1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
|
{-# LANGUAGE FlexibleContexts #-}
{-
Copyright (C) 2018 Yan Pashkovsky <yanp.bugz@gmail.com>
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
-}
{- |
Module : Text.Pandoc.Readers.Man
Copyright : Copyright (C) 2018 Yan Pashkovsky
License : GNU GPL, version 2 or above
Maintainer : Yan Pashkovsky <yanp.bugz@gmail.com>
Stability : WIP
Portability : portable
Conversion of man to 'Pandoc' document.
-}
module Text.Pandoc.Readers.Man (readMan) where
import Prelude
import Control.Monad (liftM, void, mzero)
import Control.Monad.Except (throwError)
import Data.Char (isHexDigit, chr)
import Data.Default (Default)
import Data.Maybe (catMaybes)
import qualified Data.Map as M
import Data.Set (Set, singleton)
import qualified Data.Set as S (fromList, toList)
import Data.List (intersperse, intercalate)
import qualified Data.Text as T
import Text.Pandoc.Class (PandocMonad(..), report)
import Text.Pandoc.Builder as B hiding (singleton)
import Text.Pandoc.Error (PandocError (PandocParsecError))
import Text.Pandoc.Logging (LogMessage(..))
import Text.Pandoc.Options
import Text.Pandoc.Parsing
import Text.Pandoc.Shared (crFilter, safeRead)
import Text.Parsec hiding (tokenPrim, space)
import qualified Text.Parsec as Parsec
import Text.Parsec.Pos (updatePosString)
import Text.Pandoc.GroffChar (characterCodes, combiningAccents)
--
-- Data Types
--
data FontKind = Bold | Italic | Monospace | Regular deriving (Show, Eq, Ord)
type MacroKind = String
type Font = Set FontKind
type RoffStr = (String, Font)
-- TODO parse tables (see man tbl)
data ManToken = MStr RoffStr
| MLine [RoffStr]
| MMaybeLink String
| MEmptyLine
| MMacro MacroKind [RoffStr]
| MComment String
deriving Show
data RoffState = RoffState { fontKind :: Font
} deriving Show
instance Default RoffState where
def = RoffState {fontKind = singleton Regular}
type ManLexer m = ParserT [Char] RoffState m
type ManParser m = ParserT [ManToken] ParserState m
-- | Read man (troff) from an input string and return a Pandoc document.
readMan :: PandocMonad m => ReaderOptions -> T.Text -> m Pandoc
readMan opts txt = do
eithertokens <- readWithM lexMan def (T.unpack $ crFilter txt)
case eithertokens of
Left e -> throwError e
Right tokenz -> do
let state = def {stateOptions = opts} :: ParserState
eitherdoc <- readWithMTokens parseMan state tokenz
either throwError return eitherdoc
where
readWithMTokens :: PandocMonad m
=> ParserT [ManToken] ParserState m a -- ^ parser
-> ParserState -- ^ initial state
-> [ManToken] -- ^ input
-> m (Either PandocError a)
readWithMTokens parser state input =
let leftF = PandocParsecError . (intercalate "\n") $ show <$> input
in mapLeft leftF `liftM` runParserT parser state "source" input
mapLeft :: (a -> c) -> Either a b -> Either c b
mapLeft f (Left x) = Left $ f x
mapLeft _ (Right r) = Right r
--
-- String -> ManToken function
--
lexMan :: PandocMonad m => ManLexer m [ManToken]
lexMan = many (lexComment <|> lexMacro <|> lexLine <|> lexEmptyLine)
parseMan :: PandocMonad m => ManParser m Pandoc
parseMan = do
bs <- many parseBlock
meta <- stateMeta <$> getState
let (Pandoc _ blocks) = doc $ mconcat bs
return $ Pandoc meta blocks
parseBlock :: PandocMonad m => ManParser m Blocks
parseBlock = choice [ parseList
, parseTitle
, parsePara
, parseSkippedContent
, parseCodeBlock
, parseHeader
, parseSkipMacro
]
eofline :: Stream s m Char => ParsecT s u m ()
eofline = void newline <|> eof
spacetab :: Stream s m Char => ParsecT s u m Char
spacetab = char ' ' <|> char '\t'
characterCodeMap :: M.Map String Char
characterCodeMap =
M.fromList $ map (\(x,y) -> (y,x)) $ characterCodes ++ combiningAccents
escapeLexer :: PandocMonad m => ManLexer m String
escapeLexer = do
char '\\'
twoCharGlyph <|> bracketedGlyph <|> escFont <|> escStar <|> escSingle
where
twoCharGlyph = do
char '('
cs <- count 2 anyChar
case M.lookup cs characterCodeMap of
Just c -> return [c]
Nothing -> escUnknown ('(':cs)
bracketedGlyph =
char '[' *>
( ucharCode `sepBy1` (char '_')
<|> charCode `sepBy1` (many1 Parsec.space)
) <* char ']'
ucharCode = do
char 'u'
cs <- many1 (satisfy isHexDigit)
case chr <$> safeRead ('0':'x':cs) of
Nothing -> mzero
Just c -> return c
charCode = do
cs <- many1 (noneOf ['[',']',' ','\t','\n'])
case M.lookup cs characterCodeMap of
Nothing -> mzero
Just c -> return c
escStar = do
char '*'
choice
[ ("\xae" <$ char 'R')
, ("" <$ char 'S') -- switch back to default font size
, ("\x201c" <$ try (string "(lq"))
, ("\x201d" <$ try (string "(rq"))
, ("" <$ try (string "(HF" >>
modifyState (\r -> r {fontKind = singleton Bold})))
, ("\x2122" <$ try (string "(Tm"))
]
escSingle = do
c <- anyChar
case c of
'"' -> mempty <$ skipMany (satisfy (/='\n')) -- line comment
'#' -> mempty <$ manyTill anyChar newline
'%' -> return mempty
'{' -> return mempty
'}' -> return mempty
'&' -> return mempty
'\n' -> return mempty
':' -> return mempty
'0' -> return mempty
'c' -> return mempty
'-' -> return "-"
'_' -> return "_"
' ' -> return " "
'\\' -> return "\\"
't' -> return "\t"
'e' -> return "\\"
'`' -> return "`"
'^' -> return " "
'|' -> return " "
'\'' -> return "`"
'.' -> return "`"
_ -> escUnknown [c]
escFont :: PandocMonad m => ManLexer m String
escFont = do
char 'f'
font <- choice [ singleton <$> letterFontKind
, char '(' >> anyChar >> anyChar >> return (singleton Regular)
, try lettersFont
, digit >> return (singleton Regular)
]
modifyState (\r -> r {fontKind = font})
return mempty
lettersFont :: PandocMonad m => ManLexer m Font
lettersFont = do
char '['
fs <- many letterFontKind
many letter
char ']'
return $ S.fromList fs
letterFontKind :: PandocMonad m => ManLexer m FontKind
letterFontKind = choice [
char 'B' >> return Bold
, char 'I' >> return Italic
, char 'C' >> return Monospace
, (char 'P' <|> char 'R') >> return Regular
]
escUnknown :: PandocMonad m => String -> ManLexer m String
escUnknown s = do
pos <- getPosition
report $ SkippedContent ("Unknown escape sequence " ++ s) pos
return mempty
currentFont :: PandocMonad m => ManLexer m Font
currentFont = fontKind <$> getState
-- separate function from lexMacro since real man files sometimes do not follow the rules
lexComment :: PandocMonad m => ManLexer m ManToken
lexComment = do
try $ string ".\\\""
many Parsec.space
body <- many $ noneOf "\n"
char '\n'
return $ MComment body
lexMacro :: PandocMonad m => ManLexer m ManToken
lexMacro = do
char '.' <|> char '\''
many spacetab
macroName <- many1 (letter <|> oneOf ['\\', '"', '&'])
args <- lexArgs
let joinedArgs = unwords $ fst <$> args
tok = case macroName of
x | x `elem` ["\\\"", "\\#"] -> MComment joinedArgs
"B" -> MStr (joinedArgs, singleton Bold)
"BR" -> MMaybeLink joinedArgs
x | x `elem` ["BI", "IB"] -> MStr (joinedArgs, S.fromList [Italic, Bold])
x | x `elem` ["I", "IR", "RI"] -> MStr (joinedArgs, singleton Italic)
x | x `elem` [ "P", "PP", "LP", "sp"] -> MEmptyLine
_ -> MMacro macroName args
return tok
where
-- TODO better would be [[RoffStr]], since one arg may have different fonts
lexArgs :: PandocMonad m => ManLexer m [RoffStr]
lexArgs = do
args <- many $ try oneArg
many spacetab
eofline
return args
where
oneArg :: PandocMonad m => ManLexer m RoffStr
oneArg = do
many1 spacetab
many $ try $ string "\\\n"
try quotedArg <|> plainArg -- try, because there are some erroneous files, e.g. linux/bpf.2
plainArg :: PandocMonad m => ManLexer m RoffStr
plainArg = do
indents <- many spacetab
arg <- many1 $ escapeLexer <|> many1 (noneOf " \t\n\\")
f <- currentFont
return (indents ++ mconcat arg, f)
quotedArg :: PandocMonad m => ManLexer m RoffStr
quotedArg = do
char '"'
val <- mconcat <$> many quotedChar
char '"'
val2 <- mconcat <$> many (escapeLexer <|> many1 (noneOf " \t\n"))
f <- currentFont
return (val ++ val2, f)
quotedChar :: PandocMonad m => ManLexer m String
quotedChar = escapeLexer
<|> many1 (noneOf "\"\n\\")
<|> try (string "\"\"" >> return "\"")
lexLine :: PandocMonad m => ManLexer m ManToken
lexLine = do
lnparts <- many1 (esc <|> linePart)
eofline
return $ MLine $ catMaybes lnparts
where
esc :: PandocMonad m => ManLexer m (Maybe (String, Font))
esc = do
someesc <- escapeLexer
font <- currentFont
return $ if null someesc
then Nothing
else Just (someesc, font)
linePart :: PandocMonad m => ManLexer m (Maybe (String, Font))
linePart = do
lnpart <- many1 $ noneOf "\n\\"
font <- currentFont
return $ Just (lnpart, font)
lexEmptyLine :: PandocMonad m => ManLexer m ManToken
lexEmptyLine = char '\n' >> return MEmptyLine
--
-- ManToken parsec functions
--
msatisfy :: (Show t, Stream s m t) => (t -> Bool) -> ParserT s st m t
msatisfy predic = tokenPrim show nextPos testTok
where
testTok t = if predic t then Just t else Nothing
nextPos pos _x _xs = updatePosString (setSourceColumn (setSourceLine pos $ sourceLine pos + 1) 1) ("")
mstr :: PandocMonad m => ManParser m ManToken
mstr = msatisfy isMStr where
isMStr (MStr _) = True
isMStr _ = False
mline :: PandocMonad m => ManParser m ManToken
mline = msatisfy isMLine where
isMLine (MLine _) = True
isMLine _ = False
mmaybeLink :: PandocMonad m => ManParser m ManToken
mmaybeLink = msatisfy isMMaybeLink where
isMMaybeLink (MMaybeLink _) = True
isMMaybeLink _ = False
memptyLine :: PandocMonad m => ManParser m ManToken
memptyLine = msatisfy isMEmptyLine where
isMEmptyLine MEmptyLine = True
isMEmptyLine _ = False
mmacro :: PandocMonad m => MacroKind -> ManParser m ManToken
mmacro mk = msatisfy isMMacro where
isMMacro (MMacro mk' _) | mk == mk' = True
| otherwise = False
isMMacro _ = False
mmacroAny :: PandocMonad m => ManParser m ManToken
mmacroAny = msatisfy isMMacro where
isMMacro (MMacro _ _) = True
isMMacro _ = False
mcomment :: PandocMonad m => ManParser m ManToken
mcomment = msatisfy isMComment where
isMComment (MComment _) = True
isMComment _ = False
--
-- ManToken -> Block functions
--
parseTitle :: PandocMonad m => ManParser m Blocks
parseTitle = do
(MMacro _ args) <- mmacro "TH"
let adjustMeta =
case map fst args of
(x:y:z:_) -> setMeta "title" x . setMeta "section" y . setMeta "date" z
[x,y] -> setMeta "title" x . setMeta "section" y
[x] -> setMeta "title" x
[] -> id
modifyState $ \st -> st{ stateMeta = adjustMeta $ stateMeta st }
return mempty
parseSkippedContent :: PandocMonad m => ManParser m Blocks
parseSkippedContent = mempty <$ (mcomment <|> memptyLine)
strToInlines :: RoffStr -> Inlines
strToInlines (s, fonts) = inner $ S.toList fonts where
inner :: [FontKind] -> Inlines
inner [] = text s
inner (Bold:fs) = strong $ inner fs
inner (Italic:fs) = emph $ inner fs
-- Monospace goes after Bold and Italic in ordered set
inner (Monospace:_) = code s
inner (Regular:fs) = inner fs
parsePara :: PandocMonad m => ManParser m Blocks
parsePara = para . trimInlines <$> parseInlines
parseInlines :: PandocMonad m => ManParser m Inlines
parseInlines = do
inls <- many1 (strInl <|> lineInl <|> linkInl <|> comment)
let withspaces = intersperse B.space inls
return $ mconcat withspaces
where
strInl :: PandocMonad m => ManParser m Inlines
strInl = do
(MStr rstr) <- mstr
return $ strToInlines rstr
lineInl :: PandocMonad m => ManParser m Inlines
lineInl = do
(MLine fragments) <- mline
return $ mconcat $ strToInlines <$> fragments
linkInl :: PandocMonad m => ManParser m Inlines
linkInl = do
(MMaybeLink txt) <- mmaybeLink
let inls = case runParser linkParser () "" txt of
Right lnk -> lnk
Left _ -> strong $ text txt
return inls
where
-- assuming man pages are generated from Linux-like repository
linkParser :: Parsec String () Inlines
linkParser = do
mpage <- many1 (alphaNum <|> char '_')
spacetab
char '('
mansect <- digit
char ')'
other <- many anyChar
let manurl pagename section = "../"++section++"/"++pagename++"."++section
lnkInls = link (manurl mpage [mansect]) mpage (strong $ str mpage)
return $ lnkInls <> strong (str (" ("++[mansect] ++ ")") <> text other)
comment :: PandocMonad m => ManParser m Inlines
comment = mcomment >> return mempty
bareIP :: PandocMonad m => ManParser m ManToken
bareIP = msatisfy isBareIP where
isBareIP (MMacro "IP" []) = True
isBareIP _ = False
parseCodeBlock :: PandocMonad m => ManParser m Blocks
parseCodeBlock = try $ do
optional bareIP -- some people indent their code
mmacro "nf"
toks <- many (mstr <|> mline <|> mmaybeLink <|> memptyLine <|> mcomment)
mmacro "fi"
return $ codeBlock (removeFinalNewline $
intercalate "\n" . catMaybes $
extractText <$> toks)
where
removeFinalNewline [] = []
removeFinalNewline xs = if last xs == '\n' then init xs else xs
extractText :: ManToken -> Maybe String
extractText (MStr (s, _)) = Just s
extractText (MLine ss) = Just . concat $ map fst ss -- TODO maybe unwords?
extractText (MMaybeLink s) = Just s
extractText MEmptyLine = Just "" -- string are intercalated with '\n', this prevents double '\n'
extractText _ = Nothing
parseHeader :: PandocMonad m => ManParser m Blocks
parseHeader = do
MMacro name args <- mmacro "SH" <|> mmacro "SS"
contents <- if null args
then do
MLine ils <- mline
return $ mconcat $ map strToInlines ils
else do
return $ mconcat
$ intersperse B.space
$ map strToInlines args
let lvl = if name == "SH" then 1 else 2
return $ header lvl contents
type ListBuilder = [Blocks] -> Blocks
parseList :: PandocMonad m => ManParser m Blocks
parseList = try $ do
xx <- many1 paras
let bls = map snd xx
let bldr = fst $ head xx
return $ bldr bls
where
listKind :: [RoffStr] -> ListBuilder
listKind ((cs, _):_:[]) =
let cs' = if not ('.' `elem` cs || ')' `elem` cs) then cs ++ "." else cs
in case Parsec.runParser anyOrderedListMarker defaultParserState
"list marker" cs' of
Right (start, listtype, listdelim)
| cs == cs' -> orderedListWith (start, listtype, listdelim)
| otherwise -> orderedListWith (start, listtype, DefaultDelim)
Left _ -> bulletList
listKind _ = bulletList
paras :: PandocMonad m => ManParser m (ListBuilder, Blocks)
paras = do
(MMacro _ args) <- mmacro "IP"
let lbuilder = listKind args
inls <- parseInlines
continuations <- mconcat <$> many continuation
return $ (lbuilder, para inls <> continuations)
continuation :: PandocMonad m => ManParser m Blocks
continuation = do
mmacro "RS"
bs <- mconcat <$> many (notFollowedBy (mmacro "RE") >> parseBlock)
mmacro "RE"
return bs
-- In case of weird man file it will be parsed succesfully
parseSkipMacro :: PandocMonad m => ManParser m Blocks
parseSkipMacro = do
pos <- getPosition
tok <- mmacroAny
case tok of
MMacro mkind _ -> report $ SkippedContent ('.':mkind) pos
_ -> return () -- shouldn't happen
return mempty
|