to INSTALL about variations in versions of the xhtml
19662 library that can lead to failed tests (thanks to Leif LeBaron).
19664## pandoc 0.4 (2007-01-16)
19666 [ John MacFarlane ]
19668 * Added two new output formats: groff man pages and ConTeXt. By
19669 default, output files with extensions ".ctx" and ".context" are
19670 assumed to be ConTeXt, and output files with single-digit extensions
19671 are assumed to be man pages.
19673 * Enhanced ordered lists (documented in README, under Lists):
19674 + The OrderedList block element now stores information about
19675 list number style, list number delimiter, and starting number.
19676 + The readers parse this information when possible.
19677 + The writers use this information to style ordered lists.
19678 + The enhancement can be disabled using the --strict option.
19680 * Added support for tables (with a new Table block element). Two kinds
19681 of tables are supported: a simple table with one-line rows, and a
19682 more complex variety with multiline rows. All output formats are
19683 supported, but only markdown tables are parsed at the moment. The
19684 syntax is documented in README.
19686 * Added support for definition lists (with a new DefinitionList block
19687 element). All output and input formats are supported. The syntax is
19688 documented in README.
19690 * Added support for superscripts and subscripts (with new Superscript
19691 and Subscript inline elements). All input and output
19692 formats. The syntax is documented in README.
19694 * Added support for strikeout (with a new Strikeout inline element).
19695 All input and output formats are supported. Thanks to Bradley Kuhn,
19696 who contributed a patch. The syntax is documented in README. Resolves
19699 * Added a --toc|--table-of-contents option. This causes an automatically
19700 generated table of contents (or an instruction that creates one) to
19701 be inserted at the beginning of the document. Not supported in S5,
19702 DocBook, or man page writers.
19704 * Modified the -m|--asciimathml option:
19706 + If an optional URL argument is provided, a link is inserted
19707 instead of the contents of the ASCIIMathML.js script.
19708 + Nothing is inserted unless the document actually contains
19711 * Removed Blank block element as unnecessary.
19713 * Removed Key and Note blocks from the Pandoc data structure. All
19714 links are now stored as explicit links, and note contents are
19715 stored with the (inline) notes.
19717 + All link Targets are now explicit (URL, title) pairs; there
19718 is no longer a 'Ref' target.
19719 + Markdown and RST parsers now need to extract data from key and
19720 note blocks and insert them into the relevant inline elements.
19721 Other parsers have been simplified, since there is no longer any need
19722 to construct separate key and note blocks.
19723 + Markdown, RST, and HTML writers need to construct lists of
19724 notes; Markdown and RST writers need to construct lists of link
19725 references (when the --reference-links option is specified); and
19726 the RST writer needs to construct a list of image substitution
19727 references. All writers have been rewritten to use the State monad
19728 when state is required.
19729 + Several functions (generateReference, keyTable,
19730 replaceReferenceLinks, replaceRefLinksBlockList, and some auxiliaries
19731 used by them) have been removed from Text.Pandoc.Shared, since
19732 they are no longer needed. New functions and data structures
19733 (Reference, isNoteBlock, isKeyBlock, isLineClump) have been
19734 added. The functions inTags, selfClosingTag, inTagsSimple, and
19735 inTagsIndented have been moved to the DocBook writer, since that
19736 is now the only module that uses them. NoteTable is now exported
19737 in Text.Pandoc.Shared.
19738 + Added stateKeys and stateNotes to ParserState; removed stateKeyBlocks,
19739 stateKeysUsed, stateNoteBlocks, stateNoteIdentifiers, stateInlineLinks.
19740 + Added writerNotes and writerReferenceLinks to WriterOptions.
19742 * Added Text.Pandoc module that exports basic readers, writers,
19743 definitions, and utility functions. This should export everything
19744 needed for most uses of Pandoc libraries. The haddock documentation
19745 includes a short example program.
19747 * Text.Pandoc.ASCIIMathML is no longer an exported module.
19749 * Added Text.Pandoc.Blocks module to help in printing markdown
19750 and RST tables. This module provides functions for working with
19751 fixed-width blocks of text--e.g., placing them side by side, as
19754 * Refactored to avoid reliance on Haskell's Text.Regex library, which
19755 (a) is slow, and (b) does not properly handle unicode. This fixed
19756 some strange bugs, e.g. in parsing S-cedilla, and improved performance.
19758 + Replaced 'gsub' with a general list function 'substitute'
19759 that does not rely on Text.Regex.
19760 + Rewrote extractTagType in HTML reader so that it doesn't use
19762 + In Markdown reader, replaced email regex test with a custom email
19763 autolink parser (autoLinkEmail). Also replaced selfClosingTag regex
19764 with a custom function isSelfClosingTag.
19765 + Modified Docbook writer so that it doesn't rely on Text.Regex for
19766 detecting 'mailto' links.
19767 + Removed escapePreservingRegex and reamped entity-handling
19768 functions in Text.Pandoc.Shared and Text.Pandoc.CharacterReferences to
19769 avoid reliance on Text.Regex (see below on character reference
19772 * Renamed Text.Pandoc.Entities as Text.Pandoc.CharacterReferences.
19774 * Changed handling of XML entities. Entities are now parsed (and unicode
19775 characters returned) in the Markdown and HTML readers, rather than being
19776 handled in the writers. In HTML and Docbook writers, UTF-8 is now used
19777 instead of entities for characters above 128. This makes the HTML and
19778 DocBook output much more readable and more easily editable.
19780 + Removed sgmlHexEntity, sgmlDecimalEntity, sgmlNamedEntity, and
19781 sgmlCharacterEntity regexes from Text.Pandoc.Shared.
19782 + Renamed escapeSGMLChar to escapeCharForXML. Added escapeStringForXML.
19783 Moved both functions to Text.Pandoc.Writers.Docbook.
19784 + Added characterReference parser to Text.Pandoc.CharacterReferences.
19785 This parses a string and return a unicode character.
19786 + Rewrote decodeCharacterReferences to use the new parser instead of
19788 + Added new charRef parser for Markdown and HTML, which replaces the
19789 old 'entity' parser. Added '&' as a special character in Markdown reader.
19790 + Modified HTML and Markdown readers to call decodeEntities on all raw
19791 strings (e.g. authors, dates, link titles), to ensure that no
19792 unprocessed entities are included in the native representation of
19793 the document. (In the HTML reader, most of this work is done by a
19794 change in extractAttributeName.)
19795 + In XML and Markdown output, escape unicode nonbreaking space as ' ',
19796 since a unicode non-breaking space is impossible to distinguish visually
19797 from a regular space. (Resolves Issue #3.)
19798 + Removed encodeEntitiesNumerical.
19799 + Use Data.Map for entityTable and (new) reverseEntityTable, for a
19800 slight performance boost over the old association list.
19801 + Removed unneeded decodeEntities from 'str' parser in HTML and
19804 * Text.Pandoc.UTF8: Renamed encodeUTF8 to toUTF8, decodeUTF8 to
19805 fromUTF8, for clarity.
19807 * Replaced old haskell98 module names replaced by hierarchical module
19808 names, e.g. List by Data.List. Removed haskell98 from dependencies
19809 in pandoc.cabal, and added mtl (needed for state monad). Substituted
19812 * Refactored and cleaned up character escaping in writers, using
19813 backslashEscapes and escapeStringUsing functions.
19815 * Instead of adding `\n\n` to the end of an input string in Main.hs,
19816 this is now done in the readers. This makes the libraries behave
19817 the way you'd expect from the pandoc program. Resolves Issue #10.
19819 * URLs and email addresses in autolinks are now typeset as Code.
19821 * In Main.hs, changed putStr to putStrLn -- mainly because MacOS X
19822 doesn't display the whole output unless there's a line ending.
19824 * Major code cleanup in all modules, for greater consistency, concision,
19829 + Fixed several bugs (extractTagType, attribute parsing).
19830 + Remove Null blocks in lists of blocks when possible.
19831 + Allow HTML comments as raw HTML inline.
19835 + Ordered list items may no longer begin with uppercase letters, or
19836 letters greater than 'n'. (This prevents first initials and page
19837 reference, e.g. 'p. 400', from being parsed as beginning lists.)
19838 Also, numbers beginning list items may no longer end with ')',
19839 which is now allowed only after letters. Note: These changes
19840 may cause documents to be parsed differently. Users should take
19842 + Changed autoLink parsing to conform better to Markdown.pl's
19843 behavior. `<google.com>` is not treated as a link, but
19844 `<http://google.com>`, `<ftp://google.com>`, and
19845 `<mailto:google@google.com>` are.
19846 + Cleaned up handling of embedded quotes in link titles. Now these are
19847 stored as a '"' character, not as '"'.
19848 + Use lookAhead parser for the 'first pass' (looking for reference keys),
19849 instead of parsing normally, then using setInput to reset input. This
19850 yields a slight performance boost.
19851 + Fixed several bugs in smart quote recognition.
19852 + Fixed bug in indentSpaces (which didn't properly handle
19853 cases with mixed spaces and tabs).
19854 + Consolidated 'text', 'special', and 'inline' into 'inline'.
19855 + Fixed bug which allowed URL and title to be separated by multiple blank
19856 lines in links and reference keys. They can be on separate lines but
19857 can't have blank lines between them.
19858 + Correctly handle bracketed text inside inline footnotes and links,using
19859 new function inlinesInBalanced. Resolves Issue #14.
19860 + Fixed bug in footnotes: links in footnotes were not being
19861 processed. Solution: three-stage parse. First, get all the
19862 reference keys and add information to state. Next, get all the
19863 notes and add information to state. (Reference keys may be needed
19864 at this stage.) Finally, parse everything else.
19865 + Replaced named constants like 'emphStart' with literals.
19866 + Removed an extra occurrence of escapedChar in definition of inline.
19870 + Allow the URI in a RST hyperlink target to start on the line
19871 after the reference key.
19872 + Added 'try' in front of 'string', where needed, or used a different
19873 parser. This fixes a bug where ````` would not be correctly parsed as
19875 + Fixed slow performance in parsing inline literals in RST reader. The
19876 problem was that ``#`` was seen by 'inline' as a potential link or image.
19877 Fix: inserted 'notFollowedBy (char '`')' in link parsers.
19879 + Use lookAhead instead of getInput/setInput in RST reader. Removed
19880 unneeded getState call, since lookAhead automatically saves and
19881 restores the parser state.
19882 + Allow hyperlink target URIs to be split over multiple lines, and
19883 to start on the line after the reference. Resolves Issue #7.
19884 + Fixed handling of autolinks.
19888 + Replaced `choice [(try (string ...), ...]` idiom with `oneOfStrings`,
19890 + Added clauses for tilde and caret. Tilde is `\ensuremath{\sim}`, and
19891 caret is `\^{}`, not `\^` as before.
19892 + Added parsing for `\url`.
19893 + Parse `\texttt{}` as code, provided there's nothing fancy inside.
19897 + Modified HTML writer to use the Text.XHtml library. This results
19898 in cleaner, faster code, and it makes it easier to use Pandoc in
19899 other projects, like wikis, which use Text.XHtml. Two functions are
19900 now provided, writeHtml and writeHtmlString: the former outputs an
19901 Html structure, the latter a rendered string. The S5 writer is also
19902 changed, in parallel ways (writeS5, writeS5String).
19903 + The Html header is now written programmatically, so it has been
19904 removed from the 'headers' directory. The S5 header is still
19905 needed, but the doctype and some of the meta declarations have
19906 been removed, since they are written programmatically. This change
19907 introduces a new dependency on the xhtml package.
19908 + Fixed two bugs in email obfuscation involving improper escaping
19909 of '&' in the `<noscript>` section and in `--strict` mode. Resolves
19911 + Fixed another bug in email obfuscation: If the text to be obfuscated
19912 contains an entity, this needs to be decoded before obfuscation.
19913 Thanks to thsutton for the patch. Resolves Issue #15.
19914 + Changed the way the backlink is displayed in HTML footnotes.
19915 Instead of appearing on a line by itself, it now generally
19916 appears on the last line of the note. (Exception: when the
19917 note does not end with a Plain or Para block.) This saves space
19919 + Added automatic unique identifiers to headers:
19920 - The identifier is derived from the header via a scheme
19921 documented in README.
19922 - WriterState now includes a list of header identifiers and a table
19923 of contents in addition to notes.
19924 - The function uniqueIdentifiers creates a list of unique identifiers
19925 from a list of inline lists (e.g. headers).
19926 - This list is part of WriterState and gets consumed by blockToHtml
19927 each time a header is encountered.
19928 + Include CSS for .strikethrough class in header only if strikethrough
19929 text appears in the document.
19930 + If the 'strict' option is specified, elements that do not appear in
19931 standard markdown (like definition lists) are passed through as
19933 + Simplified treatment of autolinks, using pattern matching instead of
19938 + Links in markdown output are now printed as inline links by default,
19939 rather than reference links. A --reference-links option has been added
19940 that forces links to be printed as reference links. Resolves Issue #4.
19941 + Use autolinks when possible. Instead of `[site.com](site.com)`,
19946 + Rewrote to use the State monad. The preamble now includes only those
19947 packages that are actually required, given the document's content.
19948 Thus, for example, if strikeout is not used, ulem is not required.
19949 Modified LaTeXHeader accordingly.
19950 + Modified LaTeX writer to insert `\,` between consecutive quotes.
19951 + Removed unused function tableRowColumnWidths.
19952 + Simplified code for escaping special characters.
19953 + Leave extra blank line after `\maketitle`.
19954 + Include empty `\author{}` when no author specified to avoid LaTeX
19956 + Include fancyvrb code in header only if needed -- that is, only
19957 if there is actually code in a footnote.
19958 + Use `\url{}` for autolinks.
19959 + Include [mathletters] option in ucs package, so that basic unicode
19960 Greek letters will work correctly.
19962 * RST writer: Force blank line before lists, so that sublists will
19963 be handled correctly.
19965 * Docbook writer: Fixed a bug: email links with text, like
19966 [foo](me@bar.baz), were being incorrectly treated as autolinks.
19968 * Removed Text.ParserCombinators.Pandoc and moved all its functions to
19969 Text.Pandoc.Shared.
19971 * Text.Pandoc.Shared:
19973 + Added defaultWriterOptions.
19974 + Added writerTableOfContents to WriterOptions.
19975 + Added writerIgnoreNotes option to WriterOptions. This is needed
19976 for processing header blocks for a table of contents, since notes on
19977 headers should not appear in the TOC.
19978 + Added prettyprinting for native Table format.
19979 + Removed some unneeded imports.
19980 + Moved escape and nullBlock parsers from
19981 Text.ParserCombinators.Pandoc, since the latter is for
19982 general-purpose parsers that don't depend on Text.Pandoc.Definition.
19983 + Moved isHeaderBlock from Text.Pandoc.Writers.HTML.
19984 + Moved Element, headerAtLeast, and hierarchicalize from Docbook
19985 writer, because HTML writer now uses these in constructing a table
19987 + Added clauses for new inline elements (Strikeout, Superscript,
19988 Subscript) to refsMatch.
19989 + Removed backslashEscape; added new functions escapeStringUsing and
19991 + Moved failIfStrict from markdown reader, since it is now used also
19992 by the HTML reader.
19993 + Added a 'try' to the definition of indentSpaces.
19994 + In definition of 'reference', added check to make sure it's not a note
19996 + Added functions: camelCaseToHyphenated, toRomanNumeral,
19997 anyOrderedListMarker, orderedListmarker, orderedListMarkers,
19998 charsInBalanced', withHorizDisplacement, romanNumeral
19999 + Fixed a bug in the anyLine parser. Previously it would parse an empty
20000 string "", but it should fail on an empty string, or we get an error
20001 when it is used inside "many" combinators.
20002 + Removed followedBy' parser, replacing it with the lookAhead parser from
20004 + Added some needed 'try's before multicharacter parsers, especially in
20006 + Removed the 'try' from the 'end' parser in 'enclosed', so that
20007 'enclosed' behaves like 'option', 'manyTill', etc.
20008 + Added lineClump parser, which parses a raw line block up to and
20009 including any following blank lines.
20010 + Renamed parseFromStr to parseFromString.
20011 + Added a 'try' to the 'end' parser in 'enclosed'. This makes errors in
20012 the use of 'enclosed' less likely. Removed some now-unnecessary 'try's
20014 + Removed unneeded 'try' in blanklines.
20015 + Removed endsWith function and rewrote calling functions to use
20016 isSuffixOf instead.
20017 + Added >>~ combinator.
20018 + Fixed bug in normalizeSpaces: Space:Str "":Space should compress to
20021 * Refactored runtests.pl; added separate tests for tables.
20025 + Added -asxhtml flag to tidy in html2markdown. This will
20026 perhaps help the parser, which expects closing tags.
20027 + Modified markdown2pdf to run pdflatex a second time if --toc or
20028 --table-of-contents was specified; otherwise the table of
20029 contents won't appear.
20030 + Modified markdown2pdf to print a helpful message if the 'ulem'
20031 LaTeX package is required and not found.
20033 * Changes to build process:
20035 + Dropped support for compilation with GHC 6.4. GHC 6.6 or higher
20037 + Removed cabalize and Pandoc.cabal.in. The repository now contains
20038 pandoc.cabal itself.
20039 + Pandoc.cabal has been changed to pandoc.cabal, because HackageDB
20040 likes the cabal file to have the same name as the tarball.
20041 + Expanded and revised the package description in pandoc.cabal.
20042 Revised the package synopsis.
20043 + The tarball built by 'make tarball' now contains files built from
20044 templates (including man pages and shell scripts), so pandoc can
20045 be built directly using Cabal tools, without preprocessing.
20046 + Executable binaries are now stripped before installing.
20047 + Man pages are now generated from markdown sources, using pandoc's
20049 + Use HTML version of README (instead of RTF) in Mac OS X installer.
20050 + Instead of testing for the existence of a pandoc symlink in build-exec,
20055 + Updated README and man pages with information on new features.
20056 + Updated INSTALL instructions with some useful clarifications and
20058 + Updated web content.
20060 * Added FreeBSD port.
20066 + Changed pandoc's Build-Depends to include libghc6-mtl-dev and
20067 libghc6-xhtml-dev. Removed libghc6-html-dev.
20068 + Suggest texlive-latex-recommended | tetex-extra instead of
20069 tetex-bin. This brings in fancyvrb and unicode support.
20072## pandoc 0.3 (2007-01-05)
20074 [ John MacFarlane ]
20076 * Changes in pandoc options:
20078 + Allow options to follow or precede arguments.
20079 + Changed '--smartypants' to '--smart' and adjusted symbols accordingly.
20080 + Added '--strict' option.
20081 + Added '-o/--output' option.
20082 + Added '--dump-args' and '--ignore-args' options (for use in wrappers).
20083 + Modified '-v' and '-h' output to go to STDERR, not STDOUT, and return
20084 error conditions. This is helpful for writing wrappers.
20085 + Added copyright message to '-v' output, modeled after FSF messages.
20086 + Reformatted usage message so that it doesn't wrap illegibly.
20087 + Removed extra blanks after '-h' and '-D' output.
20089 * Added docbook writer.
20091 * Added implicit setting of default input and output format based
20092 on input and output filename extensions. These defaults are
20093 overridden if explicit input and output formats are specified using
20094 '-t', '-f', '-r', or '-w' options. Documented in pandoc(1) man page
20097 * Allow ordered list items to begin with (single) letters, as well
20098 as numbers. The list item marker may now be terminated either by
20099 '.' or by ')'. This extension to standard markdown is documented
20102 * Revised footnote syntax. (See README for full details.) The
20103 '[^1]' format now standard in markdown extensions is supported,
20104 as are inline footnotes with this syntax: `^[My note.]`.
20105 The earlier footnote syntax `^(1)` is no longer supported.
20107 * Improved HTML representation of footnotes. All footnotes
20108 are now auto-numbered and appear in an ordered list at the
20109 end of the HTML document. Since the default appearance is now
20110 acceptable, the old footnote styles have been removed from the
20115 + Fixed a serious bug in the markdown, LaTeX, and RST readers.
20116 These readers ran 'runParser' on processed chunks of text to handle
20117 embedded block lists in lists and quotation blocks. But then
20118 any changes made to the parser state in these chunks was lost,
20119 as the state is local to the parser. So, for example, footnotes
20120 didn't work in quotes or list items. The fix: instead of calling
20121 runParser on some raw text, use setInput to make it the input, then
20122 parse it, then use setInput to restore the input to what it was
20123 before. This is shorter and more elegant, and it fixes the problem.
20124 + Fixed bug in notFollowedBy' combinator (adding 'try' before
20125 'parser'). Adjusted code that uses this combinator accordingly.
20126 + Fixed bug in RTF writer that caused improper indentation on
20127 footnotes occurring in indented blocks like lists.
20128 + Fixed parsing of metadata in LaTeX reader. Now the title, author,
20129 and date are parsed correctly. Everything else in the preamble
20131 + Modified escapedChar in LaTeX reader to allow a `\` at the end of a
20132 line to count as escaped whitespace.
20133 + Modified LaTeX reader to produce inline links rather than reference
20134 links. Otherwise, links in footnotes aren't handled properly.
20135 + Fixed handling of titles in links in Markdown reader, so that
20136 embedded quotation marks are now handled properly.
20137 + Fixed Markdown reader's handling of embedded brackets in links.
20138 + Fixed Markdown reader so that it only parses bracketed material
20139 as a reference link if there is actually a corresponding key.
20140 + Revised inline code parsing in Markdown reader to conform to
20141 markdown standard. Now any number of `s can begin inline code,
20142 which will end with the same number of `s. For example, to
20143 have two backticks as code, write ``` `` ```. Modified Markdown
20144 writer accordingly.
20145 + Fixed bug in text-wrapping routine in Markdown and RST writers.
20146 Now LineBreaks no longer cause wrapping problems.
20147 + Supported hexadecimal numerical entity references as well as
20149 + Fixed bug in Markdown reader's handling of underscores and other
20150 inline formatting markers inside reference labels: for example,
20151 in '[A_B]: /url/a_b', the material between underscores was being
20152 parsed as emphasized inlines.
20153 + Changed Markdown reader's handling of backslash escapes so that
20154 only non-alphanumeric characters can be escaped. Strict mode
20155 follows Markdown.pl in only allowing a select group of punctuation
20156 characters to be escaped.
20157 + Modified HTML reader to skip a newline following a `<br>` tag.
20158 Otherwise the newline will be treated as a space at the beginning
20161 * Made handling of code blocks more consistent. Previously, some
20162 readers allowed trailing newlines, while others stripped them.
20163 Now, all readers strip trailing newlines in code blocks. Writers
20164 insert a newline at the end of code blocks as needed.
20166 * Modified readers to make spacing at the end of output more consistent.
20168 * Minor improvements to LaTeX reader:
20170 + `\thanks` now treated like a footnote.
20171 + Simplified parsing of LaTeX command arguments and options.
20172 commandArgs now returns a list of arguments OR options (in
20173 whatever order they appear). The brackets are included, and
20174 a new stripFirstAndLast function is provided to strip them off
20175 when needed. This fixes a problem in dealing with \newcommand
20176 and \newenvironment.
20178 * Revised RTF writer:
20180 + Default font is now Helvetica.
20181 + An `\f0` is added to each `\pard`, so that font resizing works
20184 * Moved handling of "smart typography" from the writers to the Markdown
20185 and LaTeX readers. This allows great simplification of the writers
20186 and more accurate smart quotes, dashes, and ellipses. DocBook can
20187 now use `<quote>`. The '--smart' option now toggles an option in
20188 the parser state rather than a writer option. Several new kinds
20189 of inline elements have been added: Quoted, Ellipses, Apostrophe,
20192 * Changes in HTML writer:
20194 + Include title block in header even when title is null.
20195 + Made javascript obfuscation of emails even more obfuscatory,
20196 by combining it with entity obfuscation.
20198 * Changed default ASCIIMathML text color to black.
20202 + Added --strip-trailing-cr option to diff in runtests.pl, for
20203 compatibility with Windows.
20204 + Added regression tests with footnotes in quote blocks and lists.
20206 * Makefile changes:
20208 + osx-pkg target creates a Mac OS X package (directory). New osx
20209 directory contains files needed for construction of the package.
20210 + osx-dmg target creates a compressed disk image containing the package.
20211 + win-pkg target creates Windows binary package.
20212 + tarball target creates distribution source tarball.
20213 + website target generates pandoc's website automatically, including
20214 demos. New 'web' directory contains files needed for construction
20215 of the website (which will be created as the 'pandoc' subdirectory
20217 + Makefile checks to see if we're running Windows/Cygwin; if so,
20218 a '.exe' extension is added to each executable in EXECS.
20220 * Removed all wrappers except markdown2pdf and html2markdown.
20222 * Added new wrapper hsmarkdown, to be used as a drop-in replacement
20223 for Markdown.pl. hsmarkdown calls pandoc with the '--strict'
20224 option and disables other options.
20226 * Added code to html2markdown that tries to determine the character
20227 encoding of an HTML file, by parsing the "Content-type" meta tag.
20229 + If the encoding can't be determined, then if the content is local,
20230 the local encoding is used; if it comes from a URL, UTF-8 is used
20232 + If input is from STDIN, don't try to determine character encoding.
20233 + Encoding can be specified explicitly using '-e' option.
20235 * Improved warning messages in wrappers:
20237 + Print warning if iconv not available
20238 + More user-friendly error messages in markdown2pdf, when
20243 + Renamed 'Text/Pandoc/HtmlEntities' module to
20244 'Text/Pandoc/Entities'. Also changed function names so as
20245 not to be HTML-specific.
20246 + Refactored SGML string escaping functions from HTML and Docbook
20247 writers into Text/Pandoc/Shared. (escapeSGML, stringToSGML)
20248 + Removed 'BlockQuoteContext' from ParserContext, as it isn't
20250 + Removed splitBySpace and replaced it with a general, polymorphic
20252 + Refactored LaTeX reader for clarity (added isArg function).
20253 + Converted some CR's to LF's in src/ui/default/print.css.
20254 + Added license text to top of source files.
20255 + Added module data for haddock to source files.
20256 + Reformatted code for consistency.
20258 * Rewrote documentation and man pages. Split README into INSTALL
20261 * Split LICENSE into COPYING and COPYRIGHT.
20263 * Removed TODO, since we now maintain ToDo on the wiki.
20265 * Made COPYRIGHT in top level a symlink to debian/copyright, to avoid
20270 * Revamped build process to conform to debian standards and created
20271 a proper debian package. Closes: #391666.
20273 * Modified build process to support GHC 6.6.
20275 + The package can still be compiled using GHC 6.4.2, though because
20276 of dependencies the "make deb" target works only with GHC 6.6+.
20277 + The script 'cabalize' is used to create an appropriate
20278 'Pandoc.cabal' from 'Pandoc.cabal.in', depending on the GHC and
20281 * Refactored template processing (fillTemplates.pl).
20283 * Modified wrapper scripts to make them more robust and portable.
20284 To avoid code duplication and ensure consistency, wrappers are
20285 generated via a templating system from templates in src/wrappers.
20287 + Wrappers now accept multiple filenames, when appropriate.
20288 + Spaces and tabs allowed in filenames.
20289 + getopts shell builtin is used for portable option parsing.
20290 + Improved html2markdown's web grabber code, making it more robust,
20291 configurable and verbose. Added '-e', '-g' options.
20294## pandoc 0.2 (2006-08-14)
20296 * Fixed unicode/utf-8 translation
20298## pandoc 0.1 (2006-08-14)
20300 * Initial creation of debian package
keysesc to close
- j k
- move down, up
- enter
- open
- esc
- back, or up a level
- 1 2 3 4
- readme, files, commits, refs
- /
- filter
- y
- copy the clone command
- c
- compare the chosen commit with another (or shift-click both)
- ?
- show or hide this list