summaryrefslogtreecommitdiff
path: root/src/Text/Pandoc/Readers/LaTeX.hs
blob: 3b764584f563569f056ec75801826beaa194f132 (plain)
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
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
{-
Copyright (C) 2006-7 John MacFarlane <jgm@berkeley.edu>

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.LaTeX
   Copyright   : Copyright (C) 2006-7 John MacFarlane
   License     : GNU GPL, version 2 or above 

   Maintainer  : John MacFarlane <jgm@berkeley.edu>
   Stability   : alpha
   Portability : portable

Conversion of LaTeX to 'Pandoc' document.
-}
module Text.Pandoc.Readers.LaTeX ( 
                                  readLaTeX,
                                  rawLaTeXInline,
                                  rawLaTeXEnvironment
                                 ) where

import Text.ParserCombinators.Parsec
import Text.ParserCombinators.Pandoc
import Text.Pandoc.Definition
import Text.Pandoc.Shared 
import Data.Maybe ( fromMaybe )
import Data.Char ( chr )

-- | Parse LaTeX from string and return 'Pandoc' document.
readLaTeX :: ParserState   -- ^ Parser state, including options for parser
          -> String        -- ^ String to parse
          -> Pandoc
readLaTeX = readWith parseLaTeX

-- for testing
testString = testStringWith parseLaTeX

-- characters with special meaning
specialChars = "\\$%&^&_~#{}\n \t|<>'\"-"

--
-- utility functions
--

-- | Returns text between brackets and its matching pair.
bracketedText openB closeB = try (do
  char openB
  result <- many (choice [ oneOfStrings [ ['\\', openB], ['\\', closeB] ],
                           count 1 (noneOf [openB, closeB]),
                           bracketedText openB closeB ])
  char closeB
  return ([openB] ++ (concat result) ++ [closeB]))

-- | Returns an option or argument of a LaTeX command.
optOrArg = choice [ (bracketedText '{' '}'), (bracketedText '[' ']') ]

-- | True if the string begins with '{'.
isArg ('{':rest) = True
isArg other = False

-- | Returns list of options and arguments of a LaTeX command.
commandArgs = many optOrArg

-- | Parses LaTeX command, returns (name, star, list of options or arguments).
command = try (do
  char '\\'
  name <- many1 alphaNum
  star <- option "" (string "*")  -- some commands have starred versions
  args <- commandArgs
  return (name, star, args))

begin name = try (do
  string "\\begin{"
  string name
  char '}'
  option [] commandArgs
  spaces
  return name)

end name = try (do
  string "\\end{"
  string name
  char '}'
  spaces
  return name)

-- | Returns a list of block elements containing the contents of an
-- environment.
environment name = try (do
  begin name
  spaces
  contents <- manyTill block (end name)
  return contents)

anyEnvironment =  try (do
  string "\\begin{"
  name <- many alphaNum 
  star <- option "" (string "*") -- some environments have starred variants
  char '}'
  option [] commandArgs
  spaces
  contents <- manyTill block (end (name ++ star))
  return (BlockQuote contents))

--
-- parsing documents
--

-- | Process LaTeX preamble, extracting metadata.
processLaTeXPreamble = try (do
  manyTill (choice [bibliographic, comment, unknownCommand, nullBlock]) 
           (try (string "\\begin{document}"))
  spaces)

-- | Parse LaTeX and return 'Pandoc'.
parseLaTeX = do
  option () processLaTeXPreamble -- preamble might not be present (fragment)
  blocks <- parseBlocks
  spaces
  option "" (try (string "\\end{document}")) -- might not be present (in fragment)
  spaces
  eof
  state <- getState
  let blocks' = filter (/= Null) blocks
  let title' = stateTitle state
  let authors' = stateAuthors state
  let date' = stateDate state
  return (Pandoc (Meta title' authors' date')  blocks')

--
-- parsing blocks
--

parseBlocks = do
  spaces
  result <- many block
  return result

block = choice [ hrule, codeBlock, header, list, blockQuote, mathBlock,
                 comment, bibliographic, para, specialEnvironment,
                 itemBlock, unknownEnvironment, unknownCommand ] <?>
                 "block"

--
-- header blocks
--

header = choice (map headerLevel (enumFromTo 1 5)) <?> "header"

headerLevel n = try (do
  let subs = concat $ replicate (n - 1) "sub"
  string ("\\" ++ subs ++ "section")
  option ' ' (char '*')
  char '{'
  title <- manyTill inline (char '}')
  spaces
  return (Header n (normalizeSpaces title)))

--
-- hrule block
--

hrule = try (do
  oneOfStrings [ "\\begin{center}\\rule{3in}{0.4pt}\\end{center}\n\n", 
                 "\\newpage" ]
  spaces
  return HorizontalRule)

--
-- code blocks
--

codeBlock = try (do
  string "\\begin{verbatim}"  -- don't use begin function because it 
                              -- gobbles whitespace
  option "" blanklines        -- we want to gobble blank lines, but not 
                              -- leading space
  contents <- manyTill anyChar (try (string "\\end{verbatim}"))
  spaces
  return (CodeBlock (stripTrailingNewlines contents)))

--
-- block quotes
--

blockQuote = choice [ blockQuote1, blockQuote2 ] <?> "blockquote"

blockQuote1 = try (do
  blocks <- environment "quote"
  spaces
  return (BlockQuote blocks))

blockQuote2 = try (do
  blocks <- environment "quotation"
  spaces
  return (BlockQuote blocks))

--
-- math block
--

mathBlock = mathBlockWith (begin "equation") (end "equation") <|> 
            mathBlockWith (begin "displaymath") (end "displaymath") <|>
            mathBlockWith (string "\\[") (string "\\]") <?> "math block"

mathBlockWith start end = try (do
  start
  spaces
  result <- manyTill anyChar end
  spaces
  return (BlockQuote [Para [TeX ("$" ++ result ++ "$")]]))

--
-- list blocks
--

list = bulletList <|> orderedList <?> "list"

listItem = try (do
    ("item", _, _) <- command
    spaces
    state <- getState
    let oldParserContext = stateParserContext state
    updateState (\state -> state {stateParserContext = ListItemState})
    blocks <- many block
    updateState (\state -> state {stateParserContext = oldParserContext})
    return blocks)

orderedList = try (do
    begin "enumerate"
    spaces
    items <- many listItem
    end "enumerate"
    spaces
    return (OrderedList items))

bulletList = try (do
    begin "itemize"
    spaces
    items <- many listItem
    end "itemize"
    spaces
    return (BulletList items))

--
-- paragraph block
--

para = try (do 
  result <- many1 inline
  spaces
  return (Para (normalizeSpaces result)))

--
-- title authors date
--

bibliographic = choice [ maketitle, title, authors, date ]

maketitle = try (do
  string "\\maketitle"
  spaces
  return Null)

title = try (do
  string "\\title{"
  tit <- manyTill inline (char '}')
  spaces
  updateState (\state -> state { stateTitle = tit })
  return Null)

authors = try (do
  string "\\author{"
  authors <- manyTill anyChar (char '}')
  spaces
  let authors' = map removeLeadingTrailingSpace $ lines $
                 substitute "\\\\" "\n" authors
  updateState (\state -> state { stateAuthors = authors' })
  return Null)

date = try (do
  string "\\date{"
  date' <- manyTill anyChar (char '}')
  spaces
  updateState (\state -> state { stateDate = date' })
  return Null)

--
-- item block
-- for use in unknown environments that aren't being parsed as raw latex
--

-- this forces items to be parsed in different blocks
itemBlock = try (do
  ("item", _, args) <- command
  state <- getState
  if (stateParserContext state == ListItemState)
     then fail "item should be handled by list block"
     else if null args 
             then return Null
             else return (Plain [Str (stripFirstAndLast (head args))]))

--
-- raw LaTeX 
--

specialEnvironment = do  -- these are always parsed as raw
  lookAhead (choice (map (\name -> begin name)  ["tabular", "figure",
              "tabbing", "eqnarry", "picture", "table", "verse", "theorem"]))
  rawLaTeXEnvironment

-- | Parse any LaTeX environment and return a Para block containing
-- the whole literal environment as raw TeX.
rawLaTeXEnvironment :: GenParser Char st Block
rawLaTeXEnvironment = try (do
  string "\\begin"
  char '{'
  name <- many1 alphaNum
  star <- option "" (string "*") -- for starred variants
  let name' = name ++ star
  char '}'
  args <- option [] commandArgs
  let argStr = concat args
  contents <- manyTill (choice [(many1 (noneOf "\\")), 
                                 (do 
                                    (Para [TeX str]) <- rawLaTeXEnvironment
                                    return str),
                                 string "\\" ]) 
                       (end name')
  spaces
  return (Para [TeX ("\\begin{" ++ name' ++ "}" ++ argStr ++ 
          (concat contents) ++ "\\end{" ++ name' ++ "}")]))

unknownEnvironment = try (do
  state <- getState
  result <- if stateParseRaw state -- check whether we should include raw TeX 
               then rawLaTeXEnvironment -- if so, get whole raw environment
               else anyEnvironment      -- otherwise just the contents
  return result)

unknownCommand = try (do
  notFollowedBy' (string "\\end{itemize}")
  notFollowedBy' (string "\\end{enumerate}")
  notFollowedBy' (string "\\end{document}")
  (name, star, args) <- command
  spaces
  let argStr = concat args
  state <- getState
  if (name == "item") && ((stateParserContext state) == ListItemState)
     then fail "should not be parsed as raw"
     else string ""
  if stateParseRaw state
     then return (Plain [TeX ("\\" ++ name ++ star ++ argStr)])
     else return (Plain [Str (joinWithSep " " args)]))

-- latex comment
comment = try (do
  char '%'
  result <- manyTill anyChar newline
  spaces
  return Null)

-- 
-- inline
--

inline =  choice [ strong, emph, ref, lab, code, linebreak, spacer,
                   math, ellipses,
                   emDash, enDash, hyphen, quoted, apostrophe,
                   accentedChar, specialChar, specialInline, escapedChar,
                   unescapedChar, str, endline, whitespace ] <?> "inline"

specialInline = choice [ link, image, footnote, rawLaTeXInline ] 
                <?> "link, raw TeX, note, or image"

accentedChar = normalAccentedChar <|> specialAccentedChar

normalAccentedChar = try (do
  char '\\'
  accent <- oneOf "'`^\"~"
  character <- choice [ between (char '{') (char '}') anyChar, anyChar ]
  let table = fromMaybe [] $ lookup character accentTable 
  let result = case lookup accent table of
                 Just num  -> chr num
                 Nothing   -> '?'
  return (Str [result]))

-- an association list of letters and association list of accents
-- and decimal character numbers.
accentTable = 
  [ ('A', [('`', 192), ('\'', 193), ('^', 194), ('~', 195), ('"', 196)]),
    ('E', [('`', 200), ('\'', 201), ('^', 202), ('"', 203)]),
    ('I', [('`', 204), ('\'', 205), ('^', 206), ('"', 207)]),
    ('N', [('~', 209)]),
    ('O', [('`', 210), ('\'', 211), ('^', 212), ('~', 213), ('"', 214)]),
    ('U', [('`', 217), ('\'', 218), ('^', 219), ('"', 220)]),
    ('a', [('`', 224), ('\'', 225), ('^', 227), ('"', 228)]),
    ('e', [('`', 232), ('\'', 233), ('^', 234), ('"', 235)]),
    ('i', [('`', 236), ('\'', 237), ('^', 238), ('"', 239)]),
    ('n', [('~', 241)]),
    ('o', [('`', 242), ('\'', 243), ('^', 244), ('~', 245), ('"', 246)]),
    ('u', [('`', 249), ('\'', 250), ('^', 251), ('"', 252)]) ]

specialAccentedChar = choice [ ccedil, aring, iuml, szlig, aelig,
                              oslash, pound, euro, copyright, sect ]

ccedil = try (do
  char '\\'
  letter <- oneOfStrings ["cc", "cC"]
  let num = if letter == "cc" then 231 else 199
  return (Str [chr num]))

aring = try (do
  char '\\'
  letter <- oneOfStrings ["aa", "AA"]
  let num = if letter == "aa" then 229 else 197
  return (Str [chr num]))

iuml = try (do
  string "\\\""
  oneOfStrings ["\\i", "{\\i}"]
  return (Str [chr 239]))

icirc = try (do
  string "\\^"
  oneOfStrings ["\\i", "{\\i}"]
  return (Str [chr 238]))

szlig = try (do
  string "\\ss"
  return (Str [chr 223]))

oslash = try (do
  char '\\'
  letter <- choice [char 'o', char 'O']
  let num = if letter == 'o' then 248 else 216
  return (Str [chr num]))

aelig = try (do
  char '\\'
  letter <- oneOfStrings ["ae", "AE"]
  let num = if letter == "ae" then 230 else 198
  return (Str [chr num]))

pound = try (do 
  string "\\pounds"
  return (Str [chr 163]))

euro = try (do
  string "\\euro"
  return (Str [chr 8364]))

copyright = try (do
  string "\\copyright"
  return (Str [chr 169]))

sect = try (do
  string "\\S"
  return (Str [chr 167]))

escapedChar = do
  result <- escaped (oneOf " $%^&_#{}\n")
  return (if result == Str "\n" then Str " " else result)

unescapedChar = do  -- ignore standalone, nonescaped special characters
  oneOf "$^&_#{}|<>"
  return (Str "")

specialChar = choice [ backslash, bar, lt, gt ]

backslash = try (do 
  string "\\textbackslash"
  return (Str "\\"))

bar = try (do
  string "\\textbar"
  return (Str "\\"))

lt = try (do
  string "\\textless"
  return (Str "<"))

gt = try (do
  string "\\textgreater"
  return (Str ">"))

code = try (do 
  string "\\verb"
  marker <- anyChar
  result <- manyTill anyChar (char marker)
  let result' = removeLeadingTrailingSpace result
  return (Code result'))

emph = try (do 
  oneOfStrings [ "\\emph{", "\\textit{" ]
  result <- manyTill inline (char '}')
  return (Emph result))

apostrophe = do
  char '\''
  return Apostrophe

quoted = do
  doubleQuoted <|> singleQuoted

singleQuoted = try (do
  result <- enclosed singleQuoteStart singleQuoteEnd inline
  return $ Quoted SingleQuote $ normalizeSpaces result)

doubleQuoted = try (do
  result <- enclosed doubleQuoteStart doubleQuoteEnd inline
  return $ Quoted DoubleQuote $ normalizeSpaces result)

singleQuoteStart = char '`'

singleQuoteEnd = try (do
  char '\''
  notFollowedBy alphaNum)

doubleQuoteStart = try (string "``")

doubleQuoteEnd = try (string "''")

ellipses = try (do
  string "\\ldots"
  option "" (try (string "{}"))
  return Ellipses)

enDash = try (do
  string "--"
  notFollowedBy (char '-')
  return EnDash) 

emDash = try (do
  string "---"
  return EmDash)

hyphen = do
  char '-'
  return (Str "-")

lab = try (do
  string "\\label{"
  result <- manyTill anyChar (char '}')
  return (Str ("(" ++ result ++ ")")))

ref = try (do
  string "\\ref{"
  result <- manyTill anyChar (char '}')
  return (Str (result)))

strong = try (do 
  string "\\textbf{"
  result <- manyTill inline (char '}')
  return (Strong result))

whitespace = do
  many1 (oneOf "~ \t")
  return Space

-- hard line break
linebreak = try (do
  string "\\\\"
  return LineBreak)

spacer = try $ do
  string "\\," 
  return (Str "")

str = do 
  result <- many1 (noneOf specialChars)
  return (Str result)

-- endline internal to paragraph
endline = try (do
  newline
  notFollowedBy blankline
  return Space)

-- math
math = math1 <|> math2 <?> "math"

math1 = try (do
  char '$'
  result <- many (noneOf "$")
  char '$'
  return (TeX ("$" ++ result ++ "$")))

math2 = try (do
  string "\\("
  result <- many (noneOf "$")
  string "\\)"
  return (TeX ("$" ++ result ++ "$")))

--
-- links and images
--

link = try (do
  string "\\href{"
  url <- manyTill anyChar (char '}')
  char '{'
  label <- manyTill inline (char '}') 
  return (Link (normalizeSpaces label) (url, "")))

image = try (do
  ("includegraphics", _, args) <- command
  let args' = filter isArg args -- filter out options
  let src = if null args' then
              ("", "")
            else
              (stripFirstAndLast (head args'), "")
  return (Image [Str "image"] src))

footnote = try (do
  (name, _, (contents:[])) <- command
  if ((name == "footnote") || (name == "thanks"))
     then string ""
     else fail "not a footnote or thanks command"
  let contents' = stripFirstAndLast contents
  -- parse the extracted block, which may contain various block elements:
  rest <- getInput
  setInput $ contents'
  blocks <- parseBlocks
  setInput rest
  return (Note blocks))

-- | Parse any LaTeX command and return it in a raw TeX inline element.
rawLaTeXInline :: GenParser Char ParserState Inline
rawLaTeXInline = try (do
  (name, star, args) <- command
  let argStr = concat args
  state <- getState
  if ((name == "begin") || (name == "end") || (name == "item"))
     then fail "not an inline command" 
     else string ""
  return (TeX ("\\" ++ name ++ star ++ argStr)))