make spin() recognize #| comments as code chunks (#2320)

Co-authored-by: Yihui Xie <xie@yihui.name>

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: knitr
 Type: Package
 Title: A General-Purpose Package for Dynamic Report Generation in R
-Version: 1.45.14
+Version: 1.45.15
 Authors@R: c(
     person("Yihui", "Xie", role = c("aut", "cre"), email = "xie@yihui.name", comment = c(ORCID = "0000-0003-0645-5666")),
     person("Abhraneel", "Sarma", role = "ctb"),

NEWS.md

@@ -6,6 +6,8 @@
 
 - `knitr::spin()` now recognizes `# %%` as a valid code chunk delimiter (thanks, @kylebutts, #2307).
 
+- `knitr::spin()` also recognizes `#|` comments as code chunks now (thanks, @kylebutts, #2320).
+
 - Chunk hooks can have the `...` argument now. Previously, only arguments `before`, `options`, `envir`, and `name` are accepted. If a chunk hook function has the `...` argument, all the aforementioned four arguments are passed to the hook. This means the hook function no longer has to have the four arguments explicitly in its signature, e.g., `function(before, ...)` is also valid if only the `before` argument is used inside the hook. See <https://yihui.org/knitr/hooks/#chunk-hooks> for more information.
 
 - For package vignettes, PNG plots will be optimized by `optipng` and `pngquant` if they are installed and can be found from the system `PATH` variable. This can help reduce the package size if vignettes contain PNG plots (thanks, @nanxstats, <https://nanx.me/blog/post/rpkgs-pngquant-ragg/>).

R/spin.R

@@ -3,8 +3,7 @@
 #' This function takes a specially formatted R script and converts it to a
 #' literate programming document. By default normal text (documentation) should
 #' be written after the roxygen comment (\code{#'}) and code chunk options are
-#' written after \code{#+} or \code{# \%\%} or \code{#-} or \code{# ----} or
-#' any of these combinations replacing \code{#} with \code{--}.
+#' written after \code{#|} or \code{#+} or \code{# \%\%} or \code{# ----}.
 #'
 #' Obviously the goat's hair is the original R script, and the wool is the
 #' literate programming document (ready to be knitted).
@@ -79,7 +78,6 @@ spin = function(
 
   r = rle((is_matchable & grepl(doc, x)) | i)  # inline expressions are treated as doc instead of code
   n = length(r$lengths); txt = vector('list', n); idx = c(0L, cumsum(r$lengths))
-  p1 = gsub('\\{', '\\\\{', paste0('^', p[1L], '.*', p[2L], '$'))
 
   for (i in seq_len(n)) {
     block = x[seq(idx[i] + 1L, idx[i + 1])]
@@ -90,18 +88,24 @@ spin = function(
       # R code; #+/- indicates chunk options
       block = strip_white(block) # rm white lines in beginning and end
       if (!length(block)) next
-      rc <- '^(#|--)+(\\+|-|\\s+%%| ----+| @knitr)'
-      if (length(opt <- grep(rc, block))) {
-        opts = gsub(paste0(rc, '\\s*|-*\\s*$'), '', block[opt])
-        opts = paste0(ifelse(opts == '', '', ' '), opts)
-        block[opt] = paste0(p[1L], opts, p[2L])
+
+      rc = '^(#|--)+(\\+|-| %%| ----+| @knitr)(.*?)\\s*-*\\s*$'
+      j1 = grep(rc, block)
+      # pipe comments (#|) should start a code chunk if they are not preceded by
+      # chunk opening tokens
+      j2 = setdiff(pipe_comment_start(block), j1 + 1)
+
+      if (length(j3 <- c(j1, j2))) {
+        block[j1] = paste0(p[1], gsub(rc, '\\3', block[j1]), p[2])
+        block[j2] = paste0(p[1], p[2], '\n', block[j2])
+
         # close each chunk if there are multiple chunks in this block
-        if (any(opt > 1)) {
-          j = opt[opt > 1]
-          block[j] = paste(p[3L], block[j], sep = '\n')
+        if (any(j3 > 1)) {
+          j = j3[j3 > 1]
+          block[j] = paste0(p[3], '\n', block[j])
         }
       }
-      if (!grepl(p1, block[1L])) {
+      if (!startsWith(block[1L], p[1L])) {
         block = c(paste0(p[1L], p[2L]), block)
       }
       c('', block, p[3L], '')
@@ -155,6 +159,15 @@ spin = function(
   c(paste0(b, '{r'), '}', b, paste0(i, 'r \\1 ', i))
 }
 
+# find the position of the starting `#|` in a consecutive block of `#|` comments
+pipe_comment_start = function(x) {
+  i = startsWith(x, '#| ')
+  r = rle(i)
+  l = r$lengths
+  j = cumsum(l) - l + 1
+  j[r$values]
+}
+
 #' Spin a child R script
 #'
 #' This function is similar to \code{\link{knit_child}()} but is used in R

tests/testit/test-spin.R

@@ -1,32 +1,47 @@
 library(testit)
 
-spin_w_tempfile = function(..., format = "Rmd") {
-  tmp = tempfile(fileext = ".R")
-  writeLines(c(...), tmp)
-  spinned = spin(tmp, knit = FALSE, format = format)
-  result = readLines(spinned)
-  file.remove(c(tmp, spinned))
-  result
+spin_text = function(..., format = "Rmd") {
+  x = spin(text = c(...), knit = FALSE, format = format)
+  xfun::split_lines(x)
 }
 
 assert("spin() detects lines for documentation", {
-  (spin_w_tempfile("#' test", "1 * 1", "#' test") %==%
+  (spin_text("#' test", "1 * 1", "#' test") %==%
      c("test", "", "```{r}", "1 * 1", "```", "", "test"))
   # a multiline string literal contains the pattern of doc or inline
-  (spin_w_tempfile("code <- \"", "#' test\"") %==%
+  (spin_text("code <- \"", "#' test\"") %==%
     c("", "```{r}", "code <- \"", "#' test\"", "```", ""))
-  (spin_w_tempfile("code <- \"", "{{ 1 + 1 }}", "\"") %==%
+  (spin_text("code <- \"", "{{ 1 + 1 }}", "\"") %==%
     c("", "```{r}", "code <- \"", "{{ 1 + 1 }}", "\"", "```", ""))
   # a multiline symbol contains the pattern of doc or inline
-  (spin_w_tempfile("`", "#' test", "`") %==%
+  (spin_text("`", "#' test", "`") %==%
     c("", "```{r}", "`", "#' test", "`", "```", ""))
-  (spin_w_tempfile("`", "{{ 1 + 1 }}", "`") %==%
+  (spin_text("`", "{{ 1 + 1 }}", "`") %==%
     c("", "```{r}", "`", "{{ 1 + 1 }}", "`", "```", ""))
 })
 
 assert("spin() uses proper number of backticks", {
-  (spin_w_tempfile("{{ '`' }}") %==% c("``r  '`'  ``"))
-  (spin_w_tempfile("{{`x`}}") %==% c("``r `x` ``"))
-  (spin_w_tempfile("x <- '", "```", "'") %==%
+  (spin_text("{{ '`' }}") %==% c("``r  '`'  ``"))
+  (spin_text("{{`x`}}") %==% c("``r `x` ``"))
+  (spin_text("x <- '", "```", "'") %==%
     c("", "````{r}", "x <- '", "```", "'", "````", ""))
 })
+
+assert("spin() generates code chunks with pipe comments `#|`", {
+  (
+    spin_text("", "#| echo: false", "#| message: false", "#| include: false", "1+1", "#| eval: false", "2 + 2", "", "#' Text") %==%
+    c('', '```{r}', '#| echo: false', '#| message: false', '#| include: false', '1+1', '```', '```{r}', '#| eval: false', '2 + 2', '```', '', 'Text')
+  )
+
+  # https://github.com/yihui/knitr/issues/2314
+  (
+    spin_text('#| echo: false', '1+1', '#| label: test', '1+1') %==%
+    c('', '```{r}', '#| echo: false', '1+1', '```', '```{r}', '#| label: test', '1+1', '```', '')
+  )
+
+  # Has a `# %%` already
+  (
+    spin_text('# %%', '#| echo: false', '1+1', '#| label: test', '1+1') %==%
+    c('', '```{r}', '#| echo: false', '1+1', '```', '```{r}', '#| label: test', '1+1', '```', '')
+  )
+})