f1d83ae · typeable/pandoc

git.iko.soy/typeable/pandoc

Implement task lists (#5139)

Closes #3051

commitf1d83aea12b93b31f5218bed75bd0e9d8d373cb6 authorMauro Bieg <mb21@users.noreply.github.com> dateWed 2 Jan 2019, 20:36 (8 years ago) parent9097ec4

11 files changed+247−20

compare with master

unifiedsplit

MANUAL.txt+9−2 src/Text/Pandoc/Extensions.hs+3 src/Text/Pandoc/Readers/CommonMark.hs+5−2 src/Text/Pandoc/Readers/Markdown.hs+2−1 src/Text/Pandoc/Shared.hs+32 src/Text/Pandoc/Writers/CommonMark.hs+16−9 src/Text/Pandoc/Writers/HTML.hs+20−2 src/Text/Pandoc/Writers/LaTeX.hs+14−2 src/Text/Pandoc/Writers/Markdown.hs+4−2 test/command/gfm.md+29 test/command/tasklist.md+113

MANUAL.txt+9−2

@@ -2612,6 +2612,13 @@ If default list markers are desired, use `#.`:

26122612 #. two

26132613 #. three

26142614

2615+#### Extension: `task_lists` ####

2616+

2617+Pandoc supports task lists, using the syntax of GitHub-Flavored Markdown.

2618+

2619+ - [ ] an unchecked task list item

2620+ - [x] checked item

2621+

26152622

26162623### Definition lists ###

26172624

@@ -4223,7 +4230,7 @@ variants are supported:

42234230: `pipe_tables`, `raw_html`, `fenced_code_blocks`, `auto_identifiers`,

42244231 `gfm_auto_identifiers`, `backtick_code_blocks`,

42254232 `autolink_bare_uris`, `space_in_atx_header`,

4226− `intraword_underscores`, `strikeout`, `emoji`,

4233+ `intraword_underscores`, `strikeout`, `task_lists`, `emoji`,

42274234 `shortcut_reference_links`, `angle_brackets_escapable`,

42284235 `lists_without_preceding_blankline`.

42294236

@@ -4254,7 +4261,7 @@ only affects `gfm` output, not input.

42544261: `pipe_tables`, `raw_html`, `fenced_code_blocks`, `auto_identifiers`,

42554262 `gfm_auto_identifiers`, `backtick_code_blocks`,

42564263 `autolink_bare_uris`, `space_in_atx_header`,

4257− `intraword_underscores`, `strikeout`, `emoji`,

4264+ `intraword_underscores`, `strikeout`, `task_lists`, `emoji`,

42584265 `shortcut_reference_links`, `angle_brackets_escapable`,

42594266 `lists_without_preceding_blankline`.

42604267

src/Text/Pandoc/Extensions.hs+3

@@ -167,6 +167,7 @@ data Extension =

167167 | Ext_subscript -- ^ Subscript using ~this~ syntax

168168 | Ext_superscript -- ^ Superscript using ^this^ syntax

169169 | Ext_styles -- ^ Read styles that pandoc doesn't know

170+ | Ext_task_lists -- ^ Parse certain list items as task list items

170171 | Ext_table_captions -- ^ Pandoc-style table captions

171172 | Ext_tex_math_dollars -- ^ TeX math between $..$ or $$..$$

172173 | Ext_tex_math_double_backslash -- ^ TeX math btw \$..\$ \\[..\\]

@@ -215,6 +216,7 @@ pandocExtensions = extensionsFromList

215216 , Ext_strikeout

216217 , Ext_superscript

217218 , Ext_subscript

219+ , Ext_task_lists

218220 , Ext_auto_identifiers

219221 , Ext_header_attributes

220222 , Ext_link_attributes

@@ -274,6 +276,7 @@ githubMarkdownExtensions = extensionsFromList

274276 , Ext_space_in_atx_header

275277 , Ext_intraword_underscores

276278 , Ext_strikeout

279+ , Ext_task_lists

277280 , Ext_emoji

278281 , Ext_lists_without_preceding_blankline

279282 , Ext_shortcut_reference_links

src/Text/Pandoc/Readers/CommonMark.hs+5−2

@@ -43,7 +43,7 @@ import Text.Pandoc.Class (PandocMonad)

4343import Text.Pandoc.Definition

4444import Text.Pandoc.Emoji (emojiToInline)

4545import Text.Pandoc.Options

46−import Text.Pandoc.Shared (uniqueIdent)

46+import Text.Pandoc.Shared (uniqueIdent, taskListItemFromAscii)

4747import Text.Pandoc.Walk (walkM)

4848

4949-- | Parse a CommonMark formatted string into a 'Pandoc' structure.

@@ -111,12 +111,14 @@ addBlock _ (Node _ (CODE_BLOCK info t) _) =

111111addBlock opts (Node _ (HEADING lev) nodes) =

112112 (Header lev ("",[],[]) (addInlines opts nodes) :)

113113addBlock opts (Node _ (LIST listAttrs) nodes) =

114− (constructor (map (setTightness . addBlocks opts . children) nodes) :)

114+ (constructor (map listItem nodes) :)

115115 where constructor = case listType listAttrs of

116116 BULLET_LIST -> BulletList

117117 ORDERED_LIST -> OrderedList

118118 (start, DefaultStyle, delim)

119119 start = listStart listAttrs

120+ listItem = taskListItemFromAscii exts . setTightness

121+ . addBlocks opts . children

120122 setTightness = if listTight listAttrs

121123 then map paraToPlain

122124 else id

@@ -125,6 +127,7 @@ addBlock opts (Node _ (LIST listAttrs) nodes) =

125127 delim = case listDelim listAttrs of

126128 PERIOD_DELIM -> Period

127129 PAREN_DELIM -> OneParen

130+ exts = readerExtensions opts

128131addBlock opts (Node _ (TABLE alignments) nodes) =

129132 (Table [] aligns widths headers rows :)

130133 where aligns = map fromTableCellAlignment alignments

src/Text/Pandoc/Readers/Markdown.hs+2−1

@@ -958,7 +958,8 @@ listItem fourSpaceRule start = try $ do

958958 let raw = concat (first:continuations)

959959 contents <- parseFromString' parseBlocks raw

960960 updateState (\st -> st {stateParserContext = oldContext})

961− return contents

961+ exts <- getOption readerExtensions

962+ return $ B.fromList . taskListItemFromAscii exts . B.toList <$> contents

962963

963964orderedList :: PandocMonad m => MarkdownParser m (F Blocks)

964965orderedList = try $ do

src/Text/Pandoc/Shared.hs+32

@@ -79,6 +79,8 @@ module Text.Pandoc.Shared (

7979 headerShift,

8080 stripEmptyParagraphs,

8181 isTightList,

82+ taskListItemFromAscii,

83+ taskListItemToAscii,

8284 addMetaField,

8385 makeMeta,

8486 eastAsianLineBreakFilter,

@@ -588,6 +590,36 @@ isTightList = all firstIsPlain

588590 where firstIsPlain (Plain _ : _) = True

589591 firstIsPlain _ = False

590592

593+-- | Convert a list item containing tasklist syntax (e.g. @[x]@)

594+-- to using @U+2610 BALLOT BOX@ or @U+2612 BALLOT BOX WITH X@.

595+taskListItemFromAscii :: Extensions -> [Block] -> [Block]

596+taskListItemFromAscii = handleTaskListItem fromMd

597+ where

598+ fromMd (Str "[" : Space : Str "]" : Space : is) = (Str "☐") : Space : is

599+ fromMd (Str "[x]" : Space : is) = (Str "☒") : Space : is

600+ fromMd (Str "[X]" : Space : is) = (Str "☒") : Space : is

601+ fromMd is = is

602+

603+-- | Convert a list item containing text starting with @U+2610 BALLOT BOX@

604+-- or @U+2612 BALLOT BOX WITH X@ to tasklist syntax (e.g. @[x]@).

605+taskListItemToAscii :: Extensions -> [Block] -> [Block]

606+taskListItemToAscii = handleTaskListItem toMd

607+ where

608+ toMd (Str "☐" : Space : is) = rawMd "[ ]" : Space : is

609+ toMd (Str "☒" : Space : is) = rawMd "[x]" : Space : is

610+ toMd is = is

611+ rawMd = RawInline (Format "markdown")

612+

613+handleTaskListItem :: ([Inline] -> [Inline]) -> Extensions -> [Block] -> [Block]

614+handleTaskListItem handleInlines exts bls =

615+ if Ext_task_lists `extensionEnabled` exts

616+ then handleItem bls

617+ else bls

618+ where

619+ handleItem (Plain is : bs) = Plain (handleInlines is) : bs

620+ handleItem (Para is : bs) = Para (handleInlines is) : bs

621+ handleItem bs = bs

622+

591623-- | Set a field of a 'Meta' object. If the field already has a value,

592624-- convert it into a list with the new value appended to the old value(s).

593625addMetaField :: ToMetaValue a

src/Text/Pandoc/Writers/CommonMark.hs+16−9

@@ -46,7 +46,8 @@ import Network.HTTP (urlEncode)

4646import Text.Pandoc.Class (PandocMonad)

4747import Text.Pandoc.Definition

4848import Text.Pandoc.Options

49−import Text.Pandoc.Shared (isTightList, linesToPara, substitute, capitalize)

49+import Text.Pandoc.Shared (isTightList, taskListItemToAscii, linesToPara,

50+ substitute, capitalize)

5051import Text.Pandoc.Templates (renderTemplate')

5152import Text.Pandoc.Walk (query, walk, walkM)

5253import Text.Pandoc.Writers.HTML (writeHtml5String, tagWithAttributes)

@@ -115,24 +116,28 @@ blockToNodes opts (Para xs) ns =

115116blockToNodes opts (LineBlock lns) ns = blockToNodes opts (linesToPara lns) ns

116117blockToNodes _ (CodeBlock (_,classes,_) xs) ns = return

117118 (node (CODE_BLOCK (T.pack (unwords classes)) (T.pack xs)) [] : ns)

118−blockToNodes opts (RawBlock fmt xs) ns

119− | fmt == Format "html" && isEnabled Ext_raw_html opts

119+blockToNodes opts (RawBlock (Format f) xs) ns

120+ | f == "html" && isEnabled Ext_raw_html opts

120121 = return (node (HTML_BLOCK (T.pack xs)) [] : ns)

121− | (fmt == Format "latex" || fmt == Format "tex") && isEnabled Ext_raw_tex opts

122+ | (f == "latex" || f == "tex") && isEnabled Ext_raw_tex opts

123+ = return (node (CUSTOM_BLOCK (T.pack xs) T.empty) [] : ns)

124+ | f == "markdown"

122125 = return (node (CUSTOM_BLOCK (T.pack xs) T.empty) [] : ns)

123126 | otherwise = return ns

124127blockToNodes opts (BlockQuote bs) ns = do

125128 nodes <- blocksToNodes opts bs

126129 return (node BLOCK_QUOTE nodes : ns)

127130blockToNodes opts (BulletList items) ns = do

128− nodes <- mapM (blocksToNodes opts) items

131+ let exts = writerExtensions opts

132+ nodes <- mapM (blocksToNodes opts . taskListItemToAscii exts) items

129133 return (node (LIST ListAttributes{

130134 listType = BULLET_LIST,

131135 listDelim = PERIOD_DELIM,

132136 listTight = isTightList items,

133137 listStart = 1 }) (map (node ITEM) nodes) : ns)

134138blockToNodes opts (OrderedList (start, _sty, delim) items) ns = do

135− nodes <- mapM (blocksToNodes opts) items

139+ let exts = writerExtensions opts

140+ nodes <- mapM (blocksToNodes opts . taskListItemToAscii exts) items

136141 return (node (LIST ListAttributes{

137142 listType = ORDERED_LIST,

138143 listDelim = case delim of

@@ -292,10 +297,12 @@ inlineToNodes opts (Image alt ils (url,'f':'i':'g':':':tit)) =

292297 inlineToNodes opts (Image alt ils (url,tit))

293298inlineToNodes opts (Image _ ils (url,tit)) =

294299 (node (IMAGE (T.pack url) (T.pack tit)) (inlinesToNodes opts ils) :)

295−inlineToNodes opts (RawInline fmt xs)

296− | fmt == Format "html" && isEnabled Ext_raw_html opts

300+inlineToNodes opts (RawInline (Format f) xs)

301+ | f == "html" && isEnabled Ext_raw_html opts

297302 = (node (HTML_INLINE (T.pack xs)) [] :)

298− | (fmt == Format "latex" || fmt == Format "tex") && isEnabled Ext_raw_tex opts

303+ | (f == "latex" || f == "tex") && isEnabled Ext_raw_tex opts

304+ = (node (CUSTOM_INLINE (T.pack xs) T.empty) [] :)

305+ | f == "markdown"

299306 = (node (CUSTOM_INLINE (T.pack xs) T.empty) [] :)

300307 | otherwise = id

301308inlineToNodes opts (Quoted qt ils) =

src/Text/Pandoc/Writers/HTML.hs+20−2

@@ -365,6 +365,24 @@ defList :: PandocMonad m

365365 => WriterOptions -> [Html] -> StateT WriterState m Html

366366defList opts items = toList H.dl opts (items ++ [nl opts])

367367

368+listItemToHtml :: PandocMonad m

369+ => WriterOptions -> [Block] -> StateT WriterState m Html

370+listItemToHtml opts bls

371+ | Plain (Str "☐":Space:is) : bs <- bls = taskListItem False id is bs

372+ | Plain (Str "☒":Space:is) : bs <- bls = taskListItem True id is bs

373+ | Para (Str "☐":Space:is) : bs <- bls = taskListItem False H.p is bs

374+ | Para (Str "☒":Space:is) : bs <- bls = taskListItem True H.p is bs

375+ | otherwise = blockListToHtml opts bls

376+ where

377+ taskListItem checked constr is bs = do

378+ let checkbox = if checked

379+ then checkbox' ! A.checked ""

380+ else checkbox'

381+ checkbox' = H.input ! A.type_ "checkbox" ! A.disabled "" >> nl opts

382+ isContents <- inlineListToHtml opts is

383+ bsContents <- blockListToHtml opts bs

384+ return $ constr (checkbox >> isContents) >> bsContents

385+

368386-- | Construct table of contents from list of elements.

369387tableOfContents :: PandocMonad m => WriterOptions -> [Element]

370388 -> StateT WriterState m (Maybe Html)

@@ -824,10 +842,10 @@ blockToHtml opts (Header level attr@(_,classes,_) lst) = do

824842 6 -> H.h6 contents'

825843 _ -> H.p contents'

826844blockToHtml opts (BulletList lst) = do

827− contents <- mapM (blockListToHtml opts) lst

845+ contents <- mapM (listItemToHtml opts) lst

828846 unordList opts contents

829847blockToHtml opts (OrderedList (startnum, numstyle, _) lst) = do

830− contents <- mapM (blockListToHtml opts) lst

848+ contents <- mapM (listItemToHtml opts) lst

831849 html5 <- gets stHtml5

832850 let numstyle' = case numstyle of

833851 Example -> "decimal"

src/Text/Pandoc/Writers/LaTeX.hs+14−2

@@ -924,8 +924,20 @@ listItemToLaTeX lst

924924 -- this will keep the typesetter from throwing an error.

925925 | (Header{} :_) <- lst =

926926 blockListToLaTeX lst >>= return . (text "\\item ~" $$) . nest 2

927− | otherwise = blockListToLaTeX lst >>= return . (text "\\item" $$) .

928− nest 2

927+ | Plain (Str "☐":Space:is) : bs <- lst = taskListItem False is bs

928+ | Plain (Str "☒":Space:is) : bs <- lst = taskListItem True is bs

929+ | Para (Str "☐":Space:is) : bs <- lst = taskListItem False is bs

930+ | Para (Str "☒":Space:is) : bs <- lst = taskListItem True is bs

931+ | otherwise = blockListToLaTeX lst >>= return . (text "\\item" $$) . nest 2

932+ where

933+ taskListItem checked is bs = do

934+ let checkbox = if checked

935+ then "$\\boxtimes$"

936+ else "$\\square$"

937+ isContents <- inlineListToLaTeX is

938+ bsContents <- blockListToLaTeX bs

939+ return $ "\\item" <> brackets checkbox

940+ $$ nest 2 (isContents $+$ bsContents)

929941

930942defListItemToLaTeX :: PandocMonad m => ([Inline], [[Block]]) -> LW m Doc

931943defListItemToLaTeX (term, defs) = do

src/Text/Pandoc/Writers/Markdown.hs+4−2

@@ -765,7 +765,8 @@ itemEndsWithTightList bs =

765765-- | Convert bullet list item (list of blocks) to markdown.

766766bulletListItemToMarkdown :: PandocMonad m => WriterOptions -> [Block] -> MD m Doc

767767bulletListItemToMarkdown opts bs = do

768− contents <- blockListToMarkdown opts bs

768+ let exts = writerExtensions opts

769+ contents <- blockListToMarkdown opts $ taskListItemToAscii exts bs

769770 let sps = replicate (writerTabStop opts - 2) ' '

770771 let start = text ('-' : ' ' : sps)

771772 -- remove trailing blank line if item ends with a tight list

@@ -781,7 +782,8 @@ orderedListItemToMarkdown :: PandocMonad m

781782 -> [Block] -- ^ list item (list of blocks)

782783 -> MD m Doc

783784orderedListItemToMarkdown opts marker bs = do

784− contents <- blockListToMarkdown opts bs

785+ let exts = writerExtensions opts

786+ contents <- blockListToMarkdown opts $ taskListItemToAscii exts bs

785787 let sps = case length marker - writerTabStop opts of

786788 n | n > 0 -> text $ replicate n ' '

787789 _ -> text " "

test/command/gfm.md+29

@@ -101,3 +101,32 @@ hi

101101^D

102102[Para [Str "hi",LineBreak,Str "hi"]]

103103```

104+

105+```

106+% pandoc -f gfm -t native

107+- [ ] foo

108+- [x] bar

109+^D

110+[BulletList

111+ [[Plain [Str "\9744",Space,Str "foo"]]

112+ ,[Plain [Str "\9746",Space,Str "bar"]]]]

113+```

114+

115+```

116+% pandoc -f gfm-task_lists -t native

117+- [ ] foo

118+- [x] bar

119+^D

120+[BulletList

121+ [[Plain [Str "[",Space,Str "]",Space,Str "foo"]]

122+ ,[Plain [Str "[x]",Space,Str "bar"]]]]

123+```

124+

125+```

126+% pandoc -f gfm -t gfm

127+- [ ] foo

128+- [x] bar

129+^D

130+ - [ ] foo

131+ - [x] bar

132+```

test/command/tasklist.mdnew file+113

@@ -0,0 +1,113 @@

1+tests adapted from <https://github.github.com/gfm/#task-list-items-extension->

3+```

4+% pandoc

5+- [ ] foo

6+- [x] bar

7+^D

8+<ul>

9+<li><input type="checkbox" disabled="" />

10+foo</li>

11+<li><input type="checkbox" disabled="" checked="" />

12+bar</li>

13+</ul>

14+```

15+

16+

17+```

18+% pandoc

19+- [x] foo

20+ - [ ] bar

21+ - [x] baz

22+- [ ] bim

23+^D

24+<ul>

25+<li><input type="checkbox" disabled="" checked="" />

26+foo<ul>

27+<li><input type="checkbox" disabled="" />

28+bar</li>

29+<li><input type="checkbox" disabled="" checked="" />

30+baz</li>

31+</ul></li>

32+<li><input type="checkbox" disabled="" />

33+bim</li>

34+</ul>

35+```

36+

37+

38+custom html task list test:

39+

40+```

41+% pandoc

42+- [ ] unchecked

43+- plain item

44+- [x] checked

45+

46+paragraph

47+

48+1. [ ] ordered unchecked

49+2. [] plain item

50+3. [x] ordered checked

51+

52+paragraph

53+

54+- [ ] list item with a

55+

56+ second paragraph

57+

58+- [x] checked

59+^D

60+<ul>

61+<li><input type="checkbox" disabled="" />

62+unchecked</li>

63+<li>plain item</li>

64+<li><input type="checkbox" disabled="" checked="" />

65+checked</li>

66+</ul>

67+<p>paragraph</p>

68+<ol type="1">

69+<li><input type="checkbox" disabled="" />

70+ordered unchecked</li>

71+<li>[] plain item</li>

72+<li><input type="checkbox" disabled="" checked="" />

73+ordered checked</li>

74+</ol>

75+<p>paragraph</p>

76+<ul>

77+<li><p><input type="checkbox" disabled="" />

78+list item with a</p><p>second paragraph</p></li>

79+<li><p><input type="checkbox" disabled="" checked="" />

80+checked</p></li>

81+</ul>

82+```

83+

84+latex task list test:

85+

86+```

87+% pandoc -t latex

88+- [ ] foo bar

89+

90+ baz

91+

92+- [x] ok

93+^D

94+\begin{itemize}

95+\item[$\square$]

96+ foo bar

97+

98+ baz

99+\item[$\boxtimes$]

100+ ok

101+\end{itemize}

102+```

103+

104+round trip:

105+

106+```

107+% pandoc -f markdown -t markdown

108+- [ ] foo

109+- [x] bar

110+^D

111+- [ ] foo

112+- [x] bar

113+```