git.iko.soy/typeable/pandoc
on master help sign in
Merge branch 'groff_reader' of https://github.com/Yanpas/pandoc into Yanpas-groff_reader
5c42101John MacFarlane
pandoc.cabal+3
compare with master
unifiedsplit
@@ -206,6 +206,7 @@ extra-source-files:
206 test/haddock-reader.haddock
206 test/haddock-reader.haddock
207 test/insert
207 test/insert
208 test/lalune.jpg
208 test/lalune.jpg
209+ test/man-reader.man
209 test/movie.jpg
210 test/movie.jpg
210 test/media/rId25.jpg
211 test/media/rId25.jpg
211 test/media/rId26.jpg
212 test/media/rId26.jpg
@@ -448,6 +449,7 @@ library
448 Text.Pandoc.Readers.Odt,
449 Text.Pandoc.Readers.Odt,
449 Text.Pandoc.Readers.EPUB,
450 Text.Pandoc.Readers.EPUB,
450 Text.Pandoc.Readers.Muse,
451 Text.Pandoc.Readers.Muse,
452+ Text.Pandoc.Readers.Man,
451 Text.Pandoc.Readers.FB2,
453 Text.Pandoc.Readers.FB2,
452 Text.Pandoc.Writers,
454 Text.Pandoc.Writers,
453 Text.Pandoc.Writers.Native,
455 Text.Pandoc.Writers.Native,
@@ -669,6 +671,7 @@ test-suite test-pandoc
669 Tests.Readers.EPUB
671 Tests.Readers.EPUB
670 Tests.Readers.Muse
672 Tests.Readers.Muse
671 Tests.Readers.Creole
673 Tests.Readers.Creole
674+ Tests.Readers.Man
672 Tests.Readers.FB2
675 Tests.Readers.FB2
673 Tests.Writers.Native
676 Tests.Writers.Native
674 Tests.Writers.ConTeXt
677 Tests.Writers.ConTeXt
src/Text/Pandoc/Readers.hs+2
@@ -105,6 +105,7 @@ import Text.Pandoc.Readers.TikiWiki
105import Text.Pandoc.Readers.TWiki
105import Text.Pandoc.Readers.TWiki
106import Text.Pandoc.Readers.Txt2Tags
106import Text.Pandoc.Readers.Txt2Tags
107import Text.Pandoc.Readers.Vimwiki
107import Text.Pandoc.Readers.Vimwiki
108+import Text.Pandoc.Readers.Man
108import Text.Pandoc.Shared (mapLeft)
109import Text.Pandoc.Shared (mapLeft)
109import qualified Text.Pandoc.UTF8 as UTF8
110import qualified Text.Pandoc.UTF8 as UTF8
110import Text.Parsec.Error
111import Text.Parsec.Error
@@ -145,6 +146,7 @@ readers = [ ("native" , TextReader readNative)
145 ,("t2t" , TextReader readTxt2Tags)
146 ,("t2t" , TextReader readTxt2Tags)
146 ,("epub" , ByteStringReader readEPUB)
147 ,("epub" , ByteStringReader readEPUB)
147 ,("muse" , TextReader readMuse)
148 ,("muse" , TextReader readMuse)
149+ ,("man" , TextReader readMan)
148 ,("fb2" , TextReader readFB2)
150 ,("fb2" , TextReader readFB2)
149 ]
151 ]
150
152
src/Text/Pandoc/Readers/Man.hsnew file+559
@@ -0,0 +1,559 @@
1+{-# LANGUAGE FlexibleContexts #-}
2+{-
3+ Copyright (C) 2018 Yan Pashkovsky <yanp.bugz@gmail.com>
4+
5+This program is free software; you can redistribute it and/or modify
6+it under the terms of the GNU General Public License as published by
7+the Free Software Foundation; either version 2 of the License, or
8+(at your option) any later version.
9+
10+This program is distributed in the hope that it will be useful,
11+but WITHOUT ANY WARRANTY; without even the implied warranty of
12+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13+GNU General Public License for more details.
14+
15+You should have received a copy of the GNU General Public License
16+along with this program; if not, write to the Free Software
17+Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
18+
19+
20+
21+-}
22+
23+{- |
24+ Module : Text.Pandoc.Readers.Man
25+ Copyright : Copyright (C) 2018 Yan Pashkovsky
26+ License : GNU GPL, version 2 or above
27+
28+ Maintainer : Yan Pashkovsky <yanp.bugz@gmail.com>
29+ Stability : WIP
30+ Portability : portable
31+
32+Conversion of man to 'Pandoc' document.
33+-}
34+module Text.Pandoc.Readers.Man (readMan) where
35+
36+import Prelude
37+import Control.Monad (liftM, void)
38+import Control.Monad.Except (throwError)
39+import Data.Char (isDigit, isUpper, isLower)
40+import Data.Default (Default)
41+import Data.Map (insert)
42+import Data.Set (Set, singleton)
43+import qualified Data.Set as S (fromList, toList)
44+import Data.Maybe (catMaybes, fromMaybe, isNothing)
45+import Data.List (intersperse, intercalate)
46+import qualified Data.Text as T
47+
48+import Text.Pandoc.Class (PandocMonad(..))
49+import Text.Pandoc.Builder as B hiding (singleton)
50+import Text.Pandoc.Error (PandocError (PandocParsecError))
51+import Text.Pandoc.Logging (LogMessage(..))
52+import Text.Pandoc.Options
53+import Text.Pandoc.Parsing
54+import Text.Pandoc.Shared (crFilter)
55+import Text.Parsec hiding (tokenPrim, space)
56+import qualified Text.Parsec as Parsec
57+import Text.Parsec.Pos (updatePosString)
58+
59+--
60+-- Data Types
61+--
62+data FontKind = Bold | Italic | Monospace | Regular deriving (Show, Eq, Ord)
63+
64+data MacroKind = KTitle
65+ | KCodeBlStart
66+ | KCodeBlEnd
67+ | KTab
68+ | KTabEnd
69+ | KSubTab
70+ deriving (Show, Eq)
71+
72+type Font = Set FontKind
73+
74+type RoffStr = (String, Font)
75+
76+-- TODO parse tables (see man tbl)
77+data ManToken = MStr RoffStr
78+ | MLine [RoffStr]
79+ | MMaybeLink String
80+ | MEmptyLine
81+ | MHeader Int [RoffStr]
82+ | MMacro MacroKind [RoffStr]
83+ | MUnknownMacro String [RoffStr]
84+ | MComment String
85+ deriving Show
86+
87+data EscapeThing = EFont Font
88+ | EChar Char
89+ | ENothing
90+ deriving Show
91+
92+data RoffState = RoffState { fontKind :: Font
93+ } deriving Show
94+
95+instance Default RoffState where
96+ def = RoffState {fontKind = singleton Regular}
97+
98+type ManLexer m = ParserT [Char] RoffState m
99+type ManParser m = ParserT [ManToken] ParserState m
100+
101+---- debug functions
102+{-
103+import Text.Pandoc.Class (runIOorExplode)
104+
105+printPandoc :: Pandoc -> [Char]
106+printPandoc (Pandoc m content) =
107+ let ttl = "Pandoc: " ++ (show $ unMeta m)
108+ cnt = intercalate "\n" $ map show content
109+ in ttl ++ "\n" ++ cnt
110+
111+testStr :: String -> IO ()
112+testStr str = do
113+ pand <- runIOorExplode $ readMan def (T.pack str)
114+ putStrLn $ printPandoc pand
115+
116+
117+testFile :: FilePath -> IO ()
118+testFile fname = do
119+ cont <- readFile fname
120+ testStr cont
121+-}
122+----
123+
124+
125+-- | Read man (troff) from an input string and return a Pandoc document.
126+readMan :: PandocMonad m => ReaderOptions -> T.Text -> m Pandoc
127+readMan opts txt = do
128+ eithertokens <- readWithM lexMan def (T.unpack $ crFilter txt)
129+ case eithertokens of
130+ Left e -> throwError e
131+ Right tokenz -> do
132+ let state = def {stateOptions = opts} :: ParserState
133+ eitherdoc <- readWithMTokens parseMan state tokenz
134+ either throwError return eitherdoc
135+
136+ where
137+
138+ readWithMTokens :: PandocMonad m
139+ => ParserT [ManToken] ParserState m a -- ^ parser
140+ -> ParserState -- ^ initial state
141+ -> [ManToken] -- ^ input
142+ -> m (Either PandocError a)
143+ readWithMTokens parser state input =
144+ let leftF = PandocParsecError . (intercalate "\n") $ show <$> input
145+ in mapLeft leftF `liftM` runParserT parser state "source" input
146+
147+ mapLeft :: (a -> c) -> Either a b -> Either c b
148+ mapLeft f (Left x) = Left $ f x
149+ mapLeft _ (Right r) = Right r
150+
151+--
152+-- String -> ManToken function
153+--
154+
155+lexMan :: PandocMonad m => ManLexer m [ManToken]
156+lexMan = many (lexComment <|> lexMacro <|> lexLine <|> lexEmptyLine)
157+
158+parseMan :: PandocMonad m => ManParser m Pandoc
159+parseMan = do
160+ let parsers = [ try parseList, parseTitle, parsePara, parseSkippedContent
161+ , try parseCodeBlock, parseHeader, parseSkipMacro]
162+ bs <- many $ choice parsers
163+ let (Pandoc _ blocks) = doc $ mconcat bs
164+ meta <- stateMeta <$> getState
165+ return $ Pandoc meta blocks
166+
167+eofline :: Stream s m Char => ParsecT s u m ()
168+eofline = void newline <|> eof
169+
170+spacetab :: Stream s m Char => ParsecT s u m Char
171+spacetab = char ' ' <|> char '\t'
172+
173+-- TODO add other sequences from man (7) groff
174+escapeLexer :: PandocMonad m => ManLexer m EscapeThing
175+escapeLexer = do
176+ char '\\'
177+ choice [escChar, escFont, escUnknown]
178+ where
179+
180+ escChar :: PandocMonad m => ManLexer m EscapeThing
181+ escChar =
182+ let skipSeqs = ["%", "{", "}", "&", "\n", ":", "\"", "0", "c"]
183+ subsSeqs = [ ("-", '-'), (" ", ' '), ("\\", '\\'), ("[lq]", '“'), ("[rq]", '”')
184+ , ("[em]", '—'), ("[en]", '–'), ("*(lq", '«'), ("*(rq", '»')
185+ , ("t", '\t'), ("e", '\\'), ("`", '`'), ("^", ' '), ("|", ' ')
186+ , ("'", '`') ]
187+ substitute :: PandocMonad m => (String,Char) -> ManLexer m EscapeThing
188+ substitute (from,to) = try $ string from >> return (EChar to)
189+ skip :: PandocMonad m => String -> ManLexer m EscapeThing
190+ skip seq' = try $ string seq' >> return ENothing
191+ in choice $ (substitute <$> subsSeqs) ++
192+ (skip <$> skipSeqs) ++
193+ [ char '(' >> anyChar >> return ENothing
194+ , char '[' >> many alphaNum >> char ']' >> return ENothing
195+ ]
196+
197+ escFont :: PandocMonad m => ManLexer m EscapeThing
198+ escFont = do
199+ char 'f'
200+ font <- choice [ singleton <$> letterFontKind
201+ , char '(' >> anyChar >> anyChar >> return (singleton Regular)
202+ , try lettersFont
203+ , digit >> return (singleton Regular)
204+ ]
205+ modifyState (\r -> r {fontKind = font})
206+ return $ EFont font
207+
208+ where
209+
210+ lettersFont :: PandocMonad m => ManLexer m Font
211+ lettersFont = do
212+ char '['
213+ fs <- many letterFontKind
214+ many letter
215+ char ']'
216+ return $ S.fromList fs
217+
218+ letterFontKind :: PandocMonad m => ManLexer m FontKind
219+ letterFontKind = choice [
220+ char 'B' >> return Bold
221+ , char 'I' >> return Italic
222+ , char 'C' >> return Monospace
223+ , (char 'P' <|> char 'R') >> return Regular
224+ ]
225+
226+ escUnknown :: PandocMonad m => ManLexer m EscapeThing
227+ escUnknown = do
228+ c <- anyChar
229+ pos <- getPosition
230+ logOutput $ SkippedContent ("Unknown escape sequence \\" ++ [c]) pos
231+ return ENothing
232+
233+currentFont :: PandocMonad m => ManLexer m Font
234+currentFont = fontKind <$> getState
235+
236+-- separate function from lexMacro since real man files sometimes do not follow the rules
237+lexComment :: PandocMonad m => ManLexer m ManToken
238+lexComment = do
239+ try $ string ".\\\""
240+ many Parsec.space
241+ body <- many $ noneOf "\n"
242+ char '\n'
243+ return $ MComment body
244+
245+lexMacro :: PandocMonad m => ManLexer m ManToken
246+lexMacro = do
247+ char '.' <|> char '\''
248+ many spacetab
249+ macroName <- many1 (letter <|> oneOf ['\\', '"', '&'])
250+ args <- lexArgs
251+ let joinedArgs = unwords $ fst <$> args
252+ knownMacro mkind = MMacro mkind args
253+
254+ tok = case macroName of
255+ x | x `elem` ["\\\"", "\\#"] -> MComment joinedArgs
256+ "TH" -> knownMacro KTitle
257+ "IP" -> knownMacro KTab
258+ "TP" -> knownMacro KTab
259+ "RE" -> knownMacro KTabEnd
260+ "RS" -> knownMacro KSubTab
261+ "nf" -> knownMacro KCodeBlStart
262+ "fi" -> knownMacro KCodeBlEnd
263+ "B" -> MStr (joinedArgs, singleton Bold)
264+ "BR" -> MMaybeLink joinedArgs
265+ x | x `elem` ["BI", "IB"] -> MStr (joinedArgs, S.fromList [Italic, Bold])
266+ x | x `elem` ["I", "IR", "RI"] -> MStr (joinedArgs, singleton Italic)
267+ "SH" -> MHeader 2 args
268+ "SS" -> MHeader 3 args
269+ x | x `elem` [ "P", "PP", "LP", "sp"] -> MEmptyLine
270+ _ -> MUnknownMacro macroName args
271+ return tok
272+
273+ where
274+
275+ -- TODO better would be [[RoffStr]], since one arg may have different fonts
276+ lexArgs :: PandocMonad m => ManLexer m [RoffStr]
277+ lexArgs = do
278+ args <- many $ try oneArg
279+ many spacetab
280+ eofline
281+ return args
282+
283+ where
284+
285+ oneArg :: PandocMonad m => ManLexer m RoffStr
286+ oneArg = do
287+ many1 spacetab
288+ many $ try $ string "\\\n"
289+ try quotedArg <|> plainArg -- try, because there are some erroneous files, e.g. linux/bpf.2
290+
291+ plainArg :: PandocMonad m => ManLexer m RoffStr
292+ plainArg = do
293+ indents <- many spacetab
294+ arg <- many1 $ escChar <|> (Just <$> noneOf " \t\n")
295+ f <- currentFont
296+ return (indents ++ catMaybes arg, f)
297+
298+ quotedArg :: PandocMonad m => ManLexer m RoffStr
299+ quotedArg = do
300+ char '"'
301+ val <- many quotedChar
302+ char '"'
303+ val2 <- many $ escChar <|> (Just <$> noneOf " \t\n")
304+ f <- currentFont
305+ return (catMaybes $ val ++ val2, f)
306+
307+ quotedChar :: PandocMonad m => ManLexer m (Maybe Char)
308+ quotedChar = escChar <|> (Just <$> noneOf "\"\n") <|> (Just <$> try (string "\"\"" >> return '"'))
309+
310+ escChar :: PandocMonad m => ManLexer m (Maybe Char)
311+ escChar = do
312+ ec <- escapeLexer
313+ case ec of
314+ (EChar c) -> return $ Just c
315+ _ -> return Nothing
316+
317+lexLine :: PandocMonad m => ManLexer m ManToken
318+lexLine = do
319+ lnparts <- many1 (esc <|> linePart)
320+ eofline
321+ return $ MLine $ catMaybes lnparts
322+ where
323+
324+ esc :: PandocMonad m => ManLexer m (Maybe (String, Font))
325+ esc = do
326+ someesc <- escapeLexer
327+ font <- currentFont
328+ let rv = case someesc of
329+ EChar c -> Just ([c], font)
330+ _ -> Nothing
331+ return rv
332+
333+ linePart :: PandocMonad m => ManLexer m (Maybe (String, Font))
334+ linePart = do
335+ lnpart <- many1 $ noneOf "\n\\"
336+ font <- currentFont
337+ return $ Just (lnpart, font)
338+
339+
340+lexEmptyLine :: PandocMonad m => ManLexer m ManToken
341+lexEmptyLine = char '\n' >> return MEmptyLine
342+
343+--
344+-- ManToken parsec functions
345+--
346+
347+msatisfy :: (Show t, Stream s m t) => (t -> Bool) -> ParserT s st m t
348+msatisfy predic = tokenPrim show nextPos testTok
349+ where
350+ testTok t = if predic t then Just t else Nothing
351+ nextPos pos _x _xs = updatePosString (setSourceColumn (setSourceLine pos $ sourceLine pos + 1) 1) ("")
352+
353+mstr :: PandocMonad m => ManParser m ManToken
354+mstr = msatisfy isMStr where
355+ isMStr (MStr _) = True
356+ isMStr _ = False
357+
358+mline :: PandocMonad m => ManParser m ManToken
359+mline = msatisfy isMLine where
360+ isMLine (MLine _) = True
361+ isMLine _ = False
362+
363+mmaybeLink :: PandocMonad m => ManParser m ManToken
364+mmaybeLink = msatisfy isMMaybeLink where
365+ isMMaybeLink (MMaybeLink _) = True
366+ isMMaybeLink _ = False
367+
368+memplyLine :: PandocMonad m => ManParser m ManToken
369+memplyLine = msatisfy isMEmptyLine where
370+ isMEmptyLine MEmptyLine = True
371+ isMEmptyLine _ = False
372+
373+mheader :: PandocMonad m => ManParser m ManToken
374+mheader = msatisfy isMHeader where
375+ isMHeader (MHeader _ _) = True
376+ isMHeader _ = False
377+
378+mmacro :: PandocMonad m => MacroKind -> ManParser m ManToken
379+mmacro mk = msatisfy isMMacro where
380+ isMMacro (MMacro mk' _) | mk == mk' = True
381+ | otherwise = False
382+ isMMacro _ = False
383+
384+mmacroAny :: PandocMonad m => ManParser m ManToken
385+mmacroAny = msatisfy isMMacro where
386+ isMMacro (MMacro _ _) = True
387+ isMMacro _ = False
388+
389+munknownMacro :: PandocMonad m => ManParser m ManToken
390+munknownMacro = msatisfy isMUnknownMacro where
391+ isMUnknownMacro (MUnknownMacro _ _) = True
392+ isMUnknownMacro _ = False
393+
394+mcomment :: PandocMonad m => ManParser m ManToken
395+mcomment = msatisfy isMComment where
396+ isMComment (MComment _) = True
397+ isMComment _ = False
398+
399+--
400+-- ManToken -> Block functions
401+--
402+
403+parseTitle :: PandocMonad m => ManParser m Blocks
404+parseTitle = do
405+ (MMacro _ args) <- mmacro KTitle
406+ if null args
407+ then return mempty
408+ else do
409+ let mantitle = fst $ head args
410+ modifyState (changeTitle mantitle)
411+ return $ header 1 $ str mantitle
412+ where
413+ changeTitle title pst =
414+ let meta = stateMeta pst
415+ metaUp = Meta $ insert "title" (MetaString title) (unMeta meta)
416+ in
417+ pst {stateMeta = metaUp}
418+
419+parseSkippedContent :: PandocMonad m => ManParser m Blocks
420+parseSkippedContent = do
421+ tok <- munknownMacro <|> mcomment <|> memplyLine
422+ onToken tok
423+ return mempty
424+
425+ where
426+
427+ onToken :: PandocMonad m => ManToken -> ManParser m ()
428+ onToken (MUnknownMacro mname _) = do
429+ pos <- getPosition
430+ logMessage $ SkippedContent ("Unknown macro: " ++ mname) pos
431+ onToken _ = return ()
432+
433+strToInlines :: RoffStr -> Inlines
434+strToInlines (s, fonts) = inner $ S.toList fonts where
435+ inner :: [FontKind] -> Inlines
436+ inner [] = str s
437+ inner (Bold:fs) = strong $ inner fs
438+ inner (Italic:fs) = emph $ inner fs
439+
440+ -- Monospace goes after Bold and Italic in ordered set
441+ inner (Monospace:_) = code s
442+ inner (Regular:fs) = inner fs
443+
444+parsePara :: PandocMonad m => ManParser m Blocks
445+parsePara = para <$> parseInlines
446+
447+parseInlines :: PandocMonad m => ManParser m Inlines
448+parseInlines = do
449+ inls <- many1 (strInl <|> lineInl <|> linkInl <|> comment)
450+ let withspaces = intersperse B.space inls
451+ return $ mconcat withspaces
452+
453+ where
454+
455+ strInl :: PandocMonad m => ManParser m Inlines
456+ strInl = do
457+ (MStr rstr) <- mstr
458+ return $ strToInlines rstr
459+
460+ lineInl :: PandocMonad m => ManParser m Inlines
461+ lineInl = do
462+ (MLine fragments) <- mline
463+ return $ mconcat $ strToInlines <$> fragments
464+
465+ linkInl :: PandocMonad m => ManParser m Inlines
466+ linkInl = do
467+ (MMaybeLink txt) <- mmaybeLink
468+ let inls = case runParser linkParser () "" txt of
469+ Right lnk -> lnk
470+ Left _ -> strong $ str txt
471+ return inls
472+
473+ where
474+
475+ -- assuming man pages are generated from Linux-like repository
476+ linkParser :: Parsec String () Inlines
477+ linkParser = do
478+ mpage <- many1 (alphaNum <|> char '_')
479+ spacetab
480+ char '('
481+ mansect <- digit
482+ char ')'
483+ other <- many anyChar
484+ let manurl pagename section = "../"++section++"/"++pagename++"."++section
485+ lnkInls = link (manurl mpage [mansect]) mpage (strong $ str mpage)
486+ return $ lnkInls <> strong (str (" ("++[mansect] ++ ")") <> str other)
487+
488+ comment :: PandocMonad m => ManParser m Inlines
489+ comment = mcomment >> return mempty
490+
491+
492+parseCodeBlock :: PandocMonad m => ManParser m Blocks
493+parseCodeBlock = do
494+ mmacro KCodeBlStart
495+ toks <- many (mstr <|> mline <|> mmaybeLink <|> memplyLine <|> munknownMacro <|> mcomment)
496+ mmacro KCodeBlEnd
497+ return $ codeBlock (intercalate "\n" . catMaybes $ extractText <$> toks)
498+
499+ where
500+
501+ extractText :: ManToken -> Maybe String
502+ extractText (MStr (s, _)) = Just s
503+ extractText (MLine ss) = Just . concat $ map fst ss -- TODO maybe unwords?
504+ extractText (MMaybeLink s) = Just s
505+ extractText MEmptyLine = Just "" -- string are intercalated with '\n', this prevents double '\n'
506+ extractText _ = Nothing
507+
508+parseHeader :: PandocMonad m => ManParser m Blocks
509+parseHeader = do
510+ (MHeader lvl ss) <- mheader
511+ return $ header lvl (mconcat $ intersperse B.space $ strToInlines <$> ss)
512+
513+type ListBuilder = [Blocks] -> Blocks
514+
515+parseList :: PandocMonad m => ManParser m Blocks
516+parseList = do
517+ xx <- many1 paras
518+ let bls = map snd xx
519+ let bldr = fst $ head xx
520+ return $ bldr bls
521+
522+ where
523+
524+ macroIPInl :: [RoffStr] -> Inlines
525+ macroIPInl (x:_:[]) = strToInlines x <> B.space
526+ macroIPInl _ = mempty
527+
528+ listKind :: [RoffStr] -> Maybe ListBuilder
529+ listKind (((c:_), _):_:[]) =
530+ let params style = orderedListWith (1, style, DefaultDelim)
531+ in case c of
532+ _ | isDigit c -> Just $ params Decimal
533+ _ | isUpper c -> Just $ params UpperAlpha
534+ _ | isLower c -> Just $ params LowerAlpha
535+ _ -> Nothing
536+
537+ listKind _ = Nothing
538+
539+ paras :: PandocMonad m => ManParser m (ListBuilder, Blocks)
540+ paras = do
541+ (MMacro _ args) <- mmacro KTab
542+ let lbuilderOpt = listKind args
543+ lbuilder = fromMaybe bulletList lbuilderOpt
544+ macroinl = macroIPInl args
545+ inls <- parseInlines
546+ let parainls = if isNothing lbuilderOpt then macroinl <> inls else inls
547+ subls <- mconcat <$> many sublist
548+ return $ (lbuilder, plain parainls <> subls)
549+
550+ sublist :: PandocMonad m => ManParser m Blocks
551+ sublist = do
552+ mmacro KSubTab
553+ bl <- parseList
554+ mmacro KTabEnd
555+ return bl
556+
557+-- In case of weird man file it will be parsed succesfully
558+parseSkipMacro :: PandocMonad m => ManParser m Blocks
559+parseSkipMacro = mmacroAny >> mempty
test/Tests/Old.hs+4
@@ -171,6 +171,10 @@ tests = [ testGroup "markdown"
171 , test "tables" ["-f", "native", "-t", "../data/sample.lua"]
171 , test "tables" ["-f", "native", "-t", "../data/sample.lua"]
172 "tables.native" "tables.custom"
172 "tables.native" "tables.custom"
173 ]
173 ]
174+ , testGroup "man"
175+ [ test "reader" ["-r", "man", "-w", "native", "-s"]
176+ "man-reader.man" "man-reader.native"
177+ ]
174 ]
178 ]
175
179
176-- makes sure file is fully closed after reading
180-- makes sure file is fully closed after reading
test/Tests/Readers/Man.hsnew file+83
@@ -0,0 +1,83 @@
1+{-# LANGUAGE OverloadedStrings #-}
2+module Tests.Readers.Man (tests) where
3+
4+import Prelude
5+import Data.Text (Text)
6+import Test.Tasty
7+import Tests.Helpers
8+import Text.Pandoc
9+import Text.Pandoc.Arbitrary ()
10+import Text.Pandoc.Builder
11+import Text.Pandoc.Readers.Man
12+
13+man :: Text -> Pandoc
14+man = purely $ readMan def
15+
16+infix 4 =:
17+(=:) :: ToString c
18+ => String -> (Text, c) -> TestTree
19+(=:) = test man
20+
21+tests :: [TestTree]
22+tests = [
23+ -- .SH "HEllo bbb" "aaa"" as"
24+ testGroup "Macros" [
25+ "Bold" =:
26+ ".B foo"
27+ =?> (para $ strong "foo")
28+ , "Italic" =:
29+ ".I bar\n"
30+ =?> (para $ emph "bar")
31+ , "BoldItalic" =:
32+ ".BI foo bar"
33+ =?> (para $ strong $ emph $ str "foo bar")
34+ , "H1" =:
35+ ".SH The header\n"
36+ =?> header 2 (str "The" <> space <> str "header")
37+ , "H2" =:
38+ ".SS \"The header 2\""
39+ =?> header 3 (str "The header 2")
40+ , "Macro args" =:
41+ ".B \"single arg with \"\"Q\"\"\""
42+ =?> (para $ strong $ str "single arg with \"Q\"")
43+ , "comment" =:
44+ ".\\\"bla\naaa"
45+ =?> (para $ space <> str "aaa")
46+ , "link" =:
47+ ".BR aa (1)"
48+ =?> (para $ link "../1/aa.1" "aa" (strong $ str "aa") <> (strong $ str " (1)"))
49+ ],
50+ testGroup "Escapes" [
51+ "fonts" =:
52+ "aa\\fIbb\\fRcc"
53+ =?> (para $ str "aa" <> (emph $ str "bb") <> str "cc")
54+ , "skip" =:
55+ "a\\%\\{\\}\\\n\\:b\\0"
56+ =?> (para $ str "ab")
57+ , "replace" =:
58+ "\\-\\ \\\\\\[lq]\\[rq]\\[em]\\[en]\\*(lq\\*(rq"
59+ =?> (para $ str "- \\“”—–«»")
60+ , "replace2" =:
61+ "\\t\\e\\`\\^\\|\\'"
62+ =?> (para $ str "\t\\` `")
63+ ],
64+ testGroup "Lists" [
65+ "bullet" =:
66+ ".IP\nfirst\n.IP\nsecond"
67+ =?> bulletList [plain $ str "first", plain $ str "second"]
68+ , "odrered" =:
69+ ".IP 1 a\nfirst\n.IP 2 a\nsecond"
70+ =?> orderedListWith (1,Decimal,DefaultDelim) [plain $ str "first", plain $ str "second"]
71+ , "upper" =:
72+ ".IP A a\nfirst\n.IP B a\nsecond"
73+ =?> orderedListWith (1,UpperAlpha,DefaultDelim) [plain $ str "first", plain $ str "second"]
74+ , "nested" =:
75+ ".IP\nfirst\n.RS\n.IP\n1a\n.IP\n1b\n.RE"
76+ =?> bulletList [(plain $ str "first") <> (bulletList [plain $ str "1a", plain $ str "1b"])]
77+ ],
78+ testGroup "CodeBlocks" [
79+ "cb1"=:
80+ ".nf\naa\n\tbb\n.fi"
81+ =?> codeBlock "aa\n\tbb"
82+ ]
83+ ]
test/grofftest.shnew file+22
@@ -0,0 +1,22 @@
1+#!/bin/bash
2+
3+# iterates over specified directory, containing "\w+\.\d"-like files,
4+# executes pandoc voer them and prints stderr on nonzero return code
5+
6+if [ $# -ne 2 ]; then
7+ echo "Not enough arguments"
8+ exit 1
9+fi
10+
11+PANDOC=$1
12+DIR=$2
13+
14+$PANDOC --version > /dev/null || { echo "pandoc executable error" >&2 ; exit 1 ; }
15+
16+ls $2 | egrep "^.+\.[0-9].?$" | while read f ; do
17+ FILE="$DIR/$f"
18+ $PANDOC -f man -t native < $FILE 2>&1 > /dev/null
19+ if [ $? -ne 0 ]; then
20+ echo "Failed to convert $FILE"
21+ fi
22+done
test/man-reader.mannew file+189
@@ -0,0 +1,189 @@
1+.TH "Pandoc Man tests" "" "Oct 17, 2018" "" ""
2+.PP
3+This is a set of tests for pandoc.
4+.PP
5+ * * * * *
6+.SH Headers
7+.SH Level 1
8+.SS Level 2
9+
10+ * * * * *
11+.SH Paragraphs
12+.PP
13+Here's a regular paragraph.
14+.PP
15+Another paragraph
16+In Markdown 1.0.0 and earlier.
17+Version 8.
18+This line turns into a list item.
19+Because a hard\-wrapped line in the middle of a paragraph looked like a list
20+item.
21+.PP
22+There should be a hard line break
23+.PD 0
24+.P
25+.PD
26+here.
27+.PP
28+ * * * * *
29+.SH Block Quotes
30+Code in a block quote:
31+.IP
32+.nf
33+\f[C]
34+sub\ status\ {
35+\ \ \ \ print\ "working";
36+}
37+\f[]
38+.fi
39+.PP
40+A list:
41+.IP "1." 3
42+item one
43+.IP "2." 3
44+item two
45+.PP
46+.SH Code Blocks
47+.PP
48+Code:
49+.IP
50+.nf
51+\f[C]
52+\-\-\-\-\ (should\ be\ four\ hyphens)
53+
54+sub\ status\ {
55+\ \ \ \ print\ "working";
56+}
57+
58+\f[]
59+.fi
60+.PP
61+And:
62+.IP
63+.nf
64+\f[C]
65+\tthis\ code\ line is\ indented\ by\ one\ tab
66+
67+These\ should\ not\ be\ escaped:\ \ \\$\ \\\\\ \\>\ \\[\ \\{
68+\f[]
69+.fi
70+.PP
71+ * * * * *
72+.SH Lists
73+.SS Unordered
74+.PP
75+Asterisks:
76+.IP \[bu] 2
77+asterisk 1
78+.IP \[bu] 2
79+asterisk 2
80+.IP \[bu] 2
81+asterisk 3
82+.PP
83+.SS Ordered
84+.IP "1." 3
85+First
86+.IP "2." 3
87+Second
88+.IP "3." 3
89+Third
90+.PP
91+.SS Nested
92+.IP \[bu] 2
93+Tab
94+.RS 2
95+.IP \[bu] 2
96+Tab
97+.RS 2
98+.IP \[bu] 2
99+Tab
100+.RE
101+.RE
102+.PP
103+Here's another:
104+.IP "1." 3
105+First
106+.IP "2." 3
107+Second:
108+.RS 4
109+.IP \[bu] 2
110+Fee
111+.IP \[bu] 2
112+Fie
113+.IP \[bu] 2
114+Foe
115+.RE
116+.IP "3." 3
117+Third
118+.PP
119+Same thing:
120+.IP "1." 3
121+First
122+.IP "2." 3
123+Second:
124+.RS 4
125+.IP \[bu] 2
126+Fee
127+.IP \[bu] 2
128+Fie
129+.IP \[bu] 2
130+Foe
131+.RE
132+.IP "3." 3
133+Third
134+.SS different styles:
135+.IP "A." 3
136+Upper Alpha
137+.RS 4
138+.IP "I." 3
139+Upper Roman.
140+.RS 4
141+.IP "(6)" 4
142+Decimal start with 6
143+.RS 4
144+.IP "c)" 3
145+Lower alpha with paren
146+.RE
147+.RE
148+.RE
149+.PP
150+ * * * * *
151+.SH Special Characters
152+AT&T has an ampersand in their name.
153+.PP
154+4 < 5.
155+.PP
156+6 > 5.
157+.PP
158+Backslash: \\
159+.PP
160+Backtick: `
161+.PP
162+Asterisk: *
163+.PP
164+Underscore: _
165+.PP
166+Left brace: {
167+.PP
168+Right brace: }
169+.PP
170+Left bracket: [
171+.PP
172+Right bracket: ]
173+.PP
174+Left paren: (
175+.PP
176+Right paren: )
177+.PP
178+Greater\-than: >
179+.PP
180+Hash: #
181+.PP
182+Period: .
183+.PP
184+Bang: !
185+.PP
186+Plus: +
187+.PP
188+Minus: \-
189+.PP
test/man-reader.nativenew file+94
@@ -0,0 +1,94 @@
1+Pandoc (Meta {unMeta = fromList [("title",MetaString "Pandoc Man tests")]})
2+[Header 1 ("",[],[]) [Str "Pandoc Man tests"]
3+,Para [Str "This is a set of tests for pandoc."]
4+,Para [Str " * * * * *"]
5+,Header 2 ("",[],[]) [Str "Headers"]
6+,Header 2 ("",[],[]) [Str "Level",Space,Str "1"]
7+,Header 3 ("",[],[]) [Str "Level",Space,Str "2"]
8+,Para [Str " * * * * *"]
9+,Header 2 ("",[],[]) [Str "Paragraphs"]
10+,Para [Str "Here's a regular paragraph."]
11+,Para [Str "Another paragraph",Space,Str "In Markdown 1.0.0 and earlier.",Space,Str "Version 8.",Space,Str "This line turns into a list item.",Space,Str "Because a hard-wrapped line in the middle of a paragraph looked like a list",Space,Str "item."]
12+,Para [Str "There should be a hard line break"]
13+,Para [Str "here."]
14+,Para [Str " * * * * *"]
15+,Header 2 ("",[],[]) [Str "Block",Space,Str "Quotes"]
16+,Para [Str "Code in a block quote:"]
17+,CodeBlock ("",[],[]) "\nsub status {\n print \"working\";\n}\n"
18+,Para [Str "A list:"]
19+,OrderedList (1,Decimal,DefaultDelim)
20+ [[Plain [Str "item one"]]
21+ ,[Plain [Str "item two"]]]
22+,Header 2 ("",[],[]) [Str "Code",Space,Str "Blocks"]
23+,Para [Str "Code:"]
24+,CodeBlock ("",[],[]) "\n---- (should be four hyphens)\n\nsub status {\n print \"working\";\n}\n\n"
25+,Para [Str "And:"]
26+,CodeBlock ("",[],[]) "\n\tthis code line is indented by one tab\n\nThese should not be escaped: \\$ \\\\ \\> \\[ \\{\n"
27+,Para [Str " * * * * *"]
28+,Header 2 ("",[],[]) [Str "Lists"]
29+,Header 3 ("",[],[]) [Str "Unordered"]
30+,Para [Str "Asterisks:"]
31+,BulletList
32+ [[Plain [Str "",Space,Str "asterisk 1"]]
33+ ,[Plain [Str "",Space,Str "asterisk 2"]]
34+ ,[Plain [Str "",Space,Str "asterisk 3"]]]
35+,Header 3 ("",[],[]) [Str "Ordered"]
36+,OrderedList (1,Decimal,DefaultDelim)
37+ [[Plain [Str "First"]]
38+ ,[Plain [Str "Second"]]
39+ ,[Plain [Str "Third"]]]
40+,Header 3 ("",[],[]) [Str "Nested"]
41+,BulletList
42+ [[Plain [Str "",Space,Str "Tab"]
43+ ,BulletList
44+ [[Plain [Str "",Space,Str "Tab"]
45+ ,BulletList
46+ [[Plain [Str "",Space,Str "Tab"]]]]]]]
47+,Para [Str "Here's another:"]
48+,OrderedList (1,Decimal,DefaultDelim)
49+ [[Plain [Str "First"]]
50+ ,[Plain [Str "Second:"]
51+ ,BulletList
52+ [[Plain [Str "",Space,Str "Fee"]]
53+ ,[Plain [Str "",Space,Str "Fie"]]
54+ ,[Plain [Str "",Space,Str "Foe"]]]]
55+ ,[Plain [Str "Third"]]]
56+,Para [Str "Same thing:"]
57+,OrderedList (1,Decimal,DefaultDelim)
58+ [[Plain [Str "First"]]
59+ ,[Plain [Str "Second:"]
60+ ,BulletList
61+ [[Plain [Str "",Space,Str "Fee"]]
62+ ,[Plain [Str "",Space,Str "Fie"]]
63+ ,[Plain [Str "",Space,Str "Foe"]]]]
64+ ,[Plain [Str "Third"]]]
65+,Header 3 ("",[],[]) [Str "different",Space,Str "styles:"]
66+,OrderedList (1,UpperAlpha,DefaultDelim)
67+ [[Plain [Str "Upper Alpha"]
68+ ,OrderedList (1,UpperAlpha,DefaultDelim)
69+ [[Plain [Str "Upper Roman."]
70+ ,BulletList
71+ [[Plain [Str "(6)",Space,Str "Decimal start with 6"]
72+ ,OrderedList (1,LowerAlpha,DefaultDelim)
73+ [[Plain [Str "Lower alpha with paren"]]]]]]]]]
74+,Para [Str " * * * * *"]
75+,Header 2 ("",[],[]) [Str "Special",Space,Str "Characters"]
76+,Para [Str "AT&T has an ampersand in their name."]
77+,Para [Str "4 < 5."]
78+,Para [Str "6 > 5."]
79+,Para [Str "Backslash: \\"]
80+,Para [Str "Backtick: `"]
81+,Para [Str "Asterisk: *"]
82+,Para [Str "Underscore: _"]
83+,Para [Str "Left brace: {"]
84+,Para [Str "Right brace: }"]
85+,Para [Str "Left bracket: ["]
86+,Para [Str "Right bracket: ]"]
87+,Para [Str "Left paren: ("]
88+,Para [Str "Right paren: )"]
89+,Para [Str "Greater-than: >"]
90+,Para [Str "Hash: #"]
91+,Para [Str "Period: ."]
92+,Para [Str "Bang: !"]
93+,Para [Str "Plus: +"]
94+,Para [Str "Minus: -"]]
test/test-pandoc.hs+2
@@ -22,6 +22,7 @@ import qualified Tests.Readers.Odt
22import qualified Tests.Readers.Org
22import qualified Tests.Readers.Org
23import qualified Tests.Readers.RST
23import qualified Tests.Readers.RST
24import qualified Tests.Readers.Txt2Tags
24import qualified Tests.Readers.Txt2Tags
25+import qualified Tests.Readers.Man
25import qualified Tests.Shared
26import qualified Tests.Shared
26import qualified Tests.Writers.AsciiDoc
27import qualified Tests.Writers.AsciiDoc
27import qualified Tests.Writers.ConTeXt
28import qualified Tests.Writers.ConTeXt
@@ -76,6 +77,7 @@ tests = testGroup "pandoc tests" [ Tests.Command.tests
76 , testGroup "EPUB" Tests.Readers.EPUB.tests
77 , testGroup "EPUB" Tests.Readers.EPUB.tests
77 , testGroup "Muse" Tests.Readers.Muse.tests
78 , testGroup "Muse" Tests.Readers.Muse.tests
78 , testGroup "Creole" Tests.Readers.Creole.tests
79 , testGroup "Creole" Tests.Readers.Creole.tests
80+ , testGroup "Man" Tests.Readers.Man.tests
79 , testGroup "FB2" Tests.Readers.FB2.tests
81 , testGroup "FB2" Tests.Readers.FB2.tests
80 ]
82 ]
81 , testGroup "Lua filters" Tests.Lua.tests
83 , testGroup "Lua filters" Tests.Lua.tests