aboutsummaryrefslogtreecommitdiff
path: root/src/Text/Pandoc/Parsing.hs
blob: ce2523d12c6aec0eeddc8dcd5550c1ef89b149c8 (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
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
{-# LANGUAGE
  FlexibleContexts
, GeneralizedNewtypeDeriving
, TypeSynonymInstances
, MultiParamTypeClasses
, FlexibleInstances
, IncoherentInstances #-}

{-
Copyright (C) 2006-2017 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.Parsing
   Copyright   : Copyright (C) 2006-2017 John MacFarlane
   License     : GNU GPL, version 2 or above

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

A utility library with parsers used in pandoc readers.
-}
module Text.Pandoc.Parsing ( anyLine,
                             anyLineNewline,
                             many1Till,
                             notFollowedBy',
                             oneOfStrings,
                             oneOfStringsCI,
                             spaceChar,
                             nonspaceChar,
                             skipSpaces,
                             blankline,
                             blanklines,
                             enclosed,
                             stringAnyCase,
                             parseFromString,
                             lineClump,
                             charsInBalanced,
                             romanNumeral,
                             emailAddress,
                             uri,
                             mathInline,
                             mathDisplay,
                             withHorizDisplacement,
                             withRaw,
                             escaped,
                             characterReference,
                             anyOrderedListMarker,
                             orderedListMarker,
                             charRef,
                             lineBlockLines,
                             tableWith,
                             widthsFromIndices,
                             gridTableWith,
                             gridTableWith',
                             readWith,
                             readWithM,
                             testStringWith,
                             guardEnabled,
                             guardDisabled,
                             updateLastStrPos,
                             notAfterString,
                             logMessage,
                             reportLogMessages,
                             ParserState (..),
                             HasReaderOptions (..),
                             HasHeaderMap (..),
                             HasIdentifierList (..),
                             HasMacros (..),
                             HasLogMessages (..),
                             HasLastStrPosition (..),
                             HasIncludeFiles (..),
                             defaultParserState,
                             HeaderType (..),
                             ParserContext (..),
                             QuoteContext (..),
                             HasQuoteContext (..),
                             NoteTable,
                             NoteTable',
                             KeyTable,
                             SubstTable,
                             Key (..),
                             toKey,
                             registerHeader,
                             smartPunctuation,
                             singleQuoteStart,
                             singleQuoteEnd,
                             doubleQuoteStart,
                             doubleQuoteEnd,
                             ellipses,
                             apostrophe,
                             dash,
                             nested,
                             citeKey,
                             macro,
                             applyMacros',
                             Parser,
                             ParserT,
                             F,
                             Future(..),
                             runF,
                             askF,
                             asksF,
                             returnF,
                             trimInlinesF,
                             token,
                             (<+?>),
                             extractIdClass,
                             insertIncludedFile,
                             insertIncludedFileF,
                             -- * Re-exports from Text.Pandoc.Parsec
                             Stream,
                             runParser,
                             runParserT,
                             parse,
                             anyToken,
                             getInput,
                             setInput,
                             unexpected,
                             char,
                             letter,
                             digit,
                             alphaNum,
                             skipMany,
                             skipMany1,
                             spaces,
                             space,
                             anyChar,
                             satisfy,
                             newline,
                             string,
                             count,
                             eof,
                             noneOf,
                             oneOf,
                             lookAhead,
                             notFollowedBy,
                             many,
                             many1,
                             manyTill,
                             (<|>),
                             (<?>),
                             choice,
                             try,
                             sepBy,
                             sepBy1,
                             sepEndBy,
                             sepEndBy1,
                             endBy,
                             endBy1,
                             option,
                             optional,
                             optionMaybe,
                             getState,
                             setState,
                             updateState,
                             SourcePos,
                             getPosition,
                             setPosition,
                             sourceColumn,
                             sourceLine,
                             setSourceColumn,
                             setSourceLine,
                             newPos
                             )
where

import Text.Pandoc.Definition
import Text.Pandoc.Options
import Text.Pandoc.Builder (Blocks, Inlines, rawBlock, HasMeta(..), trimInlines)
import qualified Text.Pandoc.Builder as B
import Text.Pandoc.XML (fromEntities)
import qualified Text.Pandoc.UTF8 as UTF8 (putStrLn)
import Text.Parsec hiding (token)
import Text.Parsec.Pos (newPos)
import Data.Char ( toLower, toUpper, ord, chr, isAscii, isAlphaNum,
                   isHexDigit, isSpace, isPunctuation )
import Data.List ( intercalate, transpose, isSuffixOf )
import Text.Pandoc.Shared
import qualified Data.Map as M
import Text.TeXMath.Readers.TeX.Macros (applyMacros, Macro, pMacroDefinition)
import Text.HTML.TagSoup.Entity ( lookupEntity )
import Text.Pandoc.Asciify (toAsciiChar)
import Data.Monoid ((<>))
import Text.Pandoc.Class (PandocMonad, readFileFromDirs, report)
import Text.Pandoc.Logging
import Data.Default
import qualified Data.Set as Set
import Control.Monad.Reader
import Control.Monad.Identity
import Data.Maybe (catMaybes)

import Text.Pandoc.Error
import Control.Monad.Except

type Parser t s = Parsec t s

type ParserT = ParsecT

-- | Reader monad wrapping the parser state. This is used to possibly delay
-- evaluation until all relevant information has been parsed and made available
-- in the parser state.
newtype Future s a = Future { runDelayed :: Reader s a }
  deriving (Monad, Applicative, Functor)

type F = Future ParserState

runF :: Future s a -> s -> a
runF = runReader . runDelayed

askF :: Future s s
askF = Future ask

asksF :: (s -> a) -> Future s a
asksF f = Future $ asks f

returnF :: Monad m => a -> m (Future s a)
returnF = return . return

trimInlinesF :: Future s Inlines -> Future s Inlines
trimInlinesF = liftM trimInlines

instance Monoid a => Monoid (Future s a) where
  mempty = return mempty
  mappend = liftM2 mappend
  mconcat = liftM mconcat . sequence

-- | Parse any line of text
anyLine :: Stream [Char] m Char => ParserT [Char] st m [Char]
anyLine = do
  -- This is much faster than:
  -- manyTill anyChar newline
  inp <- getInput
  pos <- getPosition
  case break (=='\n') inp of
       (this, '\n':rest) -> do
         -- needed to persuade parsec that this won't match an empty string:
         anyChar
         setInput rest
         setPosition $ incSourceLine (setSourceColumn pos 1) 1
         return this
       _ -> mzero

-- | Parse any line, include the final newline in the output
anyLineNewline :: Stream [Char] m Char => ParserT [Char] st m [Char]
anyLineNewline = (++ "\n") <$> anyLine

-- | Like @manyTill@, but reads at least one item.
many1Till :: Stream s m t
          => ParserT s st m a
          -> ParserT s st m end
          -> ParserT s st m [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, Stream s m a) => ParserT s st m b -> ParserT s st m ()
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.)

oneOfStrings' :: Stream s m Char => (Char -> Char -> Bool) -> [String] -> ParserT s st m String
oneOfStrings' _ []   = fail "no strings"
oneOfStrings' matches strs = try $ do
  c <- anyChar
  let strs' = [xs | (x:xs) <- strs, x `matches` c]
  case strs' of
       []  -> fail "not found"
       _   -> (c:) <$> oneOfStrings' matches strs'
               <|> if "" `elem` strs'
                      then return [c]
                      else fail "not found"

-- | Parses one of a list of strings.  If the list contains
-- two strings one of which is a prefix of the other, the longer
-- string will be matched if possible.
oneOfStrings :: Stream s m Char => [String] -> ParserT s st m String
oneOfStrings = oneOfStrings' (==)

-- | Parses one of a list of strings (tried in order), case insensitive.
oneOfStringsCI :: Stream s m Char => [String] -> ParserT s st m String
oneOfStringsCI = oneOfStrings' ciMatch
  where ciMatch x y = toLower' x == toLower' y
        -- this optimizes toLower by checking common ASCII case
        -- first, before calling the expensive unicode-aware
        -- function:
        toLower' c | c >= 'A' && c <= 'Z' = chr (ord c + 32)
                   | isAscii c = c
                   | otherwise = toLower c

-- | Parses a space or tab.
spaceChar :: Stream s m Char => ParserT s st m Char
spaceChar = satisfy $ \c -> c == ' ' || c == '\t'

-- | Parses a nonspace, nonnewline character.
nonspaceChar :: Stream s m Char => ParserT s st m Char
nonspaceChar = satisfy $ flip notElem ['\t', '\n', ' ', '\r']

-- | Skips zero or more spaces or tabs.
skipSpaces :: Stream s m Char => ParserT s st m ()
skipSpaces = skipMany spaceChar

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

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

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

-- | Parse string, case insensitive.
stringAnyCase :: Stream s m Char => [Char] -> ParserT s st m 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 :: Monad m => ParserT String st m a -> String -> ParserT String st m a
parseFromString parser str = do
  oldPos <- getPosition
  oldInput <- getInput
  setInput str
  result <- parser
  spaces
  eof
  setInput oldInput
  setPosition oldPos
  return result

-- | Parse raw line block up to and including blank lines.
lineClump :: Stream [Char] m Char => ParserT [Char] st m 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 '(' ')' anyChar@ will parse "(hello (there))"
-- and return "hello (there)".
charsInBalanced :: Stream s m Char => Char -> Char -> ParserT s st m Char
                -> ParserT s st m String
charsInBalanced open close parser = try $ do
  char open
  let isDelim c = c == open || c == close
  raw <- many $  many1 (notFollowedBy (satisfy isDelim) >> parser)
             <|> (do res <- charsInBalanced open close parser
                     return $ [open] ++ res ++ [close])
  char close
  return $ concat raw

-- old charsInBalanced would be:
-- charsInBalanced open close (noneOf "\n" <|> char '\n' >> notFollowedBy blankline)
-- old charsInBalanced' would be:
-- charsInBalanced open close anyChar

-- 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 :: Stream s m Char => Bool                  -- ^ Uppercase if true
             -> ParserT s st m 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

-- | Parses an email address; returns original and corresponding
-- escaped mailto: URI.
emailAddress :: Stream s m Char => ParserT s st m (String, String)
emailAddress = try $ toResult <$> mailbox <*> (char '@' *> domain)
 where toResult mbox dom = let full = fromEntities $ mbox ++ '@':dom
                           in  (full, escapeURI $ "mailto:" ++ full)
       mailbox           = intercalate "." <$> (emailWord `sepby1` dot)
       domain            = intercalate "." <$> (subdomain `sepby1` dot)
       dot               = char '.'
       subdomain         = many1 $ alphaNum <|> innerPunct
       -- this excludes some valid email addresses, since an
       -- email could contain e.g. '__', but gives better results
       -- for our purposes, when combined with markdown parsing:
       innerPunct        = try (satisfy (\c -> isEmailPunct c || c == '@')
                                 <* notFollowedBy space
                                 <* notFollowedBy (satisfy isPunctuation))
       -- technically an email address could begin with a symbol,
       -- but allowing this creates too many problems.
       -- See e.g. https://github.com/jgm/pandoc/issues/2940
       emailWord         = do x <- satisfy isAlphaNum
                              xs <- many (satisfy isEmailChar)
                              return (x:xs)
       isEmailChar c     = isAlphaNum c || isEmailPunct c
       isEmailPunct c    = c `elem` "!\"#$%&'*+-/=?^_{|}~;"
       -- note: sepBy1 from parsec consumes input when sep
       -- succeeds and p fails, so we use this variant here.
       sepby1 p sep      = (:) <$> p <*> (many (try $ sep >> p))


-- Schemes from http://www.iana.org/assignments/uri-schemes.html plus
-- the unofficial schemes coap, doi, javascript, isbn, pmid
schemes :: [String]
schemes = ["coap","doi","javascript","aaa","aaas","about","acap","cap","cid",
           "crid","data","dav","dict","dns","file","ftp","geo","go","gopher",
           "h323","http","https","iax","icap","im","imap","info","ipp","iris",
           "iris.beep","iris.xpc","iris.xpcs","iris.lwz","ldap","mailto","mid",
           "msrp","msrps","mtqp","mupdate","news","nfs","ni","nih","nntp",
           "opaquelocktoken","pop","pres","rtsp","service","session","shttp","sieve",
           "sip","sips","sms","snmp","soap.beep","soap.beeps","tag","tel","telnet",
           "tftp","thismessage","tn3270","tip","tv","urn","vemmi","ws","wss","xcon",
           "xcon-userid","xmlrpc.beep","xmlrpc.beeps","xmpp","z39.50r","z39.50s",
           "adiumxtra","afp","afs","aim","apt","attachment","aw","beshare","bitcoin",
           "bolo","callto","chrome","chrome-extension","com-eventbrite-attendee",
           "content", "cvs","dlna-playsingle","dlna-playcontainer","dtn","dvb",
           "ed2k","facetime","feed","finger","fish","gg","git","gizmoproject",
           "gtalk","hcp","icon","ipn","irc","irc6","ircs","itms","jar","jms",
           "keyparc","lastfm","ldaps","magnet","maps","market","message","mms",
           "ms-help","msnim","mumble","mvn","notes","oid","palm","paparazzi",
           "platform","proxy","psyc","query","res","resource","rmi","rsync",
           "rtmp","secondlife","sftp","sgn","skype","smb","soldat","spotify",
           "ssh","steam","svn","teamspeak","things","udp","unreal","ut2004",
           "ventrilo","view-source","webcal","wtai","wyciwyg","xfire","xri",
           "ymsgr", "isbn", "pmid"]

uriScheme :: Stream s m Char => ParserT s st m String
uriScheme = oneOfStringsCI schemes

-- | Parses a URI. Returns pair of original and URI-escaped version.
uri :: Stream [Char] m Char => ParserT [Char] st m (String, String)
uri = try $ do
  scheme <- uriScheme
  char ':'
  -- Avoid parsing e.g. "**Notes:**" as a raw URI:
  notFollowedBy (oneOf "*_]")
  -- We allow sentence punctuation except at the end, since
  -- we don't want the trailing '.' in 'http://google.com.' We want to allow
  -- http://en.wikipedia.org/wiki/State_of_emergency_(disambiguation)
  -- as a URL, while NOT picking up the closing paren in
  -- (http://wikipedia.org). So we include balanced parens in the URL.
  let isWordChar c = isAlphaNum c || c `elem` "#$%*+/@\\_-"
  let wordChar = satisfy isWordChar
  let percentEscaped = try $ char '%' >> skipMany1 (satisfy isHexDigit)
  let entity = () <$ characterReference
  let punct = skipMany1 (char ',')
          <|> () <$ (satisfy (\c -> not (isSpace c) && c /= '<' && c /= '>'))
  let uriChunk =  skipMany1 wordChar
              <|> percentEscaped
              <|> entity
              <|> (try $ punct >>
                    lookAhead (void (satisfy isWordChar) <|> percentEscaped))
  str <- snd <$> withRaw (skipMany1 ( () <$
                                         (enclosed (char '(') (char ')') uriChunk
                                         <|> enclosed (char '{') (char '}') uriChunk
                                         <|> enclosed (char '[') (char ']') uriChunk)
                                         <|> uriChunk))
  str' <- option str $ char '/' >> return (str ++ "/")
  let uri' = scheme ++ ":" ++ fromEntities str'
  return (uri', escapeURI uri')

mathInlineWith :: Stream s m Char  => String -> String -> ParserT s st m String
mathInlineWith op cl = try $ do
  string op
  notFollowedBy space
  words' <- many1Till (count 1 (noneOf " \t\n\\")
                   <|> (char '\\' >>
                           -- This next clause is needed because \text{..} can
                           -- contain $, \(\), etc.
                           (try (string "text" >>
                                 (("\\text" ++) <$> inBalancedBraces 0 ""))
                            <|>  (\c -> ['\\',c]) <$> anyChar))
                   <|> do (blankline <* notFollowedBy' blankline) <|>
                             (oneOf " \t" <* skipMany (oneOf " \t"))
                          notFollowedBy (char '$')
                          return " "
                    ) (try $ string cl)
  notFollowedBy digit  -- to prevent capture of $5
  return $ concat words'
 where
  inBalancedBraces :: Stream s m Char => Int -> String -> ParserT s st m String
  inBalancedBraces 0 "" = do
    c <- anyChar
    if c == '{'
       then inBalancedBraces 1 "{"
       else mzero
  inBalancedBraces 0 s = return $ reverse s
  inBalancedBraces numOpen ('\\':xs) = do
    c <- anyChar
    inBalancedBraces numOpen (c:'\\':xs)
  inBalancedBraces numOpen xs = do
    c <- anyChar
    case c of
         '}' -> inBalancedBraces (numOpen - 1) (c:xs)
         '{' -> inBalancedBraces (numOpen + 1) (c:xs)
         _   -> inBalancedBraces numOpen (c:xs)

mathDisplayWith :: Stream s m Char => String -> String -> ParserT s st m String
mathDisplayWith op cl = try $ do
  string op
  many1Till (noneOf "\n" <|> (newline <* notFollowedBy' blankline)) (try $ string cl)

mathDisplay :: (HasReaderOptions st, Stream s m Char)
            => ParserT s st m String
mathDisplay =
      (guardEnabled Ext_tex_math_dollars >> mathDisplayWith "$$" "$$")
  <|> (guardEnabled Ext_tex_math_single_backslash >>
       mathDisplayWith "\\[" "\\]")
  <|> (guardEnabled Ext_tex_math_double_backslash >>
       mathDisplayWith "\\\\[" "\\\\]")

mathInline :: (HasReaderOptions st , Stream s m Char)
           => ParserT s st m String
mathInline =
      (guardEnabled Ext_tex_math_dollars >> mathInlineWith "$" "$")
  <|> (guardEnabled Ext_tex_math_single_backslash >>
       mathInlineWith "\\(" "\\)")
  <|> (guardEnabled Ext_tex_math_double_backslash >>
       mathInlineWith "\\\\(" "\\\\)")

-- | 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 :: Stream s m Char
                      => ParserT s st m a  -- ^ Parser to apply
                      -> ParserT s st m (a, Int) -- ^ (result, displacement)
withHorizDisplacement parser = do
  pos1 <- getPosition
  result <- parser
  pos2 <- getPosition
  return (result, sourceColumn pos2 - sourceColumn pos1)

-- | Applies a parser and returns the raw string that was parsed,
-- along with the value produced by the parser.
withRaw :: Stream [Char] m Char => ParsecT [Char] st m a -> ParsecT [Char] st m (a, [Char])
withRaw parser = do
  pos1 <- getPosition
  inp <- getInput
  result <- parser
  pos2 <- getPosition
  let (l1,c1) = (sourceLine pos1, sourceColumn pos1)
  let (l2,c2) = (sourceLine pos2, sourceColumn pos2)
  let inplines = take ((l2 - l1) + 1) $ lines inp
  let raw = case inplines of
                []   -> ""
                [l]  -> take (c2 - c1) l
                ls   -> unlines (init ls) ++ take (c2 - 1) (last ls)
  return (result, raw)

-- | Parses backslash, then applies character parser.
escaped :: Stream s m Char
        => ParserT s st m Char  -- ^ Parser for character to escape
        -> ParserT s st m Char
escaped parser = try $ char '\\' >> parser

-- | Parse character entity.
characterReference :: Stream s m Char => ParserT s st m Char
characterReference = try $ do
  char '&'
  ent <- many1Till nonspaceChar (char ';')
  let ent' = case ent of
                  '#':'X':xs -> '#':'x':xs  -- workaround tagsoup bug
                  '#':_  -> ent
                  _      -> ent ++ ";"
  case lookupEntity ent' of
       Just (c : _)  -> return c
       _             -> fail "entity not found"

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

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

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

-- | Parses a '@' and optional label and
-- returns (DefaultStyle, [next example number]).  The next
-- example number is incremented in parser state, and the label
-- (if present) is added to the label table.
exampleNum :: Stream s m Char
           => ParserT s ParserState m (ListNumberStyle, Int)
exampleNum = do
  char '@'
  lab <- many (alphaNum <|> satisfy (\c -> c == '_' || c == '-'))
  st <- getState
  let num = stateNextExample st
  let newlabels = if null lab
                     then stateExamples st
                     else M.insert lab num $ stateExamples st
  updateState $ \s -> s{ stateNextExample = num + 1
                       , stateExamples    = newlabels }
  return (Example, num)

-- | Parses a '#' returns (DefaultStyle, 1).
defaultNum :: Stream s m Char => ParserT s st m (ListNumberStyle, Int)
defaultNum = do
  char '#'
  return (DefaultStyle, 1)

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

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

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

-- | Parses an ordered list marker and returns list attributes.
anyOrderedListMarker :: Stream s m Char => ParserT s ParserState m ListAttributes
anyOrderedListMarker = choice $
  [delimParser numParser | delimParser <- [inPeriod, inOneParen, inTwoParens],
                           numParser <- [decimal, exampleNum, defaultNum, romanOne,
                           lowerAlpha, lowerRoman, upperAlpha, upperRoman]]

-- | Parses a list number (num) followed by a period, returns list attributes.
inPeriod :: Stream s m Char
         => ParserT s st m (ListNumberStyle, Int)
         -> ParserT s st m 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 :: Stream s m Char
           => ParserT s st m (ListNumberStyle, Int)
           -> ParserT s st m 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 :: Stream s m Char
            => ParserT s st m (ListNumberStyle, Int)
            -> ParserT s st m 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 :: Stream s m Char
                  => ListNumberStyle
                  -> ListNumberDelim
                  -> ParserT s ParserState m Int
orderedListMarker style delim = do
  let num = defaultNum <|>  -- # can continue any kind of list
            case style of
               DefaultStyle -> decimal
               Example      -> exampleNum
               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 :: Stream s m Char => ParserT s st m Inline
charRef = do
  c <- characterReference
  return $ Str [c]

lineBlockLine :: Stream [Char] m Char => ParserT [Char] st m String
lineBlockLine = try $ do
  char '|'
  char ' '
  white <- many (spaceChar >> return '\160')
  notFollowedBy newline
  line <- anyLine
  continuations <- many (try $ char ' ' >> anyLine)
  return $ white ++ unwords (line : continuations)

blankLineBlockLine :: Stream [Char] m Char => ParserT [Char] st m Char
blankLineBlockLine = try (char '|' >> blankline)

-- | Parses an RST-style line block and returns a list of strings.
lineBlockLines :: Stream [Char] m Char => ParserT [Char] st m [String]
lineBlockLines = try $ do
  lines' <- many1 (lineBlockLine <|> ((:[]) <$> blankLineBlockLine))
  skipMany1 $ blankline <|> blankLineBlockLine
  return lines'

-- | Parse a table using 'headerParser', 'rowParser',
-- 'lineParser', and 'footerParser'.
tableWith :: (Stream s m Char, HasReaderOptions st,
              Functor mf, Applicative mf, Monad mf)
          => ParserT s st m (mf [Blocks], [Alignment], [Int])
          -> ([Int] -> ParserT s st m (mf [Blocks]))
          -> ParserT s st m sep
          -> ParserT s st m end
          -> ParserT s st m (mf Blocks)
tableWith headerParser rowParser lineParser footerParser = try $ do
  (aligns, widths, heads, rows) <- tableWith' headerParser rowParser
                                                lineParser footerParser
  return $ B.table mempty (zip aligns widths) <$> heads <*> rows

type TableComponents mf = ([Alignment], [Double], mf [Blocks], mf [[Blocks]])

tableWith' :: (Stream s m Char, HasReaderOptions st,
               Functor mf, Applicative mf, Monad mf)
           => ParserT s st m (mf [Blocks], [Alignment], [Int])
           -> ([Int] -> ParserT s st m (mf [Blocks]))
           -> ParserT s st m sep
           -> ParserT s st m end
           -> ParserT s st m (TableComponents mf)
tableWith' headerParser rowParser lineParser footerParser = try $ do
    (heads, aligns, indices) <- headerParser
    lines' <- sequence <$> rowParser indices `sepEndBy1` lineParser
    footerParser
    numColumns <- getOption readerColumns
    let widths = if (indices == [])
                    then replicate (length aligns) 0.0
                    else widthsFromIndices numColumns indices
    return $ (aligns, widths, heads, lines')

-- Calculate relative widths of table columns, based on indices
widthsFromIndices :: Int      -- Number of columns on terminal
                  -> [Int]    -- Indices
                  -> [Double] -- Fractional relative sizes of columns
widthsFromIndices _ [] = []
widthsFromIndices numColumns' indices =
  let numColumns = max numColumns' (if null indices then 0 else last indices)
      lengths' = zipWith (-) indices (0:indices)
      lengths  = reverse $
                 case reverse lengths' of
                      []       -> []
                      [x]      -> [x]
                      -- compensate for the fact that intercolumn
                      -- spaces are counted in widths of all columns
                      -- but the last...
                      (x:y:zs) -> if x < y && y - x <= 2
                                     then y:y:zs
                                     else x:y:zs
      totLength = sum lengths
      quotient = if totLength > numColumns
                   then fromIntegral totLength
                   else fromIntegral numColumns
      fracs = map (\l -> (fromIntegral l) / quotient) lengths in
  tail fracs

---

-- Parse a grid table:  starts with row of '-' on top, then header
-- (which may be grid), then the rows,
-- which may be grid, separated by blank lines, and
-- ending with a footer (dashed line followed by blank line).
gridTableWith :: (Stream [Char] m Char, HasReaderOptions st,
                  Functor mf, Applicative mf, Monad mf)
              => ParserT [Char] st m (mf Blocks)  -- ^ Block list parser
              -> Bool                             -- ^ Headerless table
              -> ParserT [Char] st m (mf Blocks)
gridTableWith blocks headless =
  tableWith (gridTableHeader headless blocks) (gridTableRow blocks)
            (gridTableSep '-') gridTableFooter

gridTableWith' :: (Stream [Char] m Char, HasReaderOptions st,
                   Functor mf, Applicative mf, Monad mf)
               => ParserT [Char] st m (mf Blocks)  -- ^ Block list parser
               -> Bool                             -- ^ Headerless table
               -> ParserT [Char] st m (TableComponents mf)
gridTableWith' blocks headless =
  tableWith' (gridTableHeader headless blocks) (gridTableRow blocks)
             (gridTableSep '-') gridTableFooter

gridTableSplitLine :: [Int] -> String -> [String]
gridTableSplitLine indices line = map removeFinalBar $ tail $
  splitStringByIndices (init indices) $ trimr line

gridPart :: Stream s m Char => Char -> ParserT s st m ((Int, Int), Alignment)
gridPart ch = do
  leftColon <- option False (True <$ char ':')
  dashes <- many1 (char ch)
  rightColon <- option False (True <$ char ':')
  char '+'
  let lengthDashes = length dashes + (if leftColon then 1 else 0) +
                       (if rightColon then 1 else 0)
  let alignment = case (leftColon, rightColon) of
                       (True, True)   -> AlignCenter
                       (True, False)  -> AlignLeft
                       (False, True)  -> AlignRight
                       (False, False) -> AlignDefault
  return ((lengthDashes, lengthDashes + 1), alignment)

gridDashedLines :: Stream s m Char => Char -> ParserT s st m [((Int, Int), Alignment)]
gridDashedLines ch = try $ char '+' >> many1 (gridPart ch) <* blankline

removeFinalBar :: String -> String
removeFinalBar =
  reverse . dropWhile (`elem` " \t") . dropWhile (=='|') . reverse

-- | Separator between rows of grid table.
gridTableSep :: Stream s m Char => Char -> ParserT s st m Char
gridTableSep ch = try $ gridDashedLines ch >> return '\n'

-- | Parse header for a grid table.
gridTableHeader :: (Stream [Char] m Char, Functor mf, Applicative mf, Monad mf)
                => Bool -- ^ Headerless table
                -> ParserT [Char] st m (mf Blocks)
                -> ParserT [Char] st m (mf [Blocks], [Alignment], [Int])
gridTableHeader headless blocks = try $ do
  optional blanklines
  dashes <- gridDashedLines '-'
  rawContent  <- if headless
                    then return $ repeat ""
                    else many1
                         (notFollowedBy (gridTableSep '=') >> char '|' >>
                           many1Till anyChar newline)
  underDashes <- if headless
                    then return dashes
                    else gridDashedLines '='
  guard $ length dashes == length underDashes
  let lines'   = map (snd . fst) underDashes
  let indices  = scanl (+) 0 lines'
  let aligns   = map snd underDashes
  let rawHeads = if headless
                    then replicate (length underDashes) ""
                    else map (unlines . map trim) $ transpose
                       $ map (gridTableSplitLine indices) rawContent
  heads <- fmap sequence $ mapM (parseFromString blocks . trim) rawHeads
  return (heads, aligns, indices)

gridTableRawLine :: Stream s m Char => [Int] -> ParserT s st m [String]
gridTableRawLine indices = do
  char '|'
  line <- many1Till anyChar newline
  return (gridTableSplitLine indices line)

-- | Parse row of grid table.
gridTableRow :: (Stream [Char]  m Char, Functor mf, Applicative mf, Monad mf)
             => ParserT [Char] st m (mf Blocks)
             -> [Int]
             -> ParserT [Char] st m (mf [Blocks])
gridTableRow blocks indices = do
  colLines <- many1 (gridTableRawLine indices)
  let cols = map ((++ "\n") . unlines . removeOneLeadingSpace) $
               transpose colLines
      compactifyCell bs = case compactify [bs] of
                            []  -> mempty
                            x:_ -> x
  cells <- sequence <$> mapM (parseFromString blocks) cols
  return $ fmap (map compactifyCell) cells

removeOneLeadingSpace :: [String] -> [String]
removeOneLeadingSpace xs =
  if all startsWithSpace xs
     then map (drop 1) xs
     else xs
   where startsWithSpace ""     = True
         startsWithSpace (y:_) = y == ' '

-- | Parse footer for a grid table.
gridTableFooter :: Stream s m Char => ParserT s st m [Char]
gridTableFooter = blanklines

---

-- | Removes the ParsecT layer from the monad transformer stack
readWithM :: (Monad m)
          => ParserT [Char] st m a       -- ^ parser
          -> st                       -- ^ initial state
          -> String                   -- ^ input
          -> m (Either PandocError a)
readWithM parser state input =
    mapLeft (PandocParsecError input) `liftM` runParserT parser state "source" input


-- | Parse a string with a given parser and state
readWith :: Parser [Char] st a
         -> st
         -> String
         -> Either PandocError a
readWith p t inp = runIdentity $ readWithM p t inp

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

-- | Parsing options.
data ParserState = ParserState
    { stateOptions         :: ReaderOptions, -- ^ User options
      stateParserContext   :: ParserContext, -- ^ Inside list?
      stateQuoteContext    :: QuoteContext,  -- ^ Inside quoted environment?
      stateAllowLinks      :: Bool,          -- ^ Allow parsing of links
      stateMaxNestingLevel :: Int,           -- ^ Max # of nested Strong/Emph
      stateLastStrPos      :: Maybe SourcePos, -- ^ Position after last str parsed
      stateKeys            :: KeyTable,      -- ^ List of reference keys
      stateHeaderKeys      :: KeyTable,      -- ^ List of implicit header ref keys
      stateSubstitutions   :: SubstTable,    -- ^ List of substitution references
      stateNotes           :: NoteTable,     -- ^ List of notes (raw bodies)
      stateNotes'          :: NoteTable',    -- ^ List of notes (parsed bodies)
      stateMeta            :: Meta,          -- ^ Document metadata
      stateMeta'           :: F Meta,        -- ^ Document metadata
      stateCitations       :: M.Map String String, -- ^ RST-style citations
      stateHeaderTable     :: [HeaderType],  -- ^ Ordered list of header types used
      stateHeaders         :: M.Map Inlines String, -- ^ List of headers and ids (used for implicit ref links)
      stateIdentifiers     :: Set.Set String, -- ^ Header identifiers used
      stateNextExample     :: Int,           -- ^ Number of next example
      stateExamples        :: M.Map String Int, -- ^ Map from example labels to numbers
      stateMacros          :: [Macro],       -- ^ List of macros defined so far
      stateRstDefaultRole  :: String,        -- ^ Current rST default interpreted text role
      stateRstCustomRoles  :: M.Map String (String, Maybe String, Attr), -- ^ Current rST custom text roles
      -- Triple represents: 1) Base role, 2) Optional format (only for :raw:
      -- roles), 3) Additional classes (rest of Attr is unused)).
      stateCaption         :: Maybe Inlines, -- ^ Caption in current environment
      stateInHtmlBlock     :: Maybe String,  -- ^ Tag type of HTML block being parsed
      stateContainers      :: [String],      -- ^ parent include files
      stateLogMessages     :: [LogMessage],  -- ^ log messages
      stateMarkdownAttribute :: Bool         -- ^ True if in markdown=1 context
    }

instance Default ParserState where
  def = defaultParserState

instance HasMeta ParserState where
  setMeta field val st =
    st{ stateMeta = setMeta field val $ stateMeta st }
  deleteMeta field st =
    st{ stateMeta = deleteMeta field $ stateMeta st }

class HasReaderOptions st where
  extractReaderOptions :: st -> ReaderOptions
  getOption            :: (Stream s m t) => (ReaderOptions -> b) -> ParserT s st m b
  -- default
  getOption  f         = (f . extractReaderOptions) <$> getState

instance HasReaderOptions ParserState where
  extractReaderOptions = stateOptions

class HasQuoteContext st m where
  getQuoteContext :: (Stream s m t) => ParsecT s st m QuoteContext
  withQuoteContext :: QuoteContext -> ParsecT s st m a -> ParsecT s st m a

instance Monad m => HasQuoteContext ParserState m where
  getQuoteContext = stateQuoteContext <$> getState
  withQuoteContext context parser = do
    oldState <- getState
    let oldQuoteContext = stateQuoteContext oldState
    setState oldState { stateQuoteContext = context }
    result <- parser
    newState <- getState
    setState newState { stateQuoteContext = oldQuoteContext }
    return result

class HasHeaderMap st where
  extractHeaderMap  :: st -> M.Map Inlines String
  updateHeaderMap   :: (M.Map Inlines String -> M.Map Inlines String) ->
                       st -> st

instance HasHeaderMap ParserState where
  extractHeaderMap     = stateHeaders
  updateHeaderMap f st = st{ stateHeaders = f $ stateHeaders st }

class HasIdentifierList st where
  extractIdentifierList  :: st -> Set.Set String
  updateIdentifierList   :: (Set.Set String -> Set.Set String) -> st -> st

instance HasIdentifierList ParserState where
  extractIdentifierList     = stateIdentifiers
  updateIdentifierList f st = st{ stateIdentifiers = f $ stateIdentifiers st }

class HasMacros st where
  extractMacros         :: st -> [Macro]
  updateMacros          :: ([Macro] -> [Macro]) -> st -> st

instance HasMacros ParserState where
  extractMacros        = stateMacros
  updateMacros f st    = st{ stateMacros = f $ stateMacros st }

class HasLastStrPosition st where
  setLastStrPos  :: SourcePos -> st -> st
  getLastStrPos  :: st -> Maybe SourcePos

instance HasLastStrPosition ParserState where
  setLastStrPos pos st = st{ stateLastStrPos = Just pos }
  getLastStrPos st     = stateLastStrPos st

class HasLogMessages st where
  addLogMessage :: LogMessage -> st -> st
  getLogMessages :: st -> [LogMessage]

instance HasLogMessages ParserState where
  addLogMessage msg st = st{ stateLogMessages = msg : stateLogMessages st }
  getLogMessages st = reverse $ stateLogMessages st

class HasIncludeFiles st where
  getIncludeFiles :: st -> [String]
  addIncludeFile :: String -> st -> st
  dropLatestIncludeFile :: st -> st

instance HasIncludeFiles ParserState where
  getIncludeFiles = stateContainers
  addIncludeFile f s = s{ stateContainers = f : stateContainers s }
  dropLatestIncludeFile s = s { stateContainers = drop 1 $ stateContainers s }

defaultParserState :: ParserState
defaultParserState =
    ParserState { stateOptions         = def,
                  stateParserContext   = NullState,
                  stateQuoteContext    = NoQuote,
                  stateAllowLinks      = True,
                  stateMaxNestingLevel = 6,
                  stateLastStrPos      = Nothing,
                  stateKeys            = M.empty,
                  stateHeaderKeys      = M.empty,
                  stateSubstitutions   = M.empty,
                  stateNotes           = [],
                  stateNotes'          = [],
                  stateMeta            = nullMeta,
                  stateMeta'           = return nullMeta,
                  stateCitations       = M.empty,
                  stateHeaderTable     = [],
                  stateHeaders         = M.empty,
                  stateIdentifiers     = Set.empty,
                  stateNextExample     = 1,
                  stateExamples        = M.empty,
                  stateMacros          = [],
                  stateRstDefaultRole  = "title-reference",
                  stateRstCustomRoles  = M.empty,
                  stateCaption         = Nothing,
                  stateInHtmlBlock     = Nothing,
                  stateContainers      = [],
                  stateLogMessages     = [],
                  stateMarkdownAttribute = False
                  }

-- | Add a log message.
logMessage :: (Stream s m a, HasLogMessages st)
           => LogMessage -> ParserT s st m ()
logMessage msg = updateState (addLogMessage msg)

-- | Report all the accumulated log messages, according to verbosity level.
reportLogMessages :: (PandocMonad m, HasLogMessages st) => ParserT s st m ()
reportLogMessages = do
  msgs <- getLogMessages <$> getState
  mapM_ report msgs

-- | Succeed only if the extension is enabled.
guardEnabled :: (Stream s m a,  HasReaderOptions st) => Extension -> ParserT s st m ()
guardEnabled ext = getOption readerExtensions >>= guard . extensionEnabled ext

-- | Succeed only if the extension is disabled.
guardDisabled :: (Stream s m a, HasReaderOptions st) => Extension -> ParserT s st m ()
guardDisabled ext = getOption readerExtensions >>= guard . not . extensionEnabled ext

-- | Update the position on which the last string ended.
updateLastStrPos :: (Stream s m a, HasLastStrPosition st) => ParserT s st m ()
updateLastStrPos = getPosition >>= updateState . setLastStrPos

-- | Whether we are right after the end of a string.
notAfterString :: (Stream s m a, HasLastStrPosition st) => ParserT s st m Bool
notAfterString = do
  pos <- getPosition
  st  <- getState
  return $ getLastStrPos st /= Just pos

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 NoteTable' = [(String, F Blocks)]  -- used in markdown reader

newtype Key = Key String deriving (Show, Read, Eq, Ord)

toKey :: String -> Key
toKey = Key . map toLower . unwords . words . unbracket
  where unbracket ('[':xs) | "]" `isSuffixOf` xs = take (length xs - 1) xs
        unbracket xs       = xs

type KeyTable = M.Map Key (Target, Attr)

type SubstTable = M.Map Key Inlines

--  | Add header to the list of headers in state, together
--  with its associated identifier.  If the identifier is null
--  and the auto_identifers extension is set, generate a new
--  unique identifier, and update the list of identifiers
--  in state.  Issue a warning if an explicit identifier
--  is encountered that duplicates an earlier identifier
--  (explict or automatically generated).
registerHeader :: (Stream s m a, HasReaderOptions st,
                    HasHeaderMap st, HasLogMessages st, HasIdentifierList st)
               => Attr -> Inlines -> ParserT s st m Attr
registerHeader (ident,classes,kvs) header' = do
  ids <- extractIdentifierList <$> getState
  exts <- getOption readerExtensions
  let insert' = M.insertWith (\_new old -> old)
  if null ident && Ext_auto_identifiers `extensionEnabled` exts
     then do
       let id' = uniqueIdent (B.toList header') ids
       let id'' = if Ext_ascii_identifiers `extensionEnabled` exts
                     then catMaybes $ map toAsciiChar id'
                     else id'
       updateState $ updateIdentifierList $ Set.insert id'
       updateState $ updateIdentifierList $ Set.insert id''
       updateState $ updateHeaderMap $ insert' header' id'
       return (id'',classes,kvs)
     else do
        unless (null ident) $ do
          when (ident `Set.member` ids) $ do
            pos <- getPosition
            logMessage $ DuplicateIdentifier ident pos
          updateState $ updateIdentifierList $ Set.insert ident
          updateState $ updateHeaderMap $ insert' header' ident
        return (ident,classes,kvs)

smartPunctuation :: (HasReaderOptions st, HasLastStrPosition st, HasQuoteContext st m, Stream s m Char)
                 => ParserT s st m Inlines
                 -> ParserT s st m Inlines
smartPunctuation inlineParser = do
  guardEnabled Ext_smart
  choice [ quoted inlineParser, apostrophe, dash, ellipses ]

apostrophe :: Stream s m Char => ParserT s st m Inlines
apostrophe = (char '\'' <|> char '\8217') >> return (B.str "\x2019")

quoted :: (HasLastStrPosition st, HasQuoteContext st m, Stream s m Char)
       => ParserT s st m Inlines
       -> ParserT s st m Inlines
quoted inlineParser = doubleQuoted inlineParser <|> singleQuoted inlineParser

singleQuoted :: (HasLastStrPosition st, HasQuoteContext st m, Stream s m Char)
             => ParserT s st m Inlines
             -> ParserT s st m Inlines
singleQuoted inlineParser = try $ do
  singleQuoteStart
  withQuoteContext InSingleQuote $ many1Till inlineParser singleQuoteEnd >>=
    return . B.singleQuoted . mconcat

doubleQuoted :: (HasQuoteContext st m, Stream s m Char)
             => ParserT s st m Inlines
             -> ParserT s st m Inlines
doubleQuoted inlineParser = try $ do
  doubleQuoteStart
  withQuoteContext InDoubleQuote $ manyTill inlineParser doubleQuoteEnd >>=
    return . B.doubleQuoted . mconcat

failIfInQuoteContext :: (HasQuoteContext st m, Stream s m t)
                     => QuoteContext
                     -> ParserT s st m ()
failIfInQuoteContext context = do
  context' <- getQuoteContext
  if context' == context
     then fail "already inside quotes"
     else return ()

charOrRef :: Stream s m Char => String -> ParserT s st m Char
charOrRef cs =
  oneOf cs <|> try (do c <- characterReference
                       guard (c `elem` cs)
                       return c)

singleQuoteStart :: (HasLastStrPosition st, HasQuoteContext st m, Stream s m Char)
                 => ParserT s st m ()
singleQuoteStart = do
  failIfInQuoteContext InSingleQuote
  -- single quote start can't be right after str
  guard =<< notAfterString
  () <$ charOrRef "'\8216\145"

singleQuoteEnd :: Stream s m Char
               => ParserT s st m ()
singleQuoteEnd = try $ do
  charOrRef "'\8217\146"
  notFollowedBy alphaNum

doubleQuoteStart :: (HasQuoteContext st m, Stream s m Char)
                 => ParserT s st m ()
doubleQuoteStart = do
  failIfInQuoteContext InDoubleQuote
  try $ do charOrRef "\"\8220\147"
           notFollowedBy . satisfy $ flip elem [' ', '\t', '\n']

doubleQuoteEnd :: Stream s m Char
               => ParserT s st m ()
doubleQuoteEnd = void (charOrRef "\"\8221\148")

ellipses :: Stream s m Char
         => ParserT s st m Inlines
ellipses = try (string "..." >> return (B.str "\8230"))

dash :: (HasReaderOptions st, Stream s m Char)
     => ParserT s st m Inlines
dash = try $ do
  oldDashes <- extensionEnabled Ext_old_dashes <$> getOption readerExtensions
  if oldDashes
     then do
       char '-'
       (char '-' >> return (B.str "\8212"))
         <|> (lookAhead digit >> return (B.str "\8211"))
     else do
       string "--"
       (char '-' >> return (B.str "\8212"))
         <|> return (B.str "\8211")

-- This is used to prevent exponential blowups for things like:
-- a**a*a**a*a**a*a**a*a**a*a**a*a**
nested :: Stream s m a
       => ParserT s ParserState m a
       -> ParserT s ParserState m a
nested p = do
  nestlevel <- stateMaxNestingLevel <$>  getState
  guard $ nestlevel > 0
  updateState $ \st -> st{ stateMaxNestingLevel = stateMaxNestingLevel st - 1 }
  res <- p
  updateState $ \st -> st{ stateMaxNestingLevel = nestlevel }
  return res

citeKey :: (Stream s m Char, HasLastStrPosition st)
        => ParserT s st m (Bool, String)
citeKey = try $ do
  guard =<< notAfterString
  suppress_author <- option False (char '-' *> return True)
  char '@'
  firstChar <- alphaNum <|> char '_' <|> char '*' -- @* for wildcard in nocite
  let regchar = satisfy (\c -> isAlphaNum c || c == '_')
  let internal p = try $ p <* lookAhead regchar
  rest <- many $ regchar <|> internal (oneOf ":.#$%&-+?<>~/") <|>
                 try (oneOf ":/" <* lookAhead (char '/'))
  let key = firstChar:rest
  return (suppress_author, key)


token :: (Stream s m t)
      => (t -> String)
      -> (t -> SourcePos)
      -> (t -> Maybe a)
      -> ParsecT s st m a
token pp pos match = tokenPrim pp (\_ t _ -> pos t) match

--
-- Macros
--

-- | Parse a \newcommand or \newenviroment macro definition.
macro :: (Stream [Char] m Char, HasMacros st, HasReaderOptions st)
      => ParserT [Char] st m Blocks
macro = do
  apply <- getOption readerApplyMacros
  (m, def') <- withRaw pMacroDefinition
  if apply
     then do
       updateState $ \st -> updateMacros (m:) st
       return mempty
     else return $ rawBlock "latex" def'

-- | Apply current macros to string.
applyMacros' :: (HasReaderOptions st, HasMacros st, Stream [Char] m Char)
             => String
             -> ParserT [Char] st m String
applyMacros' target = do
  apply <- getOption readerApplyMacros
  if apply
     then do macros <- extractMacros <$> getState
             return $ applyMacros macros target
     else return target

infixr 5 <+?>
(<+?>) :: (Monoid a) => ParserT s st m a -> ParserT s st m a -> ParserT s st m a
a <+?> b = a >>= flip fmap (try b <|> return mempty) . (<>)

extractIdClass :: Attr -> Attr
extractIdClass (ident, cls, kvs) = (ident', cls', kvs')
  where
    ident' = case (lookup "id" kvs) of
               Just v  -> v
               Nothing -> ident
    cls'   = case (lookup "class" kvs) of
               Just cl -> words cl
               Nothing -> cls
    kvs'  = filter (\(k,_) -> k /= "id" || k /= "class") kvs

insertIncludedFile' :: (PandocMonad m, HasIncludeFiles st,
                        Functor mf, Applicative mf, Monad mf)
                    => ParserT String st m (mf Blocks)
                    -> [FilePath] -> FilePath
                    -> ParserT String st m (mf Blocks)
insertIncludedFile' blocks dirs f = do
  oldPos <- getPosition
  oldInput <- getInput
  containers <- getIncludeFiles <$> getState
  when (f `elem` containers) $
    throwError $ PandocParseError $ "Include file loop at " ++ show oldPos
  updateState $ addIncludeFile f
  mbcontents <- readFileFromDirs dirs f
  contents <- case mbcontents of
                   Just s -> return s
                   Nothing -> do
                     report $ CouldNotLoadIncludeFile f oldPos
                     return ""
  setPosition $ newPos f 1 1
  setInput contents
  bs <- blocks
  setInput oldInput
  setPosition oldPos
  updateState dropLatestIncludeFile
  return bs

-- | Parse content of include file as blocks. Circular includes result in an
-- @PandocParseError@.
insertIncludedFile :: (PandocMonad m, HasIncludeFiles st)
                   => ParserT String st m Blocks
                   -> [FilePath] -> FilePath
                   -> ParserT String st m Blocks
insertIncludedFile blocks dirs f =
  runIdentity <$> insertIncludedFile' (Identity <$> blocks) dirs f

-- | Parse content of include file as future blocks. Circular includes result in
-- an @PandocParseError@.
insertIncludedFileF :: (PandocMonad m, HasIncludeFiles st)
                    => ParserT String st m (Future st Blocks)
                    -> [FilePath] -> FilePath
                    -> ParserT String st m (Future st Blocks)
insertIncludedFileF = insertIncludedFile'