'R Markdown' Format for converting from 'R Markdown' document to an MS Word document.
The function enhances the output offered by rmarkdown::word_document()
with
advanced formatting features.
Arguments
- base_format
a scalar character, the format to be used as a base document for 'officedown'. Default to word_document but can also be
word_document2()
from bookdown.When the
base_format
used isbookdown::word_document2
, thenumber_sections
parameter is automatically set toFALSE
. Indeed, if you want numbered titles, you are asked to use a Word document template with auto-numbered titles (the title styles of the default `rdocx_document' template are already set to FALSE).- tables
see section 'Tables' below.
- plots
see section 'Plots' below.
- lists
see section 'Lists' below.
- mapstyles
a named list of style to be replaced in the generated document.
list("Normal" = c("Author", "Date"))
will result in a document where all paragraphs styled with stylename "Date" and "Author" will be then styled with stylename "Normal".- page_size, page_margins
default page and margins dimensions. If not null (the default), these values are used to define the default Word section. See
officer::page_size()
andofficer::page_mar()
.- reference_num
if
TRUE
, text for references to sections will be the section number (e.g. '3.2'). If FALSE, text for references to sections will be the text (e.g. 'section title').- ...
arguments used by word_document
Value
R Markdown output format to pass to render.
Tables
a list that can contain few items to style tables and table captions. Missing items will be replaced by default values. Possible items are the following:
style
: the Word stylename to use for tables.layout
: 'autofit' or 'fixed' algorithm. Seetable_layout
.width
: value of the preferred width of the table in percent (base 1).topcaption
: caption will appear before (on top of) the table,tab.lp
: caption table sequence identifier. All table captions are supposed to have the same identifier. It makes possible to insert list of tables. It is also used to prefix your 'bookdown' cross-reference call; iftab.lp
is set to "tab:", a cross-reference to table with id "xxxxx" is written as\@ref(tab:xxxxx)
. It is possible to set the value to your default Word value (in French for example it is "Tableau", in German it is "Tabelle"), you can then add manually a list of tables (go to the "References" tab and select menu "Insert Table of Figures").caption
; caption options, i.e.:style
: Word stylename to use for table captions.pre
: prefix for numbering chunk (default to "Table ").sep
: suffix for numbering chunk (default to ": ").tnd
: (only applies if positive. )Inserts the number of the last title of leveltnd
(i.e. 4.3-2 for figure 2 of chapter 4.3).tns
: separator to use between title number and table number. Default is "-".fp_text
: text formatting properties to apply to caption prefix - seeofficer::fp_text_lite()
.
conditional
: a list of named logical values:first_row
andlast_row
: apply or remove formatting from the first or last row in the tablefirst_column
andlast_column
: apply or remove formatting from the first or last column in the tableno_hband
andno_vband
: don't display odd and even rows or columns with alternating shading for ease of reading.
Default value is (in YAML format):
style: Table
layout: autofit
width: 1.0
topcaption: true
tab.lp: 'tab:'
caption:
style: Table Caption
pre: 'Table'
sep: ':'
tnd: 0
tns: '-'
fp_text: !expr officer::fp_text_lite(bold = TRUE)
conditional:
first_row: true
first_column: false
last_row: false
last_column: false
no_hband: false
no_vband: true
Plots
Argument plot is expected to be a list. It can contain few items to style figures and figure captions.
You don't have to provide values for each items of the list, missing items are replaced by default values. Possible items are:
style
: the name of the "Word" paragraph style to use for the paragraphs that will contain a graphic.align
: alignment of paragraphs containing graphics (possible values are 'left', 'right' and 'center').topcaption
: specify whether the legend should appear above the graphic (TRUE
) or below the graphic (FALSE
).fig.lp
: Sequence identifier for figure legends. All graphics' captions are supposed to have the same identifier. This value is used for different features:It is used to insert a table of figures.
It is also used to prefix your 'bookdown' cross-reference call; if
fig.lp
is set to "fig:", a cross-reference to the figure with id "xxxxx" is written as\@ref(fig:xxxxx)
.
It is possible to set the Word default value (in French for example, it is "Figure"), you can then manually add a list of figures (go to the "References" tab and select the "Insert a table of figures" menu).
caption
: a list for captions' options, i.e.style
: the name of the "Word" paragraph style to use for the paragraphs that will contain figures' captions.pre
: prefix for numbering chunk (default to "Figure ").sep
: suffix for numbering chunk (default to ": ").tnd
: (only applies if positive). Inserts the number of the last title of leveltnd
(i.e. 4.3-2 for figure 2 of chapter 4.3).tns
: separator to use between title number and figure number. Default is "-".fp_text
: text formatting properties to apply to caption prefix - seeofficer::fp_text_lite()
.
Default value is (in YAML format):
Lists
The parameter lists is a list that can contain two named items:
ol.style
: style to use for ordered lists.ul.style
: style to use for unordered lists.
Default values are list(ol.style = NULL, ul.style = NULL)
.
Expected values are the stylenames to be used to replace the style of ordered and unordered lists created by pandoc. If NULL, no replacement is made.
These values in YAML format are:
They can have values corresponding to existing stylenames (of type 'numbering'). With package 'officer', we can read these values with styles_info().
library(officer)
docx_file <- system.file(
package = "officedown", "examples",
"bookdown", "template.docx"
)
doc <- read_docx(docx_file)
styles_info(doc, type = "numbering")[, 1:6]
#> style_type style_id style_name base_on is_custom is_default
#> 13 numbering Aucuneliste No List <NA> FALSE TRUE
#> 40 numbering Defaultul Default ul Aucuneliste TRUE FALSE
#> 41 numbering Defaultol Default ol Aucuneliste TRUE FALSE
From the above available values, the possible configuration is possible:
Finding stylenames
You can access them in the Word template used. Function styles_info() can let you read these styles.
You need 'officer' to read the stylenames (to get information
from a specific "reference_docx", change ref_docx_default
in the example below.
library(officer)
docx_file <- system.file(package = "officer", "template", "template.docx")
doc <- read_docx(docx_file)
To read paragraph
stylenames:
styles_info(doc, type = "paragraph")[, 1:6]
#> style_type style_id style_name base_on is_custom is_default
#> 1 paragraph Normal Normal <NA> FALSE TRUE
#> 2 paragraph Titre1 heading 1 Normal FALSE FALSE
#> 3 paragraph Titre2 heading 2 Normal FALSE FALSE
#> 4 paragraph Titre3 heading 3 Normal FALSE FALSE
#> 9 paragraph centered centered Normal TRUE FALSE
#> 15 paragraph ImageCaption Image Caption Normal TRUE FALSE
#> 16 paragraph TableCaption Table Caption ImageCaption TRUE FALSE
#> 18 paragraph TM1 toc 1 Normal FALSE FALSE
#> 19 paragraph TM2 toc 2 Normal FALSE FALSE
#> 20 paragraph Textedebulles Balloon Text Normal FALSE FALSE
#> 23 paragraph graphictitle graphic title ImageCaption TRUE FALSE
#> 24 paragraph tabletitle table title TableCaption TRUE FALSE
To read table
stylenames:
styles_info(doc, type = "table")[, 1:6]
#> style_type style_id style_name base_on is_custom
#> 6 table TableauNormal Normal Table <NA> FALSE
#> 10 table tabletemplate table_template TableauNormal TRUE
#> 11 table Listeclaire-Accent2 Light List Accent 2 TableauNormal FALSE
#> 17 table Tableauprofessionnel Table Professional TableauNormal FALSE
#> is_default
#> 6 TRUE
#> 10 FALSE
#> 11 FALSE
#> 17 FALSE
To read list
stylenames:
styles_info(doc, type = "numbering")[, 1:6]
#> style_type style_id style_name base_on is_custom is_default
#> 7 numbering Aucuneliste No List <NA> FALSE TRUE
R Markdown yaml
The following demonstrates how to pass arguments in the R Markdown yaml:
---
output:
officedown::rdocx_document:
reference_docx: pandoc_template.docx
tables:
style: Table
layout: autofit
width: 1.0
topcaption: true
tab.lp: 'tab:'
caption:
style: Table Caption
pre: 'Table '
sep: ': '
tnd: 0
tns: '-'
fp_text: !expr officer::fp_text_lite(bold = TRUE)
conditional:
first_row: true
first_column: false
last_row: false
last_column: false
no_hband: false
no_vband: true
plots:
style: Normal
align: center
fig.lp: 'fig:'
topcaption: false
caption:
style: Image Caption
pre: 'Figure '
sep: ': '
tnd: 0
tns: '-'
fp_text: !expr officer::fp_text_lite(bold = TRUE)
lists:
ol.style: null
ul.style: null
mapstyles:
Normal: ['First Paragraph', 'Author', 'Date']
page_size:
width: 8.3
height: 11.7
orient: "portrait"
page_margins:
bottom: 1
top: 1
right: 1.25
left: 1.25
header: 0.5
footer: 0.5
gutter: 0.5
reference_num: true
---
Known limitations
When using knitr::include_graphics()
in 'rdocx_document',
the image must be local, while with standard 'R Markdown'
output documents, images located on Internet are supported.
Examples
# rdocx_document basic example -----
library(rmarkdown)
library(officedown)
if (pandoc_available() && pandoc_version() >= numeric_version("2.0")) {
# minimal example -----
example <- system.file(
package = "officedown",
"examples/minimal_word.Rmd"
)
rmd_file <- tempfile(fileext = ".Rmd")
file.copy(example, to = rmd_file)
docx_file <- tempfile(fileext = ".docx")
render(rmd_file, output_file = docx_file, quiet = TRUE)
}