summaryrefslogtreecommitdiff
path: root/src/Text/Pandoc/Readers/Markdown.hs
blob: 8c48550765c2e17f7f9baff78dc7524f62f5ec27 (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
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
{-# LANGUAGE RelaxedPolyRec #-} -- needed for inlinesBetween on GHC < 7
{-
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.Readers.Markdown
   Copyright   : Copyright (C) 2006-2010 John MacFarlane
   License     : GNU GPL, version 2 or above

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

Conversion of markdown-formatted plain text to 'Pandoc' document.
-}
module Text.Pandoc.Readers.Markdown ( readMarkdown,
                                      readMarkdownWithWarnings ) where

import Data.List ( transpose, sortBy, findIndex, intercalate )
import qualified Data.Map as M
import Data.Ord ( comparing )
import Data.Char ( isAlphaNum, toLower )
import Data.Maybe
import Text.Pandoc.Definition
import qualified Text.Pandoc.Builder as B
import Text.Pandoc.Builder (Inlines, Blocks, trimInlines, (<>))
import Text.Pandoc.Options
import Text.Pandoc.Shared
import Text.Pandoc.Parsing hiding (tableWith)
import Text.Pandoc.Readers.LaTeX ( rawLaTeXInline, rawLaTeXBlock )
import Text.Pandoc.Readers.HTML ( htmlTag, htmlInBalanced, isInlineTag, isBlockTag,
                                  isTextTag, isCommentTag )
import Text.Pandoc.XML ( fromEntities )
import Text.Pandoc.Biblio (processBiblio)
import qualified Text.CSL as CSL
import Data.Monoid (mconcat, mempty)
import Control.Applicative ((<$>), (<*), (*>), (<$))
import Control.Monad
import Text.HTML.TagSoup
import Text.HTML.TagSoup.Match (tagOpen)
import qualified Data.Set as Set

type MarkdownParser = Parser [Char] ParserState

-- | Read markdown from an input string and return a Pandoc document.
readMarkdown :: ReaderOptions -- ^ Reader options
             -> String        -- ^ String to parse (assuming @'\n'@ line endings)
             -> Pandoc
readMarkdown opts s =
  (readWith parseMarkdown) def{ stateOptions = opts } (s ++ "\n\n")

-- | Read markdown from an input string and return a pair of a Pandoc document
-- and a list of warnings.
readMarkdownWithWarnings :: ReaderOptions -- ^ Reader options
                         -> String        -- ^ String to parse (assuming @'\n'@ line endings)
                         -> (Pandoc, [String])
readMarkdownWithWarnings opts s =
  (readWith parseMarkdownWithWarnings) def{ stateOptions = opts } (s ++ "\n\n")
 where parseMarkdownWithWarnings = do
         doc <- parseMarkdown
         warnings <- stateWarnings <$> getState
         return (doc, warnings)

trimInlinesF :: F Inlines -> F Inlines
trimInlinesF = liftM trimInlines

--
-- Constants and data structure definitions
--

isBulletListMarker :: Char -> Bool
isBulletListMarker '*' = True
isBulletListMarker '+' = True
isBulletListMarker '-' = True
isBulletListMarker _   = False

isHruleChar :: Char -> Bool
isHruleChar '*' = True
isHruleChar '-' = True
isHruleChar '_' = True
isHruleChar _   = False

setextHChars :: String
setextHChars = "=-"

isBlank :: Char -> Bool
isBlank ' '  = True
isBlank '\t' = True
isBlank '\n' = True
isBlank _    = False

--
-- auxiliary functions
--

isNull :: F Inlines -> Bool
isNull ils = B.isNull $ runF ils def

spnl :: Parser [Char] st ()
spnl = try $ do
  skipSpaces
  optional newline
  skipSpaces
  notFollowedBy (char '\n')

indentSpaces :: MarkdownParser String
indentSpaces = try $ do
  tabStop <- getOption readerTabStop
  count tabStop (char ' ') <|>
    string "\t" <?> "indentation"

nonindentSpaces :: MarkdownParser String
nonindentSpaces = do
  tabStop <- getOption readerTabStop
  sps <- many (char ' ')
  if length sps < tabStop
     then return sps
     else unexpected "indented line"

skipNonindentSpaces :: MarkdownParser ()
skipNonindentSpaces = do
  tabStop <- getOption readerTabStop
  atMostSpaces (tabStop - 1)

atMostSpaces :: Int -> MarkdownParser ()
atMostSpaces 0 = notFollowedBy (char ' ')
atMostSpaces n = (char ' ' >> atMostSpaces (n-1)) <|> return ()

litChar :: MarkdownParser Char
litChar = escapedChar'
       <|> noneOf "\n"
       <|> (newline >> notFollowedBy blankline >> return ' ')

-- | Parse a sequence of inline elements between square brackets,
-- including inlines between balanced pairs of square brackets.
inlinesInBalancedBrackets :: MarkdownParser (F Inlines)
inlinesInBalancedBrackets = charsInBalancedBrackets >>=
  parseFromString (trimInlinesF . mconcat <$> many inline)

charsInBalancedBrackets :: MarkdownParser [Char]
charsInBalancedBrackets = do
  char '['
  result <- manyTill (  many1 (noneOf "`[]\n")
                    <|> (snd <$> withRaw code)
                    <|> ((\xs -> '[' : xs ++ "]") <$> charsInBalancedBrackets)
                    <|> count 1 (satisfy (/='\n'))
                    <|> (newline >> notFollowedBy blankline >> return "\n")
                     ) (char ']')
  return $ concat result

--
-- document structure
--

titleLine :: MarkdownParser (F Inlines)
titleLine = try $ do
  char '%'
  skipSpaces
  res <- many $ (notFollowedBy newline >> inline)
             <|> try (endline >> whitespace)
  newline
  return $ trimInlinesF $ mconcat res

authorsLine :: MarkdownParser (F [Inlines])
authorsLine = try $ do
  char '%'
  skipSpaces
  authors <- sepEndBy (many (notFollowedBy (satisfy $ \c ->
                                c == ';' || c == '\n') >> inline))
                       (char ';' <|>
                        try (newline >> notFollowedBy blankline >> spaceChar))
  newline
  return $ sequence $ filter (not . isNull) $ map (trimInlinesF . mconcat) authors

dateLine :: MarkdownParser (F Inlines)
dateLine = try $ do
  char '%'
  skipSpaces
  trimInlinesF . mconcat <$> manyTill inline newline

titleBlock :: MarkdownParser (F Inlines, F [Inlines], F Inlines)
titleBlock = pandocTitleBlock <|> mmdTitleBlock

pandocTitleBlock :: MarkdownParser (F Inlines, F [Inlines], F Inlines)
pandocTitleBlock = try $ do
  guardEnabled Ext_pandoc_title_block
  title <- option mempty titleLine
  author <- option (return []) authorsLine
  date <- option mempty dateLine
  optional blanklines
  return (title, author, date)

mmdTitleBlock :: MarkdownParser (F Inlines, F [Inlines], F Inlines)
mmdTitleBlock = try $ do
  guardEnabled Ext_mmd_title_block
  kvPairs <- many1 kvPair
  blanklines
  let title = maybe mempty return $ lookup "title" kvPairs
  let author = maybe mempty (\x -> return [x]) $ lookup "author" kvPairs
  let date = maybe mempty return $ lookup "date" kvPairs
  return (title, author, date)

kvPair :: MarkdownParser (String, Inlines)
kvPair = try $ do
  key <- many1Till (alphaNum <|> oneOf "_- ") (char ':')
  val <- manyTill anyChar
          (try $ newline >> lookAhead (blankline <|> nonspaceChar))
  let key' = concat $ words $ map toLower key
  let val' = trimInlines $ B.text val
  return (key',val')

parseMarkdown :: MarkdownParser Pandoc
parseMarkdown = do
  -- markdown allows raw HTML
  updateState $ \state -> state { stateOptions =
                let oldOpts = stateOptions state in
                    oldOpts{ readerParseRaw = True } }
  (title, authors, date) <- option (mempty,return [],mempty) titleBlock
  blocks <- parseBlocks
  st <- getState
  mbsty <- getOption readerCitationStyle
  refs <- getOption readerReferences
  return $ processBiblio mbsty refs
         $ B.setTitle (runF title st)
         $ B.setAuthors (runF authors st)
         $ B.setDate (runF date st)
         $ B.doc $ runF blocks st

addWarning :: Maybe SourcePos -> String -> MarkdownParser ()
addWarning mbpos msg =
  updateState $ \st -> st{
    stateWarnings = (msg ++ maybe "" (\pos -> " " ++ show pos) mbpos) :
                     stateWarnings st }

referenceKey :: MarkdownParser (F Blocks)
referenceKey = try $ do
  pos <- getPosition
  skipNonindentSpaces
  (_,raw) <- reference
  char ':'
  skipSpaces >> optional newline >> skipSpaces >> notFollowedBy (char '[')
  let sourceURL = liftM unwords $ many $ try $ do
                    notFollowedBy' referenceTitle
                    skipMany spaceChar
                    optional $ newline >> notFollowedBy blankline
                    skipMany spaceChar
                    notFollowedBy' (() <$ reference)
                    many1 $ escapedChar' <|> satisfy (not . isBlank)
  let betweenAngles = try $ char '<' >>
                       manyTill (escapedChar' <|> litChar) (char '>')
  src <- try betweenAngles <|> sourceURL
  tit <- option "" referenceTitle
  blanklines
  let target = (escapeURI $ trimr src,  tit)
  st <- getState
  let oldkeys = stateKeys st
  let key = toKey raw
  case M.lookup key oldkeys of
    Just _  -> addWarning (Just pos) $ "Duplicate link reference `" ++ raw ++ "'"
    Nothing -> return ()
  updateState $ \s -> s { stateKeys = M.insert key target oldkeys }
  return $ return mempty

referenceTitle :: MarkdownParser String
referenceTitle = try $ do
  skipSpaces >> optional newline >> skipSpaces
  tit <-    (charsInBalanced '(' ')' litChar >>= return . unwords . words)
        <|> do delim <- char '\'' <|> char '"'
               manyTill litChar (try (char delim >> skipSpaces >>
                                      notFollowedBy (noneOf ")\n")))
  return $ fromEntities tit

-- | PHP Markdown Extra style abbreviation key.  Currently
-- we just skip them, since Pandoc doesn't have an element for
-- an abbreviation.
abbrevKey :: MarkdownParser (F Blocks)
abbrevKey = do
  guardEnabled Ext_abbreviations
  try $ do
    char '*'
    reference
    char ':'
    skipMany (satisfy (/= '\n'))
    blanklines
    return $ return mempty

noteMarker :: MarkdownParser String
noteMarker = string "[^" >> many1Till (satisfy $ not . isBlank) (char ']')

rawLine :: MarkdownParser String
rawLine = try $ do
  notFollowedBy blankline
  notFollowedBy' $ try $ skipNonindentSpaces >> noteMarker
  optional indentSpaces
  anyLine

rawLines :: MarkdownParser String
rawLines = do
  first <- anyLine
  rest <- many rawLine
  return $ unlines (first:rest)

noteBlock :: MarkdownParser (F Blocks)
noteBlock = try $ do
  pos <- getPosition
  skipNonindentSpaces
  ref <- noteMarker
  char ':'
  optional blankline
  optional indentSpaces
  first <- rawLines
  rest <- many $ try $ blanklines >> indentSpaces >> rawLines
  let raw = unlines (first:rest) ++ "\n"
  optional blanklines
  parsed <- parseFromString parseBlocks raw
  let newnote = (ref, parsed)
  oldnotes <- stateNotes' <$> getState
  case lookup ref oldnotes of
    Just _  -> addWarning (Just pos) $ "Duplicate note reference `" ++ ref ++ "'"
    Nothing -> return ()
  updateState $ \s -> s { stateNotes' = newnote : oldnotes }
  return mempty

--
-- parsing blocks
--

parseBlocks :: MarkdownParser (F Blocks)
parseBlocks = mconcat <$> manyTill block eof

block :: MarkdownParser (F Blocks)
block = choice [ codeBlockFenced
               , codeBlockBackticks
               , guardEnabled Ext_latex_macros *> (mempty <$ macro)
               , header
               , rawTeXBlock
               , htmlBlock
               , table
               , codeBlockIndented
               , lhsCodeBlock
               , blockQuote
               , hrule
               , bulletList
               , orderedList
               , definitionList
               , noteBlock
               , referenceKey
               , abbrevKey
               , para
               , plain
               ] <?> "block"

--
-- header blocks
--

header :: MarkdownParser (F Blocks)
header = setextHeader <|> atxHeader <?> "header"

addToHeaderList :: F Inlines -> MarkdownParser ()
addToHeaderList text =
  updateState $ \st -> st{ stateHeaders = B.toList (runF text defaultParserState)
                                          : stateHeaders st }

atxHeader :: MarkdownParser (F Blocks)
atxHeader = try $ do
  level <- many1 (char '#') >>= return . length
  notFollowedBy (char '.' <|> char ')') -- this would be a list
  skipSpaces
  text <- trimInlinesF . mconcat <$> manyTill inline atxClosing
  addToHeaderList text
  return $ B.header level <$> text

atxClosing :: Parser [Char] st String
atxClosing = try $ skipMany (char '#') >> blanklines

setextHeader :: MarkdownParser (F Blocks)
setextHeader = try $ do
  -- This lookahead prevents us from wasting time parsing Inlines
  -- unless necessary -- it gives a significant performance boost.
  lookAhead $ anyLine >> many1 (oneOf setextHChars) >> blankline
  text <- trimInlinesF . mconcat <$> many1Till inline newline
  underlineChar <- oneOf setextHChars
  many (char underlineChar)
  blanklines
  let level = (fromMaybe 0 $ findIndex (== underlineChar) setextHChars) + 1
  addToHeaderList text
  return $ B.header level <$> text

--
-- hrule block
--

hrule :: Parser [Char] st (F Blocks)
hrule = try $ do
  skipSpaces
  start <- satisfy isHruleChar
  count 2 (skipSpaces >> char start)
  skipMany (spaceChar <|> char start)
  newline
  optional blanklines
  return $ return B.horizontalRule

--
-- code blocks
--

indentedLine :: MarkdownParser String
indentedLine = indentSpaces >> manyTill anyChar newline >>= return . (++ "\n")

blockDelimiter :: (Char -> Bool)
               -> Maybe Int
               -> Parser [Char] st Int
blockDelimiter f len = try $ do
  c <- lookAhead (satisfy f)
  case len of
      Just l  -> count l (char c) >> many (char c) >> return l
      Nothing -> count 3 (char c) >> many (char c) >>=
                 return . (+ 3) . length

attributes :: Parser [Char] st (String, [String], [(String, String)])
attributes = try $ do
  char '{'
  spnl
  attrs <- many (attribute >>~ spnl)
  char '}'
  let (ids, classes, keyvals) = unzip3 attrs
  let firstNonNull [] = ""
      firstNonNull (x:xs) | not (null x) = x
                          | otherwise    = firstNonNull xs
  return (firstNonNull $ reverse ids, concat classes, concat keyvals)

attribute :: Parser [Char] st (String, [String], [(String, String)])
attribute = identifierAttr <|> classAttr <|> keyValAttr

identifier :: Parser [Char] st String
identifier = do
  first <- letter
  rest <- many $ alphaNum <|> oneOf "-_:."
  return (first:rest)

identifierAttr :: Parser [Char] st (String, [a], [a1])
identifierAttr = try $ do
  char '#'
  result <- identifier
  return (result,[],[])

classAttr :: Parser [Char] st (String, [String], [a])
classAttr = try $ do
  char '.'
  result <- identifier
  return ("",[result],[])

keyValAttr :: Parser [Char] st (String, [a], [(String, String)])
keyValAttr = try $ do
  key <- identifier
  char '='
  val <- enclosed (char '"') (char '"') anyChar
     <|> enclosed (char '\'') (char '\'') anyChar
     <|> many nonspaceChar
  return ("",[],[(key,val)])

codeBlockFenced :: MarkdownParser (F Blocks)
codeBlockFenced = try $ do
  guardEnabled Ext_fenced_code_blocks
  size <- blockDelimiter (=='~') Nothing
  skipMany spaceChar
  attr <- option ([],[],[]) $
            guardEnabled Ext_fenced_code_attributes >> attributes
  blankline
  contents <- manyTill anyLine (blockDelimiter (=='~') (Just size))
  blanklines
  return $ return $ B.codeBlockWith attr $ intercalate "\n" contents

codeBlockBackticks :: MarkdownParser (F Blocks)
codeBlockBackticks = try $ do
  guardEnabled Ext_backtick_code_blocks
  blockDelimiter (=='`') (Just 3)
  skipMany spaceChar
  cls <- many1 alphaNum
  blankline
  contents <- manyTill anyLine $ blockDelimiter (=='`') (Just 3)
  blanklines
  return $ return $ B.codeBlockWith ("",[cls],[]) $ intercalate "\n" contents

codeBlockIndented :: MarkdownParser (F Blocks)
codeBlockIndented = do
  contents <- many1 (indentedLine <|>
                     try (do b <- blanklines
                             l <- indentedLine
                             return $ b ++ l))
  optional blanklines
  classes <- getOption readerIndentedCodeClasses
  return $ return $ B.codeBlockWith ("", classes, []) $
           stripTrailingNewlines $ concat contents

lhsCodeBlock :: MarkdownParser (F Blocks)
lhsCodeBlock = do
  guardEnabled Ext_literate_haskell
  (return . B.codeBlockWith ("",["sourceCode","literate","haskell"],[]) <$>
          (lhsCodeBlockBird <|> lhsCodeBlockLaTeX))
    <|> (return . B.codeBlockWith ("",["sourceCode","haskell"],[]) <$>
          lhsCodeBlockInverseBird)

lhsCodeBlockLaTeX :: MarkdownParser String
lhsCodeBlockLaTeX = try $ do
  string "\\begin{code}"
  manyTill spaceChar newline
  contents <- many1Till anyChar (try $ string "\\end{code}")
  blanklines
  return $ stripTrailingNewlines contents

lhsCodeBlockBird :: MarkdownParser String
lhsCodeBlockBird = lhsCodeBlockBirdWith '>'

lhsCodeBlockInverseBird :: MarkdownParser String
lhsCodeBlockInverseBird = lhsCodeBlockBirdWith '<'

lhsCodeBlockBirdWith :: Char -> MarkdownParser String
lhsCodeBlockBirdWith c = try $ do
  pos <- getPosition
  when (sourceColumn pos /= 1) $ fail "Not in first column"
  lns <- many1 $ birdTrackLine c
  -- if (as is normal) there is always a space after >, drop it
  let lns' = if all (\ln -> null ln || take 1 ln == " ") lns
                then map (drop 1) lns
                else lns
  blanklines
  return $ intercalate "\n" lns'

birdTrackLine :: Char -> Parser [Char] st String
birdTrackLine c = try $ do
  char c
  -- allow html tags on left margin:
  when (c == '<') $ notFollowedBy letter
  manyTill anyChar newline

--
-- block quotes
--

emailBlockQuoteStart :: MarkdownParser Char
emailBlockQuoteStart = try $ skipNonindentSpaces >> char '>' >>~ optional (char ' ')

emailBlockQuote :: MarkdownParser [String]
emailBlockQuote = try $ do
  emailBlockQuoteStart
  let emailLine = many $ nonEndline <|> try
                         (endline >> notFollowedBy emailBlockQuoteStart >>
                         return '\n')
  let emailSep = try (newline >> emailBlockQuoteStart)
  first <- emailLine
  rest <- many $ try $ emailSep >> emailLine
  let raw = first:rest
  newline <|> (eof >> return '\n')
  optional blanklines
  return raw

blockQuote :: MarkdownParser (F Blocks)
blockQuote = do
  raw <- emailBlockQuote
  -- parse the extracted block, which may contain various block elements:
  contents <- parseFromString parseBlocks $ (intercalate "\n" raw) ++ "\n\n"
  return $ B.blockQuote <$> contents

--
-- list blocks
--

bulletListStart :: MarkdownParser ()
bulletListStart = try $ do
  optional newline -- if preceded by a Plain block in a list context
  skipNonindentSpaces
  notFollowedBy' (() <$ hrule)     -- because hrules start out just like lists
  satisfy isBulletListMarker
  spaceChar
  skipSpaces

anyOrderedListStart :: MarkdownParser (Int, ListNumberStyle, ListNumberDelim)
anyOrderedListStart = try $ do
  optional newline -- if preceded by a Plain block in a list context
  skipNonindentSpaces
  notFollowedBy $ string "p." >> spaceChar >> digit  -- page number
  (guardDisabled Ext_fancy_lists >>
       do many1 digit
          char '.'
          spaceChar
          return (1, DefaultStyle, DefaultDelim))
   <|> do (num, style, delim) <- anyOrderedListMarker
          -- if it could be an abbreviated first name, insist on more than one space
          if delim == Period && (style == UpperAlpha || (style == UpperRoman &&
             num `elem` [1, 5, 10, 50, 100, 500, 1000]))
             then char '\t' <|> (try $ char ' ' >> spaceChar)
             else spaceChar
          skipSpaces
          return (num, style, delim)

listStart :: MarkdownParser ()
listStart = bulletListStart <|> (anyOrderedListStart >> return ())

-- parse a line of a list item (start = parser for beginning of list item)
listLine :: MarkdownParser String
listLine = try $ do
  notFollowedBy blankline
  notFollowedBy' (do indentSpaces
                     many (spaceChar)
                     listStart)
  chunks <- manyTill (liftM snd (htmlTag isCommentTag) <|> count 1 anyChar) newline
  return $ concat chunks ++ "\n"

-- parse raw text for one list item, excluding start marker and continuations
rawListItem :: MarkdownParser a
            -> MarkdownParser String
rawListItem start = try $ do
  start
  first <- listLine
  rest <- many (notFollowedBy listStart >> listLine)
  blanks <- many blankline
  return $ concat (first:rest)  ++ blanks

-- continuation of a list item - indented and separated by blankline
-- or (in compact lists) endline.
-- note: nested lists are parsed as continuations
listContinuation :: MarkdownParser String
listContinuation = try $ do
  lookAhead indentSpaces
  result <- many1 listContinuationLine
  blanks <- many blankline
  return $ concat result ++ blanks

listContinuationLine :: MarkdownParser String
listContinuationLine = try $ do
  notFollowedBy blankline
  notFollowedBy' listStart
  optional indentSpaces
  result <- manyTill anyChar newline
  return $ result ++ "\n"

listItem :: MarkdownParser a
         -> MarkdownParser (F Blocks)
listItem start = try $ do
  first <- rawListItem start
  continuations <- many listContinuation
  -- parsing with ListItemState forces markers at beginning of lines to
  -- count as list item markers, even if not separated by blank space.
  -- see definition of "endline"
  state <- getState
  let oldContext = stateParserContext state
  setState $ state {stateParserContext = ListItemState}
  -- parse the extracted block, which may contain various block elements:
  let raw = concat (first:continuations)
  contents <- parseFromString parseBlocks raw
  updateState (\st -> st {stateParserContext = oldContext})
  return contents

orderedList :: MarkdownParser (F Blocks)
orderedList = try $ do
  (start, style, delim) <- lookAhead anyOrderedListStart
  unless ((style == DefaultStyle || style == Decimal || style == Example) &&
          (delim == DefaultDelim || delim == Period)) $
    guardEnabled Ext_fancy_lists
  when (style == Example) $ guardEnabled Ext_example_lists
  items <- fmap sequence $ many1 $ listItem
                 ( try $ do
                     optional newline -- if preceded by Plain block in a list
                     skipNonindentSpaces
                     orderedListMarker style delim )
  start' <- option 1 $ guardEnabled Ext_startnum >> return start
  return $ B.orderedListWith (start', style, delim) <$> fmap compactify' items

bulletList :: MarkdownParser (F Blocks)
bulletList = do
  items <- fmap sequence $ many1 $ listItem  bulletListStart
  return $ B.bulletList <$> fmap compactify' items

-- definition lists

defListMarker :: MarkdownParser ()
defListMarker = do
  sps <- nonindentSpaces
  char ':' <|> char '~'
  tabStop <- getOption readerTabStop
  let remaining = tabStop - (length sps + 1)
  if remaining > 0
     then count remaining (char ' ') <|> string "\t"
     else mzero
  return ()

definitionListItem :: MarkdownParser (F (Inlines, [Blocks]))
definitionListItem = try $ do
  guardEnabled Ext_definition_lists
  -- first, see if this has any chance of being a definition list:
  lookAhead (anyLine >> optional blankline >> defListMarker)
  term <- trimInlinesF . mconcat <$> manyTill inline newline
  optional blankline
  raw <- many1 defRawBlock
  state <- getState
  let oldContext = stateParserContext state
  -- parse the extracted block, which may contain various block elements:
  contents <- mapM (parseFromString parseBlocks) raw
  updateState (\st -> st {stateParserContext = oldContext})
  return $ liftM2 (,) term (sequence contents)

defRawBlock :: MarkdownParser String
defRawBlock = try $ do
  defListMarker
  firstline <- anyLine
  rawlines <- many (notFollowedBy blankline >> indentSpaces >> anyLine)
  trailing <- option "" blanklines
  cont <- liftM concat $ many $ do
            lns <- many1 $ notFollowedBy blankline >> indentSpaces >> anyLine
            trl <- option "" blanklines
            return $ unlines lns ++ trl
  return $ firstline ++ "\n" ++ unlines rawlines ++ trailing ++ cont

definitionList :: MarkdownParser (F Blocks)
definitionList = do
  items <- fmap sequence $ many1 definitionListItem
  return $ B.definitionList <$> fmap compactify'DL items

compactify'DL :: [(Inlines, [Blocks])] -> [(Inlines, [Blocks])]
compactify'DL items =
  let defs = concatMap snd items
      defBlocks = reverse $ concatMap B.toList defs
      isPara (Para _) = True
      isPara _        = False
  in  case defBlocks of
           (Para x:_) -> if not $ any isPara (drop 1 defBlocks)
                            then let (t,ds) = last items
                                     lastDef = B.toList $ last ds
                                     ds' = init ds ++
                                          [B.fromList $ init lastDef ++ [Plain x]]
                                  in init items ++ [(t, ds')]
                            else items
           _          -> items

--
-- paragraph block
--

{-
isHtmlOrBlank :: Inline -> Bool
isHtmlOrBlank (RawInline "html" _) = True
isHtmlOrBlank (Space)         = True
isHtmlOrBlank (LineBreak)     = True
isHtmlOrBlank _               = False
-}

para :: MarkdownParser (F Blocks)
para = try $ do
  result <- trimInlinesF . mconcat <$> many1 inline
  -- TODO remove this if not really needed?  and remove isHtmlOrBlank
  -- guard $ not $ F.all isHtmlOrBlank result
  option (B.plain <$> result) $ try $ do
              newline
              (blanklines >> return mempty)
                <|> (guardDisabled Ext_blank_before_blockquote >> lookAhead blockQuote)
                <|> (guardDisabled Ext_blank_before_header >> lookAhead header)
              return $ B.para <$> result

plain :: MarkdownParser (F Blocks)
plain = fmap B.plain . trimInlinesF . mconcat <$> many1 inline <* spaces

--
-- raw html
--

htmlElement :: MarkdownParser String
htmlElement = strictHtmlBlock <|> liftM snd (htmlTag isBlockTag)

htmlBlock :: MarkdownParser (F Blocks)
htmlBlock = do
  guardEnabled Ext_raw_html
  res <- (guardEnabled Ext_markdown_in_html_blocks >> rawHtmlBlocks)
          <|> htmlBlock'
  return $ return $ B.rawBlock "html" res

htmlBlock' :: MarkdownParser String
htmlBlock' = try $ do
    first <- htmlElement
    finalSpace <- many spaceChar
    finalNewlines <- many newline
    return $ first ++ finalSpace ++ finalNewlines

strictHtmlBlock :: MarkdownParser String
strictHtmlBlock = htmlInBalanced (not . isInlineTag)

rawVerbatimBlock :: MarkdownParser String
rawVerbatimBlock = try $ do
  (TagOpen tag _, open) <- htmlTag (tagOpen (\t ->
                              t == "pre" || t == "style" || t == "script")
                              (const True))
  contents <- manyTill anyChar (htmlTag (~== TagClose tag))
  return $ open ++ contents ++ renderTags [TagClose tag]

rawTeXBlock :: MarkdownParser (F Blocks)
rawTeXBlock = do
  guardEnabled Ext_raw_tex
  result <- (B.rawBlock "latex" <$> rawLaTeXBlock)
        <|> (B.rawBlock "context" <$> rawConTeXtEnvironment)
  spaces
  return $ return result

rawHtmlBlocks :: MarkdownParser String
rawHtmlBlocks = do
  htmlBlocks <- many1 $ try $ do
                          s <- rawVerbatimBlock <|> try (
                                do (t,raw) <- htmlTag isBlockTag
                                   exts <- getOption readerExtensions
                                   -- if open tag, need markdown="1" if
                                   -- markdown_attributes extension is set
                                   case t of
                                        TagOpen _ as
                                          | Ext_markdown_attribute `Set.member`
                                              exts ->
                                                if "markdown" `notElem`
                                                   map fst as
                                                   then mzero
                                                   else return $
                                                     stripMarkdownAttribute raw
                                          | otherwise -> return raw
                                        _ -> return raw )
                          sps <- do sp1 <- many spaceChar
                                    sp2 <- option "" (blankline >> return "\n")
                                    sp3 <- many spaceChar
                                    sp4 <- option "" blanklines
                                    return $ sp1 ++ sp2 ++ sp3 ++ sp4
                          -- note: we want raw html to be able to
                          -- precede a code block, when separated
                          -- by a blank line
                          return $ s ++ sps
  let combined = concat htmlBlocks
  return $ if last combined == '\n' then init combined else combined

-- remove markdown="1" attribute
stripMarkdownAttribute :: String -> String
stripMarkdownAttribute s = renderTags' $ map filterAttrib $ parseTags s
  where filterAttrib (TagOpen t as) = TagOpen t
                                        [(k,v) | (k,v) <- as, k /= "markdown"]
        filterAttrib              x = x

--
-- Tables
--

-- Parse a dashed line with optional trailing spaces; return its length
-- and the length including trailing space.
dashedLine :: Char
           -> Parser [Char] st (Int, Int)
dashedLine ch = do
  dashes <- many1 (char ch)
  sp     <- many spaceChar
  return $ (length dashes, length $ dashes ++ sp)

-- Parse a table header with dashed lines of '-' preceded by
-- one (or zero) line of text.
simpleTableHeader :: Bool  -- ^ Headerless table
                  -> MarkdownParser (F [Blocks], [Alignment], [Int])
simpleTableHeader headless = try $ do
  rawContent  <- if headless
                    then return ""
                    else anyLine
  initSp      <- nonindentSpaces
  dashes      <- many1 (dashedLine '-')
  newline
  let (lengths, lines') = unzip dashes
  let indices  = scanl (+) (length initSp) lines'
  -- If no header, calculate alignment on basis of first row of text
  rawHeads <- liftM (tail . splitStringByIndices (init indices)) $
              if headless
                 then lookAhead anyLine
                 else return rawContent
  let aligns   = zipWith alignType (map (\a -> [a]) rawHeads) lengths
  let rawHeads' = if headless
                     then replicate (length dashes) ""
                     else rawHeads
  heads <- fmap sequence
           $ mapM (parseFromString (mconcat <$> many plain))
           $ map trim rawHeads'
  return (heads, aligns, indices)

-- Returns an alignment type for a table, based on a list of strings
-- (the rows of the column header) and a number (the length of the
-- dashed line under the rows.
alignType :: [String]
          -> Int
          -> Alignment
alignType [] _ = AlignDefault
alignType strLst len =
  let nonempties = filter (not . null) $ map trimr strLst
      (leftSpace, rightSpace) =
           case sortBy (comparing length) nonempties of
                 (x:_)  -> (head x `elem` " \t", length x < len)
                 []     -> (False, False)
  in  case (leftSpace, rightSpace) of
        (True,  False)   -> AlignRight
        (False, True)    -> AlignLeft
        (True,  True)    -> AlignCenter
        (False, False)   -> AlignDefault

-- Parse a table footer - dashed lines followed by blank line.
tableFooter :: MarkdownParser String
tableFooter = try $ skipNonindentSpaces >> many1 (dashedLine '-') >> blanklines

-- Parse a table separator - dashed line.
tableSep :: MarkdownParser Char
tableSep = try $ skipNonindentSpaces >> many1 (dashedLine '-') >> char '\n'

-- Parse a raw line and split it into chunks by indices.
rawTableLine :: [Int]
             -> MarkdownParser [String]
rawTableLine indices = do
  notFollowedBy' (blanklines <|> tableFooter)
  line <- many1Till anyChar newline
  return $ map trim $ tail $
           splitStringByIndices (init indices) line

-- Parse a table line and return a list of lists of blocks (columns).
tableLine :: [Int]
          -> MarkdownParser (F [Blocks])
tableLine indices = rawTableLine indices >>=
  fmap sequence . mapM (parseFromString (mconcat <$> many plain))

-- Parse a multiline table row and return a list of blocks (columns).
multilineRow :: [Int]
             -> MarkdownParser (F [Blocks])
multilineRow indices = do
  colLines <- many1 (rawTableLine indices)
  let cols = map unlines $ transpose colLines
  fmap sequence $ mapM (parseFromString (mconcat <$> many plain)) cols

-- Parses a table caption:  inlines beginning with 'Table:'
-- and followed by blank lines.
tableCaption :: MarkdownParser (F Inlines)
tableCaption = try $ do
  guardEnabled Ext_table_captions
  skipNonindentSpaces
  string ":" <|> string "Table:"
  trimInlinesF . mconcat <$> many1 inline <* blanklines

-- Parse a simple table with '---' header and one line per row.
simpleTable :: Bool  -- ^ Headerless table
            -> MarkdownParser ([Alignment], [Double], F [Blocks], F [[Blocks]])
simpleTable headless = do
  (aligns, _widths, heads', lines') <-
       tableWith (simpleTableHeader headless) tableLine
              (return ())
              (if headless then tableFooter else tableFooter <|> blanklines)
  -- Simple tables get 0s for relative column widths (i.e., use default)
  return (aligns, replicate (length aligns) 0, heads', lines')

-- Parse a multiline table:  starts with row of '-' on top, then header
-- (which may be multiline), then the rows,
-- which may be multiline, separated by blank lines, and
-- ending with a footer (dashed line followed by blank line).
multilineTable :: Bool -- ^ Headerless table
               -> MarkdownParser ([Alignment], [Double], F [Blocks], F [[Blocks]])
multilineTable headless =
  tableWith (multilineTableHeader headless) multilineRow blanklines tableFooter

multilineTableHeader :: Bool -- ^ Headerless table
                     -> MarkdownParser (F [Blocks], [Alignment], [Int])
multilineTableHeader headless = try $ do
  if headless
     then return '\n'
     else tableSep >>~ notFollowedBy blankline
  rawContent  <- if headless
                    then return $ repeat ""
                    else many1
                         (notFollowedBy tableSep >> many1Till anyChar newline)
  initSp      <- nonindentSpaces
  dashes      <- many1 (dashedLine '-')
  newline
  let (lengths, lines') = unzip dashes
  let indices  = scanl (+) (length initSp) lines'
  rawHeadsList <- if headless
                     then liftM (map (:[]) . tail .
                              splitStringByIndices (init indices)) $ lookAhead anyLine
                     else return $ transpose $ map
                           (\ln -> tail $ splitStringByIndices (init indices) ln)
                           rawContent
  let aligns   = zipWith alignType rawHeadsList lengths
  let rawHeads = if headless
                    then replicate (length dashes) ""
                    else map (intercalate " ") rawHeadsList
  heads <- fmap sequence $
           mapM (parseFromString (mconcat <$> many plain)) $
             map trim rawHeads
  return (heads, aligns, indices)

-- 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).
gridTable :: Bool -- ^ Headerless table
          -> MarkdownParser ([Alignment], [Double], F [Blocks], F [[Blocks]])
gridTable headless =
  tableWith (gridTableHeader headless) gridTableRow
            (gridTableSep '-') gridTableFooter

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

gridPart :: Char -> Parser [Char] st (Int, Int)
gridPart ch = do
  dashes <- many1 (char ch)
  char '+'
  return (length dashes, length dashes + 1)

gridDashedLines :: Char -> Parser [Char] st [(Int,Int)]
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 :: Char -> MarkdownParser Char
gridTableSep ch = try $ gridDashedLines ch >> return '\n'

-- | Parse header for a grid table.
gridTableHeader :: Bool -- ^ Headerless table
                -> MarkdownParser (F [Blocks], [Alignment], [Int])
gridTableHeader headless = try $ do
  optional blanklines
  dashes <- gridDashedLines '-'
  rawContent  <- if headless
                    then return $ repeat ""
                    else many1
                         (notFollowedBy (gridTableSep '=') >> char '|' >>
                           many1Till anyChar newline)
  if headless
     then return ()
     else gridTableSep '=' >> return ()
  let lines'   = map snd dashes
  let indices  = scanl (+) 0 lines'
  let aligns   = replicate (length lines') AlignDefault
  -- RST does not have a notion of alignments
  let rawHeads = if headless
                    then replicate (length dashes) ""
                    else map (intercalate " ") $ transpose
                       $ map (gridTableSplitLine indices) rawContent
  heads <- fmap sequence $ mapM (parseFromString block) $
               map trim rawHeads
  return (heads, aligns, indices)

gridTableRawLine :: [Int] -> MarkdownParser [String]
gridTableRawLine indices = do
  char '|'
  line <- many1Till anyChar newline
  return (gridTableSplitLine indices line)

-- | Parse row of grid table.
gridTableRow :: [Int]
             -> MarkdownParser (F [Blocks])
gridTableRow indices = do
  colLines <- many1 (gridTableRawLine indices)
  let cols = map ((++ "\n") . unlines . removeOneLeadingSpace) $
               transpose colLines
  fmap compactify' <$> fmap sequence (mapM (parseFromString block) cols)

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 :: MarkdownParser [Char]
gridTableFooter = blanklines

pipeTable :: MarkdownParser ([Alignment], [Double], F [Blocks], F [[Blocks]])
pipeTable = try $ do
  let pipeBreak = nonindentSpaces *> optional (char '|') *>
                      pipeTableHeaderPart `sepBy1` sepPipe <*
                      optional (char '|') <* blankline
  (heads,aligns) <- try ( pipeBreak >>= \als ->
                     return (return $ replicate (length als) mempty, als))
                  <|> ( pipeTableRow >>= \row -> pipeBreak >>= \als ->

                          return (row, als) )
  lines' <- sequence <$> many1 pipeTableRow
  blanklines
  let widths = replicate (length aligns) 0.0
  return $ (aligns, widths, heads, lines')

sepPipe :: MarkdownParser ()
sepPipe = try $ do
  char '|' <|> char '+'
  notFollowedBy blankline

-- parse a row, also returning probable alignments for org-table cells
pipeTableRow :: MarkdownParser (F [Blocks])
pipeTableRow = do
  nonindentSpaces
  optional (char '|')
  let cell = mconcat <$>
                 many (notFollowedBy (blankline <|> char '|') >> inline)
  first <- cell
  sepPipe
  rest <- cell `sepBy1` sepPipe
  optional (char '|')
  blankline
  let cells  = sequence (first:rest)
  return $ do
    cells' <- cells
    return $ map
        (\ils ->
           case trimInlines ils of
                 ils' | B.isNull ils' -> mempty
                      | otherwise   -> B.plain $ ils') cells'

pipeTableHeaderPart :: Parser [Char] st Alignment
pipeTableHeaderPart = do
  left <- optionMaybe (char ':')
  many1 (char '-')
  right <- optionMaybe (char ':')
  return $
    case (left,right) of
      (Nothing,Nothing) -> AlignDefault
      (Just _,Nothing)  -> AlignLeft
      (Nothing,Just _)  -> AlignRight
      (Just _,Just _)   -> AlignCenter

-- Succeed only if current line contains a pipe.
scanForPipe :: Parser [Char] st ()
scanForPipe = lookAhead (manyTill (satisfy (/='\n')) (char '|')) >> return ()

-- | Parse a table using 'headerParser', 'rowParser',
-- 'lineParser', and 'footerParser'.  Variant of the version in
-- Text.Pandoc.Parsing.
tableWith :: MarkdownParser (F [Blocks], [Alignment], [Int])
          -> ([Int] -> MarkdownParser (F [Blocks]))
          -> MarkdownParser sep
          -> MarkdownParser end
          -> MarkdownParser ([Alignment], [Double], F [Blocks], F [[Blocks]])
tableWith headerParser rowParser lineParser footerParser = try $ do
    (heads, aligns, indices) <- headerParser
    lines' <- fmap 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')

table :: MarkdownParser (F Blocks)
table = try $ do
  frontCaption <- option Nothing (Just <$> tableCaption)
  (aligns, widths, heads, lns) <-
         try (guardEnabled Ext_pipe_tables >> scanForPipe >> pipeTable) <|>
         try (guardEnabled Ext_multiline_tables >>
                multilineTable False) <|>
         try (guardEnabled Ext_simple_tables >>
                (simpleTable True <|> simpleTable False)) <|>
         try (guardEnabled Ext_multiline_tables >>
                multilineTable True) <|>
         try (guardEnabled Ext_grid_tables >>
                (gridTable False <|> gridTable True)) <?> "table"
  optional blanklines
  caption <- case frontCaption of
                  Nothing  -> option (return mempty) tableCaption
                  Just c   -> return c
  return $ do
    caption' <- caption
    heads' <- heads
    lns' <- lns
    return $ B.table caption' (zip aligns widths) heads' lns'

--
-- inline
--

inline :: MarkdownParser (F Inlines)
inline = choice [ whitespace
                , bareURL
                , str
                , endline
                , code
                , fours
                , strong
                , emph
                , note
                , cite
                , link
                , image
                , math
                , strikeout
                , superscript
                , subscript
                , inlineNote  -- after superscript because of ^[link](/foo)^
                , autoLink
                , rawHtmlInline
                , escapedChar
                , rawLaTeXInline'
                , exampleRef
                , smart
                , return . B.singleton <$> charRef
                , symbol
                , ltSign
                ] <?> "inline"

escapedChar' :: MarkdownParser Char
escapedChar' = try $ do
  char '\\'
  (guardEnabled Ext_all_symbols_escapable >> satisfy (not . isAlphaNum))
     <|> oneOf "\\`*_{}[]()>#+-.!~"

escapedChar :: MarkdownParser (F Inlines)
escapedChar = do
  result <- escapedChar'
  case result of
       ' '   -> return $ return $ B.str "\160" -- "\ " is a nonbreaking space
       '\n'  -> guardEnabled Ext_escaped_line_breaks >>
                return (return B.linebreak)  -- "\[newline]" is a linebreak
       _     -> return $ return $ B.str [result]

ltSign :: MarkdownParser (F Inlines)
ltSign = do
  guardDisabled Ext_raw_html
    <|> guardDisabled Ext_markdown_in_html_blocks
    <|> (notFollowedBy' rawHtmlBlocks >> return ())
  char '<'
  return $ return $ B.str "<"

exampleRef :: MarkdownParser (F Inlines)
exampleRef = try $ do
  guardEnabled Ext_example_lists
  char '@'
  lab <- many1 (alphaNum <|> oneOf "-_")
  return $ do
    st <- askF
    return $ case M.lookup lab (stateExamples st) of
                  Just n    -> B.str (show n)
                  Nothing   -> B.str ('@':lab)

symbol :: MarkdownParser (F Inlines)
symbol = do
  result <- noneOf "<\\\n\t "
         <|> try (do lookAhead $ char '\\'
                     notFollowedBy' (() <$ rawTeXBlock)
                     char '\\')
  return $ return $ B.str [result]

-- parses inline code, between n `s and n `s
code :: MarkdownParser (F Inlines)
code = try $ do
  starts <- many1 (char '`')
  skipSpaces
  result <- many1Till (many1 (noneOf "`\n") <|> many1 (char '`') <|>
                       (char '\n' >> notFollowedBy' blankline >> return " "))
                      (try (skipSpaces >> count (length starts) (char '`') >>
                      notFollowedBy (char '`')))
  attr <- option ([],[],[]) (try $ guardEnabled Ext_inline_code_attributes >>
                                   optional whitespace >> attributes)
  return $ return $ B.codeWith attr $ trim $ concat result

math :: MarkdownParser (F Inlines)
math =  (return . B.displayMath <$> (mathDisplay >>= applyMacros'))
     <|> (return . B.math <$> (mathInline >>= applyMacros'))

mathDisplay :: MarkdownParser String
mathDisplay =
      (guardEnabled Ext_tex_math_dollars >> mathDisplayWith "$$" "$$")
  <|> (guardEnabled Ext_tex_math_single_backslash >>
       mathDisplayWith "\\[" "\\]")
  <|> (guardEnabled Ext_tex_math_double_backslash >>
       mathDisplayWith "\\\\[" "\\\\]")

mathDisplayWith :: String -> String -> MarkdownParser String
mathDisplayWith op cl = try $ do
  string op
  many1Till (noneOf "\n" <|> (newline >>~ notFollowedBy' blankline)) (try $ string cl)

mathInline :: MarkdownParser String
mathInline =
      (guardEnabled Ext_tex_math_dollars >> mathInlineWith "$" "$")
  <|> (guardEnabled Ext_tex_math_single_backslash >>
       mathInlineWith "\\(" "\\)")
  <|> (guardEnabled Ext_tex_math_double_backslash >>
       mathInlineWith "\\\\(" "\\\\)")

mathInlineWith :: String -> String -> MarkdownParser String
mathInlineWith op cl = try $ do
  string op
  notFollowedBy space
  words' <- many1Till (count 1 (noneOf "\n\\")
                   <|> (char '\\' >> anyChar >>= \c -> return ['\\',c])
                   <|> count 1 newline <* notFollowedBy' blankline
                       *> return " ")
              (try $ string cl)
  notFollowedBy digit  -- to prevent capture of $5
  return $ concat words'

-- to avoid performance problems, treat 4 or more _ or * or ~ or ^ in a row
-- as a literal rather than attempting to parse for emph/strong/strikeout/super/sub
fours :: Parser [Char] st (F Inlines)
fours = try $ do
  x <- char '*' <|> char '_' <|> char '~' <|> char '^'
  count 2 $ satisfy (==x)
  rest <- many1 (satisfy (==x))
  return $ return $ B.str (x:x:x:rest)

-- | Parses a list of inlines between start and end delimiters.
inlinesBetween :: (Show b)
               => MarkdownParser a
               -> MarkdownParser b
               -> MarkdownParser (F Inlines)
inlinesBetween start end =
  (trimInlinesF . mconcat) <$> try (start >> many1Till inner end)
    where inner      = innerSpace <|> (notFollowedBy' (() <$ whitespace) >> inline)
          innerSpace = try $ whitespace >>~ notFollowedBy' end

emph :: MarkdownParser (F Inlines)
emph = fmap B.emph <$> nested
  (inlinesBetween starStart starEnd <|> inlinesBetween ulStart ulEnd)
    where starStart = char '*' >> lookAhead nonspaceChar
          starEnd   = notFollowedBy' (() <$ strong) >> char '*'
          ulStart   = char '_' >> lookAhead nonspaceChar
          ulEnd     = notFollowedBy' (() <$ strong) >> char '_'

strong :: MarkdownParser (F Inlines)
strong = fmap B.strong <$> nested
  (inlinesBetween starStart starEnd <|> inlinesBetween ulStart ulEnd)
    where starStart = string "**" >> lookAhead nonspaceChar
          starEnd   = try $ string "**"
          ulStart   = string "__" >> lookAhead nonspaceChar
          ulEnd     = try $ string "__"

strikeout :: MarkdownParser (F Inlines)
strikeout = fmap B.strikeout <$>
 (guardEnabled Ext_strikeout >> inlinesBetween strikeStart strikeEnd)
    where strikeStart = string "~~" >> lookAhead nonspaceChar
                        >> notFollowedBy (char '~')
          strikeEnd   = try $ string "~~"

superscript :: MarkdownParser (F Inlines)
superscript = fmap B.superscript <$> try (do
  guardEnabled Ext_superscript
  char '^'
  mconcat <$> many1Till (notFollowedBy spaceChar >> inline) (char '^'))

subscript :: MarkdownParser (F Inlines)
subscript = fmap B.subscript <$> try (do
  guardEnabled Ext_subscript
  char '~'
  mconcat <$> many1Till (notFollowedBy spaceChar >> inline) (char '~'))

whitespace :: MarkdownParser (F Inlines)
whitespace = spaceChar >> return <$> (lb <|> regsp) <?> "whitespace"
  where lb = spaceChar >> skipMany spaceChar >> option B.space (endline >> return B.linebreak)
        regsp = skipMany spaceChar >> return B.space

nonEndline :: Parser [Char] st Char
nonEndline = satisfy (/='\n')

str :: MarkdownParser (F Inlines)
str = do
  isSmart <- readerSmart . stateOptions <$> getState
  a <- alphaNum
  as <- many $ alphaNum
            <|> (guardEnabled Ext_intraword_underscores >>
                 try (char '_' >>~ lookAhead alphaNum))
            <|> if isSmart
                   then (try $ satisfy (\c -> c == '\'' || c == '\x2019') >>
                         lookAhead alphaNum >> return '\x2019')
                         -- for things like l'aide
                   else mzero
  pos <- getPosition
  updateState $ \s -> s{ stateLastStrPos = Just pos }
  let result = a:as
  let spacesToNbr = map (\c -> if c == ' ' then '\160' else c)
  if isSmart
     then case likelyAbbrev result of
               []        -> return $ return $ B.str result
               xs        -> choice (map (\x ->
                               try (string x >> oneOf " \n" >>
                                    lookAhead alphaNum >>
                                    return (return $ B.str
                                                  $ result ++ spacesToNbr x ++ "\160"))) xs)
                           <|> (return $ return $ B.str result)
     else return $ return $ B.str result

-- | if the string matches the beginning of an abbreviation (before
-- the first period, return strings that would finish the abbreviation.
likelyAbbrev :: String -> [String]
likelyAbbrev x =
  let abbrevs = [ "Mr.", "Mrs.", "Ms.", "Capt.", "Dr.", "Prof.",
                  "Gen.", "Gov.", "e.g.", "i.e.", "Sgt.", "St.",
                  "vol.", "vs.", "Sen.", "Rep.", "Pres.", "Hon.",
                  "Rev.", "Ph.D.", "M.D.", "M.A.", "p.", "pp.",
                  "ch.", "sec.", "cf.", "cp."]
      abbrPairs = map (break (=='.')) abbrevs
  in  map snd $ filter (\(y,_) -> y == x) abbrPairs

-- an endline character that can be treated as a space, not a structural break
endline :: MarkdownParser (F Inlines)
endline = try $ do
  newline
  notFollowedBy blankline
  guardEnabled Ext_blank_before_blockquote <|> notFollowedBy emailBlockQuoteStart
  guardEnabled Ext_blank_before_header <|> notFollowedBy (char '#') -- atx header
  -- parse potential list-starts differently if in a list:
  st <- getState
  when (stateParserContext st == ListItemState) $ do
     notFollowedBy' bulletListStart
     notFollowedBy' anyOrderedListStart
  (guardEnabled Ext_hard_line_breaks >> return (return B.linebreak))
    <|> (return $ return B.space)

--
-- links
--

-- a reference label for a link
reference :: MarkdownParser (F Inlines, String)
reference = do notFollowedBy' (string "[^")   -- footnote reference
               withRaw $ trimInlinesF <$> inlinesInBalancedBrackets

-- source for a link, with optional title
source :: MarkdownParser (String, String)
source =
  (try $ charsInBalanced '(' ')' litChar >>= parseFromString source') <|>
  -- the following is needed for cases like:  [ref](/url(a).
  (enclosed (char '(') (char ')') litChar >>= parseFromString source')

-- auxiliary function for source
source' :: MarkdownParser (String, String)
source' = do
  skipSpaces
  let nl = char '\n' >>~ notFollowedBy blankline
  let sourceURL = liftM unwords $ many $ try $ do
                    notFollowedBy' linkTitle
                    skipMany spaceChar
                    optional nl
                    skipMany spaceChar
                    many1 $ escapedChar' <|> satisfy (not . isBlank)
  let betweenAngles = try $
         char '<' >> manyTill (escapedChar' <|> noneOf ">\n" <|> nl) (char '>')
  src <- try betweenAngles <|> sourceURL
  tit <- option "" linkTitle
  skipSpaces
  eof
  return (escapeURI $ trimr src, tit)

linkTitle :: MarkdownParser String
linkTitle = try $ do
  (many1 spaceChar >> option '\n' newline) <|> newline
  skipSpaces
  delim <- oneOf "'\""
  tit <-   manyTill litChar (try (char delim >> skipSpaces >> eof))
  return $ fromEntities tit

link :: MarkdownParser (F Inlines)
link = try $ do
  st <- getState
  guard $ stateAllowLinks st
  setState $ st{ stateAllowLinks = False }
  (lab,raw) <- reference
  setState $ st{ stateAllowLinks = True }
  regLink B.link lab <|> referenceLink B.link (lab,raw)

regLink :: (String -> String -> Inlines -> Inlines)
        -> F Inlines -> MarkdownParser (F Inlines)
regLink constructor lab = try $ do
  (src, tit) <- source
  return $ constructor src tit <$> lab

-- a link like [this][ref] or [this][] or [this]
referenceLink :: (String -> String -> Inlines -> Inlines)
              -> (F Inlines, String) -> MarkdownParser (F Inlines)
referenceLink constructor (lab, raw) = do
  (ref,raw') <- try (optional (char ' ') >>
                 optional (newline >> skipSpaces) >>
                 reference) <|> return (mempty, "")
  let labIsRef = raw' == "" || raw' == "[]"
  let key = toKey $ if labIsRef then raw else raw'
  let dropRB (']':xs) = xs
      dropRB xs = xs
  let dropLB ('[':xs) = xs
      dropLB xs = xs
  let dropBrackets = reverse . dropRB . reverse . dropLB
  fallback <- parseFromString (mconcat <$> many inline) $ dropBrackets raw
  implicitHeaderRefs <- option False $
                         True <$ guardEnabled Ext_implicit_header_references
  return $ do
    keys <- asksF stateKeys
    case M.lookup key keys of
       Nothing        -> do
         headers <- asksF stateHeaders
         let ref' = B.toList $ runF (if labIsRef then lab else ref)
                               defaultParserState
         if implicitHeaderRefs && ref' `elem` headers
            then do
              let src = '#' : uniqueIdent ref' []
              constructor src "" <$> lab
            else (\x -> B.str "[" <> x <> B.str "]" <> B.str raw') <$> fallback
       Just (src,tit) -> constructor src tit <$> lab

bareURL :: MarkdownParser (F Inlines)
bareURL = try $ do
  guardEnabled Ext_autolink_urls
  (orig, src) <- uri <|> emailAddress
  return $ return $ B.link src "" (B.str orig)

autoLink :: MarkdownParser (F Inlines)
autoLink = try $ do
  char '<'
  (orig, src) <- uri <|> emailAddress
  char '>'
  return $ return $ B.link src "" (B.str orig)

image :: MarkdownParser (F Inlines)
image = try $ do
  char '!'
  (lab,raw) <- reference
  regLink B.image lab <|> referenceLink B.image (lab,raw)

note :: MarkdownParser (F Inlines)
note = try $ do
  guardEnabled Ext_footnotes
  ref <- noteMarker
  return $ do
    notes <- asksF stateNotes'
    case lookup ref notes of
        Nothing       -> return $ B.str $ "[^" ++ ref ++ "]"
        Just contents -> do
          st <- askF
          -- process the note in a context that doesn't resolve
          -- notes, to avoid infinite looping with notes inside
          -- notes:
          let contents' = runF contents st{ stateNotes' = [] }
          return $ B.note contents'

inlineNote :: MarkdownParser (F Inlines)
inlineNote = try $ do
  guardEnabled Ext_inline_notes
  char '^'
  contents <- inlinesInBalancedBrackets
  return $ B.note . B.para <$> contents

rawLaTeXInline' :: MarkdownParser (F Inlines)
rawLaTeXInline' = try $ do
  guardEnabled Ext_raw_tex
  lookAhead $ char '\\' >> notFollowedBy' (string "start") -- context env
  RawInline _ s <- rawLaTeXInline
  return $ return $ B.rawInline "tex" s
  -- "tex" because it might be context or latex

rawConTeXtEnvironment :: Parser [Char] st String
rawConTeXtEnvironment = try $ do
  string "\\start"
  completion <- inBrackets (letter <|> digit <|> spaceChar)
               <|> (many1 letter)
  contents <- manyTill (rawConTeXtEnvironment <|> (count 1 anyChar))
                       (try $ string "\\stop" >> string completion)
  return $ "\\start" ++ completion ++ concat contents ++ "\\stop" ++ completion

inBrackets :: (Parser [Char] st Char) -> Parser [Char] st String
inBrackets parser = do
  char '['
  contents <- many parser
  char ']'
  return $ "[" ++ contents ++ "]"

rawHtmlInline :: MarkdownParser (F Inlines)
rawHtmlInline = do
  guardEnabled Ext_raw_html
  mdInHtml <- option False $
                guardEnabled Ext_markdown_in_html_blocks >> return True
  (_,result) <- if mdInHtml
                   then htmlTag isInlineTag
                   else htmlTag (not . isTextTag)
  return $ return $ B.rawInline "html" result

-- Citations

cite :: MarkdownParser (F Inlines)
cite = do
  guardEnabled Ext_citations
  getOption readerReferences >>= guard . not . null
  citations <- textualCite <|> normalCite
  return $ flip B.cite mempty <$> citations

textualCite :: MarkdownParser (F [Citation])
textualCite = try $ do
  (_, key) <- citeKey
  let first = Citation{ citationId      = key
                      , citationPrefix  = []
                      , citationSuffix  = []
                      , citationMode    = AuthorInText
                      , citationNoteNum = 0
                      , citationHash    = 0
                      }
  mbrest <- option Nothing $ try $ spnl >> Just <$> normalCite
  case mbrest of
       Just rest  -> return $ (first:) <$> rest
       Nothing    -> option (return [first]) $ bareloc first

bareloc :: Citation -> MarkdownParser (F [Citation])
bareloc c = try $ do
  spnl
  char '['
  suff <- suffix
  rest <- option (return []) $ try $ char ';' >> citeList
  spnl
  char ']'
  return $ do
    suff' <- suff
    rest' <- rest
    return $ c{ citationSuffix = B.toList suff' } : rest'

normalCite :: MarkdownParser (F [Citation])
normalCite = try $ do
  char '['
  spnl
  citations <- citeList
  spnl
  char ']'
  return citations

citeKey :: MarkdownParser (Bool, String)
citeKey = try $ do
  suppress_author <- option False (char '-' >> return True)
  char '@'
  first <- letter
  let internal p = try $ p >>~ lookAhead (letter <|> digit)
  rest <- many $ letter <|> digit <|> internal (oneOf ":.#$%&-_?<>~")
  let key = first:rest
  citations' <- map CSL.refId <$> getOption readerReferences
  guard $ key `elem` citations'
  return (suppress_author, key)

suffix :: MarkdownParser (F Inlines)
suffix = try $ do
  hasSpace <- option False (notFollowedBy nonspaceChar >> return True)
  spnl
  rest <- trimInlinesF . mconcat <$> many (notFollowedBy (oneOf ";]") >> inline)
  return $ if hasSpace
              then (B.space <>) <$> rest
              else rest

prefix :: MarkdownParser (F Inlines)
prefix = trimInlinesF . mconcat <$>
  manyTill inline (char ']' <|> liftM (const ']') (lookAhead citeKey))

citeList :: MarkdownParser (F [Citation])
citeList = fmap sequence $ sepBy1 citation (try $ char ';' >> spnl)

citation :: MarkdownParser (F Citation)
citation = try $ do
  pref <- prefix
  (suppress_author, key) <- citeKey
  suff <- suffix
  return $ do
    x <- pref
    y <- suff
    return $ Citation{ citationId      = key
                     , citationPrefix  = B.toList x
                     , citationSuffix  = B.toList y
                     , citationMode    = if suppress_author
                                            then SuppressAuthor
                                            else NormalCitation
                     , citationNoteNum = 0
                     , citationHash    = 0
                     }

smart :: MarkdownParser (F Inlines)
smart = do
  getOption readerSmart >>= guard
  doubleQuoted <|> singleQuoted <|>
    choice (map (return . B.singleton <$>) [apostrophe, dash, ellipses])

singleQuoted :: MarkdownParser (F Inlines)
singleQuoted = try $ do
  singleQuoteStart
  withQuoteContext InSingleQuote $
    fmap B.singleQuoted . trimInlinesF . mconcat <$>
      many1Till inline singleQuoteEnd

doubleQuoted :: MarkdownParser (F Inlines)
doubleQuoted = try $ do
  doubleQuoteStart
  withQuoteContext InDoubleQuote $
    fmap B.doubleQuoted . trimInlinesF . mconcat <$>
      many1Till inline doubleQuoteEnd