-
Since rmarkdown 1.16, Pandoc's fenced
Div
's are converted to LaTeX environments when the output format is LaTeX, e.g.,::: {.center data-latex=""}
is converted to\begin{center}
. The attributedata-latex
of theDiv
was mandatory, even if it is empty. In rmarkdown 2.2, we silently drop this requirement, which means::: {.center}
is converted to\begin{center}
. This turns out to be a bad idea, because users have no control over which Div's to be converted to LaTeX environments. Previously, they could opt-in by thedata-latex
attribute, but with rmarkdown 2.3, all Div's are converted to LaTeX environments unconditionally. What's more, this change led to bugs like https://stackoverflow.com/q/62340425/559676 and rstudio/bookdown#883. Therefore thedata-latex
attribute became mandatory again in this version. If the LaTeX environment does not need arguments, you may usedata-latex=""
. -
The two Lua fitlers
pagebreak.lua
andlatex-div.lua
(introduced in rmarkdown 1.16) are also applied to the output formatbeamer_presentation
now (thanks, @XiangyunHuang, #1815).
-
Addressed an accessibility issue in highlighted code blocks of HTML output for screen reader users: screen readers no longer read out an unnecessary code line id values (thanks, @jooyoungseo and @atusy, #1833).
-
Added
file_scope
option to output format definition. This enables handling of duplicate numeric footnote identifiers (e.g. across bookdown chapters) via the pandoc--file-scope
option (#1837). -
Added the customizable
lang
attribute toioslides_presentation
output (thanks, @jooyoungseo, #1841). -
Added
publish_site()
function for "one-button" publishing of R Markdown websites. -
When the
df_print
option iskable
and the output format is not HTML,<div class="kable-table">
is no longer added to thekable()
output, because recent versions of Pandoc will convert thediv
to a LaTeX environment when the output format is LaTeX (thanks, Laurens, https://stackoverflow.com/q/62340425/559676). -
html_vignette()
only warns against differences in the vignette title and the vignette index entry for R >= 3.6.0 (thanks, @krlmlr, #1832). -
html_document()
can applycode_folding
on any chunk engines if thefoldable
class is added to a source code block via the chunk options (class.source
orattr.source
). You may apply this feature to all the source code blocks, for example, by settingknitr::opts_chunk$set(class.source = "foldable")
at the beginning of your document (thanks, @atusy, #1835).
-
Exported the internal function
find_pandoc()
, and also added two arguments,dir
andversion
, so that users can provide a custom directory under which this function may find Pandoc, as well as an expected version of Pandoc to be found (thanks, @connorp, #1785). -
pandoc_metadata_arg()
is a new Pandoc helper function to generate--metadata
argument for Pandoc command line (thanks, @cderv, #1789). -
The output format
html_vignette()
now warns against differences in the vignette title specified in thetitle
field in the YAML metadata and the one specified inside\VignetteIndexEntry{}
. Normally they are expected to be identical (#1789). -
Fixed a bug with encoding when rendering
html_notebook
containing HTML widgets (thanks, @cderv, #1799). -
TOC title can now be specified for
html_document
via the top-level optiontoc-title
in the YAML frontmatter (thanks, @atusy, #1771). -
Floating TOC can now distinguish upper/lower-cases (thanks, @atusy, #1783).
-
When
code_folding='show'
for the output formathtml_document
, code blocks can be individually hidden initially by specifying the chunk optionclass.source='fold-hide'
(thanks, @atusy, #1798). -
For LaTeX/PDF output formats
pdf_document
,beamer_presentation
, andcontext_document
, the argumentcitation_package = 'none'
was deprecated, andcitation_package = 'default'
should be used instead if citations are to be processed bypandoc-citeproc
(thanks, @njbart, rstudio/bookdown#754). -
output_format()
can now inheritkeep_md
andclean_supporting
frombase_format
whenNULL
is passed to these arguments. Previously, you must explicitly specifykeep_md
and/orclean_supporting
asTRUE
orFALSE
inoutput_format()
since they could not inherit the corresponding options ofbase_format
. This behavior was not consisent with other arguments ofoutput_format()
(thanks, @atusy, #1823). -
The
smart
argument of most output formats has been removed, because Pandoc'ssmart
extension is enabled by default, and settingsmart: false
for an output format did not really have any effect (which could be considered a bug, but we want to get rid of this option since it existed only for a historical reason for Pandoc 1.x, and Pandoc 2.x has been released for more than two years). If you want to disable thesmart
extension, you can use the optionmd_extensions: -smart
of the output format (thanks, @atusy, #1774). -
pdf_document()
should not specify thegeometry
variable when thedocumentclass
variable is passed to Pandoc (thanks, @jpcirrus, #1782). -
render()
now respects the YAML metadata in the R script when rendering the script with Pandoc 2.8 or later (thanks, @nsoranzo #1740, @cderv #1741). -
For
pandoc_convert()
, when the argumentto = 'pdf'
, it will be changed to'latex'
internally (thanks, @JohannesFriedrich, #1802). -
render(run_pandoc = FALSE)
no longer cleans up the Markdown file (typically knitted from Rmd) (thanks, @BrianDiggs, #1812).
-
Added the returned output from
shiny::runApp()
withinrmarkdown::run()
(thanks, @schloerke, #1760). -
YAML header is now correctly parsed in
html_notebook
's intermediate.knit.md
file so that features like adding bibliography works again (thanks, @everdark, @cderv, #1747). -
ioslides_presentation
template no longer generates an empty<h2>
tag whensubtitle
is not specified in YAML (thanks, @jooyoungseo #1735, @cgrudz #1663). -
No longer center the
#header
element in thehtml_vignette()
output (thanks, @EmilHvitfeldt, #1742). -
Ensure the
tempdir()
exists (viatempdir(TRUE)
) when writing HTML dependencies to a temporary file, because this directory might be erased by accident (thanks, Kurt Hornik, and also @karawoo #1743). -
Added the
slide_level
argument toslidy_presentation()
(https://stackoverflow.com/q/59157211/559676). -
Removed the jQuery dependency in
html_document_base()
(#1723). To avoid bugs like #1723, Pandoc 2.8 users have to upgrade to Pandoc 2.9+. -
For
pdf_document
, horizontal rules generated by Pandoc (before v2.8) stopped working in recent versions of TeX Live, and the same fix as the one to jgm/pandoc#5801 (i.e., hard-code\linethickness
to0.5pt
) was applied in rmarkdown (thanks, @cderv, https://stackoverflow.com/a/58646915/559676).
-
For the output format
pdf_document()
, we no longer adjust the vertical spacing of the title area. This means there will be a larger spacing above the document title in PDF. If you prefers the old (smaller) spacing, please download https://github.com/rstudio/rmarkdown/blob/f6961af/inst/rmd/latex/compact-title.tex and include it to the preamble via theincludes
option ofpdf_document
. However, please note that this means you won't be able to have multiple authors in theauthor
field of the YAML frontmatter, unless you use a custom LaTeX template. With the default LaTeX template, you will run into the error in #1716. Besides, thecompact-title
option in YAML is no longer supported. -
R code in the
header-includes
field in the YAML frontmatter stopped working in the previous version of rmarkdown. The code should be evaluated before passing to Pandoc (thanks, @mcol #1709, @cderv #1710). -
The
encoding
argument is no longer passed to theintermediates_generator
of R Markdown output formats. Theintermediates_generator
function can only accept argumentsinput_file
andintermediates_dir
now (see?rmarkdown::output_format
). This is a breaking change to developers. If you are an output format developer, you have to remove theencoding
argument in yourintermediates_generator
if your output format uses this function. -
The
encoding
argument is no longer passed to therender
element of the site generator (see?rmarkdown::render_site
). -
The
encoding
argument has been removed from many of the internal functions in the rmarkdown package. Now all input files are assumed to be encoded in UTF-8. If you see an error message like "Error inFUN(arg = ...)
: unused argument (arg = ...
)", please file an issue (with a reproducible example) to https://github.com/rstudio/rmarkdown. -
Added a new output format
context_document()
to support ConTeXT output (thanks, @RLesur #1725, @jooyoungseo #1713). -
render_site()
can render R scripts in addition to Rmd files if you setautospin: true
in_site.yml
(thanks, @zeehio, #1564). -
Added
ext
argument tomd_document()
. Its default value is ".md". This argument is intended to be used together withvariant
argument (e.g.,variant = "context"
andext = ".pdf"
) (thanks, @atusy, #1715). -
ioslides_presentation()
stylesheet is updated for printing. Browsers are notified that the presentation should preferentially be printed in landscape orientation and without margin (thanks, @RLesur, #1718). -
Reverted the fix #1703 and applied an alternative fix to #1700, because the original fix brought a new bug #1714 (thanks, @pablobernabeu @cderv @everron @aronatkins).
-
Tabsets don't work with Pandoc 2.8 (thanks, @mnazarov, #1723).
-
The
pdf_document
format failed to work if theheader-includes
field in YAML is an empty list (thanks, @cderv, #1728).
-
For
pdf_document()
, now we patch Pandoc's built-in LaTeX template to include the document subtitle (unnecessary with pandoc 2.6 onwards) and reduce the vertical spacing before title using--include-in-header
rather than overwriting the built-in template, avoiding compability problems with newer versions of Pandoc (thanks, @adunning, #1563). -
find_external_resources()
works now when multiple files are specified in theincludes
option of the output format (thanks, @andrie, #1677). -
find_external_resources()
can find external resources specified in the output format'sreference_doc
orreference_docx
option now (thanks, @jmcphers, #1696). -
rmarkdown::run(file = NULL, dir = "foo/")
failed to run Rmd files under thefoo/
directory (thanks, @jenzopr, #1703). -
Reverted the fix for #1692 since it is no longer necessary (rstudio/tinytex#152 (comment)).
-
The
header-includes
field in the YAML metadata will no longer be overwritten by the command-line option--include-in-header
(thanks, @crsh @mnazarov, #1359). -
Removed the
xmlns
attribute in the<html>
tag in the default HTML template (thanks, @grady #1640, @spgarbet #995).
-
html_vignette()
passesself_contained
argument value to base format (thanks, @cderv, #1668). -
find_external_resources()
works for thehtml_vignette
type again, this fixes rendering vignettes with external resources in pkgdown (regression introduced in rmarkdown 1.16, #1668). -
render(..., clean = TRUE)
may fail to clean the*_files
directory when the output format isprettydoc::html_pretty
(thanks, @yixuan, #1664). -
For
ioslides_presentation
, images with atributes (e.g.,![](sample.png){width=80%}
) can be correctly embedded in the self-contained mode now (thanks, @hadley, #1687). -
Fixed the Pandoc LaTeX templates to avoid the error
File
grffile.sty' not found`. This is because the LaTeX grffile is no longer available in TeX Live (thanks, @cderv #1691, @smmurphy #1692, @JacobD05 rstudio/tinytex#152).
-
The
pandoc-citeproc
binary can now be found correctly on Windows. This fixes an issue withpandoc_citeproc_convert()
(thanks @cderv, #1651). -
Added
self_contained
argument tohtml_vignette
to keep intermediate directory ifself_contained = FALSE
(thanks, @cderv, #1641). -
It is now possible to add pagebreak in HTML, Word, LaTeX, and ODT documents using the
\newpage
or\pagebreak
command in an Rmd file. This is possible thanks to the Pandoc's pagebreak lua filter. Seevignette("lua-filters", package = "rmarkdown")
(thanks, @cderv, #1626). -
The Pandoc extension
ascii_identifiers
is no longer enabled by default. If you still need it, you may use the argumentmd_extensions = "+ascii_identifiers"
in the output format function. However, please note that this will trigger an error in a future version of Pandoc. -
Output formats can be configured by arbitrary YAML files, which used to be restricted to
_output.yml
or_output.yaml
. They can be specified via theoutput_yaml
argument ofrender()
or theoutput_yaml
top-level parameter of YAML front matter, and the first existing one will be used. Ifoutput_yaml
is specified both forrender()
and YAML front matter, thenrender()
has the priority. If none are found, then_output.yml
or_output.yaml
will be used if they exist (thanks, @atusy, #1634). -
Added a Pandoc lua filter to convert fenced Divs to LaTeX environments when the output format is
latex
orbeamer
. Basically a fenced Div::: {.NAME data-latex="[OPTIONS]"}
is converted to\begin{NAME}[OPTIONS] \end{NAME}
in LaTeX. The attributedata-latex
must be provided, even if it is an empty string (meaning that the LaTeX environment does not have any optional arguments). For example,::: {.verbatim data-latex=""}
generates averbatim
environment, and::: {.minipage data-latex="{.5\textwidth}"}
generates\begin{minipage}{.5\textwidth}
. This lua filter was originally written by @RLesur at yihui/bookdown-crc#1. It will allow users to create custom blocks that work for both HTML and LaTeX output (e.g., info boxes or warning boxes). -
Added
keep_html
argument togithub_document
so to save a preview HTML file in a working directory (thanks, @atusy, #1650).
-
Exclude
README.R?md
from files processed byrender_site()
, -
html_document
withcode_folding: hide
supports showing individual source code chunks if they are assigned thefold-show
class via the chunk optionclass.source="fold-show"
(thanks, @atusy, #1602). -
The
extra_dependencies
argument only works withtemplate: default
inpdf_document
. Now it works with any Pandoc LaTeX templates as long as the template uses theheader-includes
variable.
-
Fixed a regression in
ioslides_presentation
that background colors via thedata-background
attribute on slides stopped working (thanks, @ShKlinkenberg, #1265). -
Fixed the bug #1577 introduced in rmarkdown v1.12: tabsets, floating TOC, and code folding in the
html_document
format no longer work with theshiny
runtime (thanks, @RLesur for the fix #1587, and @fawda123 @ColinChisholm @JasonAizkalns for the bug report). -
Added the
keep_md
argument topdf_document()
to keep the intermediate.md
output file (thanks, @broomej, #1001). -
For
render()
, if the input filename contains special characters such as spaces or question marks (as defined inrmarkdown:::.shell_chars_regex
), the file will be temporarily renamed with the special characters replaced by-
(dash) instead of_
(underscore, as in previous versions of rmarkdown). This change will affect users who render such files with caching (cache will be invalidated and regenerated). The change is due to the fact that-
is generally a safer character than_
, especially for LaTeX output (#1396). -
Added a pkgdown site for the rmarkdown package: https://rmarkdown.rstudio.com/docs/ (thanks, @apreshill, #1574).
-
Fixed the bug #1593: in HTML documents, when a MathJax URL is used with a custom template, the source code of the MathJax library is included in the document. This bug was first declared in bookdown (thanks, @topepo for the bug report rstudio/bookdown#683, and @RLesur for the fix #1594).
-
For
pdf_document()
, do not override margins to 1 inch when a custom document class or geometry settings are specified in the YAML front matter (thanks, @adunning, #1550) -
The default value of the
encoding
argument in all functions in this package (such asrender()
andrender_site()
) has been changed fromgetOption("encoding")
toUTF-8
. We have been hoping to support UTF-8 only in rmarkdown, knitr, and other related packages in the future. For more info, you may read https://yihui.org/en/2018/11/biggest-regret-knitr/. -
The option
toc_float: true
forhtml_document
now preserves the text formatting (thanks, @codetrainee, #1548). -
For the
output_file
argument ofrender()
, a file extension will be automatically added if the filename does not contain an extension (e.g.,render('foo.Rmd', 'html_document', output_file = 'bar')
will generatebar.html
); see the help page?rmarkdown::render
for details (thanks, @apreshill, #1551). -
TOC items are not correctly indented when
toc_float
is enabled for thehtml_document
format (thanks, @carolynwclayton #1235 and @RLesur #1243). -
Fixed rstudio/shiny#2307 where the second execution of a
shiny_prerendred
document withhref
dependencies would cause a prerender check error (thanks, @schloerke, #1562). -
The
*_files
directory is not properly cleared due to the inappropriate fix for #1503 and #1472 in the last version (thanks, @wxli0 #1553, @cderv #1566). -
Added an
output_extensions
argument topdf_document()
to make it possible to enable/disable Pandoc extensions for the LaTeX output format (thanks, @hongyuanjia, rstudio/bookdown#687).
-
Fixed file extensions of output files when using non-markdown Pandoc extensions such as
docx+styles
(#1494, @noamross). -
Added a new argument
extra_lines
tolatex_dependency()
to allow users to add extra lines of LaTeX code after\usepackage{}
. Also added a helper functionlatex_dependency_tikz()
based onlatex_dependency()
(#1502, @malcolmbarrett). -
Fixed #1529: when the path of an Rmd file contains Unicode characters that cannot be represented in the system native encoding (especially on Windows),
rmarkdown::render()
may fail. -
Applied a correct fix to an old plotly issue plotly/plotly.R#463.
-
HTML widgets used to be hidden when printing ioslides to PDF in Chrome. Now they will be printed correctly.
-
render(output_format = 'all')
may delete the figure directories that are still needed by certain output formats when one output format doesn't need its figure directory (thanks, @rmcd1024 #1472, @cderv #1503). -
The
<em>
tags in the subtitle, date, and author are removed from the default HTML template (thanks, @royfrancis, #1544).
-
Fixed #1483, which prevented the triangle to be displayed in Firefox when
<details><summary>...</summary></details>
was used (#1485, @bisaloo) -
Provided
rmarkdown.pandoc.args
as a knitr package option inknitr::opts_knit
(#1468, @noamross). -
Added the exported function
pandoc_exec()
, which returns the path of the pandoc binary used by the package (#1465, #1466 @noamross). -
new_session: true
in_site.yml
causesrender_site()
to render each file in a new R session, eliminating some cross-file difficulties, such as function masking (#1326, #1443 @jennybc). -
Added the LaTeX command
\passthrough
in the default LaTeX template for the--listings
flag of Pandoc (rstudio/bookdown#591). -
The icons in
flexdashboard::valueBox()
are not of the full sizes due to the upgrade of FontAwesome in #1340 in the previous version (#1388, rstudio/flexdashboard#189). -
Added the ability to generate tabset dropdowns, usable by adding the
.tabset-dropdown
class to a header (e.g.,# Heading {.tabset .tabset-dropdown}
) (#1405). Thanks @stefanfritsch for contributing the necessary code for this (#1116). -
The
darkly
theme (a darker variant of the Bootswatchflatly
theme) has been added tohtml_document
andhtml_notebook
(#1409, #889). -
Fixed a regression that caused scrollbars on code blocks when the syntax highlighting theme is not the default (#654, #1399).
-
Fixed #1407: reactive expressions can break the section headers of Shiny R Markdown documents.
-
Fixed #1431:
render()
with theintermediates_dir
argument when the output format ispowerpoint_presentation
with a customreference_doc
fails to find the reference document. -
Fixed the website navbar not being able to display submenus properly (#721, #1426).
-
Added checks for shiny-prerendered documents to find all html dependencies, match all execution packages, and match the major R version (#1420).
-
Added an argument
cache = TRUE
to the internal functionrmarkdown:::find_pandoc()
, so that users can invalidate the cached path of Pandoc viarmarkdown:::find_pandoc(cache = FALSE)
(thanks, @hammer, #1482). -
Added an RStudio project template for simple R Markdown websites, so that users can create such websites from RStudio:
New Project -> New Directory -> Simple R Markdown Website
(thanks, @kevinushey, #1470). -
Fixed #1471: Pandoc's (version 2.x) syntax highlighting themes don't work well with the Bootstrap style (thanks, @gponce-ars #1471, @cderv #1489).
-
Fixed the warnings in #1224 and #1288 when calling
render()
with an absoluteoutput_dir
orintermediates_dir
. -
Fixed #1300: calling
render()
withintermediates_dir
may fail when the intermediate dir is on a difference device or filesystem. -
Fixed #1358: calling
render()
withintermediates_dir
will fail if the Rmd document contains bibliography files that are dynamically generated.
-
Added a new argument
slide_level
topowerpoint_presentation()
(#1270). -
The tinytex package has become a required dependency (to build R Markdown to PDF).
-
Added
compact-title
variable to the LaTeX default templates to control use of LaTeXtitling
package; defaults totrue
(#1284). -
pdf_document(template = NULL)
does not work (#1295). -
Restore ability to use any HTML format with R Markdown Websites (#1328).
-
Add
options
argument topaged_table()
to enable explicit passing of display options. -
Add
pandoc_citeproc_convert()
function for conversion of bibliography files (e.g. BibTeX files). -
Update to Font Awesome version 5.0.13 (#1340).
-
Add
site_resources()
function for computing resource files required for a website. -
Export
default_site_generator()
function. -
The
latex_document()
format should not clean up the figure directory (thanks, @emiltb, rstudio/bookdown#582). -
Enable post processors that change the output file to specify that the base post processor should still be applied to the original output file.
- Added a new (experimental) output format
powerpoint_presentation
. If you want to test it, you will need Pandoc >= 2.1 (#1231).
- If the tinytex package is installed, PDF output is built through
tinytex::latexmk()
, otherwise it is generated byrmarkdown:::latexmk()
, which has been factored out and improved in the tinytex package, so it is recommended that you install the tinytex package (#1222).
-
Temporary files created in
render()
may be cleaned up prematurely, which can cause problems with Shiny R Markdown documents (#1184). -
Further improvements regarding compatibility with Pandoc 2.0, e.g. tabbed sections don't work (https://community.rstudio.com/t/3019).
-
When
preserve_yaml = TRUE
inmd_document()
,toc = TRUE
fails to create the table of contents (thanks, @stla, #1216). -
Suppress confusing error messages from
knitr::purl()
duringrmarkdown::find_external_resources()
(thanks, @aronatkins #1247, and @paulobrecht #1154). -
Fixed the obscure error
Error: path for html_dependency not found:
, which was due to the HTML dependency of highlight.js (thanks, @bborgesr, #1213).
-
render_site()
does not work with_site.yml
that does not have theoutput
setting (#1189). -
The variables
input
andoutput
do not work in Shiny R Markdown documents (#1193). -
ioslides_presentation
fails to embed images (#1197). -
With Pandoc 2.x,
github_document()
generates the wrong filename extension.gfm-ascii_identifiers
instead of.md
, and line height of code blocks in the HTML preview is too big (#1200).
-
Fixed an issue with
df_print: paged
where row names where not printed and added support forrownames.print
option to control when they print. -
Add
smart
option forword_document()
format. -
Save render intermediates when generating beamer presentations (fixes #1106).
-
Fixed issues when specifying NULL/null/empty parameter values (#729 and #762).
-
Better error message when unable to prerender a document. (#1125)
-
shiny::renderText()
does not work in Markdown section headings (#133). -
The
value
argument ofpandoc_variable_arg()
can be missing now (#287). -
Background colors and images are supported for ioslides presentations (#687).
-
HTML widgets in an Rmd document cannot be rendered if another Rmd document is rendered via
rmarkdown::render()
in this document (#993). -
Try harder to clean up temporary files created during
render()
(#820). -
Wrong environment for evaluating R code chunks in Shiny R Markdown docs (#1162, #1124).
-
Do not call
bibtex
to create the bibliography when there are no citations in the document and the output format ispdf_document()
withcitation_package = 'natbib'
(#1113). -
render()
will stop if the output format is PDF but there are any errors during building the index or bibliography (#1166). -
beamer_presentation()
doesn't work whencitation_package != 'none'
(#1161). -
File-based inputs don't work in parameterized documents (#919).
-
rmarkdown is compatible with Pandoc 2.0 now (#1120).
-
render()
withintermediates_dir
fails with R plots (#500). -
Added two new output formats
latex_document()
andlatex_fragment()
(#626). -
Relative paths of images in HTML output should not be resolved to absolute paths (#808).
-
render_site()
does not support multiple output formats for a single Rmd (#793). -
Unicode characters may be scrambled when downloading the Rmd source file using the download button generated by
html_document(code_download = TRUE)
(#722). -
Upgraded highlight.js from v1.1 to v9.12.0 (#988, #907).
-
The argument
keep_md = TRUE
actually preserves the Markdown output file fromknitr::knit()
now (as documented). Previously, it generates a new Markdown file by concatenating the YAML metadata (title, author, date) with the body of the original Markdown output file (#450). -
For
md_document()
, whenvariant == 'markdown'
andperserve_yaml = TRUE
, the Pandoc argument--standalone
should not be used (#656).
-
Fixed an issue where headers with non-ASCII text would not be linked to correctly in the table of contents.
-
Support code folding for bash, python, sql, Rcpp, and Stan chunks.
-
Provide rmarkdown.pandoc.id_prefix as Knit option
-
Fixed two issue with
df_print: paged
, one would prevent rendering data with lists of lists and the other where the column type would get cut off. -
Better support for
citation_package: biblatex
inpdf_document()
(#1062). -
On certain Windows platforms, compiling LaTeX to PDF may fail because
system2(stdout = FALSE)
is not supported, in which case the defaultsystem2()
will be used (#1061). -
Allow paged tables to render even when page load / visibility has a long delay
-
Fixed an issue where code within Shiny pre-rendered documents was not rendered correctly.
-
Add
includes
parameter tohtml_fragment
format. -
Use RStudio redirection URL to replace deprecated MathJax CDN
-
data.table
expressions involving:=
are no longer automatically printed within R Markdown documents. (#829) -
Fix #910: the extra_dependencies argument of pdf_document() does not work when no code chunks contain LaTeX dependencies.
-
The extra_dependencies of pdf_document() can also take a character vector of LaTeX package names, or a named list of LaTeX package options (with names being package names), which makes it much easier to express LaTeX dependencies via YAML.
-
Automatically ignore packrat directory for render_site
-
Fix #943: escape end tags in shiny_prerendered code contexts
-
Add support for sticky tabs in html_document via tabset-sticky class
-
Process Rmd files with lowercase extension (.rmd) in render_site
-
Fix stack space consumption issues with large JS payloads in chunks
-
Add
section_divs
argument to html_document and html_fragment to control generation of section tags/divs. -
Remove data-context="(data|server|server-start)" chunks from HTML served to client in shiny_prerendered
-
Fix v1.2 regression in ordering of CSS for ioslides_presentation.
-
Fix rendering of pagedtables within html_notebook format.
-
Ensure that html_prerendered UI is never cached.
-
Add
citeproc
argument to YAML header; controls whether pandoc-citeproc is used (#831)
-
Add support for df_print to handle additional dplyr classes: grouped_df, rowwise_df and tbl_sql.
-
Add new
runtime: shiny_prerendered
mode for interactive documents. -
Prepend "section-" to ids in runtime: shiny[_prerendered] to eliminate potential conflicts with shiny output ids
-
Use html_dependencies for highlight.js, pagedtables, slidy, ioslides, & navigation (improved dependency behavior for runtime: shiny[_prerendered])
-
Serialize runtime: shiny[_prerendered] dependencies to JSON rather than .rds
-
ioslides: check for null before concatenating attr["class"] (#836)
-
Add rmarkdown.onKnit/onKnitCompleted package hooks
-
Non-ASCII keys in yaml file should be marked to UTF8 as well, when read into R as the name of a list (#841)
-
Remove key-column special-case left alignment in pagedtables (#829)
-
Replace backslashes in floating TOC headings (#849)
-
Suggests rather than Imports for tibble (R 3.0 compatibility)
-
Add
paged_table
function for printing paged tables within HTML documents -
Support
{.active}
attribute for setting initially active tab (#789) -
Add
knit_root_dir
argument torender()
and YAML header, a convenience for setting knitr's root.dir option -
Improve alignment of text in sub-topics for floating TOC
-
Bibliography file paths in YAML containing forward slashes could not be rendered (#875)
-
Fixed an issue where attempts to render an R Notebook could fail if the path contained multibyte characters.
-
Fixed an issue where the default Beamer template did not provide vertical padding between paragraphs with certain versions of pandoc (<= 1.17.2).
-
Try to install latexmk automatically on Windows
-
Added
df_print
option to html_document format for optionally printing data frames usingknitr::kable
, thetibble
package, or an arbitrary function -
Fix for render_site not showing Chinese characters correctly
-
Fix for ignoring knit_meta that is explicitly passed to render
-
Parameter editing: don't allow NULL to overwrite previous state
-
Parameter editing: fix incorrect name for parameters with expressions
-
Parameter editing: allow multiple values when the parameter is configured to use a "multiple" selector
-
Switched the order in which format dependencies are added for
html_document
so thatextra_dependencies
are added at the end, after bootstrap, etc. (#737) -
pdf_document(keep_tex = TRUE)
will generate the .tex document even if PDF conversion fails (#779). -
Move latex header includes to just before \begin{document}
-
Special 'global' chunk label for runtime: shiny which designates a chunk to be run once and only once in the global environment (startup performance improvement for multi-user shiny documents)
-
Ensure supporting files are writeable (#800)
-
Make the "show code" buttons more CSS-friendly (#795)
-
Exclude
output_dir
from site files (#803) -
Export
navbar_html
andyaml_front_matter
functions
-
toc_float
no longer automatically setstoc = TRUE
-
Added an argument
error
topandoc_available()
to signal an error when (iferror = TRUE
) pandoc with the required version is not found. -
Added
html_notebook
format for creating HTML documents that include source code and output. -
Added
resolve_output_format
function (useful for front ends that need to mirror the default format resolution logic ofrender
). -
Added
code_download
option tohtml_document
to provide an option to embed a downloadable copy of the Rmd source code within the document. -
Added
slide_level
option to ioslides_presentation to set the level of heading used for individual slides. -
Added
hard_line_breaks
option togithub_document
to deal with change in behavior of GitHub's markdown renderer with respect to line breaks. -
Use "markdown_strict" rather than "markdown" for
pandoc_self_contained_html
when pandoc >= 1.17 (pandoc hanging bug was fixed in this version) -
Default highlighting engine for
html_document
now highlights bash, c++, css, ini, javascript, perl, python, r, ruby, scala, stan and xml -
Added
print
sub-option totoc_float
to control whether the table of contents appears when user prints out the HTML page. -
Added
readme
option tohtml_vignette
which automatically creates a package level README.md (for GitHub) in addition to rendering the vignette. -
Support for
keep_md
inhtml_vignette
format. -
Try to install the latexmk package automatically on Windows if the executable latexmk.exe exists.
-
Ability to set
opts_hooks
inknitr_options()
(#672) -
Added
render_site
and related functions for rendering collections of documents within a directory as a website. -
Ability to define html_document navigation bar using simple yaml format.
-
Added
pre_knit
andpost_knit
hooks for output formats. -
Discover
LaTeX
dependencies and add them to the.tex
preamble (#647) -
Added new
all_output_formats
function to enumerate all output formats registered for an Rmd. -
Change
fig_caption
default to TRUE for all formats -
Change
fig_retina
to 2 for HTML formats (no longer contingent onfig_caption
) -
Ensure pandoc binary exists before binding to pandoc directory (#632)
-
Handle relative paths for 'default_output_format' (#638)
-
Eliminate duplicate viewport meta tag from html_document
-
Added biblatex biblio-style support to the LaTeX template for Pandoc 1.15.2 (#643)
-
Allow override of header font-size in html_document custom css (#652)
-
Fix for horizontal scrollbars appearing w/ code folding (#654)
-
Specifying
toc_float
inhtml_document
now automatically setstoc = TRUE
-
Enable per-header opt-out of
toc-float
via {.toc-ignore} attribute -
Correctly handle soft line breaks in ioslides_presentation (#661)
-
Don't use text-muted for code folding btns (text visibility in non-default themes)
-
Fix for rendering non-HTML formats from .md files (resolve runtime before knit)
-
html_dependency_bootstrap
now accepts theme = "default" argument -
Use pandoc 1.17.0.2 compatible LaTeX template when pandoc >= 1.17.0.2
-
Support custom template for
ioslides_presentation
-
Added
analytics
option forioslides_presentation
for Google Analytics -
Removed the extra tag
<p></p>
around HTML output (typically generated by htmltools) from code chunks, to avoid invalid HTML like<p><div>...</div><p>
(#685)
-
Added odt_document format for OpenDocument Text output
-
Added rtf_document format for Rich Text Format output
-
Added github_document format for GitHub Flavored Markdown output
-
Only apply white background for themed HTML documents (#588)
-
Added to the default HTML template to make it work better with mobile browsers. (#589)
-
Specify --filter pandoc-citeproc after custom pandoc args
-
Long lines in code blocks will be wrapped in the
html_vignette()
output (#595) -
Added new arguments
run_pandoc = TRUE
andknit_meta = NULL
torender()
. See the help page ofrender
for details. (#594) -
The
tufte_handout
format now delegates to thetufte
package and no longer provides a base template. -
Use pandoc 1.15.2 compatible LaTeX template when pandoc >= 1.15.2
-
Fix issue with Beamer template and pandoc 1.15.2
-
Updated embedded JQuery to v1.11.3 and Bootstrap to v3.3.5.
-
Expose core HTML dependencies for use by custom R Markdown formats.
-
New
html_document
themes: "lumen", "paper", "sandstone", "simplex", & "yeti". -
Ability to include bootstrap navbar for multi-page
html_document
websites -
Added support for
abstract
field tohtml_document
format -
Added support for floating table of contents (via tocify) to
html_document
-
Added support for tabsets via use of {.tabset} class on top-level headings
-
Added support for folding/unfolding of R code chunks in
html_document
-
Support
url
references in CSS files forruntime: shiny
-
Change name of common options file to
_output.yml
-
Tweak pandoc conversion used in
pandoc_self_contained_html
to prevent hanging with large script elements (use "markdown" rather than "markdown-strict" as input format) -
The filename extension .bib will be removed before bibliography files are passed to Pandoc when the output is LaTeX/PDF and the citation package natbib or biblatex is used on Windows. This is because bibtex in MikTeX will always add the extension .bib to bibliography files, e.g. it treats foo.bib as foo.bib.bib. (#623)
-
Render Shiny documents in a clean environment; fixes issue in which code in Shiny documents could access internal R Markdown state
- Added a fix to #580 for Windows users.
-
Fix for a bug causing certain files to be deleted as intermediate files. (#580)
-
PDF/LaTeX output no longer uses natbib as the citation package by default. If you do want to use natbib or biblatex, you may still use the argument citation_package = 'natbib' or 'biblatex'. (#577)
-
Fix for JS exception in slidy_presentation when served from the filesystem (don't call pushState for file:// urls)
-
Escape single quotes in file paths
-
Fix for missing resources when rendering a filename with shell characters
-
For PDF/LaTeX output, citations are processed via natbib or biblatex instead of pandoc-citeproc. The ciation package natbib or biblatex can be chosen using the argument
citation_package
inpdf_document()
,beamer_presentation()
, andtufte_handout()
. LaTeX is compiled to PDF via latexmk (https://www.ctan.org/pkg/latexmk/); if it is not installed, a simple emulation will be used (run pdflatex/xelatex/lualatex, bibtex, and makeindex a few times). You are recommended to install latexmk, and please note latexmk requires a Perl installation (this is important especially for Windows users). -
Always use the graphics package for PDF output
-
Fix for the error "cannot change value of locked binding for 'metadata'" when one call of rmarkdown::render() is nested in another one (#248)
-
Fix for an issue causing image paths to be rendered incorrectly in Windows when rendering an
html_document
withself_contained: false
and a path is passed in argumentoutput_dir
. (#551) -
Add always_allow_html option for preventing errors when html_dependencies are rendered in non-HTML formats (e.g. pdf_document or word_document).
-
Fix for an issue causing resources not to be discovered in some documents containing an empty quoted string (
""
) in an R chunk.
-
Support for table of contents in word_document (requires pandoc >= 1.14)
-
Support for opts_template in knitr options
-
Don't implicitly discover directories when scanning for dependent resources
-
Fix for slide numbers not showing up in ioslides when served from the filesystem (don't call pushState for file:// urls)
-
Remove inlining of bootstrap CSS (was workaround for bug now fixed in pandoc)
-
Allow specifying an R file in calls to find_all_resources
-
Add support for keep_md to word_document
-
Increase pandoc stack size to 512M (often required for base64 encoding e.g. larger embedded leaflet maps). Stack size can also be controlled by the pandoc.stack.size option.
-
Allow YAML front matter to be terminated with ...
-
Automatically generate a user-interface (via a Shiny application) for user specification of values in parameterized reports.
-
Add tightlist macro for compatibility with pandoc >= 1.14
-
Bugfix: Don't merge render params recursively with knit_params
-
Bugfix: Handle slashes correctly on Windows for slidy_presentation when self_contained = FALSE
-
Add latex_engine option to beamer_presentation format
-
Ensure that when LANG=en_US pandoc receives en_US.UTF-8 (prevent hang)
-
Generate MathJax compatible output when using html_fragment format.
-
Use pandoc built-in template for Beamer
-
Use pandoc 1.14 compatible LaTeX template when pandoc >= 1.14
-
Inline bootstrap.min.css to workaround pandoc 1.14 base64 encoding issue
-
Add support for discovering references to external resources when the document output format is PDF
-
Fix several issues causing pandoc errors when an intermediates directory is used, including during render for Shiny documents
-
Support for parameterized reports. Parameter names and default values are defined in YAML and can be specified via the 'params' argument to the render function
-
'md_extensions' option to enable/disable markdown extensions for input files
-
Automatically discover and include dependent resources (e.g. images, css, etc.) for interactive documents.
-
Added pandoc_version function.
-
Use VignetteEncoding directive in html_vignette format
-
Fix issues related to use of non-ASCII characters in ioslides_presentation
-
Non-ASCII characters in the YAML data are not marked with the UTF-8 encoding when they are read into R, so character strings in
rmarkdown::metadata
may be displayed incorrectly (#420) -
Various improvements to tufte_handout format
-
Add 'dev' option to output formats to specify output device for figures
-
Enable use of footnotes in titles of LaTeX documents
-
Various improvements related to directory detection/handling on Windows
-
Sync to the latest LaTeX and Beamer templates from pandoc-templates
-
Switched from the Bootstrap 2 web framework to Bootstrap 3. This is designed to work with Shiny >= 0.10.3, which has made the same switch
-
Add CSS to restore responsive image behavior from Bootstrap 2
-
Use a more subtle treatment for inline code in Bootstrap themed documents
-
Improved support for multiple authors in ioslides
-
Workaround for poor rendering of ioslides multi-columns lists in Safari 8
-
Serve index.html as fallback default file for rmarkdown::run
Initial release to CRAN