aboutsummaryrefslogtreecommitdiff
path: root/src/Text/Pandoc/Readers/LaTeX.hs
blob: 986547f4249b3846380b6ac0e4a3e1c1b68ffd6f (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
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
{-# LANGUAGE OverloadedStrings   #-}
{-# LANGUAGE PatternGuards       #-}
{-# LANGUAGE FlexibleInstances   #-}
{-# LANGUAGE MultiParamTypeClasses #-}
{-# LANGUAGE ScopedTypeVariables #-}
{-
Copyright (C) 2006-2017 John MacFarlane <jgm@berkeley.edu>

This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
-}

{- |
   Module      : Text.Pandoc.Readers.LaTeX
   Copyright   : Copyright (C) 2006-2017 John MacFarlane
   License     : GNU GPL, version 2 or above

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

Conversion of LaTeX to 'Pandoc' document.

-}
module Text.Pandoc.Readers.LaTeX ( readLaTeX,
                                   applyMacros,
                                   rawLaTeXInline,
                                   rawLaTeXBlock,
                                   inlineCommand
                                 ) where

import Control.Applicative (many, optional, (<|>))
import Control.Monad
import Control.Monad.Except (throwError)
import Control.Monad.Trans (lift)
import Data.Char (chr, isAlphaNum, isLetter, ord, isDigit)
import Data.Default
import Data.Text (Text)
import qualified Data.Text as T
import Data.List (intercalate, isPrefixOf)
import qualified Data.Map as M
import qualified Data.Set as Set
import Data.Maybe (fromMaybe, maybeToList)
import Safe (minimumDef)
import System.FilePath (addExtension, replaceExtension, takeExtension)
import Text.Pandoc.Builder
import Text.Pandoc.Class (PandocMonad, PandocPure, lookupEnv,
                          readFileFromDirs, report, setResourcePath,
                          getResourcePath, setTranslations, translateTerm)
import qualified Text.Pandoc.Translations as Translations
import Text.Pandoc.BCP47 (Lang(..), renderLang)
import Text.Pandoc.Highlighting (fromListingsLanguage, languagesByExtension)
import Text.Pandoc.ImageSize (numUnit, showFl)
import Text.Pandoc.Logging
import Text.Pandoc.Options
import Text.Pandoc.Parsing hiding (many, optional, withRaw,
                            mathInline, mathDisplay,
                            space, (<|>), spaces, blankline)
import Text.Pandoc.Shared
import Text.Pandoc.Readers.LaTeX.Types (Macro(..), ExpansionPoint(..), Tok(..),
                            TokType(..))
import Text.Pandoc.Walk
import Text.Pandoc.Error (PandocError(PandocParsecError, PandocMacroLoop))

-- for debugging:
-- import Text.Pandoc.Extensions (getDefaultExtensions)
-- import Text.Pandoc.Class (runIOorExplode, PandocIO)
-- import Debug.Trace (traceShowId)

-- | Parse LaTeX from string and return 'Pandoc' document.
readLaTeX :: PandocMonad m
          => ReaderOptions -- ^ Reader options
          -> Text        -- ^ String to parse (assumes @'\n'@ line endings)
          -> m Pandoc
readLaTeX opts ltx = do
  parsed <- runParserT parseLaTeX def{ sOptions = opts } "source"
               (tokenize (crFilter ltx))
  case parsed of
    Right result -> return result
    Left e       -> throwError $ PandocParsecError (T.unpack ltx) e

parseLaTeX :: PandocMonad m => LP m Pandoc
parseLaTeX = do
  bs <- blocks
  eof
  st <- getState
  let meta = sMeta st
  let doc' = doc bs
  let headerLevel (Header n _ _) = [n]
      headerLevel _ = []
  let bottomLevel = minimumDef 1 $ query headerLevel doc'
  let adjustHeaders m (Header n attr ils) = Header (n+m) attr ils
      adjustHeaders _ x = x
  let (Pandoc _ bs') =
       -- handle the case where you have \part or \chapter
       (if bottomLevel < 1
           then walk (adjustHeaders (1 - bottomLevel))
           else id) $
       walk (resolveRefs (sLabels st)) $ doc'
  return $ Pandoc meta bs'

resolveRefs :: M.Map String [Inline] -> Inline -> Inline
resolveRefs labels x@(Span (ident,classes,kvs) _) =
  case (lookup "reference-type" kvs,
        lookup "reference" kvs) of
        (Just "ref", Just lab) ->
          case M.lookup lab labels of
               Just txt -> Span (ident,classes,kvs)
                             [Link nullAttr txt ('#':lab, "")]
               Nothing  -> x
        _ -> x
resolveRefs _ x = x


-- testParser :: LP PandocIO a -> Text -> IO a
-- testParser p t = do
--   res <- runIOorExplode (runParserT p defaultLaTeXState{
--             sOptions = def{ readerExtensions =
--               enableExtension Ext_raw_tex $
--                 getDefaultExtensions "latex" }} "source" (tokenize t))
--   case res of
--        Left e  -> error (show e)
--        Right r -> return r

newtype HeaderNum = HeaderNum [Int]
  deriving (Show)

renderHeaderNum :: HeaderNum -> String
renderHeaderNum (HeaderNum xs) =
  intercalate "." (map show xs)

incrementHeaderNum :: Int -> HeaderNum -> HeaderNum
incrementHeaderNum level (HeaderNum ns) = HeaderNum $
  case reverse (take level (ns ++ repeat 0)) of
       (x:xs) -> reverse (x+1 : xs)
       []     -> []  -- shouldn't happen

data LaTeXState = LaTeXState{ sOptions       :: ReaderOptions
                            , sMeta          :: Meta
                            , sQuoteContext  :: QuoteContext
                            , sMacros        :: M.Map Text Macro
                            , sContainers    :: [String]
                            , sHeaders       :: M.Map Inlines String
                            , sLogMessages   :: [LogMessage]
                            , sIdentifiers   :: Set.Set String
                            , sVerbatimMode  :: Bool
                            , sCaption       :: Maybe Inlines
                            , sInListItem    :: Bool
                            , sInTableCell   :: Bool
                            , sLastHeaderNum :: HeaderNum
                            , sLabels        :: M.Map String [Inline]
                            }
     deriving Show

defaultLaTeXState :: LaTeXState
defaultLaTeXState = LaTeXState{ sOptions       = def
                              , sMeta          = nullMeta
                              , sQuoteContext  = NoQuote
                              , sMacros        = M.empty
                              , sContainers    = []
                              , sHeaders       = M.empty
                              , sLogMessages   = []
                              , sIdentifiers   = Set.empty
                              , sVerbatimMode  = False
                              , sCaption       = Nothing
                              , sInListItem    = False
                              , sInTableCell   = False
                              , sLastHeaderNum = HeaderNum []
                              , sLabels        = M.empty
                              }

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

instance HasLogMessages LaTeXState where
  addLogMessage msg st = st{ sLogMessages = msg : sLogMessages st }
  getLogMessages st = reverse $ sLogMessages st

instance HasIdentifierList LaTeXState where
  extractIdentifierList     = sIdentifiers
  updateIdentifierList f st = st{ sIdentifiers = f $ sIdentifiers st }

instance HasIncludeFiles LaTeXState where
  getIncludeFiles = sContainers
  addIncludeFile f s = s{ sContainers = f : sContainers s }
  dropLatestIncludeFile s = s { sContainers = drop 1 $ sContainers s }

instance HasHeaderMap LaTeXState where
  extractHeaderMap     = sHeaders
  updateHeaderMap f st = st{ sHeaders = f $ sHeaders st }

instance HasMacros LaTeXState where
  extractMacros  st  = sMacros st
  updateMacros f st  = st{ sMacros = f (sMacros st) }

instance HasReaderOptions LaTeXState where
  extractReaderOptions = sOptions

instance HasMeta LaTeXState where
  setMeta field val st =
    st{ sMeta = setMeta field val $ sMeta st }
  deleteMeta field st =
    st{ sMeta = deleteMeta field $ sMeta st }

instance Default LaTeXState where
  def = defaultLaTeXState

type LP m = ParserT [Tok] LaTeXState m

withVerbatimMode :: PandocMonad m => LP m a -> LP m a
withVerbatimMode parser = do
  updateState $ \st -> st{ sVerbatimMode = True }
  result <- parser
  updateState $ \st -> st{ sVerbatimMode = False }
  return result

rawLaTeXParser :: (PandocMonad m, HasMacros s, HasReaderOptions s)
               => LP m a -> ParserT String s m String
rawLaTeXParser parser = do
  inp <- getInput
  let toks = tokenize $ T.pack inp
  pstate <- getState
  let lstate = def{ sOptions = extractReaderOptions pstate }
  res <- lift $ runParserT ((,) <$> try (snd <$> withRaw parser) <*> getState)
            lstate "source" toks
  case res of
       Left _    -> mzero
       Right (raw, st) -> do
         updateState (updateMacros ((sMacros st) <>))
         takeP (T.length (untokenize raw))

applyMacros :: (PandocMonad m, HasMacros s, HasReaderOptions s)
            => String -> ParserT String s m String
applyMacros s = (guardDisabled Ext_latex_macros >> return s) <|>
   do let retokenize = doMacros 0 *>
             (toksToString <$> many (satisfyTok (const True)))
      pstate <- getState
      let lstate = def{ sOptions = extractReaderOptions pstate
                      , sMacros  = extractMacros pstate }
      res <- runParserT retokenize lstate "math" (tokenize (T.pack s))
      case res of
           Left e -> fail (show e)
           Right s' -> return s'

rawLaTeXBlock :: (PandocMonad m, HasMacros s, HasReaderOptions s)
              => ParserT String s m String
rawLaTeXBlock = do
  lookAhead (try (char '\\' >> letter))
  rawLaTeXParser (environment <|> macroDef <|> blockCommand)

rawLaTeXInline :: (PandocMonad m, HasMacros s, HasReaderOptions s)
               => ParserT String s m String
rawLaTeXInline = do
  lookAhead (try (char '\\' >> letter) <|> char '$')
  rawLaTeXParser (inlineEnvironment <|> inlineCommand')

inlineCommand :: PandocMonad m => ParserT String ParserState m Inlines
inlineCommand = do
  lookAhead (try (char '\\' >> letter) <|> char '$')
  inp <- getInput
  let toks = tokenize $ T.pack inp
  let rawinline = do
         (il, raw) <- try $ withRaw (inlineEnvironment <|> inlineCommand')
         st <- getState
         return (il, raw, st)
  pstate <- getState
  let lstate = def{ sOptions = extractReaderOptions pstate
                  , sMacros  = extractMacros pstate }
  res <- runParserT rawinline lstate "source" toks
  case res of
       Left _ -> mzero
       Right (il, raw, s) -> do
         updateState $ updateMacros (const $ sMacros s)
         takeP (T.length (untokenize raw))
         return il

tokenize :: Text -> [Tok]
tokenize = totoks (1, 1)

totoks :: (Line, Column) -> Text -> [Tok]
totoks (lin,col) t =
  case T.uncons t of
       Nothing        -> []
       Just (c, rest)
         | c == '\n' ->
           Tok (lin, col) Newline "\n"
           : totoks (lin + 1,1) rest
         | isSpaceOrTab c ->
           let (sps, rest') = T.span isSpaceOrTab t
           in  Tok (lin, col) Spaces sps
               : totoks (lin, col + T.length sps) rest'
         | isAlphaNum c ->
           let (ws, rest') = T.span isAlphaNum t
           in  Tok (lin, col) Word ws
               : totoks (lin, col + T.length ws) rest'
         | c == '%' ->
           let (cs, rest') = T.break (== '\n') rest
           in  Tok (lin, col) Comment ("%" <> cs)
               : totoks (lin, col + 1 + T.length cs) rest'
         | c == '\\' ->
           case T.uncons rest of
                Nothing -> [Tok (lin, col) Symbol (T.singleton c)]
                Just (d, rest')
                  | isLetterOrAt d ->
                      -- \makeatletter is common in macro defs;
                      -- ideally we should make tokenization sensitive
                      -- to \makeatletter and \makeatother, but this is
                      -- probably best for now
                      let (ws, rest'') = T.span isLetterOrAt rest
                          (ss, rest''') = T.span isSpaceOrTab rest''
                      in  Tok (lin, col) (CtrlSeq ws) ("\\" <> ws <> ss)
                          : totoks (lin,
                                 col + 1 + T.length ws + T.length ss) rest'''
                  | d == '\t' || d == '\n' ->
                      Tok (lin, col) Symbol ("\\")
                      : totoks (lin, col + 1) rest
                  | otherwise  ->
                      Tok (lin, col) (CtrlSeq (T.singleton d)) (T.pack [c,d])
                      : totoks (lin, col + 2) rest'
         | c == '#' ->
           let (t1, t2) = T.span (\d -> d >= '0' && d <= '9') rest
           in  case safeRead (T.unpack t1) of
                    Just i ->
                       Tok (lin, col) (Arg i) ("#" <> t1)
                       : totoks (lin, col + 1 + T.length t1) t2
                    Nothing ->
                       Tok (lin, col) Symbol ("#")
                       : totoks (lin, col + 1) t2
         | c == '^' ->
           case T.uncons rest of
                Just ('^', rest') ->
                  case T.uncons rest' of
                       Just (d, rest'')
                         | isLowerHex d ->
                           case T.uncons rest'' of
                                Just (e, rest''') | isLowerHex e ->
                                  Tok (lin, col) Esc2 (T.pack ['^','^',d,e])
                                  : totoks (lin, col + 4) rest'''
                                _ ->
                                  Tok (lin, col) Esc1 (T.pack ['^','^',d])
                                  : totoks (lin, col + 3) rest''
                         | d < '\128' ->
                                  Tok (lin, col) Esc1 (T.pack ['^','^',d])
                                  : totoks (lin, col + 3) rest''
                       _ -> [Tok (lin, col) Symbol ("^"),
                             Tok (lin, col + 1) Symbol ("^")]
                _ -> Tok (lin, col) Symbol ("^")
                     : totoks (lin, col + 1) rest
         | otherwise ->
           Tok (lin, col) Symbol (T.singleton c) : totoks (lin, col + 1) rest

  where isSpaceOrTab ' '  = True
        isSpaceOrTab '\t' = True
        isSpaceOrTab _    = False
        isLetterOrAt '@'  = True
        isLetterOrAt c    = isLetter c

isLowerHex :: Char -> Bool
isLowerHex x = x >= '0' && x <= '9' || x >= 'a' && x <= 'f'

untokenize :: [Tok] -> Text
untokenize = mconcat . map untoken

untoken :: Tok -> Text
untoken (Tok _ _ t) = t

satisfyTok :: PandocMonad m => (Tok -> Bool) -> LP m Tok
satisfyTok f =
  try $ do
    res <- tokenPrim (T.unpack . untoken) updatePos matcher
    doMacros 0 -- apply macros on remaining input stream
    return res
  where matcher t | f t       = Just t
                  | otherwise = Nothing
        updatePos :: SourcePos -> Tok -> [Tok] -> SourcePos
        updatePos spos _ (Tok (lin,col) _ _ : _) =
          setSourceColumn (setSourceLine spos lin) col
        updatePos spos _ [] = spos

doMacros :: PandocMonad m => Int -> LP m ()
doMacros n = do
  verbatimMode <- sVerbatimMode <$> getState
  when (not verbatimMode) $ do
    inp <- getInput
    case inp of
         Tok spos (CtrlSeq "begin") _ : Tok _ Symbol "{" :
          Tok _ Word name : Tok _ Symbol "}" : ts
            -> handleMacros spos name ts
         Tok spos (CtrlSeq "end") _ : Tok _ Symbol "{" :
          Tok _ Word name : Tok _ Symbol "}" : ts
            -> handleMacros spos ("end" <> name) ts
         Tok spos (CtrlSeq name) _ : ts
            -> handleMacros spos name ts
         _ -> return ()
  where handleMacros spos name ts = do
                macros <- sMacros <$> getState
                case M.lookup name macros of
                     Nothing -> return ()
                     Just (Macro expansionPoint numargs optarg newtoks) -> do
                       setInput ts
                       let getarg = spaces >> braced
                       args <- case optarg of
                                    Nothing -> count numargs getarg
                                    Just o  ->
                                       (:) <$> option o bracketedToks
                                           <*> count (numargs - 1) getarg
                       let addTok (Tok _ (Arg i) _) acc | i > 0
                                                        , i <= numargs =
                                 map (setpos spos) (args !! (i - 1)) ++ acc
                           addTok t acc = setpos spos t : acc
                       ts' <- getInput
                       setInput $ foldr addTok ts' newtoks
                       case expansionPoint of
                            ExpandWhenUsed ->
                              if n > 20  -- detect macro expansion loops
                                 then throwError $ PandocMacroLoop (T.unpack name)
                                 else doMacros (n + 1)
                            ExpandWhenDefined -> return ()

setpos :: (Line, Column) -> Tok -> Tok
setpos spos (Tok _ tt txt) = Tok spos tt txt

anyControlSeq :: PandocMonad m => LP m Tok
anyControlSeq = satisfyTok isCtrlSeq
  where isCtrlSeq (Tok _ (CtrlSeq _) _) = True
        isCtrlSeq _                     = False

anySymbol :: PandocMonad m => LP m Tok
anySymbol = satisfyTok isSym
  where isSym (Tok _ Symbol _) = True
        isSym _                = False

spaces :: PandocMonad m => LP m ()
spaces = skipMany (satisfyTok (tokTypeIn [Comment, Spaces, Newline]))

spaces1 :: PandocMonad m => LP m ()
spaces1 = skipMany1 (satisfyTok (tokTypeIn [Comment, Spaces, Newline]))

tokTypeIn :: [TokType] -> Tok -> Bool
tokTypeIn toktypes (Tok _ tt _) = tt `elem` toktypes

controlSeq :: PandocMonad m => Text -> LP m Tok
controlSeq name = satisfyTok isNamed
  where isNamed (Tok _ (CtrlSeq n) _) = n == name
        isNamed _ = False

symbol :: PandocMonad m => Char -> LP m Tok
symbol c = satisfyTok isc
  where isc (Tok _ Symbol d) = case T.uncons d of
                                    Just (c',_) -> c == c'
                                    _ -> False
        isc _ = False

symbolIn :: PandocMonad m => [Char] -> LP m Tok
symbolIn cs = satisfyTok isInCs
  where isInCs (Tok _ Symbol d) = case T.uncons d of
                                       Just (c,_) -> c `elem` cs
                                       _ -> False
        isInCs _ = False

sp :: PandocMonad m => LP m ()
sp = whitespace <|> endline

whitespace :: PandocMonad m => LP m ()
whitespace = () <$ satisfyTok isSpaceTok
  where isSpaceTok (Tok _ Spaces _) = True
        isSpaceTok _ = False

newlineTok :: PandocMonad m => LP m ()
newlineTok = () <$ satisfyTok isNewlineTok

isNewlineTok :: Tok -> Bool
isNewlineTok (Tok _ Newline _) = True
isNewlineTok _ = False

comment :: PandocMonad m => LP m ()
comment = () <$ satisfyTok isCommentTok
  where isCommentTok (Tok _ Comment _) = True
        isCommentTok _ = False

anyTok :: PandocMonad m => LP m Tok
anyTok = satisfyTok (const True)

endline :: PandocMonad m => LP m ()
endline = try $ do
  newlineTok
  lookAhead anyTok
  notFollowedBy blankline

blankline :: PandocMonad m => LP m ()
blankline = try $ skipMany whitespace *> newlineTok

primEscape :: PandocMonad m => LP m Char
primEscape = do
  Tok _ toktype t <- satisfyTok (tokTypeIn [Esc1, Esc2])
  case toktype of
       Esc1 -> case T.uncons (T.drop 2 t) of
                    Just (c, _)
                      | c >= '\64' && c <= '\127' -> return (chr (ord c - 64))
                      | otherwise                 -> return (chr (ord c + 64))
                    Nothing -> fail "Empty content of Esc1"
       Esc2 -> case safeRead ('0':'x':T.unpack (T.drop 2 t)) of
                    Just x -> return (chr x)
                    Nothing -> fail $ "Could not read: " ++ T.unpack t
       _    -> fail "Expected an Esc1 or Esc2 token" -- should not happen

bgroup :: PandocMonad m => LP m Tok
bgroup = try $ do
  skipMany sp
  symbol '{' <|> controlSeq "bgroup" <|> controlSeq "begingroup"

egroup :: PandocMonad m => LP m Tok
egroup = (symbol '}' <|> controlSeq "egroup" <|> controlSeq "endgroup")

grouped :: (PandocMonad m,  Monoid a) => LP m a -> LP m a
grouped parser = try $ do
  bgroup
  -- first we check for an inner 'grouped', because
  -- {{a,b}} should be parsed the same as {a,b}
  try (grouped parser <* egroup) <|> (mconcat <$> manyTill parser egroup)

braced :: PandocMonad m => LP m [Tok]
braced = bgroup *> braced' 1
  where braced' (n :: Int) =
          handleEgroup n <|> handleBgroup n <|> handleOther n
        handleEgroup n = do
          t <- egroup
          if n == 1
             then return []
             else (t:) <$> braced' (n - 1)
        handleBgroup n = do
          t <- bgroup
          (t:) <$> braced' (n + 1)
        handleOther n = do
          t <- anyTok
          (t:) <$> braced' n

bracketed :: PandocMonad m => Monoid a => LP m a -> LP m a
bracketed parser = try $ do
  symbol '['
  mconcat <$> manyTill parser (symbol ']')

dimenarg :: PandocMonad m => LP m Text
dimenarg = try $ do
  ch  <- option False $ True <$ symbol '='
  Tok _ _ s <- satisfyTok isWordTok
  guard $ (T.take 2 (T.reverse s)) `elem`
           ["pt","pc","in","bp","cm","mm","dd","cc","sp"]
  let num = T.take (T.length s - 2) s
  guard $ T.length num > 0
  guard $ T.all isDigit num
  return $ T.pack ['=' | ch] <> s

-- inline elements:

word :: PandocMonad m => LP m Inlines
word = (str . T.unpack . untoken) <$> satisfyTok isWordTok

regularSymbol :: PandocMonad m => LP m Inlines
regularSymbol = (str . T.unpack . untoken) <$> satisfyTok isRegularSymbol
  where isRegularSymbol (Tok _ Symbol t) = not $ T.any isSpecial t
        isRegularSymbol _ = False
        isSpecial c = c `Set.member` specialChars

specialChars :: Set.Set Char
specialChars = Set.fromList "#$%&~_^\\{}"

isWordTok :: Tok -> Bool
isWordTok (Tok _ Word _) = True
isWordTok _ = False

inlineGroup :: PandocMonad m => LP m Inlines
inlineGroup = do
  ils <- grouped inline
  if isNull ils
     then return mempty
     else return $ spanWith nullAttr ils
          -- we need the span so we can detitlecase bibtex entries;
          -- we need to know when something is {C}apitalized

doLHSverb :: PandocMonad m => LP m Inlines
doLHSverb =
  (codeWith ("",["haskell"],[]) . T.unpack . untokenize)
    <$> manyTill (satisfyTok (not . isNewlineTok)) (symbol '|')

mkImage :: PandocMonad m => [(String, String)] -> String -> LP m Inlines
mkImage options src = do
   let replaceTextwidth (k,v) =
         case numUnit v of
              Just (num, "\\textwidth") -> (k, showFl (num * 100) ++ "%")
              _ -> (k, v)
   let kvs = map replaceTextwidth
             $ filter (\(k,_) -> k `elem` ["width", "height"]) options
   let attr = ("",[], kvs)
   let alt = str "image"
   case takeExtension src of
        "" -> do
              defaultExt <- getOption readerDefaultImageExtension
              return $ imageWith attr (addExtension src defaultExt) "" alt
        _  -> return $ imageWith attr src "" alt

doxspace :: PandocMonad m => LP m Inlines
doxspace = do
  (space <$ lookAhead (satisfyTok startsWithLetter)) <|> return mempty
  where startsWithLetter (Tok _ Word t) =
          case T.uncons t of
               Just (c, _) | isLetter c -> True
               _ -> False
        startsWithLetter _ = False


-- converts e.g. \SI{1}[\$]{} to "$ 1" or \SI{1}{\euro} to "1 €"
dosiunitx :: PandocMonad m => LP m Inlines
dosiunitx = do
  skipopts
  value <- tok
  valueprefix <- option "" $ bracketed tok
  unit <- tok
  let emptyOr160 "" = ""
      emptyOr160 _  = "\160"
  return . mconcat $ [valueprefix,
                      emptyOr160 valueprefix,
                      value,
                      emptyOr160 unit,
                      unit]

lit :: String -> LP m Inlines
lit = pure . str

removeDoubleQuotes :: Text -> Text
removeDoubleQuotes t =
  maybe t id $ T.stripPrefix "\"" t >>= T.stripSuffix "\""

doubleQuote :: PandocMonad m => LP m Inlines
doubleQuote = do
       quoted' doubleQuoted (try $ count 2 $ symbol '`')
                            (void $ try $ count 2 $ symbol '\'')
   <|> quoted' doubleQuoted ((:[]) <$> symbol '“') (void $ symbol '”')
   -- the following is used by babel for localized quotes:
   <|> quoted' doubleQuoted (try $ sequence [symbol '"', symbol '`'])
                            (void $ try $ sequence [symbol '"', symbol '\''])
   <|> quoted' doubleQuoted ((:[]) <$> symbol '"')
                            (void $ symbol '"')

singleQuote :: PandocMonad m => LP m Inlines
singleQuote = do
       quoted' singleQuoted ((:[]) <$> symbol '`')
                            (try $ symbol '\'' >>
                                  notFollowedBy (satisfyTok startsWithLetter))
   <|> quoted' singleQuoted ((:[]) <$> symbol '‘')
                            (try $ symbol '’' >>
                                  notFollowedBy (satisfyTok startsWithLetter))
  where startsWithLetter (Tok _ Word t) =
          case T.uncons t of
               Just (c, _) | isLetter c -> True
               _ -> False
        startsWithLetter _ = False

quoted' :: PandocMonad m
        => (Inlines -> Inlines)
        -> LP m [Tok]
        -> LP m ()
        -> LP m Inlines
quoted' f starter ender = do
  startchs <- (T.unpack . untokenize) <$> starter
  smart <- extensionEnabled Ext_smart <$> getOption readerExtensions
  if smart
     then do
       ils <- many (notFollowedBy ender >> inline)
       (ender >> return (f (mconcat ils))) <|>
            (<> mconcat ils) <$>
                    lit (case startchs of
                              "``" -> "“"
                              "`"  -> "‘"
                              cs   -> cs)
     else lit startchs

enquote :: PandocMonad m => LP m Inlines
enquote = do
  skipopts
  quoteContext <- sQuoteContext <$> getState
  if quoteContext == InDoubleQuote
     then singleQuoted <$> withQuoteContext InSingleQuote tok
     else doubleQuoted <$> withQuoteContext InDoubleQuote tok

doverb :: PandocMonad m => LP m Inlines
doverb = do
  Tok _ Symbol t <- anySymbol
  marker <- case T.uncons t of
              Just (c, ts) | T.null ts -> return c
              _ -> mzero
  withVerbatimMode $
    (code . T.unpack . untokenize) <$>
      manyTill (verbTok marker) (symbol marker)

verbTok :: PandocMonad m => Char -> LP m Tok
verbTok stopchar = do
  t@(Tok (lin, col) toktype txt) <- satisfyTok (not . isNewlineTok)
  case T.findIndex (== stopchar) txt of
       Nothing -> return t
       Just i  -> do
         let (t1, t2) = T.splitAt i txt
         inp <- getInput
         setInput $ Tok (lin, col + i) Symbol (T.singleton stopchar)
                  : (totoks (lin, col + i + 1) (T.drop 1 t2)) ++ inp
         return $ Tok (lin, col) toktype t1

dolstinline :: PandocMonad m => LP m Inlines
dolstinline = do
  options <- option [] keyvals
  let classes = maybeToList $ lookup "language" options >>= fromListingsLanguage
  Tok _ Symbol t <- anySymbol
  marker <- case T.uncons t of
              Just (c, ts) | T.null ts -> return c
              _ -> mzero
  let stopchar = if marker == '{' then '}' else marker
  withVerbatimMode $
    (codeWith ("",classes,[]) . T.unpack . untokenize) <$>
      manyTill (verbTok stopchar) (symbol stopchar)

keyval :: PandocMonad m => LP m (String, String)
keyval = try $ do
  Tok _ Word key <- satisfyTok isWordTok
  let isSpecSym (Tok _ Symbol t) = t `elem` [".",":","-","|","\\"]
      isSpecSym _ = False
  val <- option [] $ do
           symbol '='
           braced <|> (many1 (satisfyTok isWordTok <|> satisfyTok isSpecSym
                               <|> anyControlSeq))
  optional sp
  optional (symbol ',')
  optional sp
  return (T.unpack key, T.unpack . untokenize $ val)

keyvals :: PandocMonad m => LP m [(String, String)]
keyvals = try $ symbol '[' >> manyTill keyval (symbol ']')

accent :: (Char -> String) -> Inlines -> LP m Inlines
accent f ils =
  case toList ils of
       (Str (x:xs) : ys) -> return $ fromList (Str (f x ++ xs) : ys)
       []                -> mzero
       _                 -> return ils

grave :: Char -> String
grave 'A' = "À"
grave 'E' = "È"
grave 'I' = "Ì"
grave 'O' = "Ò"
grave 'U' = "Ù"
grave 'a' = "à"
grave 'e' = "è"
grave 'i' = "ì"
grave 'o' = "ò"
grave 'u' = "ù"
grave c   = [c]

acute :: Char -> String
acute 'A' = "Á"
acute 'E' = "É"
acute 'I' = "Í"
acute 'O' = "Ó"
acute 'U' = "Ú"
acute 'Y' = "Ý"
acute 'a' = "á"
acute 'e' = "é"
acute 'i' = "í"
acute 'o' = "ó"
acute 'u' = "ú"
acute 'y' = "ý"
acute 'C' = "Ć"
acute 'c' = "ć"
acute 'L' = "Ĺ"
acute 'l' = "ĺ"
acute 'N' = "Ń"
acute 'n' = "ń"
acute 'R' = "Ŕ"
acute 'r' = "ŕ"
acute 'S' = "Ś"
acute 's' = "ś"
acute 'Z' = "Ź"
acute 'z' = "ź"
acute c   = [c]

circ :: Char -> String
circ 'A' = "Â"
circ 'E' = "Ê"
circ 'I' = "Î"
circ 'O' = "Ô"
circ 'U' = "Û"
circ 'a' = "â"
circ 'e' = "ê"
circ 'i' = "î"
circ 'o' = "ô"
circ 'u' = "û"
circ 'C' = "Ĉ"
circ 'c' = "ĉ"
circ 'G' = "Ĝ"
circ 'g' = "ĝ"
circ 'H' = "Ĥ"
circ 'h' = "ĥ"
circ 'J' = "Ĵ"
circ 'j' = "ĵ"
circ 'S' = "Ŝ"
circ 's' = "ŝ"
circ 'W' = "Ŵ"
circ 'w' = "ŵ"
circ 'Y' = "Ŷ"
circ 'y' = "ŷ"
circ c   = [c]

tilde :: Char -> String
tilde 'A' = "Ã"
tilde 'a' = "ã"
tilde 'O' = "Õ"
tilde 'o' = "õ"
tilde 'I' = "Ĩ"
tilde 'i' = "ĩ"
tilde 'U' = "Ũ"
tilde 'u' = "ũ"
tilde 'N' = "Ñ"
tilde 'n' = "ñ"
tilde c   = [c]

umlaut :: Char -> String
umlaut 'A' = "Ä"
umlaut 'E' = "Ë"
umlaut 'I' = "Ï"
umlaut 'O' = "Ö"
umlaut 'U' = "Ü"
umlaut 'a' = "ä"
umlaut 'e' = "ë"
umlaut 'i' = "ï"
umlaut 'o' = "ö"
umlaut 'u' = "ü"
umlaut c   = [c]

hungarumlaut :: Char -> String
hungarumlaut 'A' = "A̋"
hungarumlaut 'E' = "E̋"
hungarumlaut 'I' = "I̋"
hungarumlaut 'O' = "Ő"
hungarumlaut 'U' = "Ű"
hungarumlaut 'Y' = "ӳ"
hungarumlaut 'a' = "a̋"
hungarumlaut 'e' = "e̋"
hungarumlaut 'i' = "i̋"
hungarumlaut 'o' = "ő"
hungarumlaut 'u' = "ű"
hungarumlaut 'y' = "ӳ"
hungarumlaut c   = [c]

dot :: Char -> String
dot 'C' = "Ċ"
dot 'c' = "ċ"
dot 'E' = "Ė"
dot 'e' = "ė"
dot 'G' = "Ġ"
dot 'g' = "ġ"
dot 'I' = "İ"
dot 'Z' = "Ż"
dot 'z' = "ż"
dot c   = [c]

macron :: Char -> String
macron 'A' = "Ā"
macron 'E' = "Ē"
macron 'I' = "Ī"
macron 'O' = "Ō"
macron 'U' = "Ū"
macron 'a' = "ā"
macron 'e' = "ē"
macron 'i' = "ī"
macron 'o' = "ō"
macron 'u' = "ū"
macron c   = [c]

cedilla :: Char -> String
cedilla 'c' = "ç"
cedilla 'C' = "Ç"
cedilla 's' = "ş"
cedilla 'S' = "Ş"
cedilla 't' = "ţ"
cedilla 'T' = "Ţ"
cedilla 'e' = "ȩ"
cedilla 'E' = "Ȩ"
cedilla 'h' = "ḩ"
cedilla 'H' = "Ḩ"
cedilla 'o' = "o̧"
cedilla 'O' = "O̧"
cedilla c   = [c]

hacek :: Char -> String
hacek 'A' = "Ǎ"
hacek 'a' = "ǎ"
hacek 'C' = "Č"
hacek 'c' = "č"
hacek 'D' = "Ď"
hacek 'd' = "ď"
hacek 'E' = "Ě"
hacek 'e' = "ě"
hacek 'G' = "Ǧ"
hacek 'g' = "ǧ"
hacek 'H' = "Ȟ"
hacek 'h' = "ȟ"
hacek 'I' = "Ǐ"
hacek 'i' = "ǐ"
hacek 'j' = "ǰ"
hacek 'K' = "Ǩ"
hacek 'k' = "ǩ"
hacek 'L' = "Ľ"
hacek 'l' = "ľ"
hacek 'N' = "Ň"
hacek 'n' = "ň"
hacek 'O' = "Ǒ"
hacek 'o' = "ǒ"
hacek 'R' = "Ř"
hacek 'r' = "ř"
hacek 'S' = "Š"
hacek 's' = "š"
hacek 'T' = "Ť"
hacek 't' = "ť"
hacek 'U' = "Ǔ"
hacek 'u' = "ǔ"
hacek 'Z' = "Ž"
hacek 'z' = "ž"
hacek c   = [c]

breve :: Char -> String
breve 'A' = "Ă"
breve 'a' = "ă"
breve 'E' = "Ĕ"
breve 'e' = "ĕ"
breve 'G' = "Ğ"
breve 'g' = "ğ"
breve 'I' = "Ĭ"
breve 'i' = "ĭ"
breve 'O' = "Ŏ"
breve 'o' = "ŏ"
breve 'U' = "Ŭ"
breve 'u' = "ŭ"
breve c   = [c]

toksToString :: [Tok] -> String
toksToString = T.unpack . untokenize

mathDisplay :: String -> Inlines
mathDisplay = displayMath . trim

mathInline :: String -> Inlines
mathInline = math . trim

dollarsMath :: PandocMonad m => LP m Inlines
dollarsMath = do
  symbol '$'
  display <- option False (True <$ symbol '$')
  contents <- trim . toksToString <$>
               many (notFollowedBy (symbol '$') >> anyTok)
  if display
     then do
       mathDisplay contents <$ try (symbol '$' >> symbol '$')
        <|> (guard (null contents) >> return (mathInline ""))
     else mathInline contents <$ (symbol '$')

-- citations

addPrefix :: [Inline] -> [Citation] -> [Citation]
addPrefix p (k:ks) = k {citationPrefix = p ++ citationPrefix k} : ks
addPrefix _ _      = []

addSuffix :: [Inline] -> [Citation] -> [Citation]
addSuffix s ks@(_:_) =
  let k = last ks
  in  init ks ++ [k {citationSuffix = citationSuffix k ++ s}]
addSuffix _ _ = []

simpleCiteArgs :: PandocMonad m => LP m [Citation]
simpleCiteArgs = try $ do
  first  <- optionMaybe $ toList <$> opt
  second <- optionMaybe $ toList <$> opt
  keys <- try $ bgroup *> (manyTill citationLabel egroup)
  let (pre, suf) = case (first  , second ) of
        (Just s , Nothing) -> (mempty, s )
        (Just s , Just t ) -> (s , t )
        _                  -> (mempty, mempty)
      conv k = Citation { citationId      = k
                        , citationPrefix  = []
                        , citationSuffix  = []
                        , citationMode    = NormalCitation
                        , citationHash    = 0
                        , citationNoteNum = 0
                        }
  return $ addPrefix pre $ addSuffix suf $ map conv keys

citationLabel :: PandocMonad m => LP m String
citationLabel  = do
  optional sp
  toksToString <$>
    (many1 (satisfyTok isWordTok <|> symbolIn bibtexKeyChar)
          <* optional sp
          <* optional (symbol ',')
          <* optional sp)
  where bibtexKeyChar = ".:;?!`'()/*@_+=-[]" :: [Char]

cites :: PandocMonad m => CitationMode -> Bool -> LP m [Citation]
cites mode multi = try $ do
  cits <- if multi
             then many1 simpleCiteArgs
             else count 1 simpleCiteArgs
  let cs = concat cits
  return $ case mode of
        AuthorInText -> case cs of
                             (c:rest) -> c {citationMode = mode} : rest
                             []       -> []
        _            -> map (\a -> a {citationMode = mode}) cs

citation :: PandocMonad m => String -> CitationMode -> Bool -> LP m Inlines
citation name mode multi = do
  (c,raw) <- withRaw $ cites mode multi
  return $ cite c (rawInline "latex" $ "\\" ++ name ++ (toksToString raw))

handleCitationPart :: Inlines -> [Citation]
handleCitationPart ils =
  let isCite Cite{} = True
      isCite _      = False
      (pref, rest) = break isCite (toList ils)
  in case rest of
          (Cite cs _:suff) -> addPrefix pref $ addSuffix suff cs
          _                -> []

complexNatbibCitation :: PandocMonad m => CitationMode -> LP m Inlines
complexNatbibCitation mode = try $ do
  (cs, raw) <-
    withRaw $ concat <$> do
      bgroup
      items <- mconcat <$>
                many1 (notFollowedBy (symbol ';') >> inline)
                  `sepBy1` (symbol ';')
      egroup
      return $ map handleCitationPart items
  case cs of
       []       -> mzero
       (c:cits) -> return $ cite (c{ citationMode = mode }:cits)
                      (rawInline "latex" $ "\\citetext" ++ toksToString raw)

inNote :: Inlines -> Inlines
inNote ils =
  note $ para $ ils <> str "."

inlineCommand' :: PandocMonad m => LP m Inlines
inlineCommand' = try $ do
  Tok _ (CtrlSeq name) cmd <- anyControlSeq
  guard $ name /= "begin" && name /= "end"
  star <- option "" ("*" <$ symbol '*' <* optional sp)
  let name' = name <> star
  let names = ordNub [name', name] -- check non-starred as fallback
  let raw = do
       guard $ isInlineCommand name || not (isBlockCommand name)
       rawcommand <- getRawCommand name (cmd <> star)
       (guardEnabled Ext_raw_tex >> return (rawInline "latex" rawcommand))
         <|> ignore rawcommand
  lookupListDefault raw names inlineCommands

tok :: PandocMonad m => LP m Inlines
tok = grouped inline <|> inlineCommand' <|> singleChar'
  where singleChar' = do
          Tok _ _ t <- singleChar
          return (str (T.unpack t))

singleChar :: PandocMonad m => LP m Tok
singleChar = try $ do
  Tok (lin,col) toktype t <- satisfyTok (tokTypeIn [Word, Symbol])
  guard $ not $ toktype == Symbol &&
                T.any (`Set.member` specialChars) t
  if T.length t > 1
     then do
       let (t1, t2) = (T.take 1 t, T.drop 1 t)
       inp <- getInput
       setInput $ (Tok (lin, col + 1) toktype t2) : inp
       return $ Tok (lin,col) toktype t1
     else return $ Tok (lin,col) toktype t

opt :: PandocMonad m => LP m Inlines
opt = bracketed inline

rawopt :: PandocMonad m => LP m Text
rawopt = do
  symbol '['
  inner <- untokenize <$> manyTill anyTok (symbol ']')
  optional sp
  return $ "[" <> inner <> "]"

skipopts :: PandocMonad m => LP m ()
skipopts = skipMany rawopt

-- opts in angle brackets are used in beamer
rawangle :: PandocMonad m => LP m ()
rawangle = try $ do
  symbol '<'
  () <$ manyTill anyTok (symbol '>')

skipangles :: PandocMonad m => LP m ()
skipangles = skipMany rawangle

ignore :: (Monoid a, PandocMonad m) => String -> ParserT s u m a
ignore raw = do
  pos <- getPosition
  report $ SkippedContent raw pos
  return mempty

withRaw :: PandocMonad m => LP m a -> LP m (a, [Tok])
withRaw parser = do
  inp <- getInput
  result <- parser
  nxt <- option (Tok (0,0) Word "") (lookAhead anyTok)
  let raw = takeWhile (/= nxt) inp
  return (result, raw)

inBrackets :: Inlines -> Inlines
inBrackets x = str "[" <> x <> str "]"

unescapeURL :: String -> String
unescapeURL ('\\':x:xs) | isEscapable x = x:unescapeURL xs
  where isEscapable c = c `elem` ("#$%&~_^\\{}" :: String)
unescapeURL (x:xs) = x:unescapeURL xs
unescapeURL [] = ""

mathEnvWith :: PandocMonad m
            => (Inlines -> a) -> Maybe Text -> Text -> LP m a
mathEnvWith f innerEnv name = f . mathDisplay . inner <$> mathEnv name
   where inner x = case innerEnv of
                        Nothing -> x
                        Just y  -> "\\begin{" ++ T.unpack y ++ "}\n" ++ x ++
                                   "\\end{" ++ T.unpack y ++ "}"

mathEnv :: PandocMonad m => Text -> LP m String
mathEnv name = do
  skipopts
  optional blankline
  res <- manyTill anyTok (end_ name)
  return $ stripTrailingNewlines $ T.unpack $ untokenize res

inlineEnvironment :: PandocMonad m => LP m Inlines
inlineEnvironment = try $ do
  controlSeq "begin"
  name <- untokenize <$> braced
  M.findWithDefault mzero name inlineEnvironments

inlineEnvironments :: PandocMonad m => M.Map Text (LP m Inlines)
inlineEnvironments = M.fromList [
    ("displaymath", mathEnvWith id Nothing "displaymath")
  , ("math", math <$> mathEnv "math")
  , ("equation", mathEnvWith id Nothing "equation")
  , ("equation*", mathEnvWith id Nothing "equation*")
  , ("gather", mathEnvWith id (Just "gathered") "gather")
  , ("gather*", mathEnvWith id (Just "gathered") "gather*")
  , ("multline", mathEnvWith id (Just "gathered") "multline")
  , ("multline*", mathEnvWith id (Just "gathered") "multline*")
  , ("eqnarray", mathEnvWith id (Just "aligned") "eqnarray")
  , ("eqnarray*", mathEnvWith id (Just "aligned") "eqnarray*")
  , ("align", mathEnvWith id (Just "aligned") "align")
  , ("align*", mathEnvWith id (Just "aligned") "align*")
  , ("alignat", mathEnvWith id (Just "aligned") "alignat")
  , ("alignat*", mathEnvWith id (Just "aligned") "alignat*")
  ]

inlineCommands :: PandocMonad m => M.Map Text (LP m Inlines)
inlineCommands = M.fromList $
  [ ("emph", extractSpaces emph <$> tok)
  , ("textit", extractSpaces emph <$> tok)
  , ("textsl", extractSpaces emph <$> tok)
  , ("textsc", extractSpaces smallcaps <$> tok)
  , ("textsf", extractSpaces (spanWith ("",["sans-serif"],[])) <$> tok)
  , ("textmd", extractSpaces (spanWith ("",["medium"],[])) <$> tok)
  , ("textrm", extractSpaces (spanWith ("",["roman"],[])) <$> tok)
  , ("textup", extractSpaces (spanWith ("",["upright"],[])) <$> tok)
  , ("texttt", ttfamily)
  , ("sout", extractSpaces strikeout <$> tok)
  , ("textsuperscript", extractSpaces superscript <$> tok)
  , ("textsubscript", extractSpaces subscript <$> tok)
  , ("textbackslash", lit "\\")
  , ("backslash", lit "\\")
  , ("slash", lit "/")
  , ("textbf", extractSpaces strong <$> tok)
  , ("textnormal", extractSpaces (spanWith ("",["nodecor"],[])) <$> tok)
  , ("ldots", lit "…")
  , ("vdots", lit "\8942")
  , ("dots", lit "…")
  , ("mdots", lit "…")
  , ("sim", lit "~")
  , ("textgreek", tok)
  , ("sep", lit ",")
  , ("label", rawInlineOr "label" dolabel)
  , ("ref", rawInlineOr "ref" $ doref "ref")
  , ("cref", rawInlineOr "cref" $ doref "ref")       -- from cleveref.sty
  , ("vref", rawInlineOr "vref" $ doref "ref+page")  -- from varioref.sty
  , ("eqref", rawInlineOr "eqref" $ doref "eqref")   -- from amsmath.sty
  , ("(", mathInline . toksToString <$> manyTill anyTok (controlSeq ")"))
  , ("[", mathDisplay . toksToString <$> manyTill anyTok (controlSeq "]"))
  , ("ensuremath", mathInline . toksToString <$> braced)
  , ("texorpdfstring", (\_ x -> x) <$> tok <*> tok)
  , ("P", lit "¶")
  , ("S", lit "§")
  , ("$", lit "$")
  , ("%", lit "%")
  , ("&", lit "&")
  , ("#", lit "#")
  , ("_", lit "_")
  , ("{", lit "{")
  , ("}", lit "}")
  -- old TeX commands
  , ("em", extractSpaces emph <$> inlines)
  , ("it", extractSpaces emph <$> inlines)
  , ("sl", extractSpaces emph <$> inlines)
  , ("bf", extractSpaces strong <$> inlines)
  , ("rm", inlines)
  , ("itshape", extractSpaces emph <$> inlines)
  , ("slshape", extractSpaces emph <$> inlines)
  , ("scshape", extractSpaces smallcaps <$> inlines)
  , ("bfseries", extractSpaces strong <$> inlines)
  , ("/", pure mempty) -- italic correction
  , ("aa", lit "å")
  , ("AA", lit "Å")
  , ("ss", lit "ß")
  , ("o", lit "ø")
  , ("O", lit "Ø")
  , ("L", lit "Ł")
  , ("l", lit "ł")
  , ("ae", lit "æ")
  , ("AE", lit "Æ")
  , ("oe", lit "œ")
  , ("OE", lit "Œ")
  , ("pounds", lit "£")
  , ("euro", lit "€")
  , ("copyright", lit "©")
  , ("textasciicircum", lit "^")
  , ("textasciitilde", lit "~")
  , ("H", try $ tok >>= accent hungarumlaut)
  , ("`", option (str "`") $ try $ tok >>= accent grave)
  , ("'", option (str "'") $ try $ tok >>= accent acute)
  , ("^", option (str "^") $ try $ tok >>= accent circ)
  , ("~", option (str "~") $ try $ tok >>= accent tilde)
  , ("\"", option (str "\"") $ try $ tok >>= accent umlaut)
  , (".", option (str ".") $ try $ tok >>= accent dot)
  , ("=", option (str "=") $ try $ tok >>= accent macron)
  , ("c", option (str "c") $ try $ tok >>= accent cedilla)
  , ("v", option (str "v") $ try $ tok >>= accent hacek)
  , ("u", option (str "u") $ try $ tok >>= accent breve)
  , ("i", lit "i")
  , ("\\", linebreak <$ (do inTableCell <- sInTableCell <$> getState
                            guard $ not inTableCell
                            optional (bracketed inline)
                            spaces))
  , (",", lit "\8198")
  , ("@", pure mempty)
  , (" ", lit "\160")
  , ("ps", pure $ str "PS." <> space)
  , ("TeX", lit "TeX")
  , ("LaTeX", lit "LaTeX")
  , ("bar", lit "|")
  , ("textless", lit "<")
  , ("textgreater", lit ">")
  , ("thanks", note <$> grouped block)
  , ("footnote", note <$> grouped block)
  , ("verb", doverb)
  , ("lstinline", dolstinline)
  , ("Verb", doverb)
  , ("url", ((unescapeURL . T.unpack . untokenize) <$> braced) >>= \url ->
                  pure (link url "" (str url)))
  , ("href", (unescapeURL . toksToString <$>
                 braced <* optional sp) >>= \url ->
                   tok >>= \lab -> pure (link url "" lab))
  , ("includegraphics", do options <- option [] keyvals
                           src <- unescapeURL . T.unpack .
                                    removeDoubleQuotes . untokenize <$> braced
                           mkImage options src)
  , ("enquote", enquote)
  , ("figurename", doTerm Translations.Figure)
  , ("prefacename", doTerm Translations.Preface)
  , ("refname", doTerm Translations.References)
  , ("bibname", doTerm Translations.Bibliography)
  , ("chaptername", doTerm Translations.Chapter)
  , ("partname", doTerm Translations.Part)
  , ("contentsname", doTerm Translations.Contents)
  , ("listfigurename", doTerm Translations.ListOfFigures)
  , ("listtablename", doTerm Translations.ListOfTables)
  , ("indexname", doTerm Translations.Index)
  , ("abstractname", doTerm Translations.Abstract)
  , ("tablename", doTerm Translations.Table)
  , ("enclname", doTerm Translations.Encl)
  , ("ccname", doTerm Translations.Cc)
  , ("headtoname", doTerm Translations.To)
  , ("pagename", doTerm Translations.Page)
  , ("seename", doTerm Translations.See)
  , ("seealsoname", doTerm Translations.SeeAlso)
  , ("proofname", doTerm Translations.Proof)
  , ("glossaryname", doTerm Translations.Glossary)
  , ("lstlistingname", doTerm Translations.Listing)
  , ("cite", citation "cite" NormalCitation False)
  , ("Cite", citation "Cite" NormalCitation False)
  , ("citep", citation "citep" NormalCitation False)
  , ("citep*", citation "citep*" NormalCitation False)
  , ("citeal", citation "citeal" NormalCitation False)
  , ("citealp", citation "citealp" NormalCitation False)
  , ("citealp*", citation "citealp*" NormalCitation False)
  , ("autocite", citation "autocite" NormalCitation False)
  , ("smartcite", citation "smartcite" NormalCitation False)
  , ("footcite", inNote <$> citation "footcite" NormalCitation False)
  , ("parencite", citation "parencite" NormalCitation False)
  , ("supercite", citation "supercite" NormalCitation False)
  , ("footcitetext", inNote <$> citation "footcitetext" NormalCitation False)
  , ("citeyearpar", citation "citeyearpar" SuppressAuthor False)
  , ("citeyear", citation "citeyear" SuppressAuthor False)
  , ("autocite*", citation "autocite*" SuppressAuthor False)
  , ("cite*", citation "cite*" SuppressAuthor False)
  , ("parencite*", citation "parencite*" SuppressAuthor False)
  , ("textcite", citation "textcite" AuthorInText False)
  , ("citet", citation "citet" AuthorInText False)
  , ("citet*", citation "citet*" AuthorInText False)
  , ("citealt", citation "citealt" AuthorInText False)
  , ("citealt*", citation "citealt*" AuthorInText False)
  , ("textcites", citation "textcites" AuthorInText True)
  , ("cites", citation "cites" NormalCitation True)
  , ("autocites", citation "autocites" NormalCitation True)
  , ("footcites", inNote <$> citation "footcites" NormalCitation True)
  , ("parencites", citation "parencites" NormalCitation True)
  , ("supercites", citation "supercites" NormalCitation True)
  , ("footcitetexts", inNote <$> citation "footcitetexts" NormalCitation True)
  , ("Autocite", citation "Autocite" NormalCitation False)
  , ("Smartcite", citation "Smartcite" NormalCitation False)
  , ("Footcite", citation "Footcite" NormalCitation False)
  , ("Parencite", citation "Parencite" NormalCitation False)
  , ("Supercite", citation "Supercite" NormalCitation False)
  , ("Footcitetext", inNote <$> citation "Footcitetext" NormalCitation False)
  , ("Citeyearpar", citation "Citeyearpar" SuppressAuthor False)
  , ("Citeyear", citation "Citeyear" SuppressAuthor False)
  , ("Autocite*", citation "Autocite*" SuppressAuthor False)
  , ("Cite*", citation "Cite*" SuppressAuthor False)
  , ("Parencite*", citation "Parencite*" SuppressAuthor False)
  , ("Textcite", citation "Textcite" AuthorInText False)
  , ("Textcites", citation "Textcites" AuthorInText True)
  , ("Cites", citation "Cites" NormalCitation True)
  , ("Autocites", citation "Autocites" NormalCitation True)
  , ("Footcites", citation "Footcites" NormalCitation True)
  , ("Parencites", citation "Parencites" NormalCitation True)
  , ("Supercites", citation "Supercites" NormalCitation True)
  , ("Footcitetexts", inNote <$> citation "Footcitetexts" NormalCitation True)
  , ("citetext", complexNatbibCitation NormalCitation)
  , ("citeauthor", (try (tok *> optional sp *> controlSeq "citetext") *>
                        complexNatbibCitation AuthorInText)
                   <|> citation "citeauthor" AuthorInText False)
  , ("nocite", mempty <$ (citation "nocite" NormalCitation False >>=
                          addMeta "nocite"))
  , ("hypertarget", braced >> tok)
  -- siuntix
  , ("SI", dosiunitx)
  -- hyphenat
  , ("bshyp", lit "\\\173")
  , ("fshyp", lit "/\173")
  , ("dothyp", lit ".\173")
  , ("colonhyp", lit ":\173")
  , ("hyp", lit "-")
  , ("nohyphens", tok)
  , ("textnhtt", ttfamily)
  , ("nhttfamily", ttfamily)
  -- LaTeX colors
  , ("textcolor", coloredInline "color")
  , ("colorbox", coloredInline "background-color")
  -- fontawesome
  , ("faCheck", lit "\10003")
  , ("faClose", lit "\10007")
  -- xspace
  , ("xspace", doxspace)
  -- etoolbox
  , ("ifstrequal", ifstrequal)
  ]

doTerm :: PandocMonad m => Translations.Term -> LP m Inlines
doTerm term = str <$> translateTerm term

ifstrequal :: PandocMonad m => LP m Inlines
ifstrequal = do
  str1 <- tok
  str2 <- tok
  ifequal <- braced
  ifnotequal <- braced
  if str1 == str2
     then getInput >>= setInput . (ifequal ++)
     else getInput >>= setInput . (ifnotequal ++)
  return mempty

coloredInline :: PandocMonad m => String -> LP m Inlines
coloredInline stylename = do
  skipopts
  color <- braced
  spanWith ("",[],[("style",stylename ++ ": " ++ toksToString color)]) <$> tok

ttfamily :: PandocMonad m => LP m Inlines
ttfamily = (code . stringify . toList) <$> tok

rawInlineOr :: PandocMonad m => Text -> LP m Inlines -> LP m Inlines
rawInlineOr name' fallback = do
  parseRaw <- extensionEnabled Ext_raw_tex <$> getOption readerExtensions
  if parseRaw
     then rawInline "latex" <$> getRawCommand name' ("\\" <> name')
     else fallback

getRawCommand :: PandocMonad m => Text -> Text -> LP m String
getRawCommand name txt = do
  (_, rawargs) <- withRaw $
      case name of
           "write" -> do
             void $ satisfyTok isWordTok -- digits
             void braced
           "titleformat" -> do
             void braced
             skipopts
             void $ count 4 braced
           "def" -> do
             void $ manyTill anyTok braced
           _ -> do
             skipangles
             skipopts
             option "" (try (optional sp *> dimenarg))
             void $ many braced
  return $ T.unpack (txt <> untokenize rawargs)

isBlockCommand :: Text -> Bool
isBlockCommand s =
  s `M.member` (blockCommands :: M.Map Text (LP PandocPure Blocks))
  || s `Set.member` treatAsBlock

treatAsBlock :: Set.Set Text
treatAsBlock = Set.fromList
   [ "let", "def"
   , "newcommand", "renewcommand"
   , "newenvironment", "renewenvironment"
   , "providecommand", "provideenvironment"
     -- newcommand, etc. should be parsed by macroDef, but we need this
     -- here so these aren't parsed as inline commands to ignore
   , "special", "pdfannot", "pdfstringdef"
   , "bibliographystyle"
   , "maketitle", "makeindex", "makeglossary"
   , "addcontentsline", "addtocontents", "addtocounter"
      -- \ignore{} is used conventionally in literate haskell for definitions
      -- that are to be processed by the compiler but not printed.
   , "ignore"
   , "hyperdef"
   , "markboth", "markright", "markleft"
   , "hspace", "vspace"
   , "newpage"
   , "clearpage"
   , "pagebreak"
   , "titleformat"
   ]

isInlineCommand :: Text -> Bool
isInlineCommand s =
  s `M.member` (inlineCommands :: M.Map Text (LP PandocPure Inlines))
  || s `Set.member` treatAsInline

treatAsInline :: Set.Set Text
treatAsInline = Set.fromList
  [ "index"
  , "hspace"
  , "vspace"
  , "noindent"
  , "newpage"
  , "clearpage"
  , "pagebreak"
  ]

dolabel :: PandocMonad m => LP m Inlines
dolabel = do
  v <- braced
  return $ spanWith ("",[],[("label", toksToString v)])
    $ inBrackets $ str $ toksToString v

doref :: PandocMonad m => String -> LP m Inlines
doref cls = do
  v <- braced
  let refstr = toksToString v
  return $ spanWith ("",[],[ ("reference-type", cls)
                           , ("reference", refstr)])
    $ inBrackets $ str refstr

lookupListDefault :: (Show k, Ord k) => v -> [k] -> M.Map k v -> v
lookupListDefault d = (fromMaybe d .) . lookupList
  where lookupList l m = msum $ map (`M.lookup` m) l

inline :: PandocMonad m => LP m Inlines
inline = (mempty <$ comment)
     <|> (space  <$ whitespace)
     <|> (softbreak <$ endline)
     <|> word
     <|> inlineCommand'
     <|> inlineEnvironment
     <|> inlineGroup
     <|> (symbol '-' *>
           option (str "-") (symbol '-' *>
             option (str "–") (str "—" <$ symbol '-')))
     <|> doubleQuote
     <|> singleQuote
     <|> (str "”" <$ try (symbol '\'' >> symbol '\''))
     <|> (str "”" <$ symbol '”')
     <|> (str "’" <$ symbol '\'')
     <|> (str "’" <$ symbol '’')
     <|> (str "\160" <$ symbol '~')
     <|> dollarsMath
     <|> (guardEnabled Ext_literate_haskell *> symbol '|' *> doLHSverb)
     <|> (str . (:[]) <$> primEscape)
     <|> regularSymbol
     <|> (do res <- symbolIn "#^'`\"[]"
             pos <- getPosition
             let s = T.unpack (untoken res)
             report $ ParsingUnescaped s pos
             return $ str s)

inlines :: PandocMonad m => LP m Inlines
inlines = mconcat <$> many inline

-- block elements:

begin_ :: PandocMonad m => Text -> LP m ()
begin_ t = (try $ do
  controlSeq "begin"
  spaces
  txt <- untokenize <$> braced
  guard (t == txt)) <?> ("\\begin{" ++ T.unpack t ++ "}")

end_ :: PandocMonad m => Text -> LP m ()
end_ t = (try $ do
  controlSeq "end"
  spaces
  txt <- untokenize <$> braced
  guard $ t == txt) <?> ("\\end{" ++ T.unpack t ++ "}")

preamble :: PandocMonad m => LP m Blocks
preamble = mempty <$ many preambleBlock
  where preambleBlock =  spaces1
                     <|> void include
                     <|> void macroDef
                     <|> void blockCommand
                     <|> void braced
                     <|> (notFollowedBy (begin_ "document") >> void anyTok)

paragraph :: PandocMonad m => LP m Blocks
paragraph = do
  x <- trimInlines . mconcat <$> many1 inline
  if x == mempty
     then return mempty
     else return $ para x

include :: PandocMonad m => LP m Blocks
include = do
  (Tok _ (CtrlSeq name) _) <-
                    controlSeq "include" <|> controlSeq "input" <|>
                    controlSeq "subfile" <|> controlSeq "usepackage"
  skipMany $ bracketed inline -- skip options
  fs <- (map trim . splitBy (==',') . T.unpack . untokenize) <$> braced
  let fs' = if name == "usepackage"
               then map (maybeAddExtension ".sty") fs
               else map (maybeAddExtension ".tex") fs
  dirs <- (splitBy (==':') . fromMaybe ".") <$> lookupEnv "TEXINPUTS"
  mconcat <$> mapM (insertIncludedFile blocks (tokenize . T.pack) dirs) fs'

maybeAddExtension :: String -> FilePath -> FilePath
maybeAddExtension ext fp =
  if null (takeExtension fp)
     then addExtension fp ext
     else fp

addMeta :: PandocMonad m => ToMetaValue a => String -> a -> LP m ()
addMeta field val = updateState $ \st ->
   st{ sMeta = addMetaField field val $ sMeta st }

authors :: PandocMonad m => LP m ()
authors = try $ do
  bgroup
  let oneAuthor = mconcat <$>
       many1 (notFollowedBy' (controlSeq "and") >>
               (inline <|> mempty <$ blockCommand))
               -- skip e.g. \vspace{10pt}
  auths <- sepBy oneAuthor (controlSeq "and")
  egroup
  addMeta "author" (map trimInlines auths)

macroDef :: PandocMonad m => LP m Blocks
macroDef = do
  mempty <$ ((commandDef <|> environmentDef) <* doMacros 0)
  where commandDef = do
          (name, macro') <- newcommand <|> letmacro <|> defmacro
          guardDisabled Ext_latex_macros <|>
           updateState (\s -> s{ sMacros = M.insert name macro' (sMacros s) })
        environmentDef = do
          (name, macro1, macro2) <- newenvironment
          guardDisabled Ext_latex_macros <|>
            do updateState $ \s -> s{ sMacros =
                M.insert name macro1 (sMacros s) }
               updateState $ \s -> s{ sMacros =
                M.insert ("end" <> name) macro2 (sMacros s) }
        -- @\newenvironment{envname}[n-args][default]{begin}{end}@
        -- is equivalent to
        -- @\newcommand{\envname}[n-args][default]{begin}@
        -- @\newcommand{\endenvname}@

letmacro :: PandocMonad m => LP m (Text, Macro)
letmacro = do
  controlSeq "let"
  Tok _ (CtrlSeq name) _ <- anyControlSeq
  optional $ symbol '='
  spaces
  contents <- braced <|> ((:[]) <$> (anyControlSeq <|> singleChar))
  return (name, Macro ExpandWhenDefined 0 Nothing contents)

defmacro :: PandocMonad m => LP m (Text, Macro)
defmacro = try $ do
  controlSeq "def"
  Tok _ (CtrlSeq name) _ <- anyControlSeq
  numargs <- option 0 $ argSeq 1
  contents <- withVerbatimMode braced
  return (name, Macro ExpandWhenUsed numargs Nothing contents)

-- Note: we don't yet support fancy things like #1.#2
argSeq :: PandocMonad m => Int -> LP m Int
argSeq n = do
  Tok _ (Arg i) _ <- satisfyTok isArgTok
  guard $ i == n
  argSeq (n+1) <|> return n

isArgTok :: Tok -> Bool
isArgTok (Tok _ (Arg _) _) = True
isArgTok _ = False

newcommand :: PandocMonad m => LP m (Text, Macro)
newcommand = do
  pos <- getPosition
  Tok _ (CtrlSeq mtype) _ <- controlSeq "newcommand" <|>
                             controlSeq "renewcommand" <|>
                             controlSeq "providecommand"
  optional $ symbol '*'
  Tok _ (CtrlSeq name) txt <- withVerbatimMode $ anyControlSeq <|>
    (symbol '{' *> spaces *> anyControlSeq <* spaces <* symbol '}')
  spaces
  numargs <- option 0 $ try bracketedNum
  spaces
  optarg <- option Nothing $ Just <$> try bracketedToks
  spaces
  contents <- withVerbatimMode braced
  -- we use withVerbatimMode, because macros are to be expanded
  -- at point of use, not point of definition
  when (mtype == "newcommand") $ do
    macros <- sMacros <$> getState
    case M.lookup name macros of
         Just _ -> report $ MacroAlreadyDefined (T.unpack txt) pos
         Nothing -> return ()
  return (name, Macro ExpandWhenUsed numargs optarg contents)

newenvironment :: PandocMonad m => LP m (Text, Macro, Macro)
newenvironment = do
  pos <- getPosition
  Tok _ (CtrlSeq mtype) _ <- controlSeq "newenvironment" <|>
                             controlSeq "renewenvironment" <|>
                             controlSeq "provideenvironment"
  optional $ symbol '*'
  spaces
  name <- untokenize <$> braced
  spaces
  numargs <- option 0 $ try bracketedNum
  spaces
  optarg <- option Nothing $ Just <$> try bracketedToks
  spaces
  startcontents <- withVerbatimMode braced
  spaces
  endcontents <- withVerbatimMode braced
  when (mtype == "newenvironment") $ do
    macros <- sMacros <$> getState
    case M.lookup name macros of
         Just _ -> report $ MacroAlreadyDefined (T.unpack name) pos
         Nothing -> return ()
  return (name, Macro ExpandWhenUsed numargs optarg startcontents,
             Macro ExpandWhenUsed 0 Nothing endcontents)

bracketedToks :: PandocMonad m => LP m [Tok]
bracketedToks = do
  symbol '['
  manyTill anyTok (symbol ']')

bracketedNum :: PandocMonad m => LP m Int
bracketedNum = do
  ds <- untokenize <$> bracketedToks
  case safeRead (T.unpack ds) of
       Just i -> return i
       _      -> return 0

setCaption :: PandocMonad m => LP m Blocks
setCaption = do
  ils <- tok
  mblabel <- option Nothing $
               try $ spaces >> controlSeq "label" >> (Just <$> tok)
  let ils' = case mblabel of
                  Just lab -> ils <> spanWith
                                ("",[],[("label", stringify lab)]) mempty
                  Nothing  -> ils
  updateState $ \st -> st{ sCaption = Just ils' }
  return mempty

looseItem :: PandocMonad m => LP m Blocks
looseItem = do
  inListItem <- sInListItem <$> getState
  guard $ not inListItem
  skipopts
  return mempty

resetCaption :: PandocMonad m => LP m ()
resetCaption = updateState $ \st -> st{ sCaption = Nothing }

section :: PandocMonad m => Bool -> Attr -> Int -> LP m Blocks
section starred (ident, classes, kvs) lvl = do
  skipopts
  contents <- grouped inline
  lab <- option ident $
          try (spaces >> controlSeq "label"
               >> spaces >> toksToString <$> braced)
  let classes' = if starred then "unnumbered" : classes else classes
  unless starred $ do
    hn <- sLastHeaderNum <$> getState
    let num = incrementHeaderNum lvl hn
    updateState $ \st -> st{ sLastHeaderNum = num }
    updateState $ \st -> st{ sLabels = M.insert lab
                            [Str (renderHeaderNum num)]
                            (sLabels st) }
  attr' <- registerHeader (lab, classes', kvs) contents
  return $ headerWith attr' lvl contents

blockCommand :: PandocMonad m => LP m Blocks
blockCommand = try $ do
  Tok _ (CtrlSeq name) txt <- anyControlSeq
  guard $ name /= "begin" && name /= "end"
  star <- option "" ("*" <$ symbol '*' <* optional sp)
  let name' = name <> star
  let names = ordNub [name', name]
  let rawDefiniteBlock = do
        guard $ isBlockCommand name
        rawBlock "latex" <$> getRawCommand name (txt <> star)
  -- heuristic:  if it could be either block or inline, we
  -- treat it if block if we have a sequence of block
  -- commands followed by a newline.  But we stop if we
  -- hit a \startXXX, since this might start a raw ConTeXt
  -- environment (this is important because this parser is
  -- used by the Markdown reader).
  let startCommand = try $ do
        Tok _ (CtrlSeq n) _ <- anyControlSeq
        guard $ "start" `T.isPrefixOf` n
  let rawMaybeBlock = try $ do
        guard $ not $ isInlineCommand name
        curr <- rawBlock "latex" <$> getRawCommand name (txt <> star)
        rest <- many $ notFollowedBy startCommand *> blockCommand
        lookAhead $ blankline <|> startCommand
        return $ curr <> mconcat rest
  let raw = rawDefiniteBlock <|> rawMaybeBlock
  lookupListDefault raw names blockCommands

closing :: PandocMonad m => LP m Blocks
closing = do
  contents <- tok
  st <- getState
  let extractInlines (MetaBlocks [Plain ys]) = ys
      extractInlines (MetaBlocks [Para ys ]) = ys
      extractInlines _                       = []
  let sigs = case lookupMeta "author" (sMeta st) of
                  Just (MetaList xs) ->
                    para $ trimInlines $ fromList $
                      intercalate [LineBreak] $ map extractInlines xs
                  _ -> mempty
  return $ para (trimInlines contents) <> sigs

blockCommands :: PandocMonad m => M.Map Text (LP m Blocks)
blockCommands = M.fromList $
   [ ("par", mempty <$ skipopts)
   , ("parbox",  braced >> grouped blocks)
   , ("title", mempty <$ (skipopts *>
                             (grouped inline >>= addMeta "title")
                         <|> (grouped block >>= addMeta "title")))
   , ("subtitle", mempty <$ (skipopts *> tok >>= addMeta "subtitle"))
   , ("author", mempty <$ (skipopts *> authors))
   -- -- in letter class, temp. store address & sig as title, author
   , ("address", mempty <$ (skipopts *> tok >>= addMeta "address"))
   , ("signature", mempty <$ (skipopts *> authors))
   , ("date", mempty <$ (skipopts *> tok >>= addMeta "date"))
   -- Koma-script metadata commands
   , ("dedication", mempty <$ (skipopts *> tok >>= addMeta "dedication"))
   -- sectioning
   , ("part", section False nullAttr (-1))
   , ("part*", section True nullAttr (-1))
   , ("chapter", section False nullAttr 0)
   , ("chapter*", section True ("",["unnumbered"],[]) 0)
   , ("section", section False nullAttr 1)
   , ("section*", section True ("",["unnumbered"],[]) 1)
   , ("subsection", section False nullAttr 2)
   , ("subsection*", section True ("",["unnumbered"],[]) 2)
   , ("subsubsection", section False nullAttr 3)
   , ("subsubsection*", section True ("",["unnumbered"],[]) 3)
   , ("paragraph", section False nullAttr 4)
   , ("paragraph*", section True ("",["unnumbered"],[]) 4)
   , ("subparagraph", section False nullAttr 5)
   , ("subparagraph*", section True ("",["unnumbered"],[]) 5)
   -- beamer slides
   , ("frametitle", section False nullAttr 3)
   , ("framesubtitle", section False nullAttr 4)
   -- letters
   , ("opening", (para . trimInlines) <$> (skipopts *> tok))
   , ("closing", skipopts *> closing)
   -- memoir
   , ("plainbreak", braced >> pure horizontalRule)
   , ("plainbreak*", braced >> pure horizontalRule)
   , ("fancybreak", braced >> pure horizontalRule)
   , ("fancybreak*", braced >> pure horizontalRule)
   , ("plainfancybreak", braced >> braced >> braced >> pure horizontalRule)
   , ("plainfancybreak*", braced >> braced >> braced >> pure horizontalRule)
   , ("pfbreak", pure horizontalRule)
   , ("pfbreak*", pure horizontalRule)
   --
   , ("hrule", pure horizontalRule)
   , ("strut", pure mempty)
   , ("rule", skipopts *> tok *> tok *> pure horizontalRule)
   , ("item", looseItem)
   , ("documentclass", skipopts *> braced *> preamble)
   , ("centerline", (para . trimInlines) <$> (skipopts *> tok))
   , ("caption", skipopts *> setCaption)
   , ("bibliography", mempty <$ (skipopts *> braced >>=
         addMeta "bibliography" . splitBibs . toksToString))
   , ("addbibresource", mempty <$ (skipopts *> braced >>=
         addMeta "bibliography" . splitBibs . toksToString))
   -- includes
   , ("lstinputlisting", inputListing)
   , ("graphicspath", graphicsPath)
   -- polyglossia
   , ("setdefaultlanguage", setDefaultLanguage)
   , ("setmainlanguage", setDefaultLanguage)
   -- hyperlink
   , ("hypertarget", try $ braced >> grouped block)
   -- LaTeX colors
   , ("textcolor", coloredBlock "color")
   , ("colorbox", coloredBlock "background-color")
   ]


environments :: PandocMonad m => M.Map Text (LP m Blocks)
environments = M.fromList
   [ ("document", env "document" blocks)
   , ("abstract", mempty <$ (env "abstract" blocks >>= addMeta "abstract"))
   , ("letter", env "letter" letterContents)
   , ("minipage", env "minipage" $
          skipopts *> spaces *> optional braced *> spaces *> blocks)
   , ("figure", env "figure" $ skipopts *> figure)
   , ("subfigure", env "subfigure" $ skipopts *> tok *> figure)
   , ("center", env "center" blocks)
   , ("longtable",  env "longtable" $
          resetCaption *> simpTable "longtable" False >>= addTableCaption)
   , ("table",  env "table" $
          resetCaption *> skipopts *> blocks >>= addTableCaption)
   , ("tabular*", env "tabular" $ simpTable "tabular*" True)
   , ("tabularx", env "tabularx" $ simpTable "tabularx" True)
   , ("tabular", env "tabular"  $ simpTable "tabular" False)
   , ("quote", blockQuote <$> env "quote" blocks)
   , ("quotation", blockQuote <$> env "quotation" blocks)
   , ("verse", blockQuote <$> env "verse" blocks)
   , ("itemize", bulletList <$> listenv "itemize" (many item))
   , ("description", definitionList <$> listenv "description" (many descItem))
   , ("enumerate", orderedList')
   , ("alltt", alltt <$> env "alltt" blocks)
   , ("code", guardEnabled Ext_literate_haskell *>
       (codeBlockWith ("",["sourceCode","literate","haskell"],[]) <$>
         verbEnv "code"))
   , ("comment", mempty <$ verbEnv "comment")
   , ("verbatim", codeBlock <$> verbEnv "verbatim")
   , ("Verbatim", fancyverbEnv "Verbatim")
   , ("BVerbatim", fancyverbEnv "BVerbatim")
   , ("lstlisting", do attr <- parseListingsOptions <$> option [] keyvals
                       codeBlockWith attr <$> verbEnv "lstlisting")
    , ("minted", minted)
   , ("obeylines", obeylines)
   , ("displaymath", mathEnvWith para Nothing "displaymath")
   , ("equation", mathEnvWith para Nothing "equation")
   , ("equation*", mathEnvWith para Nothing "equation*")
   , ("gather", mathEnvWith para (Just "gathered") "gather")
   , ("gather*", mathEnvWith para (Just "gathered") "gather*")
   , ("multline", mathEnvWith para (Just "gathered") "multline")
   , ("multline*", mathEnvWith para (Just "gathered") "multline*")
   , ("eqnarray", mathEnvWith para (Just "aligned") "eqnarray")
   , ("eqnarray*", mathEnvWith para (Just "aligned") "eqnarray*")
   , ("align", mathEnvWith para (Just "aligned") "align")
   , ("align*", mathEnvWith para (Just "aligned") "align*")
   , ("alignat", mathEnvWith para (Just "aligned") "alignat")
   , ("alignat*", mathEnvWith para (Just "aligned") "alignat*")
   , ("tikzpicture", rawVerbEnv "tikzpicture")
   ]

environment :: PandocMonad m => LP m Blocks
environment = do
  controlSeq "begin"
  name <- untokenize <$> braced
  M.findWithDefault mzero name environments
    <|> rawEnv name

env :: PandocMonad m => Text -> LP m a -> LP m a
env name p = p <* end_ name

rawEnv :: PandocMonad m => Text -> LP m Blocks
rawEnv name = do
  exts <- getOption readerExtensions
  let parseRaw = extensionEnabled Ext_raw_tex exts
  rawOptions <- mconcat <$> many rawopt
  let beginCommand = "\\begin{" <> name <> "}" <> rawOptions
  pos1 <- getPosition
  (bs, raw) <- withRaw $ env name blocks
  if parseRaw
     then return $ rawBlock "latex"
                 $ T.unpack $ beginCommand <> untokenize raw
     else do
       unless parseRaw $ do
         report $ SkippedContent (T.unpack beginCommand) pos1
       pos2 <- getPosition
       report $ SkippedContent ("\\end{" ++ T.unpack name ++ "}") pos2
       return bs

rawVerbEnv :: PandocMonad m => Text -> LP m Blocks
rawVerbEnv name = do
  pos <- getPosition
  (_, raw) <- withRaw $ verbEnv name
  let raw' = "\\begin{tikzpicture}" ++ toksToString raw
  exts <- getOption readerExtensions
  let parseRaw = extensionEnabled Ext_raw_tex exts
  if parseRaw
     then return $ rawBlock "latex" raw'
     else do
       report $ SkippedContent raw' pos
       return mempty

verbEnv :: PandocMonad m => Text -> LP m String
verbEnv name = withVerbatimMode $ do
  skipopts
  optional blankline
  res <- manyTill anyTok (end_ name)
  return $ stripTrailingNewlines $ toksToString res

fancyverbEnv :: PandocMonad m => Text -> LP m Blocks
fancyverbEnv name = do
  options <- option [] keyvals
  let kvs = [ (if k == "firstnumber"
                  then "startFrom"
                  else k, v) | (k,v) <- options ]
  let classes = [ "numberLines" |
                  lookup "numbers" options == Just "left" ]
  let attr = ("",classes,kvs)
  codeBlockWith attr <$> verbEnv name

obeylines :: PandocMonad m => LP m Blocks
obeylines = do
  para . fromList . removeLeadingTrailingBreaks .
     walk softBreakToHard . toList <$> env "obeylines" inlines
  where softBreakToHard SoftBreak = LineBreak
        softBreakToHard x         = x
        removeLeadingTrailingBreaks = reverse . dropWhile isLineBreak .
                                      reverse . dropWhile isLineBreak
        isLineBreak LineBreak     = True
        isLineBreak _             = False

minted :: PandocMonad m => LP m Blocks
minted = do
  options <- option [] keyvals
  lang <- toksToString <$> braced
  let kvs = [ (if k == "firstnumber"
                  then "startFrom"
                  else k, v) | (k,v) <- options ]
  let classes = [ lang | not (null lang) ] ++
                [ "numberLines" |
                  lookup "linenos" options == Just "true" ]
  let attr = ("",classes,kvs)
  codeBlockWith attr <$> verbEnv "minted"

letterContents :: PandocMonad m => LP m Blocks
letterContents = do
  bs <- blocks
  st <- getState
  -- add signature (author) and address (title)
  let addr = case lookupMeta "address" (sMeta st) of
                  Just (MetaBlocks [Plain xs]) ->
                     para $ trimInlines $ fromList xs
                  _ -> mempty
  return $ addr <> bs -- sig added by \closing

figure :: PandocMonad m => LP m Blocks
figure = try $ do
  resetCaption
  blocks >>= addImageCaption

addImageCaption :: PandocMonad m => Blocks -> LP m Blocks
addImageCaption = walkM go
  where go (Image attr alt (src,tit))
            | not ("fig:" `isPrefixOf` tit) = do
          mbcapt <- sCaption <$> getState
          return $ case mbcapt of
               Just ils -> Image attr (toList ils) (src, "fig:" ++ tit)
               Nothing  -> Image attr alt (src,tit)
        go x = return x

coloredBlock :: PandocMonad m => String -> LP m Blocks
coloredBlock stylename = try $ do
  skipopts
  color <- braced
  notFollowedBy (grouped inline)
  let constructor = divWith ("",[],[("style",stylename ++ ": " ++ toksToString color)])
  constructor <$> grouped block

graphicsPath :: PandocMonad m => LP m Blocks
graphicsPath = do
  ps <- map toksToString <$> (bgroup *> manyTill braced egroup)
  getResourcePath >>= setResourcePath . (++ ps)
  return mempty

splitBibs :: String -> [Inlines]
splitBibs = map (str . flip replaceExtension "bib" . trim) . splitBy (==',')

alltt :: Blocks -> Blocks
alltt = walk strToCode
  where strToCode (Str s)   = Code nullAttr s
        strToCode Space     = RawInline (Format "latex") "\\ "
        strToCode SoftBreak = LineBreak
        strToCode x         = x

parseListingsOptions :: [(String, String)] -> Attr
parseListingsOptions options =
  let kvs = [ (if k == "firstnumber"
                  then "startFrom"
                  else k, v) | (k,v) <- options ]
      classes = [ "numberLines" |
                  lookup "numbers" options == Just "left" ]
             ++ maybeToList (lookup "language" options
                     >>= fromListingsLanguage)
  in  (fromMaybe "" (lookup "label" options), classes, kvs)

inputListing :: PandocMonad m => LP m Blocks
inputListing = do
  pos <- getPosition
  options <- option [] keyvals
  f <- filter (/='"') . toksToString <$> braced
  dirs <- (splitBy (==':') . fromMaybe ".") <$> lookupEnv "TEXINPUTS"
  mbCode <- readFileFromDirs dirs f
  codeLines <- case mbCode of
                      Just s -> return $ lines s
                      Nothing -> do
                        report $ CouldNotLoadIncludeFile f pos
                        return []
  let (ident,classes,kvs) = parseListingsOptions options
  let language = case lookup "language" options >>= fromListingsLanguage of
                      Just l -> [l]
                      Nothing -> take 1 $ languagesByExtension (takeExtension f)
  let firstline = fromMaybe 1 $ lookup "firstline" options >>= safeRead
  let lastline = fromMaybe (length codeLines) $
                       lookup "lastline" options >>= safeRead
  let codeContents = intercalate "\n" $ take (1 + lastline - firstline) $
                       drop (firstline - 1) codeLines
  return $ codeBlockWith (ident,ordNub (classes ++ language),kvs) codeContents

-- lists

item :: PandocMonad m => LP m Blocks
item = void blocks *> controlSeq "item" *> skipopts *> blocks

descItem :: PandocMonad m => LP m (Inlines, [Blocks])
descItem = do
  blocks -- skip blocks before item
  controlSeq "item"
  optional sp
  ils <- opt
  bs <- blocks
  return (ils, [bs])

listenv :: PandocMonad m => Text -> LP m a -> LP m a
listenv name p = try $ do
  oldInListItem <- sInListItem `fmap` getState
  updateState $ \st -> st{ sInListItem = True }
  res <- env name p
  updateState $ \st -> st{ sInListItem = oldInListItem }
  return res

orderedList' :: PandocMonad m => LP m Blocks
orderedList' = try $ do
  spaces
  let markerSpec = do
        symbol '['
        ts <- toksToString <$> manyTill anyTok (symbol ']')
        case runParser anyOrderedListMarker def "option" ts of
             Right r -> return r
             Left _  -> do
               pos <- getPosition
               report $ SkippedContent ("[" ++ ts ++ "]") pos
               return (1, DefaultStyle, DefaultDelim)
  (_, style, delim) <- option (1, DefaultStyle, DefaultDelim) markerSpec
  spaces
  optional $ try $ controlSeq "setlength"
                   *> grouped (count 1 $ controlSeq "itemindent")
                   *> braced
  spaces
  start <- option 1 $ try $ do pos <- getPosition
                               controlSeq "setcounter"
                               ctr <- toksToString <$> braced
                               guard $ "enum" `isPrefixOf` ctr
                               guard $ all (`elem` ['i','v']) (drop 4 ctr)
                               optional sp
                               num <- toksToString <$> braced
                               case safeRead num of
                                    Just i -> return (i + 1 :: Int)
                                    Nothing -> do
                                      report $ SkippedContent
                                        ("\\setcounter{" ++ ctr ++
                                         "}{" ++ num ++ "}") pos
                                      return 1
  bs <- listenv "enumerate" (many item)
  return $ orderedListWith (start, style, delim) bs

-- tables

hline :: PandocMonad m => LP m ()
hline = try $ do
  spaces
  controlSeq "hline" <|>
    -- booktabs rules:
    controlSeq "toprule" <|>
    controlSeq "bottomrule" <|>
    controlSeq "midrule" <|>
    controlSeq "endhead" <|>
    controlSeq "endfirsthead"
  spaces
  optional $ bracketed inline
  return ()

lbreak :: PandocMonad m => LP m Tok
lbreak = (controlSeq "\\" <|> controlSeq "tabularnewline") <* spaces

amp :: PandocMonad m => LP m Tok
amp = symbol '&'

-- Split a Word into individual Symbols (for parseAligns)
splitWordTok :: PandocMonad m => LP m ()
splitWordTok = do
  inp <- getInput
  case inp of
       (Tok spos Word t : rest) -> do
         setInput $ map (Tok spos Symbol . T.singleton) (T.unpack t) ++ rest
       _ -> return ()

parseAligns :: PandocMonad m => LP m [(Alignment, Double, ([Tok], [Tok]))]
parseAligns = try $ do
  let maybeBar = skipMany $
        sp <|> () <$ symbol '|' <|> () <$ (symbol '@' >> braced)
  let cAlign = AlignCenter <$ symbol 'c'
  let lAlign = AlignLeft <$ symbol 'l'
  let rAlign = AlignRight <$ symbol 'r'
  let parAlign = AlignLeft <$ symbol 'p'
  -- aligns from tabularx
  let xAlign = AlignLeft <$ symbol 'X'
  let mAlign = AlignLeft <$ symbol 'm'
  let bAlign = AlignLeft <$ symbol 'b'
  let alignChar = splitWordTok *> (  cAlign <|> lAlign <|> rAlign <|> parAlign
                                 <|> xAlign <|> mAlign <|> bAlign )
  let alignPrefix = symbol '>' >> braced
  let alignSuffix = symbol '<' >> braced
  let colWidth = try $ do
        symbol '{'
        ds <- trim . toksToString <$> manyTill anyTok (controlSeq "linewidth")
        spaces
        symbol '}'
        case safeRead ds of
              Just w  -> return w
              Nothing -> return 0.0
  let alignSpec = try $ do
        spaces
        pref <- option [] alignPrefix
        spaces
        al <- alignChar
        width <- colWidth <|> option 0.0 (do s <- toksToString <$> braced
                                             pos <- getPosition
                                             report $ SkippedContent s pos
                                             return 0.0)
        spaces
        suff <- option [] alignSuffix
        return (al, width, (pref, suff))
  bgroup
  spaces
  maybeBar
  aligns' <- many (alignSpec <* maybeBar)
  spaces
  egroup
  spaces
  return aligns'

parseTableRow :: PandocMonad m
              => Text   -- ^ table environment name
              -> [([Tok], [Tok])] -- ^ pref/suffixes
              -> LP m [Blocks]
parseTableRow envname prefsufs = do
  notFollowedBy (spaces *> end_ envname)
  let cols = length prefsufs
  -- add prefixes and suffixes in token stream:
  let celltoks (pref, suff) = do
        prefpos <- getPosition
        contents <- many (notFollowedBy
                         (() <$ amp <|> () <$ lbreak <|> end_ envname)
                         >> anyTok)
        suffpos <- getPosition
        option [] (count 1 amp)
        return $ map (setpos (sourceLine prefpos, sourceColumn prefpos)) pref
                 ++ contents ++
                 map (setpos (sourceLine suffpos, sourceColumn suffpos)) suff
  rawcells <- sequence (map celltoks prefsufs)
  oldInput <- getInput
  cells <- sequence $ map (\ts -> setInput ts >> parseTableCell) rawcells
  setInput oldInput
  spaces
  let numcells = length cells
  guard $ numcells <= cols && numcells >= 1
  guard $ cells /= [mempty]
  -- note:  a & b in a three-column table leaves an empty 3rd cell:
  return $ cells ++ replicate (cols - numcells) mempty

parseTableCell :: PandocMonad m => LP m Blocks
parseTableCell = do
  let plainify bs = case toList bs of
                         [Para ils] -> plain (fromList ils)
                         _          -> bs
  updateState $ \st -> st{ sInTableCell = True }
  cells <- plainify <$> blocks
  updateState $ \st -> st{ sInTableCell = False }
  return cells

simpTable :: PandocMonad m => Text -> Bool -> LP m Blocks
simpTable envname hasWidthParameter = try $ do
  when hasWidthParameter $ () <$ (spaces >> tok)
  skipopts
  colspecs <- parseAligns
  let (aligns, widths, prefsufs) = unzip3 colspecs
  let cols = length colspecs
  optional $ controlSeq "caption" *> skipopts *> setCaption
  optional lbreak
  spaces
  skipMany hline
  spaces
  header' <- option [] $ try (parseTableRow envname prefsufs <*
                                   lbreak <* many1 hline)
  spaces
  rows <- sepEndBy (parseTableRow envname prefsufs)
                    (lbreak <* optional (skipMany hline))
  spaces
  optional $ controlSeq "caption" *> skipopts *> setCaption
  optional lbreak
  spaces
  let header'' = if null header'
                    then replicate cols mempty
                    else header'
  lookAhead $ controlSeq "end" -- make sure we're at end
  return $ table mempty (zip aligns widths) header'' rows

addTableCaption :: PandocMonad m => Blocks -> LP m Blocks
addTableCaption = walkM go
  where go (Table c als ws hs rs) = do
          mbcapt <- sCaption <$> getState
          return $ case mbcapt of
               Just ils -> Table (toList ils) als ws hs rs
               Nothing  -> Table c als ws hs rs
        go x = return x


block :: PandocMonad m => LP m Blocks
block = (mempty <$ spaces1)
    <|> environment
    <|> include
    <|> macroDef
    <|> blockCommand
    <|> paragraph
    <|> grouped block

blocks :: PandocMonad m => LP m Blocks
blocks = mconcat <$> many block

setDefaultLanguage :: PandocMonad m => LP m Blocks
setDefaultLanguage = do
  o <- option "" $ (T.unpack . T.filter (\c -> c /= '[' && c /= ']'))
                <$> rawopt
  polylang <- toksToString <$> braced
  case polyglossiaLangToBCP47 polylang o of
       Nothing -> return mempty -- TODO mzero? warning?
       Just l -> do
         setTranslations l
         updateState $ setMeta "lang" $ str (renderLang l)
         return mempty

polyglossiaLangToBCP47 :: String -> String -> Maybe Lang
polyglossiaLangToBCP47 s o =
  case (s, filter (/=' ') o) of
       ("arabic", "locale=algeria") -> Just $ Lang "ar" "" "DZ" []
       ("arabic", "locale=mashriq") -> Just $ Lang "ar" "" "SY" []
       ("arabic", "locale=libya") -> Just $ Lang "ar" "" "LY" []
       ("arabic", "locale=morocco") -> Just $ Lang "ar" "" "MA" []
       ("arabic", "locale=mauritania") -> Just $ Lang "ar" "" "MR" []
       ("arabic", "locale=tunisia") -> Just $ Lang "ar" "" "TN" []
       ("german", "spelling=old") -> Just $ Lang "de" "" "DE" ["1901"]
       ("german", "variant=austrian,spelling=old")
                                  -> Just $ Lang "de" "" "AT" ["1901"]
       ("german", "variant=austrian") -> Just $ Lang "de" "" "AT" []
       ("german", "variant=swiss,spelling=old")
                                  -> Just $ Lang "de" "" "CH" ["1901"]
       ("german", "variant=swiss") -> Just $ Lang "de" "" "CH" []
       ("german", _) -> Just $ Lang "de" "" "" []
       ("lsorbian", _) -> Just $ Lang "dsb" "" "" []
       ("greek", "variant=poly") -> Just $ Lang "el" "" "polyton" []
       ("english", "variant=australian") -> Just $ Lang "en" "" "AU" []
       ("english", "variant=canadian") -> Just $ Lang "en" "" "CA" []
       ("english", "variant=british") -> Just $ Lang "en" "" "GB" []
       ("english", "variant=newzealand") -> Just $ Lang "en" "" "NZ" []
       ("english", "variant=american") -> Just $ Lang "en" "" "US" []
       ("greek", "variant=ancient") -> Just $ Lang "grc" "" "" []
       ("usorbian", _) -> Just $ Lang "hsb" "" "" []
       ("latin", "variant=classic") -> Just $ Lang "la" "" "" ["x-classic"]
       ("slovenian", _) -> Just $ Lang "sl" "" "" []
       ("serbianc", _) -> Just $ Lang "sr" "cyrl" "" []
       ("pinyin", _) -> Just $ Lang "zh" "Latn" "" ["pinyin"]
       ("afrikaans", _) -> Just $ Lang "af" "" "" []
       ("amharic", _) -> Just $ Lang "am" "" "" []
       ("arabic", _) -> Just $ Lang "ar" "" "" []
       ("assamese", _) -> Just $ Lang "as" "" "" []
       ("asturian", _) -> Just $ Lang "ast" "" "" []
       ("bulgarian", _) -> Just $ Lang "bg" "" "" []
       ("bengali", _) -> Just $ Lang "bn" "" "" []
       ("tibetan", _) -> Just $ Lang "bo" "" "" []
       ("breton", _) -> Just $ Lang "br" "" "" []
       ("catalan", _) -> Just $ Lang "ca" "" "" []
       ("welsh", _) -> Just $ Lang "cy" "" "" []
       ("czech", _) -> Just $ Lang "cs" "" "" []
       ("coptic", _) -> Just $ Lang "cop" "" "" []
       ("danish", _) -> Just $ Lang "da" "" "" []
       ("divehi", _) -> Just $ Lang "dv" "" "" []
       ("greek", _) -> Just $ Lang "el" "" "" []
       ("english", _) -> Just $ Lang "en" "" "" []
       ("esperanto", _) -> Just $ Lang "eo" "" "" []
       ("spanish", _) -> Just $ Lang "es" "" "" []
       ("estonian", _) -> Just $ Lang "et" "" "" []
       ("basque", _) -> Just $ Lang "eu" "" "" []
       ("farsi", _) -> Just $ Lang "fa" "" "" []
       ("finnish", _) -> Just $ Lang "fi" "" "" []
       ("french", _) -> Just $ Lang "fr" "" "" []
       ("friulan", _) -> Just $ Lang "fur" "" "" []
       ("irish", _) -> Just $ Lang "ga" "" "" []
       ("scottish", _) -> Just $ Lang "gd" "" "" []
       ("ethiopic", _) -> Just $ Lang "gez" "" "" []
       ("galician", _) -> Just $ Lang "gl" "" "" []
       ("hebrew", _) -> Just $ Lang "he" "" "" []
       ("hindi", _) -> Just $ Lang "hi" "" "" []
       ("croatian", _) -> Just $ Lang "hr" "" "" []
       ("magyar", _) -> Just $ Lang "hu" "" "" []
       ("armenian", _) -> Just $ Lang "hy" "" "" []
       ("interlingua", _) -> Just $ Lang "ia" "" "" []
       ("indonesian", _) -> Just $ Lang "id" "" "" []
       ("icelandic", _) -> Just $ Lang "is" "" "" []
       ("italian", _) -> Just $ Lang "it" "" "" []
       ("japanese", _) -> Just $ Lang "jp" "" "" []
       ("khmer", _) -> Just $ Lang "km" "" "" []
       ("kurmanji", _) -> Just $ Lang "kmr" "" "" []
       ("kannada", _) -> Just $ Lang "kn" "" "" []
       ("korean", _) -> Just $ Lang "ko" "" "" []
       ("latin", _) -> Just $ Lang "la" "" "" []
       ("lao", _) -> Just $ Lang "lo" "" "" []
       ("lithuanian", _) -> Just $ Lang "lt" "" "" []
       ("latvian", _) -> Just $ Lang "lv" "" "" []
       ("malayalam", _) -> Just $ Lang "ml" "" "" []
       ("mongolian", _) -> Just $ Lang "mn" "" "" []
       ("marathi", _) -> Just $ Lang "mr" "" "" []
       ("dutch", _) -> Just $ Lang "nl" "" "" []
       ("nynorsk", _) -> Just $ Lang "nn" "" "" []
       ("norsk", _) -> Just $ Lang "no" "" "" []
       ("nko", _) -> Just $ Lang "nqo" "" "" []
       ("occitan", _) -> Just $ Lang "oc" "" "" []
       ("panjabi", _) -> Just $ Lang "pa" "" "" []
       ("polish", _) -> Just $ Lang "pl" "" "" []
       ("piedmontese", _) -> Just $ Lang "pms" "" "" []
       ("portuguese", _) -> Just $ Lang "pt" "" "" []
       ("romansh", _) -> Just $ Lang "rm" "" "" []
       ("romanian", _) -> Just $ Lang "ro" "" "" []
       ("russian", _) -> Just $ Lang "ru" "" "" []
       ("sanskrit", _) -> Just $ Lang "sa" "" "" []
       ("samin", _) -> Just $ Lang "se" "" "" []
       ("slovak", _) -> Just $ Lang "sk" "" "" []
       ("albanian", _) -> Just $ Lang "sq" "" "" []
       ("serbian", _) -> Just $ Lang "sr" "" "" []
       ("swedish", _) -> Just $ Lang "sv" "" "" []
       ("syriac", _) -> Just $ Lang "syr" "" "" []
       ("tamil", _) -> Just $ Lang "ta" "" "" []
       ("telugu", _) -> Just $ Lang "te" "" "" []
       ("thai", _) -> Just $ Lang "th" "" "" []
       ("turkmen", _) -> Just $ Lang "tk" "" "" []
       ("turkish", _) -> Just $ Lang "tr" "" "" []
       ("ukrainian", _) -> Just $ Lang "uk" "" "" []
       ("urdu", _) -> Just $ Lang "ur" "" "" []
       ("vietnamese", _) -> Just $ Lang "vi" "" "" []
       _ -> Nothing