@@ -1,6 +1,3655 @@
1# Revision history for pandoc
1# Revision history for pandoc
3−## pandoc 2.8 PROVISIONAL (YYYY-MM-DD)
3+## pandoc 2.14.0.3 (2021-06-22)
5+ * Text.Pandoc.MediaBag `insertMediaBag`: ensure we get a sane mediaPath
6+ for URLs (#7391). In earlier 2.14.x versions, we'd get
7+ incorrect paths for resources downloaded from URLs when the
8+ media are extracted (including in PDF production).
9+ * Text.Pandoc.Parsing: improve `emailAddress` (#7398).
10+ Previously the parser would accept characters in domains
11+ that are illegal in domains, and this sometimes caused it
12+ to gobble bits of the following text.
13+ * txt2tags reader: modify the email address parser so
14+ it still includes form parameters, even after the change to
15+ `emailAddress` in Text.Pandoc.Parsing.
16+ * Text.Pandoc.Readers.Metadata: Fix regression with comment-only YAML
17+ metadata blocks (#7400).
18+ * reveal.js writer and template: better handling of options. Previously
19+ it was impossible to specify false values for options that default to
20+ true (e.g. `center`); setting the option to false just caused the portion
21+ of the template setting the option to be omitted. Now we prepopulate
22+ all the variables with their default values, including them all
23+ unconditionally and allowing them to be overridden.
24+ * Markdown writer: Fix regression in code blocks with attributes (#7397).
25+ Code blocks with a single class but nonempty attributes
26+ were having attributes drop as a result of #7242.
28+ + Add strut at end of minipage if it contains line breaks.
29+ Without them, the last line is not as tall as it should be in
31+ + Always use a minipage for cells with line breaks, when
32+ width information is available (#7393). Otherwise the way we treat them
33+ can lead to content that overflows a cell.
34+ + Use `\strut` instead of `~` before `\\` in empty line.
35+ * Use lts-18.0 stack resolver.
36+ * Require skylighting 0.10.5.2 (adding support for Swift).
37+ * Require commonmark 0.2.1.
38+ * Rephrase section on unsafe HTML in manual.
41+## pandoc 2.14.0.2 (2021-06-13)
43+ * Fix MediaBag regressions (#7345). iIn the 2.14 release `--extract-media`
44+ stopped working as before; there could be mismatches between the
45+ paths in the rendered document and the extracted media.
46+ This patch makes several changes that restore the earlier behavior
47+ (while keeping the same API). The `mediaPath` in 2.14 was always
48+ constructed from the SHA1 hash of the media contents. Now, we
49+ preserve the original path unless it's an absolute path or contains
50+ `..` segments (in that case we use a path based on the SHA1 hash of
53+ In Text.Pandoc.MediaBag, `mediaDirectory` and `mediaItems` now use the
54+ `mediaPath`, rather than the mediabag key, for the first component of the
55+ tuple. This makes more sense, I think, and fits with the documentation of
56+ these functions; eventually, though, we should rework the API so that
57+ `mediaItems` returns both the keys and the MediaItems.
59+ In Text.Pandoc.Class.IO, rewriting of source paths in `extractMedia` has
62+ In Text.Pandoc.Class.PandocMonad, `fillMediaBag` has been modified so that
63+ it doesn't modify image paths (that was part of the problem in #7345).
65+ We now do path normalization (e.g. `\` separators on Windows) in
70+ + Text.Pandoc.PDF: Fix regression in 2.14 for generation of PDFs with
72+ + Only print relevant part of environment on `--verbose`. Since
73+ `--verbose` output might be put in an issue, we want to avoid
74+ spilling out secrets in environment variables.
76+ * Markdown reader: fix pipe table regression in 2.11.4 (#7343).
77+ Previously pipe tables with empty headers (that is, a header
78+ line with all empty cells) would be rendered as headerless
79+ tables. This broke in 2.11.4. The fix here is to produce an
80+ AST with an empty table head when a pipe table has all empty
83+ * LaTeX reader: don't allow optional `*` on symbol control sequences
84+ (#7340). Generally we allow optional starred variants of LaTeX commands
85+ (since many allow them, and if we don't accept these explicitly,
86+ ignoring the star usually gives acceptable results). But we
87+ don't want to do this for `\(*\)` and similar cases.
89+ * Docx reader: handle absolute URIs in Relationship Target (#7374).
91+ * Docx writer: fix handling of empty table headers (Albert Krewinkel,
92+ #7369). A table header which does not contain any cells is now treated as
95+ * LaTeX writer: Fix regression in table header position (#7347).
96+ In recent versions the table headers were no longer bottom-aligned
97+ (if more than one line). This patch fixes that by using minipages
98+ for table headers in non-simple tables.
100+ * CommonMark writer:
102+ + Do not use simple class for fenced-divs (Jan Tojnar, amends #7242.)
103+ + Do not throw away attributes when `Ext_attributes` is enabled.
104+ `Ext_attributes` covers at least the following:
105+ `Ext_fenced_code_attributes`, `Ext_header_attributes`,
106+ `Ext_inline_code_attributes`, `Ext_link_attributes`.
110+ + Allow `pipe_tables` to be disabled for commonmark formats
111+ (`commonmark_x`, `gfm`) (#7375).
112+ + Re-use functions from Text.Pandoc.Markdown.Inline (Jan Tojnar).
114+ * DocBook writer: Remove non-existent admonitions (Jan Tojnar).
115+ `attention`, `error` and `hint` are reStructuredText specific.
117+ * HTML writer: Don't omit width attribute on div (#7342).
119+ * Text.Pandoc.MIME, `extensionFromMimeType`: add a few special cases.
120+ When we do a reverse lookup in the MIME table, we just get the
121+ last match, so when the same mime type is associated with several
122+ different extensions, we sometimes got weird results, e.g. `.vs`
123+ for `text/plain`. These special cases help us get the most standard
124+ extensions for mime types like `text/plain`.
126+ * Lua utils: fix handling of table headers in `from_simple_table` (Albert
127+ Krewinkel, #7369). Passing an empty list of header cells now results
128+ in an empty table header.
130+ * Text.Pandoc.Citeproc:
132+ + Avoid duplicate classes and attributes on references div.
133+ + Fix regression in citeproc processing (#7376). If inline
134+ references are used (in the metadata `references` field), we
135+ should still only include in the bibliography items that are
136+ actually cited (unless `nocite` is used).
138+ * Require citeproc 0.4.0.1. This fixes a bug which led to doubled
139+ "et al." in some (rare) circumstances.
143+ + Mention GladTeX for EPUB export (Sebastian Humenda).
144+ This updates the manual and the web site about the GladTeX usage.
145+ + More details and a useful link for YAML syntax.
147+ * CONTRIBUTING.md: update modules overview (Albert Krewinkel).
149+ * using-the-pandoc-api.md: switch from String to Text (Albert Krewinkel).
152+## pandoc 2.14.0.1 (2021-06-01)
154+ * Commonmark reader: Fix regression in 2.14 with YAML metdata block parsing,
155+ which could cause the document body to be omitted after metadata (#7339).
157+ * HTML reader: fix column width regression in 2.14 (#7334).
158+ Column widths specified with a style attribute were off by a factor of 100.
160+ * Markdown reader: in `rebasePaths`, check for both Windows and Posix
161+ absolute paths. Previously Windows pandoc was treating
162+ `/foo/bar.jpg` as non-absolute.
164+ * Text.Pandoc.Logging: In rendering `LoadedResource`, use relative paths.
166+ * Docx writer: fix regression on captions (#7328). The "Table Caption"
167+ style was no longer getting applied. (It was overwritten by "Compact.")
169+ * Use commonmark-extensions 0.2.1.2
171+## pandoc 2.14 (2021-05-28)
173+ * Change reader types, allowing better tracking of source positions
174+ [API change]. Previously, when multiple file arguments were provided,
175+ pandoc simply concatenated them and passed the contents to the readers,
176+ which took a Text argument. As a result, the readers had no way of knowing
177+ which file was the source of any particular bit of text. This meant that
178+ we couldn't report accurate source positions on errors or include accurate
179+ source positions as attributes in the AST. More seriously, it meant that
180+ we couldn't resolve resource paths relative to the files containing them
181+ (see e.g. #5501, #6632, #6384, #3752).
183+ * Add `rebase_relative_paths` extension (#3752). When enabled, this
184+ extension rewrites relative image and link paths by prepending
185+ the (relative) directory of the containing file. This
186+ behavior is useful when your input sources are split
187+ into multiple files, across several directories, with files
188+ referring to images stored in the same directory. The
189+ extension can be enabled for all markdown and commonmark-based formats.
191+ * Add Text.Pandoc.Sources (exported module), with a `Sources` type and a
192+ `ToSources` class. A `Sources` wraps a list of `(SourcePos, Text)` pairs
193+ [API change]. A parsec `Stream` instance is provided for `Sources`. The
194+ module also exports versions of parsec's `satisfy` and other Char parsers
195+ that track source positions accurately from a `Sources` stream (or any
196+ instance of the new `UpdateSourcePos` class).
198+ * Text.Pandoc.Parsing
200+ + Export the modified Char parsers defined in Text.Pandoc.Sources
201+ instead of the ones parsec provides. Modified parsers to use a
202+ `Sources` as stream [API change].
203+ + Improve include file functions [API change]. Remove old
204+ `insertIncludedFileF`. Give `insertIncludedFile` a more general type,
205+ allowing it to be used where `insertIncludedFileF` was.
206+ + Add parameter to the `citeKey` parser from
207+ Text.Pandoc.Parsing, which controls whether the `@{..}`
208+ syntax is allowed [API change].
210+ * Text.Pandoc.Error: Modified the constructor `PandocParsecError` to take a
211+ `Sources` rather than a `Text` as first argument, so parse error locations
212+ can be accurately reported.
214+ * Fix source position reporting for YAML bibliographies (#7273).
216+ * Issue error message when reader or writer format is malformed
217+ (#7231). Previously we exited with an error status but (due to a bug)
220+ * Smarter smart quotes (#7216, #2103). Treat a leading `"` with no
221+ closing `"` as a left curly quote. This supports the practice, in
222+ fiction, of continuing paragraphs quoting the same speaker without an
223+ end quote. It also helps with quotes that break over lines in line blocks.
227+ + Use MetaInlines not MetaBlocks for multimarkdown metadata fields.
228+ This gives better results in converting to e.g. pandoc markdown.
229+ + Implement curly-brace syntax for Markdown citation keys (#6026).
230+ The change provides a way to use citation keys that contain
231+ special characters not usable with the standard citation key syntax.
232+ Example: `@{foo_bar{x}'}` for the key `foo_bar{x}`. It also allows
233+ separating citation keys from immediately following text, e.g. `@{foo}A`.
237+ + Seek include files in the directory of the file
238+ containing the include directive, as RST requires (#6632).
239+ + Use `insertIncludedFile` from Text.Pandoc.Parsing
240+ instead of reproducing much of its code.
242+ * Org reader: Resolve org includes relative to the directory containing the
243+ file containing the INCLUDE directive (#5501).
245+ * ODT reader: Treat tabs as spaces (#7185, niszet).
249+ + Add handling of vml image objects (#7257, mbrackeantidot).
250+ + Support new table features (Emily Bourke, #6316): column
251+ spans, row spans, multiple header rows, table description
252+ (parsed as a simple caption), captions, column widths.
256+ + Improved siunitx support (#6658, #6620).
257+ + Better support for `\xspace` (#7299).
258+ + Improve parsing of `\def` macros. We previously set "verbatim mode"
259+ even for parsing the initial `\def`; this caused problems
260+ for `\def` nested inside another `\def`.
261+ + Implement `\newif`.
263+ * ConTeXt writer: improve ordered lists (#5016, Denis Maier).
264+ Change ordered list from itemize to enumerate. Add new
265+ itemgroup for ordered lists. Remove manual insertion of
266+ width attributes. Use tabular figures in ordered list
271+ + Don't fail on unmatched closing "script" tag (Albert Krenkel, #7282).
272+ + Keep h1 tags as normal headers (#2293, Albert Krewinkel).
273+ The tags `<title>` and `<h1 class="title">` often contain the same
274+ information, so the latter was dropped from the document. However, as
275+ this can lead to loss of information, the heading is now always
276+ retained. Use `--shift-heading-level-by=-1` to turn the `<h1>`
277+ into the document title, or a filter to restore the previous behavior.
278+ + Handle relative lengths (e.g. `2*`) in HTML column widths (#4063).
279+ See <https://www.w3.org/TR/html4/types.html#h-6.6>.
281+ * DocBook/JATS readers:
283+ + Fix mathml regression caused by the switch in XML libraries (#7173).
284+ + Fix "phrase" in DocBook: take classes from "role" not "class" (#7195).
286+ * DocBook reader: ensure that first and last names are separated (#6541).
288+ * Jira reader (Albert Krewinkel, #7218):
290+ + Support "smart" links: `[alias|https://example.com|smart-card]` syntax.
291+ + Allow spaces and most unicode characters in attachment links.
292+ + No longer require a newline character after `{noformat}`.
293+ + Only allow URI path segment characters in bare links.
294+ + The `file:` schema is no longer allowed in bare links; these
297+ * Plain writer: handle superscript unicode minus (#7276).
301+ + Better handling of line breaks in simple tables (#7272).
302+ Now we also handle the case where they're embedded in other elements,
304+ + For beamer output, support `exampleblock` and `alertblock` (#7278).
305+ A block will be rendered as an `exampleblock` if the heading
306+ has class `example` and an `alertblock` if it has class `alert`.
307+ + Separate successive quote chars with thin space (#6958,
308+ Albert Krewinkel). Successive quote characters are separated with
309+ a thin space to improve readability and to prevent unwanted ligatures.
310+ Detection of these quotes sometimes had failed if the second quote
311+ was nested in a span element.
312+ + Separate successive quote chars with thin space (#6958, Albert
315+ * EPUB Writer: Fix belongs-to-collection XML id choice (#7267, nuew).
316+ The epub writer previously used the same XML id for both the book
317+ identifier and the epub collection. This causes an error on epubcheck.
319+ * BibTeX/BibLaTeX writer: Handle `annote` field (#7266).
321+ * ZimWiki writer: allow links and emphasis in headers (#6605,
326+ + Support blank lines in line blocks (#6564, Albert Krewinkel,
327+ thanks to @denismaier).
328+ + Use span identifiers as reference anchors (#7246, Albert Krewinkel).
332+ + Keep attributes from code nested below `pre` tag (#7221,
333+ Albert Krewinkel). If a code block is defined with `<pre><code
334+ class="language-x">…</code></pre>`, where the `<pre>` element has no
335+ attributes, then the attributes from the `<code>` element are used
336+ instead. Any leading `language-` prefix is dropped in the code's
337+ `class` attribute are dropped to improve syntax highlighting.
338+ + Ensure headings only have valid attribs in HTML4 (#5944, Albert
340+ + Parse `<header>` as a Div (Albert Krewinkel).
344+ + Inline latex envs need newlines (#7252, tecosaur).
345+ As specified in https://orgmode.org/manual/LaTeX-fragments.html, an
346+ inline \begin{}...\end{} LaTeX block must start on a new line.
347+ + Use LaTeX style maths deliminators (#7196, tecosaur).
349+ * JATS writer (Albert Krewinkel):
351+ + Use either styled-content or named-content for spans (#7211).
352+ If the element has a content-type attribute, or at least one class,
353+ then that value is used as `content-type` and the span is put inside
354+ a `<named-content>` element. Otherwise a `<styled-content>` element
356+ + Reduce unnecessary use of `<p>` elements for wrapping (#7227).
357+ The `<p>` element is used for wrapping in cases were the contents
358+ would otherwise not be allowed in a certain context. Unnecessary
359+ wrapping is avoided, especially around quotes (`<disp-quote>` elements).
360+ + Convert spans to `<named-content>` elements (#7211). Spans with
361+ attributes are converted to `<named-content>` elements instead of
362+ being wrapped with `<milestone-start/>` and `<milestone-end>`
363+ elements. Milestone elements are not allowed in documents using the
364+ articleauthoring tag set, so this change ensures the creation of valid
366+ + Add footnote number as label in backmatter (#7210). Footnotes in the
367+ backmatter are given the footnote's number as a label. The
368+ articleauthoring output is unaffected from this change, as footnotes
369+ are placed inline there.
370+ + Escape disallows chars in identifiers. XML identifiers must start
371+ with an underscore or letter, and can contain only a limited set
372+ of punctuation characters. Any IDs not adhering to these rules are
373+ rewritten by writing the offending characters as `Uxxxx`,
374+ where `xxxx` is the character's hex code.
376+ * Jira writer: use `{color}` when span has a color attribute
377+ (Albert Krewinkel, tarleb/jira-wiki-markup#10).
381+ + Autoset table width if no column has an explicit width (Albert
383+ + Extract Table handling into separate module (Albert Krewinkel).
384+ + Support colspans and rowspans in tables (Albert Krewinkel, #6315).
385+ + Support multirow table headers (Albert Krewinkel).
386+ + Improve integration of settings from reference.docx (#1209).
387+ This change allows users to create a reference.docx that
388+ sets `w:proofState` for spelling or grammar to `dirty`,
389+ so that spell/grammar checking will be triggered on the
391+ + Copy over more settings from reference.docx (#7240). From settings.xml
392+ in the reference-doc, we now include: `zoom`, `embedSystemFonts`,
393+ `doNotTrackMoves`, `defaultTabStop`, `drawingGridHorizontalSpacing`,
394+ `drawingGridVerticalSpacing`, `displayHorizontalDrawingGridEvery`,
395+ `displayVerticalDrawingGridEvery`, `characterSpacingControl`,
396+ `savePreviewPicture`, `mathPr`, `themeFontLang`, `decimalSymbol`,
397+ `listSeparator`, `autoHyphenation`, `compat`.
398+ + Set zoom to 100% by default in settings.xml.
399+ + Align math options more with current Word defaults (e.g. Cambria Math
401+ + Remove `rsid`s from default settings.xml. Word will add these
402+ when revisions are made.
404+ * Ms writer: Handle tables with multiple paragraphs (#7288).
405+ Previously they overflowed the table cell width. We now set line lengths
406+ per-cell and restore them after the table has been written.
410+ + Use cleaner braceless syntax for code blocks with a
411+ single class (#7242, Jan Tojnar).
412+ + Add quotes properly in markdown YAML metadata fields (#7245).
413+ This fixes a bug, which caused the writer to look at the *last*
414+ rather than the *first* character in determining whether quotes
415+ were needed. So we got spurious quotes in some cases and
416+ didn't get necessary quotes in others.
417+ + Use `@{..}` syntax for citations when needed.
418+ + Use fewer unneeded escapes for `#` (see #6259).
419+ + Improve escaping of `@`. We need to escape literal `@` before
420+ `{` because of the new citation syntax.
422+ * Commonmark writer: Use backslash escapes for `<` and `|`...
423+ instead of entities (#7208).
425+ * Powerpoint writer: allow `monofont` to be specified in metadata
430+ + Use non-starred names for xcolor color names (#6109).
431+ This should make svgnames and x11names work properly.
432+ + Fix bad vertical spacing after bibliography (#7234, badumont).
433+ + List of figures before list of tables (#7235, Julien Dutant).
434+ + Move CSL macro definitions before header-includes so they can be
436+ + Improve treatment of CSL `entry-spacing` (#7296).
437+ Previously with the default template settings (`indent` variable
438+ not set), we would get interparagraph spaces separating bib
439+ entries even with `entry-spacing="0"`. On the other hand,
440+ setting `entry-spacing="2"` gave ridiculously large spacing.
441+ This change makes the spacing caused by `entry-spacing` a multiple
442+ of `\parskip` by default, which gives aesthetically reasonable
443+ output. Those who want a larger or smaller unit (e.g. because
444+ they use `indent` which sets `\parskip` to 0) may
445+ `\setlength{\cslentryspacingunit}{10pt}` in header-includes
446+ to override the defaults.
447+ + Move title, author, date up to top of preamble (#7295).
448+ This allows header-includes to use them, and puts them
449+ in a position where you can see them immediately.
450+ + Define commands for zero width non-joiner character
451+ (#6639, Albert Krewinkel). The zero-width non-joiner character
452+ is used to avoid ligatures (e.g. in German).
456+ + Define `enumerate` itemgroup (#5016, Denis Maier).
457+ + List of figures before list of tables (#7235, Julien Dutant).
459+ * reveal.js template:
461+ + Support `toc-title` (#7171, Florian Kohrt).
462+ + Use `hash: true` by default rather than `history: true` (#6968).
464+ * HTML-based slide shows: add support for `institute` (#7289, Thomas
467+ * Text.Pandoc.Extensions: Add constructor `Ext_rebase_relative_paths` to
468+ `Extensions` [API change].
470+ * Text.Pandoc.XML.Light: add Eq, Ord instances for Content,
471+ Element, Attr, CDataKind [API change].
473+ * Text.Pandoc.MediaBag:
475+ + Change type to use a `Text` key instead of `[FilePath]`.
476+ We normalize the path and use `/` separators for consistency.
477+ + Export `MediaItem` type [API change].
478+ + Change `MediaBag` type to a map from Text to MediaItem [API change].
479+ + `lookupMedia` now returns a `MediaItem` [API change].
480+ + Change `insertMedia` so it sets the `mediaPath` to a filename based on
481+ the SHA1 hash of the contents. This will be used when contents
484+ * Text.Pandoc.Class.PandocMonad:
486+ + Remove `fetchMediaResource` [API change]. Use `fetchItem` to get
487+ resources in `fillMediaBag`.
488+ + Add informational message in `downloadOrRead` indicating what path
489+ local resources have been loaded from.
491+ * Text.Pandoc.Logging:
493+ + Remove single quotes around paths in messages.
494+ + Add LoadedResource constructor to LogMessage [API change].
495+ This is for INFO-level messages telling where image data has been
496+ loaded from. (This can vary because of the resource path.)
498+ * Text.Pandoc.Asciify: simplify code and export `toAsciiText` [API change].
499+ Instead of encoding a giant (and incomplete) map, we now
500+ just use unicode-transforms to normalize the text to
501+ a canonical decomposition, and manipulate the result.
503+ * App: allow tabs expansion even if file-scope is used (Albert Krewinkel,
504+ #6709). Tabs in plain-text inputs are now handled correctly, even if
505+ the `--file-scope` flag is used.
507+ * Add new internal module Text.Pandoc.Writers.GridTable (Albert Krewinkel).
509+ * Text.Pandoc.Highlighting: Change type of `languagesByExtension`, adding
510+ a parameter for a `SyntaxMap` [API change] (Jan Tojnar, #7241).
511+ Languages defined using `--syntax-definition` were not recognized by
512+ `languagesByExtension`. This patch corrects that, allowing the writers
513+ to see all custom definitions. The LaTeX writer still uses the default
514+ syntax map, but that's okay in that context, since
515+ `--syntax-definition` won't create new listings styles.
517+ * Text.Pandoc.Citeproc:
519+ + Ensure that CSL-related attributes are passed on to a Div with id
520+ 'refs'. Otherwise things like `entry-spacing` won't work when
522+ + Use metadata's `lang` for the lang parameter of citeproc, overriding
524+ + Recognize locators spelled with a capital letter (#7323).
525+ + Add a comma and a space in front of the suffix if it doesn't start
526+ with space or punctuation (#7324).
527+ + Don't detect math elements as locators (#7321).
529+ * Remove Text.Pandoc.BCP47 module [API change]. Use types and functions
530+ from UnicodeCollation.Lang instead. This is a richer implementation
533+ * Text.Pandoc.Shared:
535+ + Fix regression in grid tables for wide characters (#7214).
536+ In the translation from String to Text, a char-width-sensitive
537+ `splitAt'` was dropped. This commit reinstates it and uses it to make
538+ `splitTextByInstances` char-width sensitive.
539+ + Add `getLang` (formerly in the now-removed BCP47) [API change].
541+ * Text.Pandoc.SelfContained: use `application/octet-stream`
542+ for unknown mime types instead of halting with an error (#7202).
544+ * Lua filters: respect Inlines/Blocks filter functions in `pandoc.walk_*`
547+ * Add text as build-depend for trypandoc (#7193, Roman Beránek).
549+ * Bump upper-bounds for network-uri, time, attoparsec.
553+ * Use texmath 0.12.3.
555+ * Use jira-wiki-markup 1.3.5 (Albert Krewinkel).
557+ * Require latest skylighting (fixes a bug in XML syntax highlighting).
559+ * Use latest xml-conduit.
561+ * Use latest commonmark, commonmark-extensions, commonmark-pandoc.
563+ * Use haddock-library-1.10.0 (Albert Krewinkel).
565+ * Allow compilation with base 4.15 (Albert Krewinkel).
569+ + Add information about `lang` and bibliography sorting.
570+ + Add info about YAML escape sequences, link to spec (#7152,
572+ + Note that `institute` variable works for HTML-based slides.
573+ + Update documentation on citation syntax.
574+ + Add citation example for locators and suffixes (Tristan Stenner)
576+ * Updated and fixed typos in documentation (Charanjit Singh,
577+ Anti-Distinctlyminty, Tatiana Porras, obcat).
579+ * Add instructions for installing pandoc-types before compiling filter.
581+ * INSTALL: add note that parallel installations should be avoided
584+ * Remove `biblatex-nussbaum.md` test. It is basically the same
585+ as `biblaetx-quotes.md`.
587+ * Command tests: fail if a file contains no tests---and fix a
588+ test that failed in that way!
590+ * Use smaller images in tests, reducing the size of the source tarball by 8 MB.
593+## pandoc 2.13 (2021-03-21)
595+ * Support `yaml_metadata_block` extension for `commonmark`, `gfm` (#6537).
596+ This supported is a bit more limited than with pandoc's
597+ `markdown`. The YAML block must be the first thing in the input,
598+ and the leaf notes are parsed in isolation from the rest of
599+ the document. So, for example, you can't use reference
600+ links if the references are defined later in the document.
602+ * Fix fallback to default partials when custom templates are used.
603+ If the directory containing a template does not contain the partial,
604+ it should be sought in the default templates, but this was not
605+ working properly (#7164).
607+ * Handle `nocite` better with `--biblatex` and `--natbib` (#4585).
608+ Previously the nocite metadata field was ignored with these formats.
609+ Now it populates a `nocite-ids` template variable and causes a
610+ `\nocite` command to be issued.
612+ * Text.Pandoc.Citeproc: apply `fixLinks` correctly (#7130). This is code
613+ that incorporates a prefix like `https://doi.org/` into a following link
616+ * Text.Pandoc.Shared:
618+ + Remove `backslashEscapes`, `escapeStringUsing` [API change]. Replace
619+ these inefficient association list lookups with more efficient escaping
620+ functions in the writers that used them (for a 10-25% performance boost
621+ in org, haddock, rtf, texinfo writers).
622+ + Remove `ToString`, `ToText` typeclasses [API change]. These were needed
623+ for the transition from String to Text, but they are no longer used and
624+ may clash with other things.
625+ + Simplify `compactDL`.
627+ * Text.Pandoc.Parsing:
629+ + Change type of `readWithM` so that it is no longer polymorphic
630+ [API change]. The `ToText` class has been removed, and now that we've
631+ completed the transition to Text we no longer need this to operate
633+ + Remove `F` type synonym [API change]. Muse and Org were defining their
636+ * Text.Pandoc.Readers.Metadata:
638+ + Export `yamlMetaBlock` [API change].
639+ + Make `yamlBsToMeta`, `yamlBsToRefs` polymorphic on the parser state
642+ * Markdown reader: Fix regression with `tex_math_backslash` (#7155).
644+ * MediaWiki reader: Allow block-level content in notes (ref) (#7145).
646+ * Jira reader (Albert Krewinkel):
648+ + Fixed parsing of autolinks (i.e., of bare URLs in the text).
649+ Previously an autolink would take up the rest of a line, as spaces
650+ were allowed characters in these items.
651+ + Emoji character sequences no longer cause parsing failures. This was
652+ due to missing backtracking when emoji parsing fails.
653+ + Mark divs created from panels with class "panel".
655+ * RST reader: fix logic for ending comments (#7134). Previously comments
656+ sometimes got extended too far.
658+ * DocBook writer: include Header attributes as XML attributes on
659+ section (Erik Rask). Attributes with key names that are not allowed
660+ as XML attributes are dropped, as are attributes with invalid values
661+ and `xml:id` (DocBook 5) and `id` (DocBook 4).
665+ + Make `nsid` in `abstractNum` deterministic. Previously we assigned
666+ a random number, but we don't need random values, so now we just
667+ assign a value based on the list marker.
668+ + Use integral values for `w:tblW` (#7141).
670+ * Jira writer (Albert Krewinkel):
672+ + Block quotes are only rendered as `bq.` if they do not contain a
674+ + Jira writer: improve div/panel handling. Include div attributes in
675+ panels, always render divs with class `panel` as panels, and
676+ avoid nesting of panels.
678+ * HTML writer: Add warnings on duplicate attribute values.
679+ This prevents emitting invalid HTML. Ultimately it would be good to
680+ prevent this in the types themselves, but this is better for now.
682+ * Org writer: Prevent unintended creation of ordered list items (#7132,
683+ Albert Krewinkel). Adjust line wrapping if default wrapping would cause
684+ a line to be read as an ordered list item.
686+ * JATS templates: support 'equal-contrib' attrib for authors (Albert
687+ Krewinkel). Authors who contributed equally to a paper may be marked
688+ with `equal-contrib`.
690+ * reveal.js template: replace JS comment with HTML (#7154, Florian Kohrt).
692+ * Text.Pandoc.Logging: Add `DuplicateAttribute` constructor to `LogMessage`.
695+ * Use `-j4` for linux release build. This speeds up the build dramatically
698+ * cabal.project: remove ghcoptions. Move flags to top level, so they can
699+ be set differently on the command line.
701+ * Require latest texmath, skylighting, citeproc, jira-wiki-markup.
702+ (The latest skylighting fixes a bad bug with Haskell syntax highlighting.)
703+ Narrow version bounds for texmath, skylighting, and citeproc, since
704+ the test output depend on them.
706+ * Use doclayout 0.3.0.2. This significantly reduces the time and memory
707+ needed to compile pandoc.
709+ * Use `foldl'` instead of `foldl` everywhere.
711+ * Update bounds for random (#7156, Alexey Kuleshevich).
713+ * Remove uses of some partial functions.
715+ * Don't bake in a larger stack size for the executable.
717+ * Test improvements:
719+ + Use `getExecutablePath` from base, avoiding the dependency on
721+ + Factor out `setupEnvironment` in Helpers, to avoid code duplication.
722+ + Fix finding of data files by setting teh `pandoc_datadir` environment
723+ variable when we shell out to pandoc. This avoids the need to use
724+ `--data-dir` for the tests, which caused problems finding `pandoc.lua`
725+ when compiling without the `embed_data_files` flag (#7163).
727+ * Benchmark improvements:
729+ + Build `+RTS -A8m -RTS` into default ghc-options for benchmark.
730+ This is necessary to get accurate benchmark results; otherwise we
731+ are largely measuring garbage collecting, some not related to the
733+ + Allow specifying BASELINE file in 'make bench' for comparison
734+ (otherwise the latest benchmark is chosen by default).
735+ + Force `readFile` in benchmarks early (Bodigrim).
737+ * CONTRIBUTING: suggest using a `cabal.project.local` file (#7153,
740+ * Add ghcid-test to Makefile. This loads the test suite in ghcid.
743+## pandoc 2.12 (2021-03-08)
745+ * `--resource-path` now accumulates if specified multiple
746+ times (#6152). Resource paths specified later on the command line are
747+ prepended to those specified earlier. Thus,
748+ `--resource-path foo --resource-path bar:baz` is equivalent to
749+ `--resource-path bar:bas:foo`. (The previous behavior was
750+ for the last `--resource-path` to replace all the rest.)
751+ `resource-path` in defaults files behaves the same way: it
752+ will be prepended to the resource path set by earlier
753+ command line options or defaults files. This change
754+ facilitates the use of multiple defaults files: each can
755+ specify a directory containing resources it refers to
756+ without clobbering the resource paths set by the others.
758+ * Allow defaults files to refer to the home directory, the
759+ user data directory, and the directory containing the defaults file
760+ itself (#5871, #5982, #5977). In fields that expect file paths
761+ (and only in these fields),
763+ + `${VARIABLE}` will expand to the value of the environment variable
764+ `VARIABLE` (and in particular `${HOME}` will expand to the path
765+ of the home directory). A warning will be raised for undefined
767+ + `${USERDATA}` will expand to the path of the user data
768+ directory in force when the defaults file is being processed.
769+ + `${.}` will expand to the directory containing the defaults file.
770+ (This allows default files to be placed in a directory containing
771+ resources they make use of.)
773+ * When downloading content from URL arguments, be sensitive to
774+ the character encoding (#5600). We can properly handle UTF-8 and latin1
775+ (ISO-8859-1); for others we raise an error. Fall back to latin1 if
776+ no charset is given in the mime type and UTF-8 decoding fails.
778+ * Allow abbreviations that don't end in a period to be
779+ specified using `--abbreviations` (#7124).
781+ * Add new unexported module Text.Pandoc.XML.Light, as well
782+ as Text.Pandoc.XML.Light.Types, Text.Pantoc.XML.Light.Proc,
783+ Text.Pandoc.XML.Light.Output. (Closes #6001, #6565, #7091).
785+ This module exports definitions of `Element` and `Content`
786+ that are isomorphic to xml-light's, but with Text
787+ instead of String. This allows us to keep most of the code in existing
788+ readers that use xml-light, but avoid lots of unnecessary allocation.
790+ We also add versions of the functions from xml-light's
791+ Text.XML.Light.Output and Text.XML.Light.Proc that operate on our
792+ modified XML types, and functions that convert xml-light types to our
793+ types (since some of our dependencies, like texmath, use xml-light).
795+ We export functions that use xml-conduit's parser to produce an
796+ `Element` or `[Content]`. This allows existing pandoc code to use
797+ a better parser without much modification.
799+ The new parser is used in all places where xml-light's parser was
800+ previously used. Benchmarks show a significant performance improvement
801+ in parsing XML-based formats (with docbook, opml, jats, and docx
802+ almost twice as fast, odt and fb2 more than twice as fast).
804+ In addition, the new parser gives us better error reporting than
805+ xml-light. We report XML errors, when possible, using the new
806+ `PandocXMLError` constructor in `PandocError`.
808+ These changes revealed the need for some changes in the tests. The
809+ docbook-reader.docbook test lacked definitions for the entities it used;
810+ these have been added. And the docx golden tests have been updated,
811+ because the new parser does not preserve the order of attributes.
815+ + Avoid expensive tree normalization step, as it is not necessary
816+ with the new XML parser.
817+ + Support `informalfigure` (#7079) (Nils Carlson).
821+ + Use Map instead of list for Namespaces. This gives a speedup of
822+ about 5-10%. With this and the XML parsing changes, the docx reader
823+ is now about twice as fast as in the previous release.
827+ + Small performance tweaks.
828+ + Also, remove exported class `NamedTag(..)` [API change]. This was just
829+ intended to smooth over the transition from String to Text and is no
831+ + As a result, the functions `isInlineTag` and `isBlockTag`
832+ are no longer polymorphic; they apply to a `Tag Text` [API change].
833+ + Do a lookahead to find the right parser to use. This takes
834+ benchmarks from 34ms to 23ms, with less allocation.
835+ + Fix bad handling of empty `src` attribute in `iframe` (#7099).
836+ If `src` is empty, we simply skip the `iframe`.
837+ If `src` is invalid or cannot be fetched, we issue a warning
838+ nd skip instead of failing with an error.
842+ + Avoid tree normalization, which is no longer necessary given the
847+ + Don't export `tokenize`, `untokenize` [API change]. These are internal
848+ implementation details, which were only exported for testing.
849+ They don't belong in the public API.
850+ + Improved efficiency of the parser. With these changes the reader
851+ is almost twice as fast as in the last release in our benchmarks.
852+ + Code cleanup, removing some unnecessary things.
853+ + Rewrite `withRaw` so it doesn't rely on fragile assumptions
854+ about token positions (which break when macros are expanded)
855+ (#7092). This requires the addition of `sEnableWithRaw` and
856+ `sRawTokens` in `LaTeXState`, and a new combinator `disablingWithRaw`
857+ to disable collecting of raw tokens in certain contexts.
858+ Add `parseFromToks` to Text.Pandoc.Readers.LaTeX.Parsing.
859+ Fix parsing of single character tokens so it doesn't mess
860+ up the new raw token collecting. These changes slightly increase
861+ allocations and have a small performance impact.
862+ + Handle some bibtex/biblatex-specific commands that used to be
863+ dealt with in pandoc-citeproc (#7049).
864+ + Optimize `satisfyTok`, avoiding unnecessary macro expansion steps.
865+ Benchmarks after this change show 2/3 of the run time and 2/3 of the
866+ allocation of the Feb. 10 benchmarks.
867+ + Removed `sExpanded` in state. This isn't actually needed and checking
868+ it doesn't change anything.
869+ + Improve `braced'`. Remove the parameter, have it parse the
870+ opening brace, and make it more efficient.
871+ + Factor out pieces of the LaTeX reader to make the module smaller.
872+ This reduces memory demands when compiling. Created
873+ Text.Pandoc.Readers.{LaTeX,Math,Citation,Table,Macro,Inline}.
874+ Changed Text.Pandoc.Readers.LaTeX.SIunitx to export a command map
875+ instead of individual commands.
876+ + Handle table cells containing `&` in `\verb` (#7129).
878+ * Make Text.Pandoc.Readers.LaTeX.Types an unexported module [API change].
882+ + Improved handling of mmd link attributes in references (#7080).
883+ Previously they only worked for links that had titles.
884+ + Improved efficiency of the parser (benchmarks show a 15% speedup).
888+ + Avoid tree normalization, which is no longer necessary with the
893+ + Finer-grained errors on parse failure (#7091).
894+ + Give more information if the zip container can't be unpacked.
898+ + Support `task_lists` extension (Albert Krewinkel, #6336).
899+ + Fix bug in org-ref citation parsing (Albert Krewinkel, #7101).
900+ The org-ref syntax allows to list multiple citations separated by
901+ comma. Previously commas were accepted as part of the citation id,
902+ so all citation lists were parsed as one single citation.
906+ + Use `getTimestamp` instead of `getCurrentTime` to fetch timestamp.
907+ Setting `SOURCE_DATE_EPOCH` will allow reproducible builds.
908+ + RST reader: fix handling of header in CSV tables (#7064).
909+ The interpretation of this line is not affected by the delim option.
913+ + Modified the Doc parser to skip leading blank lines. This fixes
914+ parsing of documents which start with multiple blank lines (Albert
916+ + Prevent URLs within link aliases to be treated as autolinks (Albert
919+ * Text.Pandoc.Shared
921+ + Remove formerly exported functions that are no longer used in the
922+ code base: `splitByIndices`, `splitStringByIndicies`, `substitute`,
923+ and `underlineSpan` (which had been deprecated in April 2020)
925+ + Export `handleTaskListItem` (Albert Krewinkel) [API change].
926+ + Change `defaultUserDataDirs` to `defaultUserDataDir` [API
927+ change]. We determine what is the default user data directory
928+ by seeing whether the XDG directory and/or legacy
933+ + BibTeX writer: use doclayout and doctemplate. This change allows
934+ bibtex/biblatex output to wrap as other formats do,
935+ depending on the settings of `--wrap` and `--columns` (#7068).
939+ + Output `[]` if no references in input, instead of raising a
940+ PandocAppError as before.
944+ + Use `getTimestamp` instead of `getCurrentTime` for timestamp.
945+ Setting `SOURCE_DATE_EPOCH` will allow reproducible builds.
949+ + Use `getTimestamp` instead of `getCurrentTime` for timestamp.
950+ Setting `SOURCE_DATE_EPOCH` will allow reproducible builds (#7093).
951+ This does not suffice to fully enable reproducible in EPUB, since
952+ a unique id is still being generated for each build.
953+ + Support `belongs-to-collection` metadata (#7063) (Nick Berendsen).
957+ + Escape special chars in reference elements (Albert Krewinkel).
958+ Prevents the generation of invalid markup if a citation element
959+ contains an ampersand or another character with a special meaning
964+ + Use Span identifiers as anchors (Albert Krewinkel).
965+ + Use `{noformat}` instead of `{code}` for unknown languages (Albert
966+ Krewinkel). Code blocks which are not marked as a language supported
967+ by Jira are rendered as preformatted text via `{noformat}` blocks.
971+ + Adjust hypertargets to beginnings of paragraphs (#7078).
972+ Use `\vadjust pre` so that the hypertarget takes you to the beginning
973+ of the paragraph rather than one line down.
974+ This makes a particular difference for links to citations using
975+ `--citeproc` and `link-citations: true`.
976+ + Change BCP47 lang tag from `jp` to `ja` (Mauro Bieg, #7047).
977+ + Use function instead of map for accent lookup (should be
979+ + Split the module to make it easier to compile on low-memory
980+ systems: added Text.Pandoc.Writers.LaTeX.{Util,Citation,Lang}.
984+ + Handle math right before digit. We insert an HTML comment to
985+ avoid a `$` right before a digit, which pandoc will not recognize
986+ as a math delimiter.
987+ + Split the module to make it easier to compile on low-memory
988+ systems: added Text.Pandoc.Writers.Markdown.{Types,Inline}.
992+ + Use `getTimestamp` instead of `getCurrentTime` for timestamp.
993+ Setting `SOURCE_DATE_EPOCH` will allow reproducible builds.
994+ + Update default ODT style (Lorenzo). Previously, the "First paragraph"
995+ style inherited from "Standard" but not from "Text body." Now
996+ it is adjusted to inherit from "Text body", to avoid some ugly
997+ spacing issues. It may be necessary to update a custom `reference.odt`
998+ in light of this change.
1002+ + Support `task_lists` extension (Albert Krewinkel, #6336).
1006+ + Use `getTimestamp` instead of `getCurrentTime` for timestamp.
1007+ Setting `SOURCE_DATE_EPOCH` will allow reproducible builds.
1009+ * JATS templates: tag `author.name` as `string-name` (Albert Krewinkel).
1010+ The partitioning the components of a name into surname, given names,
1011+ etc. is not always possible or not available. Using `author.name`
1012+ allows to give the full name as a fallback to be used when
1013+ `author.surname` is not available.
1015+ * Add default templates for bibtex and biblatex, so that
1016+ the variables `header-include`, `include-before`, `include-after`
1017+ (or alternatively the command line options
1018+ `--include-in-header`, `--include-before-body`, `--include-after-body`)
1023+ + Update to iftex package (#7073) (Andrew Dunning)
1024+ + Wrap url colours in braces (#7121) (Loïc Grobol).
1026+ * revealjs template: Add 'center' option for vertical slide centering.
1029+ * Text.Pandoc.XML: Improve efficiency of `fromEntities`.
1033+ + Add exported function `getCharset` [API change].
1035+ * Text.Pandoc.UTF8: change IO functions to return Text, not String
1036+ [API change]. This affects `readFile`, `getContents`, `writeFileWith`,
1037+ `writeFile`, `putStrWith`, `putStr`, `putStrLnWith`, `putStrLn`.
1038+ `hPutStrWith`, `hPutStr`, `hPutStrLnWith`, `hPutStrLn`, `hGetContents`.
1039+ This avoids the need to uselessly create a linked list of characters
1040+ when emiting output.
1044+ + Add `parseOptionsFromArgs` [API change, new exported function].
1045+ + Add fields for CSL options to `Opt` [API change]:
1046+ `optCSL`, `optbibliography`, `optCitationAbbreviations`.
1048+ * Text.Pandoc.Citeproc.BibTeX
1050+ + `Text.Pandoc.Citeproc.writeBibTeXString` now returns
1051+ `Doc Text` instead of `Text` (#7068).
1052+ + Correctly handle `pages` (= `page` in CSL) (#7067).
1053+ + Correctly handle BibLaTeX `langid` (= `language` in CSL, #7067).
1054+ + In BibTeX output, protect foreign titles since there's no language
1056+ + Clean up BibTeX parsing (#7049). Previously there was a messy code
1057+ path that gave strange results in some cases, not passing through raw
1058+ tex but trying to extract a string content. This was an artefact of
1059+ trying to handle some special bibtex-specific commands in the BibTeX
1060+ reader. Now we just handle these in the LaTeX reader and simplify
1061+ parsing in the BibTeX reader. This does mean that more raw tex will
1062+ be passed through (and currently this is not sensitive to the
1063+ `raw_tex` extension; this should be fixed).
1065+ * Text.Pandoc.Citeproc.MetaValue
1067+ + Correctly parse "raw" date value in markdown references metadata.
1068+ (See jgm/citeproc#53.)
1070+ * Text.Pandoc.Citeproc
1072+ + Use https URLs for links (Salim B, #7122).
1074+ * Text.Pandoc.Class
1076+ + Add `getTimestamp` [API change]. This attempts to read the
1077+ `SOURCE_DATE_EPOCH` environment variable and parse a UTC time
1078+ from it (treating it as a unix date stamp, see
1079+ https://reproducible-builds.org/specs/source-date-epoch/). If the
1080+ variable is not set or can't be parsed as a unix date stamp, then the
1081+ function returns the current date.
1083+ * Text.Pandoc.Error
1085+ + Add `PandocUnsupportedCharsetError` constructor for
1086+ `PandocError` [API change].
1087+ + Export `renderError` [API change].
1088+ + Refactor `handleError` to use `renderError`. This allows us render
1089+ error messages without exiting.
1091+ * Text.Pandoc.Extensions
1093+ + `Ext_task_lists` is now supported by org (and turned
1094+ on by default) (Albert Krewinkel, #6336).
1095+ + Remove `Ext_fenced_code_attributes` from allowed commonmark attributes
1096+ (#7097). This attribute was listed as allowed, but it didn't actually
1097+ do anything. Use `attributes` for code attributes and more.
1101+ + Always load built-in Lua scripts from default data-dir (Albert
1102+ Krewinkel). The Lua modules `pandoc` and `pandoc.List` are now always
1103+ loaded from the system's default data directory. Loading from a
1104+ different directory by overriding the default path, e.g. via
1105+ `--data-dir`, is no longer supported to avoid unexpected behavior
1106+ and to address security concerns.
1107+ + Add module "pandoc.path" (Albert Krewinkel, #6001, #6565).
1108+ The module allows to work with file paths in a convenient and
1109+ platform-independent manner.
1110+ + Use strict evaluation when retrieving AST value from the stack
1111+ (Albert Krewinkel, #6674).
1115+ + Disable `smart` extension when building PDF via LaTeX.
1116+ This is to prevent accidental creation of ligatures like
1117+ `` ?` `` and `` !` `` (especially in languages with quotations like
1118+ German), and similar ligature issues. (See jgm/citeproc#54.)
1122+ + Fix parsing of unquoted values (#7112). Previously we didn't allow
1123+ unescaped quotes in unquoted values, but they are allowed
1128+ + Use a more robust method for testing the executable. Many
1129+ of our tests require running the pandoc executable. This
1130+ is problematic for a few different reasons. First,
1131+ cabal-install will sometimes run the test suite after
1132+ building the library but before building the executable,
1133+ which means the executable isn't in place for the tests.
1134+ One can work around that by first building, then building
1135+ and running the tests, but that's fragile. Second, we
1136+ have to find the executable. So far, we've done that using
1137+ a function `findPandoc` that attempts to locate it
1138+ relative to the test executable (which can be located
1139+ using findExecutablePath). But the logic here is delicate
1140+ and work with every combination of options. To solve both
1141+ problems, we add an `--emulate` option to the
1142+ `test-pandoc` executable. When `--emulate` occurs as the
1143+ first argument passed to `test-pandoc`, the program simply
1144+ emulates the regular pandoc executable, using the rest of
1145+ the arguments (after `--emulate`). Thus, `test-pandoc
1146+ --emulate -f markdown -t latex` is just like `pandoc -f
1147+ markdown -t latex`. Since all the work is done by library
1148+ functions, implementing this emulation just takes a couple
1149+ lines of code and should be entirely reliable. With this
1150+ change, we can test the pandoc executable by running the
1151+ test program itself (locatable using `findExecutablePath`)
1152+ with the `--emulate` option. This removes the need for the
1153+ fragile `findPandoc` step, and it means we can run our
1154+ integration tests even when we're just building the
1155+ library, not the executable. [Note: part of this change
1156+ involved simplifying some complex handling to set
1157+ environment variables for dynamic library paths. I have
1158+ tested a build with `--enable-dynamic-executable`, and it
1159+ works, but further testing may be needed.]
1160+ + Print accurate location if a test fails (Albert
1161+ Krewinkel). Ensures that tasty-hunit reports the location
1162+ of the failing test instead of the location of the helper
1165+ * Documentation: Update URLs and use `https` where possible (#7122,
1168+ * Add `doc/libraries.md`, a description of libraries that support pandoc.
1172+ + MANUAL: block-level formatting is not allowed in line blocks (#7107).
1173+ + Clarify `tex_math_dollars` extension. Note that no blank lines
1174+ are allowed between the delimiters in display math.
1175+ + Add MANUAL section on reproducible builds.
1176+ + Document no template fallback for absolute path (#7077, Nixon
1178+ + Improve docs for cite-method.
1179+ + Update README and man page.
1181+ * Makefile: in `make bench`, create CSV files for comparison and compare
1182+ against previous benchmark run. Add timestamp to CSV filenames.
1184+ * cabal.project: don't explicitly set -trypandoc.
1185+ If we do, this can't be overridden on the cabal command line.
1187+ * doc/lua-filters.md: improve documentation for
1188+ `pandoc.mediabag.insert`, `pandoc.mediabag.fetch`,
1189+ `directory`, `normalize` (Albert Krewinkel).
1191+ * Allow base64-bytestring-1.2.* (Dmitrii Kovanikov)
1193+ * Require jira-wiki-markup 1.3.3 (Albert Krewinkel)
1195+ * Require citeproc 0.3.0.8, which correctly titlecases when titles
1196+ contain non-ASCII characters.
1198+ * Use skylighting 0.10.4. This version of skylighting uses xml-conduit
1199+ rather than hxt. This speeds up parsing of XML syntax definitions
1200+ fourfold, and removes four packages from pandoc's dependency graph:
1201+ hxt-charproperties, hxt-unicode, hxt-regex-xmlschema, hxt.
1203+ * Add script `tools/parseTimings.pl` to help pin down which
1204+ modules take the most time and memory to compile.
1206+ * Avoid unnecessary use of NoImplicitPrelude pragma (#7089) (Albert
1211+ + Use the lighter-weight tasty-bench instead of criterion.
1212+ + Run writer benchmarks for binary formats too.
1213+ + Alphabetize benchmarks.
1214+ + Don't run benchmarks for bibliography formats
1215+ (yet; we need a special input for them).
1216+ + Show allocation data
1217+ + Clean up benchmark code.
1218+ + Allow specifying patterns using `-p blah'.
1220+ * trypandoc: add 2 second timeout.
1222+ * Use `-split-sections` in creating linux release binary.
1223+ This reduces executable size significantly (by about 30%).
1225+ * Remove `weigh-pandoc`. It's not really useful any more, now that our
1226+ regular benchmarks include data on allocation.
1228+ * Improve linux package build process and add script to
1229+ automate building an arm64 binary package.
1232+## pandoc 2.11.4 (2021-01-22)
1234+ * Add `biblatex`, `bibtex` as output formats (closes #7040).
1236+ * Recognize more extensions as markdown by default (#7034):
1237+ `mkdn`, `mkd`, `mdwn`, `mdown`, `Rmd`.
1239+ * Implement defaults file inheritance (#6924, David Martschenko).
1240+ Allow defaults files to inherit options from other defaults files by
1241+ specifying them with the following syntax:
1242+ `defaults: [list of defaults files or single defaults file]`.
1244+ * Fix infinite HTTP requests when writing epubs from URL source (#7013).
1245+ Due to a bug in code added to avoid overwriting the cover image
1246+ if it had the form `fileX.YYY`, pandoc made an endless sequence
1247+ of HTTP requests when writing epub with input from a URL.
1251+ + Allow multiple pipe chars in todo sequences (Albert Krewinkel, #7014).
1252+ Additional pipe chars, used to separate "action" state from "no further
1253+ action" states, are ignored. E.g., for the following sequence, both
1254+ `DONE` and `FINISHED` are states with no further action required:
1255+ `#+TODO: UNFINISHED | DONE | FINISHED`.
1256+ + Restructure output of captioned code blocks (Albert Krewinkel, #6977).
1257+ The Div wrapper of code blocks with captions now has the class
1258+ "captioned-content". The caption itself is added as a Plain block
1259+ inside a Div of class "caption". This makes it easier to write filters
1260+ which match on captioned code blocks. Existing filters will need to be
1262+ + Mark verbatim code with class `verbatim` (Dimitri Sabadie, #6998).
1266+ + Handle `filecontents` environment (#7003).
1267+ + Put contents of unknown environments in a Div when `raw_tex` is not
1268+ enabled (#6997). (When `raw_tex` is enabled, the whole environment is
1269+ parsed as a raw block.) The class name is the name of the environment.
1270+ Previously, we just included the contents without the surrounding Div,
1271+ but having a record of the environment's boundaries and name can be
1274+ * Mediawiki reader:
1276+ + Allow space around storng/emph delimiters (#6993).
1278+ * New module Text.Pandoc.Writers.BibTeX, exporting
1279+ writeBibTeX and writeBibLaTeX. [API change]
1283+ + Revert table line height increase in 2.11.3 (#6996).
1284+ In 2.11.3 we started adding `\addlinespace`, which produced less dense
1285+ tables. This wasn't an intentional change; I misunderstood a comment in
1286+ the discussion leading up to the change. This commit restores the earlier
1287+ default table appearance. Note that if you want a less dense table, you
1288+ can use something like `\def\arraystretch{1.5}` in your header.
1292+ + Adjust internal links to identifiers defined in raw HTML sections
1293+ after splitting into chapters (#7000).
1294+ + Recognize `Format "html4"`, `Format "html5"` as raw HTML.
1295+ + Adjust internal links to images, links, and tables after splitting into
1296+ chapters. Previously we only did this for Div and Span and Header
1297+ elements (see #7000).
1301+ + Don't justify text inside table cells.
1305+ + Use `<element-citation>` if `element_citations`
1306+ extension is enabled (Albert Krewinkel).
1307+ + Fix citations (Albert Krewinkel, #7018). By default
1308+ we use formatted citations.
1309+ + Ensure that `<disp-quote>` is always wrapped in `<p>` (#7041).
1313+ + Cleaned up raw formats. We now react appropriately
1314+ to `gfm`, `commonmark`, and `commonmark_x` as raw formats.
1318+ + Fix bug with dropped content from inside spans with a class in
1319+ some cases (#7039).
1323+ + Handle table header using styles (#7008). Instead of hard-coding
1324+ the border and header cell vertical alignment, we now let this
1325+ be determined by the Table style, making use of Word's
1326+ "conditional formatting" for the table's first row. For
1327+ headerless tables, we use the tblLook element to tell Word
1328+ not to apply conditional first-row formatting.
1330+ * Commonmark writer:
1332+ + Implement start number on ordered lists (#7009). Previously they always
1333+ started at 1, but according to the spec the start number is respected.
1337+ + Fix implicit_figure at end of footnotes (#7006).
1339+ * ConTeXt template: Remove `\setupthinrules` from default template.
1340+ The width parameter this used is not actually supported,
1341+ and the command didn't do anything.
1343+ * Text.Pandoc.Extensions:
1345+ + Add `Ext_element_citations` constructor (Albert Krewinkel).
1347+ * Text.Pandoc.Citeproc.BibTeX: New unexported function
1348+ `writeBibtexString`.
1350+ * Text.Pandoc.Citeproc:
1352+ + Use finer grained imports (Albert Krewinkel).
1353+ + Factor out and export `getStyle` [API change].
1354+ + Export `getReferences` [API change, #7106].
1355+ + Factor out getLang.
1357+ * Text.Pandoc.Parsing: modify `gridTableWith'` for headerless tables.
1358+ If the table lacks a header, the header row should be an empty
1359+ list. Previously we got a list of empty cells, which caused
1360+ an empty header to be emitted instead of no header. In LaTeX/PDF
1361+ output that meant we got a double top line with space between.
1363+ * ImageSize: use `viewBox` for SVG if no length, width attributes (#7045).
1364+ This change allows pandoc to extract size information from more SVGs.
1366+ * Add simple default.nix.
1368+ * Use commonmark 0.1.1.3.
1370+ * Use citeproc 0.3.0.5.
1372+ * Update default CSL to use latest chicago-author-date.csl.
1374+ * CONTRIBUTING.md: add note on GNU xargs.
1378+ + Update description of `-L`/`--lua-filter`.
1379+ + Document use of citations in note styles (#6828).
1381+## pandoc 2.11.3.2 (2020-12-29)
1383+ * HTML reader: use renderTags' from Text.Pandoc.Shared (Albert Krewinkel).
1384+ A side effect of this change is that empty `<col>` elements are written
1385+ as self-closing tags in raw HTML blocks.
1387+ * Asciidoc writer: Add support for writing nested tables (#6972, timo-a).
1388+ Asciidoc supports one level of nesting. If deeper tables are to be
1389+ written, they are omitted and a warning is issued.
1391+ * Docx writer: fix nested tables with captions (#6983).
1392+ Previously we got unreadable content, because docx seems
1393+ to want a `<w:p>` element (even an empty one) at the end of
1396+ * Powerpoint writer: allow arbitrary OOXML in raw inline elements
1397+ (Albert Krewinkel). The raw text is now included verbatim in the
1398+ output. Previously is was parsed into XML elements, which prevented
1399+ the inclusion of partial XML snippets.
1401+ * LaTeX writer: support colspans and rowspans in tables (#6950,
1402+ Albert Krewinkel). Note that the multirow package is needed for
1403+ rowspans. It is included in the latex template under a variable,
1404+ so that it won't be used unless needed for a table.
1406+ * HTML writer: don't include p tags in CSL bibliography entries
1407+ (#6966). Fixes a regression in 2.11.3.
1409+ * Add `meta-description` variable to HTML templates (#6982). This
1410+ is populated by the writer by stringifying the `description`
1411+ field of metadata (Jerry Sky). The `description` meta tag will
1412+ make the generated HTML documents more complete and SEO-friendly.
1414+ * Citeproc: fix handling of empty URL variables (`DOI`, etc.).
1415+ The `linkifyVariables` function was changing these to links
1416+ which then got treated as non-empty by citeproc, leading
1417+ to wrong results (e.g. ignoring nonempty URL when empty DOI is present).
1418+ See jgm/citeproc#41.
1420+ * Use citeproc 0.3.0.3. Fixes an issue in author-only citations when
1421+ both an author and translator are present, and an issue with
1422+ citation group delimiters.
1424+ * Require texmath 0.12.1. This improves siunitx support in math,
1425+ fixes bugs with `\*mod` family operators and arrays, and avoids
1426+ italicizing symbols and operator names in docx output.
1428+ * Ensure that the perl interpreter used for filters with `.pl`
1431+ * MANUAL: note that textarea content is never parsed as Markdown
1432+ (Albert Krewinkel).
1435+## pandoc 2.11.3.1 (2020-12-18)
1437+ * Added some missing files to extra-source-files and data
1438+ files, so they are included in the sdist tarball. Closes #6961.
1439+ Cleaned up some extraneous data and test files, and added
1440+ a CI check to ensure that the test and data files included
1441+ in the sdist match what is in the git repository.
1443+ * Use citeproc 0.3.0.1, which avoids removing nonbreaking
1444+ space at the end of the `initialize-with` attribute. (Some
1445+ journals require nonbreaking space after initials, and this
1446+ makes that possible.)
1448+## pandoc 2.11.3 (2020-12-17)
1450+ * With `--bibliography` (or `bibliography` in metadata), a
1451+ URL may now be provided, and pandoc will fetch the resource.
1452+ In addition, if a file path is provided and it is not
1453+ found relative to the working directory, the resource
1454+ path will be searched (#6940).
1456+ * Add `sourcepos` extension for `commonmark`, `gfm`, `commonmark_x`
1457+ (#4565). With the `sourcepos` extension set set, `data-pos`
1458+ attributes are added to the AST by the commonmark reader. No other
1459+ readers are affected. The `data-pos` attributes are put on elements
1460+ that accept attributes; for other elements, an enlosing Div or Span
1461+ is added to hold the attributes.
1463+ * Change extensions for `commonmark_x`: replace `auto_identifiers`
1464+ with `gfm_auto_identifiers` (#6863). `commonmark_x` never actually
1465+ supported `auto_identifiers` (it didn't do anything), because the
1466+ underlying library implements gfm-style identifiers only. Attempts
1467+ to add the `auto_identifiers` extension to `commonmark` will now
1468+ fail with an error.
1472+ + Split module into several submodules (Albert Krewinkel). Reducing
1473+ module size should reduce memory use during compilation.
1474+ + Support advanced table features (Albert Krewinkel):
1475+ block level content in captions, row and colspans,
1476+ body headers, row head columns, footers, attributes.
1477+ + Disable round-trip testing for tables. Information for cell
1478+ alignment in a column is not preserved during round-trips (Albert
1480+ + Allow finer grained options for tag omission (Albert Krewinkel).
1481+ + Simplify list attribute handling (Albert Krewinkel).
1482+ + Pay attention to `lang` attributes on body element (#6938).
1483+ These (as well as `lang` attributes on the html element) should update
1485+ + Retain attribute prefixes and avoid duplicates (#6938).
1486+ Previously we stripped attribute prefixes, reading `xml:lang` as
1487+ `lang` for example. This resulted in two duplicate `lang`
1488+ attributes when `xml:lang` and `lang` were both used. This commit
1489+ causes the prefixes to be retained, and also avoids invald
1490+ duplicate attributes.
1492+ * Commonmark reader:
1494+ + Refactor `specFor`.
1495+ + Set input name to `""` to avoid clutter in sourcepos output.
1499+ + Parse `#+LANGUAGE` into `lang` metadata field (#6845, Albert
1501+ + Preserve targets of spurious links (#6916, Albert
1502+ Krewinkel). Links with (internal) targets that the reader doesn't
1503+ know about are converted into emphasized text. Information on the
1504+ link target is now preserved by wrapping the text in a Span of class
1505+ `spurious-link`, with an attribute `target` set to the link's
1506+ original target. This allows to recover and fix broken or unknown
1507+ links with filters.
1511+ + Table text width support (#6791, Nils Carlson).
1512+ Table width in relation to text width is not natively supported
1513+ by docbook but is by the docbook `fo` stylesheets through an XML
1514+ processing instruction, `<?dbfo table-width="50%"?>`.
1518+ + Improve parsing of command options (#6869, #6873).
1519+ In cases where we run into trouble parsing inlines til the
1520+ closing `]`, e.g. quotes, we return a plain string with the
1521+ option contents. Previously we mistakenly included the brackets
1523+ + Preserve center environment (#6852, Igor Pashev).
1524+ The contents of the `center` environment are put in a `Div`
1525+ with class `center`.
1526+ + Don't parse `\rule` with width 0 as horizontal rule. These are
1527+ sometimes used as spacers in LaTeX.
1528+ + Don't apply theorem default styling to a figure inside (#6925).
1529+ If we put an image in italics, then when rendering to Markdown
1530+ we no longer get an implicit figure.
1534+ + Handle unknown interwiki links better (#6932).
1535+ DokuWiki lets the user define his own Interwiki links. Previously
1536+ pandoc reacted to these by emitting a google search link, which is
1537+ not helpful. Instead, we now just emit the full URL including the
1538+ wikilink prefix, e.g. `faquk>FAQ-mathml`. This at least gives users
1539+ the ability to modify the links using filters.
1543+ + Properly handle boolean values in writing YAML metadata (#6388).
1544+ + Ensure that a new csl-block begins on a new line (#6921).
1545+ This just looks better and doesn't affect the semantics.
1549+ + Better image handling (#6948). An image alone in its paragraph
1550+ (but not a figure) is now rendered as an independent image, with an
1551+ `alt` attribute if a description is supplied. An inline image that
1552+ is not alone in its paragraph will be rendered, as before, using a
1553+ substitution. Such an image cannot have a "center", "left", or
1554+ "right" alignment, so the classes `align-center`, `align-left`, or
1555+ `align-right` are ignored. However, `align-top`, `align-middle`,
1556+ `align-bottom` will generate a corresponding `align` attribute.
1560+ + Keep raw openxml strings verbatim (#6933, Albert Krewinkel).
1561+ + Use Content instead of Element. This allows us to inject
1562+ raw OpenXML into the document without reparsing it into an
1563+ Element, which is necessary if you want to inject an open
1565+ + Fix bullets/lists indentation, so that the first level is slightly
1566+ indented to the right instead of right on the margin (cholonam).
1567+ + Support bold and italic in "complex script" (#6911).
1568+ Previously bold and italics didn't work properly in LTR
1569+ text. This commit causes the w:bCs and w:iCs attributes
1570+ to be used, in addition to w:b and w:i, for bold and
1571+ italics respectively.
1575+ + Fix image bounding box for custom widths/heighta (Mauro Bieg, #6936).
1579+ + Improve table spacing (#6842, #6860).
1580+ Remove the `\strut` that was added at the end of minipage
1581+ environments in cells. Replace `\tabularnewline` with
1582+ `\\ \addlinespace`.
1583+ + Improve calculation of column spacing (#6883).
1584+ + Extract table handling into separate module (Albert Krewinkel).
1585+ + Fix bug with nested `csl-` display Spans (#6921).
1586+ + Improve longtable output (#6883). Don't create minipages for
1587+ regular paragraphs. Put width and alignment information in the
1588+ longtable column descriptors.
1590+ * OpenDocument writer:
1592+ + Support for table width as a percentage of text width
1593+ (#6792, Nils Carson).
1594+ + Implement Div and Span ident support (#6755, Nils Carson).
1595+ Spans and Divs containing an ident in the Attr will become bookmarks
1596+ or sections with idents in OpenDocument format.
1597+ + Add two extensions, `xrefs_name` and `xrefs_number` (#6774, Nils
1598+ Carlson). Links to headings, figures and tables inside the
1599+ document are substituted with cross-references that will use the
1600+ name or caption of the referenced item for `xrefs_name` or the
1601+ number for `xrefs_number`. For the `xrefs_number` to be useful
1602+ heading numbers must be enabled in the generated document and
1603+ table and figure captions must be enabled using for example the
1604+ `native_numbering` extension. In order for numbers and reference
1605+ text to be updated the generated document must be refreshed.
1609+ + Support advanced table features (Albert Krewinkel).
1610+ + Support author affiliations (#6687, Albert Krewinkel).
1614+ + Use correct id attribute consistently (Jan Tojnar).
1615+ DocBook5 should always use `xml:id` instead of `id`.
1616+ + Handle admonition titles better (Jan Tojnar).
1617+ Docbook reader produces a `Div` with `title` class for `<title>`
1618+ element within an “admonition” element. Markdown writer then turns
1619+ this into a fenced div with `title` class attribute. Since fenced
1620+ divs are block elements, their content is recognized as a
1621+ paragraph by the Markdown reader. This is an issue for Docbook
1622+ writer because it would produce an invalid DocBook document from
1623+ such AST – the `<title>` element can only contain “inline”
1624+ elements. Handle this special case separately by unwrapping
1625+ the paragraph before creating the `<title>` element.
1626+ + Add XML namespaces to top-level elements (#6923, Jan Tojnar).
1627+ Previously, we only added `xmlns` attributes to chapter
1628+ elements, even when running with `--top-level-division=section`.
1629+ These namespaces are now added to part and section elements too,
1630+ when they are the selected top-level divisions.
1631+ We do not need to add namespaces to documents produced with
1632+ `--standalone` flag, since those will already have xmlns attribute on
1633+ the root element in the template.
1637+ + Fix handling of nested `csl-` display spans (#6921).
1638+ Previously inner Spans used to represent CSL display attributes were
1639+ not rendered as div tags as intended.
1643+ + Include title page in landmarks (#6919).
1644+ Note that the toc is also included if `--toc` is specified.
1645+ + Add frontmatter type on body element for nav.xhtml (#6918).
1647+ * EPUB templates: use preserveAspectRatio="xMidYMid" for cover image (#6895,
1648+ Shin Sang-jae). This change affects both the epub2 and the epub3
1649+ templates. It avoids distortion of the cover image by requiring that the
1650+ aspect ratio be preserved.
1654+ + Include `csquotes` package if `csquotes` variable set.
1655+ + Put back `amssymb`. We need it for checkboxes in todo lists,
1656+ and maybe for other things. In this location it seems compatible
1657+ with the cases that prompted #6469 and PR #6762.
1658+ + Disable language-specific shorthands in babel (#6817, #6887).
1659+ Babel defines "shorthands" for some languages, and these can
1660+ produce unexpected results. For example, in Spanish, `1.22`
1661+ gets rendered as `122`, and `et~al.` as `etal`.
1662+ One would think that babel's `shorthands=off` option (which
1663+ we were using) would disable these, but it doesn't. So we
1664+ remove `shorthands=off` and add some code that redefines
1665+ the shorthands macro. Eventually this will be fixed in babel,
1666+ I hope, and we can revert to something simpler.
1668+ * JATS template: allow array of persistent institute ids in `pid`
1669+ (Albert Krewinkel).
1671+ * Text.Pandoc.Parsing: minor code and efficiency improvements.
1673+ * Text.Pandoc.Extension:
1675+ + Add `Ext_sourcepos` constructor for `Extension` [API change].
1676+ + Add `Ext_xrefs_name` and `Ext_xrefs_number` constructors for
1677+ `Extension` (Nils Carson) [API change].
1679+ * Text.Pandoc.Citeproc:
1681+ + Fix truncation of `[Citation]` list in `Cite` inside footnotes (#6890).
1682+ This affected author-in-text citations in footnotes. It didn't cause
1683+ problems for the printed output, but for filters that expected the
1684+ citation id and other information.
1685+ + Allow the use of both inline and external references (#6951),
1686+ as with pandoc-citeproc. References defined in the document's
1687+ metadata take priority over references with the same id defined in
1688+ an external bibliography.
1689+ + Use `fetchItem` to get external bibliography (#6940).
1690+ + Ensure that BCP47 lang codes can be used. We ignore the variants
1691+ and just use the base lang code and country code when passing off
1693+ + Citeproc BibTeX parser: revert change in `getRawField`
1694+ which was made (for reasons forgotten) when transferring
1695+ this code from pandoc-citeproc. The change led to `--` in
1696+ URLs being interpreted as en-dashes, which is unwanted (#6874).
1698+ * Text.Pandoc.ImageSize:
1700+ + Default to DPI 72 if the format specifies DPI of 0 (#6880).
1701+ This shouldn't happen, in general, but it can happen with
1702+ JPEGs that don't conform to the spec. Having a DPI of 0
1703+ will blow up size calculations (division by 0).
1704+ + ImageSize: use JuicyPixels to determine size for png, jpeg, and
1705+ gif, instead of doing our own binary parsing (#6936). This
1706+ gives more reliable results.
1710+ + Remove `foldOrElse` (internal module) (Albert Krewinkel).
1712+ * Use skylighting 0.10.2 (#6625).
1714+ * Use citeproc 0.3. This fixes issues with references with
1715+ duplicate ids (jgm/citeproc#36).
1717+ * Use doctemplates 0.9. This fixes issues with boolean
1718+ metadata values in the Markdown writer (#6388)
1719+ and in `meta-json` (#6650). It also fixes
1720+ issues with nested for loops in templates.
1722+ * Add translations zh-Hans.yaml and zh-Hant.yaml (#6904, #6909,
1723+ Kolen Cheung, taotieren).
1725+ * Add tests: True to cabal.project.
1726+ This fixes some CI failures for cabal.
1728+ * Normalize test/tables/*.native (#6888, Kolen Cheung).
1730+ * Move executable to `app` directory to avoid problems with cabal repl.
1732+ * CONTRIBUTING: add section "How can I help?" (#6892, Albert Krewinkel).
1733+ Also adds a paragraph aimed at highlighting the importance of feature
1736+ * MANUAL: Document that --number-sections works in `ms` (#6935).
1739+## pandoc 2.11.2 (2020-11-19)
1741+ * Default to using ATX (`##`-style) headings for Markdown output
1742+ (#6662, Aner Lucero). Previously we used Setext (underlined) headings
1743+ by default for levels 1--2.
1745+ * Add option `--markdown-headings=atx|setext`, and deprecate
1746+ `--atx-headers` (#6662, Aner Lucero).
1748+ * Support `markdown-headings` in defaults files.
1750+ * Fix corner case in YAML metadata parsing (#6823). Previously YAML
1751+ metadata would sometimes not get recognized if a field ended with a
1752+ newline followed by spaces.
1754+ * `--self-contained`: increase coverage (#6854).
1755+ Previously we only self-contained attributes for certain tag names
1756+ (`img`, `embed`, `video`, `input`, `audio`, `source`, `track`,
1757+ `section`). Now we self-contain any occurrence of `src`,
1758+ `data-src`, `poster`, or `data-background-image`, on any tag; and
1759+ also `href` on `link` tags.
1763+ + Fix detection of locators following in-text citations.
1764+ Prevously, if we had `@foo [p. 33; @bar]`, the `p. 33` would be
1765+ incorrectly parsed as a prefix of `@bar` rather than a suffix of
1767+ + Improve period suppression algorithm for citations in notes
1768+ in note citation styles (#6835).
1769+ + Don't increment `stateNoteNumber` for example list references.
1770+ This helps with #6836 (a bug in which example list references
1771+ disturb calculation of citation note number and affect when
1772+ `ibid` is triggered).
1776+ + Move `getNextNumber` from Readers.LaTeX to Readers.LaTeX.Parsing.
1777+ + Fix negative numbers in siunitx commands. A change in pandoc 2.11
1778+ broke negative numbers, e.g. `\SI{-33}{\celcius}` or `\num{-3}`.
1779+ This fixes the regression.
1781+ * DocBook reader: drop period in formalpara title
1782+ and put it in a div with class `formalpara-title`, so that
1783+ people can reformat with filters (#6562).
1785+ * Man reader: improve handling of `.IP` (#6858). We now better handle
1786+ `.IP` when it is used with non-bullet, non-numbered lists, creating a
1787+ definition list. We also skip blank lines like groff itself.
1789+ * Bibtex reader: fall back on `en-US` if locale for LANG not found.
1790+ This reproduces earlier pandoc-citeproc behavior (jgm/citeproc#26).
1794+ + Wrap all tables (Albert Krewinkel).
1795+ All `<table>` elements are put inside `<table-wrap>` elements, as the
1796+ former are not valid as immediate child elements of `<body>`.
1797+ + Move Table handling to separate module (Albert Krewinkel).
1798+ Adds two new unexported modules:
1799+ Text.Pandoc.Writers.JATS.Types, Text.Pandoc.Writers.JATS.Table.
1803+ + Replace org #+KEYWORDS with #+keywords (TEC).
1804+ As of ~2 years ago, lower case keywords became the standard (though
1805+ they are handled case insensitive, as always).
1806+ + Update org supported languages and identifiers according to the
1807+ current list contained in
1808+ <https://orgmode.org/worg/org-contrib/babel/languages/index.html>
1811+ * Only use `filterIpynbOutput` if input format is ipynb (#6841).
1812+ Before this change content could go missing from divs with class
1813+ `output`, even when non-ipynb was being converted.
1815+ * When checking reader/writer name, check base name now that we permit
1816+ extensions on formats other than markdown.
1818+ * Text.Pandoc.PDF: Fix `changePathSeparators` for Windows (#6173).
1819+ Previously a path beginning with a drive, like `C:\foo\bar`, was
1820+ translated to `C:\/foo/bar`, which caused problems.
1821+ With this fix, the backslashes are removed.
1823+ * Text.Pandoc.Logging:
1825+ + Add constructor `ATXHeadingInLHS` to `LogMessage` [API change].
1826+ + Add constructor `EnvironmentVariableUndefined` to
1827+ `LogMessage` [API change].
1829+ * Fix error that is given when people specify `doc` output (#6834,
1832+ * LaTeX template: add a `\break` after parbox in `CSLRightInline`.
1833+ This should fix spacing problems between entries with numeric styles.
1834+ Also fix number of params on `CSLReferences`.
1836+ * reveal.js template: Put quotes around `controlsLayout`,
1837+ `controlsBackArrows`, and `display`, since these require strings.
1838+ Add `showSlideNumber`, `hashOneBasedIndex`, `pause`.
1840+ * Use citeproc 0.2. This fixes a bug with title case around parentheses.
1842+ * pandoc.cabal: remove 'static' flag.
1843+ This isn't really necessary and can be misleading (e.g. on macOS,
1844+ where a fully static build isn't possible). cabal's new option
1845+ `--enable-executable-static` does the same. On stack you can add
1846+ something like this to the options for your executable in package.yaml:
1848+ ld-options: -static -pthread
1850+ * Remove obsolete bibutils flag setting in `linux/make_artifacts.sh`.
1854+ + Correct `link-citation` -> `link-citations`.
1855+ + Add a sentence about `pagetitle` for HTML (#6843, Alex Toldaiev).
1857+ * INSTALL.md: Remove references to `pandoc-citeproc` (#6857).
1859+ * CONTRIBUTING: describe hlint and how it's used (#6840, Albert
1862+## pandoc 2.11.1.1 (2020-11-07)
1864+ * Citeproc: improve punctuation in in-text note citations (#6813).
1865+ Previously in-text note citations inside a footnote would sometimes have
1866+ the final period stripped, even if it was needed (e.g. on the end of
1869+ * Use citeproc 0.1.1.1. This improves the decision about when
1870+ to use `ibid` in cases where citations are used inside
1871+ a footnote (#6813).
1873+ * Support `nocase` spans for `csljson` output.
1875+ * Require latest commonmark, commonmark-extensions.
1876+ This fixes a bug with `autolink_bare_uris` and commonmark.
1878+ * LaTeX reader: better handling of `\\` inside math in table cells (#6811).
1880+ * DokuWiki writer: translate language names for code elements
1881+ and improve whitespace (#6807).
1883+ * MediaWiki writer: use `syntaxhighlight` tag instead of deprecated
1884+ `source` for highlighted code (#6810). Also support `startFrom`
1885+ attribute and `numberLines`.
1887+ * Lint code in PRs and when committing to master (#6790,
1890+ * doc/filters.md: describe technical details of filter invocations (#6815,
1894+## pandoc 2.11.1 (2020-11-03)
1896+ * DocBook Reader: fix duplicate bibliography bug (#6773, Nils Carlson).
1900+ + Parse contents of iframes (#6770).
1901+ + Parse inline svg as image unless `raw_html` is set in the reader (in
1902+ which case the svg is passed through as raw HTML) (#6770).
1906+ + Fix bug parsing macro arguments (#6796). If `\cL` is
1907+ defined as `\mathcal{L}`, and `\til` as `\tilde{#1}`, then
1908+ `\til\cL` should expand to `\tilde{\mathcal{L}}`, but
1909+ pandoc was expanding it to `\tilde\mathcal{L}`. This is
1910+ fixed by parsing the arguments in "verbatim mode" when the
1911+ macro expands arguments at the point of use.
1912+ + Properly support optional (cite) argument for
1913+ `\blockquote` from `csquotes` (#6802).
1915+ * LaTeX writer: Improved calculation of table column widths.
1916+ We now have LaTeX do the calculation, using `\tabcolsep`.
1917+ So we should now have accurate relative column widths no
1918+ matter what the text width. The default template has been modified to load
1919+ the calc package if tables are used.
1921+ * HTML writer: Fix duplicate "class" attribute for table
1922+ rows (Andy Morris).
1924+ * Text.Pandoc.Filter: allow shorter YAML representation of Citeproc
1925+ (Albert Krewinkel). The map-based YAML representation of filters expects
1926+ `type` and `path` fields. The path field had to be present for all filter
1927+ types, but is not used for citeproc filters. The field can now be omitted
1928+ when type is "citeproc", as described in the MANUAL.
1930+ * Text.Pandoc.Error: Add `PandocBibliographyError` constructor
1931+ for `PandocError` [API change]. This ensures that bibliography parsing
1932+ errors generate messages that include the bibliography file name --
1933+ otherwise it can be quite mysterious where it is coming from.
1935+ * Citeproc: properly handle `csl` field with `data:` URI (#6783).
1936+ This is used with the JATS writer, so this fixes a regression
1937+ in pandoc 2.11 with JATS output and citeproc.
1939+ * Allow `citation-abbreviations` in defaults file.
1941+ * JATS templates: ensure `jats_publishing` output is valid
1942+ (Albert Krewinkel).
1944+ * LaTeX template: Fix `CSLRightInline`, so that it does not
1945+ run over the right margin.
1947+ * HTML template: default CSS tweaks (Mauro Bieg and John
1950+ - Fix margin before codeblock
1951+ - Add `monobackgroundcolor` variable, making the background color
1952+ and padding of code optional.
1953+ - Ensure that backgrounds from highlighting styles take precedence over
1954+ monobackgroundcolor
1955+ - Remove list markers from TOC
1956+ - Add margin-bottom where needed
1957+ - Remove italics from blockquote styling
1958+ - Change borders and spacing in tables to be more consistent with other
1961+ - Set font-size for print media to 12pt.
1962+ - Reduce interline space.
1963+ - Reduce interparagraph space.
1964+ - Reduce line width.
1965+ - Remove the special `line-height: 1` for table cells.
1966+ - Remove the special line-height for pre.
1967+ - Ensure that there is a bit more space before a heading
1969+ - Slightly reduced space after title header.
1970+ - Add CSS example to MANUAL
1972+ * man template: Change comment that triggers `tbl` from
1973+ `.\"t` to `'\" t`, as specified in groff_man(7) (#6803).
1975+ * Use latest commonmark, commonmark-extensions.
1976+ This fixes a bug with nested blocks in footnotes with the
1977+ `footnote` extension to `commonmark`. See jgm/commonmark-hs#63.
1979+ * Citeproc: use comma for in-text citations inside footnotes.
1980+ When an author-in-text citation like `@foo` occurs in a footnote,
1981+ we now render it with: `AUTHOR NAME + COMMA + SPACE + REST`.
1982+ Previously we rendered: `AUTHOR NAME + SPACE + "(" + REST + ")"`.
1983+ This gives better results. Note that normal citations are still
1984+ rendered in parentheses.
1986+ * Use latest citeproc:
1988+ + citeproc no longer capitalizes notes, so we do it
1989+ in pandoc when appropriate.
1992+ * Clarify manual on `--track-changes` (#6801).
1994+ * Add `doc/jats.md` to document pandoc's handling of JATS
1995+ (#6794, Albert Krewinkel).
1997+ * Fix code example in lua-filters.md (#6795).
1999+## pandoc 2.11.0.4 (2020-10-21)
2001+ * Commonmark writer: fix regression with fenced divs (#6768).
2002+ Starting with 2.10.1, fenced divs no longer render with
2003+ HTML div tags in commonmark output. This is a regression
2004+ due to our transition from cmark-gfm. This commit fixes it.
2006+ * Use released version of citeproc. (This fixes a mis-step
2007+ in the 2.11.0.3 release, which is now deprecated.)
2009+ * Use latest sylighting, with support for `groovy`.
2011+ * Document that --html-q-tags requires the smart extension on the reader
2014+## pandoc 2.11.0.3 (2020-10-20)
2016+ * Use latest citeproc (closes #6765). This fixes a problem with
2017+ author-in-text citations for references including both an author
2018+ and an editor. Previously, both were included in the text, but only
2019+ the author should be.
2021+ * With `--citeproc`, ensure that the final period is removed when
2022+ citations that occur in notes in note-based styles get put in
2023+ parentheses. See jgm/citeproc#20.
2025+ * Normalize rewritten image paths with `--extract-media` (#6761).
2026+ This change will avoid mixed paths like this one when
2027+ `--extract-media` is used with a Word file:
2028+ ``. Instead we'll get
2029+ ``.
2031+ * Modify `--version` output. Use space more efficiently and report the
2032+ citeproc and ipynb versions, along with skylighting, texmath, and
2033+ pandoc-types. Drop the word "default" before "user data directory."
2035+ * DocBook reader: bibliomisc and anchor support (#6754, Nils Carlson).
2036+ Also ensure that bibliodiv without a title no longer results in
2039+ * ConTeXt template: adds `\setupinterlinespace` to fonts larger
2040+ than normal (#6763, Denis Maier).
2042+ * LaTeX template: Do not load amssymb if not needed (#6469, Angelo Peronio).
2043+ See <https://tex.stackexchange.com/a/549938>.
2045+ * Relax upper bound on hslua, allow hslua-1.3.* (Albert Krewinkel).
2049+ + Improve explanation of "indent" variable (#6767, Cyrus Yip).
2050+ + Remove org from list of input formats supporting raw_tex (#6753,
2054+## pandoc 2.11.0.2 (2020-10-15)
2056+ * Fix handling of `xdata` in bibtex/biblatex bibliographies (#6752).
2058+ * Fix some small typos in the API documentation (#6751, Michael Hoffmann).
2060+ * Require citeproc 0.1.0.2. This fixes a regression from pandoc-citeproc
2061+ involving spacing between components of a reference in certain styles
2064+ * Fix typos in comments, doc strings, error messages, and tests
2065+ (Albert Krewinkel, #6738).
2067+## pandoc 2.11.0.1 (2020-10-13)
2069+ * LaTeX reader: support more acronym commands (#6746):
2070+ `\acl`, `\aclp`, and capitalized versions of already
2071+ supported commands.
2073+ * Commonmark reader: add `pipe_table` extension *after* defaults (#6739).
2074+ Otherwise we get bad results for non-table, non-paragraph
2075+ lines containing pipe characters.
2077+ * Markdown writer: Fix autolinks rendering for gfm (#6740).
2078+ Previously, autolinks rendered as raw HTML, due to the
2079+ `class="uri"` added by pandoc's markdown reader.
2083+ + Escape option values in lstlistings environment (#6742).
2084+ + Fix handling of `lang` `pt-BR` (#2953). For polyglossia we now
2085+ use `\setmainlanguage[variant=brazilian]{portuguese}` and for babel
2086+ `\usepackage[shorthands=off,main=brazilian]{babel}`.
2088+ * Depend on latest citeproc (0.1.0.1).
2090+ + This fixes the citation number issue with ieee.csl and other
2091+ styles that do not explicitly sort bibliographies (#6741). (Pandoc
2092+ was numbering them by their order in the bibliography file,
2093+ rather than the order cited, as required by the CSL spec.)
2094+ + Fixes groupin/collapsing with citation items with prefixes.
2096+ * default.latex: fix `CSLReference` macro definition.
2098+ * Fix MANUAL.txt CSL JSON conversion examples.
2100+ * Fix spelling errors in changelog, MANUAL.txt, `doc/org.md` (#6738).
2103+## pandoc 2.11 (2020-10-11)
2105+ * Add `--citeproc` (`-C`) option to trigger built-in citation processing.
2106+ It is no longer necessary to use the external `pandoc-citeproc`
2107+ filter. `--citeproc` behaves like a filter and can be positioned
2108+ relative to other filters as they appear on the command line.
2110+ The new built-in citation processing uses the citeproc library,
2111+ a reimplementation of CSL citation processing that fixes many
2112+ of the shortcomings of pandoc-citeproc. In general, citation
2113+ processing should work much the same as it did with pandoc-citeproc,
2114+ but with greater fidelity to CSL styles and better performance.
2115+ (The tests from the pandoc-citeproc package have been carried
2116+ over to pandoc.) The following differences should be noted:
2118+ - At this point, only some of the writers (HTML, ms, LaTeX) properly
2119+ interpret CSL display styles. You should get decent output in
2120+ all formats (at least as good as with pandoc-citeproc), but
2121+ indentation and block-alignment may not be right.
2123+ - pandoc-citeproc searches the `~/.csl` directory for `.csl`
2124+ styles. Pandoc instead searches the `csl` subdirectory of
2125+ the pandoc user data directory (e.g., `~/.pandoc/csl` or
2126+ `~/.local/share/pandoc/csl`). Users who already keep
2127+ CSL styles in `~/.csl` may find it convenient to add a
2130+ - Some of the bibliography formats supported by pandoc-citeproc (via
2131+ hs-bibutils) are no longer supported: Copac, EndNote,
2132+ ISI, MEDLINE, MODS, and RIS. If you use one of these formats,
2133+ you may use the `bibutils` utility to convert to BibLaTeX.
2134+ We now support only BibTeX, BibLaTeX, CSL JSON,
2135+ and pandoc's YAML/Markdown analogue of CSL JSON.
2137+ - pandoc-citeproc would always retrieve the independent parent
2138+ of a dependent style by doing an HTTP request. pandoc will
2139+ now first seek the independent parent locally (in the resource
2140+ path or in the `csl` subdirectory of the pandoc user data
2141+ directory) before resorting to HTTP. In addition, you may
2142+ omit the `.csl` extension, e.g. `--csl zoology`.
2144+ - Using the `--bibliography` option (or including
2145+ `bibliography` in YAML metadata) no longer triggers
2146+ citation processing implicitly: one must always use the
2147+ `--citeproc` option if citation processing is wanted.
2149+ * Add `csljson` as and input and output format. This allows pandoc
2150+ to convert between `csljson` and other bibliography formats
2151+ (e.g. `-f csljson -t markdown -s` or `-f bibtex -t csljson`),
2152+ and to generate formatted versions of CSL JSON bibliographies
2153+ (e.g., `pandoc -f csljson --citeproc pl.json -o pl.pdf`).
2155+ * Added `bibtex`, `biblatex` as input formats. This allows pandoc
2156+ to convert between BibLaTeX and BibTeX and other bibliography formats,
2157+ and to generated formatted versions of BibTeX/BibLaTeX bibliographies
2158+ (e.g., `pandoc -f biblatex --citeproc pl.bib -o pl.pdf`).
2160+ * Raise informative errors when YAML metadata parsing fails (#6730).
2161+ Previously the command would succeed, returning empty metadata,
2162+ with no errors or warnings.
2164+ * Sort languages in `--list-highlight-languages` output (#6718,
2165+ Albert Krewinkel). Languages were previously sorted by their
2166+ long name, which leads to unexpected results).
2168+ * Add CSS to default HTML template (#6601, Mauro Bieg). This
2169+ greatly improves the default typography in pandoc's HTML
2170+ output. The CSS is sensitive to a number of variables
2171+ (e.g. `mainfont`, `fontsize`, `linestretch`): see the manual for
2172+ details. To restore the earlier, more spartan output, you can
2173+ disable this with `-M document-css=false`.
2175+ * Support `--toc-depth` option for ODT writer (#6696, niszet).
2177+ * Fix issues with Windows UNC paths with some options (#5127).
2179+ * Remove `fenced_code_blocks` and `backtick_code_blocks` from allowed
2180+ `commonmark` and `gfm` extensions. These shouldn't really be counted
2181+ as extensions, because they can't be disabled in commonmark.
2182+ Adjust markdown writer to check for the commonmark variant in addition
2185+ * Add these extensions to `gfm` and `commonmark`:
2186+ `fenced_code_blocks`, `backtick_code_blocks`, `fenced_code_attributes`.
2187+ These can't really be disabled in the reader, but they need to be enabled
2188+ in the writer or we just get indented code.
2190+ * Make sure proper set of extensions is recognized for `commonmark_x`.
2192+ * Allow `gfm_auto_identifiers`, `ascii_identifiers` extensions for `docx`.
2196+ + Add `Maybe FilePath` parameter to `yamlToMeta` [API change].
2197+ + Export `yamlToRefs` [API change], a version of `yamlToMeta`
2198+ specialized to references.
2199+ + Set `citationNoteNum` accurately in citations.
2200+ + Revise abbreviation support. Don't insert a nonbreaking space after a
2201+ potential abbreviation if it comes right before a note or citation.
2202+ This causes problems for citeproc's moving of note citations.
2206+ + Support missing siunitx commands (#6658).
2207+ + Support `squared`, `cubed`, `tothe` in siunitx (#6657).
2208+ + Improved uncertainty handling in slunitx.
2209+ + Factored out siunitx stuff into separate unexposed module.
2210+ + Fix improper empty cell filtering (#6689, Christian Despres).
2211+ + Fix parsing of "show name" in `\newtheorem` (#6734).
2212+ Previously we were just treating it as a string and
2213+ ignoring accents and formatting.
2214+ + Prevent wrong nesting of `\multirow` and `\multicolumn` table
2215+ cells (#6603, Laurent P. René de Cotret).
2216+ + Table cell parser not consuming spaces correctly (#6596,
2217+ Laurent P. René de Cotret).
2218+ + Change `SIRange` to `SIrange` (#6617, Emerson Harkin).
2219+ + Allow blank lines inside `\author` (#6324).
2223+ + Don't squelch space at end of emphasis and other inline elements;
2224+ instead, move it outside the element (#6719).
2225+ + Implement table cell alignment (#6698, Nils Carlson).
2226+ + Implement column span support for tables (#6492, Nils Carlson).
2227+ + Update list of block level tags (#6610).
2232+ + Don't squelch space at end of emphasis and other inline elements;
2233+ instead, move it outside the element (#6719).
2237+ + Apply `.. class::` directly to following Header rather than creating
2238+ a surrounding Div (#6699).
2242+ + Allow empty dates in comments and tracked changes (#6726, Diego
2247+ + Be less aggressive about using quotes for YAML values,
2248+ allowing e.g. a quotation mark or bracket as long as it's not at the
2249+ beginning of the line.
2250+ + Use double quotes for YAML metadata (#6727).
2251+ + Sort YAML metadata keys in Markdown output case-insensitive.
2256+ + Support asciidoctor's block figures (#6538, argent0).
2260+ + Fix spacing issue with list in definition list.
2261+ When a list occurs at the beginning of a definition list definition,
2262+ it can start on the same line as the label, which looks bad.
2263+ Fix that by starting such lists with an `\item[]`.
2267+ + Support intermediate table headers (#5314, Albert Krewinkel).
2268+ + Support attributes on all table elements (Albert Krewinkel).
2269+ + Render table footers if present (#6314, Albert Krewinkel).
2270+ + Fix addition of `doc-biblioentry` role.
2271+ + Support colspans and rowspans in HTML tables (#6312, Albert Krewinkel).
2275+ + Support internal document links (#5541, Leonard Rosenthol).
2276+ + Changed default link state to invisible (#6676, Leonard Rosenthol).
2280+ + Better handle list items whose contents are lists (#5948, Michael
2281+ Hoffmann). If the first element of a bulleted or ordered list is
2282+ another list, then that first item would previously disappear if the
2283+ target format is docx.
2284+ + Separate adjacent tables (#4315). Word combines adjacent tables,
2285+ so to prevent this we insert an empty paragraph between two
2290+ + Don't force blank line after headings (#6554).
2292+ * OpenDocument writer:
2294+ + Implement table cell alignment (#6700 Nils Carson, Mauro Bieg).
2295+ + New table cell support with row and column spans (#6682, Nils Carson).
2296+ + Syntax highlighting for inline code (#6711, niszet).
2298+ * Add Text.Pandoc.Citeproc module, exporting `processCitations`
2299+ [API change]. This depends on several other, unexported
2300+ modules under Text.Pandoc.Citeproc.
2302+ * Add module Text.Pandoc.Writers.CslJson, exporting `writeCslJson`.
2305+ * Add module Text.Pandoc.Readers.CslJson, exporting `readCslJson`.
2308+ * Add module Text.Pandoc.Readers.BibTeX, exporting `readBibTeX` and
2309+ `readBibLaTeX`. [API change]
2311+ * Text.Pandoc.Filter: Add `CiteprocFilter` constructor to Filter.
2312+ [API change] This runs the processCitations transformation.
2313+ We need to treat it like a filter so it can be placed
2314+ in the sequence of filter runs (after some, before others).
2315+ In FromYAML, this is parsed from `citeproc` or `{type: citeproc}`,
2316+ so this special filter may be specified either way in a defaults file
2317+ (or by `citeproc: true`, though this gives no control of positioning
2318+ relative to other filters).
2320+ * Add new exported module Text.Pandoc.Writers.AnnotatedTable [API change]
2321+ (#6655, Christian Despres). This module (which should generally
2322+ be imported qualified to avoid name conflicts) provides a
2323+ `Table` type that mirrors the structure of a pandoc `Table`,
2324+ but with added inferred information so that the writers do not have to
2325+ lay out tables themselves. The `toTable` and `fromTable` functions convert
2326+ between an annotated `Table` and a regular pandoc `Table`. In addition to
2327+ producing a `Table` with coherent and well-formed annotations, the
2328+ `toTable` function also normalizes its input table like the table
2329+ builder does. Tests ensure that `toTable` normalizes tables exactly
2330+ like the table builder, and that its annotations are coherent.
2332+ * Text.Pandoc.Logging:
2334+ + Remove unused `CouldNotParseYamlMetadata` constructor for `LogMessage`
2336+ + Add `CiteprocWarning` constructor to `LogMessage` [API change].
2338+ * Text.Pandoc.Readers.Metadata: export `yamlBsToRefs` [API change].
2339+ These allow specifying an id filter so we parse only references
2340+ that are used in the document.
2342+ * Text.Pandoc.Parsing:
2344+ + Export ParseError [API change].
2345+ + Add `stateInNote` and `stateNoteNumber` to `ParserState`
2346+ [API change]. These are used to populate note numbers for citation
2349+ * Fix apparent typos in sample.lua (#6729, William Lupton).
2350+ Also make the writer less aggressive in escaping quotes.
2352+ * Text.Pandoc.Options:
2354+ + `defaultMathJaxURL`: use `tex-chtml-full` instead of `tex-mml-chtml`
2355+ (#6599, Kolen Cheung). This drops the MathML support (which we
2356+ don't need for HTML math rendering) and includes the full JavaScript,
2357+ which makes it possible to use `--self-contained` (though there may
2358+ still be issues if the required math fonts aren't available). This
2359+ change should also reduce latency in pages with lots of formulas.
2360+ + Add `/tex-chtml-full.js` to `defaultMathJaxURL` (#6593) Previously we
2361+ added this in processing command line options, but not in processing
2362+ defaults files, which was inconsistent.
2364+ * epub.css: Fix cover page selectors and add note explaining their use
2367+ * Add data files needed for Text.Pandoc.Citeproc: these include
2368+ `default.csl` in the data directory and a `citeproc` directory that
2369+ is only used at compile-time for biblatex localizations. Note that we've
2370+ added `file-embed` as a mandatory rather than a conditional
2371+ dependency, because of the biblatex localization files.
2375+ + Add SimpleTable for backwards compatibility (#6575, Albert Krewinkel).
2376+ A new type `SimpleTable` is made available to Lua filters. It is
2377+ similar to the `Table` type in pandoc versions before 2.10;
2378+ conversion functions from and to the new Table type are provided.
2379+ Old filters using tables now require minimal changes and can use, e.g.,
2381+ if PANDOC_VERSION > {2,10,1} then
2382+ pandoc.Table = pandoc.SimpleTable
2387+ function Table (tbl)
2388+ tbl = pandoc.utils.to_simple_table(tbl)
2390+ return pandoc.utils.from_simple_table(tbl)
2393+ to work with the current pandoc version.
2395+ + Make `attr` argument optional in `Table` constructor (Albert Krewinkel).
2396+ This changes the Lua API. It is highly unlikely for this change to affect
2397+ existing filters, since the documentation for the new Table constructor
2398+ (and type) was incomplete and partly wrong before. The Lua API is now
2399+ more consistent, as all constructors for elements with attributes now
2400+ take attributes as the last parameter.
2404+ * Add a dedicated Citations section which consolidates the information
2405+ the manual used to contain about citation processing, and incorporates
2406+ some information formerly found in the pandoc-citeproc man page.
2407+ + Add note about lualatex using `selnolig`.
2408+ + Remove duplicate `seriespage` (#6568, Blake Eryx).
2409+ + Remove lists of support extensions for markdown variants (#6604).
2410+ Instead, offer the advice to use `--list-extensions=FORMAT`.
2411+ + Fix position of attributes in header (Albert Krewinkel).
2412+ + Delete obsolete section on compact and loose lists (#6684).
2415+ * doc/lua-filters.md:
2417+ + Add info on how to debug Lua filters (#6732, Ian Max Andolina).
2418+ + Document Underline type and constructor (Albert Krewinkel).
2419+ + Document `body` field (Albert Krewinkel).
2420+ + Add missing header attribute
2421+ + Add missing Link.title field (Albert Krewinkel).
2422+ + Make the setting-the-date example conditional (the-solipsist).
2423+ This makes the example a bit more realistic.
2424+ + Remove outdated link table example.
2428+ + Add section on tables (Albert Krewinkel).
2429+ + Add section on handling of unknown directives (Albert Krewinkel).
2431+ * CONTRIBUTING.md: fix typo (#6584, Dmitry Volodin).
2433+ * Use golden test framework for command tests. This means that
2434+ `--accept` can be used to update expected output.
2436+ * Use the `smart` extension when generating pandoc's man page (#6613).
2438+ * Release-candidate: don't build windows i386.
2439+ So far we haven't been able to figure out how to get
2440+ stack to use a 32-bit ghc.
2442+ * Use `null` instead of deprecated `Builder.isNull`.
2446+ + Fix macospkg target to fetch target from S3 artifacts.
2447+ + Fix pandoc-templates target to include all partials.
2449+ * Remove duplicated dependency in pandoc.cabal (#6591, Felix Yan).
2451+ * Sort build depends in pandoc.cabal alphabetically (#6691,
2454+ * Add .travis.yml for macos release candidate build (#6622).
2455+ We need to build the release candidate on Travis rather
2456+ than GitHub actions, because GH has macos 10.15, and
2457+ binaries compiled on that OS will not work with 10.13.
2458+ This build is only triggered on `rc/*` branches.
2460+ * Remove instructions for building pandoc-citeproc from CI and
2461+ release binary build instructions. We will no longer distribute
2464+ * Fix math rendering in trypandoc (this broke after
2465+ commit d8ad766d17603784b86fc5c2e1b22864125d04d1).
2467+ * Use latest versions of skylighting, commonmark (#6589),
2468+ comonmark-extensions, commonmark-pandoc, texmath.
2470+ * Relax version bounds for hslua, hslua-module-text, bytestring.
2472+ * Use released pandoc-types 1.22. This changes the JSON
2473+ encoding slightly for the new table types introduced
2474+ in 1.21, so they're more consistent with the rest.
2475+ Developers of libraries for pandoc filters will want
2478+ * Fix hlint suggestions, update hlint.yaml (#6680, Christian Despres).
2480+ * Code cleanup (#6678, Joseph C. Sible).
2482+ * Add haddocks to functions in Text.Pandoc.Writers.Shared (Albert
2485+ * Remove duplicate `tshow` definition.
2487+ * Linux release candidate build: use ghc-musl container. This simplifies
2488+ our build process (over using a customized alpine container).
2491+## pandoc 2.10.1 (2020-07-23)
2493+ * Add `commonmark_x` output format. This is `commonmark` with a number
2494+ of useful pandoc extensions enabled.
2496+ * Many more extensions now work with `commonmark` and `gfm`.
2498+ * Add generic `attributes` extension. This allows attributes to
2499+ be added to any block or inline element in a uniform way. Since
2500+ the Pandoc AST doesn't include attributes on each element type,
2501+ the attributes will sometimes be added by creating a surrounding
2502+ Div or Span container. Currently this extension is only
2503+ compatible with the `commonmark` and `gfm` readers.
2505+ To add an attribute to a block-level element, e.g. a paragraph,
2506+ put it before the block:
2509+ This is a paragraph.
2511+ Multiple attributes may be used and will be combined:
2514+ {.blue .warning key="val"}
2515+ This is a paragraph.
2517+ To add an attribute to an inline-level element, put it
2518+ immediately after the element:
2520+ *emphasized text*{.special}
2523+ * Support `--number-sections` for docx output (#1413).
2527+ + Support `\SIRange` reader (#6418, Emerson Harkin).
2528+ + Support table col-span and row-span (#6311, Laurent P. René de Cotret).
2529+ Supports `\multirow` and `\multicolumn`.
2530+ + Support amsthm: `\newtheorem`, `\theoremstyle`, and theorem and
2531+ proof environments, including labels and references. The only thing
2532+ that is unsupported is the second optional argument, which causes
2533+ numbering to be reset after the specified series is incremented.
2534+ + Moved some code to T.P.LaTeX.Parsing. We need to reduce the size
2535+ of the LaTeX reader to ease compilation on resource-limited systems.
2539+ + Fix csv tables with multiline cells (#6549).
2540+ + Fix spurious newlines in some attributes from directives.
2541+ + Avoid extra newline in included code blocks.
2543+ * Commonmark reader:
2545+ + Switch from cmark-gfm to commonmark-hs for commonmark and gfm
2546+ parsing. This avoids depending on a C library
2547+ and allows us to support more pandoc extensions for
2548+ `commonmark` and `gfm`.
2552+ + Parse releaseinfo as metadata (#6542).
2556+ + Only use `bCs/iCs` on runs with `rtl` or `cs` property (#6514, Nikolay
2558+ + Code cleanup/refactoring (Nikolay Yakimov).
2560+ * Org reader (Albert Krewinkel):
2562+ + Respect export setting which disables entities
2563+ MathML-like entities, e.g., `\alpha`, can be disabled with the
2564+ `#+OPTION: e:nil` export setting (Albert Krewinkel).
2565+ + Respect export setting disabling footnotes. Footnotes can be removed
2566+ from the final document with the `#+OPTION: f:nil` export setting.
2567+ + Respect tables-excluding export setting. Tables can be removed from
2568+ the final document with the `#+OPTION: |:nil` export setting.
2572+ + Move `asciify` out of `escapeString`. Otherwise `unsmartify`
2573+ doesn't catch quotes that have already been turned to entities.
2574+ + Add `writeCommonmark` (new exported function, API change).
2575+ + Use unicode super/subscript characters when possible if the
2576+ `superscript` or `subscript` extension or `raw_html` aren't available.
2577+ + Render caption as following paragraph when `table_caption` extension
2579+ + Use numerical labels for reference links that are longer
2580+ than 999 characters or contain square brackets, for conformity
2581+ with commonmark (#6560).
2583+ * Commonmark writer:
2585+ + Instead of using cmark-gfm, use `writeCommonmark` from the
2586+ Markdown writer. This function calls the markdown writer
2587+ with appropriate extensions and a few small modifications
2588+ (e.g. not requiring backslashes before spaces inside
2589+ super/subscripts). With this change `comonmark` and
2590+ `gfm` output can be used with a wider selection of
2593+ * Jira writer: keep image caption as alt attribute (#6529, Albert
2598+ + Improve alt-text/caption handling for HTML5 (#6491, Albert Krewinkel).
2599+ Screen readers read an image's `alt` attribute and the figure caption,
2600+ both of which come from the same source in pandoc. The figure caption is
2601+ hidden from screen readers with the `aria-hidden` attribute. This
2602+ improves accessibility. For HTML4, where `aria-hidden` is not allowed,
2603+ pandoc still uses an empty `alt` attribute to avoid duplicate contents.
2607+ + Fix code highlighting with blank lines. Previously blank lines
2608+ were simply omitted from highlighted code.
2609+ + Escape starting periods in ms writer code blocks (#6505, Michael
2610+ Hoffmann). If a line of ms code block output starts with a period (.),
2611+ it should be prepended by `\&` so that it is not interpreted as a roff
2614+ * Text.Pandoc.Extensions:
2616+ + Add `raw_markdown` extension (which only affects `ipynb`
2618+ + Trim down `githubMarkdownExtensions`.
2619+ Previously it included all of the following, which make
2620+ sense for the legacy `markdown_github` but not for `gfm`,
2621+ since they are part of base commonmark and thus
2622+ can't be turned off in `gfm`:
2624+ - `Ext_all_symbols_escapable`
2625+ - `Ext_backtick_code_blocks`
2626+ - `Ext_fenced_code_blocks`
2627+ - `Ext_space_in_atx_header`
2628+ - `Ext_intraword_underscores`
2629+ - `Ext_lists_without_preceding_blankline`
2630+ - `Ext_shortcut_reference_links`
2632+ These have been removed from `githubMarkdownExtensions`, though
2633+ they're still turned on for legacy `markdown_github`.
2634+ + Add `Ext_attributes` constructor for `Extension` [API change].
2636+ * LaTeX template: use selnolig to selectively suppress ligatures with
2639+ * Benchmark bytestring readers (Nikolay Yakimov).
2643+ + Update using-the-pandoc-api.md (favonia).
2644+ + Fix Typos in lua-filters.md (tajmone).
2645+ + Rewrite Raw HTML/TeX section in MANUAL.txt to avoid duplicate
2646+ headings for the extensions.
2647+ + Fix typo in MANUAL.txt (Benjamin Wuethrich).
2648+ + Remove duplicate 'titlepage' in MANUAL.txt (Blake Eryx).
2649+ + CONTRIBUTING.md: Advertise the official nightlies in GitHub actions.
2650+ Replaces #6500, thanks to @ickc.
2654+## pandoc 2.10 (2020-06-29)
2656+ * Use pandoc-types 1.21. This adds two things:
2658+ + A native Underline constructor for Inline (#6277, Vaibhav Sagar).
2659+ + More expressive types for tables (#1024, Christian Despres).
2660+ Tables can now take attributes; and rowspans and colspans,
2661+ column headers, multiple row headers, table head and foot
2662+ can all be represented. (Note, however, that reader and
2663+ writer support for these features is still lacking, so
2664+ most users won't see any differences in table conversion
2665+ yet. These changes just lay the foundation for further
2668+ * Support new Underline element in readers and writers (#6277,
2671+ * Support new Table type (Christian Despres).
2672+ The Builder.simpleTable now only adds a row to the TableHead
2673+ when the given header row is not null. This uncovered an
2674+ inconsistency in the readers: some would unconditionally
2675+ emit a header filled with empty cells, even if the header
2676+ was not present. Now every reader has the conditional
2677+ behaviour. Only the XWiki writer depended on the header row
2678+ being always present; it now pads its head as necessary.
2680+ * Add an option to disable certificate validation (#6156, Cédric Couralet,
2681+ Cécile Chemin, Juliette Fourcot). This commit adds the option
2682+ `--no-check-certificate`, which disables certificate checking when
2683+ resources are fetched by HTTP.
2685+ * Unify defaults and markdown metadata parsers (#6328, Nikolay
2686+ Yakimov). Clean up code in Text.Pandoc.Readers.Metadata and
2687+ properly handle errors in `yamlToMeta`. This fixes parsing
2688+ of Boolean fields in metadata withinin defaults files and reduces
2693+ + Implement `<procedure>` (#6442, Mathieu Boespflug).
2694+ + Implement `<phrase>` (#6438, Mathieu Boespflug).
2695+ + Treat envar and systemitem like code (#6435, Mathieu Boespflug).
2696+ + Implement `<replaceable>` (#6437, Mathieu Boespflug)
2697+ + Map `<simplesect>` to unnumbered section (#6436, Mathieu Boespflug).
2701+ + Handle "label" element in section title (#6288).
2702+ + Parse abstract element into metadata field of same name
2703+ (#6480, Albert Krewinkel).
2705+ * Jira reader (Albert Krewinkel):
2707+ + Resolve multiple parsing problems, including issues with empty
2708+ table cells, faulty recognition of closing emphasis characters, and
2709+ parsing of image attributes (#6212, #6219, #6220).
2710+ + Two consecutive markup chars are now parsed verbatim (#6343);
2711+ styled text must not be empty.
2712+ + Newlines are no longer allowed within styled text (#6325).
2713+ + Links to anchors are now parsed as links (#6407).
2714+ + Retain image attributes (#6234). Jira images attributes as in
2715+ `!image.jpg|align=right!` are retained as key-value pairs. Thumbnail
2716+ images, such as `!example.gif|thumbnail!`, are marked by a
2717+ `thumbnail` class in their attributes.
2718+ + Use Underline for inserted text (#6237). Previously, the span was
2719+ marked with the non-standard class `inserted`.
2720+ + Improve icon conversion for `(/)`, `(x)`, `(!)`, `(?)`
2721+ `(+)`, `(-)`, `(off)`, `(*)`. (#6236, #6264).
2722+ + Support citations, attachment links, and user links (#6231, #6238,
2724+ + Resolve parsing issues of blockquote, color (#6233, #6235).
2728+ + Parse attributes into table attributes.
2729+ + Support `<bdo>` (#5794, Tristan de Cacqueray).
2730+ + Add `summary` to list of block-level HTML tags (#6385).
2731+ This improves support for summary/details inside Markdown.
2732+ NOTE: you need to include a blank line before the closing
2733+ `</details>`, if you want the last part of the content to
2734+ be parsed as a paragraph.
2735+ + Fix parsing unclosed th elements in a table (#6247).
2737+ * Commonmark reader: Implement `implicit_figures` extension (#6350).
2741+ + Fix inline code in lists (#6284, Nikolay Yakimov).
2742+ Previously inline code containing list markers was sometimes
2743+ parsed incorrectly.
2744+ + Don't require blank line after grid table (#6481).
2745+ This allows grid tables to be enclosed in fenced divs with no
2746+ intervening blank lines.
2750+ + Don't parse beyond `\end{document}` (#6380).
2751+ This required some internal changes to `\subfile` handling.
2752+ + Better handling of `\lettrine`. SmallCaps instead of Span
2753+ for the part after the initial capital. Ensure that both
2754+ arguments are parsed, so that in Markdown both
2755+ are treated as raw LateX. (Closes #6258.)
2757+ * Org reader (Albert Krewinkel):
2759+ + Recognize images with uppercase extensions (#6472).
2760+ + Keep unknown keyword lines as raw org. The lines of unknown
2761+ keywords, like `#+SOMEWORD: value` are no longer read as metadata,
2762+ but kept as raw `org` blocks. This ensures that more information
2763+ is retained when round-tripping org-mode files; additionally,
2764+ this change makes it possible to support non-standard org
2765+ extensions via filters.
2766+ + Unify keyword handling. Handling of export settings and other
2767+ keywords (like `#+LINK`) has been combined and unified.
2768+ + Support `LATEX_HEADER_EXTRA` and `HTML_HEAD_EXTRA`
2769+ settings. These export settings are treated like their non-extra
2770+ counterparts, i.e., the values are added to the `header-includes`
2772+ + Allow multiple `#+SUBTITLE` export settings. The values of all
2773+ lines are read as inlines and collected in the `subtitle`
2775+ + Read `#+INSTITUTE` values as text with markup. The value is
2776+ stored in the `institute` metadata field and used in the
2777+ default beamer presentation template.
2778+ + The behavior of the `#+AUTHOR` and `#+KEYWORD` export
2779+ settings has changed: Org now allows multiple such lines
2780+ and adds a space between the contents of each line. Pandoc
2781+ now always parses these settings as meta inlines; setting
2782+ values are no longer treated as comma-separated lists.
2783+ Note that a Lua filter can be used to restore the previous
2785+ + Read description lines as inlines (#6485). `#+DESCRIPTION` lines
2786+ are now treated as text with markup. If multiple such
2787+ lines are given, then all lines are read and separated by soft
2789+ + Honor tex export option (#4070). The `tex` export option can be set
2790+ with `#+OPTION: tex:nil` and allows three settings:
2791+ `t` (the default) causes LaTeX fragments to be parsed as TeX or added
2792+ as raw TeX. `nil` removes all LaTeX fragments from the document.
2793+ `verbatim` treats LaTeX as text.
2798+ + Pass arbitrary attributes through in code blocks (#6465).
2799+ Exceptions: name (which becomes the id), class (which becomes the
2800+ classes), and number-lines (which is treated specially to fit
2801+ with pandoc highlighting).
2802+ + Handle `date::` directive (#6276).
2804+ * Textile reader: support `pre.` for code blocks (#6454).
2808+ + Implement `raw_markdown` extension (#5408). Specifying
2809+ `-f ipynb+raw_markdown` will cause Markdown cells
2810+ to be represented as raw Markdown blocks, instead of being
2811+ parsed. This is not what you want when going from `ipynb`
2812+ to other formats, but it may be useful when going from `ipynb`
2813+ to Markdown or to `ipynb`, to avoid semantically insignificant
2814+ changes in the contents of the Markdown cells that might
2815+ otherwise be introduced.
2816+ + Handle application/pdf output as image (#6430).
2817+ + Properly handle image/svg+xml as an image (#6430).
2821+ + Distinguish between docx parsing and docx container unpacking errors.
2823+ * MediaWiki reader:
2825+ + Fix `gfm_auto_identifiers` so that `-` is not replaced by `_` (#6335).
2829+ + Add nested syntax highlighting (#6256, Vlad Hanciuta).
2830+ Nested syntaxes are specified like this:
2833+ SELECT * FROM table
2836+ The preformatted code block parser has been extended to check if the
2837+ first attribute of the block is not a `key=value` pair, and in that
2838+ case it will be considered as a class.
2841+ * Jira writer (Albert Krewinkel):
2843+ + Always escape braces (#6478). Braces are now always escaped, even
2844+ within words or when surrounded by whitespace. Jira and
2845+ Confluence treat braces specially.
2846+ + Convert Underline to inserted text (`+inserted+`).
2847+ + Add image attributes (#6234). Image attributes are added
2848+ to the output as image parameters. If the image has a
2849+ class "thumbnail", then a thumbnail image is generated;
2850+ all other attributes are discarded in this case.
2854+ + Ensure that `-M csquotes` works even in fragment mode (#6265).
2855+ + Escape `^` specially for listings (#6460).
2856+ + Create hypertarget for links with identifier (#6360).
2857+ + Distinguish between single and double quotes when using enquote
2858+ package (#6457, dbecher-ito).
2859+ + Add support for customizable alignment of columns in beamer (#6331,
2861+ + Add support for customizable alignment of columns in beamer
2862+ (#4805, #4150, andrebauer).
2866+ + Use CSS in favor of `<br>` for display math (#6372)
2867+ Some CSS to ensure that display math is
2868+ displayed centered and on a new line is now included
2869+ in the default HTML-based templates; this may be
2870+ overridden if the user wants a different behavior.
2874+ + Clean-up Div handling (Albert Krewinkel).
2878+ + Enable column and row bands for tables (#6371).
2879+ This change will not have any effect with the default style.
2880+ However, it enables users to use a style (via a reference.docx)
2881+ that turns on row and/or column bands.
2883+ * OpenDocument (and ODT) writer:
2885+ + Add custom-style "Abstract" in metadata abstract.
2886+ This ensures that the abstract is rendered with style Abstract.
2887+ + Enable custom-style attribute on a Div.
2888+ This allows you to apply a custom style to contained paragraphs.
2892+ + Add id of figure to enclosed image.
2893+ + Add personname element to docbook author (#6244).
2897+ + Properly handle cover-image containing spaces (#6391).
2901+ + Ensure consistent padding for pipe tables (#6240).
2902+ + Avoid unnecessary escapes before intraword `_` when
2903+ `intraword_underscores` extension is enabled (#6296).
2907+ + Properly handle images with same alt text (#6194).
2908+ Previously we created duplicate references for these in rendering RST.
2912+ + Add blank line after Div (#6308).
2916+ + Support Haddock tables (Joe Hermaszewski).
2917+ See this PR on Haddock for details on the table format:
2918+ <https://github.com/haskell/haddock/pull/718>.
2920+ * PowerPoint writer (Jesse Rosenthal):
2922+ + Write math input verbatim in speaker notes (#6301).
2923+ OMML in speaker notes would lead to corrupt PowerPoint output. We now
2924+ output the OMML verbatim as LaTeX in the speaker notes.
2926+ * LaTeX template: Make polyglossia package options list-aware
2927+ (#6444, Frederik Elwert).
2929+ * Reveal.js template:
2931+ + Update template for reveal.js 4.0.0 (#6390, Salim B).
2932+ + Update template with newly available options (#6347, Jake Zimmerman).
2933+ + Use CDN version of revealjs v4 by default (#6408).
2935+ * opendocument template: Add abstract and subtitle to opendocument
2938+ * reference.odt: clean up styles. Add Abstract.
2939+ Change Author, Date to centered paragraphs with no character
2942+ * epub.css: wrap overlong lines in highlighted code blocks (#6242).
2943+ This fixes a problem in iBooks v2.4 with our earlier
2944+ horizontally scrolling code blocks. The problem seems to
2945+ be a bug in iBooks, not pandoc, but since iBooks is a major
2946+ target we're changing pandoc's default behavior so that
2947+ pandoc-produced epubs work on that platform.
2951+ + Use `--enable-local-file-access` in invoking `wkhtmltopdf` (#6474).
2952+ `wkhtmltopdf` changed in recent versions to require this for
2953+ access to local files. This fixes PDF via HTML5 with `--css`.
2954+ + Send verbose output to stderr, not stdout (#6483).
2956+ * Text.Pandoc.MIME: Fix MIME type for TrueType fonts in EPUBs
2957+ (#6464, Michael Reed).
2959+ * Text.Pandoc.Shared:
2961+ + `makeSections`: omit number attribute when unnumbered class
2962+ is present (#6339). Previously the attribute was included but given
2963+ an empty value, and this caused the table of contents creation
2964+ functions in Text.Pandoc.Writers.Shared to think these items had
2965+ numbers, which meant that they were included in the TOC even if the
2966+ `unlisted` class was used.
2967+ + Deprecate `underlineSpan` in Shared in favor of
2968+ `Text.Pandoc.Builder.underline` (Vaibhav Sagar).
2969+ + `renderTags'`: use self-closing tag for col element (#6295).
2971+ * Text.Pandoc.UUID: Fix `getRandomUUID`, which previously would
2972+ return the same value twice in a row. Make `getRandomUUID`
2973+ polymorphic in PandocMonad. Remove `getUUID` (#6228, Joseph C. Sible).
2975+ * Text.Pandoc.Class: Generalize `PandocIO` functions to `MonadIO`.
2977+ * Fixed Katex standalone script (#6399, Lucas Escot).
2978+ Global macros are now persistent when using the HTML Writer with the
2981+ * Lua subsystem (Albert Krewinkel):
2983+ + Use new type PandocLua for all pandoc Lua operations (API change).
2984+ The new type `PandocLua` is an instance of the `PandocMonad` typeclass
2985+ and can thus be used in a way similar to `PandocIO`.
2986+ + Use PandocError for exceptions in Lua subsystem (API change).
2987+ The PandocError type is used throughout the Lua subsystem. All Lua
2988+ functions throw an exception of this type if an error occurs. The
2989+ `LuaException` type is removed and no longer exported from
2990+ `Text.Pandoc.Lua`. In its place, a new constructor `PandocLuaError`
2991+ is added to PandocError.
2993+ * Lua filters: improve error messages for failing filters (#6332,
2994+ Albert Krewinkel). Print the Lua error properly instead of
2995+ displaying their `show` string.
2997+ * Use latest skylighting. This fixes a bug with lua multiline
2998+ comments (and may improve handling of other syntaxes as well).
2999+ IT also adds `aria-hidden="true"` to the empty a elements, which
3000+ helps people who use screen readers.
3002+ * Use latest texmath.
3004+ * Require latest doctemplates 0.8.2.
3005+ This adds support for template pipes `first`, `rest`,
3006+ `last`, `allbutlast`.
3008+ * Revert 0e48a02 and dependency on base-noprelude, which hasn't
3009+ been updated for ghc 8.10 (see #6187).
3011+ * Dependency adjustments:
3013+ + Allow haddock-library 1.9.x.
3014+ + Allow hslua 1.1 (#6243, Felix Yan).
3015+ + Allow base64-bytestring 1.1.
3016+ + Use latest jira-wiki-markup.
3017+ + Allow http-client 0.7.
3018+ + Allow tasty 1.3.x.
3019+ + Allow aeson 1.5 (#6400, Felix Yan).
3020+ + Remove unused dependency `vector` (#6462, Laurent P. René de Cotret).
3021+ + Bump QuickCheck upper bound.
3023+ * Significant code cleanup and simplification (Joseph C. Sible, #6223,
3024+ #6209, #6225, #6229, #6226, #6340).
3026+ * Remove unnecessary hlint ignores (#6341, Joseph C. Sible).
3028+ * Remove obsolete RelaxedPolyRec extension (#6487, Nikolay Yakimov).
3030+ * trypandoc improvements (Mike Tzou):
3032+ + Add standalone option to the command text (#6210).
3033+ + Update third party libraries.
3037+ + Clarify template partial naming (#6476, Mauro Bieg).
3038+ + Describe `jira` as "Jira/Confluence wiki markup" (#6351, Albert
3039+ Krewinkel). In the past, Jira's wiki markup was also used by -- and
3040+ could be imported into -- Atlassian Confluence.
3041+ + Add link to print-css.rocks (#6272, Mauro Bieg).
3042+ + Clarify pipe table column width adjustment (#6254).
3043+ + Fix ATX header syntax.
3044+ + Fix misleading note about image size conversions (#6353).
3045+ + Update links to reveal.js documentation (#6386, Salim B).
3046+ + Separate adjacent verbatim code blocks (#6307, tom-audm).
3050+ + Document behavior of `smart` extension (#4387, Albert Krewinkel).
3051+ + Describe all supported export options in detail.
3055+ + Fix description of BulletList Lua type (Levi Gruspe).
3056+ + Use pandoc.system module in TikZ example (Albert
3057+ Krewinkel). Showcase temporary directory handling with
3058+ `with_temporary_directory` and `with_working_directory`.
3060+ * INSTALL.md: fix FreeBSD port link (#6422, Mo).
3061+ The FreeBSD port was renamed from pandoc to hs-pandoc in 2010.
3062+ The old pandoc port is still at version 1.5.1.1
3064+ * Propagate `(DY)LD_LIBRARY_PATH` in tests (#6376, Lila).
3066+ * Bump `cabal-version` to 2.2 (#6377).
3068+ * Make it possible to compile using Stack on NixOS (#6439, Mathieu
3071+ * CI action to check for commit message length (Nikolay Yakimov, #6398).
3074+## pandoc 2.9.2.1 (2020-03-23)
3076+ * Markdown reader: Fix table alignment when heading begins with t (#6153).
3077+ Due to a typo (`t` instead of `\t`) we were center aligning column
3078+ headings that begin with a lowercase `t`!
3080+ * Text.Pandoc.Readers.Roff:
3082+ + Fix parsing of `\.` in man/ms readers (#6175).
3083+ Previously due to a typo it was being parsed as `` ` ``.
3084+ + Fix parsing of `\'` in man/ms readers (#6175). It was being parsed
3087+ * Jira reader (Albert Krewinkel):
3089+ + Fix parsing of tables without preceding blankline (#6198).
3090+ A bug was fixed which caused faulty parsing if a table was not
3091+ preceded by a newline and the first table cell had no space
3092+ after the initial `|` characters.
3093+ + Fix parsing of strikeout, emphasis (#6196). A bug was fixed which
3094+ caused non-emphasized text containing digits and/or non-special
3095+ symbols (like dots) to sometimes be parsed incorrectly.
3096+ + Support colored inline text, indented lists (#6183, #6184).
3100+ + Fix definition lists so indent even when paragraph indent is
3101+ set to 0 (as is the default). Also ensure indent for display math
3102+ that falls back to TeX.
3103+ + Use `.QS/.QE` instead of `.RS/.RE` for block quotes.
3105+ * EPUB writer: fix regression on detection of front/back/bodymatter
3106+ (#6170). This bug caused sections with epub:type `dedication` to be
3107+ misplaced in bodymatter instead of frontmatter as specified
3108+ in the manual. The same problem would affect other epub:types.
3109+ The pattern matching needed to be changed with the use of
3112+ * AsciiDoc writer: remove redundant `otherwise` guard in
3113+ `inlineToAsciiDoc` (#6146, Ryan Scott).
3115+ * Text.Pandoc.Class:
3117+ + Fix missing import when data files are not embedded (Albert Krewinkel).
3118+ + Subdivide Text.Pandoc.Class into small unexported modules
3119+ and ensure that all functions have Haddock documentation (#6106,
3121+ + Finer grained imports of Text.Pandoc.Class submodules (#6203, Albert
3124+ * Text.Pandoc.XML: Add to list of HTML5 attributes:
3125+ `allow`, `autocapitalize`, `decoding`, `enterkeyhint`,
3126+ `imagesizes`, `imagesrcset`, `loading`.
3128+ * Use implicit Prelude (#6187, Albert Krewinkel).
3129+ The previous behavior was introduced as a fix for #4464. It seems that
3130+ this change alone did not fix the issue, and `stack ghci` and `cabal
3131+ repl` only work with GHC 8.4.1 or newer, as no custom Prelude is loaded
3132+ for these versions. Given this, it seems cleaner to revert to the
3135+ * Always use custom prelude (#6187, Albert Krewinkel).
3136+ Previously, the custom prelude was used only with older GHC versions, as
3137+ a workaround for problems with ghci. The ghci problems are resolved by
3138+ replacing package `base` with `base-noprelude`, allowing for consistent
3139+ use of the custom prelude across all GHC versions.
3141+ * Remove outdated checks for no longer supported base versions
3142+ (Albert Krewinkel).
3144+ * PDF via wkhtmltopdf: put user-specified options last (#6171).
3145+ Certain options (e.g. `cover`) need to come after flags on
3148+ * Text.Pandoc.App: set resource path at the beginning so it can affect
3149+ things like include-in-header (#5982).
3151+ * Change macOS release candidate CI process so that notarized
3152+ packages can be produced (#6169).
3154+ * Make MANUAL more explicit about nonbreaking space handling by
3155+ `all_symbols_escapable` (#6154, Fabien Schurter).
3157+ * trypandoc (Mike Tzou):
3159+ + Add checkbox for standalone option (#6189).
3160+ + Use strict mode for JavaScript code (#6188).
3161+ + Fetch resources over https (#6188).
3162+ + Remove unnecessary attributes on style, script elements (#6188).
3164+ * Use details tag to make GitHub releases changelog collapsible.
3166+ * Update filter code in doc/filters.md so it works with latest pandoc
3169+ * linux/Dockerfile: upgrade to alpine 3.11 (#6180, Albert Krewinkel).
3170+ This is used to build the static linux binaries.
3173+## pandoc 2.9.2 (2020-02-15)
3175+ * Add `csv` as an input format (#6100). The CSV table is converted into a
3176+ pandoc simple table. A new module Text.Pandoc.Readers.CSV
3177+ exports `readCSV` [API change].
3179+ * Introduce new format variants for JATS writer (#6014, Albert Krewinkel):
3181+ - `jats_archiving` for the "Archiving and Interchange Tag Set",
3182+ - `jats_publishing` for the "Journal Publishing Tag Set", and
3183+ - `jats_articleauthoring` for the "Article Authoring Tag Set."
3185+ The `jats` output format is now an alias for `jats_archiving`.
3186+ The module Text.Pandoc.Writers.JATS now exports
3187+ `writeJatsArchiving`, `writeJatsPublishing`, and
3188+ `writeJatsArticleAuthoring`, as well as the legacy
3189+ `writeJATS` [API change].
3191+ * `--defaults`: Support `bibliography` and `csl` fields.
3192+ Move `addMeta` from Text.Pandoc.App.CommandLineOptions to
3193+ Text.Pandoc.App.Opt (internal change).
3195+ * Add timing info for filters in `--verbose` mode (#6112).
3196+ When verbose mode is specified (verbosity == INFO), print a
3197+ notice when running a filter and when a filter completes (including
3202+ + Allow `&` in LaTeX citation keys (#6110).
3203+ + Improve caption and label parsing.
3204+ + Don't emit empty Span elements for labels.
3205+ + Put tables with labels in a surrounding Div.
3206+ + Resolve `\ref` to table numbers (#6137).
3207+ + Skip comments in more places where this is needed (#6114).
3208+ + Allow beamer overlays for all commands in all raw tex (#6043).
3209+ This affects parsing of raw tex in LaTeX and in Markdown and
3211+ + Improve parsing of raw environments (#6034). If parsing fails
3212+ in a raw environment (e.g. due to special characters like unescaped
3213+ `_`), try again as a verbatim environment, which is less sensitive to
3214+ special characters. This allows us to capture special environments
3215+ that change catcodes as raw tex when `-f latex+raw_tex` is used.
3219+ + Add highlight directive (#6140, Lucas Escot).
3221+ * MediaWiki writer:
3223+ + Prevent triple `[[[` which confuses MediaWiki (#6119).
3227+ + Don't parse `data-id` as `id` attribute. And similarly don't
3228+ parse any `data-X` as `X` when `X` is a valid HTML attribute.
3232+ + Simplify parsing of sub- and superscripts (#6127, Albert Krewinkel).
3233+ Speeds up parsing of single-word, markup-less sub- and superscripts.
3237+ + Group biblatex citations even with prefix and suffix (#5849, Ethan
3238+ Riley). Previously biblatex citations were only grouped if there
3239+ was no prefix. This patch allows them to be grouped in subgroups split
3240+ by prefixes and suffixes, which allows better citation sorting.
3241+ + Fix regression in handling of columns in beamer slides (#6033).
3242+ Columns in title slides were causing problems with
3244+ + Fix duplicate frame classes in LaTeX/Beamer output (#6107).
3248+ + Fix duplicate attributes on headings (#6062), regression from 2.7.x.
3249+ + Fix `--number-offset` with HTML TOC. Eventually it would be worth
3250+ adding a parameter to `makeSections` so this could be done at that
3251+ level; then it would also affect other writers that construct
3253+ + reveal.js: restore old behavior for 2D nesting (#6032).
3254+ The fix to #6030 actually changed behavior, so that the
3255+ 2D nesting occurred at slide level N-1 and N, instead of
3256+ at the top-level section. This commit restores the v2.7.3 behavior.
3257+ If there are more than 2 levels, the top level is horizontal
3258+ and the rest are collapsed to vertical.
3259+ + reveal.js: ensure that pauses work even in title slides (#5819).
3263+ + Fix regression: spurious dots in markdown_mmd metadata output (#6133).
3267+ + Fix regression with Compact style on tight lists (#6072).
3268+ Starting in 2.8, the docx writer no longer distinguishes between tight
3269+ and loose lists, since the Compact style is omitted. This is a
3270+ side-effect of the fix to #5670, as explained in the changelog. This
3271+ patch fixes the problem by extending the exception currently offered to
3272+ Plain blocks inside tables to Plain blocks inside list items.
3276+ + Fix output of table headers (Albert Krewinkel, #6035).
3278+ * Add Text.Pandoc.Image with unexported svgToPng.
3280+ * Text.Pandoc.XML: Export `html5Attributes`, `html4Attributes`,
3281+ `rdfaAttributes` (formerly unexported in Text.Pandoc.Writers.HTML).
3284+ * Text.Pandoc.Shared: Export a new function `findM` (#6125,
3287+ * Text.Pandoc.Logging: Add `RunningFilter`, `FilterCompleted`
3288+ constructors to LogMessage [API change].
3290+ * Text.Pandoc.CSV: fix bug in CSV parser; previously an extra blank record
3291+ would sometimes be inserted at the end.
3293+ * LaTeX template: add space option to xeCJK with PassOptionsToPackage
3294+ (#6002). Otherwise we can get a clash with documentclasses that
3295+ already load the package.
3299+ + Allow filtering of element lists (#6038, Albert Krewinkel). Lists of
3300+ Inline and Block elements can now be filtered via `Inlines` and
3301+ `Blocks` functions, respectively. This is helpful if a filter
3302+ conversion depends on the order of elements rather than a single
3303+ element. For example, the following filter can be used to remove all
3304+ spaces before a citation:
3306+ function isSpaceBeforeCite (spc, cite)
3307+ return spc and spc.t == 'Space'
3308+ and cite and cite.t == 'Cite'
3311+ function Inlines (inlines)
3312+ for i = #inlines-1,1,-1 do
3313+ if isSpaceBeforeCite(inlines[i], inlines[i+1]) then
3320+ + Add methods `insert`, `remove`, and `sort` to pandoc.List
3321+ (Albert Krewinkel). Example of use:
3323+ local numbers = pandoc.List {2, 3, 1}
3324+ numbers:sort() -- numbers is now {1, 2, 3}
3325+ + Make `pandoc.List` a callable constructor (Albert Krewinkel).
3326+ It is now possible to construct a new List via
3327+ `pandoc.List()` instead of `pandoc.List:new()`.
3328+ + Add tests for pandoc.List module (Albert Krewinkel).
3330+ * Text.Pandoc.App.CommandLineOptions: Change `setVariable` to use `Text`
3331+ instead of `String`. This avoids some unnecessary unpacking.
3333+ * Use versioned directory for windows release zipfile.
3334+ Also remove old `make-windows-installer.bat`, superseded by GitHub
3335+ actions workflow, and modify `pandoc.wxs` for new paths.
3337+ * Extensive code cleanup (#6141, #6128, #6129, #6130, #6123,
3338+ #6105, 6102, #6117, #6124, #6115, #6116, #6111, Joseph C. Sible).
3340+ * Fix hlint warnings (Albert Krewinkel).
3342+ * Use latest doclayout, doctemplates (#6031). The new version of
3343+ doclayout fixes a memory leak that affected `--include-in-header` with
3344+ large files (and possibly other cases involving extremely long lines).
3346+ * Use latest texmath.
3348+ * Use latest skylighting and fix test suite (#6086).
3350+ * sample.lua: Fix typo in descriptive comments (#6136, Caleb Maclennan).
3351+ Fix typo in error message (#6135).
3353+ * Add Docker and GH Actions instructions/links to INSTALL.md.
3355+ * Update filter documentation (#6065). Improve cabal v2 instructions.
3356+ Remove example using pandoc API directly (we have other
3357+ docs for that and it was outdated).
3361+ + Cross-link constructors and types (Albert Krewinkel).
3362+ Thanks to @bpj for the idea.
3363+ + Sort pandoc.List methods alphabetically (Albert Krewinkel).
3364+ + Unify, fix anchors and internal links (#6061, Albert Krewinkel).
3365+ Links and anchors now follow consistent conventions, like
3366+ lowercase-only anchor names. This breaks some links to specific
3367+ sections in the document, but will make it much easier to link
3368+ documentation in the future.
3369+ + Clarify filter function execution order (#6059, Albert Krewinkel).
3371+ * In docs, update URLs and use `https:` wherever possible (#6090,
3374+## pandoc 2.9.1.1 (2020-01-05)
3378+ + Fix parsing bug affected indented code after raw HTML (#6009, #5360).
3382+ + Fix regression in beamer slide structure with certain slide levels
3384+ + Allow framebreaks for beamer's TOC (Heiko Schlittermann, #6012)
3385+ + Properly handle unnumbered headings level 4+ (#6018).
3386+ Previously the `\paragraph` command was used instead of
3387+ `\paragraph*` for unnumbered level 4 headings.
3391+ + Fix revealjs slide structure regression with certain slide levels
3393+ + Add newlines to make slide show output more readable.
3397+ + Remove extra spaces from table cells (Albert Krewinkel, #6024).
3399+ * JATS template: Update JATS dtd (Arfon Smith, #6020). Use the archiving
3400+ and interchange DTD rather than the more restrictive journal publishing
3401+ DTD (which doesn't permit ext-link as a valid child).
3403+ * Text.Pandoc.PDF: Fix `runTeXProgram` so that the input source is always
3404+ overwritten (#6027). Previously it wasn't overridden if the file already
3405+ existed, which led to bad results on subsequent runs when
3406+ `pdf-engine-opt=-output-directory=` was used to specify an explicit temp
3409+ * Text.Pandoc.BCP47: Change `getLang` to handle block-level contents
3410+ (#6008). Some readers (e.g. RST) will populate the `lang` metadata field
3411+ with block-level content. `getLang` has been modified to handle this.
3412+ Previously in these cases the LaTeX writer would not properly set the
3413+ "main language" of the document.
3415+ * Fix `test/tables.org` (Albert Krewinkel).
3417+ * Use HTTPS in copyright message (Felix Yan, #6010)
3420+## pandoc 2.9.1 (2019-12-23)
3422+ * Add Jira reader (Albert Krewinkel, #5556).
3424+ * Jira writer: use jira-wiki-markup renderer (Albert Krewinkel,
3425+ #5926). The following improvements are included in this change:
3427+ + non-jira raw blocks are fully discarded instead of showing
3429+ + table cells can contain multiple blocks;
3430+ + unnecessary blank lines are removed from the output;
3431+ + markup chars within words are properly surrounded by
3433+ + preserving soft linebreaks via `--wrap=preserve` is
3436+ Note that backslashes are rendered as HTML entities, as there
3437+ appears no alternative to produce a plain backslash if it is
3438+ followed by markup. This may cause problems when used with
3439+ confluence, where rendering seems to fail in this case.
3441+ * Fix regression with `--number-sections`. Starting with 2.8,
3442+ `--number-sections` also had the effect of `--section-divs`,
3443+ even if `--section-divs` was not specified.
3445+ * Improved table of contents generation in markdown, RTF,
3446+ commonmark, better handling cases where section headings are
3449+ * Ensure that later default file values for `variable` replace
3450+ earlier ones (5988).
3452+ * HTML reader: Add `nav` to list of block-level tags.
3454+ * Org reader (Albert Krewinkel):
3456+ + Wrap named table in Div, using name as id (#5984).
3457+ Tables which are given a name via `#+NAME:` or `#+LABEL:`
3458+ are wrapped in an additional Div, with the name set as the
3460+ + Report parsing errors properly.
3461+ + Fix parsing problem for colons in headline (#5993).
3463+ * Text.Pandoc.PDF: Ensure UTF8 when printing source in
3464+ `--verbose` mode, avoiding an error on platforms that
3465+ default to something other than UTF-8 (#5997).
3467+ * Text.Pandoc.Templates: Strip directory before trying to find
3468+ partial in data files (#5987).
3470+ * Text.Pandoc.Shared: Improve `makeSections` so we don't get
3471+ doubled "number" attributes in EPUB output (or anywhere
3474+ * Added tests for `--toc` and `--section-divs`.
3476+ * Text.Pandoc.MIME: Added glsl MIME type for WebGL maps (#6000,
3479+ * MANUAL: A bit clearer explanation for `--base-header-level`.
3480+ We now say exactly how to translate between the deprecated
3481+ `--base-header-level` and `--shift-heading-level-by`.
3485+ + Remove spurious dot in title (#5996, Mauro Bieg).
3486+ + Replace metadata example with image centering (#6004,
3487+ Albert Krewinkel). Thanks to @efx for proposing this filter.
3490+## pandoc 2.9 (2019-12-11)
3492+ * Text.Pandoc.Templates [API change]
3494+ + Add Monad wrappers `WithDefaultPartials` and `WithPartials`.
3495+ Wrapping these around an instance of `PandocMonad` gives
3496+ us different instances of `TemplateMonad`, with different
3497+ search behavior in retrieving partials.
3498+ To compile a template and limit partial search to pandoc's
3499+ data files, use `runWithDefaultPartials (compileTemplate ...)`.
3500+ To compile a template and allow partials to be found locally
3501+ (either on the file system or via HTTP, in the event that
3502+ the main template has an absolute URL), ue
3503+ `runWithPartials (compileTemplate ...)`.
3504+ + Export `getTemplate`, which seeks a template locally,
3505+ or via HTTP if the template has an absolute URL, falling
3506+ back to the data files if not found.
3507+ + Export `compileDefaultTemplate` -- does `getDefaultTemplate`
3508+ and compiles the result, raising an error on failure.
3510+ * Text.Pandoc.Class [API change]
3512+ + Remove `TemplateMonad` instances for `PandocIO` and `PandocPure`.
3513+ These were too limiting and caused a bug whereby a local
3514+ partial could be used even when the default template was requested.
3515+ We now rely on instances provided in the Templates module.
3517+ * Text.Pandoc.App.OutputSettings: Simplify template retrieval code.
3519+ * ConTeXt template: Adjust to title formatting (#5949, Denis Maier).
3520+ Add `\setupinterlinespace` to `title`, `subtitle`, `date` and `author`
3521+ elements: otherwise longer titles that run over multiple lines will look
3522+ squashed as `\tfd` etc. won't adapt the line spacing to the font size.
3524+ * reveal.js template: Add title-slide-attributes variable (#5981,
3527+ * More informative JSON parse error (#5973).
3529+ * Use external emojis package (forked from pandoc). Removed emoji data
3530+ in Text.Pandoc.Emoji.
3532+ * Fix regression in `makeSections` (#5965).
3533+ Previously `hierarchicalize` (the ancestor of `makeSections`) would put
3534+ header attributes on the containing Div. In 2.8 this behavior changed,
3535+ which broke some tools depending on pandoc. Here we roll back this change,
3536+ so that attributes again migrate from the header to the containing Div when
3537+ `makeSections` is run. Note that attributes are retained on the header as
3538+ well (unlike before) -- with the exception of the `id` attribute, which of
3539+ course cannot be duplicated.
3541+ * Fix `--toc-depth` regression in 2.8 (#5967).
3543+ * Use doctemplates 0.8. Rename template 'filters' as 'pipes'
3544+ to avoid confusion with the other notion of filter used by pandoc.
3546+ * Fix README.md so that relative links from manual become absolute.
3547+ Previously they'd be broken links when viewed on GitHub or Hackage.
3548+ So we add the base URL for the pandoc manual.
3550+ * Document display math syntax in manual.
3552+## pandoc 2.8.1 (2019-12-05)
3554+ * Add `ascii_identifiers` as a supported extension for `markdown`.
3555+ This fixes a regression in 2.8.
3557+ * Fix regression with behavior of `--variable` (#5962).
3558+ Previously `-Vfoo=1 -Vfoo=2` would produce a list value for foo;
3559+ with 2.8 it produced just `2`. This commit restores the earlier
3562+ * Roll back part of of `--shift-heading-level-by` change (#5957).
3563+ With positive heading shifts, starting in 2.8 this option
3564+ caused metadata titles to be removed and changed to regular
3565+ headings. This behavior is incompatible with the old
3566+ behavior of `--base-header-level` and breaks old workflows,
3567+ so we have rolled back the change. Note that there is now an
3568+ asymmetry in positive and negative heading level shifts:
3569+ With positive shifts, the metadata title stays the same and
3570+ does not get changed to a heading in the body, while with
3571+ negative shifts, a heading can be converted into the
3574+ * Text.Pandoc.Shared: Fix `makeSections` so it doesn't turn
3575+ column Divs into sections.
3577+ * HTML writer: add task-list class to ul if all elements are
3578+ task list items. This will allow styling unordered task
3579+ lists in a way that omits the bullet.
3581+ * HTML-based templates: Add CSS to suppress bullet on unordered task lists.
3583+ * ConTeXt template: Fix `\startcslreferences` and use ConTeXt syntax
3584+ conventions (#5945, Denis Maier). The old version had a too large
3585+ a skip at the beginning of the reference list.
3587+ * LaTeX template: keep the `\author{}` command even if author is not
3588+ specified (#5961, Yihui Xie). Avoids a LaTeX warning.
3590+ * Generate Emoji module with TH.
3592+ + Add Text.Pandoc.Emoji.TH.
3593+ + Replace long literal list in Text.Pandoc.Emoji with one-liner
3594+ generating it from `emoji.json` using TH.
3595+ + Add Makefile target to download `emoji.json`.
3596+ + Remove `tools/emoji.hs`.
3598+ * Increase GC allocation space for compilation in cabal.project.
3600+ * Clean up manual on PDF generation backend options (#5940).
3602+ * Update release checklist to include code signing step and update
3603+ Windows release-candidate builds (#5950).
3606+## pandoc 2.8.0.1 (2019-11-26)
3608+ * List `pdf` in `--list-output-formats`.
3609+ * EPUB writer: Fix regression with `--css` (#5937). In 2.8 `--css`
3610+ would not have an effect on EPUB output.
3611+ * RST writer: Use grid tables for one-column tables, since
3612+ simple tables clash with heading syntax in this case (#5936).
3613+ * Add unexported module Text.Pandoc.Readers.Metadata (see #5914).
3614+ * Use doctemplates 0.7.2, which adds the `nowrap` filter to
3616+ * Update default man template using `nowrap` for .TH heading (#5929).
3617+ * HTML templates: Add support for `toc-title` variable (#5930,
3619+ * Remove `grffile` (LaTeX package) requirement in MANUAL.txt
3620+ (#5927, Ian Max Andolina).
3621+ * Use skylighting 0.8.3.
3623+## pandoc 2.8 (2019-11-22)
3625+ * Improvements in templates system (from doctemplates):
3627+ + Pandoc templates now support a number of new features that
3628+ have been added in doctemplates: notably, `elseif`, `it`,
3629+ partials, filters, and syntax to control nesting and reflowing of
3630+ text. These changes make pandoc more suitable out of the
3631+ box for generating plain-text documents from data in YAML
3632+ metadata. It can create enumerated lists and even tabular
3634+ + We now used templates parameterized on doclayout Doc types.
3635+ The main impact of this change is better reflowing of
3636+ content interpolated into templates. Previously,
3637+ interpolated variables were rendered independently and
3638+ interpolated as strings, which could lead to overly long
3639+ lines. Now the templates interpolated as Doc values which
3640+ may include breaking spaces, and reflowing occurs after
3641+ template interpolation rather than before.
3642+ + Remove code from the LaTeX, Docbook, and JATS writers that
3643+ looked in the template for strings to determine whether it
3644+ is a book or an article, or whether csquotes is used. This
3645+ was always kludgy and unreliable.
3646+ + Change template code to use new API for doctemplates.
3648+ * Add `--defaults`/`-d` option. This adds the ability to specify
3649+ a collection of default values for options in a YAML file. For
3650+ example, one might define a set of defaults for letters,
3651+ and then do `pandoc -d letter myletter.md -o myletter.pdf`.
3652+ See the documentation of this feature in MANUAL.txt.
5 * Raise error on unsupported extensions (#4338).
3654 * Raise error on unsupported extensions (#4338).
@@ -20,11 +3669,27 @@
21 * Allow `--metadata-file` to be used repeatedly to include
3670 * Allow `--metadata-file` to be used repeatedly to include
22 multiple metadata files (Owen McGrath, #5702). Values in
3671 multiple metadata files (Owen McGrath, #5702). Values in
23− files specified first will be used instead of those in later
3672+ files specified first will be overridden by those in later
26 * `--ascii` now uses numerical hex character references (#5718).
3675 * `--ascii` now uses numerical hex character references (#5718).
3677+ * Allow PDF output to stdout (#5751). PDF output now behaves like other
3678+ binary formats: it will not be output to the terminal, but can be
3679+ sent to stdout using either `-o -` or a pipe. The intermediate format
3680+ will be determined based on the setting of `--pdf-engine`.
3682+ * Make some writers sensitive to 'unlisted' class on headings (#1762).
3683+ If this is present on a heading with the 'unnumbered' class,
3684+ the heading won't appear in the TOC. This class has no
3685+ effect if 'unnumbered' is not also specified. This affects HTML-based
3686+ writers (including slide shows and EPUB), LateX (including beamer), RTF,
3687+ and PowerPoint. Other writers do not yet support `unlisted`.
3689+ * Fix `gfm_auto_identifiers` behavior with emojis (#5813). Note that
3690+ we also now use emoji names for emojis when `ascii_identifiers`
28 * When `--ipynb-output` is used with the default "best" format, strip
3693 * When `--ipynb-output` is used with the default "best" format, strip
29 ANSI escape codes for non-ipynb output (#5633). These cause problems
3694 ANSI escape codes for non-ipynb output (#5633). These cause problems
30 in many formats, including LaTeX.
3695 in many formats, including LaTeX.
@@ -34,6 +3699,17 @@
34 URL was used for the input file, instead of taking it from the
3699 URL was used for the input file, instead of taking it from the
35 data directory.
3700 data directory.
3702+ * Allow combining `-Vheader-includes` and `--include-in-header` (#5904).
3703+ Previously `header-includes` set as a variable would be
3704+ clobbered by material included using `--include-in-header`.
3706+ * Change merge behavior for metadata. Previously, if a document
3707+ contained two YAML metadata blocks that set the same field, the
3708+ conflict would be resolved in favor of the first. Now it is resolved
3709+ in favor of the second (due to a change in pandoc-types).
3710+ This makes the behavior more uniform with other things in pandoc
3711+ (such as reference links and `--metadata-file`).
37 * Don't add a newline to fragment output if there's already one.
3713 * Don't add a newline to fragment output if there's already one.
39 * Change exit codes and document in MANUAL.txt:
3715 * Change exit codes and document in MANUAL.txt:
@@ -42,6 +3718,80 @@
42 + `PandocOptionError` was 2, is now 6
3718 + `PandocOptionError` was 2, is now 6
43 + `PandocMakePDFError` was 65, is now 66
3719 + `PandocMakePDFError` was 65, is now 66
3721+ * Switch to new pandoc-types and use Text instead of String [API change].
3722+ (Christian Despres, #5884).
3726+ + Better handling of `<q>` with cite attribute (#5798, Ole Martin Ruud).
3727+ If a `<q>` tag has a `cite` attribute, we interpret it as a Quoted
3728+ element with an inner Span.
3729+ + Add support for HTML `<samp>` element (#5792, Amogh Rathore).
3730+ The `<samp>` element is parsed as Code with class `sample`.
3731+ + Add support for HTML `<var>` element (#5799, Amogh Rathore).
3732+ The `<var>` element is parsed as Code with class `variable`.
3733+ + Add support for `<mark>` elements (Florian B, #5797). Parse
3734+ `<mark>` elements from HTML as Spans with class `mark`.
3735+ + Add support for `<kbd>` elements, parsing them as Span with class
3736+ `kbd` (Daniele D'Orazio, #5796).
3737+ + Add support for `<dfn>`, parsing this as a Span with class `dfn`
3738+ (#5882, Florian Beeres).
3742+ + Headers: don't parse content over newline boundary (#5714).
3743+ + Handle inline code more eagerly within lists (Brian Leung, #5627).
3744+ + Removed some needless lookaheads.
3745+ + Don't parse footnote body unless extension enabled.
3746+ + Fix small super/subscript issue (#5878). Superscripts and subscripts
3747+ cannot contain spaces, but newlines were previously allowed
3748+ (unintentionally). This led to bad interactions in some cases
3749+ with footnotes. With this change newlines are also not allowed inside
3751+ + Use `take1WhileP` for `str`, table row. This yields a small but
3752+ measurable performance improvement.
3757+ + Fix parsing of optional arguments that contain braced text (#5740).
3758+ + Don't try to parse includes if `raw_tex` is set (#5673).
3759+ When the `raw_tex` extension is set, we just carry through
3760+ `\usepackage`, `\input`, etc. verbatim as raw LaTeX.
3761+ + Properly handle optional arguments for macros (#5682).
3762+ + Fix `\\` in `\parbox` inside a table cell (#5711).
3763+ + Improve `withRaw` so it can handle cases where the token string is
3764+ modified by a parser (e.g. accent when it only takes part of a Word
3765+ token) (#5686). This fixes a bug that caused the ends of
3766+ certain documents to be dropped.
3767+ + Handle `\passthrough` macro used by latex writer (#5659).
3768+ + Support tex `\tt` command (#5654).
3769+ + Search for image with list of extensions like latex does, if an
3770+ extension is not provided (#4933).
3771+ + Handle `\looseness` command values better (#4439).
3772+ + Add `mbox` and `hbox` handling (Vasily Alferov, #5586).
3773+ When `+raw_tex` is enabled, these are passed through literally.
3774+ Otherwise, they are handled in a way that emulates LaTeX's behavior.
3775+ + Properly handle `\providecommand` and `\provideenvironment` (#5635).
3776+ They are now ignored if the corresponding command or environment
3777+ is already defined.
3778+ + Support epigraph command in LaTeX Reader (oquechy, #3523).
3779+ + Ensure that expanded macros in raw LaTeX end with a space
3780+ if the original did (#4442).
3781+ + Treat `ly` environment from lilypond as verbatim (Urs Liska, #5671).
3782+ + Add `tikzcd` to list of special environments (Eigil Rischel).
3783+ This allows it to be processed by filters, in the same way that
3784+ one can do for `tikzpicture`.
3788+ + Better support for `while`.
3789+ + More improvements in parsing conditionals.
3790+ + Fix problem parsing comments before macro.
3791+ + Improve handling of groups.
3792+ + Better parsing of groups (#5410). We now allow groups
3793+ where the closing `\\}` isn't at the beginning of a line.
45 * RST reader:
3795 * RST reader:
47 + Keep `name` property in `imgAttr` (Brian Leung, #5619).
3797 + Keep `name` property in `imgAttr` (Brian Leung, #5619).
@@ -50,6 +3800,8 @@
50 + Use title, not admonition-title, for admonition title.
3800 + Use title, not admonition-title, for admonition title.
51 This puts RST reader into alignment with docbook reader.
3801 This puts RST reader into alignment with docbook reader.
52 + Don't strip final underscore from absolute URI (#5763).
3802 + Don't strip final underscore from absolute URI (#5763).
3803+ + Avoid spurious warning when resolving links to internal anchors
3804+ ending with `_` (#5763).
54 * Org reader:
3806 * Org reader:
@@ -62,20 +3814,35 @@
62 Code blocks in Org lists must have their `#+BEGIN_` aligned in a
3814 Code blocks in Org lists must have their `#+BEGIN_` aligned in a
63 reasonable way, but their other components can be
3815 reasonable way, but their other components can be
64 positioned otherwise.
3816 positioned otherwise.
3817+ + Fix parsing of empty comment lines (#5856, Albert Krewinkel).
3818+ Comment lines in Org-mode can be completely empty.
66− * Muse reader:
3820+ * Muse reader (Alexander Krotov):
68− + Add RTL support (Alexander Krotov, #5551).
3822+ + Add RTL support (#5551).
3823+ + Do not allow closing asterisks to be followed by `*`.
3824+ + Do not split series of asterisks into symbols and emphasis (#5821).
3825+ + Do not terminate emphasis on `*` not followed by space.
70− * JIRA writer:
3827+ * DokuWiki reader:
72− + Remove escapeStringForJira for code blocks (Jan-Otto Kröpke).
3829+ + Parse markup inside monospace ('') (#5916, Alexander Krotov).
74 * Docx reader:
3831 * Docx reader:
76 + Move style-parsing-specific code to a new unexported module,
3833 + Move style-parsing-specific code to a new unexported module,
77− Text.Pandoc.Readers.Docx.Parse.Styles.
3834+ Text.Pandoc.Readers.Docx.Parse.Styles (Nikolay Yakimov).
78− + Move StyleMap to docx writer.
3835+ + Move StyleMap to docx writer (Nikolay Yakimov).
3836+ + Only use LTR when it is overriding BiDi setting (#5723, Jesse
3837+ Rosenthal). The left-to-right direction setting in docx is used
3838+ in the spec only for overriding an explicit right-to-left setting.
3839+ We only process it when it happens in a paragraph set with BiDi.
3840+ This is especially important for docs exported from Google Docs,
3841+ which explicitly (and unnecessarily) set `rtl=0` for every paragraph.
3842+ + Fix list number resumption for sublists (#4324).
3843+ The first list item of a sublist should not resume numbering
3844+ from the number of the last sublist item of the same level,
3845+ if that sublist was a sublist of a different list item.
80 * Docbook reader:
3847 * Docbook reader:
@@ -83,11 +3850,35 @@
83 Instead of parsing admonitions as blockquotes, we now parse
3850 Instead of parsing admonitions as blockquotes, we now parse
84 them as Divs with an appropriate class. We also handle titles
3851 them as Divs with an appropriate class. We also handle titles
85 for admonitions as a nested Div with the "title" class.
3852 for admonitions as a nested Div with the "title" class.
3853+ + Fix nesting of chapters and sections (#5864, Florian Klink,
3854+ Félix Baylac-Jacqué).
3855+ + Fix bug with entities in mathphrase element (#5885).
87 * MediaWiki reader:
3857 * MediaWiki reader:
89 + Skip optional `{{table}}` template (#5757).
3859 + Skip optional `{{table}}` template (#5757).
3863+ + Fix dollar-math parsing to ensure that space is left between a
3864+ control sequence and a following letter (#5836).
3865+ + In `untokenize`, ensure space between control sequence
3866+ and following letter (#5836).
3867+ + Don't omit macro definitions defined in the preamble.
3868+ These were formerly omitted (though they still affected macro
3869+ resolution if `latex_macros` was set). Now they are included in
3871+ + Parse macro definitions as raw LaTeX when `latex_macros` is
3872+ disabled. (When `latex_macros` is enabled, we omit them, since
3873+ pandoc is applying the macros itself.)
3874+ + Fix a hang/memory leak in certain circumstances (#5845).
3875+ + Text.Pandoc.Readers.LaTeX.Parsing: add `[Tok]` parameter to
3876+ `rawLaTeXParser`. This allows us to repeat retokenizing
3877+ unnecessarily in e.g. `rawLaTeXBlock`.
3878+ + Add KOMA-Script metadata commands (#5910, Andrew Dunning).
3879+ Add all titling commands to existing definition for `\dedication`.
3880+ + Parse `\micro` siunitx unit command (#5921, Jose Luis Duran).
91 * Markdown writer:
3882 * Markdown writer:
93 + Ensure proper nesting when we have long ordered list markers (#5705).
3884 + Ensure proper nesting when we have long ordered list markers (#5705).
@@ -105,6 +3896,9 @@
105 and LaTeX, and even when `raw_html` and `raw_tex` are
3896 and LaTeX, and even when `raw_html` and `raw_tex` are
106 enabled, as they are by default. To get the old behavior,
3897 enabled, as they are by default. To get the old behavior,
107 disable `raw_attribute` in the writer.
3898 disable `raw_attribute` in the writer.
3899+ + Prefer `pipe_tables` to raw HTML even when we must
3900+ lose width information (#2608, #4497).
3901+ + Improve escaping in list items (#5918).
109 * AsciiDoc writer:
3903 * AsciiDoc writer:
@@ -135,6 +3929,9 @@
135 `csquotes` in your variables or metadata. To specify a book
3929 `csquotes` in your variables or metadata. To specify a book
136 style, use the `documentclass` variable or
3930 style, use the `documentclass` variable or
137 `--top-level-division`.
3931 `--top-level-division`.
3932+ + Fix horizontal rule (#5801). We change to use 0.5pt rather than
3933+ `\linethickness`, which apparently only ever worked "by accident"
3934+ and no longer works with recent updates to texlive.
139 * ConTeXt writer:
3936 * ConTeXt writer:
@@ -149,6 +3946,10 @@
149 using `pdfaiccprofile` and `pdfaintent`.
3946 using `pdfaiccprofile` and `pdfaintent`.
150 + Unit tests: adjust code property to avoid an irrelevant
3947 + Unit tests: adjust code property to avoid an irrelevant
151 failure involving inline code with two consecutive newlines.
3948 failure involving inline code with two consecutive newlines.
3949+ + Set `csl-hanging-indent` variable if needed.
3950+ + Use special environment for CSL references.
3951+ + Use braces, not start/stop, for inline language tags.
3952+ This prevents unwanted gobbling of spaces.
153 * HTML writer:
3954 * HTML writer:
@@ -166,6 +3967,18 @@
166 Pesca, #5655). We add `\textstyle` to the beginning of the
3967 Pesca, #5655). We add `\textstyle` to the beginning of the
167 formula to ensure it will be rendered in inline style.
3968 formula to ensure it will be rendered in inline style.
168 + Pass through `aria-` attributes to HTML5 (#5642).
3969 + Pass through `aria-` attributes to HTML5 (#5642).
3970+ + Render a Quoted element with an inner Span with
3971+ `cite` attribute using a `<q>` tag (#5798, Ole Martin Ruud).
3972+ + Render a Span with class `mark` using the `<mark>` element
3973+ (Florian Beeres, #5797).
3974+ + Render Span with class `dfn` using `<dfn>` element
3975+ (Florian Beeres, #5882).
3976+ + Render Span with class `kbd` using `<kbd>` element (Daniele
3978+ + Render Code with class `variable` using `<var>` element
3979+ (Amogh Rathore, #5799).
3980+ + Render Code with class `sample` using `<samp>` element
3981+ (Amogh Rathore, #5799).
170 * EPUB writer:
3983 * EPUB writer:
@@ -182,10 +3995,22 @@
182 + Add support for EPUB2 covers (blmage, #3992).
3995 + Add support for EPUB2 covers (blmage, #3992).
183 + Do not override existing "fileN" medias when writing to EPUB format
3996 + Do not override existing "fileN" medias when writing to EPUB format
184 (blmage, #4206).
3997 (blmage, #4206).
3998+ + Ensure that `lang` variable is set on all chapters (so that it
3999+ will add an `xml:lang` attribute on the `html` element).
186 * RST writer:
4001 * RST writer:
188 + Removed remnants of `admonition-title`.
4003 + Removed remnants of `admonition-title`.
4004+ + Fix handling of `:align:` on figures and images (#4420).
4005+ When the image has the `align-right` (etc.) class, we now use
4006+ an `:align:` attribute.
4007+ + Improve spacing for tables with no width information (#5899).
4008+ If a simple table would be too wide, we use a grid table.
4009+ + Fix backslash escaping after strings (Albert Krewinkel, #5906).
4010+ The check whether a complex inline element following a string must
4011+ be escaped, now depends on the last character of the string instead
4013+ + Ensure there's a blank line before tables (#5898).
190 * Dokuwiki writer:
4015 * Dokuwiki writer:
@@ -212,6 +4037,7 @@
212 * Ms writer:
4037 * Ms writer:
214 + Use `.LP` instead of `.PP` for line block (#5588).
4039 + Use `.LP` instead of `.PP` for line block (#5588).
4040+ + Use boldface for definition terms in DefinitionLists.
216 * JATS writer:
4042 * JATS writer:
@@ -221,12 +4047,25 @@
221 name-part elements. (#5629)
4047 name-part elements. (#5629)
222 + Add `abstract` to template (Mauro Bieg).
4048 + Add `abstract` to template (Mauro Bieg).
4052+ + Don't strip hash from internal links (#5922).
4056+ + Remove escapeStringForJira for code blocks (Jan-Otto Kröpke).
4057+ + Remove extraneous newline after single-line block quotes
4058+ (#5858, Albert Krewinkel).
4059+ + Improve escaping of special characters, using backslash escapes
4060+ instead of HTML entities (#5858, Albert Krewinkel).
224 * OpenDocument writer:
4062 * OpenDocument writer:
226 + Avoid duplicate attributes (#4634). We use the innermost
4064 + Avoid duplicate attributes (#4634). We use the innermost
227 attribute in nested cases.
4065 attribute in nested cases.
228 + If `native_numbering` extension is set, use native OpenDocument
4066 + If `native_numbering` extension is set, use native OpenDocument
229 enumeration for figures and tables (Nils Carlson).
4067 enumeration for figures and tables (Nils Carlson).
4068+ + Place caption before table (#5681, Dmitry Pogodin).
231 * ODT writer:
4070 * ODT writer:
@@ -275,12 +4114,16 @@
275 a margin and font size according to their level.
4114 a margin and font size according to their level.
276 Furthermore this commit allows changing the font with which code is
4115 Furthermore this commit allows changing the font with which code is
277 formatted via the `monofont` option.
4116 formatted via the `monofont` option.
278− + Start numbering at appopriate numbers (Jesse Rosenthal, #5709).
4117+ + Start numbering at appropriate numbers (Jesse Rosenthal, #5709).
279 Starting numbers for ordered lists were previously ignored. Now
4118 Starting numbers for ordered lists were previously ignored. Now
280 we specify the number if it is something other than 1.
4119 we specify the number if it is something other than 1.
282 * Text.Pandoc.Parsing:
4121 * Text.Pandoc.Parsing:
4123+ + Add `manyChar`, `many1Char`, `manyTillChar`, `many1TillChar`,
4124+ `many1Till`, `manyUntil`, `mantyUntilChar`: these are like their
4125+ unsuffixed counterparts but pack some or all of their output
4126+ (Christian Despres, #5884).
284 + Add `stateAllowLineBreaks` to `ParserState` [API change].
4127 + Add `stateAllowLineBreaks` to `ParserState` [API change].
285 + Fix inline parsing in grid table cells (#5708).
4128 + Fix inline parsing in grid table cells (#5708).
286 + Change type of `setLastStrPos` so it takes a `Maybe SourcePos`
4129 + Change type of `setLastStrPos` so it takes a `Maybe SourcePos`
@@ -289,12 +4132,16 @@
289 `gridTableWith'` polymorphic in the parser state,
4132 `gridTableWith'` polymorphic in the parser state,
290 constraining it with `HasLastStrPosition` [API change].
4133 constraining it with `HasLastStrPosition` [API change].
291 + `parseFromString'`: reset `stateLastStrPos` to `Nothing` before parse.
4134 + `parseFromString'`: reset `stateLastStrPos` to `Nothing` before parse.
4135+ + Rename takeWhileP -> take1WhileP and clean it up.
4136+ (It doesn't match the empty sequence.)
293 * Text.Pandoc.PDF:
4138 * Text.Pandoc.PDF:
295 + For PDFs via HTML, ensure temp file is deleted even if the pdf
4140 + For PDFs via HTML, ensure temp file is deleted even if the pdf
296 program is not found (#5720).
4141 program is not found (#5720).
297 + Better detection of a Cygwin environment (#5451).
4142 + Better detection of a Cygwin environment (#5451).
4143+ + Don't assume tex log file is UTF8-encoded (#5872).
4144+ Fall back to latin1 if it can't be read as UTF-8.
299 * Text.Pandoc.Extensions:
4146 * Text.Pandoc.Extensions:
@@ -395,52 +4242,6 @@
395 + Pass value of `--dpi` to `rsvg-convert` when converting SVG to PDF
4242 + Pass value of `--dpi` to `rsvg-convert` when converting SVG to PDF
396 in the process of creating a PDF (#5721).
4243 in the process of creating a PDF (#5721).
400− + Headers: don't parse content over newline boundary (#5714).
401− + Handle inline code more eagerly within lists (Brian Leung, #5627).
402− + Removed some needless lookaheads.
406− + Fix parsing of optional arguments that contain braced text (#5740).
407− + Don't try to parse includes if `raw_tex` is set (#5673).
408− When the `raw_tex` extension is set, we just carry through
409− `\usepackage`, `\input`, etc. verbatim as raw LaTeX.
410− + Properly handle optional arguments for macros (#5682).
411− + Fix `\\` in `\parbox` inside a table cell (#5711).
412− + Improve `withRaw` so it can handle cases where the token string is
413− modified by a parser (e.g. accent when it only takes part of a Word
414− token) (#5686). This fixes a bug that caused the ends of
415− certain documents to be dropped.
416− + Handle `\passthrough` macro used by latex writer (#5659).
417− + Support tex `\tt` command (#5654).
418− + Search for image with list of extensions like latex does, if an
419− extension is not provided (#4933).
420− + Handle `\looseness` command values better (#4439).
421− + Add `mbox` and `hbox` handling (Vasily Alferov, #5586).
422− When `+raw_tex` is enabled, these are passed through literally.
423− Otherwise, they are handled in a way that emulates LaTeX's behavior.
424− + Properly handle `\providecommand` and `\provideenvironment` (#5635).
425− They are now ignored if the corresponding command or environment
427− + Support epigraph command in LaTeX Reader (oquechy, #3523).
428− + Ensure that expanded macros in raw LaTeX end with a space
429− if the original did (#4442).
430− + Treat `ly` environment from lilypond as verbatim (Urs Liska, #5671).
431− + Add `tikzcd` to list of special environments (Eigil Rischel).
432− This allows it to be processed by filters, in the same way that
433− one can do for `tikzpicture`.
437− + Better support for `while`.
438− + More improvements in parsing conditionals.
439− + Fix problem parsing comments before macro.
440− + Improve handling of groups.
441− + Better parsing of groups (#5410). We now allow groups
442− where the closing `\\}` isn't at the beginning of a line.
444 * Text.Pandoc.Shared:
4245 * Text.Pandoc.Shared:
446 + Replace `Element` and `makeHierarchical` with `makeSections`.
4247 + Replace `Element` and `makeHierarchical` with `makeSections`.
@@ -457,6 +4258,14 @@
457 + Make `filterIpynbOutput` strip ANSI escapes from code in output
4258 + Make `filterIpynbOutput` strip ANSI escapes from code in output
458 for non-ipynb formats, when the default "best" option is used with
4259 for non-ipynb formats, when the default "best" option is used with
459 `--ipynb-output` (#5633).
4260 `--ipynb-output` (#5633).
4261+ + Fix `camelCaseToHyphenated` so it handles `ABCDef` better.
4262+ + Improve `isTight` (#5857). If a list has an empty item,
4263+ this should not count against its being a tight list.
4264+ + Export `htmlSpanLikeElements` [API change] (Daniele D'Orazio, #5796).
4265+ This is a mapping of HTML span-like elements that are internally
4266+ represented as a Span with a single class.
4267+ + Change the implementation of `htmlSpanLikeElements` to retain
4268+ classes and attributes (#5882, Florian Beeres).
461 * Text.Pandoc.Slides: recognize content in Divs when determining
4270 * Text.Pandoc.Slides: recognize content in Divs when determining
462 slide level.
4271 slide level.
@@ -476,17 +4285,45 @@
476 + Modify `addVariablesToContext`, `defField`, `setField`, `getField`,
4285 + Modify `addVariablesToContext`, `defField`, `setField`, `getField`,
477 `resetField` to work with Context rather than JSON values. [API change]
4286 `resetField` to work with Context rather than JSON values. [API change]
478 + Export new function `endsWithPlain` [API change].
4287 + Export new function `endsWithPlain` [API change].
4288+ + Change `gridTables` so it does better at keeping the widths of
4289+ columns (#4320) and does better at figuring out column widths
4290+ when no widths are given (#5899).
480 * Text.Pandoc.Options
4292 * Text.Pandoc.Options
482 + Change type of `writerTemplate` to `Maybe Template` instead
4294 + Change type of `writerTemplate` to `Maybe Template` instead
483 of `Maybe String`.
4295 of `Maybe String`.
4296+ + Change To/FromJSON instances for `HTMLMathMethod`, `CiteMethod`,
4297+ `ObfuscationMethod`, `TrackChanges`, `WrapOption`,
4298+ `TopLevelDivision`, `ReferenceLocation`, `HTMLSlideVariant` (#5790).
4299+ In each case we use lowercase (or hyphenated lowercase) for
4300+ constructors to line up more closely with command-line option
4301+ values. This is a breaking change for those who manually decode or
4302+ encode JSON for these data types (e.g. for `ReaderOptions` or
4305+ * Text.Pandoc.Filters:
4307+ + Add `FromYAML` instance for `Filter`.
4308+ + `applyFilters`: Add and apply filters in order (not reversed)
4309+ This changes `applyFilters` from Text.Pandoc.Filter so
4310+ that it does a left fold rather than a right fold, applying
4311+ the filters in the order listed.
485 * Text.Pandoc.XML:
4313 * Text.Pandoc.XML:
487 + Change `toEntities` to emit numerical hex character references
4315 + Change `toEntities` to emit numerical hex character references
488 (#5718).
4316 (#5718).
4318+ * Text.Pandoc.Highlighting:
4320+ + Add additional listings languages (Wandmalfarbe).
4322+ * Text.Pandoc.MediaBag:
4324+ + Some of the types using Strings were switched to use FilePath instead
4325+ (Christian Despres, #5884).
490 * Text.Pandoc.Templates:
4327 * Text.Pandoc.Templates:
492 + Don't import/export `varListToJSON` [API change]. It is removed in
4329 + Don't import/export `varListToJSON` [API change]. It is removed in
@@ -495,25 +4332,38 @@
495 now Text rather than being polymorphic. This makes room for
4332 now Text rather than being polymorphic. This makes room for
496 upcoming removal of the `TemplateTarget` class from doctemplates.
4333 upcoming removal of the `TemplateTarget` class from doctemplates.
498− * Improvements in templates system (from doctemplates-0.6.1):
4335+ * Text.Pandoc.App.Opt [API changes]:
500− + Pandoc templates now support a number of new features that
4337+ + More convenient To/FromJSON instances. Make the field names
501− have been added in doctemplates: notably, `elseif`, `it`,
4338+ like `strip-empty-paragraphs` rather than `optStripEmptyParagraphs`.
502− partials and keywords to control nesting and reflowing of
4339+ Anyone who is using JSON serialization of Opt will need to adjust
503− text.
4340+ things accordingly.
504− + We now used templates parameterized on doclayout Doc types.
4341+ + Change `optHighlightStyle` to a `Maybe String` instead of
505− The main impact of this change is better reflowing of
4342+ `Maybe Style`. Do the parsing/loading of themes later, after
506− content interpolated into templates. Previously,
4343+ option parsing.
507− interpolated variables were rendered independently and
4344+ + Remove `optBaseHeaderLevel` from `Opt`. We now just use
508− intepolated as strings, which could lead to overly long
4345+ `optShiftHeadingLevelBy`, to avoid redundancy.
509− lines. Now the templates interpolated as Doc values which
4346+ + Change `optShiftHeadingLevel` to `optShiftHeadingLevelBy` to
510− may include breaking spaces, and reflowing occurs after
4347+ match the option.
511− template interpolation rather than before.
4348+ + Custom FromJSON instance for `LineEnding`, so either `CRLF`
512− + Remove code from the LaTeX, Docbook, and JATS writers that
4349+ or `crlf` will work.
513− looked in the template for strings to determine whether it
4350+ + Change `optVariables` from `[(String, String)]` to `Context Text`.
514− is a book or an article, or whether csquotes is used. This
4351+ + Change `optMetadata` to `Meta`, to allow structured values.
515− was always kludgy and unreliable.
4352+ The current behavior of the `--metadata` option stays the same.
516− + Change template code to use new API for doctemplates.
4353+ + Rename `optReader`, `optWriter` as `optFrom`, `optTo`.
4354+ + Add `FromYAML` instances to `Opt` and to all subsidiary types.
4355+ + Rename `optMetadataFile` to `optMetadataFiles`.
4356+ + Rename `optPDFEngineArgs` to `optPDFEngineOpts`.
4357+ + Rename `optWrapText` to `optWrap`.
4358+ + Add `IpynbOutput` enumerated type: use this instead of
4359+ a string for `optIpynbOutput`.
4360+ + Change optInputFiles to a `Maybe [FilePath]` (#5888) [API change].
4361+ `Nothing` means: nothing specified.
4362+ `Just []` means: an empty list specified (e.g. in defaults).
4363+ + List fields in Opt so they aren't reversed (#5881) [API change].
4364+ Previously `optIncludeInHeader`, etc. were in reverse order.
4365+ + The `sourcefile` variable is now always a list. It used to be
4366+ sometimes a string, sometimes a list (when there was more than one).
518 * Template changes:
4368 * Template changes:
@@ -522,13 +4372,43 @@
522 variable (set automatically if there is a `hanging-ident`
4372 variable (set automatically if there is a `hanging-ident`
523 class on the references Div) controls whether contents of this
4373 class on the references Div) controls whether contents of this
524 environment receive a hanging indent.
4374 environment receive a hanging indent.
4375+ + default.latex: Add `space` as default option for xeCJK, so that
4376+ spaces between words are preserved (#5855, jeongminkim-islab).
4377+ This is necessary for Korean.
4378+ + default.latex: Remove include of `grffile` (#5848).
4379+ This package used to be needed for proper handling of image filenames
4380+ containing periods (in addition to the period before the extension).
4381+ It no longer works with the latest LaTeX kernel and graphicx,
4382+ so we have removed it. Future versions of graphicx will handle
4383+ these filenames without the need for `grffile`.
4384+ + default.context: add a saner default for page numbers.
4385+ Previously they appeared centered at the top of the page;
4386+ now we put them centered at the bottom, unless the `pagenumbering`
4388+ + default.context: define a start-stop-pair `cslreferences` to
4389+ allow for hanging indents in the bibliography (#5875, Denis Maier).
4390+ + default.ms: update defaults. Use Palatino font, use slightly
4391+ wider interparagraph space, don't indent paragraphs,
4392+ and put page numbers on the bottom. This brings ms output
4393+ closer to default LaTeX output.
525− + revealjs template: add navigationMode (Mauro Bieg, #5657).
4394+ + default.revealjs: add navigationMode (Mauro Bieg, #5657).
526− + muse template: handle multiple authors better.
4395+ + default.muse: handle multiple authors better.
527 + docbook4, docbook5 templates: add indentation to body.
4396 + docbook4, docbook5 templates: add indentation to body.
528 + HTML-based templates: use `styles.html` partial to avoid
4397 + HTML-based templates: use `styles.html` partial to avoid
529 code duplication.
4398 code duplication.
530 + HTML-based templates: change indentation of styles in template.
4399 + HTML-based templates: change indentation of styles in template.
4401+ * reference.docx (#5820):
4403+ + Change Block Text (block quote) style so that the same font
4404+ is used as in the body text, and the block text is indented
4406+ + All headings now have a uniform color.
4407+ + Level-1 headings no longer set `w:themeShade="B5"`.
4408+ + Level-2 headings are now 14 point rather than 16 point.
4409+ + Level-3 headings are now 12 point rather than 14 point.
4410+ + Level-4 headings are italic rather than bold.
532 * epub.css: Add CSS for hanging-indent div to support pandoc-citeproc's
4412 * epub.css: Add CSS for hanging-indent div to support pandoc-citeproc's
533 new hanging indents.
4413 new hanging indents.
@@ -541,9 +4421,11 @@
541 + Add cabal.project to extra-source-files (Albert Krewinkel).
4421 + Add cabal.project to extra-source-files (Albert Krewinkel).
542 + Add dependency on skylighting-core (#5729). Even though it shouldn't
4422 + Add dependency on skylighting-core (#5729). Even though it shouldn't
543 be needed, some tools require it.
4423 be needed, some tools require it.
544− + Require latest pandoc-types, texmath, skylighting.
4424+ + Require latest pandoc-types, texmath, skylighting, haddock-library.
545 + Ensure TemplateHaskell is added to other-extensions when it is
4425 + Ensure TemplateHaskell is added to other-extensions when it is
546 used (Vanessa McHale, #5728).
4426 used (Vanessa McHale, #5728).
4427+ + Remove `derive_json_via_th` flag; always use TH. This cuts down
4428+ on code duplication and reduces the chance for errors. See #4083.
548 * Makefile:
4430 * Makefile:
@@ -553,6 +4435,11 @@
554 * Benchmarks: fix failure on ipynb.
4436 * Benchmarks: fix failure on ipynb.
4438+ * Use MathJax 3 (zorkow).
4440+ * KaTeX math: respect `classoption=fleqn` variable,
4441+ bump KaTeX version to 0.11.1 (#5815, Mauro Bieg).
556 * Fix redundant constraint compiler warnings (Pete Ryland, #5625).
4443 * Fix redundant constraint compiler warnings (Pete Ryland, #5625).
558 * Use throwError instead of fail when appropriate.
4445 * Use throwError instead of fail when appropriate.
@@ -567,10 +4454,19 @@
567 in `test/fb2/reader/` don't have newlines converted. This should fix
4454 in `test/fb2/reader/` don't have newlines converted. This should fix
568 a test failure on GitHub CI with Windows.
4455 a test failure on GitHub CI with Windows.
4457+ * Reorder options in `--help`.
570 * Revise code for HsYAML-0.2.0.0 (@vijayphoenix, #5704).
4459 * Revise code for HsYAML-0.2.0.0 (@vijayphoenix, #5704).
572 * Remove blank line in code example in Haddocks (Leif Metcalf, #5679).
4461 * Remove blank line in code example in Haddocks (Leif Metcalf, #5679).
4463+ * Fix trypandoc with `getReader`/`getWriter` changes.
4465+ * Allow building pandoc with GHC 8.8.
4467+ * linux tarball: add architecture `-amd64` to filename.
4468+ Now it will be: `pandoc-VERSION-linux-ARCH.tar.gz`.
574 * MANUAL.txt:
4470 * MANUAL.txt:
576 + Add section for exit codes.
4472 + Add section for exit codes.
@@ -579,11 +4475,39 @@
579 + reveal.js flags (Mauro Bieg, #5653).
4475 + reveal.js flags (Mauro Bieg, #5653).
580 + Document addition of `data-` prefix to unknown attributes in HTML5.
4476 + Document addition of `data-` prefix to unknown attributes in HTML5.
581 + Link to YAML spec (Mauro Bieg, #5687).
4477 + Link to YAML spec (Mauro Bieg, #5687).
4478+ + Fix capitalization of "Linux" (#5859, Marcus Stollsteimer).
4479+ + Use a table for exit codes.
4480+ + Put all template variable docs into one section.
4481+ + Use ATX headers consistently.
4482+ + Add fuller documentation of templates (#5779), including
4483+ new template syntax, partials, etc.
4484+ + Add documentation for the variable `hyperrefoptions` (Wandmalfarbe).
4485+ + Clarify when macro definitions are passed as raw latex.
4486+ In Markdown input, they are always passed through. In LaTeX, only if
4487+ `latex_macros` is disabled.
4488+ + Clarify that `--dpi` provides a default and doesn't override
4489+ dpi values specified in the images themselves (#5721).
4490+ + Document how to use custom writers with `--standalone` (#5866).
4491+ + Clarify `--preserve-tabs` default.
4495+ + Fix instructions for libicu.
4496+ + Add Void Linux instructions (Volodymyr Kozachnyskyi).
4500+ + Add information on tests (Agustín Martín Barbero, #5652).
4501+ + Add information about command test naming to CONTRIBUTING (Florian B).
583− * INSTALL.md: fix instructions for libicu.
4503+ * Fix typos in changelog and comments (#5896, Brian Wignall).
585− * CONTRIBUTING.md: add information on tests (Agustín Martín
4505+ * doc/lua-filters.md:
586− Barbero, #5652).
4506+
4507+ + Fix mistakes in mediabag module docs (#5851, Albert Krewinkel).
4508+ + Improve metadata replacement example in lua-filters doc (#5851).
4509+ + Mention which Lua version is shipped with pandoc (Albert Krewinkel,
588## pandoc 2.7.3 (2019-06-11)
4512## pandoc 2.7.3 (2019-06-11)
@@ -693,14 +4617,14 @@
693 screen-readers would read it twice, see #4737.
4617 screen-readers would read it twice, see #4737.
694 + Don't add variation selector if it's already there.
4618 + Don't add variation selector if it's already there.
695 This fixes round-trip failures.
4619 This fixes round-trip failures.
696− + Prevent gratuitious emojification on iOS (#5469).
4620+ + Prevent gratuitous emojification on iOS (#5469).
697 iOS chooses to render a number of Unicode entities, including '↩', as
4621 iOS chooses to render a number of Unicode entities, including '↩', as
698− big colorful emoji. This can be defeated by appending Unicode VARIATION
4622+ big colorful emoji. This can be defeated by appending Unicode
699− SELECTOR-15'/'VARIATION SELECTOR-16'. So we now append this character
4623+ VARIATION SELECTOR-15'/'VARIATION SELECTOR-16'. So we now append this
700− when escaping strings, for both '↩' and '↔'. If other characters prove
4624+ character when escaping strings, for both '↩' and '↔'. If other
701− problematic, they can simply be added to `needsVariationSelector`.
4625+ characters prove problematic, they can simply be added to
702− + Add `class="heading"` to level 7+ Headers rendered as `<p>` elements
4626+ `needsVariationSelector`. + Add `class="heading"` to level 7+ Headers
703− (#5457).
4627+ rendered as `<p>` elements (#5457).
705 * RST writer: treat Span with no attributes as transparent (#5446).
4629 * RST writer: treat Span with no attributes as transparent (#5446).
706 Previously an Emph inside a Span was being treated as
4630 Previously an Emph inside a Span was being treated as
@@ -1007,7 +4931,7 @@
1007 + Properly escape attributes in Markdown writer (#5369).
4931 + Properly escape attributes in Markdown writer (#5369).
1008 + Be sure implicit figures work in list contexts (#5368).
4932 + Be sure implicit figures work in list contexts (#5368).
1009 Previously they would sometimes not work: e.g., when they
4933 Previously they would sometimes not work: e.g., when they
1010− occured in final paragraphs in lists that were originally
4934+ occurred in final paragraphs in lists that were originally
1011 parsed as Plain and converted later using PlainToPara.
4935 parsed as Plain and converted later using PlainToPara.
1013 * Docx writer: Use `w:br` without attributes for line breaks (#5377).
4937 * Docx writer: Use `w:br` without attributes for line breaks (#5377).
@@ -1167,7 +5091,7 @@
1167 the title slide (for beamer). This change makes possible
5091 the title slide (for beamer). This change makes possible
1168 2D reveal.js slideshows with content in the top slide on
5092 2D reveal.js slideshows with content in the top slide on
1169 each stack (#4317, #5237).
5093 each stack (#4317, #5237).
1171 * Add command line option `--ipynb-output=all|none|best` (#5339).
5095 * Add command line option `--ipynb-output=all|none|best` (#5339).
1172 Output cells in ipynb notebooks often contain several different
5096 Output cells in ipynb notebooks often contain several different
1173 versions of an output, with different MIME types, e.g. an HTML
5097 versions of an output, with different MIME types, e.g. an HTML
@@ -1314,7 +5238,7 @@
1314 `utils`, and `text`, respectively.
5238 `utils`, and `text`, respectively.
1316 * Text.Pandoc.Lua (Albert Krewinkel):
5240 * Text.Pandoc.Lua (Albert Krewinkel):
1318 + Split `StackInstances` into smaller Marshaling modules.
5242 + Split `StackInstances` into smaller Marshaling modules.
1319 + Get `CommonState` from Lua global. This allows more control over
5243 + Get `CommonState` from Lua global. This allows more control over
1320 the common state from within Lua scripts.
5244 the common state from within Lua scripts.
@@ -1509,7 +5433,7 @@
1509 * Man/Ms writers: Don't escape `-` as `\-`. The `\-` gets rendered
5433 * Man/Ms writers: Don't escape `-` as `\-`. The `\-` gets rendered
1510 in HTML and PDF as a unicode minus sign.
5434 in HTML and PDF as a unicode minus sign.
1512− * Ms writer: Ensure we have a newline after .EN in disply math (#5251).
5436+ * Ms writer: Ensure we have a newline after .EN in display math (#5251).
1514 * RST writer: Don't wrap simple table header lines (#5128).
5438 * RST writer: Don't wrap simple table header lines (#5128).
@@ -1641,7 +5565,7 @@
1641 + Set default listings language for lua, assembler (#5227, John
5565 + Set default listings language for lua, assembler (#5227, John
1642 MacFarlane). Otherwise we get an error when trying to compile code
5566 MacFarlane). Otherwise we get an error when trying to compile code
1643 with lua or assembler code. To change the default dialect
5567 with lua or assembler code. To change the default dialect
1644− (currenty 5.3 for lua and x86masm for assembler), you can use
5568+ (currently 5.3 for lua and x86masm for assembler), you can use
1645 `--include-in-header` to inject something like
5569 `--include-in-header` to inject something like
1646 `\lstset{defaultdialect=[5.2]Lua}`.
5570 `\lstset{defaultdialect=[5.2]Lua}`.
@@ -1651,7 +5575,7 @@
1652 * Text.Pandoc.Writers: Changed types of `writeJSON`; it now runs
5576 * Text.Pandoc.Writers: Changed types of `writeJSON`; it now runs
1653 in an instance of PandocMonad, like the other readers and
5577 in an instance of PandocMonad, like the other readers and
1654− writers. [API change]
5578+ writers. [API change]
1656 * Text.Pandoc.Error: Added `PandocUTF8DecodingError` constructor
5580 * Text.Pandoc.Error: Added `PandocUTF8DecodingError` constructor
1657 for `PandocError`. [API change]
5581 for `PandocError`. [API change]
@@ -1961,7 +5885,7 @@
1962 * `--ascii` now works with LaTeX output. 100% ASCII output can't be
5886 * `--ascii` now works with LaTeX output. 100% ASCII output can't be
1963 guaranteed, but the writer will use commands like `\"{a}` and `\l`
5887 guaranteed, but the writer will use commands like `\"{a}` and `\l`
1964− whenever possible, to avoid emiting a non-ASCII character.
5888+ whenever possible, to avoid emitting a non-ASCII character.
1966 * For HTML5 output, `--ascii` now uses HTML5 character reference
5890 * For HTML5 output, `--ascii` now uses HTML5 character reference
1967 entities rather than numerical entities.
5891 entities rather than numerical entities.
@@ -1990,7 +5914,7 @@
1990 used when `raw_html` is enabled.
5914 used when `raw_html` is enabled.
1992 * Powerpoint writer: support raw openxml (Jesse Rosenthal, #4976).
5916 * Powerpoint writer: support raw openxml (Jesse Rosenthal, #4976).
1993− This allows raw openxml blocks and inlines to be used in the pptx
5917+ This allows raw openxml blocks and inlines to be used in the pptx
1994 writer. Caveats: (1) It's up to the user to write
5918 writer. Caveats: (1) It's up to the user to write
1995 well-formed openxml. The chances for corruption, especially with
5919 well-formed openxml. The chances for corruption, especially with
1996 such a brittle format as pptx, is high. (2) Because of
5920 such a brittle format as pptx, is high. (2) Because of
@@ -2067,7 +5991,7 @@
2067 using a generic `for` loop with pairs`:
5991 using a generic `for` loop with pairs`:
2069 for field_name, field_content in pairs(element) do
5993 for field_name, field_content in pairs(element) do
2073 Raw table fields of AST elements should be considered an
5997 Raw table fields of AST elements should be considered an
@@ -3493,7 +7417,7 @@
3493 * JATS writer: Remove extraneous, significant whitespace (#4335,
7417 * JATS writer: Remove extraneous, significant whitespace (#4335,
3494 Nokome Bentley).
7418 Nokome Bentley).
3496− * html2pdf: inject base tag wih current working directory (#4413, Mauro
7420+ * html2pdf: inject base tag with current working directory (#4413, Mauro
3497 Bieg). This helps ensure that linked resources are included.
7421 Bieg). This helps ensure that linked resources are included.
3499 * Add Semigroup instances for everything for which we defined a
7423 * Add Semigroup instances for everything for which we defined a
@@ -5338,7 +9262,7 @@
5339 * Removed `hard_line_breaks` extension from `markdown_github` (#3594).
9263 * Removed `hard_line_breaks` extension from `markdown_github` (#3594).
5340 GitHub has two Markdown modes, one for long-form documents like READMEs
9264 GitHub has two Markdown modes, one for long-form documents like READMEs
5341− and one for short things like issue coments. In issue comments, a line
9265+ and one for short things like issue comments. In issue comments, a line
5342 break is treated as a hard line break. In README, wikis, etc., it is
9266 break is treated as a hard line break. In README, wikis, etc., it is
5343 treated as a space as in regular Markdown. Since pandoc is more likely to
9267 treated as a space as in regular Markdown. Since pandoc is more likely to
5344 be used to convert long-form documents from GitHub Markdown,
9268 be used to convert long-form documents from GitHub Markdown,
@@ -5534,7 +9458,7 @@
5534 * HTML reader: parse a span with class `smallcaps` as `SmallCaps`.
9458 * HTML reader: parse a span with class `smallcaps` as `SmallCaps`.
5536 * LaTeX reader:
9460 * LaTeX reader:
5538 + Implemented `\graphicspath` (#736).
9462 + Implemented `\graphicspath` (#736).
5539 + Properly handle column prefixes/suffixes. For example, in
9463 + Properly handle column prefixes/suffixes. For example, in
5540 `\begin{tabular}{>{$}l<{$}>{$}l<{$} >{$}l<{$}}`
9464 `\begin{tabular}{>{$}l<{$}>{$}l<{$} >{$}l<{$}}`
@@ -5691,7 +9615,7 @@
5691 top-level `Text.Pandoc` module.
9615 top-level `Text.Pandoc` module.
5693 + Changed `StringWriter` -> `TextWriter`.
9617 + Changed `StringWriter` -> `TextWriter`.
5694− + `getWriter` now retuns a pair of a reader and
9618+ + `getWriter` now returns a pair of a reader and
5695 `Extensions`, instead of building the extensions into the
9619 `Extensions`, instead of building the extensions into the
5696 reader (#3659). The calling code must explicitly set
9620 reader (#3659). The calling code must explicitly set
5697 `readerExtensions` using the `Extensions` returned. The
9621 `readerExtensions` using the `Extensions` returned. The
@@ -6104,7 +10028,7 @@
6104 of item disappear or are misplaced. Use `\texttt` instead.
10028 of item disappear or are misplaced. Use `\texttt` instead.
6105 + Fix problem with escaping in `lstinline` (#1629). Previously the
10029 + Fix problem with escaping in `lstinline` (#1629). Previously the
6106 LaTeX writer created invalid LaTeX when `--listings` was specified and
10030 LaTeX writer created invalid LaTeX when `--listings` was specified and
6107− a code span occured inside emphasis or another construction.
10031+ a code span occurred inside emphasis or another construction.
6108 + Fix error with line breaks after empty content (#2874). LaTeX
10032 + Fix error with line breaks after empty content (#2874). LaTeX
6109 requires something before a line break, so we insert a `~` if no
10033 requires something before a line break, so we insert a `~` if no
6110 printable content has yet been emitted.
10034 printable content has yet been emitted.
@@ -6211,7 +10135,7 @@
6212 + Don't drop smartTag contents (#2242).
10136 + Don't drop smartTag contents (#2242).
6213 + Handle local namespace declarations (#3365). Previously we didn't
10137 + Handle local namespace declarations (#3365). Previously we didn't
6214− recognize math, for example, when the xmlns declaration occured on
10138+ recognize math, for example, when the xmlns declaration occurred on
6215 the element and not the root.
10139 the element and not the root.
6216 + More efficient trimSps (#1530). Replacing `trimLineBreaks`. This
10140 + More efficient trimSps (#1530). Replacing `trimLineBreaks`. This
6217 does the work of `normalizeSpaces` as well, so we avoid the need for
10141 does the work of `normalizeSpaces` as well, so we avoid the need for
@@ -6445,7 +10369,7 @@
6445 + Allow `]` inside group in option brackets (#3857).
10369 + Allow `]` inside group in option brackets (#3857).
6446 + lstinline with braces can be used (verb cannot be used with braces)
10370 + lstinline with braces can be used (verb cannot be used with braces)
6447 (Marc Schreiber, #3535).
10371 (Marc Schreiber, #3535).
6448− + Fix keyval funtion: pandoc did not parse options in braces correctly
10372+ + Fix keyval function: pandoc did not parse options in braces correctly
6449 (Marc Schreiber, #3642).
10373 (Marc Schreiber, #3642).
6450 + When parsing raw LaTeX commands, include trailing space (#1773).
10374 + When parsing raw LaTeX commands, include trailing space (#1773).
6451 Otherwise things like `\noindent foo` break and turn into
10375 Otherwise things like `\noindent foo` break and turn into
@@ -6537,7 +10461,7 @@
6537 parse, the parser was applied too often, consuming too much of the
10461 parse, the parser was applied too often, consuming too much of the
6538 input. This only affects `many1Till p end` where `p` matches on a
10462 input. This only affects `many1Till p end` where `p` matches on a
6539 prefix of `end`.
10463 prefix of `end`.
6540− + Provide `parseFromString` (#3690). This is a verison of
10464+ + Provide `parseFromString` (#3690). This is a version of
6541 `parseFromString` specialied to ParserState, which resets
10465 `parseFromString` specialied to ParserState, which resets
6542 `stateLastStrPos` at the end. This is almost always what we want.
10466 `stateLastStrPos` at the end. This is almost always what we want.
6543 This fixes a bug where `_hi_` wasn't treated as emphasis in the
10467 This fixes a bug where `_hi_` wasn't treated as emphasis in the
@@ -6643,7 +10567,7 @@
6643 + Use `unicode-math` (Vaclav Haisman). Use `mathspec` with only
10567 + Use `unicode-math` (Vaclav Haisman). Use `mathspec` with only
6644 XeLaTeX on request.
10568 XeLaTeX on request.
6645 + Don't load `fontspec` before `unicode-math` (over there).
10569 + Don't load `fontspec` before `unicode-math` (over there).
6646− The `unicode-math` package loads `fontspec` so explict loading of
10570+ The `unicode-math` package loads `fontspec` so explicit loading of
6647 `fontspec` before `unicode-math` is not necessary.
10571 `fontspec` before `unicode-math` is not necessary.
6648 + Use `unicode-math` by default in default.latex template. mathspec will
10572 + Use `unicode-math` by default in default.latex template. mathspec will
6649 be used in xelatex if the `mathspec` variable is set; otherwise
10573 be used in xelatex if the `mathspec` variable is set; otherwise
@@ -6812,7 +10736,7 @@
6812 This is needed for dependency version numbers to be available,
10736 This is needed for dependency version numbers to be available,
6813 with Cabal > 2.
10737 with Cabal > 2.
6815−## pandoc 1.19.2.3 (2017-09-09)
10739+## pandoc 1.19.2.3 (2017-09-09)
6817 * Add CPP to Setup.hs so it works with Cabal >= 2 and < 2.
10741 * Add CPP to Setup.hs so it works with Cabal >= 2 and < 2.
@@ -7324,7 +11248,7 @@
7324 is now used in parsing RST and Markdown line blocks, DocBook
11248 is now used in parsing RST and Markdown line blocks, DocBook
7325 `linegroup`/`line` combinations, and Org-mode `VERSE` blocks.
11249 `linegroup`/`line` combinations, and Org-mode `VERSE` blocks.
7326 Previously `Para` blocks with hard linebreaks were used. `LineBlock`s
11250 Previously `Para` blocks with hard linebreaks were used. `LineBlock`s
7327− are handled specially in the following ouput formats: AsciiDoc
11251+ are handled specially in the following output formats: AsciiDoc
7328 (as `[verse]` blocks), ConTeXt (`\startlines`/`\endlines`),
11252 (as `[verse]` blocks), ConTeXt (`\startlines`/`\endlines`),
7329 HTML (`div` with a style), Markdown (line blocks if `line_blocks`
11253 HTML (`div` with a style), Markdown (line blocks if `line_blocks`
7330 is enabled), Org-mode (`VERSE` blocks), RST (line blocks). In
11254 is enabled), Org-mode (`VERSE` blocks), RST (line blocks). In
@@ -7471,7 +11395,7 @@
7471 comma-separated list.
11395 comma-separated list.
7472 + Give precedence to later meta lines. The last meta-line of any given
11396 + Give precedence to later meta lines. The last meta-line of any given
7473 type is the significant line. Previously the value of the first line
11397 type is the significant line. Previously the value of the first line
7474− was kept, even if more lines of the same type were encounterd.
11398+ was kept, even if more lines of the same type were encountered.
7475 + Read LaTeX_header as header-includes. LaTeX-specific header commands
11399 + Read LaTeX_header as header-includes. LaTeX-specific header commands
7476 can be defined in `#+LaTeX_header` lines. They are parsed as
11400 can be defined in `#+LaTeX_header` lines. They are parsed as
7477 format-specific inlines to ensure that they will only show up in LaTeX
11401 format-specific inlines to ensure that they will only show up in LaTeX
@@ -7519,7 +11443,7 @@
7519 be able to figure out internal links to a header in a docx if the
11443 be able to figure out internal links to a header in a docx if the
7520 anchor span was empty. We change that to read the inlines out of the
11444 anchor span was empty. We change that to read the inlines out of the
7521 first anchor span in a header.
11445 first anchor span in a header.
7522− + Let headers use exisiting id. Previously we always generated an id for
11446+ + Let headers use existing id. Previously we always generated an id for
7523 headers (since they wouldn't bring one from Docx). Now we let it use an
11447 headers (since they wouldn't bring one from Docx). Now we let it use an
7524 existing one if possible. This should allow us to recurs through anchor
11448 existing one if possible. This should allow us to recurs through anchor
@@ -7623,7 +11547,7 @@
7623 + Don't emit HTML for tables unless `raw_html` extension is set (#3154).
11547 + Don't emit HTML for tables unless `raw_html` extension is set (#3154).
7624 Emit `[TABLE]` if no suitable table formats are enabled and raw HTML
11548 Emit `[TABLE]` if no suitable table formats are enabled and raw HTML
7625 is disabled.
11549 is disabled.
7626− + Check for the `raw_html` extension before emiting a raw HTML block.
11550+ + Check for the `raw_html` extension before emitting a raw HTML block.
7627 + Abstract out note/ref function (Jesse Rosenthal).
11551 + Abstract out note/ref function (Jesse Rosenthal).
7628 + Add ReaderT monad for environment variables (Jesse Rosenthal).
11552 + Add ReaderT monad for environment variables (Jesse Rosenthal).
@@ -8098,7 +12022,7 @@
8098 character. Empty rows where parsed as alignment rows and dropped from
12022 character. Empty rows where parsed as alignment rows and dropped from
8099 the output.
12023 the output.
8100 + Fix spacing after LaTeX-style symbols.
12024 + Fix spacing after LaTeX-style symbols.
8101− The org-reader was droping space after unescaped LaTeX-style symbol
12025+ The org-reader was dropping space after unescaped LaTeX-style symbol
8102 commands: `\ForAll \Auml` resulted in `∀Ä` but should give `∀ Ä`
12026 commands: `\ForAll \Auml` resulted in `∀Ä` but should give `∀ Ä`
8103 instead. This seems to be because the LaTeX-reader treats the
12027 instead. This seems to be because the LaTeX-reader treats the
8104 command-terminating space as part of the command. Dropping the trailing
12028 command-terminating space as part of the command. Dropping the trailing
@@ -8246,7 +12170,7 @@
8246 benefit that footnotes and links can be in different files, but for
12170 benefit that footnotes and links can be in different files, but for
8247 some purposes it is useful to parse the individual files first
12171 some purposes it is useful to parse the individual files first
8248 and then combine their outputs (e.g. when the files use footnotes
12172 and then combine their outputs (e.g. when the files use footnotes
8249− or links with the same labels). The `--file-scope` option causes
12173+ or links with the same labels). The `--file-scope` option causes
8250 pandoc to parse the files first, and then combine the parsed output,
12174 pandoc to parse the files first, and then combine the parsed output,
8251 instead of combining before parsing. `--file-scope` is selected
12175 instead of combining before parsing. `--file-scope` is selected
8252 automatically for binary input files (which cannot be concatenated)
12176 automatically for binary input files (which cannot be concatenated)
@@ -8433,7 +12357,7 @@
8434 * Entity handling fixes: improved handling of entities like
12358 * Entity handling fixes: improved handling of entities like
8435 `⟨` that require a trailing semicolon. Allow uppercase
12359 `⟨` that require a trailing semicolon. Allow uppercase
8436− `x` in numerical hexidecimal character references, working
12360+ `x` in numerical hexadecimal character references, working
8437 around a tagsoup bug.
12361 around a tagsoup bug.
8439 * `stack.yaml` - use lts-4.0, but with older aeson to avoid excessive
12363 * `stack.yaml` - use lts-4.0, but with older aeson to avoid excessive
@@ -9918,7 +13842,7 @@
9918 * `Text.Pandoc.Shared`:
13842 * `Text.Pandoc.Shared`:
9920 + Make safeRead safe (#1801, Matthew Pickering).
13844 + Make safeRead safe (#1801, Matthew Pickering).
9921− + Addded `mapLeft`, `hush` (Matthew Pickering).
13845+ + Added `mapLeft`, `hush` (Matthew Pickering).
9923 * `Text.Pandoc.Pretty`:
13847 * `Text.Pandoc.Pretty`:
@@ -9986,7 +13910,7 @@
9987 + Use `text:p` instead of `text:h` for title.
13911 + Use `text:p` instead of `text:h` for title.
9988 Using `text:h` causes problems with numbering. Closes #2059.
13912 Using `text:h` causes problems with numbering. Closes #2059.
9989− Thansk to @nkalvi for diagnosing this.
13913+ Thanks to @nkalvi for diagnosing this.
9991 * reveal.js template:
13915 * reveal.js template:
@@ -10091,7 +14015,7 @@
10091 + Parse RST class directives. The class directive accepts one or more
14015 + Parse RST class directives. The class directive accepts one or more
10092 class names, and creates a Div value with those classes. If the
14016 class names, and creates a Div value with those classes. If the
10093 directive has an indented body, the body is parsed as the children of
14017 directive has an indented body, the body is parsed as the children of
10094− the Div. If not, the first block folowing the directive is made a
14018+ the Div. If not, the first block following the directive is made a
10095 child of the Div. This differs from the behavior of rst2xml, which
14019 child of the Div. This differs from the behavior of rst2xml, which
10096 does not create a Div element. Instead, the specified classes are
14020 does not create a Div element. Instead, the specified classes are
10097 applied to each child of the directive. However, most Pandoc Block
14021 applied to each child of the directive. However, most Pandoc Block
@@ -10252,7 +14176,7 @@
10252 * RST writer:
14176 * RST writer:
10254 + Ensure blank line after figure.
14178 + Ensure blank line after figure.
10255− + Avoid exces whitespace after last list item (#1777).
14179+ + Avoid excess whitespace after last list item (#1777).
10256 + Wrap line blocks with spaces before continuations (#1656).
14180 + Wrap line blocks with spaces before continuations (#1656).
10257 + Fixed double-rendering of footnotes in RST tables (#1769).
14181 + Fixed double-rendering of footnotes in RST tables (#1769).
@@ -10314,7 +14238,7 @@
10314 + Renumber header and footer relationships to avoid collisions (Jesse
14238 + Renumber header and footer relationships to avoid collisions (Jesse
10315 Rosenthal). We previously took the old relationship names of the
14239 Rosenthal). We previously took the old relationship names of the
10316 headers and footer in secptr. That led to collisions. We now make
14240 headers and footer in secptr. That led to collisions. We now make
10317− a map of availabl names in the relationships file, and then rename
14241+ a map of available names in the relationships file, and then rename
10318 in secptr.
14242 in secptr.
10320 * ConTeXt writer:
14244 * ConTeXt writer:
@@ -11037,7 +14961,7 @@
11037 different versions of the `directory` library.
14961 different versions of the `directory` library.
11039 + Added `Text.Pandoc.Compat.Except` to allow building against
14963 + Added `Text.Pandoc.Compat.Except` to allow building against
11040− different verions of `mtl`.
14964+ different versions of `mtl`.
11042 * Code cleanup in some writers, using Reader monad to avoid
14966 * Code cleanup in some writers, using Reader monad to avoid
11043 passing options parameter around (Matej Kollar).
14967 passing options parameter around (Matej Kollar).
@@ -11693,7 +15617,7 @@
11693 + Parse contents of curly quotes or matched `"` as quotes.
15617 + Parse contents of curly quotes or matched `"` as quotes.
11694 + Support `\textnormal` as span with class `nodecor`.
15618 + Support `\textnormal` as span with class `nodecor`.
11695 This is needed for pandoc-citeproc.
15619 This is needed for pandoc-citeproc.
11696− + Improved citation parsing. This fixes a run-time error that occured
15620+ + Improved citation parsing. This fixes a run-time error that occurred
11697 with `\citet{}` (empty list of keys). It also ensures that empty keys
15621 with `\citet{}` (empty list of keys). It also ensures that empty keys
11698 don't get produced.
15622 don't get produced.
@@ -11987,7 +15911,7 @@
11987 to make that the current directory before running pandoc. (#942)
15911 to make that the current directory before running pandoc. (#942)
11989 * Better error reporting in some readers, due to changes in `readWith`:
15913 * Better error reporting in some readers, due to changes in `readWith`:
11990− the line in which the error occured is printed, with a caret pointing
15914+ the line in which the error occurred is printed, with a caret pointing
11991 to the column.
15915 to the column.
11993 * All slide formats now support incremental slide view for definition lists.
15917 * All slide formats now support incremental slide view for definition lists.
@@ -12041,7 +15965,7 @@
12041 * A new generic block container (`Div`) has been added to `Block`,
15965 * A new generic block container (`Div`) has been added to `Block`,
12042 and a generic inline container (`Span`) has been added to `Inline`.
15966 and a generic inline container (`Span`) has been added to `Inline`.
12043 These can take attributes. They will render in HTML, Textile,
15967 These can take attributes. They will render in HTML, Textile,
12044− MediaWiki, Org, RST and and Markdown (with `markdown_in_html`
15968+ MediaWiki, Org, RST and Markdown (with `markdown_in_html`
12045 extension) as HTML `<div>` and `<span>` elements; in other formats
15969 extension) as HTML `<div>` and `<span>` elements; in other formats
12046 they will simply pass through their contents. But they can be
15970 they will simply pass through their contents. But they can be
12047 targeted by scripts.
15971 targeted by scripts.
@@ -12676,7 +16600,7 @@
12676 + Include HTML TOC, even in epub2. The TOC is included in `<spine>`,
16600 + Include HTML TOC, even in epub2. The TOC is included in `<spine>`,
12677 but `linear` is set to `no` unless the `--toc` option is specified.
16601 but `linear` is set to `no` unless the `--toc` option is specified.
12678 Include `<guide>` element in OPF. This should allow the TOC to
16602 Include `<guide>` element in OPF. This should allow the TOC to
12679− be useable in Kindles when converted with kindlegen. Closes #773.
16603+ be usable in Kindles when converted with kindlegen. Closes #773.
12681 * `Text.Pandoc.Parsing`: Optimized `oneOfStringsCI`.
16605 * `Text.Pandoc.Parsing`: Optimized `oneOfStringsCI`.
12682 This dramatically reduces the speed penalty that comes from enabling the
16606 This dramatically reduces the speed penalty that comes from enabling the
@@ -13038,7 +16962,7 @@
13038### API changes
16962### API changes
13040 * `Text.Pandoc.Definition`: Added `Attr` field to `Header`.
16964 * `Text.Pandoc.Definition`: Added `Attr` field to `Header`.
13041− Previously header identifers were autogenerated by the writers.
16965+ Previously header identifiers were autogenerated by the writers.
13042 Now they are added in the readers (either automatically or explicitly).
16966 Now they are added in the readers (either automatically or explicitly).
13044 * `Text.Pandoc.Builder`:
16968 * `Text.Pandoc.Builder`:
@@ -13272,10 +17196,10 @@
13272 be a parenthesized emphasized "hello".
17196 be a parenthesized emphasized "hello".
13273 The new list is taken from the PHP source of textile 2.4.
17197 The new list is taken from the PHP source of textile 2.4.
13274 + Fixed autolinks. Previously the textile reader and writer
17198 + Fixed autolinks. Previously the textile reader and writer
13275− incorrectly implented RST-style autolinks for URLs and email
17199+ incorrectly implemented RST-style autolinks for URLs and email
13276 addresses. This has been fixed. Now an autolink is done this way:
17200 addresses. This has been fixed. Now an autolink is done this way:
13277 `"$":http://myurl.com`.
17201 `"$":http://myurl.com`.
13278− + Fixed footnotes bug in textile. This affected notes occuring
17202+ + Fixed footnotes bug in textile. This affected notes occurring
13279 before punctuation, e.g. `foo[1].`. Closes #518.
17203 before punctuation, e.g. `foo[1].`. Closes #518.
13281 * LaTeX reader:
17205 * LaTeX reader:
@@ -14335,7 +18259,7 @@
14335 Items are no longer installed as root.
18259 Items are no longer installed as root.
14336 Man pages are zipped and given proper permissions.
18260 Man pages are zipped and given proper permissions.
14338− * Modified windows installer generater to use cabal-dev.
18262+ * Modified windows installer generator to use cabal-dev.
14340 * Setup: Making man pages now works with cabal-dev (at least on OSX). In
18264 * Setup: Making man pages now works with cabal-dev (at least on OSX). In
14341 Setup.hs we now invoke 'runghc' in a way that points it to the correct
18265 Setup.hs we now invoke 'runghc' in a way that points it to the correct