aboutsummaryrefslogtreecommitdiff
path: root/src/Text/Pandoc/Shared.hs
blob: 274c969ca238087db00dd692d3bfdc0d52c1ae70 (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
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
{-# LANGUAGE DeriveDataTypeable #-}
{-
Copyright (C) 2006-2010 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.Shared
   Copyright   : Copyright (C) 2006-2010 John MacFarlane
   License     : GNU GPL, version 2 or above 

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

Utility functions and definitions used by the various Pandoc modules.
-}
module Text.Pandoc.Shared (
                     -- * List processing
                     splitBy,
                     splitByIndices,
                     substitute,
                     -- * Text processing
                     backslashEscapes,
                     escapeStringUsing,
                     stripTrailingNewlines,
                     removeLeadingTrailingSpace,
                     removeLeadingSpace,
                     removeTrailingSpace,
                     stripFirstAndLast,
                     camelCaseToHyphenated,
                     toRomanNumeral,
                     escapeURI,
                     unescapeURI,
                     wrapped,
                     wrapIfNeeded,
                     wrappedTeX,
                     wrapTeXIfNeeded,
                     BlockWrapper (..),
                     wrappedBlocksToDoc,
                     tabFilter,
                     -- * Parsing
                     (>>~),
                     anyLine,
                     many1Till,
                     notFollowedBy',
                     oneOfStrings,
                     spaceChar,
                     skipSpaces,
                     blankline,
                     blanklines,
                     enclosed,
                     stringAnyCase,
                     parseFromString,
                     lineClump,
                     charsInBalanced,
                     charsInBalanced',
                     romanNumeral,
                     emailAddress,
                     uri,
                     withHorizDisplacement,
                     nullBlock,
                     failIfStrict,
                     failUnlessLHS,
                     escaped,
                     anyOrderedListMarker,
                     orderedListMarker,
                     charRef,
                     readWith,
                     testStringWith,
                     ParserState (..),
                     defaultParserState,
                     HeaderType (..),
                     ParserContext (..),
                     QuoteContext (..),
                     NoteTable,
                     KeyTable,
                     lookupKeySrc,
                     refsMatch,
                     -- * Prettyprinting
                     hang',
                     prettyPandoc,
                     -- * Pandoc block and inline list processing
                     orderedListMarkers,
                     normalizeSpaces,
                     compactify,
                     Element (..),
                     hierarchicalize,
                     uniqueIdent,
                     isHeaderBlock,
                     -- * Writer options
                     HTMLMathMethod (..),
                     ObfuscationMethod (..),
                     WriterOptions (..),
                     defaultWriterOptions,
                     -- * File handling
                     inDirectory,
                     readDataFile
                    ) where

import Text.Pandoc.Definition
import qualified Text.Pandoc.UTF8 as UTF8 (readFile, putStrLn)
import Text.ParserCombinators.Parsec
import Text.PrettyPrint.HughesPJ ( Doc, fsep, ($$), (<>), empty, isEmpty, text, nest )
import qualified Text.PrettyPrint.HughesPJ as PP
import Text.Pandoc.CharacterReferences ( characterReference )
import Data.Char ( toLower, toUpper, ord, isLower, isUpper, isAlpha, isAscii,
                   isLetter, isDigit )
import Data.List ( find, isPrefixOf, intercalate )
import Network.URI ( parseURI, URI (..), isAllowedInURI, escapeURIString, unEscapeString )
import Codec.Binary.UTF8.String ( encodeString, decodeString )
import System.Directory
import System.FilePath ( (</>) )
import Data.Generics
import qualified Control.Monad.State as S
import Control.Monad (join)
import Paths_pandoc (getDataFileName)

--
-- List processing
--

-- | Split list by groups of one or more sep.
splitBy :: (Eq a) => a -> [a] -> [[a]]
splitBy _ [] = []
splitBy sep lst = 
  let (first, rest) = break (== sep) lst
      rest'         = dropWhile (== sep) rest
  in  first:(splitBy sep rest')

-- | Split list into chunks divided at specified indices.
splitByIndices :: [Int] -> [a] -> [[a]]
splitByIndices [] lst = [lst]
splitByIndices (x:xs) lst =
    let (first, rest) = splitAt x lst in
    first:(splitByIndices (map (\y -> y - x)  xs) rest)

-- | Replace each occurrence of one sublist in a list with another.
substitute :: (Eq a) => [a] -> [a] -> [a] -> [a]
substitute _ _ [] = []
substitute [] _ lst = lst
substitute target replacement lst = 
    if target `isPrefixOf` lst
       then replacement ++ (substitute target replacement $ drop (length target) lst)
       else (head lst):(substitute target replacement $ tail lst)

--
-- Text processing
--

-- | Returns an association list of backslash escapes for the
-- designated characters.
backslashEscapes :: [Char]    -- ^ list of special characters to escape
                 -> [(Char, String)]
backslashEscapes = map (\ch -> (ch, ['\\',ch]))

-- | Escape a string of characters, using an association list of
-- characters and strings.
escapeStringUsing :: [(Char, String)] -> String -> String
escapeStringUsing _ [] = ""
escapeStringUsing escapeTable (x:xs) = 
  case (lookup x escapeTable) of
       Just str  -> str ++ rest
       Nothing   -> x:rest
  where rest = escapeStringUsing escapeTable xs

-- | Strip trailing newlines from string.
stripTrailingNewlines :: String -> String
stripTrailingNewlines = reverse . dropWhile (== '\n') . reverse

-- | Remove leading and trailing space (including newlines) from string.
removeLeadingTrailingSpace :: String -> String
removeLeadingTrailingSpace = removeLeadingSpace . removeTrailingSpace

-- | Remove leading space (including newlines) from string.
removeLeadingSpace :: String -> String
removeLeadingSpace = dropWhile (`elem` " \n\t")

-- | Remove trailing space (including newlines) from string.
removeTrailingSpace :: String -> String
removeTrailingSpace = reverse . removeLeadingSpace . reverse

-- | Strip leading and trailing characters from string
stripFirstAndLast :: String -> String
stripFirstAndLast str =
  drop 1 $ take ((length str) - 1) str

-- | Change CamelCase word to hyphenated lowercase (e.g., camel-case). 
camelCaseToHyphenated :: String -> String
camelCaseToHyphenated [] = ""
camelCaseToHyphenated (a:b:rest) | isLower a && isUpper b =
  a:'-':(toLower b):(camelCaseToHyphenated rest)
camelCaseToHyphenated (a:rest) = (toLower a):(camelCaseToHyphenated rest)

-- | Convert number < 4000 to uppercase roman numeral.
toRomanNumeral :: Int -> String
toRomanNumeral x =
  if x >= 4000 || x < 0
     then "?"
     else case x of
              _ | x >= 1000 -> "M" ++ toRomanNumeral (x - 1000)
              _ | x >= 900  -> "CM" ++ toRomanNumeral (x - 900)
              _ | x >= 500  -> "D" ++ toRomanNumeral (x - 500)
              _ | x >= 400  -> "CD" ++ toRomanNumeral (x - 400)
              _ | x >= 100  -> "C" ++ toRomanNumeral (x - 100)
              _ | x >= 90   -> "XC" ++ toRomanNumeral (x - 90)
              _ | x >= 50   -> "L"  ++ toRomanNumeral (x - 50)
              _ | x >= 40   -> "XL" ++ toRomanNumeral (x - 40)
              _ | x >= 10   -> "X" ++ toRomanNumeral (x - 10)
              _ | x >= 9    -> "IX" ++ toRomanNumeral (x - 5)
              _ | x >= 5    -> "V" ++ toRomanNumeral (x - 5)
              _ | x >= 4    -> "IV" ++ toRomanNumeral (x - 4)
              _ | x >= 1    -> "I" ++ toRomanNumeral (x - 1)
              _             -> ""

-- | Escape unicode characters in a URI.  Characters that are
-- already valid in a URI, including % and ?, are left alone.
escapeURI :: String -> String
escapeURI = escapeURIString isAllowedInURI . encodeString

-- | Unescape unicode and some special characters in a URI, but
-- without introducing spaces.
unescapeURI :: String -> String
unescapeURI = escapeURIString (\c -> isAllowedInURI c || not (isAscii c)) .
               decodeString . unEscapeString

-- | Wrap inlines to line length.
wrapped :: Monad m => ([Inline] -> m Doc) -> [Inline] -> m Doc
wrapped listWriter sect = (mapM listWriter $ splitBy Space sect) >>= 
                          return . fsep

-- | Wrap inlines if the text wrap option is selected.
wrapIfNeeded :: Monad m => WriterOptions -> ([Inline] -> m Doc) -> 
                           [Inline] -> m Doc
wrapIfNeeded opts = if writerWrapText opts
                       then wrapped 
                       else ($)

-- auxiliary function for wrappedTeX
isNote :: Inline -> Bool
isNote (Note _) = True
isNote _ = False

-- | Wrap inlines to line length, treating footnotes in a way that
-- makes sense in LaTeX and ConTeXt.
wrappedTeX :: Monad m 
           => Bool
           -> ([Inline] -> m Doc) 
           -> [Inline] 
           -> m Doc
wrappedTeX includePercent listWriter sect = do
  let (firstpart, rest) = break isNote sect
  firstpartWrapped <- wrapped listWriter firstpart
  if null rest
     then return firstpartWrapped
     else do let (note:rest') = rest
             let (rest1, rest2) = break (== Space) rest'
             -- rest1 is whatever comes between the note and a Space.
             -- if the note is followed directly by a Space, rest1 is null.
             -- rest1 is printed after the note but before the line break,
             -- to avoid spurious blank space the note and immediately
             -- following punctuation.
             rest1Out <- if null rest1
                            then return empty
                            else listWriter rest1
             rest2Wrapped <- if null rest2
                                then return empty
                                else wrappedTeX includePercent listWriter (tail rest2)
             noteText <- listWriter [note]
             return $ (firstpartWrapped <> if includePercent then PP.char '%' else empty) $$ 
                      (noteText <> rest1Out) $$
                      rest2Wrapped

-- | Wrap inlines if the text wrap option is selected, specialized 
-- for LaTeX and ConTeXt.
wrapTeXIfNeeded :: Monad m 
                => WriterOptions
                -> Bool
                -> ([Inline] -> m Doc) 
                -> [Inline] 
                -> m Doc
wrapTeXIfNeeded opts includePercent = if writerWrapText opts
                                         then wrappedTeX includePercent
                                         else ($)

-- | Indicates whether block should be surrounded by blank lines (@Pad@) or not (@Reg@).
data BlockWrapper = Pad Doc | Reg Doc

-- | Converts a list of wrapped blocks to a Doc, with appropriate spaces around blocks.
wrappedBlocksToDoc :: [BlockWrapper] -> Doc
wrappedBlocksToDoc = foldr addBlock empty
     where addBlock (Pad d) accum | isEmpty accum = d
           addBlock (Pad d) accum = d $$ text "" $$ accum
           addBlock (Reg d) accum = d $$ accum

-- | Convert tabs to spaces and filter out DOS line endings.
-- Tabs will be preserved if tab stop is set to 0.
tabFilter :: Int       -- ^ Tab stop
          -> String    -- ^ Input
          -> String
tabFilter tabStop =
  let go _ [] = ""
      go _ ('\n':xs) = '\n' : go tabStop xs
      go _ ('\r':'\n':xs) = '\n' : go tabStop xs
      go _ ('\r':xs) = '\n' : go tabStop xs
      go spsToNextStop ('\t':xs) =
        if tabStop == 0
           then '\t' : go tabStop xs
           else replicate spsToNextStop ' ' ++ go tabStop xs
      go 1 (x:xs) =
        x : go tabStop xs
      go spsToNextStop (x:xs) =
        x : go (spsToNextStop - 1) xs
  in  go tabStop

--
-- Parsing
--

-- | Like >>, but returns the operation on the left.
-- (Suggested by Tillmann Rendel on Haskell-cafe list.)
(>>~) :: (Monad m) => m a -> m b -> m a
a >>~ b = a >>= \x -> b >> return x

-- | Parse any line of text
anyLine :: GenParser Char st [Char]
anyLine = manyTill anyChar newline

-- | Like @manyTill@, but reads at least one item.
many1Till :: GenParser tok st a
	     -> GenParser tok st end
	     -> GenParser tok st [a]
many1Till p end = do
         first <- p
         rest <- manyTill p end
         return (first:rest)

-- | A more general form of @notFollowedBy@.  This one allows any 
-- type of parser to be specified, and succeeds only if that parser fails.
-- It does not consume any input.
notFollowedBy' :: Show b => GenParser a st b -> GenParser a st ()
notFollowedBy' p  = try $ join $  do  a <- try p
                                      return (unexpected (show a))
                                  <|>
                                  return (return ())
-- (This version due to Andrew Pimlott on the Haskell mailing list.)

-- | Parses one of a list of strings (tried in order).  
oneOfStrings :: [String] -> GenParser Char st String
oneOfStrings listOfStrings = choice $ map (try . string) listOfStrings

-- | Parses a space or tab.
spaceChar :: CharParser st Char
spaceChar = char ' ' <|> char '\t'

-- | Skips zero or more spaces or tabs.
skipSpaces :: GenParser Char st ()
skipSpaces = skipMany spaceChar

-- | Skips zero or more spaces or tabs, then reads a newline.
blankline :: GenParser Char st Char
blankline = try $ skipSpaces >> newline

-- | Parses one or more blank lines and returns a string of newlines.
blanklines :: GenParser Char st [Char]
blanklines = many1 blankline

-- | Parses material enclosed between start and end parsers.
enclosed :: GenParser Char st t   -- ^ start parser
	    -> GenParser Char st end  -- ^ end parser
	    -> GenParser Char st a    -- ^ content parser (to be used repeatedly)
	    -> GenParser Char st [a]
enclosed start end parser = try $ 
  start >> notFollowedBy space >> many1Till parser end

-- | Parse string, case insensitive.
stringAnyCase :: [Char] -> CharParser st String
stringAnyCase [] = string ""
stringAnyCase (x:xs) = do
  firstChar <- char (toUpper x) <|> char (toLower x)
  rest <- stringAnyCase xs
  return (firstChar:rest)

-- | Parse contents of 'str' using 'parser' and return result.
parseFromString :: GenParser tok st a -> [tok] -> GenParser tok st a
parseFromString parser str = do
  oldPos <- getPosition
  oldInput <- getInput
  setInput str
  result <- parser
  setInput oldInput
  setPosition oldPos
  return result

-- | Parse raw line block up to and including blank lines.
lineClump :: GenParser Char st String
lineClump = blanklines 
          <|> (many1 (notFollowedBy blankline >> anyLine) >>= return . unlines)

-- | Parse a string of characters between an open character
-- and a close character, including text between balanced
-- pairs of open and close, which must be different. For example,
-- @charsInBalanced '(' ')'@ will parse "(hello (there))"
-- and return "hello (there)".  Stop if a blank line is
-- encountered.
charsInBalanced :: Char -> Char -> GenParser Char st String
charsInBalanced open close = try $ do
  char open
  raw <- many $     (many1 (noneOf [open, close, '\n']))
                <|> (do res <- charsInBalanced open close
                        return $ [open] ++ res ++ [close])
                <|> try (string "\n" >>~ notFollowedBy' blanklines)
  char close
  return $ concat raw

-- | Like @charsInBalanced@, but allow blank lines in the content.
charsInBalanced' :: Char -> Char -> GenParser Char st String
charsInBalanced' open close = try $ do
  char open
  raw <- many $       (many1 (noneOf [open, close]))
                  <|> (do res <- charsInBalanced' open close
                          return $ [open] ++ res ++ [close])
  char close
  return $ concat raw

-- Auxiliary functions for romanNumeral:

lowercaseRomanDigits :: [Char]
lowercaseRomanDigits = ['i','v','x','l','c','d','m']

uppercaseRomanDigits :: [Char]
uppercaseRomanDigits = map toUpper lowercaseRomanDigits

-- | Parses a roman numeral (uppercase or lowercase), returns number.
romanNumeral :: Bool                  -- ^ Uppercase if true
             -> GenParser Char st Int
romanNumeral upperCase = do
    let romanDigits = if upperCase 
                         then uppercaseRomanDigits 
                         else lowercaseRomanDigits
    lookAhead $ oneOf romanDigits 
    let [one, five, ten, fifty, hundred, fivehundred, thousand] = 
          map char romanDigits
    thousands <- many thousand >>= (return . (1000 *) . length)
    ninehundreds <- option 0 $ try $ hundred >> thousand >> return 900
    fivehundreds <- many fivehundred >>= (return . (500 *) . length)
    fourhundreds <- option 0 $ try $ hundred >> fivehundred >> return 400
    hundreds <- many hundred >>= (return . (100 *) . length)
    nineties <- option 0 $ try $ ten >> hundred >> return 90
    fifties <- many fifty >>= (return . (50 *) . length)
    forties <- option 0 $ try $ ten >> fifty >> return 40
    tens <- many ten >>= (return . (10 *) . length)
    nines <- option 0 $ try $ one >> ten >> return 9
    fives <- many five >>= (return . (5 *) . length)
    fours <- option 0 $ try $ one >> five >> return 4
    ones <- many one >>= (return . length)
    let total = thousands + ninehundreds + fivehundreds + fourhundreds +
                hundreds + nineties + fifties + forties + tens + nines +
                fives + fours + ones
    if total == 0
       then fail "not a roman numeral"
       else return total

-- Parsers for email addresses and URIs

emailChar :: GenParser Char st Char
emailChar = alphaNum <|> oneOf "-+_."

domainChar :: GenParser Char st Char
domainChar = alphaNum <|> char '-'

domain :: GenParser Char st [Char]
domain = do
  first <- many1 domainChar
  dom <- many1 $ try (char '.' >> many1 domainChar )
  return $ intercalate "." (first:dom)

-- | Parses an email address; returns original and corresponding
-- escaped mailto: URI.
emailAddress :: GenParser Char st (String, String)
emailAddress = try $ do
  firstLetter <- alphaNum
  restAddr <- many emailChar
  let addr = firstLetter:restAddr
  char '@'
  dom <- domain
  let full = addr ++ '@':dom
  return (full, escapeURI $ "mailto:" ++ full)

-- | Parses a URI. Returns pair of original and URI-escaped version.
uri :: GenParser Char st (String, String)
uri = try $ do
  let protocols = [ "http:", "https:", "ftp:", "file:", "mailto:",
                    "news:", "telnet:" ]
  lookAhead $ oneOfStrings protocols
  -- scan non-ascii characters and ascii characters allowed in a URI
  str <- many1 $ satisfy (\c -> not (isAscii c) || isAllowedInURI c)
  -- now see if they amount to an absolute URI
  case parseURI (escapeURI str) of
       Just uri' -> if uriScheme uri' `elem` protocols
                       then return (str, show uri')
                       else fail "not a URI"
       Nothing   -> fail "not a URI"

-- | Applies a parser, returns tuple of its results and its horizontal
-- displacement (the difference between the source column at the end
-- and the source column at the beginning). Vertical displacement
-- (source row) is ignored.
withHorizDisplacement :: GenParser Char st a  -- ^ Parser to apply
                      -> GenParser Char st (a, Int) -- ^ (result, displacement)
withHorizDisplacement parser = do
  pos1 <- getPosition
  result <- parser
  pos2 <- getPosition
  return (result, sourceColumn pos2 - sourceColumn pos1)

-- | Parses a character and returns 'Null' (so that the parser can move on
-- if it gets stuck).
nullBlock :: GenParser Char st Block
nullBlock = anyChar >> return Null

-- | Fail if reader is in strict markdown syntax mode.
failIfStrict :: GenParser Char ParserState ()
failIfStrict = do
  state <- getState
  if stateStrict state then fail "strict mode" else return ()

-- | Fail unless we're in literate haskell mode.
failUnlessLHS :: GenParser tok ParserState ()
failUnlessLHS = do
  state <- getState
  if stateLiterateHaskell state then return () else fail "Literate haskell feature"

-- | Parses backslash, then applies character parser.
escaped :: GenParser Char st Char  -- ^ Parser for character to escape
        -> GenParser Char st Inline
escaped parser = try $ do
  char '\\'
  result <- parser
  return (Str [result])

-- | Parses an uppercase roman numeral and returns (UpperRoman, number).
upperRoman :: GenParser Char st (ListNumberStyle, Int)
upperRoman = do
  num <- romanNumeral True
  return (UpperRoman, num)

-- | Parses a lowercase roman numeral and returns (LowerRoman, number).
lowerRoman :: GenParser Char st (ListNumberStyle, Int)
lowerRoman = do
  num <- romanNumeral False
  return (LowerRoman, num)

-- | Parses a decimal numeral and returns (Decimal, number).
decimal :: GenParser Char st (ListNumberStyle, Int)
decimal = do
  num <- many1 digit
  return (Decimal, read num)

-- | Parses a '#' returns (DefaultStyle, 1).
defaultNum :: GenParser Char st (ListNumberStyle, Int)
defaultNum = do
  char '#'
  return (DefaultStyle, 1)

-- | Parses a lowercase letter and returns (LowerAlpha, number).
lowerAlpha :: GenParser Char st (ListNumberStyle, Int)
lowerAlpha = do
  ch <- oneOf ['a'..'z']
  return (LowerAlpha, ord ch - ord 'a' + 1)

-- | Parses an uppercase letter and returns (UpperAlpha, number).
upperAlpha :: GenParser Char st (ListNumberStyle, Int)
upperAlpha = do
  ch <- oneOf ['A'..'Z']
  return (UpperAlpha, ord ch - ord 'A' + 1)

-- | Parses a roman numeral i or I
romanOne :: GenParser Char st (ListNumberStyle, Int)
romanOne = (char 'i' >> return (LowerRoman, 1)) <|>
           (char 'I' >> return (UpperRoman, 1))

-- | Parses an ordered list marker and returns list attributes.
anyOrderedListMarker :: GenParser Char st ListAttributes 
anyOrderedListMarker = choice $ 
  [delimParser numParser | delimParser <- [inPeriod, inOneParen, inTwoParens],
                           numParser <- [decimal, defaultNum, romanOne,
                           lowerAlpha, lowerRoman, upperAlpha, upperRoman]]

-- | Parses a list number (num) followed by a period, returns list attributes.
inPeriod :: GenParser Char st (ListNumberStyle, Int)
         -> GenParser Char st ListAttributes 
inPeriod num = try $ do
  (style, start) <- num
  char '.'
  let delim = if style == DefaultStyle
                 then DefaultDelim
                 else Period
  return (start, style, delim)
 
-- | Parses a list number (num) followed by a paren, returns list attributes.
inOneParen :: GenParser Char st (ListNumberStyle, Int)
           -> GenParser Char st ListAttributes 
inOneParen num = try $ do
  (style, start) <- num
  char ')'
  return (start, style, OneParen)

-- | Parses a list number (num) enclosed in parens, returns list attributes.
inTwoParens :: GenParser Char st (ListNumberStyle, Int)
            -> GenParser Char st ListAttributes 
inTwoParens num = try $ do
  char '('
  (style, start) <- num
  char ')'
  return (start, style, TwoParens)

-- | Parses an ordered list marker with a given style and delimiter,
-- returns number.
orderedListMarker :: ListNumberStyle 
                  -> ListNumberDelim 
                  -> GenParser Char st Int
orderedListMarker style delim = do
  let num = defaultNum <|>  -- # can continue any kind of list
            case style of
               DefaultStyle -> decimal
               Decimal      -> decimal
               UpperRoman   -> upperRoman
               LowerRoman   -> lowerRoman
               UpperAlpha   -> upperAlpha
               LowerAlpha   -> lowerAlpha
  let context = case delim of
               DefaultDelim -> inPeriod
               Period       -> inPeriod
               OneParen     -> inOneParen
               TwoParens    -> inTwoParens
  (start, _, _) <- context num
  return start

-- | Parses a character reference and returns a Str element.
charRef :: GenParser Char st Inline
charRef = do
  c <- characterReference
  return $ Str [c]

-- | Parse a string with a given parser and state.
readWith :: GenParser Char ParserState a      -- ^ parser
         -> ParserState                    -- ^ initial state
         -> String                         -- ^ input string
         -> a
readWith parser state input = 
    case runParser parser state "source" input of
      Left err     -> error $ "\nError:\n" ++ show err
      Right result -> result

-- | Parse a string with @parser@ (for testing).
testStringWith :: (Show a) => GenParser Char ParserState a
               -> String
               -> IO ()
testStringWith parser str = UTF8.putStrLn $ show $
                            readWith parser defaultParserState str

-- | Parsing options.
data ParserState = ParserState
    { stateParseRaw        :: Bool,          -- ^ Parse raw HTML and LaTeX?
      stateParserContext   :: ParserContext, -- ^ Inside list?
      stateQuoteContext    :: QuoteContext,  -- ^ Inside quoted environment?
      stateSanitizeHTML    :: Bool,          -- ^ Sanitize HTML?
      stateKeys            :: KeyTable,      -- ^ List of reference keys
#ifdef _CITEPROC
      stateCitations       :: [String],      -- ^ List of available citations
#endif
      stateNotes           :: NoteTable,     -- ^ List of notes
      stateTabStop         :: Int,           -- ^ Tab stop
      stateStandalone      :: Bool,          -- ^ Parse bibliographic info?
      stateTitle           :: [Inline],      -- ^ Title of document
      stateAuthors         :: [[Inline]],    -- ^ Authors of document
      stateDate            :: [Inline],      -- ^ Date of document
      stateStrict          :: Bool,          -- ^ Use strict markdown syntax?
      stateSmart           :: Bool,          -- ^ Use smart typography?
      stateLiterateHaskell :: Bool,          -- ^ Treat input as literate haskell
      stateColumns         :: Int,           -- ^ Number of columns in terminal
      stateHeaderTable     :: [HeaderType],  -- ^ Ordered list of header types used
      stateIndentedCodeClasses :: [String]   -- ^ Classes to use for indented code blocks
    }
    deriving Show

defaultParserState :: ParserState
defaultParserState = 
    ParserState { stateParseRaw        = False,
                  stateParserContext   = NullState,
                  stateQuoteContext    = NoQuote,
                  stateSanitizeHTML    = False,
                  stateKeys            = [],
#ifdef _CITEPROC
                  stateCitations       = [],
#endif
                  stateNotes           = [],
                  stateTabStop         = 4,
                  stateStandalone      = False,
                  stateTitle           = [],
                  stateAuthors         = [],
                  stateDate            = [],
                  stateStrict          = False,
                  stateSmart           = False,
                  stateLiterateHaskell = False,
                  stateColumns         = 80,
                  stateHeaderTable     = [],
                  stateIndentedCodeClasses = [] }

data HeaderType 
    = SingleHeader Char  -- ^ Single line of characters underneath
    | DoubleHeader Char  -- ^ Lines of characters above and below
    deriving (Eq, Show)

data ParserContext 
    = ListItemState   -- ^ Used when running parser on list item contents
    | NullState       -- ^ Default state
    deriving (Eq, Show)

data QuoteContext
    = InSingleQuote   -- ^ Used when parsing inside single quotes
    | InDoubleQuote   -- ^ Used when parsing inside double quotes
    | NoQuote         -- ^ Used when not parsing inside quotes
    deriving (Eq, Show)

type NoteTable = [(String, String)]

type KeyTable = [([Inline], Target)]

-- | Look up key in key table and return target object.
lookupKeySrc :: KeyTable  -- ^ Key table
             -> [Inline]  -- ^ Key
             -> Maybe Target
lookupKeySrc table key = case find (refsMatch key . fst) table of
                           Nothing       -> Nothing
                           Just (_, src) -> Just src

-- | Returns @True@ if keys match (case insensitive).
refsMatch :: [Inline] -> [Inline] -> Bool
refsMatch ((Str x):restx) ((Str y):resty) = 
    ((map toLower x) == (map toLower y)) && refsMatch restx resty
refsMatch ((Emph x):restx) ((Emph y):resty) = 
    refsMatch x y && refsMatch restx resty
refsMatch ((Strong x):restx) ((Strong y):resty) = 
    refsMatch x y && refsMatch restx resty
refsMatch ((Strikeout x):restx) ((Strikeout y):resty) = 
    refsMatch x y && refsMatch restx resty
refsMatch ((Superscript x):restx) ((Superscript y):resty) = 
    refsMatch x y && refsMatch restx resty
refsMatch ((Subscript x):restx) ((Subscript y):resty) = 
    refsMatch x y && refsMatch restx resty
refsMatch ((SmallCaps x):restx) ((SmallCaps y):resty) = 
    refsMatch x y && refsMatch restx resty
refsMatch ((Quoted t x):restx) ((Quoted u y):resty) = 
    t == u && refsMatch x y && refsMatch restx resty
refsMatch ((Code x):restx) ((Code y):resty) = 
    ((map toLower x) == (map toLower y)) && refsMatch restx resty
refsMatch ((Math t x):restx) ((Math u y):resty) = 
    ((map toLower x) == (map toLower y)) && t == u && refsMatch restx resty
refsMatch ((TeX x):restx) ((TeX y):resty) = 
    ((map toLower x) == (map toLower y)) && refsMatch restx resty
refsMatch ((HtmlInline x):restx) ((HtmlInline y):resty) = 
    ((map toLower x) == (map toLower y)) && refsMatch restx resty
refsMatch (x:restx) (y:resty) = (x == y) && refsMatch restx resty
refsMatch [] x = null x
refsMatch x [] = null x

--
-- Prettyprinting
--

-- | A version of hang that works like the version in pretty-1.0.0.0
hang' :: Doc -> Int -> Doc -> Doc
hang' d1 n d2 = d1 $$ (nest n d2)

-- | Indent string as a block.
indentBy :: Int    -- ^ Number of spaces to indent the block 
         -> Int    -- ^ Number of spaces (rel to block) to indent first line
         -> String -- ^ Contents of block to indent
         -> String
indentBy _ _ [] = ""
indentBy num first str = 
  let (firstLine:restLines) = lines str 
      firstLineIndent = num + first
  in  (replicate firstLineIndent ' ') ++ firstLine ++ "\n" ++ 
      (intercalate "\n" $ map ((replicate num ' ') ++ ) restLines)

-- | Prettyprint list of Pandoc blocks elements.
prettyBlockList :: Int       -- ^ Number of spaces to indent list of blocks
                -> [Block]   -- ^ List of blocks
                -> String
prettyBlockList indent [] = indentBy indent 0 "[]"
prettyBlockList indent blocks = indentBy indent (-2) $ "[ " ++ 
  (intercalate "\n, " (map prettyBlock blocks)) ++ " ]"

-- | Prettyprint Pandoc block element.
prettyBlock :: Block -> String
prettyBlock (BlockQuote blocks) = "BlockQuote\n  " ++ 
                                  (prettyBlockList 2 blocks) 
prettyBlock (OrderedList attribs blockLists) = 
  "OrderedList " ++ show attribs ++ "\n" ++ indentBy 2 0 ("[ " ++ 
  (intercalate ", " $ map (\blocks -> prettyBlockList 2 blocks)
  blockLists)) ++ " ]"
prettyBlock (BulletList blockLists) = "BulletList\n" ++ 
  indentBy 2 0 ("[ " ++ (intercalate ", "
  (map (\blocks -> prettyBlockList 2 blocks) blockLists))) ++ " ]" 
prettyBlock (DefinitionList items) = "DefinitionList\n" ++ 
  indentBy 2 0 ("[ " ++ (intercalate "\n, "
  (map (\(term, defs) -> "(" ++ show term ++ ",\n" ++ 
  indentBy 3 0 ("[ " ++ (intercalate ", "
  (map (\blocks -> prettyBlockList 2 blocks) defs)) ++ "]") ++
   ")") items))) ++ " ]" 
prettyBlock (Table caption aligns widths header rows) = 
  "Table " ++ show caption ++ " " ++ show aligns ++ " " ++ 
  show widths ++ "\n" ++ prettyRow header ++ " [\n" ++  
  (intercalate ",\n" (map prettyRow rows)) ++ " ]"
  where prettyRow cols = indentBy 2 0 ("[ " ++ (intercalate ", "
                         (map (\blocks -> prettyBlockList 2 blocks) 
                         cols))) ++ " ]"
prettyBlock block = show block

-- | Prettyprint Pandoc document.
prettyPandoc :: Pandoc -> String
prettyPandoc (Pandoc meta blocks) = "Pandoc " ++ "(" ++ show meta ++ 
  ")\n" ++ (prettyBlockList 0 blocks) ++ "\n"

--
-- Pandoc block and inline list processing
--

-- | Generate infinite lazy list of markers for an ordered list,
-- depending on list attributes.
orderedListMarkers :: (Int, ListNumberStyle, ListNumberDelim) -> [String]
orderedListMarkers (start, numstyle, numdelim) = 
  let singleton c = [c]
      nums = case numstyle of
                     DefaultStyle -> map show [start..]
                     Decimal      -> map show [start..]
                     UpperAlpha   -> drop (start - 1) $ cycle $ 
                                     map singleton ['A'..'Z']
                     LowerAlpha   -> drop (start - 1) $ cycle $
                                     map singleton ['a'..'z']
                     UpperRoman   -> map toRomanNumeral [start..]
                     LowerRoman   -> map (map toLower . toRomanNumeral) [start..]
      inDelim str = case numdelim of
                            DefaultDelim -> str ++ "."
                            Period       -> str ++ "."
                            OneParen     -> str ++ ")"
                            TwoParens    -> "(" ++ str ++ ")"
  in  map inDelim nums

-- | Normalize a list of inline elements: remove leading and trailing
-- @Space@ elements, collapse double @Space@s into singles, and
-- remove empty Str elements.
normalizeSpaces :: [Inline] -> [Inline]
normalizeSpaces [] = []
normalizeSpaces list = 
    let removeDoubles [] = []
        removeDoubles (Space:Space:rest) = removeDoubles (Space:rest)
        removeDoubles (Space:(Str ""):Space:rest) = removeDoubles (Space:rest)
        removeDoubles ((Str ""):rest) = removeDoubles rest 
        removeDoubles (x:rest) = x:(removeDoubles rest)
        removeLeading (Space:xs) = removeLeading xs
        removeLeading x = x
        removeTrailing [] = []
        removeTrailing lst = if (last lst == Space)
                                then init lst
                                else lst
    in  removeLeading $ removeTrailing $ removeDoubles list

-- | Change final list item from @Para@ to @Plain@ if the list contains
-- no other @Para@ blocks.
compactify :: [[Block]]  -- ^ List of list items (each a list of blocks)
           -> [[Block]]
compactify [] = []
compactify items =
  case (init items, last items) of
       (_,[])          -> items
       (others, final) ->
            case last final of
                 Para a -> case (filter isPara $ concat items) of
                                -- if this is only Para, change to Plain
                                [_] -> others ++ [init final ++ [Plain a]]
                                _   -> items
                 _      -> items

isPara :: Block -> Bool
isPara (Para _) = True
isPara _        = False

-- | Data structure for defining hierarchical Pandoc documents
data Element = Blk Block 
             | Sec Int [Int] String [Inline] [Element]
             --    lvl  num ident  label    contents
             deriving (Eq, Read, Show, Typeable, Data)

-- | Convert Pandoc inline list to plain text identifier.  HTML
-- identifiers must start with a letter, and may contain only
-- letters, digits, and the characters _-:.
inlineListToIdentifier :: [Inline] -> String
inlineListToIdentifier =
  dropWhile (not . isAlpha) . intercalate "-" . words . map toLower .
  filter (\c -> isLetter c || isDigit c || c `elem` "_-:. ") .
  concatMap extractText
    where extractText x = case x of
              Str s           -> s
              Emph lst        -> concatMap extractText lst
              Strikeout lst   -> concatMap extractText lst
              Superscript lst -> concatMap extractText lst
              SmallCaps   lst -> concatMap extractText lst
              Subscript lst   -> concatMap extractText lst
              Strong lst      -> concatMap extractText lst
              Quoted _ lst    -> concatMap extractText lst
              Cite   _ lst    -> concatMap extractText lst
              Code s          -> s
              Space           -> " "
              EmDash          -> "---"
              EnDash          -> "--"
              Apostrophe      -> ""
              Ellipses        -> "..."
              LineBreak       -> " "
              Math _ s        -> s
              TeX _           -> ""
              HtmlInline _    -> ""
              Link lst _      -> concatMap extractText lst
              Image lst _     -> concatMap extractText lst
              Note _          -> ""

-- | Convert list of Pandoc blocks into (hierarchical) list of Elements
hierarchicalize :: [Block] -> [Element]
hierarchicalize blocks = S.evalState (hierarchicalizeWithIds blocks) ([],[])

hierarchicalizeWithIds :: [Block] -> S.State ([Int],[String]) [Element]
hierarchicalizeWithIds [] = return []
hierarchicalizeWithIds ((Header level title'):xs) = do
  (lastnum, usedIdents) <- S.get
  let ident = uniqueIdent title' usedIdents
  let lastnum' = take level lastnum
  let newnum = if length lastnum' >= level
                  then init lastnum' ++ [last lastnum' + 1] 
                  else lastnum ++ replicate (level - length lastnum - 1) 0 ++ [1]
  S.put (newnum, (ident : usedIdents))
  let (sectionContents, rest) = break (headerLtEq level) xs
  sectionContents' <- hierarchicalizeWithIds sectionContents
  rest' <- hierarchicalizeWithIds rest
  return $ Sec level newnum ident title' sectionContents' : rest'
hierarchicalizeWithIds (x:rest) = do
  rest' <- hierarchicalizeWithIds rest
  return $ (Blk x) : rest'

headerLtEq :: Int -> Block -> Bool
headerLtEq level (Header l _) = l <= level
headerLtEq _ _ = False

-- | Generate a unique identifier from a list of inlines.
-- Second argument is a list of already used identifiers.
uniqueIdent :: [Inline] -> [String] -> String
uniqueIdent title' usedIdents =
  let baseIdent = case inlineListToIdentifier title' of
                        ""   -> "section"
                        x    -> x
      numIdent n = baseIdent ++ "-" ++ show n
  in  if baseIdent `elem` usedIdents
        then case find (\x -> numIdent x `notElem` usedIdents) ([1..60000] :: [Int]) of
                  Just x  -> numIdent x
                  Nothing -> baseIdent   -- if we have more than 60,000, allow repeats
        else baseIdent

-- | True if block is a Header block.
isHeaderBlock :: Block -> Bool
isHeaderBlock (Header _ _) = True
isHeaderBlock _ = False

--
-- Writer options
--

data HTMLMathMethod = PlainMath 
                    | LaTeXMathML (Maybe String)  -- url of LaTeXMathML.js
                    | JsMath (Maybe String)       -- url of jsMath load script
                    | GladTeX
                    | MimeTeX String              -- url of mimetex.cgi 
                    | MathML (Maybe String)       -- url of MathMLinHTML.js
                    deriving (Show, Read, Eq)

-- | Methods for obfuscating email addresses in HTML.
data ObfuscationMethod = NoObfuscation
                       | ReferenceObfuscation
                       | JavascriptObfuscation
                       deriving (Show, Read, Eq)

-- | Options for writers
data WriterOptions = WriterOptions
  { writerStandalone       :: Bool   -- ^ Include header and footer
  , writerTemplate         :: String -- ^ Template to use in standalone mode
  , writerVariables        :: [(String, String)] -- ^ Variables to set in template
  , writerIncludeBefore    :: String -- ^ Text to include before the body
  , writerIncludeAfter     :: String -- ^ Text to include after the body
  , writerTabStop          :: Int    -- ^ Tabstop for conversion btw spaces and tabs
  , writerTableOfContents  :: Bool   -- ^ Include table of contents
  , writerS5               :: Bool   -- ^ We're writing S5 
  , writerXeTeX            :: Bool   -- ^ Create latex suitable for use by xetex
  , writerHTMLMathMethod   :: HTMLMathMethod  -- ^ How to print math in HTML
  , writerIgnoreNotes      :: Bool   -- ^ Ignore footnotes (used in making toc)
  , writerIncremental      :: Bool   -- ^ Incremental S5 lists
  , writerNumberSections   :: Bool   -- ^ Number sections in LaTeX
  , writerStrictMarkdown   :: Bool   -- ^ Use strict markdown syntax
  , writerReferenceLinks   :: Bool   -- ^ Use reference links in writing markdown, rst
  , writerWrapText         :: Bool   -- ^ Wrap text to line length
  , writerLiterateHaskell  :: Bool   -- ^ Write as literate haskell
  , writerEmailObfuscation :: ObfuscationMethod -- ^ How to obfuscate emails
  , writerIdentifierPrefix :: String -- ^ Prefix for section & note ids in HTML
  } deriving Show

-- | Default writer options.
defaultWriterOptions :: WriterOptions
defaultWriterOptions = 
  WriterOptions { writerStandalone       = False
                , writerTemplate         = ""
                , writerVariables        = []
                , writerIncludeBefore    = ""
                , writerIncludeAfter     = ""
                , writerTabStop          = 4
                , writerTableOfContents  = False
                , writerS5               = False
                , writerXeTeX            = False
                , writerHTMLMathMethod   = PlainMath
                , writerIgnoreNotes      = False
                , writerIncremental      = False
                , writerNumberSections   = False
                , writerStrictMarkdown   = False
                , writerReferenceLinks   = False
                , writerWrapText         = True
                , writerLiterateHaskell  = False
                , writerEmailObfuscation = JavascriptObfuscation
                , writerIdentifierPrefix = ""
                }

--
-- File handling
--

-- | Perform an IO action in a directory, returning to starting directory.
inDirectory :: FilePath -> IO a -> IO a
inDirectory path action = do
  oldDir <- getCurrentDirectory
  setCurrentDirectory path
  result <- action
  setCurrentDirectory oldDir
  return result

-- | Read file from specified user data directory or, if not found there, from
-- Cabal data directory.
readDataFile :: Maybe FilePath -> FilePath -> IO String
readDataFile userDir fname =
  case userDir of
       Nothing  -> getDataFileName fname >>= UTF8.readFile
       Just u   -> catch (UTF8.readFile $ u </> fname)
                   (\_ -> getDataFileName fname >>= UTF8.readFile)