@@ -72,6 +72,16 @@
7272 which URLs it is fetching, but not giving the full header). In
7373 addition, there are better error messages when fetching a URL fails.
7474
75+ * Previously we used to store the directory of the first input file,
76+ even if it was local, and used this as a base directory for finding
77+ images in ODT, EPUB, Docx, and PDF. This has been confusing to many
78+ users. So we now look for images relative to the current
79+ working directory, even if the first file argument is in another
80+ directory. Note that this change may break some existing workflows.
81+ If you have been assuming that relative links will be interpreted
82+ relative to the directory of the first file argument, you'll need
83+ to make that the current directory before running pandoc. (#942)
84+
7585 * Better error reporting in some readers, due to changes in `readWith`:
7686 the line in which the error occured is printed, with a caret pointing
7787 to the column.
@@ -92,6 +102,13 @@
92102 pandoc will still parse it into a title, section, header, and
93103 footer. But you can also specify these elements explicitly (#885).
94104
105+ * Markdown reader
106+
107+ + Added support for YAML metadata blocks, which can come anywhere
108+ in the document (not just at the beginning). A document can contain
109+ multiple YAML metadata blocks.
110+ + HTML span and div tags are parsed as pandoc Span and Div elements.
111+
95112 * Markdown writer
96113
97114 + Allow simple tables to be printed as grid tables,
@@ -139,7 +156,26 @@
139156 + Export `varListToJSON`.
140157
141158 * `Text.Pandoc.PDF` exports `makePDF` instead of `tex2pdf`.
142−
159+
160+ * `Text.Pandoc`:
161+
162+ + Made `toJsonFilter` an alias for `toJSONFilter` from `Text.Pandoc.JSON`.
163+ + Removed `ToJsonFilter` typeclass. `ToJSONFilter` from
164+ `Text.Pandoc.JSON` should be used instead. (Compiling against
165+ pandoc-types instead of pandoc will also produce smaller executables.)
166+ * Removed the deprecated `jsonFilter` function.
167+ + Added `readJSON`, `writeJSON` to the API (#817).
168+
169+ * `Text.Pandoc.Options`:
170+
171+ + Added `Ext_lists_without_preceding_blankline`,
172+ `Ext_ascii_identifiers`, `Ext_ignore_line_breaks`,
173+ `Ext_yaml_metadataBlock` to `Extension`.
174+ + Changed `writerSourceDir` to `writerSourceURL` and changed the type to
175+ a `Maybe`. `writerSourceURL` is set to 'Just url' when the first
176+ command-line argument is an absolute URL. (So, relative links will be
177+ resolved in relation to the first page.) Otherwise, 'Nothing'.
178+
143179 [bug fixes]
144180
145181 * In markdown, don't autolink a bare URI that is followed by `</a>`
@@ -168,7 +204,11 @@
168204 temp directory path may contain tildes, which can cause
169205 problems in LaTeX (#777).
170206 + Put temporary output directory in `TEXINPUTS` (see #917).
171−
207+ + `makePDF` tries to download images that are not found locally,
208+ if the first argument is a URL (#917).
209+ + If compiling with `pdflatex` yields an encoding error, offer
210+ the suggestion to use `--latex-engine=xelatex`.
211+
172212 * Produce automatic header identifiers in parsing textile, RST,
173213 and LaTeX, unless `auto_identifiers` extension is disabled (#967).
174214
@@ -188,6 +228,18 @@
188228 Reason: some browsers have trouble with the gzipped js file,
189229 at least on the local file system (#795).
190230
231+ * Markdown reader
232+
233+ + Properly handle blank line at beginning of input (#882).
234+ + Fixed bug in unmatched reference links. The input
235+ `[*infile*] [*outfile*]` was getting improperly parsed:
236+ "infile" was emphasized, but "*outfile*" was literal (#883).
237+ + Allow internal `+` in citation identifiers (#856).
238+ + Allow `.` or `)` after `#` in ATX headers if no `fancy_lists`.
239+ + Do not generate blank title, author, or date metadata elements.
240+ Leave these out entirely if they aren't present.
241+ + Allow backtick code blocks not to be preceded by blank line (#975).
242+
191243 * Textile reader:
192244
193245 + Correctly handle entities.
@@ -409,6 +461,13 @@
409461 * Added `Text.Pandoc.Asciify` utility module.
410462 This exports functions to create ASCII-only versions of identifiers.
411463
464+ * `Text.Pandoc.Parsing`
465+
466+ + Generalized state type on `readWith` (API change).
467+ + Specialize readWith to `String` input. (API change).
468+ + In `ParserState`, replace `stateTitle`, `stateAuthors`, `stateDate`
469+ with `stateMeta` and `stateMeta'`.
470+
412471 * `Text.Pandoc.UTF8`: use strict bytestrings in reading. The use of lazy
413472 bytestrings seemed to cause problems using pandoc on 64-bit Windows
414473 7/8 (#874).
@@ -447,6 +506,18 @@
447506
448507------------------
449508
509+ TODO - metadata changes
510+
511+ TODO - json module
512+
513+ TODO - walk module
514+
515+ TODO - citation changes
516+ removed Biblio
517+ removed integrated citation support
518+ etc.
519+ include pandoc-citeproc changes
520+
450521 * citeproc support has been removed from core pandoc (API changes).
451522
452523 + The `--bibliography`, `--csl`, and `--citation-abbreviation` options are
@@ -497,15 +568,6 @@
497568 and `<span>` elements; in other formats they will simply pass through
498569 their contents. But they can be targeted by scripts.
499570
500− * `Text.Pandoc`
501−
502− + Make `toJsonFilter` an alias for `toJSONFilter` from `Text.Pandoc.JSON`.
503− + Removed `ToJsonFilter` typeclass. `ToJSONFilter` from
504− `Text.Pandoc.JSON` should be used instead. (Compiling against
505− pandoc-types instead of pandoc will also produce smaller executables.)
506− * Removed the deprecated `jsonFilter` function.
507− + Added `readJSON`, `writeJSON` to the API (#817).
508−
509571 [ TODO - systematic documentation of pandoc-types API changes,
510572 including .JSON, .Walk and changes to .Definition, .Builder.
511573 Include the new Format newtype, and the new Span and Div
@@ -547,76 +609,6 @@
547609 pandoc types. They use GHC generics and should be faster than the
548610 old JSON serialization using `Data.Aeson.Generic`.
549611
550−
551− * Markdown reader
552−
553− + Properly handle blank line at beginning of input (#882).
554− + Fixed bug in unmatched reference links. The input
555− `[*infile*] [*outfile*]` was getting improperly parsed:
556− "infile" was emphasized, but "*outfile*" was literal (#883).
557− + Check for blank lines first in blocks. (And skip them. This
558− might speed things up in some cases.)
559− + Allow internal `+` in citation identifiers (#856).
560− + Allow `.` or `)` after `#` in ATX headers if no `fancy_lists`.
561− + Do not generate blank title, author, or date metadata elements.
562− Leave these out entirely if they aren't present.
563− + Allow backtick code blocks not to be preceded by blank line (#975).
564−
565− * Markdown reader
566−
567− + Added support for YAML metadata blocks, which can come anywhere
568− in the document (not just at the beginning). A document can contain
569− multiple YAML metadata blocks.
570− + HTML span and div tags are parsed as pandoc Span and Div elements.
571−
572−
573−
574−
575−
576− * `Text.Pandoc.PDF`
577−
578− + `makePDF` walks the pandoc AST and checks for the existence of
579− images in the local directory. If they are not found, it attempts
580− to find them, either in the directory containing the first source
581− file, or at an absolute URL, or at a URL relative to the base URL
582− of the first command line argument (#917).
583− + If compiling with `pdflatex` yields an encoding error, offer
584− the suggestion to use `--latex-engine=xelatex`.
585−
586−
587− * `Text.Pandoc.Parsing`
588−
589− + Generalized state type on `readWith` (API change).
590− + Specialize readWith to `String` input. (API change).
591− + In `ParserState`, replace `stateTitle`, `stateAuthors`, `stateDate`
592− with `stateMeta` and `stateMeta'`.
593−
594− * `Text.Pandoc.Options`
595− * Added `Ext_lists_without_preceding_blankline` to
596− `Extension` in `Options`. Added this option to
597− `githubMarkdownExtensions`.
598− + Implemented `Ext_ascii_identifiers`.
599− + Ext_ignore_line_breaks.
600− + Added `Ext_yaml_metadata_block`. (API change.)
601− + Added `Ext_ascii_identifiers`. (API change.)
602− This will force `Ext_auto_identifiers` to use ASCII only.
603− Set as default for github markdown.
604− + Changed `writerSourceDir` to `writerSourceURL` and changed
605− the type to a `Maybe`. Previously we used to store the directory
606− of the first input file, even if it was local, and used this as a
607− base directory for finding images in ODT, EPUB, Docx, and PDF.
608− This has been confusing to many users. It seems better to look for
609− images relative to the current working directory, even if the first
610− file argument is in another directory. `writerSourceURL` is set
611− to 'Just url' when the first command-line argument is an absolute
612− URL. (So, relative links will be resolved in relation to the first
613− page.) Otherwise, 'Nothing'. The ODT, EPUB, Docx, and PDF
614− writers have been modified accordingly. Note that this change may
615− break some existing workflows. If you have been assuming that
616− relative links will be interpreted relative to the directory of
617− the first file argument, you'll need to make that the current
618− directory before running pandoc. API change (#942).
619−
620612pandoc (1.11.1)
621613
622614 * Markdown reader: