aboutsummaryrefslogtreecommitdiff
path: root/src/Text/Pandoc/Readers/Markdown.hs
blob: 8b34b9b4a88d7d4a42bf6a81ea615b8a134a70cc (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
{-# 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 ) where

import Data.List ( transpose, sortBy, findIndex, intercalate )
import qualified Data.Map as M
import Data.Ord ( comparing )
import Data.Char ( isAlphaNum )
import Data.Maybe
import Text.Pandoc.Definition
import Text.Pandoc.Generic
import Text.Pandoc.Options
import Text.Pandoc.Shared
import Text.Pandoc.Parsing
import Text.Pandoc.Readers.LaTeX ( rawLaTeXInline, rawLaTeXBlock )
import Text.Pandoc.Readers.HTML ( htmlTag, htmlInBalanced, isInlineTag, isBlockTag,
                                  isTextTag, isCommentTag )
import Text.Pandoc.XML ( fromEntities )
import Control.Monad (when, liftM, guard, mzero)
import Text.HTML.TagSoup
import Text.HTML.TagSoup.Match (tagOpen)

-- | 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")

--
-- 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 :: [Char]
setextHChars = "=-"

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

--
-- auxiliary functions
--

indentSpaces :: Parser [Char] ParserState [Char]
indentSpaces = try $ do
  tabStop <- getOption readerTabStop
  count tabStop (char ' ') <|>
    string "\t" <?> "indentation"

nonindentSpaces :: Parser [Char] ParserState [Char]
nonindentSpaces = do
  tabStop <- getOption readerTabStop
  sps <- many (char ' ')
  if length sps < tabStop
     then return sps
     else unexpected "indented line"

skipNonindentSpaces :: Parser [Char] ParserState ()
skipNonindentSpaces = do
  tabStop <- getOption readerTabStop
  atMostSpaces (tabStop - 1)

atMostSpaces :: Int -> Parser [Char] ParserState ()
atMostSpaces 0 = notFollowedBy (char ' ')
atMostSpaces n = (char ' ' >> atMostSpaces (n-1)) <|> return ()

litChar :: Parser [Char] ParserState 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 :: Parser [Char] ParserState Inline
                          -> Parser [Char] ParserState [Inline]
inlinesInBalancedBrackets parser = try $ do
  char '['
  result <- manyTill ( (do lookAhead $ try $ do (Str res) <- parser
                                                guard (res == "[")
                           bal <- inlinesInBalancedBrackets parser
                           return $ [Str "["] ++ bal ++ [Str "]"])
                       <|> (count 1 parser))
                     (char ']')
  return $ concat result

--
-- document structure
--

titleLine :: Parser [Char] ParserState [Inline]
titleLine = try $ do
  char '%'
  skipSpaces
  res <- many $ (notFollowedBy newline >> inline)
             <|> try (endline >> whitespace)
  newline
  return $ normalizeSpaces res

authorsLine :: Parser [Char] ParserState [[Inline]]
authorsLine = try $ do
  char '%'
  skipSpaces
  authors <- sepEndBy (many (notFollowedBy (satisfy $ \c ->
                                c == ';' || c == '\n') >> inline))
                       (char ';' <|>
                        try (newline >> notFollowedBy blankline >> spaceChar))
  newline
  return $ filter (not . null) $ map normalizeSpaces authors

dateLine :: Parser [Char] ParserState [Inline]
dateLine = try $ do
  char '%'
  skipSpaces
  date <- manyTill inline newline
  return $ normalizeSpaces date

titleBlock :: Parser [Char] ParserState ([Inline], [[Inline]], [Inline])
titleBlock = try $ do
  guardEnabled Ext_pandoc_title_blocks
  title <- option [] titleLine
  author <- option [] authorsLine
  date <- option [] dateLine
  optional blanklines
  return (title, author, date)

parseMarkdown :: Parser [Char] ParserState Pandoc
parseMarkdown = do
  -- markdown allows raw HTML
  updateState $ \state -> state { stateOptions =
                let oldOpts = stateOptions state in
                    oldOpts{ readerParseRaw = True } }
  startPos <- getPosition
  -- go through once just to get list of reference keys and notes
  -- docMinusKeys is the raw document with blanks where the keys/notes were...
  let firstPassParser = referenceKey
                     <|> (guardEnabled Ext_footnotes >> noteBlock)
                     <|> (guardEnabled Ext_delimited_code_blocks >>
                          liftM snd (withRaw codeBlockDelimited))
                     <|> lineClump
  docMinusKeys <- liftM concat $ manyTill firstPassParser eof
  setInput docMinusKeys
  setPosition startPos
  st' <- getState
  let reversedNotes = stateNotes st'
  updateState $ \s -> s { stateNotes = reverse reversedNotes }
  -- now parse it for real...
  (title, author, date) <- option ([],[],[]) titleBlock
  blocks <- parseBlocks
  let doc = Pandoc (Meta title author date) $ filter (/= Null) blocks
  -- if there are labeled examples, change references into numbers
  examples <- liftM stateExamples getState
  let handleExampleRef :: Inline -> Inline
      handleExampleRef z@(Str ('@':xs)) =
        case M.lookup xs examples of
              Just n     -> Str (show n)
              Nothing    -> z
      handleExampleRef z = z
  if M.null examples
     then return doc
     else return $ bottomUp handleExampleRef doc

--
-- initial pass for references and notes
--

referenceKey :: Parser [Char] ParserState [Char]
referenceKey = try $ do
  startPos <- getPosition
  skipNonindentSpaces
  lab <- 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
  endPos <- getPosition
  let target = (escapeURI $ removeTrailingSpace src,  tit)
  st <- getState
  let oldkeys = stateKeys st
  updateState $ \s -> s { stateKeys = M.insert (toKey lab) target oldkeys }
  -- return blanks so line count isn't affected
  return $ replicate (sourceLine endPos - sourceLine startPos) '\n'

referenceTitle :: Parser [Char] ParserState 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

noteMarker :: Parser [Char] ParserState [Char]
noteMarker = string "[^" >> many1Till (satisfy $ not . isBlank) (char ']')

rawLine :: Parser [Char] ParserState [Char]
rawLine = try $ do
  notFollowedBy blankline
  notFollowedBy' $ try $ skipNonindentSpaces >> noteMarker
  optional indentSpaces
  anyLine

rawLines :: Parser [Char] ParserState [Char]
rawLines = do
  first <- anyLine
  rest <- many rawLine
  return $ unlines (first:rest)

noteBlock :: Parser [Char] ParserState [Char]
noteBlock = try $ do
  startPos <- getPosition
  skipNonindentSpaces
  ref <- noteMarker
  char ':'
  optional blankline
  optional indentSpaces
  raw <- sepBy rawLines
             (try (blankline >> indentSpaces >>
                   notFollowedBy blankline))
  optional blanklines
  endPos <- getPosition
  let newnote = (ref, (intercalate "\n" raw) ++ "\n\n")
  st <- getState
  let oldnotes = stateNotes st
  updateState $ \s -> s { stateNotes = newnote : oldnotes }
  -- return blanks so line count isn't affected
  return $ replicate (sourceLine endPos - sourceLine startPos) '\n'

--
-- parsing blocks
--

parseBlocks :: Parser [Char] ParserState [Block]
parseBlocks = manyTill block eof

block :: Parser [Char] ParserState Block
block = choice [ codeBlockDelimited
               , guardEnabled Ext_latex_macros >> macro
               , header
               , table
               , codeBlockIndented
               , lhsCodeBlock
               , blockQuote
               , hrule
               , bulletList
               , orderedList
               , definitionList
               , rawTeXBlock
               , para
               , htmlBlock
               , plain
               , nullBlock ] <?> "block"

--
-- header blocks
--

header :: Parser [Char] ParserState Block
header = setextHeader <|> atxHeader <?> "header"

atxHeader :: Parser [Char] ParserState Block
atxHeader = try $ do
  level <- many1 (char '#') >>= return . length
  notFollowedBy (char '.' <|> char ')') -- this would be a list
  skipSpaces
  text <- manyTill inline atxClosing >>= return . normalizeSpaces
  return $ Header level text

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

setextHeader :: Parser [Char] ParserState Block
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 <- many1Till inline newline
  underlineChar <- oneOf setextHChars
  many (char underlineChar)
  blanklines
  let level = (fromMaybe 0 $ findIndex (== underlineChar) setextHChars) + 1
  return $ Header level (normalizeSpaces text)

--
-- hrule block
--

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

--
-- code blocks
--

indentedLine :: Parser [Char] ParserState [Char]
indentedLine = indentSpaces >> manyTill anyChar newline >>= return . (++ "\n")

blockDelimiter :: (Char -> Bool)
               -> Maybe Int
               -> Parser [Char] st (Int, (String, [String], [(String, String)]), Char)
blockDelimiter f len = try $ do
  c <- lookAhead (satisfy f)
  size <- 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
  many spaceChar
  attr <- option ([],[],[])
          $ attributes                                     -- ~~~ {.ruby}
         <|> (many1 alphaNum >>= \x -> return ([],[x],[])) -- github variant ```ruby
  blankline
  return (size, attr, c)

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

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

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

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

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

codeBlockDelimited :: Parser [Char] ParserState Block
codeBlockDelimited = try $ do
  guardEnabled Ext_delimited_code_blocks
  (size, attr, c) <- blockDelimiter (\c -> c == '~' || c == '`') Nothing
  contents <- manyTill anyLine (blockDelimiter (== c) (Just size))
  blanklines
  return $ CodeBlock attr $ intercalate "\n" contents

codeBlockIndented :: Parser [Char] ParserState Block
codeBlockIndented = do
  contents <- many1 (indentedLine <|>
                     try (do b <- blanklines
                             l <- indentedLine
                             return $ b ++ l))
  optional blanklines
  classes <- getOption readerIndentedCodeClasses
  return $ CodeBlock ("", classes, []) $
           stripTrailingNewlines $ concat contents

lhsCodeBlock :: Parser [Char] ParserState Block
lhsCodeBlock = do
  failUnlessLHS
  liftM (CodeBlock ("",["sourceCode","literate","haskell"],[]))
          (lhsCodeBlockBird <|> lhsCodeBlockLaTeX)
    <|> liftM (CodeBlock ("",["sourceCode","haskell"],[]))
          lhsCodeBlockInverseBird

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

lhsCodeBlockBird :: Parser [Char] ParserState String
lhsCodeBlockBird = lhsCodeBlockBirdWith '>'

lhsCodeBlockInverseBird :: Parser [Char] ParserState String
lhsCodeBlockInverseBird = lhsCodeBlockBirdWith '<'

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


--
-- block quotes
--

emailBlockQuoteStart :: Parser [Char] ParserState Char
emailBlockQuoteStart = try $ skipNonindentSpaces >> char '>' >>~ optional (char ' ')

emailBlockQuote :: Parser [Char] ParserState [[Char]]
emailBlockQuote = try $ do
  emailBlockQuoteStart
  raw <- sepBy (many (nonEndline <|>
                      (try (endline >> notFollowedBy emailBlockQuoteStart >>
                       return '\n'))))
               (try (newline >> emailBlockQuoteStart))
  newline <|> (eof >> return '\n')
  optional blanklines
  return raw

blockQuote :: Parser [Char] ParserState Block
blockQuote = do
  raw <- emailBlockQuote
  -- parse the extracted block, which may contain various block elements:
  contents <- parseFromString parseBlocks $ (intercalate "\n" raw) ++ "\n\n"
  return $ BlockQuote contents

--
-- list blocks
--

bulletListStart :: Parser [Char] ParserState ()
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 :: Parser [Char] ParserState (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
  state <- getState
  if readerStrict (stateOptions state)
     then do many1 digit
             char '.'
             spaceChar
             return (1, DefaultStyle, DefaultDelim)
     else 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 :: Parser [Char] ParserState ()
listStart = bulletListStart <|> (anyOrderedListStart >> return ())

-- parse a line of a list item (start = parser for beginning of list item)
listLine :: Parser [Char] ParserState [Char]
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 :: Parser [Char] ParserState a
            -> Parser [Char] ParserState [Char]
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 :: Parser [Char] ParserState [Char]
listContinuation = try $ do
  lookAhead indentSpaces
  result <- many1 listContinuationLine
  blanks <- many blankline
  return $ concat result ++ blanks

listContinuationLine :: Parser [Char] ParserState [Char]
listContinuationLine = try $ do
  notFollowedBy blankline
  notFollowedBy' listStart
  optional indentSpaces
  result <- manyTill anyChar newline
  return $ result ++ "\n"

listItem :: Parser [Char] ParserState a
         -> Parser [Char] ParserState [Block]
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 :: Parser [Char] ParserState Block
orderedList = try $ do
  (start, style, delim) <- lookAhead anyOrderedListStart
  items <- many1 $ listItem $ try $
             do optional newline -- if preceded by a Plain block in a list context
                skipNonindentSpaces
                orderedListMarker style delim
  return $ OrderedList (start, style, delim) $ compactify items

bulletList :: Parser [Char] ParserState Block
bulletList =
  many1 (listItem bulletListStart) >>= return . BulletList . compactify

-- definition lists

defListMarker :: Parser [Char] ParserState ()
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 :: Parser [Char] ParserState ([Inline], [[Block]])
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 <- 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 ((normalizeSpaces term), contents)

defRawBlock :: Parser [Char] ParserState [Char]
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 :: Parser [Char] ParserState Block
definitionList = do
  items <- many1 definitionListItem
  -- "compactify" the definition list:
  let defs = map snd items
  let defBlocks = reverse $ concat $ concat defs
  let isPara (Para _) = True
      isPara _        = False
  let items' = case take 1 defBlocks of
                [Para x]   -> if not $ any isPara (drop 1 defBlocks)
                                 then let (t,ds) = last items
                                          lastDef = last ds
                                          ds' = init ds ++
                                                [init lastDef ++ [Plain x]]
                                       in init items ++ [(t, ds')]
                                 else items
                _          -> items
  return $ DefinitionList items'

--
-- paragraph block
--

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

para :: Parser [Char] ParserState Block
para = try $ do
  result <- liftM normalizeSpaces $ many1 inline
  guard $ not . all isHtmlOrBlank $ result
  option (Plain result) $ try $ do
              newline
              (blanklines >> return Null)
                <|> (guardDisabled Ext_blank_before_blockquote >> lookAhead blockQuote)
                <|> (guardDisabled Ext_blank_before_header >> lookAhead header)
              return $ Para result

plain :: Parser [Char] ParserState Block
plain = many1 inline >>~ spaces >>= return . Plain . normalizeSpaces

--
-- raw html
--

htmlElement :: Parser [Char] ParserState [Char]
htmlElement = strictHtmlBlock <|> liftM snd (htmlTag isBlockTag)

htmlBlock :: Parser [Char] ParserState Block
htmlBlock = RawBlock "html" `fmap`
  ((guardEnabled Ext_markdown_in_html_blocks >> rawHtmlBlocks)
     <|> htmlBlock')

htmlBlock' :: Parser [Char] ParserState String
htmlBlock' = try $ do
    first <- htmlElement
    finalSpace <- many spaceChar
    finalNewlines <- many newline
    return $ first ++ finalSpace ++ finalNewlines

strictHtmlBlock :: Parser [Char] ParserState [Char]
strictHtmlBlock = htmlInBalanced (not . isInlineTag)

rawVerbatimBlock :: Parser [Char] ParserState 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 :: Parser [Char] ParserState Block
rawTeXBlock = do
  guardEnabled Ext_raw_tex
  result <- liftM (RawBlock "latex") rawLaTeXBlock
          <|> liftM (RawBlock "context") rawConTeXtEnvironment
  spaces
  return result

rawHtmlBlocks :: Parser [Char] ParserState String
rawHtmlBlocks = do
  htmlBlocks <- many1 $ do blk <- rawVerbatimBlock <|>
                                   liftM snd (htmlTag isBlockTag)
                           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 $ blk ++ sps
  let combined = concat htmlBlocks
  return $ if last combined == '\n' then init combined else combined

--
-- 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
                  -> Parser [Char] ParserState ([[Block]], [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 <- mapM (parseFromString (many plain)) $
             map removeLeadingTrailingSpace rawHeads'
  return (heads, aligns, indices)

-- Parse a table footer - dashed lines followed by blank line.
tableFooter :: Parser [Char] ParserState [Char]
tableFooter = try $ skipNonindentSpaces >> many1 (dashedLine '-') >> blanklines

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

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

-- Parse a table line and return a list of lists of blocks (columns).
tableLine :: [Int]
          -> Parser [Char] ParserState [[Block]]
tableLine indices = rawTableLine indices >>= mapM (parseFromString (many plain))

-- Parse a multiline table row and return a list of blocks (columns).
multilineRow :: [Int]
             -> Parser [Char] ParserState [[Block]]
multilineRow indices = do
  colLines <- many1 (rawTableLine indices)
  let cols = map unlines $ transpose colLines
  mapM (parseFromString (many plain)) cols

-- Parses a table caption:  inlines beginning with 'Table:'
-- and followed by blank lines.
tableCaption :: Parser [Char] ParserState [Inline]
tableCaption = try $ do
  guardEnabled Ext_table_captions
  skipNonindentSpaces
  string ":" <|> string "Table:"
  result <- many1 inline
  blanklines
  return $ normalizeSpaces result

-- Parse a simple table with '---' header and one line per row.
simpleTable :: Bool  -- ^ Headerless table
            -> Parser [Char] ParserState Block
simpleTable headless = do
  Table c a _w h l <- 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 $ Table c a (replicate (length a) 0) h l

-- 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
               -> Parser [Char] ParserState Block
multilineTable headless =
  tableWith (multilineTableHeader headless) multilineRow blanklines tableFooter

multilineTableHeader :: Bool -- ^ Headerless table
                     -> Parser [Char] ParserState ([[Block]], [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 <- mapM (parseFromString (many plain)) $
             map removeLeadingTrailingSpace 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 removeTrailingSpace 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

gridTable :: Bool -- ^ Headerless table
          -> Parser [Char] ParserState Block
gridTable = gridTableWith block

pipeTable :: Bool -- ^ Headerless table
           -> Parser [Char] ParserState Block
pipeTable headless = tableWith (pipeTableHeader headless)
   (\_ -> pipeTableRow) (return ()) blanklines

-- | Parse header for an pipe table.
pipeTableHeader :: Bool -- ^ Headerless table
                 -> Parser [Char] ParserState ([[Block]], [Alignment], [Int])
pipeTableHeader headless = do
  try $ do
    heads <- if headless
                then return $ repeat []
                else pipeTableRow
    aligns <- nonindentSpaces >> optional (char '|') >>
               pipeTableHeaderPart `sepBy1` sepPipe
    optional (char '|')
    newline
    let cols = length aligns
    return (take cols heads, aligns, [])

sepPipe :: Parser [Char] ParserState ()
sepPipe = try $ char '|' >> notFollowedBy blankline

pipeTableRow :: Parser [Char] ParserState [[Block]]
pipeTableRow = do
  nonindentSpaces
  optional (char '|')
  let cell = many (notFollowedBy (blankline <|> char '|') >> inline)
  first <- cell
  sepPipe
  rest <- cell `sepBy1` sepPipe
  optional (char '|')
  blankline
  return $ map (\ils ->
     if null ils
        then []
        else [Plain $ normalizeSpaces ils]) (first:rest)

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 ()

table :: Parser [Char] ParserState Block
table = try $ do
  frontCaption <- option [] tableCaption
  Table _ aligns widths heads lines' <-
           try (guardEnabled Ext_pipe_tables >> scanForPipe >>
                 (pipeTable True <|> pipeTable False)) <|>
           try (guardEnabled Ext_multiline_tables >>
                (multilineTable False <|> simpleTable True)) <|>
           try (guardEnabled Ext_simple_tables >>
                (simpleTable False <|> multilineTable True)) <|>
           try (guardEnabled Ext_grid_tables >>
                (gridTable False <|> gridTable True)) <?> "table"
  optional blanklines
  caption <- if null frontCaption
                then option [] tableCaption
                else return frontCaption
  return $ Table caption aligns widths heads lines'

--
-- inline
--

inline :: Parser [Char] ParserState Inline
inline = choice inlineParsers <?> "inline"

inlineParsers :: [Parser [Char] ParserState Inline]
inlineParsers = [ whitespace
                , str
                , endline
                , code
                , fours
                , strong
                , emph
                , note
                , link
                , cite
                , image
                , math
                , strikeout
                , superscript
                , subscript
                , inlineNote  -- after superscript because of ^[link](/foo)^
                , autoLink
                , rawHtmlInline
                , escapedChar
                , rawLaTeXInline'
                , exampleRef
                , smartPunctuation inline
                , charRef
                , symbol
                , ltSign ]

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

escapedChar :: Parser [Char] ParserState Inline
escapedChar = do
  result <- escapedChar'
  case result of
       ' '   -> return $ Str "\160" -- "\ " is a nonbreaking space
       '\n'  -> guardEnabled Ext_escaped_line_breaks >>
                return LineBreak  -- "\[newline]" is a linebreak
       _     -> return $ Str [result]

ltSign :: Parser [Char] ParserState Inline
ltSign = do
  guardDisabled Ext_markdown_in_html_blocks
    <|> (notFollowedBy' rawHtmlBlocks >> return ())
  char '<'
  return $ Str ['<']

exampleRef :: Parser [Char] ParserState Inline
exampleRef = try $ do
  char '@'
  lab <- many1 (alphaNum <|> oneOf "-_")
  -- We just return a Str. These are replaced with numbers
  -- later. See the end of parseMarkdown.
  return $ Str $ '@' : lab

symbol :: Parser [Char] ParserState Inline
symbol = do
  result <- noneOf "<\\\n\t "
         <|> try (do lookAhead $ char '\\'
                     notFollowedBy' rawTeXBlock
                     char '\\')
  return $ Str [result]

-- parses inline code, between n `s and n `s
code :: Parser [Char] ParserState Inline
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 $ Code attr $ removeLeadingTrailingSpace $ concat result

mathWord :: Parser [Char] st [Char]
mathWord = liftM concat $ many1 mathChunk

mathChunk :: Parser [Char] st [Char]
mathChunk = do char '\\'
               c <- anyChar
               return ['\\',c]
        <|> many1 (satisfy $ \c -> not (isBlank c || c == '\\' || c == '$'))

math :: Parser [Char] ParserState Inline
math =  (mathDisplay >>= applyMacros' >>= return . Math DisplayMath)
     <|> (mathInline >>= applyMacros' >>= return . Math InlineMath)

mathDisplay :: Parser [Char] ParserState String
mathDisplay = try $ do
  guardEnabled Ext_tex_math
  string "$$"
  many1Till (noneOf "\n" <|> (newline >>~ notFollowedBy' blankline)) (try $ string "$$")

mathInline :: Parser [Char] ParserState String
mathInline = try $ do
  guardEnabled Ext_tex_math
  char '$'
  notFollowedBy space
  words' <- sepBy1 mathWord (many1 (spaceChar <|> (newline >>~ notFollowedBy' blankline)))
  char '$'
  notFollowedBy digit
  return $ intercalate " " 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 Inline
fours = try $ do
  x <- char '*' <|> char '_' <|> char '~' <|> char '^'
  count 2 $ satisfy (==x)
  rest <- many1 (satisfy (==x))
  return $ Str (x:x:x:rest)

-- | Parses a list of inlines between start and end delimiters.
inlinesBetween :: (Show b)
               => Parser [Char] ParserState a
               -> Parser [Char] ParserState b
               -> Parser [Char] ParserState [Inline]
inlinesBetween start end =
  normalizeSpaces `liftM` try (start >> many1Till inner end)
    where inner      = innerSpace <|> (notFollowedBy' whitespace >> inline)
          innerSpace = try $ whitespace >>~ notFollowedBy' end

-- This is used to prevent exponential blowups for things like:
-- a**a*a**a*a**a*a**a*a**a*a**a*a**
nested :: Parser [Char] ParserState a
       -> Parser [Char] ParserState a
nested p = do
  nestlevel <- stateMaxNestingLevel `fmap` getState
  guard $ nestlevel > 0
  updateState $ \st -> st{ stateMaxNestingLevel = stateMaxNestingLevel st - 1 }
  res <- p
  updateState $ \st -> st{ stateMaxNestingLevel = nestlevel }
  return res

emph :: Parser [Char] ParserState Inline
emph = Emph `fmap` 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 :: Parser [Char] ParserState Inline
strong = Strong `liftM` nested
  (inlinesBetween starStart starEnd <|> inlinesBetween ulStart ulEnd)
    where starStart = string "**" >> lookAhead nonspaceChar
          starEnd   = try $ string "**"
          ulStart   = string "__" >> lookAhead nonspaceChar
          ulEnd     = try $ string "__"

strikeout :: Parser [Char] ParserState Inline
strikeout = Strikeout `liftM`
 (guardEnabled Ext_strikeout >> inlinesBetween strikeStart strikeEnd)
    where strikeStart = string "~~" >> lookAhead nonspaceChar
                        >> notFollowedBy (char '~')
          strikeEnd   = try $ string "~~"

superscript :: Parser [Char] ParserState Inline
superscript = guardEnabled Ext_superscript >> enclosed (char '^') (char '^')
              (notFollowedBy spaceChar >> inline) >>= -- may not contain Space
              return . Superscript

subscript :: Parser [Char] ParserState Inline
subscript = guardEnabled Ext_subscript >> enclosed (char '~') (char '~')
            (notFollowedBy spaceChar >> inline) >>=  -- may not contain Space
            return . Subscript

whitespace :: Parser [Char] ParserState Inline
whitespace = spaceChar >>
  (   (spaceChar >> skipMany spaceChar >> option Space (endline >> return LineBreak))
  <|> (skipMany spaceChar >> return Space) ) <?> "whitespace"

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

str :: Parser [Char] ParserState Inline
str = do
  smart <- (readerSmart . stateOptions) `fmap` getState
  a <- alphaNum
  as <- many $ alphaNum
            <|> (guardEnabled Ext_intraword_underscores >>
                 try (char '_' >>~ lookAhead alphaNum))
            <|> if smart
                   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 smart
     then case likelyAbbrev result of
               []        -> return $ Str result
               xs        -> choice (map (\x ->
                               try (string x >> oneOf " \n" >>
                                    lookAhead alphaNum >>
                                    return (Str $ result ++ spacesToNbr x ++ "\160"))) xs)
                           <|> (return $ Str result)
     else return $ 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 :: Parser [Char] ParserState Inline
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
  return Space

--
-- links
--

-- a reference label for a link
reference :: Parser [Char] ParserState [Inline]
reference = do notFollowedBy' (string "[^")   -- footnote reference
               result <- inlinesInBalancedBrackets inline
               return $ normalizeSpaces result

-- source for a link, with optional title
source :: Parser [Char] ParserState (String, [Char])
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' :: Parser [Char] ParserState (String, [Char])
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 $ removeTrailingSpace src, tit)

linkTitle :: Parser [Char] ParserState 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 :: Parser [Char] ParserState Inline
link = try $ do
  lab <- reference
  (src, tit) <- source <|> referenceLink lab
  return $ Link (delinkify lab) (src, tit)

delinkify :: [Inline] -> [Inline]
delinkify = bottomUp $ concatMap go
  where go (Link lab _) = lab
        go x            = [x]

-- a link like [this][ref] or [this][] or [this]
referenceLink :: [Inline]
              -> Parser [Char] ParserState (String, [Char])
referenceLink lab = do
  ref <- option [] (try (optional (char ' ') >>
                         optional (newline >> skipSpaces) >> reference))
  let ref' = if null ref then lab else ref
  state <- getState
  case lookupKeySrc (stateKeys state) (toKey ref') of
     Nothing     -> fail "no corresponding key"
     Just target -> return target

autoLink :: Parser [Char] ParserState Inline
autoLink = try $ do
  char '<'
  (orig, src) <- uri <|> emailAddress
  char '>'
  (guardEnabled Ext_autolink_code_spans >>
       return (Link [Code ("",["url"],[]) orig] (src, "")))
    <|> return (Link [Str orig] (src, ""))

image :: Parser [Char] ParserState Inline
image = try $ do
  char '!'
  lab <- reference
  (src, tit) <- source <|> referenceLink lab
  return $ Image lab (src,tit)

note :: Parser [Char] ParserState Inline
note = try $ do
  guardEnabled Ext_footnotes
  ref <- noteMarker
  state <- getState
  let notes = stateNotes state
  case lookup ref notes of
    Nothing   -> fail "note not found"
    Just raw  -> do
       -- We temporarily empty the note list while parsing the note,
       -- so that we don't get infinite loops with notes inside notes...
       -- Note references inside other notes do not work.
       updateState $ \st -> st{ stateNotes = [] }
       contents <- parseFromString parseBlocks raw
       updateState $ \st -> st{ stateNotes = notes }
       return $ Note contents

inlineNote :: Parser [Char] ParserState Inline
inlineNote = try $ do
  guardEnabled Ext_inline_notes
  char '^'
  contents <- inlinesInBalancedBrackets inline
  return $ Note [Para contents]

rawLaTeXInline' :: Parser [Char] ParserState Inline
rawLaTeXInline' = try $ do
  guardEnabled Ext_raw_tex
  lookAhead $ char '\\' >> notFollowedBy' (string "start") -- context env
  RawInline _ s <- rawLaTeXInline
  return $ 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 :: Parser [Char] ParserState Inline
rawHtmlInline = do
  mdInHtml <- option False $
                guardEnabled Ext_markdown_in_html_blocks >> return True
  (_,result) <- if mdInHtml
                   then htmlTag isInlineTag
                   else htmlTag (not . isTextTag)
  return $ RawInline "html" result

-- Citations

cite :: Parser [Char] ParserState Inline
cite = do
  guardEnabled Ext_citations
  citations <- textualCite <|> normalCite
  return $ Cite citations []

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

textualCite :: Parser [Char] ParserState [Citation]
textualCite = try $ do
  (_, key) <- citeKey
  let first = Citation{ citationId      = key
                      , citationPrefix  = []
                      , citationSuffix  = []
                      , citationMode    = AuthorInText
                      , citationNoteNum = 0
                      , citationHash    = 0
                      }
  rest <- option [] $ try $ spnl >> normalCite
  if null rest
     then option [first] $ bareloc first
     else return $ first : rest

bareloc :: Citation -> Parser [Char] ParserState [Citation]
bareloc c = try $ do
  spnl
  char '['
  suff <- suffix
  rest <- option [] $ try $ char ';' >> citeList
  spnl
  char ']'
  return $ c{ citationSuffix = suff } : rest

normalCite :: Parser [Char] ParserState [Citation]
normalCite = try $ do
  char '['
  spnl
  citations <- citeList
  spnl
  char ']'
  return citations

citeKey :: Parser [Char] ParserState (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' <- getOption readerCitations
  guard $ key `elem` citations'
  return (suppress_author, key)

suffix :: Parser [Char] ParserState [Inline]
suffix = try $ do
  hasSpace <- option False (notFollowedBy nonspaceChar >> return True)
  spnl
  rest <- liftM normalizeSpaces $ many $ notFollowedBy (oneOf ";]") >> inline
  return $ if hasSpace
              then Space : rest
              else rest

prefix :: Parser [Char] ParserState [Inline]
prefix = liftM normalizeSpaces $
  manyTill inline (char ']' <|> liftM (const ']') (lookAhead citeKey))

citeList :: Parser [Char] ParserState [Citation]
citeList = sepBy1 citation (try $ char ';' >> spnl)

citation :: Parser [Char] ParserState Citation
citation = try $ do
  pref <- prefix
  (suppress_author, key) <- citeKey
  suff <- suffix
  return $ Citation{ citationId        = key
                     , citationPrefix  = pref
                     , citationSuffix  = suff
                     , citationMode    = if suppress_author
                                            then SuppressAuthor
                                            else NormalCitation
                     , citationNoteNum = 0
                     , citationHash    = 0
                     }