implement task lists

closes jgm#3051

MANUAL.txt

@@ -2612,6 +2612,13 @@ If default list markers are desired, use `#.`:
     #.  two
     #.  three
 
+#### Extension: `tasklist` ####
+
+Pandoc supports task lists, using the syntax of GitHub-Flavored Markdown.
+
+    - [ ] an unchecked task list item
+    - [x] checked item
+
 
 ### Definition lists ###
 
@@ -4223,7 +4230,7 @@ variants are supported:
 :   `pipe_tables`, `raw_html`, `fenced_code_blocks`, `auto_identifiers`,
     `gfm_auto_identifiers`, `backtick_code_blocks`,
     `autolink_bare_uris`, `space_in_atx_header`,
-    `intraword_underscores`, `strikeout`, `emoji`,
+    `intraword_underscores`, `strikeout`, `tasklist`, `emoji`,
     `shortcut_reference_links`, `angle_brackets_escapable`,
     `lists_without_preceding_blankline`.
 

src/Text/Pandoc/Extensions.hs

@@ -167,6 +167,7 @@ data Extension =
     | Ext_subscript           -- ^ Subscript using ~this~ syntax
     | Ext_superscript         -- ^ Superscript using ^this^ syntax
     | Ext_styles              -- ^ Read styles that pandoc doesn't know
+    | Ext_tasklist            -- ^ Parse certain list items as task list items
     | Ext_table_captions      -- ^ Pandoc-style table captions
     | Ext_tex_math_dollars    -- ^ TeX math between $..$ or $$..$$
     | Ext_tex_math_double_backslash  -- ^ TeX math btw \\(..\\) \\[..\\]
@@ -215,6 +216,7 @@ pandocExtensions = extensionsFromList
   , Ext_strikeout
   , Ext_superscript
   , Ext_subscript
+  , Ext_tasklist
   , Ext_auto_identifiers
   , Ext_header_attributes
   , Ext_link_attributes
@@ -274,6 +276,7 @@ githubMarkdownExtensions = extensionsFromList
   , Ext_space_in_atx_header
   , Ext_intraword_underscores
   , Ext_strikeout
+  , Ext_tasklist
   , Ext_emoji
   , Ext_lists_without_preceding_blankline
   , Ext_shortcut_reference_links

src/Text/Pandoc/Readers/Markdown.hs

@@ -946,8 +946,17 @@ listItem :: PandocMonad m
          -> MarkdownParser m a
          -> MarkdownParser m (F Blocks)
 listItem fourSpaceRule start = try $ do
-  (first, continuationIndent) <- rawListItem fourSpaceRule start
+  (firstLine, continuationIndent) <- rawListItem fourSpaceRule start
   continuations <- many (listContinuation continuationIndent)
+  exts <- getOption readerExtensions
+  let first = if Ext_tasklist `extensionEnabled` exts
+              then let (prefx, rest) = splitAt 4 firstLine
+                   in  case prefx of
+                         "[ ] " -> "☐ " <> rest
+                         "[x] " -> "☒ " <> rest
+                         "[X] " -> "☒ " <> rest
+                         _      -> firstLine
+              else firstLine
   -- parsing with ListItemState forces markers at beginning of lines to
   -- count as list item markers, even if not separated by blank space.
   -- see definition of "endline"

src/Text/Pandoc/Writers/HTML.hs

@@ -365,6 +365,24 @@ defList :: PandocMonad m
         => WriterOptions -> [Html] -> StateT WriterState m Html
 defList opts items = toList H.dl opts (items ++ [nl opts])
 
+listItemToHtml :: PandocMonad m
+               => WriterOptions -> [Block] -> StateT WriterState m Html
+listItemToHtml opts bls
+  | (Plain (Str "\9744":Space:is) : bs) <- bls = taskListItem False id  is bs
+  | (Plain (Str "\9746":Space:is) : bs) <- bls = taskListItem True  id  is bs
+  | (Para  (Str "\9744":Space:is) : bs) <- bls = taskListItem False H.p is bs
+  | (Para  (Str "\9746":Space:is) : bs) <- bls = taskListItem True  H.p is bs
+  | otherwise = blockListToHtml opts bls
+  where
+    taskListItem checked constr is bs = do
+      let checkbox  = if checked
+                      then checkbox' ! A.checked ""
+                      else checkbox'
+          checkbox' = H.input ! A.type_ "checkbox" ! A.disabled "" >> nl opts
+      isContents <- inlineListToHtml opts is
+      bsContents <- blockListToHtml opts bs
+      return $ constr (checkbox <> isContents) <> bsContents
+
 -- | Construct table of contents from list of elements.
 tableOfContents :: PandocMonad m => WriterOptions -> [Element]
                 -> StateT WriterState m (Maybe Html)
@@ -824,10 +842,10 @@ blockToHtml opts (Header level attr@(_,classes,_) lst) = do
               6 -> H.h6 contents'
               _ -> H.p contents'
 blockToHtml opts (BulletList lst) = do
-  contents <- mapM (blockListToHtml opts) lst
+  contents <- mapM (listItemToHtml opts) lst
   unordList opts contents
 blockToHtml opts (OrderedList (startnum, numstyle, _) lst) = do
-  contents <- mapM (blockListToHtml opts) lst
+  contents <- mapM (listItemToHtml opts) lst
   html5 <- gets stHtml5
   let numstyle' = case numstyle of
                        Example -> "decimal"

src/Text/Pandoc/Writers/LaTeX.hs

@@ -922,8 +922,20 @@ listItemToLaTeX lst
   -- this will keep the typesetter from throwing an error.
   | (Header{} :_) <- lst =
     blockListToLaTeX lst >>= return . (text "\\item ~" $$) . nest 2
-  | otherwise = blockListToLaTeX lst >>= return .  (text "\\item" $$) .
-                      nest 2
+  | (Plain (Str "\9744":Space:is) : bs) <- lst = taskListItem False is bs
+  | (Plain (Str "\9746":Space:is) : bs) <- lst = taskListItem True  is bs
+  | (Para  (Str "\9744":Space:is) : bs) <- lst = taskListItem False is bs
+  | (Para  (Str "\9746":Space:is) : bs) <- lst = taskListItem True  is bs
+  | otherwise = blockListToLaTeX lst >>= return . listItem
+  where
+    listItem = (text "\\item" $$) . nest 2
+    taskListItem checked is bs = do
+      let checkbox  = if checked
+                      then "$\\boxtimes$ "
+                      else "$\\square$ "
+      isContents <- inlineListToLaTeX is
+      bsContents <- blockListToLaTeX bs
+      return $ listItem $ checkbox <> isContents $+$ bsContents
 
 defListItemToLaTeX :: PandocMonad m => ([Inline], [[Block]]) -> LW m Doc
 defListItemToLaTeX (term, defs) = do

test/command/task-list-test.md

@@ -0,0 +1,102 @@
+tests adapted from <https://github.github.com/gfm/#task-list-items-extension->
+
+```
+% pandoc
+- [ ] foo
+- [x] bar
+^D
+<ul>
+<li><input type="checkbox" disabled="" />
+foo</li>
+<li><input type="checkbox" disabled="" checked="" />
+bar</li>
+</ul>
+```
+
+
+```
+% pandoc
+- [x] foo
+  - [ ] bar
+  - [x] baz
+- [ ] bim
+^D
+<ul>
+<li><input type="checkbox" disabled="" checked="" />
+foo<ul>
+<li><input type="checkbox" disabled="" />
+bar</li>
+<li><input type="checkbox" disabled="" checked="" />
+baz</li>
+</ul></li>
+<li><input type="checkbox" disabled="" />
+bim</li>
+</ul>
+```
+
+
+custom html task list test:
+
+```
+% pandoc
+- [ ]  unchecked
+- plain item
+-  [x] checked
+
+paragraph
+
+1. [ ] ordered unchecked
+2. [] plain item
+3. [x] ordered checked
+
+paragraph
+
+- [ ] list item with a
+
+    second paragraph
+
+- [x] checked
+^D
+<ul>
+<li><input type="checkbox" disabled="" />
+unchecked</li>
+<li>plain item</li>
+<li><input type="checkbox" disabled="" checked="" />
+checked</li>
+</ul>
+<p>paragraph</p>
+<ol type="1">
+<li><input type="checkbox" disabled="" />
+ordered unchecked</li>
+<li>[] plain item</li>
+<li><input type="checkbox" disabled="" checked="" />
+ordered checked</li>
+</ol>
+<p>paragraph</p>
+<ul>
+<li><p><input type="checkbox" disabled="" />
+list item with a</p><p>second paragraph</p></li>
+<li><p><input type="checkbox" disabled="" checked="" />
+checked</p></li>
+</ul>
+```
+
+latex task list test:
+
+```
+% pandoc -t latex
+- [ ] foo bar
+
+  baz
+
+- [x] boo
+^D
+\begin{itemize}
+\item
+  $\square$ foo bar
+
+  baz
+\item
+  $\boxtimes$ boo
+\end{itemize}
+```