查看原文
其他

Bookdown文档生成教程

2017-05-25 陈同 生信宝典

bookdown是一款及其方便的编写技术文档或教材的工具,语法简洁,数据处理灵活。支持Rmarkdown或普通markdown通过pandoc软件转换为HTML或PDF。

Here lists the usage of bookdown for writing documents.

Get required information

Install required software

RstudioPandoc二选一, bookdown必须安装。

  • Install Rstudio (version>1.0.0)

  • Install Pandoc (version>1.17.0.2)或者参照here

  • In R install.packages("bookdown")

Get a demo example

Clone or download en example from https://github.com/rstudio/bookdown-demo.

Build the book

run bash _build.sh in the downloaded demo example, then the books will be in _book dir.

The content of _build.sh is:

#!/bin/sh Rscript -e "bookdown::render_book('index.Rmd', 'bookdown::gitbook')" Rscript -e "bookdown::render_book('index.Rmd', 'bookdown::pdf_book')"

Customize our bookdown

准备Rmd文件

基本规则
  • 一个典型的bookdown文档包含多个章节,每个章节在一个R Markdown文件里面 (文件的语法可以是pandoc支持的markdown语法,但后缀必须为Rmd)。

  • 每一个章节都必须以# Chapter title开头。后面可以跟一段概括性语句,概述本章的内容,方便理解,同时也防止二级标题出现在这一页。默认系统会按照文件名的顺序合并Rmd文件。

  • 另外章节的顺序也可在_bookdown.yml文件中通过rmd_files:["file1.Rmd", "file2.Rmd", ..]指定。

  • 如果有index.Rmdindex.Rmd总是出现在第一个位置。通常index.Rmd里面也需要有一章节,如果不需要对这一章节编号的话,可以写作# Preface {-}, 关键是{-}

  • 在第一个出现的Rmd文件中,可以定义Pandoc相关的YAML metadata, 比如标题、作者、日期等(去掉#及其后的内容)。

    ``` title: "My book" author: #可以写多行信息,都会被当做Author处理 - "CT" - "CY" - "chentong_biology@163.com" date: "`r Sys.Date()`" documentclass: article #可以为book或article # 如果需要引用参考文献,则添加下面三行内容 bibliography: [database.bib]  #指定存储参考文献的bib文件,endote或zotero都可以导出这种引文格式 biblio-style: apalike  #设定参考文献显示类型 link-citations: yes ``` ```{r setup, include=FALSE} knitr::opts_chunk$set(echo = FALSE, fig.align="center", out.width="95%", fig.pos='H') # knitr::opts_chunk$set(cache = FALSE, autodep=TRUE) set.seed(0304) ``` ~~~~~~
插入并引用图片(外部图片)

插入图片最好使用knitr::include_graphics,可以同时适配HTML和PDF输出。另外当目录下同时存在name1.pngname1.pdf文件时,会自动选择在HTML展示name1.png文件,在PDF输出中引入name1.pdf格式的文件。

图的标签为fig-name(不能有下划线),在引用时需使用如下格式\@ref(fig:fig-name),且fig.cap也要设置内容。

多张图可以同时展示,图的名字以vector形式传给include_graphics,需要设置out.width=1/number-picsfig.show="hold"

Insert a single pic and refer as Figure \@ref(fig:fig-name). `echo=FALSE` will hide the code block and display the output of `r` command only. These options can be set globally as indicated below. ```{r fig-name, fig.cap="Markdown supported string as caption", fig.align="center", echo=FALSE} knitr::include_graphics("images/1.png") ``` Suppose we have 3 pictures in `images` folder with names as `Fig1_a`, `Fig1_b`,  `Fig1_c`,  we can refer to them using Figure \@ref(fig:fig1). ```{r fig1, fig.cap="3 sub-plots.", fig.align="center", out.width=33%, fig.show="hold"} fig1 = list.files("images", pattern="Fig1_.*", full.names=T) knitr::include_graphics(fig1) ``` Another way of including two pics. ```{r fig-name2,  out.width="49%", fig.show="hold", fig.cap="Markdown supported string as caption",  fig.align="center", echo=FALSE} knitr::include_graphics(c("images/1.png", "images/2.png")) ``` ~~~~~~~~~~~~~~~~

如果图或表的标题中有Markdown语法,输出为HTML时是可以正确解析的,但是输出为PDF时却不可以。这时可以使用Text Reference。当图或表的标题太长时,也可以使用Text Reference引用一段话作为图和表的标题。

Here is normal text. (ref:pic-label) This line can be referred in **fig.cap** and markdown syntax is supported for both `HTML` and `PDF` output. ```{r pic-label, fig.cap="(ref:pic-label)"} knitr::include_graphics("images/1.png") ``` ~
插入并引用表格(外部表格)

外部表格的名字中必须包含tab:, 然后是表格的实际名字,格式为(\#tab:table-name); 引用时使用Table \@ref(tab:table-name)。 表格名字中不能有下划线。

Check Table \@ref(tab:seq-sum) for detail. Table: (\#tab:seq-sum) Summary of sequencing reads 测序量总结 (对于双端测序,  *\_1* 表示左端reads, *\_2* 表示右端reads) ---------------------------------------------------------------------- Sample     Total reads     Total bases Sequence length (nt)     GC content (%) Encoding               -------- ------------- --------------- ---------------------- ---------------- ----------------------- T8_1        37,106,941   5,566,036,721 138-150                              47 Sanger / Illumina 1.9   T8_2        37,106,941   5,566,034,285 138-150                              47 Sanger / Illumina 1.9   ----------------------------------------------------------------------
插入并引用表格(内部表格)

插入表格推荐使用knitr::kable,只要提供数据矩阵,用r读取就可以了。

Check Table \@ref(tab:table-id) for detail. ```{r table-id, include=FALSE} a <- as.data.frame(matrix(rnorm(20), nrow=4)) knitr::kable(a, caption="Test table",  booktabs=TRUE) ``` ~
插入脚注

text^[footnote] is used to get the footnote.

where `type` may be `article`,  `book`,  `manual`,  and so on.^[The type name is case-insensitive,  so it does not matter if it is `manual`,  `Manual`,  or `MANUAL`.]
插入引文

假如我们的bib文件中内容如下,如果我们要引用这个文章,只要写 [@chen_m6a_2015]就可以了。

@article{chen_m6a_2015,    title = {m6A {RNA} {Methylation} {Is} {Regulated} by {MicroRNAs} and {Promotes} {Reprogramming} to {Pluripotency}},    volume = {16},    issn = {1934-5909, 1875-9777},    url = {http://www.cell.com/cell-stem-cell/abstract/S1934-5909(15)00017-X},    doi = {10.1016/j.stem.2015.01.016},    language = {English},    number = {3},    urldate = {2016-12-08},    journal = {Cell Stem Cell},    author = {Chen, Tong and Hao, Ya-Juan and Zhang, Ying and Li, Miao-Miao and Wang, Meng and Han, Weifang and Wu, Yongsheng and Lv, Ying and Hao, Jie and Wang, Libin and Li, Ang and Yang, Ying and Jin, Kang-Xuan and Zhao, Xu and Li, Yuhuan and Ping, Xiao-Li and Lai, Wei-Yi and Wu, Li-Gang and Jiang, Guibin and Wang, Hai-Lin and Sang, Lisi and Wang, Xiu-Jie and Yang, Yun-Gui and Zhou, Qi},    month = mar,    year = {2015},    pmid = {25683224},    pages = {289--301}, }

准备YML配置文件

_bookdown.yml

配置输入和输出文件参数。

book_filename: "输出文件的名字" output_dir: "输出目录的名字,默认_book" language:  ui:      chapter_name: ""
_output.yml

配置产生输出文件的命令行参数。

bookdown::pdf_book:  template: ehbio.tex #使用自己定制的pandoc latex模板  includes: # or only customize part latex module    in_header: preamble.tex    before_body: latex/before_body.tex    after_body: latex/after_body.tex  latex_engine: xelatex  citation_package: natbib  keep_tex: yes  pandoc_args: --chapters  toc_depth: 3  toc_unnumbered: no  toc_appendix: yes  quote_footer: ["\\VA{", "}{}"] bookdown::epub_book:  stylesheet: css/style.css bookdown::gitbook:  css: style.css  split_by: section  config:    toc:      collapse: none      before: | #设置toc开头和结尾的链接          <li><a href="http://www.ehbio.com"><img src="ehbio_logo.png" width="100%"></a></li>      after: |          <li><a href="mailto:ct@ehbio.com" target="blank">ct@ehbio.com</a></li>    download: [pdf, epub, mobi]    edit: https://github.com/rstudio/bookdown/edit/master/inst/examples/%s    sharing:      twitter: no      github: no      facebook: no

其它定制

  • 不同的文件分别用于htmlpdf输出

    # in _bookdown.yml rmd_files:  html: ["index.Rmd", "file2.Rmd"]  latex: ["index_pdf.Rmd", "file3.Rmd"] # Different render way #!/bin/sh Rscript -e "bookdown::render_book('index.Rmd', 'bookdown::gitbook')" Rscript -e "bookdown::render_book('index_pdf.Rmd', 'bookdown::pdf_book')"
  • 配置全局变量自适应HTMLPDF输出

    ```{r setup, include=FALSE} library(knitr) output <- opts_knit$get("rmarkdown.pandoc.to") html = FALSE latex = FALSE opts_chunk$set(echo = FALSE, fig.align="center", fig.show="hold") if (output=="html") {    html = TRUE } if (output=="latex") {    opts_chunk$set(out.width="95%", out.height='0.7\\textheight', out.extra='keepaspectratio', fig.pos='H')    latex = TRUE } #knitr::opts_chunk$set(cache = FALSE,  autodep=TRUE) set.seed(0304) ``` Below text will only appear in HTML output. ```{asis, echo=html} # EHBIO Gene Technology {-} ``` Below command will only be executed and displayed in HTML output. ```{r cover, eval=html, out.width="99%"} knitr::include_graphics("ehbio/cover.png") ``` ~~~~~~~~~~~~~~~~~~~~
  • 保留生成的markdown文件

    # add below lines to last Rmd file ```{r, include=FALSE} file.rename(from="bookdown_file_name.md",  to="bookdown_file_name.saved.md") ``` ~~~~~~~~~~~
  • 包含子文件 (subfile.txt)

    ```{r child="subfile.txt"} ``` ~~~~~~
  • cahce external file ref

    ```{r mtime-func} mtime <- function(files){  lapply(Sys.glob(files), function(x) file.info(x)$mtime) } ``` ```{r mtime-usage, cache=T, cache.extra=mtime(c("file1", "file2", file3))} file3 <- paste0(dir, '/', name) data1 <- read.table("file1") data2 <- read.table("file2") ``` ~~~

预览生成的WEB文件

如果没有安装Rstudio,可以在生成的book目录(有index.html的目录)下运行python -m SimpleHTTPServer 11521(11521为端口号,一般选较大值避免冲突), 然后就可以在浏览器输入网址http://server-ip:11521来访问了。

A template Rmd

Use bookdown_init_general.py -o test -t ‘我的文档’ -a “ 作者1 - 北京;2 - 南京;1521082” -b a.bib to generate a minimal template.

References

  • https://bookdown.org/yihui/bookdown/get-started.html

  • https://github.com/rstudio/bookdown/tree/master/inst/examples

  • http://stackoverflow.com/questions/25236850/how-to-set-different-global-options-in-knitr-and-rstudio-for-word-and-html

  • Multiple output with different configs https://github.com/yihui/knitr/issues/1145

  • Multiple output with different configs https://github.com/yihui/knitr/issues/114://github.com/rstudio/rmarkdown/issues/614

  • Citation style http://rmarkdown.rstudio.com/authoring_bibliographies_and_citations.html

  • Save markdown http://stackoverflow.com/questions/19989325/knit-rmd-file-to-md-and-save-the-md-file-one-level-up-with-a-different-name

您可能也对以下帖子感兴趣

文章有问题?点此查看未经处理的缓存