switch some vignettes to litedown

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: knitr
 Type: Package
 Title: A General-Purpose Package for Dynamic Report Generation in R
-Version: 1.48.3
+Version: 1.48.4
 Authors@R: c(
     person("Yihui", "Xie", role = c("aut", "cre"), email = "[email protected]", comment = c(ORCID = "0000-0003-0645-5666")),
     person("Abhraneel", "Sarma", role = "ctb"),
@@ -134,6 +134,7 @@ Suggests:
     jpeg,
     JuliaCall (>= 0.11.1),
     magick,
+    litedown,
     markdown (>= 1.3),
     png,
     ragg,
@@ -156,7 +157,7 @@ License: GPL
 URL: https://yihui.org/knitr/
 BugReports: https://github.com/yihui/knitr/issues
 Encoding: UTF-8
-VignetteBuilder: knitr
+VignetteBuilder: litedown, knitr
 SystemRequirements: Package vignettes based on R Markdown v2 or reStructuredText require Pandoc (http://pandoc.org).
     The function rst2pdf() requires rst2pdf (https://github.com/rst2pdf/rst2pdf).
 Collate:

NEWS.md

@@ -16,6 +16,8 @@
 
 - Changed the format of the reference card from PDF to HTML so building this package will not require LaTeX. See `vignette('knitr-refcard', package = 'knitr')`.
 
+- Switched the vignette engine from **knitr** to `litedown::vignette` for some package vignettes.
+
 # CHANGES IN knitr VERSION 1.48
 
 ## BUG FIXES

R/output.R

@@ -132,7 +132,7 @@ knit = function(
   oconc = knit_concord$get(); on.exit(knit_concord$set(oconc), add = TRUE)
 
   if (child_mode()) {
-    setwd(opts_knit$get('output.dir')) # always restore original working dir
+    setwd(opts_knit$get('output.dir') %n% '.') # always restore original working dir
     # in child mode, input path needs to be adjusted
     if (in.file && !is_abs_path(input)) {
       input = paste0(opts_knit$get('child.path'), input)

vignettes/datatables.Rmd

@@ -1,40 +1,41 @@
 ---
 title: Display Tables with the JavaScript Library DataTables
 author: Yihui Xie
-date: "`r Sys.Date()`"
+date: "`{r} Sys.Date()`"
 output:
-  markdown::html_format:
+  litedown::html_format:
     meta:
-      css: ["@default", "@prism-xcode", "https://cdn.datatables.net/1.13.7/css/jquery.dataTables.min.css"]
+      css: ["@default", "https://cdn.datatables.net/1.13.7/css/jquery.dataTables.min.css"]
       js: ["@npm/[email protected]/dist/jquery.min.js", "https://cdn.datatables.net/1.13.7/js/jquery.dataTables.min.js"]
 ---
 
 <!--
-%\VignetteEngine{knitr::knitr_notangle}
+%\VignetteEngine{litedown::vignette}
 %\VignetteIndexEntry{Display Tables with the JavaScript Library DataTables}
 -->
 
 ## jQuery DataTables
 
 We can use the JavaScript library [**DataTables**](https://datatables.net) to generate enhanced tables in HTML. In the example below, we create a table for the `mtcars` data:
 
-```{r cool, results='asis'}
-library(knitr)
-kable(mtcars, 'html', table.attr='id="mtcars_table"')
+::: {#mtcars-table}
+```{r cool, print.args=list(data.frame=list(limit=NULL))}
+mtcars
 ```
+:::
 
 Note we assigned an `id` to the table, and next we use the **DataTables** library to initialize the table and you will get an interactive table.
 
 ```{js}
 window.addEventListener('load', () => {
-  $('#mtcars_table').DataTable();
+  $('#mtcars-table > table').DataTable();
 });
 ```
 
 By comparison, below is an ordinary table:
 
-```{r boring, results='asis'}
-kable(head(mtcars), 'html')
+```{r boring}
+mtcars
 ```
 
 This vignette is only a toy example. I'd recommend you to use the **DT** package instead: https://github.com/rstudio/DT

vignettes/knit_expand.Rmd

@@ -1,15 +1,15 @@
 ---
 title: Templating with knit_expand()
 author: Yihui Xie
-date: "`r Sys.Date()`"
+date: "`{r} Sys.Date()`"
 output:
-  markdown::html_format:
+  litedown::html_format:
     meta:
-      css: ["@default", "@prism-xcode"]
+      css: ["@default"]
 ---
 
 <!--
-%\VignetteEngine{knitr::knitr_notangle}
+%\VignetteEngine{litedown::vignette}
 %\VignetteIndexEntry{Templating with knit_expand()}
 -->
 
@@ -56,14 +56,12 @@ res = knit_expand(text = c('{{i=0;h2=function(x){i<<-i+1;sprintf("<h2>%d. %s</h2
 cat(res)
 ```
 
-Build regression models based on a template; loop through all variables in `mtcars`:
+Build regression models based on a template; loop through some variables in `mtcars`:
 
-```{r results='hide'}
-src = lapply(names(mtcars)[-1], function(i) {
-knit_expand(text=c("# Regression on {{i}}", '```{r lm-{{i}}}', 'lm(mpg~{{i}}, data=mtcars)', '```'))
+```{r, comment=''}
+src = lapply(names(mtcars)[2:5], function(i) {
+knit_expand(text=c("# Regression on {{i}}", '```{r lm-{{i}}}', 'lm(mpg~{{i}}, data=mtcars)', '```', ''))
 })
 # knit the source
-res = knit_child(text = unlist(src))
-res = paste('<pre><code>', gsub('^\\s*|\\s*$', '', res), '</code></pre>', sep = '')
+litedown::fuse(unlist(src), 'markdown')
 ```
-`r res`

vignettes/knit_print.Rmd

@@ -3,11 +3,11 @@ title: "Custom Print Methods"
 author: "Yihui Xie"
 date: '`r Sys.Date()`'
 output:
-  markdown::html_format:
+  litedown::html_format:
     options:
       toc: true
     meta:
-      css: ["@default", "@prism-xcode"]
+      css: ["@default"]
 vignette: >
   %\VignetteEngine{knitr::knitr_notangle}
   %\VignetteIndexEntry{Custom Print Methods}

vignettes/knitr-intro.Rmd

@@ -1,14 +1,13 @@
 ---
 title: "Not An Introduction to knitr"
 author: "Yihui Xie"
-date: "`r Sys.Date()`"
+date: "`{r} Sys.Date()`"
 output:
-  markdown::html_format:
+  litedown::html_format:
     meta:
-      css: ["@default", "@prism-xcode"]
-      js: ["@npm/@xiee/utils/js/center-img.min.js"]
+      css: ["@default"]
 vignette: >
-  %\VignetteEngine{knitr::knitr_notangle}
+  %\VignetteEngine{litedown::vignette}
   %\VignetteIndexEntry{Not an Introduction to knitr}
 ---
 
@@ -24,18 +23,18 @@ and knitr" for this package, too.
 Anyway, here is a code chunk that shows you can compile vignettes with **knitr**
 as well using R 3.0.x, which supports non-Sweave vignettes:
 
-```{r show-off, tidy=TRUE}
-options(digits=4)
+```{r show-off}
+options(digits = 4)
 rnorm(20)
-fit=lm(dist~speed, data=cars)
-b=coef(fit)
+fit = lm(dist ~ speed, data = cars)
+b = coef(fit)
 ```
 
 ```{r echo=FALSE}
-knitr::kable(summary(fit)$coefficients)
+summary(fit)$coefficients
 ```
 
-The fitted regression equation is $Y=`r b[1]`+`r b[2]`x$.
+The fitted regression equation is $Y=`{r} b[1]`+`{r} b[2]`x$.
 
 ```{r graphics, fig.cap='A scatterplot with a regression line.'}
 par(mar=c(4, 4, 1, .1))

vignettes/knitr-markdown.Rmd

@@ -1,15 +1,19 @@
 ---
-title: An R Markdown (v1) Vignette with knitr
+title: R Markdown (v1) Vignettes with knitr
 author: Yihui Xie
-date: "`r Sys.Date()`"
+date: "`{r} Sys.Date()`"
+output:
+  litedown::html_format:
+    meta:
+      css: ["@default"]
 ---
 
 <!--
-%\VignetteEngine{knitr::knitr_notangle}
+%\VignetteEngine{litedown::vignette}
 %\VignetteIndexEntry{An R Markdown Vignette with knitr}
 -->
 
-This is an example of Markdown vignettes in R. Before R 3.0.0, only Sweave/PDF vignettes were supported in R. Markdown is gaining popularity over the years due to its simplicity, and R 3.0.0 starts to support package vignettes written in R Markdown.
+Before R 3.0.0, only Sweave/PDF vignettes were supported in R. Markdown is gaining popularity over the years due to its simplicity, and R 3.0.0 starts to support package vignettes written in R Markdown.
 
 > Please note this example is for [R Markdown v1](https://cran.r-project.org/package=markdown) only. If you use [R Markdown v2](https://rmarkdown.rstudio.com), you should use the vignette engine `knitr::rmarkdown` instead of `knitr::knitr`.