2title: Pandoc User's Guide
9`pandoc` [*options*] [*input-file*]...
13Pandoc is a [Haskell] library for converting from one markup format to
14another, and a command-line tool that uses this library.
16Pandoc can convert between numerous markup and word processing formats,
17including, but not limited to, various flavors of [Markdown], [HTML],
18[LaTeX] and [Word docx]. For the full lists of input and output formats,
19see the `--from` and `--to` [options below][General options].
20Pandoc can also produce [PDF] output: see [creating a PDF], below.
22Pandoc's enhanced version of Markdown includes syntax for [tables],
23[definition lists], [metadata blocks], [footnotes], [citations], [math],
24and much more. See below under [Pandoc's Markdown].
26Pandoc has a modular design: it consists of a set of readers, which parse
27text in a given format and produce a native representation of the document
28(an _abstract syntax tree_ or AST), and a set of writers, which convert
29this native representation into a target format. Thus, adding an input
30or output format requires only adding a reader or writer. Users can also
31run custom [pandoc filters] to modify the intermediate AST.
33Because pandoc's intermediate representation of a document is less
34expressive than many of the formats it converts between, one should
35not expect perfect conversions between every format and every other.
36Pandoc attempts to preserve the structural elements of a document, but
37not formatting details such as margin size. And some document elements,
38such as complex tables, may not fit into pandoc's simple document
39model. While conversions from pandoc's Markdown to all formats aspire
40to be perfect, conversions from formats more expressive than pandoc's
41Markdown can be expected to be lossy.
45If no *input-files* are specified, input is read from *stdin*.
46Output goes to *stdout* by default. For output to a file,
49 pandoc -o output.html input.txt
51By default, pandoc produces a document fragment. To produce a standalone
52document (e.g. a valid HTML file including `<head>` and `<body>`),
53use the `-s` or `--standalone` flag:
55 pandoc -s -o output.html input.txt
57For more information on how standalone documents are produced, see
60If multiple input files are given, `pandoc` will concatenate them all (with
61blank lines between them) before parsing. (Use `--file-scope` to parse files
66The format of the input and output can be specified explicitly using
67command-line options. The input format can be specified using the
68`-f/--from` option, the output format using the `-t/--to` option.
69Thus, to convert `hello.txt` from Markdown to LaTeX, you could type:
71 pandoc -f markdown -t latex hello.txt
73To convert `hello.html` from HTML to Markdown:
75 pandoc -f html -t markdown hello.html
77Supported input and output formats are listed below under [Options]
78(see `-f` for input formats and `-t` for output formats). You
79can also use `pandoc --list-input-formats` and
80`pandoc --list-output-formats` to print lists of supported
83If the input or output format is not specified explicitly, `pandoc`
84will attempt to guess it from the extensions of the filenames.
87 pandoc -o hello.tex hello.txt
89will convert `hello.txt` from Markdown to LaTeX. If no output file
90is specified (so that output goes to *stdout*), or if the output file's
91extension is unknown, the output format will default to HTML.
92If no input file is specified (so that input comes from *stdin*), or
93if the input files' extensions are unknown, the input format will
94be assumed to be Markdown.
98Pandoc uses the UTF-8 character encoding for both input and output.
99If your local character encoding is not UTF-8, you
100should pipe input and output through [`iconv`]:
102 iconv -t utf-8 input.txt | pandoc | iconv -f utf-8
104Note that in some output formats (such as HTML, LaTeX, ConTeXt,
105RTF, OPML, DocBook, and Texinfo), information about
106the character encoding is included in the document header, which
107will only be included if you use the `-s/--standalone` option.
109[`iconv`]: https://www.gnu.org/software/libiconv/
113To produce a PDF, specify an output file with a `.pdf` extension:
115 pandoc test.txt -o test.pdf
117By default, pandoc will use LaTeX to create the PDF, which requires
118that a LaTeX engine be installed (see `--pdf-engine` below).
119Alternatively, pandoc can use ConTeXt, roff ms, or HTML as an
120intermediate format. To do this, specify an output file with a
121`.pdf` extension, as before, but add the `--pdf-engine` option
122or `-t context`, `-t html`, or `-t ms` to the command line.
123The tool used to generate the PDF from the intermediate format
124may be specified using `--pdf-engine`.
126You can control the PDF style using variables, depending on
127the intermediate format used: see [variables for LaTeX],
128[variables for ConTeXt], [variables for `wkhtmltopdf`],
129[variables for ms]. When HTML is used as an intermediate
130format, the output can be styled using `--css`.
132To debug the PDF creation, it can be useful to look at the intermediate
133representation: instead of `-o test.pdf`, use for example `-s -o test.tex`
134to output the generated LaTeX. You can then test it with `pdflatex test.tex`.
136When using LaTeX, the following packages need to be available
137(they are included with all recent versions of [TeX Live]):
138[`amsfonts`], [`amsmath`], [`lm`], [`unicode-math`],
139[`iftex`], [`listings`] (if the
140`--listings` option is used), [`fancyvrb`], [`longtable`],
141[`booktabs`], [`graphicx`] (if the document
142contains images), [`hyperref`], [`xcolor`],
143[`ulem`], [`geometry`] (with the `geometry` variable set),
144[`setspace`] (with `linestretch`), and
145[`babel`] (with `lang`). The use of `xelatex` or `lualatex` as
146the PDF engine requires [`fontspec`]. `lualatex` uses
147[`selnolig`]. `xelatex` uses [`polyglossia`] (with `lang`),
148[`xecjk`], and [`bidi`] (with the `dir` variable set). If the
149`mathspec` variable is set, `xelatex` will use [`mathspec`]
150instead of [`unicode-math`]. The [`upquote`] and [`microtype`]
151packages are used if available, and [`csquotes`] will be used
152for [typography] if the `csquotes` variable or metadata field is
153set to a true value. The [`natbib`], [`biblatex`], [`bibtex`],
154and [`biber`] packages can optionally be used for [citation
155rendering]. The following packages will be used to improve
156output quality if present, but pandoc does not require them to
157be present: [`upquote`] (for straight quotes in verbatim
158environments), [`microtype`] (for better spacing adjustments),
159[`parskip`] (for better inter-paragraph spaces), [`xurl`] (for
160better line breaks in URLs), [`bookmark`] (for better PDF
161bookmarks), and [`footnotehyper`] or [`footnote`] (to allow
164[TeX Live]: https://www.tug.org/texlive/
165[`amsfonts`]: https://ctan.org/pkg/amsfonts
166[`amsmath`]: https://ctan.org/pkg/amsmath
167[`babel`]: https://ctan.org/pkg/babel
168[`biber`]: https://ctan.org/pkg/biber
169[`biblatex`]: https://ctan.org/pkg/biblatex
170[`bibtex`]: https://ctan.org/pkg/bibtex
171[`bidi`]: https://ctan.org/pkg/bidi
172[`bookmark`]: https://ctan.org/pkg/bookmark
173[`booktabs`]: https://ctan.org/pkg/booktabs
174[`csquotes`]: https://ctan.org/pkg/csquotes
175[`fancyvrb`]: https://ctan.org/pkg/fancyvrb
176[`fontspec`]: https://ctan.org/pkg/fontspec
177[`footnote`]: https://ctan.org/pkg/footnote
178[`footnotehyper`]: https://ctan.org/pkg/footnotehyper
179[`geometry`]: https://ctan.org/pkg/geometry
180[`graphicx`]: https://ctan.org/pkg/graphicx
181[`grffile`]: https://ctan.org/pkg/grffile
182[`hyperref`]: https://ctan.org/pkg/hyperref
183[`iftex`]: https://ctan.org/pkg/iftex
184[`listings`]: https://ctan.org/pkg/listings
185[`lm`]: https://ctan.org/pkg/lm
186[`longtable`]: https://ctan.org/pkg/longtable
187[`mathspec`]: https://ctan.org/pkg/mathspec
188[`microtype`]: https://ctan.org/pkg/microtype
189[`natbib`]: https://ctan.org/pkg/natbib
190[`parskip`]: https://ctan.org/pkg/parskip
191[`polyglossia`]: https://ctan.org/pkg/polyglossia
192[`prince`]: https://www.princexml.com/
193[`setspace`]: https://ctan.org/pkg/setspace
194[`ulem`]: https://ctan.org/pkg/ulem
195[`unicode-math`]: https://ctan.org/pkg/unicode-math
196[`upquote`]: https://ctan.org/pkg/upquote
197[`weasyprint`]: https://weasyprint.org
198[`wkhtmltopdf`]: https://wkhtmltopdf.org
199[`xcolor`]: https://ctan.org/pkg/xcolor
200[`xecjk`]: https://ctan.org/pkg/xecjk
201[`xurl`]: https://ctan.org/pkg/xurl
202[`selnolig`]: https://ctan.org/pkg/selnolig
206## Reading from the Web
208Instead of an input file, an absolute URI may be given. In this case
209pandoc will fetch the content using HTTP:
211 pandoc -f html -t markdown https://www.fsf.org
213It is possible to supply a custom User-Agent string or other
214header when requesting a document from a URL:
216 pandoc -f html -t markdown --request-header User-Agent:"Mozilla/5.0" \
221## General options {.options}
223`-f` *FORMAT*, `-r` *FORMAT*, `--from=`*FORMAT*, `--read=`*FORMAT*
225: Specify input format. *FORMAT* can be:
228 - `bibtex` ([BibTeX] bibliography)
229 - `biblatex` ([BibLaTeX] bibliography)
230 - `commonmark` ([CommonMark] Markdown)
231 - `commonmark_x` ([CommonMark] Markdown with extensions)
232 - `creole` ([Creole 1.0])
233 - `csljson` ([CSL JSON] bibliography)
234 - `csv` ([CSV] table)
235 - `docbook` ([DocBook])
236 - `docx` ([Word docx])
237 - `dokuwiki` ([DokuWiki markup])
239 - `fb2` ([FictionBook2] e-book)
240 - `gfm` ([GitHub-Flavored Markdown]),
241 or the deprecated and less accurate `markdown_github`;
242 use [`markdown_github`](#markdown-variants) only
243 if you need extensions not supported in [`gfm`](#markdown-variants).
244 - `haddock` ([Haddock markup])
246 - `ipynb` ([Jupyter notebook])
247 - `jats` ([JATS] XML)
248 - `jira` ([Jira]/Confluence wiki markup)
249 - `json` (JSON version of native AST)
251 - `markdown` ([Pandoc's Markdown])
252 - `markdown_mmd` ([MultiMarkdown])
253 - `markdown_phpextra` ([PHP Markdown Extra])
254 - `markdown_strict` (original unextended [Markdown])
255 - `mediawiki` ([MediaWiki markup])
258 - `native` (native Haskell)
261 - `org` ([Emacs Org mode])
262 - `rst` ([reStructuredText])
264 - `textile` ([Textile])
265 - `tikiwiki` ([TikiWiki markup])
266 - `twiki` ([TWiki markup])
267 - `vimwiki` ([Vimwiki])
270 Extensions can be individually enabled or disabled by
271 appending `+EXTENSION` or `-EXTENSION` to the format name.
272 See [Extensions] below, for a list of extensions and
273 their names. See `--list-input-formats` and `--list-extensions`,
276`-t` *FORMAT*, `-w` *FORMAT*, `--to=`*FORMAT*, `--write=`*FORMAT*
278: Specify output format. *FORMAT* can be:
280 ::: {#output-formats}
281 - `asciidoc` ([AsciiDoc]) or `asciidoctor` ([AsciiDoctor])
282 - `beamer` ([LaTeX beamer][`beamer`] slide show)
283 - `bibtex` ([BibTeX] bibliography)
284 - `biblatex` ([BibLaTeX] bibliography)
285 - `commonmark` ([CommonMark] Markdown)
286 - `commonmark_x` ([CommonMark] Markdown with extensions)
287 - `context` ([ConTeXt])
288 - `csljson` ([CSL JSON] bibliography)
289 - `docbook` or `docbook4` ([DocBook] 4)
290 - `docbook5` (DocBook 5)
291 - `docx` ([Word docx])
292 - `dokuwiki` ([DokuWiki markup])
293 - `epub` or `epub3` ([EPUB] v3 book)
295 - `fb2` ([FictionBook2] e-book)
296 - `gfm` ([GitHub-Flavored Markdown]),
297 or the deprecated and less accurate `markdown_github`;
298 use [`markdown_github`](#markdown-variants) only
299 if you need extensions not supported in [`gfm`](#markdown-variants).
300 - `haddock` ([Haddock markup])
301 - `html` or `html5` ([HTML], i.e. [HTML5]/XHTML [polyglot markup])
302 - `html4` ([XHTML] 1.0 Transitional)
303 - `icml` ([InDesign ICML])
304 - `ipynb` ([Jupyter notebook])
305 - `jats_archiving` ([JATS] XML, Archiving and Interchange Tag Set)
306 - `jats_articleauthoring` ([JATS] XML, Article Authoring Tag Set)
307 - `jats_publishing` ([JATS] XML, Journal Publishing Tag Set)
308 - `jats` (alias for `jats_archiving`)
309 - `jira` ([Jira]/Confluence wiki markup)
310 - `json` (JSON version of native AST)
313 - `markdown` ([Pandoc's Markdown])
314 - `markdown_mmd` ([MultiMarkdown])
315 - `markdown_phpextra` ([PHP Markdown Extra])
316 - `markdown_strict` (original unextended [Markdown])
317 - `mediawiki` ([MediaWiki markup])
320 - `native` (native Haskell),
321 - `odt` ([OpenOffice text document][ODT])
323 - `opendocument` ([OpenDocument])
324 - `org` ([Emacs Org mode])
326 - `plain` (plain text),
327 - `pptx` ([PowerPoint] slide show)
328 - `rst` ([reStructuredText])
329 - `rtf` ([Rich Text Format])
330 - `texinfo` ([GNU Texinfo])
331 - `textile` ([Textile])
332 - `slideous` ([Slideous] HTML and JavaScript slide show)
333 - `slidy` ([Slidy] HTML and JavaScript slide show)
334 - `dzslides` ([DZSlides] HTML5 + JavaScript slide show),
335 - `revealjs` ([reveal.js] HTML5 + JavaScript slide show)
336 - `s5` ([S5] HTML and JavaScript slide show)
337 - `tei` ([TEI Simple])
338 - `xwiki` ([XWiki markup])
339 - `zimwiki` ([ZimWiki markup])
340 - the path of a custom Lua writer, see [Custom writers] below
343 Note that `odt`, `docx`, `epub`, and `pdf` output will not be directed
344 to *stdout* unless forced with `-o -`.
346 Extensions can be individually enabled or
347 disabled by appending `+EXTENSION` or `-EXTENSION` to the format
348 name. See [Extensions] below, for a list of extensions and their
349 names. See `--list-output-formats` and `--list-extensions`, below.
351`-o` *FILE*, `--output=`*FILE*
353: Write output to *FILE* instead of *stdout*. If *FILE* is
354 `-`, output will go to *stdout*, even if a non-textual format
355 (`docx`, `odt`, `epub2`, `epub3`) is specified.
357`--data-dir=`*DIRECTORY*
359: Specify the user data directory to search for pandoc data files.
360 If this option is not specified, the default user data directory
361 will be used. On \*nix and macOS systems this will be the `pandoc`
362 subdirectory of the XDG data directory (by default,
363 `$HOME/.local/share`, overridable by setting the `XDG_DATA_HOME`
364 environment variable). If that directory does not exist and
365 `$HOME/.pandoc` exists, it will be used (for backwards compatibility).
366 On Windows the default user data directory is
367 `C:\Users\USERNAME\AppData\Roaming\pandoc`.
368 You can find the default user data directory on your system by
369 looking at the output of `pandoc --version`.
370 Data files placed in this directory (for example, `reference.odt`,
371 `reference.docx`, `epub.css`, `templates`) will override
372 pandoc's normal defaults.
374`-d` *FILE*, `--defaults=`*FILE*
376: Specify a set of default option settings. *FILE* is a YAML
377 file whose fields correspond to command-line option
378 settings. All options for document conversion, including input
379 and output files, can be set using a defaults file. The file will
380 be searched for first in the working directory, and then in
381 the `defaults` subdirectory of the user data directory
382 (see `--data-dir`). The `.yaml` extension may be omitted.
383 See the section [Default files] for more information on the
384 file format. Settings from the defaults file may be
385 overridden or extended by subsequent options on the command
390: Generate a bash completion script. To enable bash completion
391 with pandoc, add this to your `.bashrc`:
393 eval "$(pandoc --bash-completion)"
397: Give verbose debugging output.
401: Suppress warning messages.
405: Exit with error status if there are any warnings.
409: Write log messages in machine-readable JSON format to
410 *FILE*. All messages above DEBUG level will be written,
411 regardless of verbosity settings (`--verbose`, `--quiet`).
413`--list-input-formats`
415: List supported input formats, one per line.
417`--list-output-formats`
419: List supported output formats, one per line.
421`--list-extensions`[`=`*FORMAT*]
423: List supported extensions for *FORMAT*, one per line, preceded
424 by a `+` or `-` indicating whether it is enabled by default
425 in *FORMAT*. If *FORMAT* is not specified, defaults for
426 pandoc's Markdown are given.
428`--list-highlight-languages`
430: List supported languages for syntax highlighting, one per
433`--list-highlight-styles`
435: List supported styles for syntax highlighting, one per line.
436 See `--highlight-style`.
446[Markdown]: https://daringfireball.net/projects/markdown/
447[CommonMark]: https://commonmark.org
448[PHP Markdown Extra]: https://michelf.ca/projects/php-markdown/extra/
449[GitHub-Flavored Markdown]: https://help.github.com/articles/github-flavored-markdown/
450[MultiMarkdown]: https://fletcherpenney.net/multimarkdown/
451[reStructuredText]: https://docutils.sourceforge.io/docs/ref/rst/introduction.html
452[S5]: https://meyerweb.com/eric/tools/s5/
453[Slidy]: https://www.w3.org/Talks/Tools/Slidy2/
454[Slideous]: https://goessner.net/articles/slideous/
455[HTML]: https://www.w3.org/html/
456[HTML5]: https://html.spec.whatwg.org/
457[polyglot markup]: https://www.w3.org/TR/html-polyglot/
458[XHTML]: https://www.w3.org/TR/xhtml1/
459[LaTeX]: https://www.latex-project.org/
460[`beamer`]: https://ctan.org/pkg/beamer
461[Beamer User's Guide]: http://mirrors.ctan.org/macros/latex/contrib/beamer/doc/beameruserguide.pdf
462[ConTeXt]: https://www.contextgarden.net/
463[Rich Text Format]: https://en.wikipedia.org/wiki/Rich_Text_Format
464[DocBook]: https://docbook.org
465[JATS]: https://jats.nlm.nih.gov
466[Jira]: https://jira.atlassian.com/secure/WikiRendererHelpAction.jspa?section=all
467[txt2tags]: https://txt2tags.org
468[EPUB]: http://idpf.org/epub
469[OPML]: http://dev.opml.org/spec2.html
470[OpenDocument]: http://opendocument.xml.org
471[ODT]: https://en.wikipedia.org/wiki/OpenDocument
472[Textile]: https://www.promptworks.com/textile
473[MediaWiki markup]: https://www.mediawiki.org/wiki/Help:Formatting
474[DokuWiki markup]: https://www.dokuwiki.org/dokuwiki
475[ZimWiki markup]: https://zim-wiki.org/manual/Help/Wiki_Syntax.html
476[XWiki markup]: https://www.xwiki.org/xwiki/bin/view/Documentation/UserGuide/Features/XWikiSyntax/
477[TWiki markup]: https://twiki.org/cgi-bin/view/TWiki/TextFormattingRules
478[TikiWiki markup]: https://doc.tiki.org/Wiki-Syntax-Text#The_Markup_Language_Wiki-Syntax
479[Haddock markup]: https://www.haskell.org/haddock/doc/html/ch03s08.html
480[Creole 1.0]: http://www.wikicreole.org/wiki/Creole1.0
481[CSV]: https://tools.ietf.org/html/rfc4180
482[roff man]: https://man.cx/groff_man(7)
483[roff ms]: https://man.cx/groff_ms(7)
484[Haskell]: https://www.haskell.org
485[GNU Texinfo]: https://www.gnu.org/software/texinfo/
486[Emacs Org mode]: https://orgmode.org
487[AsciiDoc]: https://www.methods.co.nz/asciidoc/
488[AsciiDoctor]: https://asciidoctor.org/
489[DZSlides]: https://paulrouget.com/dzslides/
490[Word docx]: https://en.wikipedia.org/wiki/Office_Open_XML
491[PDF]: https://www.adobe.com/pdf/
492[reveal.js]: https://revealjs.com/
493[FictionBook2]: http://www.fictionbook.org/index.php/Eng:XML_Schema_Fictionbook_2.1
494[Jupyter notebook]: https://nbformat.readthedocs.io/en/latest/
495[InDesign ICML]: https://wwwimages.adobe.com/www.adobe.com/content/dam/acom/en/devnet/indesign/sdk/cs6/idml/idml-cookbook.pdf
496[TEI Simple]: https://github.com/TEIC/TEI-Simple
497[Muse]: https://amusewiki.org/library/manual
498[PowerPoint]: https://en.wikipedia.org/wiki/Microsoft_PowerPoint
499[Vimwiki]: https://vimwiki.github.io
500[CSL JSON]: https://citeproc-js.readthedocs.io/en/latest/csl-json/markup.html
501[BibTeX]: https://ctan.org/pkg/bibtex
502[BibLaTeX]: https://ctan.org/pkg/biblatex
504## Reader options {.options}
506`--shift-heading-level-by=`*NUMBER*
508: Shift heading levels by a positive or negative integer.
509 For example, with `--shift-heading-level-by=-1`, level 2
510 headings become level 1 headings, and level 3 headings
511 become level 2 headings. Headings cannot have a level
512 less than 1, so a heading that would be shifted below level 1
513 becomes a regular paragraph. Exception: with a shift of -N,
514 a level-N heading at the beginning of the document
515 replaces the metadata title. `--shift-heading-level-by=-1`
516 is a good choice when converting HTML or Markdown documents that
517 use an initial level-1 heading for the document title and
518 level-2+ headings for sections. `--shift-heading-level-by=1`
519 may be a good choice for converting Markdown documents that
520 use level-1 headings for sections to HTML, since pandoc uses
521 a level-1 heading to render the document title.
523`--base-header-level=`*NUMBER*
525: *Deprecated. Use `--shift-heading-level-by`=X instead,
526 where X = NUMBER - 1.* Specify the base level for headings
529`--strip-empty-paragraphs`
531: *Deprecated. Use the `+empty_paragraphs` extension instead.*
532 Ignore paragraphs with no content. This option is useful
533 for converting word processing documents where users have
534 used empty paragraphs to create inter-paragraph space.
536`--indented-code-classes=`*CLASSES*
538: Specify classes to use for indented code blocks--for example,
539 `perl,numberLines` or `haskell`. Multiple classes may be separated
542`--default-image-extension=`*EXTENSION*
544: Specify a default extension to use when image paths/URLs have no
545 extension. This allows you to use the same source for formats that
546 require different kinds of images. Currently this option only affects
547 the Markdown and LaTeX readers.
551: Parse each file individually before combining for multifile
552 documents. This will allow footnotes in different files with the
553 same identifiers to work as expected. If this option is set,
554 footnotes and links will not work across files. Reading binary
555 files (docx, odt, epub) implies `--file-scope`.
557`-F` *PROGRAM*, `--filter=`*PROGRAM*
559: Specify an executable to be used as a filter transforming the
560 pandoc AST after the input is parsed and before the output is
561 written. The executable should read JSON from stdin and write
562 JSON to stdout. The JSON must be formatted like pandoc's own
563 JSON input and output. The name of the output format will be
564 passed to the filter as the first argument. Hence,
566 pandoc --filter ./caps.py -t latex
570 pandoc -t json | ./caps.py latex | pandoc -f json -t latex
572 The latter form may be useful for debugging filters.
574 Filters may be written in any language. `Text.Pandoc.JSON`
575 exports `toJSONFilter` to facilitate writing filters in Haskell.
576 Those who would prefer to write filters in python can use the
577 module [`pandocfilters`], installable from PyPI. There are also
578 pandoc filter libraries in [PHP], [perl], and
579 [JavaScript/node.js].
581 In order of preference, pandoc will look for filters in
583 1. a specified full or relative path (executable or
586 2. `$DATADIR/filters` (executable or non-executable)
587 where `$DATADIR` is the user data directory (see
588 `--data-dir`, above).
590 3. `$PATH` (executable only)
592 Filters, Lua-filters, and citeproc processing are applied in
593 the order specified on the command line.
595`-L` *SCRIPT*, `--lua-filter=`*SCRIPT*
597: Transform the document in a similar fashion as JSON filters (see
598 `--filter`), but use pandoc's built-in Lua filtering system. The given
599 Lua script is expected to return a list of Lua filters which will be
600 applied in order. Each Lua filter must contain element-transforming
601 functions indexed by the name of the AST element on which the filter
602 function should be applied.
604 The `pandoc` Lua module provides helper functions for element
605 creation. It is always loaded into the script's Lua environment.
607 See the [Lua filters documentation] for further details.
609 In order of preference, pandoc will look for Lua filters in
611 1. a specified full or relative path
613 2. `$DATADIR/filters` where `$DATADIR` is the user data
614 directory (see `--data-dir`, above).
616 Filters, Lua filters, and citeproc processing are applied in
617 the order specified on the command line.
619`-M` *KEY*[`=`*VAL*], `--metadata=`*KEY*[`:`*VAL*]
621: Set the metadata field *KEY* to the value *VAL*. A value specified
622 on the command line overrides a value specified in the document
623 using [YAML metadata blocks][Extension: `yaml_metadata_block`].
624 Values will be parsed as YAML boolean or string values. If no value is
625 specified, the value will be treated as Boolean true. Like
626 `--variable`, `--metadata` causes template variables to be set.
627 But unlike `--variable`, `--metadata` affects the metadata of the
628 underlying document (which is accessible from filters and may be
629 printed in some output formats) and metadata values will be escaped
630 when inserted into the template.
632`--metadata-file=`*FILE*
634: Read metadata from the supplied YAML (or JSON) file. This
635 option can be used with every input format, but string scalars
636 in the YAML file will always be parsed as Markdown. Generally,
637 the input will be handled the same as in [YAML metadata
638 blocks][Extension: `yaml_metadata_block`]. This option can be
639 used repeatedly to include multiple metadata files; values in
640 files specified later on the command line will be preferred
641 over those specified in earlier files. Metadata values
642 specified inside the document, or by using `-M`, overwrite
643 values specified with this option.
645`-p`, `--preserve-tabs`
647: Preserve tabs instead of converting them to spaces. (By default, pandoc
648 converts tabs to spaces before parsing its input.) Note that this will
649 only affect tabs in literal code spans and code blocks. Tabs in regular
650 text are always treated as spaces.
654: Specify the number of spaces per tab (default is 4).
656`--track-changes=accept`|`reject`|`all`
658: Specifies what to do with insertions, deletions, and comments
659 produced by the MS Word "Track Changes" feature. `accept` (the
660 default) processes all the insertions and deletions.
661 `reject` ignores them. Both `accept` and `reject` ignore comments.
662 `all` includes all insertions, deletions, and comments, wrapped
663 in spans with `insertion`, `deletion`, `comment-start`, and
664 `comment-end` classes, respectively. The author and time of
665 change is included. `all` is useful for scripting: only
666 accepting changes from a certain reviewer, say, or before a
667 certain date. If a paragraph is inserted or deleted,
668 `track-changes=all` produces a span with the class
669 `paragraph-insertion`/`paragraph-deletion` before the
670 affected paragraph break. This option only affects the docx
673`--extract-media=`*DIR*
675: Extract images and other media contained in or linked from
676 the source document to the path *DIR*, creating it if
677 necessary, and adjust the images references in the document
678 so they point to the extracted files. If the source format is
679 a binary container (docx, epub, or odt), the media is
680 extracted from the container and the original
681 filenames are used. Otherwise the media is read from the
682 file system or downloaded, and new filenames are constructed
683 based on SHA1 hashes of the contents.
685`--abbreviations=`*FILE*
687: Specifies a custom abbreviations file, with abbreviations
688 one to a line. If this option is not specified, pandoc will
689 read the data file `abbreviations` from the user data
690 directory or fall back on a system default. To see the
692 `pandoc --print-default-data-file=abbreviations`. The only
693 use pandoc makes of this list is in the Markdown reader.
694 Strings found in this list will be followed by a nonbreaking
695 space, and the period will not produce sentence-ending space
696 in formats like LaTeX. The strings may not contain spaces.
698[`pandocfilters`]: https://github.com/jgm/pandocfilters
699[PHP]: https://github.com/vinai/pandocfilters-php
700[perl]: https://metacpan.org/pod/Pandoc::Filter
701[JavaScript/node.js]: https://github.com/mvhenderson/pandoc-filter-node
702[Lua filters documentation]: https://pandoc.org/lua-filters.html
704## General writer options {.options}
708: Produce output with an appropriate header and footer (e.g. a
709 standalone HTML, LaTeX, TEI, or RTF file, not a fragment). This option
710 is set automatically for `pdf`, `epub`, `epub3`, `fb2`, `docx`, and `odt`
711 output. For `native` output, this option causes metadata to
712 be included; otherwise, metadata is suppressed.
714`--template=`*FILE*|*URL*
716: Use the specified file as a custom template for the generated document.
717 Implies `--standalone`. See [Templates], below, for a description
718 of template syntax. If no extension is specified, an extension
719 corresponding to the writer will be added, so that `--template=special`
720 looks for `special.html` for HTML output. If the template is not
721 found, pandoc will search for it in the `templates` subdirectory of
722 the user data directory (see `--data-dir`). If this option is not used,
723 a default template appropriate for the output format will be used (see
724 `-D/--print-default-template`).
726`-V` *KEY*[`=`*VAL*], `--variable=`*KEY*[`:`*VAL*]
728: Set the template variable *KEY* to the value *VAL* when rendering the
729 document in standalone mode. If no *VAL* is specified, the
730 key will be given the value `true`.
732`-D` *FORMAT*, `--print-default-template=`*FORMAT*
734: Print the system default template for an output *FORMAT*. (See `-t`
735 for a list of possible *FORMAT*s.) Templates in the user data
736 directory are ignored. This option may be used with
737 `-o`/`--output` to redirect output to a file, but
738 `-o`/`--output` must come before `--print-default-template`
741 Note that some of the default templates use partials, for
742 example `styles.html`. To print the partials, use
743 `--print-default-data-file`: for example,
744 `--print-default-data-file=templates/styles.html`.
746`--print-default-data-file=`*FILE*
748: Print a system default data file. Files in the user data directory
749 are ignored. This option may be used with `-o`/`--output` to
750 redirect output to a file, but `-o`/`--output` must come before
751 `--print-default-data-file` on the command line.
753`--eol=crlf`|`lf`|`native`
755: Manually specify line endings: `crlf` (Windows), `lf`
756 (macOS/Linux/UNIX), or `native` (line endings appropriate
757 to the OS on which pandoc is being run). The default is
762: Specify the default dpi (dots per inch) value for conversion
763 from pixels to inch/centimeters and vice versa. (Technically,
764 the correct term would be ppi: pixels per inch.) The default
765 is 96dpi. When images contain information about dpi
766 internally, the encoded value is used instead of the default
767 specified by this option.
769`--wrap=auto`|`none`|`preserve`
771: Determine how text is wrapped in the output (the source
772 code, not the rendered version). With `auto` (the default),
773 pandoc will attempt to wrap lines to the column width specified by
774 `--columns` (default 72). With `none`, pandoc will not wrap
775 lines at all. With `preserve`, pandoc will attempt to
776 preserve the wrapping from the source document (that is,
777 where there are nonsemantic newlines in the source, there
778 will be nonsemantic newlines in the output as well).
779 Automatic wrapping does not currently work in HTML output.
780 In `ipynb` output, this option affects wrapping of the
781 contents of markdown cells.
785: Specify length of lines in characters. This affects text wrapping
786 in the generated source code (see `--wrap`). It also affects
787 calculation of column widths for plain text tables (see [Tables] below).
789`--toc`, `--table-of-contents`
791: Include an automatically generated table of contents (or, in
792 the case of `latex`, `context`, `docx`, `odt`,
793 `opendocument`, `rst`, or `ms`, an instruction to create
794 one) in the output document. This option has no effect
795 unless `-s/--standalone` is used, and it has no effect
796 on `man`, `docbook4`, `docbook5`, or `jats` output.
798 Note that if you are producing a PDF via `ms`, the table
799 of contents will appear at the beginning of the
800 document, before the title. If you would prefer it to
801 be at the end of the document, use the option
802 `--pdf-engine-opt=--no-toc-relocation`.
804`--toc-depth=`*NUMBER*
806: Specify the number of section levels to include in the table
807 of contents. The default is 3 (which means that level-1, 2, and 3
808 headings will be listed in the contents).
812: Strip out HTML comments in the Markdown or Textile source,
813 rather than passing them on to Markdown, Textile or HTML
814 output as raw HTML. This does not apply to HTML comments
815 inside raw HTML blocks when the `markdown_in_html_blocks`
816 extension is not set.
820: Disables syntax highlighting for code blocks and inlines, even when
821 a language attribute is given.
823`--highlight-style=`*STYLE*|*FILE*
825: Specifies the coloring style to be used in highlighted source code.
826 Options are `pygments` (the default), `kate`, `monochrome`,
827 `breezeDark`, `espresso`, `zenburn`, `haddock`, and `tango`.
828 For more information on syntax highlighting in pandoc, see
829 [Syntax highlighting], below. See also
830 `--list-highlight-styles`.
832 Instead of a *STYLE* name, a JSON file with extension
833 `.theme` may be supplied. This will be parsed as a KDE
834 syntax highlighting theme and (if valid) used as the
837 To generate the JSON version of an existing style,
838 use `--print-highlight-style`.
840`--highlight-style-dark=`*STYLE*|*FILE*
842: Specifies the coloring style to be used in highlighted source code
843 in [dark mode] for HTML-based output formats. The default is `breezeDark`.
844 See `--highlight-style` for more information.
846`--print-highlight-style=`*STYLE*|*FILE*
848: Prints a JSON version of a highlighting style, which can
849 be modified, saved with a `.theme` extension, and used
850 with `--highlight-style`. This option may be used with
851 `-o`/`--output` to redirect output to a file, but
852 `-o`/`--output` must come before `--print-highlight-style`
855`--syntax-definition=`*FILE*
857: Instructs pandoc to load a KDE XML syntax definition file,
858 which will be used for syntax highlighting of appropriately
859 marked code blocks. This can be used to add support for
860 new languages or to use altered syntax definitions for
861 existing languages. This option may be repeated to add
862 multiple syntax definitions.
864`-H` *FILE*, `--include-in-header=`*FILE*|*URL*
866: Include contents of *FILE*, verbatim, at the end of the header.
867 This can be used, for example, to include special
868 CSS or JavaScript in HTML documents. This option can be used
869 repeatedly to include multiple files in the header. They will be
870 included in the order specified. Implies `--standalone`.
872`-B` *FILE*, `--include-before-body=`*FILE*|*URL*
874: Include contents of *FILE*, verbatim, at the beginning of the
875 document body (e.g. after the `<body>` tag in HTML, or the
876 `\begin{document}` command in LaTeX). This can be used to include
877 navigation bars or banners in HTML documents. This option can be
878 used repeatedly to include multiple files. They will be included in
879 the order specified. Implies `--standalone`.
881`-A` *FILE*, `--include-after-body=`*FILE*|*URL*
883: Include contents of *FILE*, verbatim, at the end of the document
884 body (before the `</body>` tag in HTML, or the
885 `\end{document}` command in LaTeX). This option can be used
886 repeatedly to include multiple files. They will be included in the
887 order specified. Implies `--standalone`.
889`--resource-path=`*SEARCHPATH*
891: List of paths to search for images and other resources.
892 The paths should be separated by `:` on Linux, UNIX, and
893 macOS systems, and by `;` on Windows. If `--resource-path`
894 is not specified, the default resource path is the working
895 directory. Note that, if `--resource-path` is specified,
896 the working directory must be explicitly listed or it
897 will not be searched. For example:
898 `--resource-path=.:test` will search the working directory
899 and the `test` subdirectory, in that order.
900 This option can be used repeatedly. Search path components
901 that come later on the command line will be searched before
902 those that come earlier, so
903 `--resource-path foo:bar --resource-path baz:bim` is
904 equivalent to `--resource-path baz:bim:foo:bar`.
906`--request-header=`*NAME*`:`*VAL*
908: Set the request header *NAME* to the value *VAL* when making
909 HTTP requests (for example, when a URL is given on the
910 command line, or when resources used in a document must be
911 downloaded). If you're behind a proxy, you also need to set
912 the environment variable `http_proxy` to `http://...`.
914`--no-check-certificate`
916: Disable the certificate verification to allow access to
917 unsecure HTTP resources (for example when the certificate
918 is no longer valid or self signed).
920## Options affecting specific writers {.options}
924: Produce a standalone HTML file with no external dependencies, using
925 `data:` URIs to incorporate the contents of linked scripts, stylesheets,
926 images, and videos. Implies `--standalone`. The resulting file should be
927 "self-contained," in the sense that it needs no external files and no net
928 access to be displayed properly by a browser. This option works only with
929 HTML output formats, including `html4`, `html5`, `html+lhs`, `html5+lhs`,
930 `s5`, `slidy`, `slideous`, `dzslides`, and `revealjs`. Scripts, images,
931 and stylesheets at absolute URLs will be downloaded; those at relative
932 URLs will be sought relative to the working directory (if the first source
933 file is local) or relative to the base URL (if the first source
934 file is remote). Elements with the attribute
935 `data-external="1"` will be left alone; the documents they
936 link to will not be incorporated in the document.
937 Limitation: resources that are loaded dynamically through
938 JavaScript cannot be incorporated; as a result,
939 `--self-contained` does not work with `--mathjax`, and some
940 advanced features (e.g. zoom or speaker notes) may not work
941 in an offline "self-contained" `reveal.js` slide show.
945: Use `<q>` tags for quotes in HTML. (This option only has an
946 effect if the `smart` extension is enabled for the input
951: Use only ASCII characters in output. Currently supported for XML
952 and HTML formats (which use entities instead of UTF-8 when this
953 option is selected), CommonMark, gfm, and Markdown (which use
954 entities), roff ms (which use hexadecimal escapes), and to a
955 limited degree LaTeX (which uses standard commands for accented
956 characters when possible). roff man output uses ASCII by default.
960: Use reference-style links, rather than inline links, in writing Markdown
961 or reStructuredText. By default inline links are used. The
962 placement of link references is affected by the
963 `--reference-location` option.
965`--reference-location=block`|`section`|`document`
967: Specify whether footnotes (and references, if `reference-links` is
968 set) are placed at the end of the current (top-level) block, the
969 current section, or the document. The default is
970 `document`. Currently only affects the markdown writer.
972`--markdown-headings=setext`|`atx`
974: Specify whether to use ATX-style (`#`-prefixed) or
975 Setext-style (underlined) headings for level 1 and 2
976 headings in Markdown output. (The default is `atx`.)
977 ATX-style headings are always used for levels 3+.
978 This option also affects Markdown cells in `ipynb` output.
982: *Deprecated synonym for `--markdown-headings=atx`.*
984`--top-level-division=default`|`section`|`chapter`|`part`
986: Treat top-level headings as the given division type in
987 LaTeX, ConTeXt, DocBook, and TEI output. The hierarchy
988 order is part, chapter, then section; all headings are
989 shifted such that the top-level heading becomes the
990 specified type. The default behavior is to determine the
991 best division type via heuristics: unless other conditions
992 apply, `section` is chosen. When the `documentclass`
993 variable is set to `report`, `book`, or `memoir` (unless the
994 `article` option is specified), `chapter` is implied as the
995 setting for this option. If `beamer` is the output format,
996 specifying either `chapter` or `part` will cause top-level
997 headings to become `\part{..}`, while second-level headings
998 remain as their default type.
1000`-N`, `--number-sections`
1002: Number section headings in LaTeX, ConTeXt, HTML, Docx, ms, or EPUB
1003 output. By default, sections are not numbered. Sections with class
1004 `unnumbered` will never be numbered, even if `--number-sections`
1007`--number-offset=`*NUMBER*[`,`*NUMBER*`,`*...*]
1009: Offset for section headings in HTML output (ignored in other
1010 output formats). The first number is added to the section number for
1011 top-level headings, the second for second-level headings, and so on.
1012 So, for example, if you want the first top-level heading in your
1013 document to be numbered "6", specify `--number-offset=5`.
1014 If your document starts with a level-2 heading which you want to
1015 be numbered "1.5", specify `--number-offset=1,4`.
1016 Offsets are 0 by default. Implies `--number-sections`.
1020: Use the [`listings`] package for LaTeX code blocks. The package
1021 does not support multi-byte encoding for source code. To handle UTF-8
1022 you would need to use a custom template. This issue is fully
1023 documented here: [Encoding issue with the listings package].
1025`-i`, `--incremental`
1027: Make list items in slide shows display incrementally (one by one).
1028 The default is for lists to be displayed all at once.
1030`--slide-level=`*NUMBER*
1032: Specifies that headings with the specified level create
1033 slides (for `beamer`, `s5`, `slidy`, `slideous`, `dzslides`). Headings
1034 above this level in the hierarchy are used to divide the
1035 slide show into sections; headings below this level create
1036 subheads within a slide. Note that content that is
1037 not contained under slide-level headings will not appear in
1038 the slide show. The default is to set the slide level based
1039 on the contents of the document; see [Structuring the slide
1044: Wrap sections in `<section>` tags (or `<div>` tags for `html4`),
1045 and attach identifiers to the enclosing `<section>` (or `<div>`)
1046 rather than the heading itself. See
1047 [Heading identifiers], below.
1049`--email-obfuscation=none`|`javascript`|`references`
1051: Specify a method for obfuscating `mailto:` links in HTML documents.
1052 `none` leaves `mailto:` links as they are. `javascript` obfuscates
1053 them using JavaScript. `references` obfuscates them by printing their
1054 letters as decimal or hexadecimal character references. The default
1057`--id-prefix=`*STRING*
1059: Specify a prefix to be added to all identifiers and internal links
1060 in HTML and DocBook output, and to footnote numbers in Markdown
1061 and Haddock output. This is useful for preventing duplicate
1062 identifiers when generating fragments to be included in other pages.
1064`-T` *STRING*, `--title-prefix=`*STRING*
1066: Specify *STRING* as a prefix at the beginning of the title
1067 that appears in the HTML header (but not in the title as it
1068 appears at the beginning of the HTML body). Implies `--standalone`.
1070`-c` *URL*, `--css=`*URL*
1072: Link to a CSS style sheet. This option can be used repeatedly to
1073 include multiple files. They will be included in the order specified.
1075 A stylesheet is required for generating EPUB. If none is
1076 provided using this option (or the `css` or `stylesheet`
1077 metadata fields), pandoc will look for a file `epub.css` in the
1078 user data directory (see `--data-dir`). If it is not
1079 found there, sensible defaults will be used.
1081`--reference-doc=`*FILE*
1083: Use the specified file as a style reference in producing a
1088 : For best results, the reference docx should be a modified
1089 version of a docx file produced using pandoc. The contents
1090 of the reference docx are ignored, but its stylesheets and
1091 document properties (including margins, page size, header,
1092 and footer) are used in the new docx. If no reference docx
1093 is specified on the command line, pandoc will look for a
1094 file `reference.docx` in the user data directory (see
1095 `--data-dir`). If this is not found either, sensible
1096 defaults will be used.
1098 To produce a custom `reference.docx`, first get a copy of
1099 the default `reference.docx`: `pandoc
1100 -o custom-reference.docx --print-default-data-file reference.docx`.
1101 Then open `custom-reference.docx` in Word, modify the
1102 styles as you wish, and save the file. For best
1103 results, do not make changes to this file other than
1104 modifying the styles used by pandoc:
1140 - Default Paragraph Font
1143 - Footnote Reference
1153 : For best results, the reference ODT should be a modified
1154 version of an ODT produced using pandoc. The contents of
1155 the reference ODT are ignored, but its stylesheets are used
1156 in the new ODT. If no reference ODT is specified on the
1157 command line, pandoc will look for a file `reference.odt` in
1158 the user data directory (see `--data-dir`). If this is not
1159 found either, sensible defaults will be used.
1161 To produce a custom `reference.odt`, first get a copy of
1162 the default `reference.odt`: `pandoc
1163 -o custom-reference.odt --print-default-data-file reference.odt`.
1164 Then open `custom-reference.odt` in LibreOffice, modify
1165 the styles as you wish, and save the file.
1169 : Templates included with Microsoft PowerPoint 2013 (either with
1170 `.pptx` or `.potx` extension) are known to work, as are most
1171 templates derived from these.
1173 The specific requirement is that the template should begin with
1174 the following first four layouts:
1177 2. Title and Content
1181 All templates included with a recent version of MS PowerPoint
1182 will fit these criteria. (You can click on `Layout` under the
1183 `Home` menu to check.)
1185 You can also modify the default `reference.pptx`: first run
1186 `pandoc -o custom-reference.pptx --print-default-data-file
1187 reference.pptx`, and then modify `custom-reference.pptx`
1188 in MS PowerPoint (pandoc will use the first four layout
1189 slides, as mentioned above).
1191`--epub-cover-image=`*FILE*
1193: Use the specified image as the EPUB cover. It is recommended
1194 that the image be less than 1000px in width and height. Note that
1195 in a Markdown source document you can also specify `cover-image`
1196 in a YAML metadata block (see [EPUB Metadata], below).
1198`--epub-metadata=`*FILE*
1200: Look in the specified XML file for metadata for the EPUB.
1201 The file should contain a series of [Dublin Core elements].
1204 <dc:rights>Creative Commons</dc:rights>
1205 <dc:language>es-AR</dc:language>
1207 By default, pandoc will include the following metadata elements:
1208 `<dc:title>` (from the document title), `<dc:creator>` (from the
1209 document authors), `<dc:date>` (from the document date, which should
1210 be in [ISO 8601 format]), `<dc:language>` (from the `lang`
1211 variable, or, if is not set, the locale), and `<dc:identifier
1212 id="BookId">` (a randomly generated UUID). Any of these may be
1213 overridden by elements in the metadata file.
1215 Note: if the source document is Markdown, a YAML metadata block
1216 in the document can be used instead. See below under
1219`--epub-embed-font=`*FILE*
1221: Embed the specified font in the EPUB. This option can be repeated
1222 to embed multiple fonts. Wildcards can also be used: for example,
1223 `DejaVuSans-*.ttf`. However, if you use wildcards on the command
1224 line, be sure to escape them or put the whole filename in single quotes,
1225 to prevent them from being interpreted by the shell. To use the
1226 embedded fonts, you will need to add declarations like the following
1227 to your CSS (see `--css`):
1230 font-family: DejaVuSans;
1232 font-weight: normal;
1233 src:url("DejaVuSans-Regular.ttf");
1236 font-family: DejaVuSans;
1239 src:url("DejaVuSans-Bold.ttf");
1242 font-family: DejaVuSans;
1244 font-weight: normal;
1245 src:url("DejaVuSans-Oblique.ttf");
1248 font-family: DejaVuSans;
1251 src:url("DejaVuSans-BoldOblique.ttf");
1253 body { font-family: "DejaVuSans"; }
1255`--epub-chapter-level=`*NUMBER*
1257: Specify the heading level at which to split the EPUB into separate
1258 "chapter" files. The default is to split into chapters at level-1
1259 headings. This option only affects the internal composition of the
1260 EPUB, not the way chapters and sections are displayed to users. Some
1261 readers may be slow if the chapter files are too large, so for large
1262 documents with few level-1 headings, one might want to use a chapter
1265`--epub-subdirectory=`*DIRNAME*
1267: Specify the subdirectory in the OCF container that is to hold
1268 the EPUB-specific contents. The default is `EPUB`. To put
1269 the EPUB contents in the top level, use an empty string.
1271`--ipynb-output=all|none|best`
1273: Determines how ipynb output cells are treated. `all` means
1274 that all of the data formats included in the original are
1275 preserved. `none` means that the contents of data cells
1276 are omitted. `best` causes pandoc to try to pick the
1277 richest data block in each output cell that is compatible
1278 with the output format. The default is `best`.
1280`--pdf-engine=`*PROGRAM*
1282: Use the specified engine when producing PDF output.
1283 Valid values are `pdflatex`, `lualatex`, `xelatex`, `latexmk`,
1284 `tectonic`, `wkhtmltopdf`, `weasyprint`, `prince`, `context`,
1285 and `pdfroff`. If the engine is not in your PATH, the full
1286 path of the engine may be specified here. If this option
1287 is not specified, pandoc uses the following defaults
1288 depending on the output format specified using `-t/--to`:
1290 - `-t latex` or none: `pdflatex` (other options: `xelatex`, `lualatex`,
1291 `tectonic`, `latexmk`)
1292 - `-t context`: `context`
1293 - `-t html`: `wkhtmltopdf` (other options: `prince`, `weasyprint`;
1294 see [print-css.rocks](https://print-css.rocks) for a good
1295 introduction to PDF generation from HTML/CSS.)
1296 - `-t ms`: `pdfroff`
1298`--pdf-engine-opt=`*STRING*
1300: Use the given string as a command-line argument to the `pdf-engine`.
1301 For example, to use a persistent directory `foo` for `latexmk`'s
1302 auxiliary files, use `--pdf-engine-opt=-outdir=foo`.
1303 Note that no check for duplicate options is done.
1305[Dublin Core elements]: https://www.dublincore.org/specifications/dublin-core/dces/
1306[ISO 8601 format]: https://www.w3.org/TR/NOTE-datetime
1307[Encoding issue with the listings package]:
1308 https://en.wikibooks.org/wiki/LaTeX/Source_Code_Listings#Encoding_issue
1310## Citation rendering {.options}
1314: Process the citations in the file, replacing them with
1315 rendered citations and adding a bibliography.
1316 Citation processing will not take place unless bibliographic
1317 data is supplied, either through an external file specified
1318 using the `--bibliography` option or the `bibliography`
1319 field in metadata, or via a `references` section in metadata
1320 containing a list of citations in CSL YAML format with
1321 Markdown formatting. The style is controlled by a [CSL]
1322 stylesheet specified using the `--csl` option or the `csl`
1323 field in metadata. (If no stylesheet is specified,
1324 the `chicago-author-date` style will be used by default.)
1325 The citation processing transformation may be applied before
1326 or after filters or Lua filters (see `--filter`,
1327 `--lua-filter`): these transformations are applied in the
1328 order they appear on the command line. For more
1329 information, see the section on [Citations].
1331`--bibliography=`*FILE*
1333: Set the `bibliography` field in the document's metadata to *FILE*,
1334 overriding any value set in the metadata. If you supply
1335 this argument multiple times, each *FILE* will be added to
1336 bibliography. If *FILE* is a URL, it will be fetched
1337 via HTTP. If *FILE* is not found relative to the
1338 working directory, it will be sought in the resource path
1339 (see `--resource-path`).
1343: Set the `csl` field in the document's metadata to *FILE*,
1344 overriding any value set in the metadata. (This is equivalent to
1345 `--metadata csl=FILE`.) If *FILE* is a URL, it will be
1346 fetched via HTTP. If *FILE* is not found relative to the
1347 working directory, it will be sought in the resource path
1348 (see `--resource-path`) and finally in the `csl`
1349 subdirectory of the pandoc user data directory.
1351`--citation-abbreviations=`*FILE*
1353: Set the `citation-abbreviations` field in the document's metadata to
1354 *FILE*, overriding any value set in the metadata. (This is equivalent to
1355 `--metadata citation-abbreviations=FILE`.)
1356 If *FILE* is a URL, it will be fetched via HTTP. If *FILE* is not
1357 found relative to the working directory, it will be sought
1358 in the resource path (see `--resource-path`) and finally in
1359 the `csl` subdirectory of the pandoc user data directory.
1363: Use [`natbib`] for citations in LaTeX output. This option
1364 is not for use with the `--citeproc` option or with PDF
1365 output. It is intended for use in producing a LaTeX file
1366 that can be processed with [`bibtex`].
1370: Use [`biblatex`] for citations in LaTeX output. This option
1371 is not for use with the `--citeproc` option or with PDF
1372 output. It is intended for use in producing a LaTeX file
1373 that can be processed with [`bibtex`] or [`biber`].
1375## Math rendering in HTML {.options}
1377The default is to render TeX math as far as possible using
1378Unicode characters. Formulas are put inside a `span` with
1379`class="math"`, so that they may be styled differently from the
1380surrounding text if needed. However, this gives acceptable
1381results only for basic math, usually you will want to use
1382`--mathjax` or another of the following options.
1384`--mathjax`[`=`*URL*]
1386: Use [MathJax] to display embedded TeX math in HTML output.
1387 TeX math will be put between `\(...\)` (for inline math)
1388 or `\[...\]` (for display math) and wrapped in `<span>` tags
1389 with class `math`. Then the MathJax JavaScript will render it.
1390 The *URL* should point to the `MathJax.js` load script.
1391 If a *URL* is not provided, a link to the Cloudflare CDN will
1396: Convert TeX math to [MathML] (in `epub3`, `docbook4`,
1397 `docbook5`, `jats`, `html4` and `html5`). This is the
1398 default in `odt` output. Note that currently only Firefox
1399 and Safari (and select e-book readers) natively support
1404: Convert TeX formulas to `<img>` tags that link to an external script
1405 that converts formulas to images. The formula will be URL-encoded
1406 and concatenated with the URL provided. For SVG images you can for
1407 example use `--webtex https://latex.codecogs.com/svg.latex?`.
1408 If no URL is specified, the CodeCogs URL generating PNGs
1409 will be used (`https://latex.codecogs.com/png.latex?`).
1410 Note: the `--webtex` option will affect Markdown output
1411 as well as HTML, which is useful if you're targeting a
1412 version of Markdown without native math support.
1416: Use [KaTeX] to display embedded TeX math in HTML output.
1417 The *URL* is the base URL for the KaTeX library. That directory
1418 should contain a `katex.min.js` and a `katex.min.css` file.
1419 If a *URL* is not provided, a link to the KaTeX CDN will be inserted.
1423: Enclose TeX math in `<eq>` tags in HTML output. The resulting HTML
1424 can then be processed by [GladTeX] to produce images of the typeset
1425 formulas and an HTML file with links to these images.
1426 So, the procedure is:
1428 pandoc -s --gladtex input.md -o myfile.htex
1429 gladtex -d myfile-images myfile.htex
1430 # produces myfile.html and images in myfile-images
1432[MathML]: https://www.w3.org/Math/
1433[MathJax]: https://www.mathjax.org
1434[KaTeX]: https://github.com/Khan/KaTeX
1435[GladTeX]: https://humenda.github.io/GladTeX/
1437## Options for wrapper scripts {.options}
1441: Print information about command-line arguments to *stdout*, then exit.
1442 This option is intended primarily for use in wrapper scripts.
1443 The first line of output contains the name of the output file specified
1444 with the `-o` option, or `-` (for *stdout*) if no output file was
1445 specified. The remaining lines contain the command-line arguments,
1446 one per line, in the order they appear. These do not include regular
1447 pandoc options and their arguments, but do include any options appearing
1448 after a `--` separator at the end of the line.
1452: Ignore command-line arguments (for use in wrapper scripts).
1453 Regular pandoc options are not ignored. Thus, for example,
1455 pandoc --ignore-args -o foo.html -s foo.txt -- -e latin1
1459 pandoc -o foo.html -s
1463If pandoc completes successfully, it will return exit code 0.
1464Nonzero exit codes have the following meanings:
1467----- ------------------------------------
1468 3 PandocFailOnWarningError
1470 5 PandocTemplateError
1472 21 PandocUnknownReaderError
1473 22 PandocUnknownWriterError
1474 23 PandocUnsupportedExtensionError
1475 24 PandocCiteprocError
1476 31 PandocEpubSubdirectoryError
1479 47 PandocPDFProgramNotFoundError
1481 62 PandocShouldNeverHappenError
1484 65 PandocParsecError
1485 66 PandocMakePDFError
1486 67 PandocSyntaxMapError
1487 83 PandocFilterError
1489 92 PandocUTF8DecodingError
1490 93 PandocIpynbDecodingError
1491 94 PandocUnsupportedCharsetError
1492 97 PandocCouldNotFindDataFileError
1493 99 PandocResourceNotFound
1494----- ------------------------------------
1498The `--defaults` option may be used to specify a package
1499of options. Here is a sample defaults file demonstrating all of
1500the fields that may be used:
1504# reader: may be used instead of from:
1506# writer: may be used instead of to:
1508# leave blank for output to stdout:
1510# leave blank for input from stdin, use [] for no input:
1514# or you may use input-file: with a single value
1516# Include options from the specified defaults files.
1517# The files will be searched for first in the working directory
1518# and then in the defaults subdirectory of the user data directory.
1519# The files are included in the same order in which they appear in
1520# the list. Options specified in this defaults file always have
1521# priority over the included ones.
1528self-contained: false
1530# note that structured variables may be specified:
1537# metadata values specified here are parsed as literal
1538# string text, not markdown:
1545# or you may use metadata-file: with a single value
1547# Note that these take files, not their contents:
1548include-before-body: []
1549include-after-body: []
1550include-in-header: []
1553# turn on built-in citation processing. Note that if you need
1554# control over when the citeproc processing is done relative
1555# to other filters, you should instead use `citeproc` in the
1556# list of `filters` (see below).
1562citation-abbreviations: abbrevs.json
1564# Filters will be assumed to be Lua filters if they have
1565# the .lua extension, and json filters otherwise. But
1566# the filter type can also be specified explicitly, as shown.
1567# Filters are run in the order specified.
1568# To include the built-in citeproc filter, use either `citeproc`
1569# or `{type: citeproc}`.
1579# ERROR, WARNING, or INFO
1583# citeproc, natbib, or biblatex. This only affects LaTeX
1584# output. If you want to use citeproc to format citations,
1585# you should also set 'citeproc: true' (see above).
1586cite-method: citeproc
1588# part, chapter, section, or default:
1589top-level-division: chapter
1595# you may also use pdf-engine-opt: with a single option
1596# pdf-engine-opt: "-shell-escape"
1598# auto, preserve, or none
1603extract-media: mediadir
1605table-of-contents: true
1607number-sections: false
1608# a list of offsets at each heading level
1609number-offset: [0,0,0,0,0,0]
1610# toc: may also be used instead of table-of-contents:
1611shift-heading-level-by: 1
1613identifier-prefix: foo
1615strip-empty-paragraphs: true
1616# lf, crlf, or native
1618strip-comments: false
1619indented-code-classes: []
1621default-image-extension: ".jpg"
1623# either a style name of a style definition file:
1624highlight-style: pygments
1627# or you may use syntax-definition: with a single value
1630reference-doc: myref.docx
1632# method is plain, webtex, gladtex, mathml, mathjax, katex
1633# you may specify a url with webtex, mathjax, katex
1636 url: "https://cdn.jsdelivr.net/npm/mathjax@3/es5/tex-mml-chtml.js"
1637# none, references, or javascript
1638email-obfuscation: javascript
1646epub-subdirectory: EPUB
1647epub-metadata: meta.xml
1650epub-chapter-level: 1
1651epub-cover-image: cover.jpg
1653reference-links: true
1654# block, section, or document
1655reference-location: block
1656markdown-headings: setext
1658# accept, reject, or all
1659track-changes: accept
1668# A list of two-element lists
1670- ["User-Agent", "Mozilla/5.0"]
1672fail-if-warnings: false
1678Fields that are omitted will just have their regular
1679default values. So a defaults file can be as simple as
1686In fields that expect a file path (or list of file paths), the
1687following syntax may be used to interpolate environment variables:
1690csl: ${HOME}/mycsldir/special.csl
1693`${USERDATA}` may also be used; this will always resolve to the
1694user data directory that is current when the defaults file is
1695parsed, regardless of the setting of the environment
1698`${.}` will resolve to the directory containing the default
1699file itself. This allows you to refer to resources contained
1703epub-cover-image: ${.}/cover.jpg
1704epub-metadata: ${.}/meta.xml
1706- . # the working directory from which pandoc is run
1707- ${.}/images # the images subdirectory of the directory
1708 # containing this defaults file
1711This environment variable interpolation syntax *only* works in
1712fields that expect file paths.
1714Default files can be placed in the `defaults` subdirectory of
1715the user data directory and used from any directory. For
1716example, one could create a file specifying defaults for writing
1717letters, save it as `letter.yaml` in the `defaults` subdirectory
1718of the user data directory, and then invoke these defaults
1719from any directory using `pandoc --defaults letter`
1720or `pandoc -dletter`.
1722When multiple defaults are used, their contents will be combined.
1724Note that, where command-line arguments may be repeated
1725(`--metadata-file`, `--css`, `--include-in-header`,
1726`--include-before-body`, `--include-after-body`, `--variable`,
1727`--metadata`, `--syntax-definition`), the values specified on
1728the command line will combine with values specified in the
1729defaults file, rather than replacing them.
1733When the `-s/--standalone` option is used, pandoc uses a template to
1734add header and footer material that is needed for a self-standing
1735document. To see the default template that is used, just type
1739where *FORMAT* is the name of the output format. A custom template
1740can be specified using the `--template` option. You can also override
1741the system default templates for a given output format *FORMAT*
1742by putting a file `templates/default.*FORMAT*` in the user data
1743directory (see `--data-dir`, above). *Exceptions:*
1745- For `odt` output, customize the `default.opendocument`
1747- For `pdf` output, customize the `default.latex` template
1748 (or the `default.context` template, if you use `-t context`,
1749 or the `default.ms` template, if you use `-t ms`, or the
1750 `default.html` template, if you use `-t html`).
1751- `docx` and `pptx` have no template (however, you can use
1752 `--reference-doc` to customize the output).
1754Templates contain *variables*, which allow for the inclusion of
1755arbitrary information at any point in the file. They may be set at the
1756command line using the `-V/--variable` option. If a variable is not set,
1757pandoc will look for the key in the document's metadata, which can be set
1758using either [YAML metadata blocks][Extension: `yaml_metadata_block`]
1759or with the `-M/--metadata` option. In addition, some variables
1760are given default values by pandoc. See [Variables] below for
1761a list of variables used in pandoc's default templates.
1763If you use custom templates, you may need to revise them as pandoc
1764changes. We recommend tracking the changes in the default templates,
1765and modifying your custom templates accordingly. An easy way to do this
1766is to fork the [pandoc-templates] repository and merge in
1767changes after each pandoc release.
1769 [pandoc-templates]: https://github.com/jgm/pandoc-templates
1775Anything between the sequence `$--` and the end of the
1776line will be treated as a comment and omitted from the output.
1780To mark variables and control structures in the template,
1781either `$`...`$` or `${`...`}` may be used as delimiters.
1782The styles may also be mixed in the same template, but the
1783opening and closing delimiter must match in each case. The
1784opening delimiter may be followed by one or more spaces
1785or tabs, which will be ignored. The closing delimiter may
1786be followed by one or more spaces or tabs, which will be
1789To include a literal `$` in the document, use `$$`.
1791### Interpolated variables
1793A slot for an interpolated variable is a variable name surrounded
1794by matched delimiters. Variable names must begin with a letter
1795and can contain letters, numbers, `_`, `-`, and `.`. The
1796keywords `it`, `if`, `else`, `endif`, `for`, `sep`, and `endfor` may
1797not be used as variable names. Examples:
1810Variable names with periods are used to get at structured
1811variable values. So, for example, `employee.salary` will return the
1812value of the `salary` field of the object that is the value of
1813the `employee` field.
1815- If the value of the variable is simple value, it will be
1816 rendered verbatim. (Note that no escaping is done;
1817 the assumption is that the calling program will escape
1818 the strings appropriately for the output format.)
1819- If the value is a list, the values will be concatenated.
1820- If the value is a map, the string `true` will be rendered.
1821- Every other value will be rendered as the empty string.
1825A conditional begins with `if(variable)` (enclosed in
1826matched delimiters) and ends with `endif` (enclosed in matched
1827delimiters). It may optionally contain an `else` (enclosed in
1828matched delimiters). The `if` section is used if
1829`variable` has a non-empty value, otherwise the `else`
1830section is used (if present). Examples:
1845${if(foo)}bar${endif}
1858The keyword `elseif` may be used to simplify complex nested
1873A for loop begins with `for(variable)` (enclosed in
1874matched delimiters) and ends with `endfor` (enclosed in matched
1877- If `variable` is an array, the material inside the loop will
1878 be evaluated repeatedly, with `variable` being set to each
1879 value of the array in turn, and concatenated.
1880- If `variable` is a map, the material inside will be set to
1882- If the value of the associated variable is not an array or
1883 a map, a single iteration will be performed on its value.
1888$for(foo)$$foo$$sep$, $endfor$
1891 - $foo.last$, $foo.first$
1895 - ${ foo.bar.last }, ${ foo.bar.first }
1899$it.name$: $it.office$
1903You may optionally specify a separator between consecutive
1904values using `sep` (enclosed in matched delimiters). The
1905material between `sep` and the `endfor` is the separator.
1908${ for(foo) }${ foo }${ sep }, ${ endfor }
1911Instead of using `variable` inside the loop, the special
1912anaphoric keyword `it` may be used.
1916 - ${ it.last }, ${ it.first }
1922Partials (subtemplates stored in different files) may be
1923included by using the name of the partial, followed
1924by `()`, for example:
1930Partials will be sought in the directory containing
1931the main template. The file name will be assumed to
1932have the same extension as the main template if it
1933lacks an extension. When calling the partial, the
1934full name including file extension can also be used:
1940(If a partial is not found in the directory of the
1941template and the template path is given as a relative
1942path, it will also be sought in the `templates`
1943subdirectory of the user data directory.)
1945Partials may optionally be applied to variables using
1951${ articles:bibentry() }
1954If `articles` is an array, this will iterate over its
1955values, applying the partial `bibentry()` to each one. So the
1956second example above is equivalent to
1964Note that the anaphoric keyword `it` must be used when
1965iterating over partials. In the above examples,
1966the `bibentry` partial should contain `it.title`
1967(and so on) instead of `articles.title`.
1969Final newlines are omitted from included partials.
1971Partials may include other partials.
1973A separator between values of an array may be specified
1974in square brackets, immediately after the variable name
1980${articles:bibentry()[; ]$
1983The separator in this case is literal and (unlike with `sep`
1984in an explicit `for` loop) cannot contain interpolated
1985variables or other template directives.
1989To ensure that content is "nested," that is, subsequent lines
1990indented, use the `^` directive:
1993$item.number$ $^$$item.description$ ($item.price$)
1996In this example, if `item.description` has multiple lines,
1997they will all be indented to line up with the first line:
200000123 A fine bottle of 18-year old
2001 Oban whiskey. ($148)
2004To nest multiple lines to the same level, align them
2005with the `^` directive in the template. For example:
2008$item.number$ $^$$item.description$ ($item.price$)
2009 (Available til $item.sellby$.)
201500123 A fine bottle of 18-year old
2016 Oban whiskey. ($148)
2017 (Available til March 30, 2020.)
2020If a variable occurs by itself on a line, preceded by whitespace
2021and not followed by further text or directives on the same line,
2022and the variable's value contains multiple lines, it will be
2023nested automatically.
2027Normally, spaces in the template itself (as opposed to values of
2028the interpolated variables) are not breakable, but they can be
2029made breakable in part of the template by using the `~` keyword
2030(ended with another `~`).
2033$~$This long line may break if the document is rendered
2034with a short line length.$~$
2039A pipe transforms the value of a variable or partial. Pipes are
2040specified using a slash (`/`) between the variable name (or partial)
2041and the pipe name. Example:
2048$for(metadata/pairs)$
2049- $it.key$: $it.value$
2052$employee:name()/uppercase$
2055Pipes may be chained:
2058$for(employees/pairs)$
2059$it.key/alpha/uppercase$. $it.name$
2063Some pipes take parameters:
2066|----------------------|------------|
2068$it.name.first/uppercase/left 20 "| "$$it.name.salary/right 10 " | " " |"$
2070|----------------------|------------|
2073Currently the following pipes are predefined:
2075- `pairs`: Converts a map or array to an array of maps,
2076 each with `key` and `value` fields. If the original
2077 value was an array, the `key` will be the array index,
2080- `uppercase`: Converts text to uppercase.
2082- `lowercase`: Converts text to lowercase.
2084- `length`: Returns the length of the value: number
2085 of characters for a textual value, number of elements
2088- `reverse`: Reverses a textual value or array,
2089 and has no effect on other values.
2091- `first`: Returns the first value of an array, if
2092 applied to a non-empty array; otherwise returns
2095- `last`: Returns the last value of an array, if
2096 applied to a non-empty array; otherwise returns
2099- `rest`: Returns all but the first value of an array, if
2100 applied to a non-empty array; otherwise returns
2103- `allbutlast`: Returns all but the last value of an array, if
2104 applied to a non-empty array; otherwise returns
2107- `chomp`: Removes trailing newlines (and breakable space).
2109- `nowrap`: Disables line wrapping on breakable spaces.
2111- `alpha`: Converts textual values that can be
2112 read as an integer into lowercase alphabetic
2113 characters `a..z` (mod 26). This can be used to get lettered
2114 enumeration from array indices. To get uppercase
2115 letters, chain with `uppercase`.
2117- `roman`: Converts textual values that can be
2118 read as an integer into lowercase roman numerials.
2119 This can be used to get lettered enumeration from array indices.
2120 To get uppercase roman, chain with `uppercase`.
2122- `left n "leftborder" "rightborder"`: Renders a textual value
2123 in a block of width `n`, aligned to the left, with an optional
2124 left and right border. Has no effect on other values. This
2125 can be used to align material in tables. Widths are positive
2126 integers indicating the number of characters. Borders
2127 are strings inside double quotes; literal `"` and `\` characters
2128 must be backslash-escaped.
2130- `right n "leftborder" "rightborder"`: Renders a textual value
2131 in a block of width `n`, aligned to the right, and has no
2132 effect on other values.
2134- `center n "leftborder" "rightborder"`: Renders a textual
2135 value in a block of width `n`, aligned to the center, and has
2136 no effect on other values.
2141### Metadata variables
2143`title`, `author`, `date`
2144: allow identification of basic aspects of the document. Included
2145 in PDF metadata through LaTeX and ConTeXt. These can be set
2146 through a [pandoc title block][Extension: `pandoc_title_block`],
2147 which allows for multiple authors, or through a
2148 [YAML metadata block][Extension: `yaml_metadata_block`]:
2156 Note that if you just want to set PDF or HTML metadata, without
2157 including a title block in the document itself, you can
2158 set the `title-meta`, `author-meta`, and `date-meta`
2159 variables. (By default these are set automatically, based
2160 on `title`, `author`, and `date`.) The page title in HTML
2161 is set by `pagetitle`, which is equal to `title` by default.
2164: document subtitle, included in HTML, EPUB, LaTeX, ConTeXt, and docx
2168: document summary, included in LaTeX, ConTeXt, AsciiDoc, and docx
2172: list of keywords to be included in HTML, PDF, ODT, pptx, docx
2173 and AsciiDoc metadata; repeat as for `author`, above
2176: document subject, included in ODT, PDF, docx and pptx metadata
2179: document description, included in ODT, docx and pptx metadata. Some
2180 applications show this as `Comments` metadata.
2183: document category, included in docx and pptx metadata
2186any root-level string metadata, not included in ODT, docx
2187or pptx metadata is added as a *custom property*.
2188The following [YAML] metadata block for instance:
2191 title: 'This is the title'
2192 subtitle: "This is the subtitle"
2200 It consists of two paragraphs
2203will include `title`, `author` and `description` as standard document
2204properties and `subtitle` as a custom property when converting to docx,
2207### Language variables
2210: identifies the main language of the document using IETF language
2211 tags (following the [BCP 47] standard), such as `en` or `en-GB`.
2212 The [Language subtag lookup] tool can look up or verify these tags.
2213 This affects most formats, and controls hyphenation in PDF output
2214 when using LaTeX (through [`babel`] and [`polyglossia`]) or ConTeXt.
2216 Use native pandoc [Divs and Spans] with the `lang` attribute to
2217 switch the language:
2223 Text in the main document language (British English).
2226 > Cette citation est écrite en français canadien.
2229 More text in English. ['Zitat auf Deutsch.']{lang=de}
2232: the base script direction, either `rtl` (right-to-left)
2233 or `ltr` (left-to-right).
2235 For bidirectional documents, native pandoc `span`s and
2236 `div`s with the `dir` attribute (value `rtl` or `ltr`) can
2237 be used to override the base direction in some output
2238 formats. This may not always be necessary if the final
2239 renderer (e.g. the browser, when generating HTML) supports
2240 the [Unicode Bidirectional Algorithm].
2242 When using LaTeX for bidirectional documents, only the
2243 `xelatex` engine is fully supported (use
2244 `--pdf-engine=xelatex`).
2246[BCP 47]: https://tools.ietf.org/html/bcp47
2247[Unicode Bidirectional Algorithm]: https://www.w3.org/International/articles/inline-bidi-markup/uba-basics
2248[Language subtag lookup]: https://r12a.github.io/app-subtags/
2250### Variables for HTML
2253: Enables inclusion of most of the [CSS] in the `styles.html`
2254 [partial](#partials) (have a look with
2255 `pandoc --print-default-data-file=templates/styles.html`).
2256 Unless you use [`--css`](#option--css), this variable
2257 is set to `true` by default. You can disable it with
2258 e.g. `pandoc -M document-css=false`.
2261: sets the CSS `font-family` property on the `html` element.
2264: sets the base CSS `font-size`, which you'd usually set
2265 to e.g. `20px`, but it also accepts `pt`
2266 (12pt = 16px in most browsers).
2269: sets the CSS `color` property on the `html` element.
2272: sets the CSS `color` property on all links.
2275: sets the CSS `font-family` property on `code` elements.
2277`monobackgroundcolor`
2278: sets the CSS `background-color` property on `code` elements
2279 and adds extra padding.
2282: sets the CSS `line-height` property on the `html` element,
2283 which is preferred to be unitless.
2286: sets the CSS `background-color` property on the `html` element.
2288`margin-left`, `margin-right`, `margin-top`, `margin-bottom`
2289: sets the corresponding CSS `padding` properties on the `body` element.
2291To override or extend some [CSS] for just one document, include for example:
2300 background-color: #f0f0f0;
2303 padding: 0.5em 2em 0.5em 0.5em;
2306 border-bottom: none;
2311[CSS]: https://developer.mozilla.org/en-US/docs/Learn/CSS
2313### Variables for HTML math
2316: when using [KaTeX](#option--katex), you can render display
2317math equations flush left using [YAML metadata](#layout) or with
2318`-M classoption=fleqn`.
2320### Variables for HTML slides
2322These affect HTML output when [producing slide shows with pandoc].
2324All [reveal.js configuration options] are available as variables.
2325To turn off boolean flags that default to true in reveal.js, use `0`.
2328: base URL for reveal.js documents (defaults to
2329 `https://unpkg.com/reveal.js@^4/`)
2332: base URL for S5 documents (defaults to `s5/default`)
2335: base URL for Slidy documents (defaults to
2336 `https://www.w3.org/Talks/Tools/Slidy2`)
2339: base URL for Slideous documents (defaults to `slideous`)
2341`title-slide-attributes`
2342: additional attributes for the title slide of reveal.js slide shows.
2343 See [background in reveal.js and beamer] for an example.
2345[reveal.js configuration options]: https://revealjs.com/config/
2347### Variables for Beamer slides
2349These variables change the appearance of PDF slides using [`beamer`].
2352: slide aspect ratio (`43` for 4:3 [default], `169` for 16:9,
2353 `1610` for 16:10, `149` for 14:9, `141` for 1.41:1, `54` for 5:4,
2357: produce an article from Beamer slides
2360: add extra beamer option with `\setbeameroption{}`
2363: author affiliations: can be a list when there are multiple authors
2366: logo image for slides
2369: controls navigation symbols (default is `empty` for no navigation
2370 symbols; other valid values are `frame`, `vertical`, and `horizontal`)
2373: enables "title pages" for new sections (default is true)
2375`theme`, `colortheme`, `fonttheme`, `innertheme`, `outertheme`
2379: options for LaTeX beamer themes (a list).
2382: image for title slide
2384### Variables for PowerPoint
2386These variables control the visual aspects of a slide show that
2387are not easily controlled via templates.
2390: font to use for code.
2392### Variables for LaTeX
2394Pandoc uses these variables when [creating a PDF] with a LaTeX engine.
2399: make `\paragraph` and `\subparagraph` (fourth- and
2400 fifth-level headings, or fifth- and sixth-level with book
2401 classes) free-standing rather than run-in; requires further
2402 formatting to distinguish from `\subsubsection` (third- or
2403 fourth-level headings). Instead of using this option,
2404 [KOMA-Script] can adjust headings more extensively:
2407 documentclass: scrartcl
2409 \RedeclareSectionCommand[
2410 beforeskip=-10pt plus -2pt minus -1pt,
2411 afterskip=1sp plus -1sp minus 1sp,
2412 font=\normalfont\itshape]{paragraph}
2413 \RedeclareSectionCommand[
2414 beforeskip=-10pt plus -2pt minus -1pt,
2415 afterskip=1sp plus -1sp minus 1sp,
2416 font=\normalfont\scshape,
2417 indent=0pt]{subparagraph}
2421: option for document class, e.g. `oneside`; repeat for multiple options:
2430: document class: usually one of the standard classes,
2431 [`article`], [`book`], and [`report`]; the [KOMA-Script]
2432 equivalents, `scrartcl`, `scrbook`, and `scrreprt`, which
2433 default to smaller margins; or [`memoir`]
2436: option for [`geometry`] package, e.g. `margin=1in`;
2437 repeat for multiple options:
2447: option for [`hyperref`] package, e.g. `linktoc=all`;
2448 repeat for multiple options:
2454 - pdfpagemode=FullScreen
2458: if true, pandoc will use document class settings for
2459 indentation (the default LaTeX template otherwise removes
2460 indentation and adds space between paragraphs)
2463: adjusts line spacing using the [`setspace`]
2464 package, e.g. `1.25`, `1.5`
2466`margin-left`, `margin-right`, `margin-top`, `margin-bottom`
2467: sets margins if `geometry` is not used (otherwise `geometry`
2471: control `\pagestyle{}`: the default article class
2472 supports `plain` (default), `empty` (no running heads or page numbers),
2473 and `headings` (section titles in running heads)
2476: paper size, e.g. `letter`, `a4`
2479: numbering depth for sections (with `--number-sections` option
2480 or `numbersections` variable)
2485: allows font encoding to be specified through `fontenc` package (with
2486 `pdflatex`); default is `T1` (see [LaTeX font encodings guide])
2489: font package for use with `pdflatex`:
2490 [TeX Live] includes many options, documented in the [LaTeX Font Catalogue].
2491 The default is [Latin Modern][`lm`].
2494: options for package used as `fontfamily`; repeat for multiple options.
2495 For example, to use the Libertine font with proportional lowercase
2496 (old-style) figures through the [`libertinus`] package:
2499 fontfamily: libertinus
2506: font size for body text. The standard classes allow 10pt, 11pt, and
2507 12pt. To use another size, set `documentclass` to one of
2508 the [KOMA-Script] classes, such as `scrartcl` or `scrbook`.
2510`mainfont`, `sansfont`, `monofont`, `mathfont`, `CJKmainfont`
2511: font families for use with `xelatex` or
2512 `lualatex`: take the name of any system font, using the
2513 [`fontspec`] package. `CJKmainfont` uses the [`xecjk`] package.
2515`mainfontoptions`, `sansfontoptions`, `monofontoptions`, `mathfontoptions`, `CJKoptions`
2516: options to use with `mainfont`, `sansfont`, `monofont`, `mathfont`,
2517 `CJKmainfont` in `xelatex` and `lualatex`. Allow for any
2518 choices available through [`fontspec`]; repeat for multiple
2519 options. For example, to use the [TeX Gyre] version of
2520 Palatino with lowercase figures:
2523 mainfont: TeX Gyre Pagella
2526 - Numbers=Proportional
2530: options to pass to the microtype package
2535: add color to link text; automatically enabled if any of
2536 `linkcolor`, `filecolor`, `citecolor`, `urlcolor`, or `toccolor` are set
2538`linkcolor`, `filecolor`, `citecolor`, `urlcolor`, `toccolor`
2539: color for internal links, external links, citation links, linked
2540 URLs, and links in table of contents, respectively: uses options
2541 allowed by [`xcolor`], including the `dvipsnames`, `svgnames`, and
2545: causes links to be printed as footnotes
2550: include list of figures, list of tables
2553: contents of acknowledgments footnote after document title
2556: include table of contents (can also be set using
2557 `--toc/--table-of-contents`)
2560: level of section to include in table of contents
2562#### BibLaTeX Bibliographies
2564These variables function when using BibLaTeX for [citation rendering].
2567: list of options for biblatex
2570: bibliography style, when used with `--natbib` and `--biblatex`.
2573: bibliography title, when used with `--natbib` and `--biblatex`.
2576: bibliography to use for resolving references
2579: list of options for natbib
2581[KOMA-Script]: https://ctan.org/pkg/koma-script
2582[LaTeX Font Catalogue]: https://tug.org/FontCatalogue/
2583[LaTeX font encodings guide]: https://ctan.org/pkg/encguide
2584[TeX Gyre]: http://www.gust.org.pl/projects/e-foundry/tex-gyre
2585[`article`]: https://ctan.org/pkg/article
2586[`book`]: https://ctan.org/pkg/book
2587[`libertinus`]: https://ctan.org/pkg/libertinus
2588[`memoir`]: https://ctan.org/pkg/memoir
2589[`report`]: https://ctan.org/pkg/report
2591### Variables for ConTeXt
2593Pandoc uses these variables when [creating a PDF] with ConTeXt.
2596: font size for body text (e.g. `10pt`, `12pt`)
2598`headertext`, `footertext`
2599: text to be placed in running header or footer (see [ConTeXt Headers and
2600 Footers]); repeat up to four times for different placement
2603: controls indentation of paragraphs, e.g. `yes,small,next` (see
2604 [ConTeXt Indentation]); repeat for multiple options
2607: adjusts line spacing, e.g. `4ex` (using [`setupinterlinespace`]);
2608 repeat for multiple options
2611: options for page margins and text arrangement (see [ConTeXt Layout]);
2612 repeat for multiple options
2614`linkcolor`, `contrastcolor`
2615: color for links outside and inside a page, e.g. `red`, `blue` (see
2619: typeface style for links, e.g. `normal`, `bold`, `slanted`, `boldslanted`,
2620 `type`, `cap`, `small`
2623: include list of figures, list of tables
2625`mainfont`, `sansfont`, `monofont`, `mathfont`
2626: font families: take the name of any system font (see
2627 [ConTeXt Font Switching])
2629`margin-left`, `margin-right`, `margin-top`, `margin-bottom`
2630: sets margins, if `layout` is not used (otherwise `layout`
2634: page number style and location (using [`setuppagenumbering`]);
2635 repeat for multiple options
2638: paper size, e.g. `letter`, `A4`, `landscape` (see [ConTeXt Paper Setup]);
2639 repeat for multiple options
2642: adds to the preamble the setup necessary to generate PDF/A
2643 of the type specified, e.g. `1a:2005`, `2a`. If no type is
2644 specified (i.e. the value is set to True, by e.g.
2645 `--metadata=pdfa` or `pdfa: true` in a YAML metadata block),
2646 `1b:2005` will be used as default, for reasons of backwards
2647 compatibility. Using `--variable=pdfa` without specified value
2648 is not supported. To successfully generate PDF/A the required
2649 ICC color profiles have to be available and the content and all
2650 included files (such as images) have to be standard conforming.
2651 The ICC profiles and output intent may be specified using the
2652 variables `pdfaiccprofile` and `pdfaintent`. See also [ConTeXt
2653 PDFA] for more details.
2656: when used in conjunction with `pdfa`, specifies the ICC profile to use
2657 in the PDF, e.g. `default.cmyk`. If left unspecified, `sRGB.icc` is
2658 used as default. May be repeated to include multiple profiles. Note that
2659 the profiles have to be available on the system. They can be obtained
2660 from [ConTeXt ICC Profiles].
2663: when used in conjunction with `pdfa`, specifies the output intent for
2664 the colors, e.g. `ISO coated v2 300\letterpercent\space (ECI)`
2665 If left unspecified, `sRGB IEC61966-2.1` is used as default.
2668: include table of contents (can also be set using
2669 `--toc/--table-of-contents`)
2672: spacing between paragraphs, e.g. `none`, `small` (using
2673 [`setupwhitespace`])
2676: include all source documents as file attachments in the PDF file
2678[ConTeXt Paper Setup]: https://wiki.contextgarden.net/PaperSetup
2679[ConTeXt Layout]: https://wiki.contextgarden.net/Layout
2680[ConTeXt Font Switching]: https://wiki.contextgarden.net/Font_Switching
2681[ConTeXt Color]: https://wiki.contextgarden.net/Color
2682[ConTeXt Headers and Footers]: https://wiki.contextgarden.net/Headers_and_Footers
2683[ConTeXt Indentation]: https://wiki.contextgarden.net/Indentation
2684[ConTeXt ICC Profiles]: https://wiki.contextgarden.net/PDFX#ICC_profiles
2685[ConTeXt PDFA]: https://wiki.contextgarden.net/PDF/A
2686[`setupwhitespace`]: https://wiki.contextgarden.net/Command/setupwhitespace
2687[`setupinterlinespace`]: https://wiki.contextgarden.net/Command/setupinterlinespace
2688[`setuppagenumbering`]: https://wiki.contextgarden.net/Command/setuppagenumbering
2690### Variables for `wkhtmltopdf`
2692Pandoc uses these variables when [creating a PDF] with [`wkhtmltopdf`].
2693The `--css` option also affects the output.
2695`footer-html`, `header-html`
2696: add information to the header and footer
2698`margin-left`, `margin-right`, `margin-top`, `margin-bottom`
2699: set the page margins
2702: sets the PDF paper size
2704### Variables for man pages
2707: adjusts text to left (`l`), right (`r`), center (`c`),
2708 or both (`b`) margins
2711: footer in man pages
2714: header in man pages
2717: if `true` (the default), hyphenation will be used
2720: section number in man pages
2725: font family (e.g. `T` or `P`)
2728: paragraph indent (e.g. `2m`)
2731: line height (e.g. `12p`)
2734: point size (e.g. `10p`)
2736### Variables set automatically
2738Pandoc sets these variables automatically in response to [options] or
2739document contents; users can also modify them. These vary depending
2740on the output format, and include the following:
2746: the `date` variable converted to ISO 8601 YYYY-MM-DD,
2747 included in all HTML based formats (dzslides, epub,
2748 html, html4, html5, revealjs, s5, slideous, slidy).
2749 The recognized formats for `date` are: `mm/dd/yyyy`,
2750 `mm/dd/yy`, `yyyy-mm-dd` (ISO 8601), `dd MM yyyy`
2751 (e.g. either `02 Apr 2018` or `02 April 2018`),
2752 `MM dd, yyyy` (e.g. `Apr. 02, 2018` or `April 02, 2018),
2753 `yyyy[mm[dd]]]` (e.g. `20180402, `201804` or `2018`).
2756: contents specified by `-H/--include-in-header` (may have multiple
2760: contents specified by `-B/--include-before-body` (may have
2764: contents specified by `-A/--include-after-body` (may have
2768: JSON representation of all of the document's metadata. Field
2769 values are transformed to the selected output format.
2772: non-null value if `-N/--number-sections` was specified
2774`sourcefile`, `outputfile`
2775: source and destination filenames, as given on the command line.
2776 `sourcefile` can also be a list if input comes from multiple files,
2777 or empty if input is from stdin. You can use the following snippet in
2778 your template to distinguish them:
2788 Similarly, `outputfile` can be `-` if output goes to the terminal.
2790 If you need absolute paths, use e.g. `$curdir$/$sourcefile$`.
2793: working directory from which pandoc is run.
2796: non-null value if `--toc/--table-of-contents` was specified
2799: title of table of contents (works only with EPUB,
2800 HTML, opendocument, odt, docx, pptx, beamer, LaTeX)
2802[pandoc-templates]: https://github.com/jgm/pandoc-templates
2806The behavior of some of the readers and writers can be adjusted by
2807enabling or disabling various extensions.
2809An extension can be enabled by adding `+EXTENSION`
2810to the format name and disabled by adding `-EXTENSION`. For example,
2811`--from markdown_strict+footnotes` is strict Markdown with footnotes
2812enabled, while `--from markdown-footnotes-pipe_tables` is pandoc's
2813Markdown without footnotes or pipe tables.
2815The markdown reader and writer make by far the most use of extensions.
2816Extensions only used by them are therefore covered in the
2817section [Pandoc's Markdown] below (See [Markdown variants] for
2818`commonmark` and `gfm`.) In the following, extensions that also work
2819for other formats are covered.
2821Note that markdown extensions added to the `ipynb` format
2822affect Markdown cells in Jupyter notebooks (as do command-line
2823options like `--atx-headers`).
2827#### Extension: `smart` ####
2829Interpret straight quotes as curly quotes, `---` as em-dashes,
2830`--` as en-dashes, and `...` as ellipses. Nonbreaking spaces are
2831inserted after certain abbreviations, such as "Mr."
2833This extension can be enabled/disabled for the following formats:
2836: `markdown`, `commonmark`, `latex`, `mediawiki`, `org`, `rst`, `twiki`
2839: `markdown`, `latex`, `context`, `rst`
2841enabled by default in
2842: `markdown`, `latex`, `context` (both input and output)
2844Note: If you are *writing* Markdown, then the `smart` extension
2845has the reverse effect: what would have been curly quotes comes
2848In LaTeX, `smart` means to use the standard TeX ligatures
2849for quotation marks (` `` ` and ` '' ` for double quotes,
2850`` ` `` and `` ' `` for single quotes) and dashes (`--` for
2851en-dash and `---` for em-dash). If `smart` is disabled,
2852then in reading LaTeX pandoc will parse these characters
2853literally. In writing LaTeX, enabling `smart` tells pandoc
2854to use the ligatures when possible; if `smart` is disabled
2855pandoc will use unicode quotation mark and dash characters.
2857## Headings and sections
2859#### Extension: `auto_identifiers` ####
2861A heading without an explicitly specified identifier will be
2862automatically assigned a unique identifier based on the heading text.
2864This extension can be enabled/disabled for the following formats:
2867: `markdown`, `latex`, `rst`, `mediawiki`, `textile`
2872enabled by default in
2875The default algorithm used to derive the identifier from the
2878 - Remove all formatting, links, etc.
2879 - Remove all footnotes.
2880 - Remove all non-alphanumeric characters,
2881 except underscores, hyphens, and periods.
2882 - Replace all spaces and newlines with hyphens.
2883 - Convert all alphabetic characters to lowercase.
2884 - Remove everything up to the first letter (identifiers may
2885 not begin with a number or punctuation mark).
2886 - If nothing is left after this, use the identifier `section`.
2891 ------------------------------- ----------------------------
2892 `Heading identifiers in HTML` `heading-identifiers-in-html`
2893 `Maître d'hôtel` `maître-dhôtel`
2894 `*Dogs*?--in *my* house?` `dogs--in-my-house`
2895 `[HTML], [S5], or [RTF]?` `html-s5-or-rtf`
2896 `3. Applications` `applications`
2899These rules should, in most cases, allow one to determine the identifier
2900from the heading text. The exception is when several headings have the
2901same text; in this case, the first will get an identifier as described
2902above; the second will get the same identifier with `-1` appended; the
2903third with `-2`; and so on.
2905(However, a different algorithm is used if
2906`gfm_auto_identifiers` is enabled; see below.)
2908These identifiers are used to provide link targets in the table of
2909contents generated by the `--toc|--table-of-contents` option. They
2910also make it easy to provide links from one section of a document to
2911another. A link to this section, for example, might look like this:
2914 [heading identifiers](#heading-identifiers-in-html-latex-and-context).
2916Note, however, that this method of providing links to sections works
2917only in HTML, LaTeX, and ConTeXt formats.
2919If the `--section-divs` option is specified, then each section will
2920be wrapped in a `section` (or a `div`, if `html4` was specified),
2921and the identifier will be attached to the enclosing `<section>`
2922(or `<div>`) tag rather than the heading itself. This allows entire
2923sections to be manipulated using JavaScript or treated differently in
2926#### Extension: `ascii_identifiers` ####
2928Causes the identifiers produced by `auto_identifiers` to be pure ASCII.
2929Accents are stripped off of accented Latin letters, and non-Latin
2932#### Extension: `gfm_auto_identifiers` ####
2934Changes the algorithm used by `auto_identifiers` to conform to
2935GitHub's method. Spaces are converted to dashes (`-`),
2936uppercase characters to lowercase characters, and punctuation
2937characters other than `-` and `_` are removed.
2938Emojis are replaced by their names.
2942The extensions [`tex_math_dollars`](#extension-tex_math_dollars),
2943[`tex_math_single_backslash`](#extension-tex_math_single_backslash), and
2944[`tex_math_double_backslash`](#extension-tex_math_double_backslash)
2945are described in the section about Pandoc's Markdown.
2947However, they can also be used with HTML input. This is handy for
2948reading web pages formatted using MathJax, for example.
2952The following extensions are described in more detail in
2953their respective sections of [Pandoc's Markdown]:
2955- [`raw_html`](#extension-raw_html) allows HTML elements which
2956 are not representable in pandoc's AST to be parsed as raw HTML.
2957 By default, this is disabled for HTML input.
2959- [`raw_tex`](#extension-raw_tex) allows raw LaTeX, TeX, and ConTeXt
2960 to be included in a document. This extension can be enabled/disabled
2961 for the following formats (in addition to `markdown`):
2964 : `latex`, `textile`, `html` (environments, `\ref`, and
2965 `\eqref` only), `ipynb`
2968 : `textile`, `commonmark`
2970 Note: as applied to `ipynb`, `raw_html` and `raw_tex` affect not
2971 only raw TeX in markdown cells, but data with mime type
2972 `text/html` in output cells. Since the `ipynb` reader attempts
2973 to preserve the richest possible outputs when several options
2974 are given, you will get best results if you disable `raw_html`
2975 and `raw_tex` when converting to formats like `docx` which don't
2976 allow raw `html` or `tex`.
2978- [`native_divs`](#extension-native_divs) causes HTML `div`
2979 elements to be parsed as native pandoc Div blocks.
2980 If you want them to be parsed as raw HTML, use
2981 `-f html-native_divs+raw_html`.
2983- [`native_spans`](#extension-native_spans) causes HTML `span`
2984 elements to be parsed as native pandoc Span inlines.
2985 If you want them to be parsed as raw HTML, use
2986 `-f html-native_spans+raw_html`. If you want to drop all
2987 `div`s and `span`s when converting HTML to Markdown, you
2988 can use `pandoc -f html-native_divs-native_spans -t markdown`.
2990## Literate Haskell support
2992#### Extension: `literate_haskell` ####
2994Treat the document as literate Haskell source.
2996This extension can be enabled/disabled for the following formats:
2999: `markdown`, `rst`, `latex`
3002: `markdown`, `rst`, `latex`, `html`
3004If you append `+lhs` (or `+literate_haskell`) to one of the formats
3005above, pandoc will treat the document as literate Haskell source.
3008 - In Markdown input, "bird track" sections will be parsed as Haskell
3009 code rather than block quotations. Text between `\begin{code}`
3010 and `\end{code}` will also be treated as Haskell code. For
3011 ATX-style headings the character '=' will be used instead of '#'.
3013 - In Markdown output, code blocks with classes `haskell` and `literate`
3014 will be rendered using bird tracks, and block quotations will be
3015 indented one space, so they will not be treated as Haskell code.
3016 In addition, headings will be rendered setext-style (with underlines)
3017 rather than ATX-style (with '#' characters). (This is because ghc
3018 treats '#' characters in column 1 as introducing line numbers.)
3020 - In restructured text input, "bird track" sections will be parsed
3023 - In restructured text output, code blocks with class `haskell` will
3024 be rendered using bird tracks.
3026 - In LaTeX input, text in `code` environments will be parsed as
3029 - In LaTeX output, code blocks with class `haskell` will be rendered
3030 inside `code` environments.
3032 - In HTML output, code blocks with class `haskell` will be rendered
3033 with class `literatehaskell` and bird tracks.
3037 pandoc -f markdown+lhs -t html
3039reads literate Haskell source formatted with Markdown conventions and writes
3040ordinary HTML (without bird tracks).
3042 pandoc -f markdown+lhs -t html+lhs
3044writes HTML with the Haskell code in bird tracks, so it can be copied
3045and pasted as literate Haskell source.
3047Note that GHC expects the bird tracks in the first column, so indented
3048literate code blocks (e.g. inside an itemized environment) will not be
3049picked up by the Haskell compiler.
3053#### Extension: `empty_paragraphs` ####
3055Allows empty paragraphs. By default empty paragraphs are
3058This extension can be enabled/disabled for the following formats:
3064: `docx`, `odt`, `opendocument`, `html`
3066#### Extension: `native_numbering` ####
3068Enables native numbering of figures and tables. Enumeration
3071This extension can be enabled/disabled for the following formats:
3074: `odt`, `opendocument`
3076#### Extension: `xrefs_name` ####
3078Links to headings, figures and tables inside the document are
3079substituted with cross-references that will use the name or caption
3080of the referenced item. The original link text is replaced once
3081the generated document is refreshed. This extension can be combined
3082with `xrefs_number` in which case numbers will appear before the
3085Text in cross-references is only made consistent with the referenced
3086item once the document has been refreshed.
3088This extension can be enabled/disabled for the following formats:
3091: `odt`, `opendocument`
3093#### Extension: `xrefs_number` ####
3095Links to headings, figures and tables inside the document are
3096substituted with cross-references that will use the number
3097of the referenced item. The original link text is discarded.
3098This extension can be combined with `xrefs_name` in which case
3099the name or caption numbers will appear after the number.
3101For the `xrefs_number` to be useful heading numbers must be enabled
3102in the generated document, also table and figure captions must be enabled
3103using for example the `native_numbering` extension.
3105Numbers in cross-references are only visible in the final document once
3106it has been refreshed.
3108This extension can be enabled/disabled for the following formats:
3111: `odt`, `opendocument`
3113#### Extension: `styles` #### {#ext-styles}
3115When converting from docx, read all docx styles as divs (for
3116paragraph styles) and spans (for character styles) regardless
3117of whether pandoc understands the meaning of these styles.
3118This can be used with [docx custom styles](#custom-styles).
3124#### Extension: `amuse` ####
3126In the `muse` input format, this enables Text::Amuse
3127extensions to Emacs Muse markup.
3129#### Extension: `raw_markdown` ####
3131In the `ipynb` input format, this causes Markdown cells
3132to be included as raw Markdown blocks (allowing lossless
3133round-tripping) rather than being parsed. Use this only
3134when you are targeting `ipynb` or a markdown-based
3137#### Extension: `citations` {#org-citations}
3139Some aspects of [Pandoc's Markdown citation syntax](#citations)
3140are also accepted in `org` input.
3142#### Extension: `element_citations` ####
3144In the `jats` output formats, this causes reference items to
3145be replaced with `<element-citation>` elements. These
3146elements are not influenced by CSL styles, but all information
3147on the item is included in tags.
3149#### Extension: `ntb` ####
3151In the `context` output format this enables the use of [Natural Tables
3152(TABLE)](https://wiki.contextgarden.net/TABLE) instead of the default
3153[Extreme Tables (xtables)](https://wiki.contextgarden.net/xtables).
3154Natural tables allow more fine-grained global customization but come
3155at a performance penalty compared to extreme tables.
3160Pandoc understands an extended and slightly revised version of
3161John Gruber's [Markdown] syntax. This document explains the syntax,
3162noting differences from standard Markdown. Except where noted, these
3163differences can be suppressed by using the `markdown_strict` format instead
3164of `markdown`. Extensions can be enabled or disabled to specify the
3165behavior more granularly. They are described in the following. See also
3166[Extensions] above, for extensions that work also on other formats.
3170Markdown is designed to be easy to write, and, even more importantly,
3173> A Markdown-formatted document should be publishable as-is, as plain
3174> text, without looking like it's been marked up with tags or formatting
3176> -- [John Gruber](https://daringfireball.net/projects/markdown/syntax#philosophy)
3178This principle has guided pandoc's decisions in finding syntax for
3179tables, footnotes, and other extensions.
3181There is, however, one respect in which pandoc's aims are different
3182from the original aims of Markdown. Whereas Markdown was originally
3183designed with HTML generation in mind, pandoc is designed for multiple
3184output formats. Thus, while pandoc allows the embedding of raw HTML,
3185it discourages it, and provides other, non-HTMLish ways of representing
3186important document elements like definition lists, tables, mathematics, and
3191A paragraph is one or more lines of text followed by one or more blank lines.
3192Newlines are treated as spaces, so you can reflow your paragraphs as you like.
3193If you need a hard line break, put two or more spaces at the end of a line.
3195#### Extension: `escaped_line_breaks` ####
3197A backslash followed by a newline is also a hard line break.
3198Note: in multiline and grid table cells, this is the only way
3199to create a hard line break, since trailing spaces in the cells
3204There are two kinds of headings: Setext and ATX.
3206### Setext-style headings ###
3208A setext-style heading is a line of text "underlined" with a row of `=` signs
3209(for a level-one heading) or `-` signs (for a level-two heading):
3217The heading text can contain inline formatting, such as emphasis (see
3218[Inline formatting], below).
3221### ATX-style headings ###
3223An ATX-style heading consists of one to six `#` signs and a line of
3224text, optionally followed by any number of `#` signs. The number of
3225`#` signs at the beginning of the line is the heading level:
3227 ## A level-two heading
3229 ### A level-three heading ###
3231As with setext-style headings, the heading text can contain formatting:
3233 # A level-one heading with a [link](/url) and *emphasis*
3235#### Extension: `blank_before_header` ####
3237Standard Markdown syntax does not require a blank line before a heading.
3238Pandoc does require this (except, of course, at the beginning of the
3239document). The reason for the requirement is that it is all too easy for a
3240`#` to end up at the beginning of a line by accident (perhaps through line
3241wrapping). Consider, for example:
3243 I like several of their flavors of ice cream:
3244 #22, for example, and #5.
3246#### Extension: `space_in_atx_header` ####
3248Many Markdown implementations do not require a space between the
3249opening `#`s of an ATX heading and the heading text, so that
3250`#5 bolt` and `#hashtag` count as headings. With this extension,
3251pandoc does require the space.
3253### Heading identifiers ###
3255See also the [`auto_identifiers`
3256extension](#extension-auto_identifiers) above.
3258#### Extension: `header_attributes` ####
3260Headings can be assigned attributes using this syntax at the end
3261of the line containing the heading text:
3263 {#identifier .class .class key=value key=value}
3265Thus, for example, the following headings will all be assigned the identifier
3270 ## My heading ## {#foo}
3272 My other heading {#foo}
3275(This syntax is compatible with [PHP Markdown Extra].)
3277Note that although this syntax allows assignment of classes and key/value
3278attributes, writers generally don't use all of this information. Identifiers,
3279classes, and key/value attributes are used in HTML and HTML-based formats such
3280as EPUB and slidy. Identifiers are used for labels and link anchors in the
3281LaTeX, ConTeXt, Textile, Jira markup, and AsciiDoc writers.
3283Headings with the class `unnumbered` will not be numbered, even if
3284`--number-sections` is specified. A single hyphen (`-`) in an attribute
3285context is equivalent to `.unnumbered`, and preferable in non-English
3292 # My heading {.unnumbered}
3294If the `unlisted` class is present in addition to `unnumbered`,
3295the heading will not be included in a table of contents.
3296(Currently this feature is only implemented for certain
3297formats: those based on LaTeX and HTML, PowerPoint, and RTF.)
3299#### Extension: `implicit_header_references` ####
3301Pandoc behaves as if reference links have been defined for each heading.
3302So, to link to a heading
3304 # Heading identifiers in HTML
3308 [Heading identifiers in HTML]
3312 [Heading identifiers in HTML][]
3316 [the section on heading identifiers][heading identifiers in
3319instead of giving the identifier explicitly:
3321 [Heading identifiers in HTML](#heading-identifiers-in-html)
3323If there are multiple headings with identical text, the corresponding
3324reference will link to the first one only, and you will need to use explicit
3325links to link to the others, as described above.
3327Like regular reference links, these references are case-insensitive.
3329Explicit link reference definitions always take priority over
3330implicit heading references. So, in the following example, the
3331link will point to `bar`, not to `#foo`:
3341Markdown uses email conventions for quoting blocks of text.
3342A block quotation is one or more paragraphs or other block elements
3343(such as lists or headings), with each line preceded by a `>` character
3344and an optional space. (The `>` need not start at the left margin, but
3345it should not be indented more than three spaces.)
3347 > This is a block quote. This
3348 > paragraph has two lines.
3350 > 1. This is a list inside a block quote.
3353A "lazy" form, which requires the `>` character only on the first
3354line of each block, is also allowed:
3356 > This is a block quote. This
3357 paragraph has two lines.
3359 > 1. This is a list inside a block quote.
3362Among the block elements that can be contained in a block quote are
3363other block quotes. That is, block quotes can be nested:
3365 > This is a block quote.
3367 > > A block quote within a block quote.
3369If the `>` character is followed by an optional space, that space
3370will be considered part of the block quote marker and not part of
3371the indentation of the contents. Thus, to put an indented code
3372block in a block quote, you need five spaces after the `>`:
3376#### Extension: `blank_before_blockquote` ####
3378Standard Markdown syntax does not require a blank line before a
3379block quote. Pandoc does require this (except, of course, at
3380the beginning of the document). The reason for the requirement
3381is that it is all too easy for a `>` to end up at the beginning
3382of a line by accident (perhaps through line wrapping). So,
3383unless the `markdown_strict` format is used, the following does
3384not produce a nested block quote in pandoc:
3386 > This is a block quote.
3390## Verbatim (code) blocks
3392### Indented code blocks ###
3394A block of text indented four spaces (or one tab) is treated as verbatim
3395text: that is, special characters do not trigger special formatting,
3396and all spaces and line breaks are preserved. For example,
3399 moveShip(5 * gravity, DOWN);
3402The initial (four space or one tab) indentation is not considered part
3403of the verbatim text, and is removed in the output.
3405Note: blank lines in the verbatim text need not begin with four spaces.
3408### Fenced code blocks ###
3410#### Extension: `fenced_code_blocks` ####
3412In addition to standard indented code blocks, pandoc supports
3413*fenced* code blocks. These begin with a row of three or more
3414tildes (`~`) and end with a row of tildes that must be at least as long as
3415the starting row. Everything between these lines is treated as code. No
3416indentation is necessary:
3420 moveShip(5 * gravity, DOWN);
3424Like regular code blocks, fenced code blocks must be separated
3425from surrounding text by blank lines.
3427If the code itself contains a row of tildes or backticks, just use a longer
3428row of tildes or backticks at the start and end:
3432 code including tildes
3436#### Extension: `backtick_code_blocks` ####
3438Same as `fenced_code_blocks`, but uses backticks (`` ` ``) instead of tildes
3441#### Extension: `fenced_code_attributes` ####
3443Optionally, you may attach attributes to fenced or backtick code block using
3446 ~~~~ {#mycode .haskell .numberLines startFrom="100"}
3448 qsort (x:xs) = qsort (filter (< x) xs) ++ [x] ++
3449 qsort (filter (>= x) xs)
3450 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3452Here `mycode` is an identifier, `haskell` and `numberLines` are
3453classes, and `startFrom` is an attribute with value `100`. Some
3454output formats can use this information to do syntax
3455highlighting. Currently, the only output formats that uses this
3456information are HTML, LaTeX, Docx, Ms, and PowerPoint. If
3457highlighting is supported for your output format and language,
3458then the code block above will appear highlighted, with numbered
3459lines. (To see which languages are supported, type `pandoc
3460--list-highlight-languages`.) Otherwise, the code block above
3461will appear as follows:
3463 <pre id="mycode" class="haskell numberLines" startFrom="100">
3469The `numberLines` (or `number-lines`) class will cause the lines
3470of the code block to be numbered, starting with `1` or the value
3471of the `startFrom` attribute. The `lineAnchors` (or
3472`line-anchors`) class will cause the lines to be clickable
3473anchors in HTML output.
3475A shortcut form can also be used for specifying the language of
3482This is equivalent to:
3488If the `fenced_code_attributes` extension is disabled, but
3489input contains class attribute(s) for the code block, the first
3490class attribute will be printed after the opening fence as a bare
3493To prevent all highlighting, use the `--no-highlight` flag.
3494To set the highlighting style, use `--highlight-style`.
3495For more information on highlighting, see [Syntax highlighting],
3500#### Extension: `line_blocks` ####
3502A line block is a sequence of lines beginning with a vertical bar (`|`)
3503followed by a space. The division into lines will be preserved in
3504the output, as will any leading spaces; otherwise, the lines will
3505be formatted as Markdown. This is useful for verse and addresses:
3507 | The limerick packs laughs anatomical
3508 | In space that is quite economical.
3509 | But the good ones I've seen
3510 | So seldom are clean
3511 | And the clean ones so seldom are comical
3514 | Berkeley, CA 94718
3516The lines can be hard-wrapped if needed, but the continuation
3517line must begin with a space.
3519 | The Right Honorable Most Venerable and Righteous Samuel L.
3522 | Berkeley, CA 94718
3524Inline formatting (such as emphasis) is allowed in the content,
3525but not block-level formatting (such as block quotes or lists).
3527This syntax is borrowed from [reStructuredText].
3533A bullet list is a list of bulleted list items. A bulleted list
3534item begins with a bullet (`*`, `+`, or `-`). Here is a simple
3541This will produce a "compact" list. If you want a "loose" list, in which
3542each item is formatted as a paragraph, put spaces between the items:
3550The bullets need not be flush with the left margin; they may be
3551indented one, two, or three spaces. The bullet must be followed
3554List items look best if subsequent lines are flush with the first
3555line (after the bullet):
3561But Markdown also allows a "lazy" format:
3567### Block content in list items ###
3569A list item may contain multiple paragraphs and other block-level
3570content. However, subsequent paragraphs must be preceded by a blank line
3571and indented to line up with the first non-space content after
3578 * Second paragraph. With a code block, which must be indented
3583Exception: if the list marker is followed by an indented code
3584block, which must begin 5 spaces after the list marker, then
3585subsequent paragraphs must begin two columns after the last
3586character of the list marker:
3590 continuation paragraph
3592List items may include other lists. In this case the preceding blank
3593line is optional. The nested list must be indented to line up with
3594the first non-space character after the list marker of the
3595containing list item.
3607As noted above, Markdown allows you to write list items "lazily," instead of
3608indenting continuation lines. However, if there are multiple paragraphs or
3609other blocks in a list item, the first line of each must be indented.
3611 + A lazy, lazy, list
3614 + Another one; this looks
3617 Second paragraph of second
3620### Ordered lists ###
3622Ordered lists work just like bulleted lists, except that the items
3623begin with enumerators rather than bullets.
3625In standard Markdown, enumerators are decimal numbers followed
3626by a period and a space. The numbers themselves are ignored, so
3627there is no difference between this list:
3639#### Extension: `fancy_lists` ####
3641Unlike standard Markdown, pandoc allows ordered list items to be marked
3642with uppercase and lowercase letters and roman numerals, in addition to
3643Arabic numerals. List markers may be enclosed in parentheses or followed by a
3644single right-parentheses or period. They must be separated from the
3645text that follows by at least one space, and, if the list marker is a
3646capital letter with a period, by at least two spaces.[^2]
3648[^2]: The point of this rule is to ensure that normal paragraphs
3649 starting with people's initials, like
3651 B. Russell was an English philosopher.
3653 do not get treated as list items.
3655 This rule will not prevent
3659 from being interpreted as a list item. In this case, a backslash
3664The `fancy_lists` extension also allows '`#`' to be used as an
3665ordered list marker in place of a numeral:
3670#### Extension: `startnum` ####
3672Pandoc also pays attention to the type of list marker used, and to the
3673starting number, and both of these are preserved where possible in the
3674output format. Thus, the following yields a list with numbers followed
3675by a single parenthesis, starting with 9, and a sublist with lowercase
3685Pandoc will start a new list each time a different type of list
3686marker is used. So, the following will create three lists:
3693If default list markers are desired, use `#.`:
3699#### Extension: `task_lists` ####
3701Pandoc supports task lists, using the syntax of GitHub-Flavored Markdown.
3703 - [ ] an unchecked task list item
3706### Definition lists ###
3708#### Extension: `definition_lists` ####
3710Pandoc supports definition lists, using the syntax of
3711[PHP Markdown Extra] with some extensions.[^3]
3717 Term 2 with *inline markup*
3721 { some code, part of Definition 2 }
3723 Third paragraph of definition 2.
3725Each term must fit on one line, which may optionally be followed by
3726a blank line, and must be followed by one or more definitions.
3727A definition begins with a colon or tilde, which may be indented one
3730A term may have multiple definitions, and each definition may
3731consist of one or more block elements (paragraph, code block,
3732list, etc.), each indented four spaces or one tab stop. The
3733body of the definition (including the first line, aside from the
3734colon or tilde) should be indented four spaces. However, as with
3735other Markdown lists, you can "lazily" omit indentation except
3736at the beginning of a paragraph or other block element:
3741 with lazy continuation.
3743 Second paragraph of the definition.
3745If you leave space before the definition (as in the example above),
3746the text of the definition will be treated as a paragraph. In some
3747output formats, this will mean greater spacing between term/definition
3748pairs. For a more compact definition list, omit the space before the
3758Note that space between items in a definition list is required.
3759(A variant that loosens this requirement, but disallows "lazy"
3760hard wrapping, can be activated with `compact_definition_lists`: see
3761[Non-pandoc extensions], below.)
3763[^3]: I have been influenced by the suggestions of [David
3764 Wheeler](https://justatheory.com/2009/02/modest-markdown-proposal/).
3766### Numbered example lists ###
3768#### Extension: `example_lists` ####
3770The special list marker `@` can be used for sequentially numbered
3771examples. The first list item with a `@` marker will be numbered '1',
3772the next '2', and so on, throughout the document. The numbered examples
3773need not occur in a single list; each new list using `@` will take up
3774where the last stopped. So, for example:
3776 (@) My first example will be numbered (1).
3777 (@) My second example will be numbered (2).
3779 Explanation of examples.
3781 (@) My third example will be numbered (3).
3783Numbered examples can be labeled and referred to elsewhere in the
3786 (@good) This is a good example.
3788 As (@good) illustrates, ...
3790The label can be any string of alphanumeric characters, underscores,
3793Note: continuation paragraphs in example lists must always
3794be indented four spaces, regardless of the length of the
3795list marker. That is, example lists always behave as if the
3796`four_space_rule` extension is set. This is because example
3797labels tend to be long, and indenting content to the
3798first non-space character after the label would be awkward.
3801### Ending a list ###
3803What if you want to put an indented code block after a list?
3810Trouble! Here pandoc (like other Markdown implementations) will treat
3811`{ my code block }` as the second paragraph of item two, and not as
3814To "cut off" the list after item two, you can insert some non-indented
3815content, like an HTML comment, which won't produce visible output in
3821 <!-- end of list -->
3825You can use the same trick if you want two consecutive lists instead
3840A line containing a row of three or more `*`, `-`, or `_` characters
3841(optionally separated by spaces) produces a horizontal rule:
3850Four kinds of tables may be used. The first three kinds presuppose the use of
3851a fixed-width font, such as Courier. The fourth kind can be used with
3852proportionally spaced fonts, as it does not require lining up columns.
3854#### Extension: `table_captions` ####
3856A caption may optionally be provided with all 4 kinds of tables (as
3857illustrated in the examples below). A caption is a paragraph beginning
3858with the string `Table:` (or just `:`), which will be stripped off.
3859It may appear either before or after the table.
3861#### Extension: `simple_tables` ####
3863Simple tables look like this:
3865 Right Left Center Default
3866 ------- ------ ---------- -------
3871 Table: Demonstration of simple table syntax.
3873The header and table rows must each fit on one line. Column
3874alignments are determined by the position of the header text relative
3875to the dashed line below it:[^4]
3877 - If the dashed line is flush with the header text on the right side
3878 but extends beyond it on the left, the column is right-aligned.
3879 - If the dashed line is flush with the header text on the left side
3880 but extends beyond it on the right, the column is left-aligned.
3881 - If the dashed line extends beyond the header text on both sides,
3882 the column is centered.
3883 - If the dashed line is flush with the header text on both sides,
3884 the default alignment is used (in most cases, this will be left).
3886[^4]: This scheme is due to Michel Fortin, who proposed it on the
3887 [Markdown discussion list](http://six.pairlist.net/pipermail/markdown-discuss/2005-March/001097.html).
3889The table must end with a blank line, or a line of dashes followed by
3892The column header row may be omitted, provided a dashed line is used
3893to end the table. For example:
3895 ------- ------ ---------- -------
3899 ------- ------ ---------- -------
3901When the header row is omitted, column alignments are determined on the basis
3902of the first line of the table body. So, in the tables above, the columns
3903would be right, left, center, and right aligned, respectively.
3905#### Extension: `multiline_tables` ####
3907Multiline tables allow header and table rows to span multiple lines
3908of text (but cells that span multiple columns or rows of the table are
3909not supported). Here is an example:
3911 -------------------------------------------------------------
3912 Centered Default Right Left
3913 Header Aligned Aligned Aligned
3914 ----------- ------- --------------- -------------------------
3915 First row 12.0 Example of a row that
3916 spans multiple lines.
3918 Second row 5.0 Here's another one. Note
3919 the blank line between
3921 -------------------------------------------------------------
3923 Table: Here's the caption. It, too, may span
3926These work like simple tables, but with the following differences:
3928 - They must begin with a row of dashes, before the header text
3929 (unless the header row is omitted).
3930 - They must end with a row of dashes, then a blank line.
3931 - The rows must be separated by blank lines.
3933In multiline tables, the table parser pays attention to the widths of
3934the columns, and the writers try to reproduce these relative widths in
3935the output. So, if you find that one of the columns is too narrow in the
3936output, try widening it in the Markdown source.
3938The header may be omitted in multiline tables as well as simple tables:
3940 ----------- ------- --------------- -------------------------
3941 First row 12.0 Example of a row that
3942 spans multiple lines.
3944 Second row 5.0 Here's another one. Note
3945 the blank line between
3947 ----------- ------- --------------- -------------------------
3949 : Here's a multiline table without a header.
3951It is possible for a multiline table to have just one row, but the row
3952should be followed by a blank line (and then the row of dashes that ends
3953the table), or the table may be interpreted as a simple table.
3955#### Extension: `grid_tables` ####
3957Grid tables look like this:
3959 : Sample grid table.
3961 +---------------+---------------+--------------------+
3962 | Fruit | Price | Advantages |
3963 +===============+===============+====================+
3964 | Bananas | $1.34 | - built-in wrapper |
3965 | | | - bright color |
3966 +---------------+---------------+--------------------+
3967 | Oranges | $2.10 | - cures scurvy |
3969 +---------------+---------------+--------------------+
3971The row of `=`s separates the header from the table body, and can be
3972omitted for a headerless table. The cells of grid tables may contain
3973arbitrary block elements (multiple paragraphs, code blocks, lists,
3974etc.). Cells that span multiple columns or rows are not
3975supported. Grid tables can be created easily using Emacs' table-mode
3976(`M-x table-insert`).
3978Alignments can be specified as with pipe tables, by putting
3979colons at the boundaries of the separator line after the
3982 +---------------+---------------+--------------------+
3983 | Right | Left | Centered |
3984 +==============:+:==============+:==================:+
3985 | Bananas | $1.34 | built-in wrapper |
3986 +---------------+---------------+--------------------+
3988For headerless tables, the colons go on the top line instead:
3990 +--------------:+:--------------+:------------------:+
3991 | Right | Left | Centered |
3992 +---------------+---------------+--------------------+
3994##### Grid Table Limitations #####
3996Pandoc does not support grid tables with row spans or column spans.
3997This means that neither variable numbers of columns across rows nor
3998variable numbers of rows across columns are supported by Pandoc.
3999All grid tables must have the same number of columns in each row,
4000and the same number of rows in each column. For example, the
4001Docutils [sample grid tables] will not render as expected with
4004[sample grid tables]: https://docutils.sourceforge.io/docs/ref/rst/restructuredtext.html#grid-tables
4007#### Extension: `pipe_tables` ####
4009Pipe tables look like this:
4011 | Right | Left | Default | Center |
4012 |------:|:-----|---------|:------:|
4013 | 12 | 12 | 12 | 12 |
4014 | 123 | 123 | 123 | 123 |
4017 : Demonstration of pipe table syntax.
4019The syntax is identical to [PHP Markdown Extra tables]. The beginning and
4020ending pipe characters are optional, but pipes are required between all
4021columns. The colons indicate column alignment as shown. The header
4022cannot be omitted. To simulate a headerless table, include a header
4025Since the pipes indicate column boundaries, columns need not be vertically
4026aligned, as they are in the above example. So, this is a perfectly
4027legal (though ugly) pipe table:
4035The cells of pipe tables cannot contain block elements like paragraphs
4036and lists, and cannot span multiple lines. If a pipe table contains a
4037row whose Markdown content is wider than the column width (see
4038`--columns`), then the table will take up the full text width and
4039the cell contents will wrap, with the relative cell widths determined
4040by the number of dashes in the line separating the table header from
4041the table body. (For example `---|-` would make the first column 3/4
4042and the second column 1/4 of the full text width.)
4043On the other hand, if no lines are wider than column width, then
4044cell contents will not be wrapped, and the cells will be sized
4047Note: pandoc also recognizes pipe tables of the following
4048form, as can be produced by Emacs' orgtbl-mode:
4055The difference is that `+` is used instead of `|`. Other orgtbl features
4056are not supported. In particular, to get non-default column alignment,
4057you'll need to add colons as above.
4059[PHP Markdown Extra tables]: https://michelf.ca/projects/php-markdown/extra/#table
4063#### Extension: `pandoc_title_block` ####
4065If the file begins with a title block
4068 % author(s) (separated by semicolons)
4071it will be parsed as bibliographic information, not regular text. (It
4072will be used, for example, in the title of standalone LaTeX or HTML
4073output.) The block may contain just a title, a title and an author,
4074or all three elements. If you want to include an author but no
4075title, or a title and a date but no author, you need a blank line:
4088The title may occupy multiple lines, but continuation lines must
4089begin with leading space, thus:
4096If a document has multiple authors, the authors may be put on
4097separate lines with leading space, or separated by semicolons, or
4098both. So, all of the following are equivalent:
4106% Author One; Author Two
4114The date must fit on one line.
4116All three metadata fields may contain standard inline formatting
4117(italics, links, footnotes, etc.).
4119Title blocks will always be parsed, but they will affect the output only
4120when the `--standalone` (`-s`) option is chosen. In HTML output, titles
4121will appear twice: once in the document head -- this is the title that
4122will appear at the top of the window in a browser -- and once at the
4123beginning of the document body. The title in the document head can have
4124an optional prefix attached (`--title-prefix` or `-T` option). The title
4125in the body appears as an H1 element with class "title", so it can be
4126suppressed or reformatted with CSS. If a title prefix is specified with
4127`-T` and no title block appears in the document, the title prefix will
4128be used by itself as the HTML title.
4130The man page writer extracts a title, man page section number, and
4131other header and footer information from the title line. The title
4132is assumed to be the first word on the title line, which may optionally
4133end with a (single-digit) section number in parentheses. (There should
4134be no space between the title and the parentheses.) Anything after
4135this is assumed to be additional footer and header text. A single pipe
4136character (`|`) should be used to separate the footer text from the header
4141will yield a man page with the title `PANDOC` and section 1.
4143 % PANDOC(1) Pandoc User Manuals
4145will also have "Pandoc User Manuals" in the footer.
4147 % PANDOC(1) Pandoc User Manuals | Version 4.0
4149will also have "Version 4.0" in the header.
4151#### Extension: `yaml_metadata_block` ####
4153A [YAML] metadata block is a valid YAML object, delimited by a line of three
4154hyphens (`---`) at the top and a line of three hyphens (`---`) or three dots
4155(`...`) at the bottom. A YAML metadata block may occur anywhere in the
4156document, but if it is not at the beginning, it must be preceded by a blank
4157line. (Note that, because of the way pandoc concatenates input files when
4158several are provided, you may also keep the metadata in a separate YAML file
4159and pass it to pandoc as an argument, along with your Markdown files:
4161 pandoc chap1.md chap2.md chap3.md metadata.yaml -s -o book.html
4163Just be sure that the YAML file begins with `---` and ends with `---` or
4164`...`.) Alternatively, you can use the `--metadata-file` option. Using
4165that approach however, you cannot reference content (like footnotes)
4166from the main markdown input document.
4168Metadata will be taken from the fields of the YAML object and added to any
4169existing document metadata. Metadata can contain lists and objects (nested
4170arbitrarily), but all string scalars will be interpreted as Markdown. Fields
4171with names ending in an underscore will be ignored by pandoc. (They may be
4172given a role by external processors.) Field names must not be
4173interpretable as YAML numbers or boolean values (so, for
4174example, `yes`, `True`, and `15` cannot be used as field names).
4176A document may contain multiple metadata blocks. If two
4177metadata blocks attempt to set the same field, the value from
4178the second block will be taken.
4180When pandoc is used with `-t markdown` to create a Markdown document,
4181a YAML metadata block will be produced only if the `-s/--standalone`
4182option is used. All of the metadata will appear in a single block
4183at the beginning of the document.
4185Note that [YAML] escaping rules must be followed. Thus, for example,
4186if a title contains a colon, it must be quoted. The pipe character
4187(`|`) can be used to begin an indented block that will be interpreted
4188literally, without need for escaping. This form is necessary
4189when the field contains blank lines or block-level formatting:
4192 title: 'This is the title: it contains a colon'
4196 keywords: [nothing, nothingness]
4198 This is the abstract.
4200 It consists of two paragraphs.
4203Template variables will be set automatically from the metadata. Thus, for
4204example, in writing HTML, the variable `abstract` will be set to the HTML
4205equivalent of the Markdown in the `abstract` field:
4207 <p>This is the abstract.</p>
4208 <p>It consists of two paragraphs.</p>
4210Variables can contain arbitrary YAML structures, but the template must match
4211this structure. The `author` variable in the default templates expects a
4212simple list or string, but can be changed to support more complicated
4213structures. The following combination, for example, would add an affiliation
4214to the author if one is given:
4217 title: The document title
4220 affiliation: University of Somewhere
4222 affiliation: University of Nowhere
4225To use the structured authors in the example above, you would need a custom
4230 $author.name$$if(author.affiliation)$ ($author.affiliation$)$endif$
4236Raw content to include in the document's header may be specified
4237using `header-includes`; however, it is important to mark up
4238this content as raw code for a particular output format, using
4239the [`raw_attribute` extension](#extension-raw_attribute)), or it
4240will be interpreted as markdown. For example:
4245 \let\oldsection\section
4246 \renewcommand{\section}[1]{\clearpage\oldsection{#1}}
4251#### Extension: `all_symbols_escapable` ####
4253Except inside a code block or inline code, any punctuation or space
4254character preceded by a backslash will be treated literally, even if it
4255would normally indicate formatting. Thus, for example, if one writes
4265 <strong>hello</strong>
4267This rule is easier to remember than standard Markdown's rule,
4268which allows only the following characters to be backslash-escaped:
4272(However, if the `markdown_strict` format is used, the standard Markdown rule
4275A backslash-escaped space is parsed as a nonbreaking space. In TeX output,
4276it will appear as `~`. In HTML and XML output, it will appear as a
4277literal unicode nonbreaking space character (note that it will thus
4278actually look "invisible" in the generated HTML source; you can still
4279use the `--ascii` command-line option to make it appear as an explicit
4282A backslash-escaped newline (i.e. a backslash occurring at the end of
4283a line) is parsed as a hard line break. It will appear in TeX output as
4284`\\` and in HTML as `<br />`. This is a nice alternative to
4285Markdown's "invisible" way of indicating hard line breaks using
4286two trailing spaces on a line.
4288Backslash escapes do not work in verbatim contexts.
4294To *emphasize* some text, surround it with `*`s or `_`, like this:
4296 This text is _emphasized with underscores_, and this
4297 is *emphasized with asterisks*.
4299Double `*` or `_` produces **strong emphasis**:
4301 This is **strong emphasis** and __with underscores__.
4303A `*` or `_` character surrounded by spaces, or backslash-escaped,
4304will not trigger emphasis:
4306 This is * not emphasized *, and \*neither is this\*.
4308#### Extension: `intraword_underscores` ####
4310Because `_` is sometimes used inside words and identifiers,
4311pandoc does not interpret a `_` surrounded by alphanumeric
4312characters as an emphasis marker. If you want to emphasize
4313just part of a word, use `*`:
4315 feas*ible*, not feas*able*.
4320#### Extension: `strikeout` ####
4322To strikeout a section of text with a horizontal line, begin and end it
4323with `~~`. Thus, for example,
4325 This ~~is deleted text.~~
4328### Superscripts and subscripts ###
4330#### Extension: `superscript`, `subscript` ####
4332Superscripts may be written by surrounding the superscripted text by `^`
4333characters; subscripts may be written by surrounding the subscripted
4334text by `~` characters. Thus, for example,
4336 H~2~O is a liquid. 2^10^ is 1024.
4338The text between `^...^` or `~...~` may not contain spaces or
4339newlines. If the superscripted or subscripted text contains
4340spaces, these spaces must be escaped with backslashes. (This is
4341to prevent accidental superscripting and subscripting through
4342the ordinary use of `~` and `^`, and also bad interactions with
4343footnotes.) Thus, if you want the letter P with 'a cat' in
4344subscripts, use `P~a\ cat~`, not `P~a cat~`.
4348To make a short span of text verbatim, put it inside backticks:
4350 What is the difference between `>>=` and `>>`?
4352If the verbatim text includes a backtick, use double backticks:
4354 Here is a literal backtick `` ` ``.
4356(The spaces after the opening backticks and before the closing
4357backticks will be ignored.)
4359The general rule is that a verbatim span starts with a string
4360of consecutive backticks (optionally followed by a space)
4361and ends with a string of the same number of backticks (optionally
4362preceded by a space).
4364Note that backslash-escapes (and other Markdown constructs) do not
4365work in verbatim contexts:
4367 This is a backslash followed by an asterisk: `\*`.
4369#### Extension: `inline_code_attributes` ####
4371Attributes can be attached to verbatim text, just as with
4372[fenced code blocks]:
4378To write small caps, use the `smallcaps` class:
4380 [Small caps]{.smallcaps}
4382Or, without the `bracketed_spans` extension:
4384 <span class="smallcaps">Small caps</span>
4386For compatibility with other Markdown flavors, CSS is also supported:
4388 <span style="font-variant:small-caps;">Small caps</span>
4390This will work in all output formats that support small caps.
4395#### Extension: `tex_math_dollars` ####
4397Anything between two `$` characters will be treated as TeX math. The
4398opening `$` must have a non-space character immediately to its right,
4399while the closing `$` must have a non-space character immediately to its
4400left, and must not be followed immediately by a digit. Thus,
4401`$20,000 and $30,000` won't parse as math. If for some reason
4402you need to enclose text in literal `$` characters, backslash-escape
4403them and they won't be treated as math delimiters.
4405For display math, use `$$` delimiters. (In this case, the delimiters
4406may be separated from the formula by whitespace. However, there can be
4407no blank lines betwen the opening and closing `$$` delimiters.)
4409TeX math will be printed in all output formats. How it is rendered
4410depends on the output format:
4413 ~ It will appear verbatim surrounded by `\(...\)` (for inline
4414 math) or `\[...\]` (for display math).
4416Markdown, Emacs Org mode, ConTeXt, ZimWiki
4417 ~ It will appear verbatim surrounded by `$...$` (for inline
4418 math) or `$$...$$` (for display math).
4421 ~ It will appear verbatim surrounded by `{{formula}}..{{/formula}}`.
4424 ~ It will be rendered using an [interpreted text role `:math:`].
4427 ~ For AsciiDoc output format (`-t asciidoc`) it will appear verbatim
4428 surrounded by `latexmath:[$...$]` (for inline math) or
4429 `[latexmath]++++\[...\]+++` (for display math).
4430 For AsciiDoctor output format (`-t asciidoctor`) the LaTex delimiters
4431 (`$..$` and `\[..\]`) are omitted.
4434 ~ It will be rendered inside a `@math` command.
4436roff man, Jira markup
4437 ~ It will be rendered verbatim without `$`'s.
4440 ~ It will be rendered inside `<math>` tags.
4443 ~ It will be rendered inside `<span class="math">` tags.
4446 ~ It will be rendered, if possible, using Unicode characters,
4447 and will otherwise appear verbatim.
4450 ~ It will be rendered, if possible, using MathML.
4453 ~ If the `--mathml` flag is used, it will be rendered using MathML
4454 in an `inlineequation` or `informalequation` tag. Otherwise it
4455 will be rendered, if possible, using Unicode characters.
4458 ~ It will be rendered using OMML math markup.
4461 ~ If the `--webtex` option is used, formulas are rendered as images
4462 using CodeCogs or other compatible web service, downloaded
4463 and embedded in the e-book. Otherwise, they will appear verbatim.
4465HTML, Slidy, DZSlides, S5, EPUB
4466 ~ The way math is rendered in HTML will depend on the
4467 command-line options selected. Therefore see [Math rendering in HTML]
4470[interpreted text role `:math:`]: https://docutils.sourceforge.io/docs/ref/rst/roles.html#math
4474#### Extension: `raw_html` ####
4476Markdown allows you to insert raw HTML (or DocBook) anywhere in a document
4477(except verbatim contexts, where `<`, `>`, and `&` are interpreted
4478literally). (Technically this is not an extension, since standard
4479Markdown allows it, but it has been made an extension so that it can
4480be disabled if desired.)
4482The raw HTML is passed through unchanged in HTML, S5, Slidy, Slideous,
4483DZSlides, EPUB, Markdown, CommonMark, Emacs Org mode, and Textile
4484output, and suppressed in other formats.
4486For a more explicit way of including raw HTML in a Markdown
4487document, see the [`raw_attribute` extension][Extension: `raw_attribute`].
4489In the CommonMark format, if `raw_html` is enabled, superscripts,
4490subscripts, strikeouts and small capitals will be represented as HTML.
4491Otherwise, plain-text fallbacks will be used. Note that even if
4492`raw_html` is disabled, tables will be rendered with HTML syntax if
4493they cannot use pipe syntax.
4495#### Extension: `markdown_in_html_blocks` ####
4497Standard Markdown allows you to include HTML "blocks": blocks
4498of HTML between balanced tags that are separated from the surrounding text
4499with blank lines, and start and end at the left margin. Within
4500these blocks, everything is interpreted as HTML, not Markdown;
4501so (for example), `*` does not signify emphasis.
4503Pandoc behaves this way when the `markdown_strict` format is used; but
4504by default, pandoc interprets material between HTML block tags as Markdown.
4505Thus, for example, pandoc will turn
4510 <td>[a link](https://google.com)</td>
4518 <td><em>one</em></td>
4519 <td><a href="https://google.com">a link</a></td>
4523whereas `Markdown.pl` will preserve it as is.
4525There is one exception to this rule: text between `<script>`,
4526`<style>`, and `<textarea>` tags is not interpreted as Markdown.
4528This departure from standard Markdown should make it easier to mix
4529Markdown with HTML block elements. For example, one can surround
4530a block of Markdown text with `<div>` tags without preventing it
4531from being interpreted as Markdown.
4533#### Extension: `native_divs` ####
4535Use native pandoc `Div` blocks for content inside `<div>` tags.
4536For the most part this should give the same output as
4537`markdown_in_html_blocks`, but it makes it easier to write pandoc
4538filters to manipulate groups of blocks.
4540#### Extension: `native_spans` ####
4542Use native pandoc `Span` blocks for content inside `<span>` tags.
4543For the most part this should give the same output as `raw_html`,
4544but it makes it easier to write pandoc filters to manipulate groups
4547#### Extension: `raw_tex` ####
4549In addition to raw HTML, pandoc allows raw LaTeX, TeX, and ConTeXt to be
4550included in a document. Inline TeX commands will be preserved and passed
4551unchanged to the LaTeX and ConTeXt writers. Thus, for example, you can use
4552LaTeX to include BibTeX citations:
4554 This result was proved in \cite{jones.1967}.
4556Note that in LaTeX environments, like
4558 \begin{tabular}{|l|l|}\hline
4559 Age & Frequency \\ \hline
4562 36--45 & 22 \\ \hline
4565the material between the begin and end tags will be interpreted as raw
4566LaTeX, not as Markdown.
4568For a more explicit and flexible way of including raw TeX in a
4569Markdown document, see the [`raw_attribute`
4570extension][Extension: `raw_attribute`].
4572Inline LaTeX is ignored in output formats other than Markdown, LaTeX,
4573Emacs Org mode, and ConTeXt.
4575### Generic raw attribute ###
4577#### Extension: `raw_attribute` ####
4579Inline spans and fenced code blocks with a special
4580kind of attribute will be parsed as raw content with the
4581designated format. For example, the following produces a raw
4588And the following produces a raw `html` inline element:
4590 This is `<a>html</a>`{=html}
4592This can be useful to insert raw xml into `docx` documents, e.g.
4598 <w:br w:type="page"/>
4603The format name should match the target format name (see
4604`-t/--to`, above, for a list, or use `pandoc
4605--list-output-formats`). Use `openxml` for `docx` output,
4606`opendocument` for `odt` output, `html5` for `epub3` output,
4607`html4` for `epub2` output, and `latex`, `beamer`,
4608`ms`, or `html5` for `pdf` output (depending on what you
4609use for `--pdf-engine`).
4611This extension presupposes that the relevant kind of
4612inline code or fenced code block is enabled. Thus, for
4613example, to use a raw attribute with a backtick code block,
4614`backtick_code_blocks` must be enabled.
4616The raw attribute cannot be combined with regular attributes.
4620#### Extension: `latex_macros` ####
4622When this extension is enabled, pandoc will parse LaTeX
4623macro definitions and apply the resulting macros to all LaTeX
4624math and raw LaTeX. So, for example, the following will work in
4625all output formats, not just LaTeX:
4627 \newcommand{\tuple}[1]{\langle #1 \rangle}
4631Note that LaTeX macros will not be applied if they occur
4632inside a raw span or block marked with the
4633[`raw_attribute` extension](#extension-raw_attribute).
4635When `latex_macros` is disabled, the raw LaTeX and math will
4636not have macros applied. This is usually a better approach when
4637you are targeting LaTeX or PDF.
4639Macro definitions in LaTeX will be passed through as raw LaTeX
4640only if `latex_macros` is not enabled. Macro definitions in
4641Markdown source (or other formats allowing `raw_tex`) will
4642be passed through regardless of whether `latex_macros` is
4647Markdown allows links to be specified in several ways.
4649### Automatic links ###
4651If you enclose a URL or email address in pointy brackets, it
4654 <https://google.com>
4655 <sam@green.eggs.ham>
4659An inline link consists of the link text in square brackets,
4660followed by the URL in parentheses. (Optionally, the URL can
4661be followed by a link title, in quotes.)
4663 This is an [inline link](/url), and here's [one with
4664 a title](https://fsf.org "click here for a good time!").
4666There can be no space between the bracketed part and the parenthesized part.
4667The link text can contain formatting (such as emphasis), but the title cannot.
4669Email addresses in inline links are not autodetected, so they have to be
4670prefixed with `mailto`:
4672 [Write me!](mailto:sam@green.eggs.ham)
4674### Reference links ###
4676An *explicit* reference link has two parts, the link itself and the link
4677definition, which may occur elsewhere in the document (either
4678before or after the link).
4680The link consists of link text in square brackets, followed by a label in
4681square brackets. (There cannot be space between the two unless the
4682`spaced_reference_links` extension is enabled.) The link definition
4683consists of the bracketed label, followed by a colon and a space, followed by
4684the URL, and optionally (after a space) a link title either in quotes or in
4685parentheses. The label must not be parseable as a citation (assuming
4686the `citations` extension is enabled): citations take precedence over
4689Here are some examples:
4691 [my label 1]: /foo/bar.html "My title, optional"
4693 [my label 3]: https://fsf.org (The free software foundation)
4694 [my label 4]: /bar#special 'A title in single quotes'
4696The URL may optionally be surrounded by angle brackets:
4698 [my label 5]: <http://foo.bar.baz>
4700The title may go on the next line:
4702 [my label 3]: https://fsf.org
4703 "The free software foundation"
4705Note that link labels are not case sensitive. So, this will work:
4707 Here is [my link][FOO]
4711In an *implicit* reference link, the second pair of brackets is
4716 [my website]: http://foo.bar.baz
4718Note: In `Markdown.pl` and most other Markdown implementations,
4719reference link definitions cannot occur in nested constructions
4720such as list items or block quotes. Pandoc lifts this arbitrary
4721seeming restriction. So the following is fine in pandoc, though
4722not in most other implementations:
4728#### Extension: `shortcut_reference_links` ####
4730In a *shortcut* reference link, the second pair of brackets may
4735 [my website]: http://foo.bar.baz
4737### Internal links ###
4739To link to another section of the same document, use the automatically
4740generated identifier (see [Heading identifiers]). For example:
4742 See the [Introduction](#introduction).
4746 See the [Introduction].
4748 [Introduction]: #introduction
4750Internal links are currently supported for HTML formats (including
4751HTML slide shows and EPUB), LaTeX, and ConTeXt.
4755A link immediately preceded by a `!` will be treated as an image.
4756The link text will be used as the image's alt text:
4758 
4762 [movie reel]: movie.gif
4764#### Extension: `implicit_figures` ####
4766An image with nonempty alt text, occurring by itself in a
4767paragraph, will be rendered as a figure with a caption. The
4768image's alt text will be used as the caption.
4770 
4772How this is rendered depends on the output format. Some output
4773formats (e.g. RTF) do not yet support figures. In those
4774formats, you'll just get an image in a paragraph by itself, with
4777If you just want a regular inline image, just make sure it is not
4778the only thing in the paragraph. One way to do this is to insert a
4779nonbreaking space after the image:
4781 \
4783Note that in reveal.js slide shows, an image in a paragraph
4784by itself that has the `stretch` class will fill the screen,
4785and the caption and figure tags will be omitted.
4787#### Extension: `link_attributes` ####
4789Attributes can be set on links and images:
4791 An inline {#id .class width=30 height=20px}
4792 and a reference ![image][ref] with attributes.
4794 [ref]: foo.jpg "optional title" {#id .class key=val key2="val 2"}
4796(This syntax is compatible with [PHP Markdown Extra] when only `#id`
4797and `.class` are used.)
4799For HTML and EPUB, all known HTML5 attributes except `width` and
4800`height` (but including `srcset` and `sizes`) are passed through
4801as is. Unknown attributes are passed through as custom
4802attributes, with `data-` prepended. The other writers ignore
4803attributes that are not specifically supported by their output format.
4805The `width` and `height` attributes on images are treated specially. When
4806used without a unit, the unit is assumed to be pixels. However, any of
4807the following unit identifiers can be used: `px`, `cm`, `mm`, `in`, `inch`
4808and `%`. There must not be any spaces between the number and the unit.
4812{ width=50% }
4815- Dimensions may be converted to a form that is compatible with
4816 the output format (for example, dimensions given in pixels will
4817 be converted to inches when converting HTML to LaTeX). Conversion
4818 between pixels and physical measurements is affected by the
4819 `--dpi` option (by default, 96 dpi is assumed, unless the image
4820 itself contains dpi information).
4821- The `%` unit is generally relative to some available space.
4822 For example the above example will render to the following.
4823 - HTML: `<img href="file.jpg" style="width: 50%;" />`
4824 - LaTeX: `\includegraphics[width=0.5\textwidth,height=\textheight]{file.jpg}`
4825 (If you're using a custom template, you need to configure `graphicx`
4826 as in the default template.)
4827 - ConTeXt: `\externalfigure[file.jpg][width=0.5\textwidth]`
4828- Some output formats have a notion of a class
4829 ([ConTeXt](https://wiki.contextgarden.net/Using_Graphics#Multiple_Image_Settings))
4830 or a unique identifier (LaTeX `\caption`), or both (HTML).
4831- When no `width` or `height` attributes are specified, the fallback
4832 is to look at the image resolution and the dpi metadata embedded in
4837Using the `native_divs` and `native_spans` extensions
4838(see [above][Extension: `native_divs`]), HTML syntax can
4839be used as part of markdown to create native `Div` and `Span`
4840elements in the pandoc AST (as opposed to raw HTML).
4841However, there is also nicer syntax available:
4843#### Extension: `fenced_divs` ####
4845Allow special fenced syntax for native `Div` blocks. A Div
4846starts with a fence containing at least three consecutive
4847colons plus some attributes. The attributes may optionally
4848be followed by another string of consecutive colons.
4849The attribute syntax is exactly as in fenced code blocks (see
4850[Extension: `fenced_code_attributes`]). As with fenced
4851code blocks, one can use either attributes in curly braces
4852or a single unbraced word, which will be treated as a class
4853name. The Div ends with another line containing a string of at
4854least three consecutive colons. The fenced Div should be
4855separated by blank lines from preceding and following blocks.
4859 ::::: {#special .sidebar}
4860 Here is a paragraph.
4865Fenced divs can be nested. Opening fences are distinguished
4866because they *must* have attributes:
4872 This is a warning within a warning.
4876Fences without attributes are always closing fences. Unlike
4877with fenced code blocks, the number of colons in the closing
4878fence need not match the number in the opening fence. However,
4879it can be helpful for visual clarity to use fences of different
4880lengths to distinguish nested divs from their parents.
4883#### Extension: `bracketed_spans` ####
4885A bracketed sequence of inlines, as one would use to begin
4886a link, will be treated as a `Span` with attributes if it is
4887followed immediately by attributes:
4889 [This is *some text*]{.class key="val"}
4893#### Extension: `footnotes` ####
4895Pandoc's Markdown allows footnotes, using the following syntax:
4897 Here is a footnote reference,[^1] and another.[^longnote]
4899 [^1]: Here is the footnote.
4901 [^longnote]: Here's one with multiple blocks.
4903 Subsequent paragraphs are indented to show that they
4904 belong to the previous footnote.
4908 The whole paragraph can be indented, or just the first
4909 line. In this way, multi-paragraph footnotes work like
4910 multi-paragraph list items.
4912 This paragraph won't be part of the note, because it
4915The identifiers in footnote references may not contain spaces, tabs,
4916or newlines. These identifiers are used only to correlate the
4917footnote reference with the note itself; in the output, footnotes
4918will be numbered sequentially.
4920The footnotes themselves need not be placed at the end of the
4921document. They may appear anywhere except inside other block elements
4922(lists, block quotes, tables, etc.). Each footnote should be
4923separated from surrounding content (including other footnotes)
4926#### Extension: `inline_notes` ####
4928Inline footnotes are also allowed (though, unlike regular notes,
4929they cannot contain multiple paragraphs). The syntax is as follows:
4931 Here is an inline note.^[Inlines notes are easier to write, since
4932 you don't have to pick an identifier and move down to type the
4935Inline and regular footnotes may be mixed freely.
4939#### Extension: `citations` ####
4941Markdown citations go inside square brackets and are separated
4942by semicolons. Each citation must have a key, composed of '@' +
4943the citation identifier from the database, and may optionally
4944have a prefix, a locator, and a suffix. The citation key must
4945begin with a letter, digit, or `_`, and may contain
4946alphanumerics, `_`, and internal punctuation characters
4947(`:.#$%&-+?<>~/`). Here are some examples:
4949 Blah blah [see @doe99, pp. 33-35; also @smith04, chap. 1].
4951 Blah blah [@doe99, pp. 33-35, 38-39 and *passim*].
4953 Blah blah [@smith04; @doe99].
4955`pandoc` detects locator terms in the [CSL locale files].
4956Either abbreviated or unabbreviated forms are accepted. In the `en-US`
4957locale, locator terms can be written in either singular or plural forms,
4958as `book`, `bk.`/`bks.`; `chapter`, `chap.`/`chaps.`; `column`,
4959`col.`/`cols.`; `figure`, `fig.`/`figs.`; `folio`, `fol.`/`fols.`;
4960`number`, `no.`/`nos.`; `line`, `l.`/`ll.`; `note`, `n.`/`nn.`; `opus`,
4961`op.`/`opp.`; `page`, `p.`/`pp.`; `paragraph`, `para.`/`paras.`; `part`,
4962`pt.`/`pts.`; `section`, `sec.`/`secs.`; `sub verbo`, `s.v.`/`s.vv.`;
4963`verse`, `v.`/`vv.`; `volume`, `vol.`/`vols.`; `¶`/`¶¶`; `§`/`§§`. If no
4964locator term is used, "page" is assumed.
4966`pandoc` will use heuristics to distinguish the locator
4967from the suffix. In complex cases, the locator can be enclosed
4970 [@smith{ii, A, D-Z}, with a suffix]
4971 [@smith, {pp. iv, vi-xi, (xv)-(xvii)} with suffix here]
4973A minus sign (`-`) before the `@` will suppress mention of
4974the author in the citation. This can be useful when the
4975author is already mentioned in the text:
4977 Smith says blah [-@smith04].
4979You can also write an in-text citation, as follows:
4983 @smith04 [p. 33] says blah.
4986[CSL]: https://docs.citationstyles.org/en/stable/specification.html
4987[CSL markup specs]: https://docs.citationstyles.org/en/1.0/release-notes.html#rich-text-markup-within-fields
4988[Chicago Manual of Style]: https://chicagomanualofstyle.org
4989[Citation Style Language]: https://citationstyles.org
4990[Zotero Style Repository]: https://www.zotero.org/styles
4991[finding and editing styles]: https://citationstyles.org/authors/
4992[CSL locale files]: https://github.com/citation-style-language/locales
4994## Non-pandoc extensions
4996The following Markdown syntax extensions are not enabled by default
4997in pandoc, but may be enabled by adding `+EXTENSION` to the format
4998name, where `EXTENSION` is the name of the extension. Thus, for
4999example, `markdown+hard_line_breaks` is Markdown with hard line breaks.
5001#### Extension: `attributes` ####
5003Allows attributes to be attached to any inline or block-level
5004element. The syntax for the attributes is the same as that
5005used in [`header_attributes`][Extension: `header_attributes`].
5007- Attributes that occur immediately after an inline
5008 element affect that element. If they follow a space, then they
5009 belong to the space. (Hence, this option subsumes
5010 `inline_code_attributes` and `link_attributes`.)
5011- Attributes that occur immediately before a block
5012 element, on a line by themselves, affect that
5014- Consecutive attribute specifiers may be used,
5015 either for blocks or for inlines. Their attributes
5017- Attributes that occur at the end of the text of
5018 a Setext or ATX heading (separated by whitespace
5019 from the text) affect the heading element. (Hence, this
5020 option subsumes `header_attributes`.)
5021- Attributes that occur after the opening fence
5022 in a fenced code block affect the code block element. (Hence,
5023 this option subsumes `fenced_code_attributes`.)
5024- Attributes that occur at the end of a reference
5025 link definition affect links that refer to that
5028Note that pandoc's AST does not currently allow attributes
5029to be attached to arbitrary elements. Hence a Span or Div
5030container will be added if needed.
5032#### Extension: `old_dashes` ####
5034Selects the pandoc <= 1.8.2.1 behavior for parsing smart dashes:
5035`-` before a numeral is an en-dash, and `--` is an em-dash.
5036This option only has an effect if `smart` is enabled. It is
5037selected automatically for `textile` input.
5039#### Extension: `angle_brackets_escapable` ####
5041Allow `<` and `>` to be backslash-escaped, as they can be in
5042GitHub flavored Markdown but not original Markdown. This is
5043implied by pandoc's default `all_symbols_escapable`.
5045#### Extension: `lists_without_preceding_blankline` ####
5047Allow a list to occur right after a paragraph, with no intervening
5050#### Extension: `four_space_rule` ####
5052Selects the pandoc <= 2.0 behavior for parsing lists, so that
5053four spaces indent are needed for list item continuation
5056#### Extension: `spaced_reference_links` ####
5058Allow whitespace between the two components of a reference link,
5063#### Extension: `hard_line_breaks` ####
5065Causes all newlines within a paragraph to be interpreted as hard line
5066breaks instead of spaces.
5068#### Extension: `ignore_line_breaks` ####
5070Causes newlines within a paragraph to be ignored, rather than being
5071treated as spaces or as hard line breaks. This option is intended for
5072use with East Asian languages where spaces are not used between words,
5073but text is divided into lines for readability.
5075#### Extension: `east_asian_line_breaks` ####
5077Causes newlines within a paragraph to be ignored, rather than
5078being treated as spaces or as hard line breaks, when they occur
5079between two East Asian wide characters. This is a better choice
5080than `ignore_line_breaks` for texts that include a mix of East
5081Asian wide characters and other characters.
5083#### Extension: `emoji` ####
5085Parses textual emojis like `:smile:` as Unicode emoticons.
5087#### Extension: `tex_math_single_backslash` ####
5089Causes anything between `\(` and `\)` to be interpreted as inline
5090TeX math, and anything between `\[` and `\]` to be interpreted
5091as display TeX math. Note: a drawback of this extension is that
5092it precludes escaping `(` and `[`.
5094#### Extension: `tex_math_double_backslash` ####
5096Causes anything between `\\(` and `\\)` to be interpreted as inline
5097TeX math, and anything between `\\[` and `\\]` to be interpreted
5100#### Extension: `markdown_attribute` ####
5102By default, pandoc interprets material inside block-level tags as Markdown.
5103This extension changes the behavior so that Markdown is only parsed
5104inside block-level tags if the tags have the attribute `markdown=1`.
5106#### Extension: `mmd_title_block` ####
5108Enables a [MultiMarkdown] style title block at the top of
5109the document, for example:
5113 Date: September 1, 2008
5114 Comment: This is a sample mmd title block, with
5115 a field spanning multiple lines.
5117See the MultiMarkdown documentation for details. If `pandoc_title_block` or
5118`yaml_metadata_block` is enabled, it will take precedence over
5121#### Extension: `abbreviations` ####
5123Parses PHP Markdown Extra abbreviation keys, like
5125 *[HTML]: Hypertext Markup Language
5127Note that the pandoc document model does not support
5128abbreviations, so if this extension is enabled, abbreviation keys are
5129simply skipped (as opposed to being parsed as paragraphs).
5131#### Extension: `autolink_bare_uris` ####
5133Makes all absolute URIs into links, even when not surrounded by
5134pointy braces `<...>`.
5136#### Extension: `mmd_link_attributes` ####
5138Parses multimarkdown style key-value attributes on link
5139and image references. This extension should not be confused with the
5140[`link_attributes`](#extension-link_attributes) extension.
5142 This is a reference ![image][ref] with multimarkdown attributes.
5144 [ref]: https://path.to/image "Image title" width=20px height=30px
5145 id=myId class="myClass1 myClass2"
5147#### Extension: `mmd_header_identifiers` ####
5149Parses multimarkdown style heading identifiers (in square brackets,
5150after the heading but before any trailing `#`s in an ATX heading).
5152#### Extension: `compact_definition_lists` ####
5154Activates the definition list syntax of pandoc 1.12.x and earlier.
5155This syntax differs from the one described above under [Definition lists]
5158 - No blank line is required between consecutive items of the
5160 - To get a "tight" or "compact" list, omit space between consecutive
5161 items; the space between a term and its definition does not affect
5163 - Lazy wrapping of paragraphs is not allowed: the entire definition must
5164 be indented four spaces.[^6]
5166[^6]: To see why laziness is incompatible with relaxing the requirement
5167 of a blank line between items, consider the following example:
5174 Is this a single list item with two definitions of "bar," the first of
5175 which is lazily wrapped, or two list items? To remove the ambiguity
5176 we must either disallow lazy wrapping or require a blank line between
5179#### Extension: `gutenberg` ####
5181Use [Project Gutenberg] conventions for `plain` output:
5182all-caps for strong emphasis, surround by underscores
5183for regular emphasis, add extra blank space around headings.
5185 [Project Gutenberg]: https://www.gutenberg.org
5187#### Extension: `sourcepos` ####
5189Include source position attributes when parsing `commonmark`.
5190For elements that accept attributes, a `data-pos` attribute
5191is added; other elements are placed in a surrounding
5192Div or Span elemnet with a `data-pos` attribute.
5196In addition to pandoc's extended Markdown, the following Markdown
5197variants are supported:
5199- `markdown_phpextra` (PHP Markdown Extra)
5200- `markdown_github` (deprecated GitHub-Flavored Markdown)
5201- `markdown_mmd` (MultiMarkdown)
5202- `markdown_strict` (Markdown.pl)
5203- `commonmark` (CommonMark)
5204- `gfm` (Github-Flavored Markdown)
5205- `commonmark_x` (CommonMark with many pandoc extensions)
5207To see which extensions are supported for a given format,
5208and which are enabled by default, you can use the command
5210 pandoc --list-extensions=FORMAT
5212where `FORMAT` is replaced with the name of the format.
5214Note that the list of extensions for `commonmark`,
5215`gfm`, and `commonmark_x` are defined relative to default
5216commonmark. So, for example, `backtick_code_blocks`
5217does not appear as an extension, since it is enabled by
5218default and cannot be disabled.
5222When the `--citeproc` option is used, pandoc can automatically generate
5223citations and a bibliography in a number of styles. Basic usage is
5225 pandoc --citeproc myinput.txt
5227To use this feature, you will need to have
5229- a document containing citations (see [Extension: `citations`]);
5230- a source of bibliographic data: either an external bibliography
5231 file or a list of `references` in the document's YAML metadata
5232- optionally, a [CSL] citation style.
5234## Specifying bibliographic data
5236You can specify an external bibliography using the
5237`bibliography` metadata field in a YAML metadata section or the
5238`--bibliography` command line argument. If you want to use
5239multiple bibliography files, you can supply multiple
5240`--bibliography` arguments or set `bibliography` metadata field
5241to YAML array. A bibliography may have any of these formats:
5243 Format File extension
5244 ------------ --------------
5250Note that `.bib` can be used with both BibTeX and BibLaTeX files;
5251use the extension `.bibtex` to force interpretation as BibTeX.
5253In BibTeX and BibLaTeX databases, pandoc parses LaTeX markup
5254inside fields such as `title`; in CSL YAML databases, pandoc
5255Markdown; and in CSL JSON databases, an [HTML-like markup][CSL
5264`<span style="font-variant:small-caps;">...</span>` or `<sc>...</sc>`
5273`<span class="nocase">...</span>`
5274: prevent a phrase from being capitalized as title case
5276As an alternative to specifying a bibliography file using
5277`--bibliography` or the YAML metadata field `bibliography`, you
5278can include the citation data directly in the `references` field
5279of the document's YAML metadata. The field should contain an
5280array of YAML-encoded references, for example:
5284 - type: article-journal
5296 title: 'Molecular structure of nucleic acids: a structure for
5297 deoxyribose nucleic acid'
5298 title-short: Molecular structure of nucleic acids
5299 container-title: Nature
5303 DOI: 10.1038/171737a0
5304 URL: https://www.nature.com/articles/171737a0
5308If both an external bibliography and inline (YAML metadata)
5309references are provided, both will be used. In case of
5310conflicting `id`s, the inline references will take precedence.
5312Note that `pandoc` can be used to produce such a YAML metadata
5313section from a BibTeX, BibLaTeX, or CSL JSON bibliography:
5315 pandoc chem.bib -s -f biblatex -t markdown
5316 pandoc chem.json -s -f csljson -t markdown
5318Indeed, `pandoc` can convert between any of these
5321 pandoc chem.bib -s -f biblatex -t csljson
5322 pandoc chem.yaml -s -f markdown -t biblatex
5324Running pandoc on a bibliography file with the `--citeproc`
5325option will create a formatted bibliography in the format
5328 pandoc chem.bib -s --citeproc -o chem.html
5329 pandoc chem.bib -s --citeproc -o chem.pdf
5331### Capitalization in titles
5333If you are using a bibtex or biblatex bibliography, then observe
5336 - English titles should be in title case. Non-English titles should
5337 be in sentence case, and the `langid` field in biblatex should be
5338 set to the relevant language. (The following values are treated
5339 as English: `american`, `british`, `canadian`, `english`,
5340 `australian`, `newzealand`, `USenglish`, or `UKenglish`.)
5342 - As is standard with bibtex/biblatex, proper names should be
5343 protected with curly braces so that they won't be lowercased
5344 in styles that call for sentence case. For example:
5346 title = {My Dinner with {Andre}}
5348 - In addition, words that should remain lowercase (or camelCase)
5349 should be protected:
5351 title = {Spin Wave Dispersion on the {nm} Scale}
5353 Though this is not necessary in bibtex/biblatex, it is necessary
5354 with citeproc, which stores titles internally in sentence case,
5355 and converts to title case in styles that require it. Here we
5356 protect "nm" so that it doesn't get converted to "Nm" at this stage.
5358If you are using a CSL bibliography (either JSON or YAML), then observe
5361 - All titles should be in sentence case.
5363 - Use the `language` field for non-English titles to prevent their
5364 conversion to title case in styles that call for this. (Conversion
5365 happens only if `language` begins with `en` or is left empty.)
5367 - Protect words that should not be converted to title case using
5370 Spin wave dispersion on the <span class="nocase">nm</span> scale
5372### Conference Papers, Published vs. Unpublished
5374For a formally published conference paper, use the biblatex entry type
5375`inproceedings` (which will be mapped to CSL `paper-conference`).
5377For an unpublished manuscript, use the biblatex entry type
5378`unpublished` without an `eventtitle` field (this entry type
5379will be mapped to CSL `manuscript`).
5381For a talk, an unpublished conference paper, or a poster
5382presentation, use the biblatex entry type `unpublished` with an
5383`eventtitle` field (this entry type will be mapped to CSL
5384`speech`). Use the biblatex `type` field to indicate the type,
5385e.g. "Paper", or "Poster". `venue` and `eventdate` may be useful
5386too, though `eventdate` will not be rendered by most CSL styles.
5387Note that `venue` is for the event's venue, unlike `location`
5388which describes the publisher's location; do not use the latter
5389for an unpublished conference paper.
5392## Specifying a citation style
5394Citations and references can be formatted using any style supported by the
5395[Citation Style Language], listed in the [Zotero Style Repository].
5396These files are specified using the `--csl` option or the `csl`
5397(or `citation-style`) metadata field. By default, pandoc will
5398use the [Chicago Manual of Style] author-date format. (You can
5399override this default by copying a CSL style of your choice
5400to `default.csl` in your user data directory.)
5401The CSL project provides further information on [finding and
5404The `--citation-abbreviations` option (or the
5405`citation-abbreviations` metadata field) may be used to specify
5406a JSON file containing abbreviations of journals that should be
5407used in formatted bibliographies when `form="short"` is
5408specified. The format of the file can be illustrated with an
5413 "container-title": {
5414 "Lloyd's Law Reports": "Lloyd's Rep",
5415 "Estates Gazette": "EG",
5416 "Scots Law Times": "SLT"
5422## Citations in note styles
5424Pandoc's citation processing is designed to allow you to
5425move between author-date, numerical, and note styles without
5426modifying the markdown source. When you're using a note
5427style, avoid inserting footnotes manually. Instead, insert
5428citations just as you would in an author-date style---for
5431 Blah blah [@foo, p. 33].
5433The footnote will be created automatically. Pandoc will take
5434care of removing the space and moving the note before or
5435after the period, depending on the setting of
5436`notes-after-punctuation`, as described below in [Other relevant
5439In some cases you may need to put a citation inside a regular
5440footnote. Normal citations in footnotes (such as `[@foo, p.
544133]`) will be rendered in parentheses. In-text citations (such
5442as `@foo [p. 33]`) will be rendered without parentheses. (A
5443comma will be added if appropriate.) Thus:
5445 [^1]: Some studies [@foo; @bar, p. 33] show that
5446 frubulicious zoosnaps are quantical. For a survey
5447 of the literature, see @baz [chap. 1].
5449## Raw content in a style
5451To include raw content in a prefix, suffix, delimiter, or term,
5452surround it with these tags indicating the format:
5454 {{jats}}<ref>{{/jats}}
5456Without the tags, the string will be interpreted as a string
5457and escaped in the output, rather than being passed through raw.
5459This feature allows stylesheets to be customized to give
5460different output for different output formats. However,
5461stylesheets customized in this way will not be usable
5462by other CSL implementations.
5466## Placement of the bibliography
5468If the style calls for a list of works cited, it will be placed
5469in a div with id `refs`, if one exists:
5474Otherwise, it will be placed at the end of the document.
5475Generation of the bibliography can be suppressed by setting
5476`suppress-bibliography: true` in the YAML metadata.
5478If you wish the bibliography to have a section heading, you can
5479set `reference-section-title` in the metadata, or put the heading
5480at the beginning of the div with id `refs` (if you are using it)
5481or at the end of your document:
5487The bibliography will be inserted after this heading. Note that
5488the `unnumbered` class will be added to this heading, so that the
5489section will not be numbered.
5491## Including uncited items in the bibliography
5493If you want to include items in the bibliography without actually
5494citing them in the body text, you can define a dummy `nocite` metadata
5495field and put the citations there:
5504In this example, the document will contain a citation for `item3`
5505only, but the bibliography will contain entries for `item1`, `item2`, and
5508It is possible to create a bibliography with all the citations,
5509whether or not they appear in the document, by using a wildcard:
5516For LaTeX output, you can also use [`natbib`] or [`biblatex`] to
5517render the bibliography. In order to do so, specify bibliography
5518files as outlined above, and add `--natbib` or `--biblatex`
5519argument to `pandoc` invocation. Bear in mind that bibliography
5520files have to be in either BibTeX (for `--natbib`)
5521or BibLaTeX (for `--biblatex`) format.
5523## Other relevant metadata fields
5525A few other metadata fields affect bibliography formatting:
5528: If true, citations will be
5529 hyperlinked to the corresponding bibliography entries
5530 (for author-date and numerical styles only).
5533: The `lang` field will affect how the style is localized,
5534 for example in the translation of labels and the use
5535 of quotation marks. (For backwards compatibility,
5536 `locale` may be used instead of `lang`, but this use
5539`notes-after-punctuation`
5540: If true (the default), pandoc will put footnote citations
5541 after following punctuation. For example, if the source
5542 contains `blah blah [@jones99].`, the result will look like
5543 `blah blah.[^1]`, with the note moved after the period and
5544 the space collapsed. If false, the space will still be
5545 collapsed, but the footnote will not be moved after the
5557You can use pandoc to produce an HTML + JavaScript slide presentation
5558that can be viewed via a web browser. There are five ways to do this,
5559using [S5], [DZSlides], [Slidy], [Slideous], or [reveal.js].
5560You can also produce a PDF slide show using LaTeX [`beamer`], or
5561slides shows in Microsoft [PowerPoint] format.
5563Here's the Markdown source for a simple slide show, `habits.txt`:
5590 
5597To produce an HTML/JavaScript slide show, simply type
5599 pandoc -t FORMAT -s habits.txt -o habits.html
5601where `FORMAT` is either `s5`, `slidy`, `slideous`, `dzslides`, or `revealjs`.
5603For Slidy, Slideous, reveal.js, and S5, the file produced by
5604pandoc with the `-s/--standalone` option embeds a link to
5605JavaScript and CSS files, which are assumed to be available at
5606the relative path `s5/default` (for S5), `slideous` (for
5607Slideous), `reveal.js` (for reveal.js), or at the Slidy website
5608at `w3.org` (for Slidy). (These paths can be changed by setting
5609the `slidy-url`, `slideous-url`, `revealjs-url`, or `s5-url`
5610variables; see [Variables for HTML slides], above.) For
5611DZSlides, the (relatively short) JavaScript and CSS are included
5612in the file by default.
5614With all HTML slide formats, the `--self-contained` option can
5615be used to produce a single file that contains all of the data
5616necessary to display the slide show, including linked scripts,
5617stylesheets, images, and videos.
5619To produce a PDF slide show using beamer, type
5621 pandoc -t beamer habits.txt -o habits.pdf
5623Note that a reveal.js slide show can also be converted to a PDF
5624by printing it to a file from the browser.
5626To produce a Powerpoint slide show, type
5628 pandoc habits.txt -o habits.pptx
5630## Structuring the slide show
5632By default, the *slide level* is the highest heading level in
5633the hierarchy that is followed immediately by content, and not another
5634heading, somewhere in the document. In the example above, level-1 headings
5635are always followed by level-2 headings, which are followed by content,
5636so the slide level is 2. This default can be overridden using
5637the `--slide-level` option.
5639The document is carved up into slides according to the following
5642 * A horizontal rule always starts a new slide.
5644 * A heading at the slide level always starts a new slide.
5646 * Headings *below* the slide level in the hierarchy create
5647 headings *within* a slide.
5649 * Headings *above* the slide level in the hierarchy create
5650 "title slides," which just contain the section title
5651 and help to break the slide show into sections.
5652 Non-slide content under these headings will be included
5653 on the title slide (for HTML slide shows) or in a
5654 subsequent slide with the same title (for beamer).
5656 * A title page is constructed automatically from the document's title
5657 block, if present. (In the case of beamer, this can be disabled
5658 by commenting out some lines in the default template.)
5660These rules are designed to support many different styles of slide show. If
5661you don't care about structuring your slides into sections and subsections,
5662you can just use level-1 headings for all each slide. (In that case, level-1
5663will be the slide level.) But you can also structure the slide show into
5664sections, as in the example above.
5666Note: in reveal.js slide shows, if slide level is 2, a two-dimensional
5667layout will be produced, with level-1 headings building horizontally
5668and level-2 headings building vertically. It is not recommended that
5669you use deeper nesting of section levels with reveal.js.
5673By default, these writers produce lists that display "all at once."
5674If you want your lists to display incrementally (one item at a time),
5675use the `-i` option. If you want a particular list to depart from the
5676default, put it in a `div` block with class `incremental` or
5677`nonincremental`. So, for example, using the `fenced div` syntax, the
5678following would be incremental regardless of the document default:
5696While using `incremental` and `nonincremental` divs are the
5697recommended method of setting incremental lists on a per-case basis,
5698an older method is also supported: putting lists inside a blockquote
5699will depart from the document default (that is, it will display
5700incrementally without the `-i` option and all at once with the `-i`
5706Both methods allow incremental and nonincremental lists to be mixed
5707in a single document.
5709Note: Neither the `-i/--incremental` option nor any of the
5710methods described here currently works for PowerPoint output.
5714You can add "pauses" within a slide by including a paragraph containing
5715three dots, separated by spaces:
5717 # Slide with a pause
5719 content before the pause
5723 content after the pause
5725Note: this feature is not yet implemented for PowerPoint output.
5727## Styling the slides
5729You can change the style of HTML slides by putting customized CSS files
5730in `$DATADIR/s5/default` (for S5), `$DATADIR/slidy` (for Slidy),
5731or `$DATADIR/slideous` (for Slideous),
5732where `$DATADIR` is the user data directory (see `--data-dir`, above).
5733The originals may be found in pandoc's system data directory (generally
5734`$CABALDIR/pandoc-VERSION/s5/default`). Pandoc will look there for any
5735files it does not find in the user data directory.
5737For dzslides, the CSS is included in the HTML file itself, and may
5740All [reveal.js configuration options] can be set through variables.
5741For example, themes can be used by setting the `theme` variable:
5745Or you can specify a custom stylesheet using the `--css` option.
5747To style beamer slides, you can specify a `theme`, `colortheme`,
5748`fonttheme`, `innertheme`, and `outertheme`, using the `-V` option:
5750 pandoc -t beamer habits.txt -V theme:Warsaw -o habits.pdf
5752Note that heading attributes will turn into slide attributes
5753(on a `<div>` or `<section>`) in HTML slide formats, allowing you
5754to style individual slides. In beamer, the only heading attribute
5755that affects slides is the `allowframebreaks` class, which sets the
5756`allowframebreaks` option, causing multiple slides to be created
5757if the content overfills the frame. This is recommended especially for
5760 # References {.allowframebreaks}
5764Speaker notes are supported in reveal.js and PowerPoint (pptx)
5765output. You can add notes to your Markdown document thus:
5771 - It can contain Markdown
5776To show the notes window in reveal.js, press `s` while viewing the
5777presentation. Speaker notes in PowerPoint will be available, as usual,
5778in handouts and presenter view.
5780Notes are not yet supported for other slide formats, but the notes
5781will not appear on the slides themselves.
5785To put material in side by side columns, you can use a native
5786div container with class `columns`, containing two or more div
5787containers with class `column` and a `width` attribute:
5789 :::::::::::::: {.columns}
5790 ::: {.column width="40%"}
5793 ::: {.column width="60%"}
5798### Additional columns attributes in beamer
5800The div containers with classes `columns` and `column` can optionally have
5801an `align` attribute.
5802The class `columns` can optionally have a `totalwidth` attribute or an
5803`onlytextwidth` class.
5805 :::::::::::::: {.columns align=center totalwidth=8em}
5806 ::: {.column width="40%"}
5809 ::: {.column width="60%" align=bottom}
5814The `align` attributes on `columns` and `column` can be used with the
5815values `top`, `top-baseline`, `center` and `bottom` to vertically align
5816the columns. It defaults to `top` in `columns`.
5818The `totalwidth` attribute limits the width of the columns to the given value.
5820 :::::::::::::: {.columns align=top .onlytextwidth}
5821 ::: {.column width="40%" align=center}
5824 ::: {.column width="60%"}
5829The class `onlytextwidth` sets the `totalwidth` to `\textwidth`.
5831See Section 12.7 of the [Beamer User's Guide] for more details.
5833## Frame attributes in beamer
5835Sometimes it is necessary to add the LaTeX `[fragile]` option to
5836a frame in beamer (for example, when using the `minted` environment).
5837This can be forced by adding the `fragile` class to the heading
5838introducing the slide:
5840 # Fragile slide {.fragile}
5842All of the other frame attributes described in Section 8.1 of
5843the [Beamer User's Guide] may also be used: `allowdisplaybreaks`,
5844`allowframebreaks`, `b`, `c`, `t`, `environment`, `label`, `plain`,
5845`shrink`, `standout`, `noframenumbering`.
5847## Background in reveal.js and beamer
5849Background images can be added to self-contained reveal.js slideshows and
5850to beamer slideshows.
5852For the same image on every slide, use the configuration
5853option `background-image` either in the YAML metadata block
5854or as a command-line variable. (There are no other options in
5855beamer and the rest of this section concerns reveal.js slideshows.)
5857For reveal.js, you can instead use the reveal.js-native option
5858`parallaxBackgroundImage`. You can also set `parallaxBackgroundHorizontal`
5859and `parallaxBackgroundVertical` the same way and must also set
5860`parallaxBackgroundSize` to have your values take effect.
5862To set an image for a particular reveal.js slide, add
5863`{data-background-image="/path/to/image"}`
5864to the first slide-level heading on the slide (which may even be empty).
5866In reveal.js's overview mode, the parallaxBackgroundImage will show up
5867only on the first slide.
5869Other reveal.js background settings also work on individual slides, including
5870`data-background-size`, `data-background-repeat`, `data-background-color`,
5871`data-transition`, and `data-transition-speed`.
5873To add a background image to the automatically generated title slide, use the
5874`title-slide-attributes` variable in the YAML metadata block. It must contain
5875a map of attribute names and values.
5877See the [reveal.js documentation](https://revealjs.com/backgrounds/) for more
5880For example in reveal.js:
5885parallaxBackgroundImage: /path/to/my/background_image.png
5886title-slide-attributes:
5887 data-background-image: /path/to/title_image.png
5888 data-background-size: contain
5893Slide 1 has background_image.png as its background.
5895## {data-background-image="/path/to/special_image.jpg"}
5897Slide 2 has a special image for its background, even though the heading has no content.
5904EPUB metadata may be specified using the `--epub-metadata` option, but
5905if the source document is Markdown, it is better to use a [YAML metadata
5906block][Extension: `yaml_metadata_block`]. Here is an example:
5913 text: An investigation of metadata
5921 text: doi:10.234234.234/33
5923 rights: © 2007 John Smith, CC BY-NC
5928The following fields are recognized:
5931 ~ Either a string value or an object with fields `text` and
5932 `scheme`. Valid values for `scheme` are `ISBN-10`,
5933 `GTIN-13`, `UPC`, `ISMN-10`, `DOI`, `LCCN`, `GTIN-14`,
5934 `ISBN-13`, `Legal deposit number`, `URN`, `OCLC`,
5935 `ISMN-13`, `ISBN-A`, `JP`, `OLCC`.
5938 ~ Either a string value, or an object with fields `file-as` and
5939 `type`, or a list of such objects. Valid values for `type` are
5940 `main`, `subtitle`, `short`, `collection`, `edition`, `extended`.
5943 ~ Either a string value, or an object with fields `role`, `file-as`,
5944 and `text`, or a list of such objects. Valid values for `role` are
5945 [MARC relators], but
5946 pandoc will attempt to translate the human-readable versions
5947 (like "author" and "editor") to the appropriate marc relators.
5950 ~ Same format as `creator`.
5953 ~ A string value in `YYYY-MM-DD` format. (Only the year is necessary.)
5954 Pandoc will attempt to convert other common date formats.
5956`lang` (or legacy: `language`)
5957 ~ A string value in [BCP 47] format. Pandoc will default to the local
5958 language if nothing is specified.
5961 ~ A string value or a list of such values.
5981`belongs-to-collection`
5982 ~ A string value. identifies the name of a collection to which
5983 the EPUB Publication belongs.
5986 ~ The `group-position` field indicates the numeric position in which
5987 the EPUB Publication belongs relative to other works belonging to
5988 the same `belongs-to-collection` field.
5991 ~ A string value (path to cover image).
5993`css` (or legacy: `stylesheet`)
5994 ~ A string value (path to CSS stylesheet).
5996`page-progression-direction`
5997 ~ Either `ltr` or `rtl`. Specifies the `page-progression-direction`
5998 attribute for the [`spine` element].
6001 ~ iBooks-specific metadata, with the following fields:
6003 - `version`: (string)
6004 - `specified-fonts`: `true`|`false` (default `false`)
6005 - `ipad-orientation-lock`: `portrait-only`|`landscape-only`
6006 - `iphone-orientation-lock`: `portrait-only`|`landscape-only`
6007 - `binding`: `true`|`false` (default `true`)
6008 - `scroll-axis`: `vertical`|`horizontal`|`default`
6010[MARC relators]: https://loc.gov/marc/relators/relaterm.html
6011[`spine` element]: http://idpf.org/epub/301/spec/epub-publications.html#sec-spine-elem
6013## The `epub:type` attribute
6015For `epub3` output, you can mark up the heading that corresponds to an EPUB
6016chapter using the [`epub:type` attribute][epub-type]. For example, to set
6017the attribute to the value `prologue`, use this markdown:
6019 # My chapter {epub:type=prologue}
6021Which will result in:
6023 <body epub:type="frontmatter">
6024 <section epub:type="prologue">
6027Pandoc will output `<body epub:type="bodymatter">`, unless
6028you use one of the following values, in which case either
6029`frontmatter` or `backmatter` will be output.
6031`epub:type` of first section `epub:type` of body
6032---------------------------- ------------------
6035acknowledgments frontmatter
6036copyright-page frontmatter
6037dedication frontmatter
6041contributors frontmatter
6042other-credits frontmatter
6044revision-history frontmatter
6045titlepage frontmatter
6046halftitlepage frontmatter
6047seriespage frontmatter
6052bibliography backmatter
6055[epub-type]: http://www.idpf.org/epub/31/spec/epub-contentdocs.html#sec-epub-type-attribute
6059By default, pandoc will download media referenced from any `<img>`, `<audio>`,
6060`<video>` or `<source>` element present in the generated EPUB,
6061and include it in the EPUB container, yielding a completely
6062self-contained EPUB. If you want to link to external media resources
6063instead, use raw HTML in your source and add `data-external="1"` to the tag
6064with the `src` attribute. For example:
6066 <audio controls="1">
6067 <source src="https://example.com/music/toccata.mp3"
6068 data-external="1" type="audio/mpeg">
6074When creating a [Jupyter notebook], pandoc will try to infer the
6075notebook structure. Code blocks with the class `code` will be
6076taken as code cells, and intervening content will be taken as
6077Markdown cells. Attachments will automatically be created for
6078images in Markdown cells. Metadata will be taken from the
6079`jupyter` metadata field. For example:
6088 display_name: Python 2
6095 file_extension: ".py"
6096 mimetype: "text/x-python"
6098 nbconvert_exporter: "python"
6099 pygments_lexer: "ipython2"
6105**Lorem ipsum** dolor sit amet, consectetur adipiscing elit. Nunc luctus
6106bibendum felis dictum sodales.
6115from IPython.display import HTML
6118console.log("hello");
6126This image  will be
6127included as a cell attachment.
6130If you want to add cell attributes, group cells differently, or
6131add output to code cells, then you need to include divs to
6132indicate the structure. You can use either [fenced
6133divs][Extension: `fenced_divs`] or [native divs][Extension:
6134`native_divs`] for this. Here is an example:
6137:::::: {.cell .markdown}
6140**Lorem ipsum** dolor sit amet, consectetur adipiscing elit. Nunc luctus
6141bibendum felis dictum sodales.
6144:::::: {.cell .code execution_count=1}
6149::: {.output .stream .stdout}
6156:::::: {.cell .code execution_count=2}
6158from IPython.display import HTML
6161console.log("hello");
6167::: {.output .execute_result execution_count=2}
6170console.log("hello");
6179If you include raw HTML or TeX in an output cell, use the
6180[raw attribute][Extension: `fenced_attribute`], as shown
6181in the last cell of the example above. Although pandoc can
6182process "bare" raw HTML and TeX, the result is often
6183interspersed raw elements and normal textual elements, and
6184in an output cell pandoc expects a single, connected raw
6185block. To avoid using raw HTML or TeX except when
6186marked explicitly using raw attributes, we recommend
6187specifying the extensions `-raw_html-raw_tex+raw_attribute` when
6188translating between Markdown and ipynb notebooks.
6190Note that options and extensions that affect reading and
6191writing of Markdown will also affect Markdown cells in ipynb
6192notebooks. For example, `--wrap=preserve` will preserve
6193soft line breaks in Markdown cells; `--atx-headers` will
6194cause ATX-style headings to be used; and `--preserve-tabs` will
6195prevent tabs from being turned to spaces.
6197# Syntax highlighting
6199Pandoc will automatically highlight syntax in [fenced code blocks] that
6200are marked with a language name. The Haskell library [skylighting] is
6201used for highlighting. Currently highlighting is supported only for
6202HTML, EPUB, Docx, Ms, and LaTeX/PDF output. To see a list of language names
6203that pandoc will recognize, type `pandoc --list-highlight-languages`.
6205The color scheme can be selected using the `--highlight-style` option.
6206The default color scheme is `pygments`, which imitates the default color
6207scheme used by the Python library pygments (though pygments is not actually
6208used to do the highlighting). To see a list of highlight styles,
6209type `pandoc --list-highlight-styles`.
6211For HTML-based output formats, the color scheme used for [dark mode]
6212can be selected using the `--highlight-style-dark` option.
6213It defaults to `breezeDark`.
6215If you are not satisfied with the predefined styles, you can
6216use `--print-highlight-style` to generate a JSON `.theme` file which
6217can be modified and used as the argument to `--highlight-style`. To
6218get a JSON version of the `pygments` style, for example:
6220 pandoc --print-highlight-style pygments > my.theme
6222Then edit `my.theme` and use it like this:
6224 pandoc --highlight-style my.theme
6226If you are not satisfied with the built-in highlighting, or you
6227want highlight a language that isn't supported, you can use the
6228`--syntax-definition` option to load a [KDE-style XML syntax definition
6229file](https://docs.kde.org/stable5/en/applications/katepart/highlight.html).
6230Before writing your own, have a look at KDE's [repository of syntax
6231definitions](https://github.com/KDE/syntax-highlighting/tree/master/data/syntax).
6233To disable highlighting, use the `--no-highlight` option.
6235[skylighting]: https://github.com/jgm/skylighting
6236[dark mode]: https://developer.mozilla.org/en-US/docs/Web/CSS/@media/prefers-color-scheme
6240Custom styles can be used in the docx and ICML formats.
6244By default, pandoc's docx and ICML output applies a predefined set of styles
6245for blocks such as paragraphs and block quotes, and uses largely default
6246formatting (italics, bold) for inlines. This will work for most
6247purposes, especially alongside a `reference.docx` file. However, if you
6248need to apply your own styles to blocks, or match a preexisting set of
6249styles, pandoc allows you to define custom styles for blocks and text
6250using `div`s and `span`s, respectively.
6252If you define a `div` or `span` with the attribute `custom-style`,
6253pandoc will apply your specified style to the contained elements (with
6254the exception of elements whose function depends on a style, like
6255headings, code blocks, block quotes, or links). So, for example, using
6256the `bracketed_spans` syntax,
6258 [Get out]{custom-style="Emphatically"}, he said.
6260would produce a docx file with "Get out" styled with character
6261style `Emphatically`. Similarly, using the `fenced_divs` syntax,
6263 Dickinson starts the poem simply:
6265 ::: {custom-style="Poetry"}
6266 | A Bird came down the Walk---
6267 | He did not know I saw---
6270would style the two contained lines with the `Poetry` paragraph style.
6272For docx output, styles will be defined in the output file as inheriting
6273from normal text, if the styles are not yet in your reference.docx.
6274If they are already defined, pandoc will not alter the definition.
6276This feature allows for greatest customization in conjunction with
6277[pandoc filters]. If you want all paragraphs after block quotes to be
6278indented, you can write a filter to apply the styles necessary. If you
6279want all italics to be transformed to the `Emphasis` character style
6280(perhaps to change their color), you can write a filter which will
6281transform all italicized inlines to inlines within an `Emphasis`
6284For docx output, you don't need to enable any extensions for
6285custom styles to work.
6287[pandoc filters]: https://pandoc.org/filters.html
6291The docx reader, by default, only reads those styles that it can
6292convert into pandoc elements, either by direct conversion or
6293interpreting the derivation of the input document's styles.
6295By enabling the [`styles` extension](#ext-styles) in the docx reader
6296(`-f docx+styles`), you can produce output that maintains the styles
6297of the input document, using the `custom-style` class. Paragraph
6298styles are interpreted as divs, while character styles are interpreted
6301For example, using the `custom-style-reference.docx` file in the test
6302directory, we have the following different outputs:
6304Without the `+styles` extension:
6306 $ pandoc test/docx/custom-style-reference.docx -f docx -t markdown
6309 This is text with an *emphasized* text style. And this is text with a
6310 **strengthened** text style.
6312 > Here is a styled paragraph that inherits from Block Text.
6314And with the extension:
6316 $ pandoc test/docx/custom-style-reference.docx -f docx+styles -t markdown
6318 ::: {custom-style="First Paragraph"}
6322 ::: {custom-style="Body Text"}
6323 This is text with an [emphasized]{custom-style="Emphatic"} text style.
6324 And this is text with a [strengthened]{custom-style="Strengthened"}
6328 ::: {custom-style="My Block Style"}
6329 > Here is a styled paragraph that inherits from Block Text.
6332With these custom styles, you can use your input document as a
6333reference-doc while creating docx output (see below), and maintain the
6334same styles in your input and output files.
6338Pandoc can be extended with custom writers written in [Lua]. (Pandoc
6339includes a Lua interpreter, so Lua need not be installed separately.)
6341To use a custom writer, simply specify the path to the Lua script
6342in place of the output format. For example:
6344 pandoc -t data/sample.lua
6346Creating a custom writer requires writing a Lua function for each
6347possible element in a pandoc document. To get a documented example
6348which you can modify according to your needs, do
6350 pandoc --print-default-data-file sample.lua
6352Note that custom writers have no default template. If you want
6353to use `--standalone` with a custom writer, you will need to
6354specify a template manually using `--template` or add a new
6355default template with the name
6356`default.NAME_OF_CUSTOM_WRITER.lua` to the `templates`
6357subdirectory of your user data directory (see [Templates]).
6359[Lua]: https://www.lua.org
6361# Reproducible builds
6363Some of the document formats pandoc targets (such as EPUB,
6364docx, and ODT) include build timestamps in the generated document.
6365That means that the files generated on successive builds will
6366differ, even if the source does not. To avoid this, set the
6367`SOURCE_DATE_EPOCH` environment variable, and the timestamp will
6368be taken from it instead of the current time.
6369`SOURCE_DATE_EPOCH` should contain an integer unix timestamp
6370(specifying the number of second since midnight UTC January 1, 1970).
6372Some document formats also include a unique identifier. For
6373EPUB, this can be set explicitly by setting the `identifier`
6374metadata field (see [EPUB Metadata], above).
6378If you use pandoc to convert user-contributed content in a web
6379application, here are some things to keep in mind:
63811. Although pandoc itself will not create or modify any files other
6382 than those you explicitly ask it create (with the exception
6383 of temporary files used in producing PDFs), a filter or custom
6384 writer could in principle do anything on your file system. Please
6385 audit filters and custom writers very carefully before using them.
63872. If your application uses pandoc as a Haskell library (rather than
6388 shelling out to the executable), it is possible to use it in a mode
6389 that fully isolates pandoc from your file system, by running the
6390 pandoc operations in the `PandocPure` monad. See the document
6391 [Using the pandoc API](https://pandoc.org/using-the-pandoc-api.html)
63943. Pandoc's parsers can exhibit pathological performance on some
6395 corner cases. It is wise to put any pandoc operations under
6396 a timeout, to avoid DOS attacks that exploit these issues.
6397 If you are using the pandoc executable, you can add the
6398 command line options `+RTS -M512M -RTS` (for example) to limit
6399 the heap size to 512MB.
64014. The HTML generated by pandoc is not guaranteed to be safe.
6402 If `raw_html` is enabled for the Markdown input, users can
6403 inject arbitrary HTML. Even if `raw_html` is disabled,
6404 users can include dangerous content in attributes for
6405 headings, spans, and code blocks. To be safe, you should
6406 run all the generated HTML through an HTML sanitizer.
6410Copyright 2006--2021 John MacFarlane (jgm@berkeley.edu). Released
6411under the [GPL], version 2 or greater. This software carries no
6412warranty of any kind. (See COPYRIGHT for full copyright and
6413warranty notices.) For a full list of contributors, see the file
6414AUTHORS.md in the pandoc source code.
6416[GPL]: https://www.gnu.org/copyleft/gpl.html "GNU General Public License"
6417[YAML]: https://yaml.org/spec/1.2/spec.html "YAML v1.2 Spec"
keysesc to close
- j k
- move down, up
- enter
- open
- esc
- back, or up a level
- 1 2 3 4
- readme, files, commits, refs
- /
- filter
- y
- copy the clone command
- c
- compare the chosen commit with another (or shift-click both)
- ?
- show or hide this list