aboutsummaryrefslogtreecommitdiff
path: root/src/Text/Pandoc/Writers/LaTeX.hs
blob: 0ef04db3f3cda02a2e78f2c6bd5a65b1b895b6cc (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
{-# LANGUAGE NoImplicitPrelude #-}
{-# LANGUAGE OverloadedStrings   #-}
{-# LANGUAGE PatternGuards       #-}
{-# LANGUAGE ScopedTypeVariables #-}
{- |
   Module      : Text.Pandoc.Writers.LaTeX
   Copyright   : Copyright (C) 2006-2019 John MacFarlane
   License     : GNU GPL, version 2 or above

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

Conversion of 'Pandoc' format into LaTeX.
-}
module Text.Pandoc.Writers.LaTeX (
    writeLaTeX
  , writeBeamer
  ) where
import Prelude
import Control.Applicative ((<|>))
import Control.Monad.State.Strict
import Data.Aeson (FromJSON, object, (.=))
import Data.Char (isAlphaNum, isAscii, isDigit, isLetter, isSpace,
                  isPunctuation, ord, toLower)
import Data.List (foldl', intercalate, intersperse, isInfixOf, nubBy,
                  stripPrefix, (\\), uncons)
import Data.Maybe (catMaybes, fromMaybe, isJust, mapMaybe, isNothing)
import qualified Data.Map as M
import Data.Text (Text)
import qualified Data.Text as T
import Network.URI (unEscapeString)
import Text.Pandoc.BCP47 (Lang (..), getLang, renderLang)
import Text.Pandoc.Class (PandocMonad, report, toLang)
import Text.Pandoc.Definition
import Text.Pandoc.Highlighting (formatLaTeXBlock, formatLaTeXInline, highlight,
                                 styleToLaTeX, toListingsLanguage)
import Text.Pandoc.ImageSize
import Text.Pandoc.Logging
import Text.Pandoc.Options
import Text.Pandoc.Pretty
import Text.Pandoc.Shared
import Text.Pandoc.Slides
import Text.Pandoc.Templates
import Text.Pandoc.Walk
import Text.Pandoc.Writers.Shared
import qualified Text.Parsec as P
import Text.Printf (printf)
import qualified Data.Text.Normalize as Normalize

data WriterState =
  WriterState { stInNote        :: Bool          -- true if we're in a note
              , stInQuote       :: Bool          -- true if in a blockquote
              , stInMinipage    :: Bool          -- true if in minipage
              , stInHeading     :: Bool          -- true if in a section heading
              , stInItem        :: Bool          -- true if in \item[..]
              , stNotes         :: [Doc]         -- notes in a minipage
              , stOLLevel       :: Int           -- level of ordered list nesting
              , stOptions       :: WriterOptions -- writer options, so they don't have to be parameter
              , stVerbInNote    :: Bool          -- true if document has verbatim text in note
              , stTable         :: Bool          -- true if document has a table
              , stStrikeout     :: Bool          -- true if document has strikeout
              , stUrl           :: Bool          -- true if document has visible URL link
              , stGraphics      :: Bool          -- true if document contains images
              , stLHS           :: Bool          -- true if document has literate haskell code
              , stHasChapters   :: Bool          -- true if document has chapters
              , stCsquotes      :: Bool          -- true if document uses csquotes
              , stHighlighting  :: Bool          -- true if document has highlighted code
              , stIncremental   :: Bool          -- true if beamer lists should be displayed bit by bit
              , stInternalLinks :: [String]      -- list of internal link targets
              , stBeamer        :: Bool          -- produce beamer
              , stEmptyLine     :: Bool          -- true if no content on line
              }

startingState :: WriterOptions -> WriterState
startingState options = WriterState {
                  stInNote = False
                , stInQuote = False
                , stInMinipage = False
                , stInHeading = False
                , stInItem = False
                , stNotes = []
                , stOLLevel = 1
                , stOptions = options
                , stVerbInNote = False
                , stTable = False
                , stStrikeout = False
                , stUrl = False
                , stGraphics = False
                , stLHS = False
                , stHasChapters = case writerTopLevelDivision options of
                                    TopLevelPart    -> True
                                    TopLevelChapter -> True
                                    _               -> False
                , stCsquotes = False
                , stHighlighting = False
                , stIncremental = writerIncremental options
                , stInternalLinks = []
                , stBeamer = False
                , stEmptyLine = True }

-- | Convert Pandoc to LaTeX.
writeLaTeX :: PandocMonad m => WriterOptions -> Pandoc -> m Text
writeLaTeX options document =
  evalStateT (pandocToLaTeX options document) $
    startingState options

-- | Convert Pandoc to LaTeX Beamer.
writeBeamer :: PandocMonad m => WriterOptions -> Pandoc -> m Text
writeBeamer options document =
  evalStateT (pandocToLaTeX options document) $
    (startingState options){ stBeamer = True }

type LW m = StateT WriterState m

pandocToLaTeX :: PandocMonad m
              => WriterOptions -> Pandoc -> LW m Text
pandocToLaTeX options (Pandoc meta blocks) = do
  -- Strip off final 'references' header if --natbib or --biblatex
  let method = writerCiteMethod options
  let blocks' = if method == Biblatex || method == Natbib
                   then case reverse blocks of
                             Div ("refs",_,_) _:xs -> reverse xs
                             _                     -> blocks
                   else blocks
  -- see if there are internal links
  let isInternalLink (Link _ _ ('#':xs,_)) = [xs]
      isInternalLink _                     = []
  modify $ \s -> s{ stInternalLinks = query isInternalLink blocks' }
  let template = fromMaybe "" $ writerTemplate options
  let colwidth = if writerWrapText options == WrapAuto
                    then Just $ writerColumns options
                    else Nothing
  let render' :: Doc -> Text
      render' = render colwidth
  metadata <- metaToJSON options
              (fmap render' . blockListToLaTeX)
              (fmap render' . inlineListToLaTeX)
              meta
  let chaptersClasses = ["memoir","book","report","scrreprt","scrbook","extreport","extbook","tufte-book"]
  let frontmatterClasses = ["memoir","book","scrbook","extbook","tufte-book"]
  -- these have \frontmatter etc.
  beamer <- gets stBeamer
  let documentClass =
        case lookup "documentclass" (writerVariables options) `mplus`
              fmap stringify (lookupMeta "documentclass" meta) of
                 Just x -> x
                 Nothing ->
                  case P.parse pDocumentClass "template" template of
                     Right r -> r
                     Left _
                      | beamer    -> "beamer"
                      | otherwise -> case writerTopLevelDivision options of
                                       TopLevelPart    -> "book"
                                       TopLevelChapter -> "book"
                                       _               -> "article"
  when (documentClass `elem` chaptersClasses) $
     modify $ \s -> s{ stHasChapters = True }
  -- check for \usepackage...{csquotes}; if present, we'll use
  -- \enquote{...} for smart quotes:
  let headerIncludesField :: FromJSON a => Maybe a
      headerIncludesField = getField "header-includes" metadata
  let headerIncludes = fromMaybe [] $ mplus
                       (fmap return headerIncludesField)
                       headerIncludesField
  when (any (isInfixOf "{csquotes}") (template : headerIncludes)) $
    modify $ \s -> s{stCsquotes = True}
  let (blocks'', lastHeader) = if writerCiteMethod options == Citeproc then
                                 (blocks', [])
                               else case reverse blocks' of
                                 Header 1 _ il : _ -> (init blocks', il)
                                 _                 -> (blocks', [])
  blocks''' <- if beamer
                  then toSlides blocks''
                  else return blocks''
  body <- mapM (elementToLaTeX options) $ hierarchicalize blocks'''
  (biblioTitle :: Text) <- render' <$> inlineListToLaTeX lastHeader
  let main = render' $ vsep body
  st <- get
  titleMeta <- stringToLaTeX TextString $ stringify $ docTitle meta
  authorsMeta <- mapM (stringToLaTeX TextString . stringify) $ docAuthors meta
  docLangs <- catMaybes <$>
      mapM (toLang . Just) (ordNub (query (extract "lang") blocks))
  let hasStringValue x = isJust (getField x metadata :: Maybe String)
  let geometryFromMargins = intercalate [','] $ mapMaybe (\(x,y) ->
                                ((x ++ "=") ++) <$> getField y metadata)
                              [("lmargin","margin-left")
                              ,("rmargin","margin-right")
                              ,("tmargin","margin-top")
                              ,("bmargin","margin-bottom")
                              ]
  let toPolyObj lang = object [ "name"    .= T.pack name
                              , "options" .= T.pack opts ]
        where
          (name, opts) = toPolyglossia lang
  mblang <- toLang $ case getLang options meta of
                          Just l -> Just l
                          Nothing | null docLangs -> Nothing
                                  | otherwise     -> Just "en"
  -- we need a default here since lang is used in template conditionals

  let dirs = query (extract "dir") blocks

  let context  =  defField "toc" (writerTableOfContents options) $
                  defField "toc-depth" (show (writerTOCDepth options -
                                              if stHasChapters st
                                                 then 1
                                                 else 0)) $
                  defField "body" main $
                  defField "title-meta" titleMeta $
                  defField "author-meta" (intercalate "; " authorsMeta) $
                  defField "documentclass" documentClass $
                  defField "verbatim-in-note" (stVerbInNote st) $
                  defField "tables" (stTable st) $
                  defField "strikeout" (stStrikeout st) $
                  defField "url" (stUrl st) $
                  defField "numbersections" (writerNumberSections options) $
                  defField "lhs" (stLHS st) $
                  defField "graphics" (stGraphics st) $
                  defField "has-chapters" (stHasChapters st) $
                  defField "has-frontmatter" (documentClass `elem` frontmatterClasses) $
                  defField "listings" (writerListings options || stLHS st) $
                  defField "beamer" beamer $
                  (if stHighlighting st
                      then case writerHighlightStyle options of
                                Just sty ->
                                   defField "highlighting-macros"
                                      (styleToLaTeX sty)
                                Nothing -> id
                      else id) $
                  (case writerCiteMethod options of
                         Natbib   -> defField "biblio-title" biblioTitle .
                                     defField "natbib" True
                         Biblatex -> defField "biblio-title" biblioTitle .
                                     defField "biblatex" True
                         _        -> id) $
                  defField "colorlinks" (any hasStringValue
                           ["citecolor", "urlcolor", "linkcolor", "toccolor",
                            "filecolor"]) $
                  (if null dirs
                     then id
                     else defField "dir" ("ltr" :: String)) $
                  defField "section-titles" True $
                  defField "geometry" geometryFromMargins $
                  (case getField "papersize" metadata of
                        -- uppercase a4, a5, etc.
                        Just (('A':d:ds) :: String)
                          | all isDigit (d:ds) -> resetField "papersize"
                                                    (('a':d:ds) :: String)
                        _                     -> id)
                  metadata
  let context' =
          -- note: lang is used in some conditionals in the template,
          -- so we need to set it if we have any babel/polyglossia:
          maybe id (defField "lang" . renderLang) mblang
        $ maybe id (defField "babel-lang" . toBabel) mblang
        $ defField "babel-otherlangs" (map toBabel docLangs)
        $ defField "babel-newcommands" (concatMap (\(poly, babel) ->
            -- \textspanish and \textgalician are already used by babel
            -- save them as \oritext... and let babel use that
            if poly `elem` ["spanish", "galician"]
               then "\\let\\oritext" ++ poly ++ "\\text" ++ poly ++ "\n" ++
                    "\\AddBabelHook{" ++ poly ++ "}{beforeextras}" ++
                      "{\\renewcommand{\\text" ++ poly ++ "}{\\oritext"
                      ++ poly ++ "}}\n" ++
                    "\\AddBabelHook{" ++ poly ++ "}{afterextras}" ++
                      "{\\renewcommand{\\text" ++ poly ++ "}[2][]{\\foreignlanguage{"
                      ++ poly ++ "}{##2}}}\n"
               else (if poly == "latin" -- see #4161
                        then "\\providecommand{\\textlatin}{}\n\\renewcommand"
                        else "\\newcommand") ++ "{\\text" ++ poly ++
                    "}[2][]{\\foreignlanguage{" ++ babel ++ "}{#2}}\n" ++
                    "\\newenvironment{" ++ poly ++
                    "}[2][]{\\begin{otherlanguage}{" ++
                    babel ++ "}}{\\end{otherlanguage}}\n"
            )
            -- eliminate duplicates that have same polyglossia name
            $ nubBy (\a b -> fst a == fst b)
            -- find polyglossia and babel names of languages used in the document
            $ map (\l -> (fst $ toPolyglossia l, toBabel l)) docLangs
          )
        $ maybe id (defField "polyglossia-lang" . toPolyObj) mblang
        $ defField "polyglossia-otherlangs" (map toPolyObj docLangs)
        $
                  defField "latex-dir-rtl"
           (getField "dir" context == Just ("rtl" :: String)) context
  case writerTemplate options of
       Nothing  -> return main
       Just tpl -> renderTemplate' tpl context'

-- | Convert Elements to LaTeX
elementToLaTeX :: PandocMonad m => WriterOptions -> Element -> LW m Doc
elementToLaTeX _ (Blk block) = blockToLaTeX block
elementToLaTeX opts (Sec level _ (id',classes,_) title' elements) = do
  modify $ \s -> s{stInHeading = True}
  header' <- sectionHeader ("unnumbered" `elem` classes) id' level title'
  modify $ \s -> s{stInHeading = False}
  innerContents <- mapM (elementToLaTeX opts) elements
  return $ vsep (header' : innerContents)

data StringContext = TextString
                   | URLString
                   | CodeString
                   deriving (Eq)

-- escape things as needed for LaTeX
stringToLaTeX :: PandocMonad m => StringContext -> String -> LW m String
stringToLaTeX context zs = do
  opts <- gets stOptions
  return $
    foldr (go opts context) mempty $
    if writerPreferAscii opts
       then T.unpack $ Normalize.normalize Normalize.NFD $ T.pack zs
       else zs
 where
  go opts ctx x xs   =
    let ligatures = isEnabled Ext_smart opts && ctx == TextString
        isUrl = ctx == URLString
        mbAccentCmd =
          if writerPreferAscii opts && ctx == TextString
             then uncons xs >>= \(c,_) -> M.lookup c accents
             else Nothing
        emits s =
          case mbAccentCmd of
               Just cmd ->
                 cmd ++ "{" ++ s ++ "}" ++ drop 1 xs -- drop combining accent
               Nothing  -> s ++ xs
        emitc c =
          case mbAccentCmd of
               Just cmd ->
                 cmd ++ "{" ++ [c] ++ "}" ++ drop 1 xs -- drop combining accent
               Nothing  -> c : xs
        emitcseq cs = do
          case xs of
            c:_ | isLetter c
                , ctx == TextString
                             -> cs <> " " <> xs
                | isSpace c  -> cs <> "{}" <> xs
                | ctx == TextString
                             -> cs <> xs
            _ -> cs <> "{}" <> xs
    in case x of
         '{' -> emits "\\{"
         '}' -> emits "\\}"
         '`' | ctx == CodeString -> emitcseq "\\textasciigrave"
         '$' | not isUrl -> emits "\\$"
         '%' -> emits "\\%"
         '&' -> emits "\\&"
         '_' | not isUrl -> emits "\\_"
         '#' -> emits "\\#"
         '-' | not isUrl -> case xs of
                     -- prevent adjacent hyphens from forming ligatures
                     ('-':_) -> emits "-\\/"
                     _       -> emitc '-'
         '~' | not isUrl -> emitcseq "\\textasciitilde"
         '^' -> emits "\\^{}"
         '\\'| isUrl     -> emitc '/' -- NB. / works as path sep even on Windows
             | otherwise -> emitcseq "\\textbackslash"
         '|' | not isUrl -> emitcseq "\\textbar"
         '<' -> emitcseq "\\textless"
         '>' -> emitcseq "\\textgreater"
         '[' -> emits "{[}"  -- to avoid interpretation as
         ']' -> emits "{]}"  -- optional arguments
         '\'' | ctx == CodeString -> emitcseq "\\textquotesingle"
         '\160' -> emits "~"
         '\x202F' -> emits "\\,"
         '\x2026' -> emitcseq "\\ldots"
         '\x2018' | ligatures -> emits "`"
         '\x2019' | ligatures -> emits "'"
         '\x201C' | ligatures -> emits "``"
         '\x201D' | ligatures -> emits "''"
         '\x2014' | ligatures -> emits "---"
         '\x2013' | ligatures -> emits "--"
         _ | writerPreferAscii opts
             -> case x of
                  'ı' -> emitcseq "\\i"
                  'ȷ' -> emitcseq "\\j"
                  'å' -> emitcseq "\\aa"
                  'Å' -> emitcseq "\\AA"
                  'ß' -> emitcseq "\\ss"
                  'ø' -> emitcseq "\\o"
                  'Ø' -> emitcseq "\\O"
                  'Ł' -> emitcseq "\\L"
                  'ł' -> emitcseq "\\l"
                  'æ' -> emitcseq "\\ae"
                  'Æ' -> emitcseq "\\AE"
                  'œ' -> emitcseq "\\oe"
                  'Œ' -> emitcseq "\\OE"
                  '£' -> emitcseq "\\pounds"
                  '€' -> emitcseq "\\euro"
                  '©' -> emitcseq "\\copyright"
                  _   -> emitc x
           | otherwise -> emitc x

accents :: M.Map Char String
accents = M.fromList
  [ ('\779' , "\\H")
  , ('\768' , "\\`")
  , ('\769' , "\\'")
  , ('\770' , "\\^")
  , ('\771' , "\\~")
  , ('\776' , "\\\"")
  , ('\775' , "\\.")
  , ('\772' , "\\=")
  , ('\781' , "\\|")
  , ('\817' , "\\b")
  , ('\807' , "\\c")
  , ('\783' , "\\G")
  , ('\777' , "\\h")
  , ('\803' , "\\d")
  , ('\785' , "\\f")
  , ('\778' , "\\r")
  , ('\865' , "\\t")
  , ('\782' , "\\U")
  , ('\780' , "\\v")
  , ('\774' , "\\u")
  , ('\808' , "\\k")
  , ('\785' , "\\newtie")
  , ('\8413', "\\textcircled")
  ]

toLabel :: PandocMonad m => String -> LW m String
toLabel z = go `fmap` stringToLaTeX URLString z
 where go [] = ""
       go (x:xs)
         | (isLetter x || isDigit x) && isAscii x = x:go xs
         | x `elem` ("_-+=:;." :: String) = x:go xs
         | otherwise = "ux" ++ printf "%x" (ord x) ++ go xs

-- | Puts contents into LaTeX command.
inCmd :: String -> Doc -> Doc
inCmd cmd contents = char '\\' <> text cmd <> braces contents

toSlides :: PandocMonad m => [Block] -> LW m [Block]
toSlides bs = do
  opts <- gets stOptions
  let slideLevel = fromMaybe (getSlideLevel bs) $ writerSlideLevel opts
  let bs' = prepSlides slideLevel bs
  concat `fmap` mapM (elementToBeamer slideLevel) (hierarchicalize bs')

elementToBeamer :: PandocMonad m => Int -> Element -> LW m [Block]
elementToBeamer _slideLevel (Blk (Div attr bs)) = do
  -- make sure we support "blocks" inside divs
  bs' <- concat `fmap` mapM (elementToBeamer 0) (hierarchicalize bs)
  return [Div attr bs']
elementToBeamer _slideLevel (Blk b) = return [b]
elementToBeamer slideLevel  (Sec lvl _num (ident,classes,kvs) tit elts)
  | lvl >  slideLevel = do
      bs <- concat `fmap` mapM (elementToBeamer slideLevel) elts
      return $ Para ( RawInline "latex" "\\begin{block}{"
                    : tit ++ [RawInline "latex" "}"] )
             : bs ++ [RawBlock "latex" "\\end{block}"]
  | lvl <  slideLevel = do
      let isSec Sec{} = True
          isSec _     = False
      let (contentElts, secElts) = break isSec elts
      let elts' = if null contentElts
                     then secElts
                     else Sec slideLevel [] nullAttr tit contentElts :
                          secElts
      bs <- concat `fmap` mapM (elementToBeamer slideLevel) elts'
      return $ Header lvl (ident,classes,kvs) tit : bs
  | otherwise = do -- lvl == slideLevel
      -- note: [fragile] is required or verbatim breaks
      let hasCodeBlock (CodeBlock _ _) = [True]
          hasCodeBlock _               = []
      let hasCode (Code _ _) = [True]
          hasCode _          = []
      let fragile = "fragile" `elem` classes ||
                    not (null $ query hasCodeBlock elts ++ query hasCode elts)
      let frameoptions = ["allowdisplaybreaks", "allowframebreaks", "fragile",
                          "b", "c", "t", "environment",
                          "label", "plain", "shrink", "standout",
                          "noframenumbering"]
      let optionslist = ["fragile" | fragile
                                   , isNothing (lookup "fragile" kvs)
                                   , "fragile" `notElem` classes] ++
                        [k | k <- classes, k `elem` frameoptions] ++
                        [k ++ "=" ++ v | (k,v) <- kvs, k `elem` frameoptions]
      let options = if null optionslist
                       then ""
                       else "[" ++ intercalate "," optionslist ++ "]"
      let latex = RawInline (Format "latex")
      slideTitle <-
            if tit == [Str "\0"] -- marker for hrule
               then return []
               else return $ latex "{" : tit ++ [latex "}"]
      ref <- toLabel ident
      let slideAnchor = if null ident
                           then []
                           else [latex ("\n\\protect\\hypertarget{" ++
                                  ref ++ "}{}")]
      let slideStart = Para $
              RawInline "latex" ("\\begin{frame}" ++ options) :
                 slideTitle ++ slideAnchor
      let slideEnd = RawBlock "latex" "\\end{frame}"
      -- now carve up slide into blocks if there are sections inside
      bs <- concat `fmap` mapM (elementToBeamer slideLevel) elts
      return $ slideStart : bs ++ [slideEnd]

isListBlock :: Block -> Bool
isListBlock (BulletList _)     = True
isListBlock (OrderedList _ _)  = True
isListBlock (DefinitionList _) = True
isListBlock _                  = False

isLineBreakOrSpace :: Inline -> Bool
isLineBreakOrSpace LineBreak = True
isLineBreakOrSpace SoftBreak = True
isLineBreakOrSpace Space     = True
isLineBreakOrSpace _         = False

-- | Convert Pandoc block element to LaTeX.
blockToLaTeX :: PandocMonad m
             => Block     -- ^ Block to convert
             -> LW m Doc
blockToLaTeX Null = return empty
blockToLaTeX (Div (identifier,classes,kvs) bs)
  | "incremental" `elem` classes = do
      let classes' = filter ("incremental"/=) classes
      beamer <- gets stBeamer
      if beamer
        then do oldIncremental <- gets stIncremental
                modify $ \s -> s{ stIncremental = True }
                result <- blockToLaTeX $ Div (identifier,classes',kvs) bs
                modify $ \s -> s{ stIncremental = oldIncremental }
                return result
        else blockToLaTeX $ Div (identifier,classes',kvs) bs
  | "nonincremental" `elem` classes = do
      let classes' = filter ("nonincremental"/=) classes
      beamer <- gets stBeamer
      if beamer
        then do oldIncremental <- gets stIncremental
                modify $ \s -> s{ stIncremental = False }
                result <- blockToLaTeX $ Div (identifier,classes',kvs) bs
                modify $ \s -> s{ stIncremental = oldIncremental }
                return result
        else blockToLaTeX $ Div (identifier,classes',kvs) bs
  | otherwise = do
      beamer <- gets stBeamer
      linkAnchor' <- hypertarget True identifier empty
    -- see #2704 for the motivation for adding \leavevmode:
      let linkAnchor =
            case bs of
              Para _ : _
                | not (isEmpty linkAnchor')
                  -> "\\leavevmode" <> linkAnchor' <> "%"
              _ -> linkAnchor'
      let align dir txt = inCmd "begin" dir $$ txt $$ inCmd "end" dir
      lang <- toLang $ lookup "lang" kvs
      let wrapColumns = if "columns" `elem` classes
                        then \contents ->
                               inCmd "begin" "columns" <> brackets "T"
                               $$ contents
                               $$ inCmd "end" "columns"
                        else id
          wrapColumn  = if "column" `elem` classes
                        then \contents ->
                               let w = maybe "0.48" fromPct (lookup "width" kvs)
                               in  inCmd "begin" "column" <>
                                   braces (text w <> "\\textwidth")
                                   $$ contents
                                   $$ inCmd "end" "column"
                        else id
          fromPct xs =
            case reverse xs of
              '%':ds -> case safeRead (reverse ds) of
                          Just digits -> showFl (digits / 100 :: Double)
                          Nothing -> xs
              _      -> xs
          wrapDir = case lookup "dir" kvs of
                      Just "rtl" -> align "RTL"
                      Just "ltr" -> align "LTR"
                      _          -> id
          wrapLang txt = case lang of
                           Just lng -> let (l, o) = toPolyglossiaEnv lng
                                           ops = if null o
                                                 then ""
                                                 else brackets $ text o
                                       in  inCmd "begin" (text l) <> ops
                                           $$ blankline <> txt <> blankline
                                           $$ inCmd "end" (text l)
                           Nothing  -> txt
          wrapNotes txt = if beamer && "notes" `elem` classes
                          then "\\note" <> braces txt -- speaker notes
                          else linkAnchor $$ txt
      (wrapColumns . wrapColumn . wrapDir . wrapLang . wrapNotes)
        <$> blockListToLaTeX bs
blockToLaTeX (Plain lst) =
  inlineListToLaTeX $ dropWhile isLineBreakOrSpace lst
-- title beginning with fig: indicates that the image is a figure
blockToLaTeX (Para [Image attr@(ident, _, _) txt (src,'f':'i':'g':':':tit)]) = do
  (capt, captForLof, footnotes) <- getCaption True txt
  lab <- labelFor ident
  let caption = "\\caption" <> captForLof <> braces capt <> lab
  img <- inlineToLaTeX (Image attr txt (src,tit))
  innards <- hypertarget True ident $
                 "\\centering" $$ img $$ caption <> cr
  let figure = cr <> "\\begin{figure}" $$ innards $$ "\\end{figure}"
  st <- get
  return $ if stInNote st || stInMinipage st
              -- can't have figures in notes or minipage (here, table cell)
              -- http://www.tex.ac.uk/FAQ-ouparmd.html
              then "\\begin{center}" $$ img $+$ capt $$ "\\end{center}"
              else figure $$ footnotes
-- . . . indicates pause in beamer slides
blockToLaTeX (Para [Str ".",Space,Str ".",Space,Str "."]) = do
  beamer <- gets stBeamer
  if beamer
     then blockToLaTeX (RawBlock "latex" "\\pause")
     else inlineListToLaTeX [Str ".",Space,Str ".",Space,Str "."]
blockToLaTeX (Para lst) =
  inlineListToLaTeX $ dropWhile isLineBreakOrSpace lst
blockToLaTeX (LineBlock lns) =
  blockToLaTeX $ linesToPara lns
blockToLaTeX (BlockQuote lst) = do
  beamer <- gets stBeamer
  case lst of
       [b] | beamer && isListBlock b -> do
         oldIncremental <- gets stIncremental
         modify $ \s -> s{ stIncremental = not oldIncremental }
         result <- blockToLaTeX b
         modify $ \s -> s{ stIncremental = oldIncremental }
         return result
       _ -> do
         oldInQuote <- gets stInQuote
         modify (\s -> s{stInQuote = True})
         contents <- blockListToLaTeX lst
         modify (\s -> s{stInQuote = oldInQuote})
         return $ "\\begin{quote}" $$ contents $$ "\\end{quote}"
blockToLaTeX (CodeBlock (identifier,classes,keyvalAttr) str) = do
  opts <- gets stOptions
  lab <- labelFor identifier
  linkAnchor' <- hypertarget True identifier lab
  let linkAnchor = if isEmpty linkAnchor'
                      then empty
                      else linkAnchor' <> "%"
  let lhsCodeBlock = do
        modify $ \s -> s{ stLHS = True }
        return $ flush (linkAnchor $$ "\\begin{code}" $$ text str $$
                            "\\end{code}") $$ cr
  let rawCodeBlock = do
        st <- get
        env <- if stInNote st
                  then modify (\s -> s{ stVerbInNote = True }) >>
                       return "Verbatim"
                  else return "verbatim"
        return $ flush (linkAnchor $$ text ("\\begin{" ++ env ++ "}") $$
                 text str $$ text ("\\end{" ++ env ++ "}")) <> cr
  let listingsCodeBlock = do
        st <- get
        ref <- toLabel identifier
        let params = if writerListings (stOptions st)
                     then (case getListingsLanguage classes of
                                Just l  -> [ "language=" ++ mbBraced l ]
                                Nothing -> []) ++
                          [ "numbers=left" | "numberLines" `elem` classes
                             || "number" `elem` classes
                             || "number-lines" `elem` classes ] ++
                          [ (if key == "startFrom"
                                then "firstnumber"
                                else key) ++ "=" ++ mbBraced attr |
                                (key,attr) <- keyvalAttr,
                                key `notElem` ["exports", "tangle", "results"]
                                -- see #4889
                          ] ++
                          (if identifier == ""
                                then []
                                else [ "label=" ++ ref ])

                     else []
            printParams
                | null params = empty
                | otherwise   = brackets $ hcat (intersperse ", "
                      (map text params))
        return $ flush ("\\begin{lstlisting}" <> printParams $$ text str $$
                 "\\end{lstlisting}") $$ cr
  let highlightedCodeBlock =
        case highlight (writerSyntaxMap opts)
                 formatLaTeXBlock ("",classes,keyvalAttr) str of
               Left msg -> do
                 unless (null msg) $
                   report $ CouldNotHighlight msg
                 rawCodeBlock
               Right h -> do
                  st <- get
                  when (stInNote st) $ modify (\s -> s{ stVerbInNote = True })
                  modify (\s -> s{ stHighlighting = True })
                  return (flush $ linkAnchor $$ text (T.unpack h))
  case () of
     _ | isEnabled Ext_literate_haskell opts && "haskell" `elem` classes &&
         "literate" `elem` classes           -> lhsCodeBlock
       | writerListings opts                 -> listingsCodeBlock
       | not (null classes) && isJust (writerHighlightStyle opts)
                                             -> highlightedCodeBlock
       | otherwise                           -> rawCodeBlock
blockToLaTeX b@(RawBlock f x) = do
  beamer <- gets stBeamer
  if (f == Format "latex" || f == Format "tex" ||
       (f == Format "beamer" && beamer))
     then return $ text x
     else do
       report $ BlockNotRendered b
       return empty
blockToLaTeX (BulletList []) = return empty  -- otherwise latex error
blockToLaTeX (BulletList lst) = do
  incremental <- gets stIncremental
  beamer <- gets stBeamer
  let inc = if beamer && incremental then "[<+->]" else ""
  items <- mapM listItemToLaTeX lst
  let spacing = if isTightList lst
                   then text "\\tightlist"
                   else empty
  return $ text ("\\begin{itemize}" ++ inc) $$ spacing $$ vcat items $$
             "\\end{itemize}"
blockToLaTeX (OrderedList _ []) = return empty -- otherwise latex error
blockToLaTeX (OrderedList (start, numstyle, numdelim) lst) = do
  st <- get
  let inc = if stBeamer st && stIncremental st then "[<+->]" else ""
  let oldlevel = stOLLevel st
  put $ st {stOLLevel = oldlevel + 1}
  items <- mapM listItemToLaTeX lst
  modify (\s -> s {stOLLevel = oldlevel})
  let beamer = stBeamer st
  let tostyle x = case numstyle of
                       Decimal      -> "\\arabic" <> braces x
                       UpperRoman   -> "\\Roman" <> braces x
                       LowerRoman   -> "\\roman" <> braces x
                       UpperAlpha   -> "\\Alph" <> braces x
                       LowerAlpha   -> "\\alph" <> braces x
                       Example      -> "\\arabic" <> braces x
                       DefaultStyle -> "\\arabic" <> braces x
  let todelim x = case numdelim of
                       OneParen  -> x <> ")"
                       TwoParens -> parens x
                       Period    -> x <> "."
                       _         -> x <> "."
  let exemplar = case numstyle of
                       Decimal      -> "1"
                       UpperRoman   -> "I"
                       LowerRoman   -> "i"
                       UpperAlpha   -> "A"
                       LowerAlpha   -> "a"
                       Example      -> "1"
                       DefaultStyle -> "1"
  let enum = text $ "enum" ++ map toLower (toRomanNumeral oldlevel)
  let stylecommand
        | numstyle == DefaultStyle && numdelim == DefaultDelim = empty
        | beamer && numstyle == Decimal && numdelim == Period = empty
        | beamer = brackets (todelim exemplar)
        | otherwise = "\\def" <> "\\label" <> enum <>
          braces (todelim $ tostyle enum)
  let resetcounter = if start == 1 || oldlevel > 4
                        then empty
                        else "\\setcounter" <> braces enum <>
                              braces (text $ show $ start - 1)
  let spacing = if isTightList lst
                   then text "\\tightlist"
                   else empty
  return $ text ("\\begin{enumerate}" ++ inc)
         $$ stylecommand
         $$ resetcounter
         $$ spacing
         $$ vcat items
         $$ "\\end{enumerate}"
blockToLaTeX (DefinitionList []) = return empty
blockToLaTeX (DefinitionList lst) = do
  incremental <- gets stIncremental
  beamer <- gets stBeamer
  let inc = if beamer && incremental then "[<+->]" else ""
  items <- mapM defListItemToLaTeX lst
  let spacing = if all isTightList (map snd lst)
                   then text "\\tightlist"
                   else empty
  return $ text ("\\begin{description}" ++ inc) $$ spacing $$ vcat items $$
               "\\end{description}"
blockToLaTeX HorizontalRule =
            return
  "\\begin{center}\\rule{0.5\\linewidth}{\\linethickness}\\end{center}"
blockToLaTeX (Header level (id',classes,_) lst) = do
  modify $ \s -> s{stInHeading = True}
  hdr <- sectionHeader ("unnumbered" `elem` classes) id' level lst
  modify $ \s -> s{stInHeading = False}
  return hdr
blockToLaTeX (Table caption aligns widths heads rows) = do
  (captionText, captForLof, footnotes) <- getCaption False caption
  let toHeaders hs = do contents <- tableRowToLaTeX True aligns widths hs
                        return ("\\toprule" $$ contents $$ "\\midrule")
  let removeNote (Note _) = Span ("", [], []) []
      removeNote x        = x
  firsthead <- if isEmpty captionText || all null heads
                  then return empty
                  else ($$ text "\\endfirsthead") <$> toHeaders heads
  head' <- if all null heads
              then return "\\toprule"
              -- avoid duplicate notes in head and firsthead:
              else toHeaders (if isEmpty firsthead
                                 then heads
                                 else walk removeNote heads)
  let capt = if isEmpty captionText
                then empty
                else "\\caption" <> captForLof <> braces captionText
                         <> "\\tabularnewline"
  rows' <- mapM (tableRowToLaTeX False aligns widths) rows
  let colDescriptors = text $ concatMap toColDescriptor aligns
  modify $ \s -> s{ stTable = True }
  return $ "\\begin{longtable}[]" <>
              braces ("@{}" <> colDescriptors <> "@{}")
              -- the @{} removes extra space at beginning and end
         $$ capt
         $$ firsthead
         $$ head'
         $$ "\\endhead"
         $$ vcat rows'
         $$ "\\bottomrule"
         $$ "\\end{longtable}"
         $$ footnotes

getCaption :: PandocMonad m => Bool -> [Inline] -> LW m (Doc, Doc, Doc)
getCaption externalNotes txt = do
  oldIsMinipage <- gets stInMinipage
  modify $ \st -> st{ stInMinipage = externalNotes, stNotes = [] }
  capt <- inlineListToLaTeX txt
  notes <- gets stNotes
  modify $ \st -> st{ stInMinipage = oldIsMinipage, stNotes = [] }
  -- We can't have footnotes in the list of figures/tables, so remove them:
  captForLof <- if null notes
                   then return empty
                   else brackets <$> inlineListToLaTeX (walk deNote txt)
  let footnotes = if externalNotes
                     then notesToLaTeX notes
                     else empty
  return (capt, captForLof, footnotes)

toColDescriptor :: Alignment -> String
toColDescriptor align =
  case align of
         AlignLeft    -> "l"
         AlignRight   -> "r"
         AlignCenter  -> "c"
         AlignDefault -> "l"

blockListToLaTeX :: PandocMonad m => [Block] -> LW m Doc
blockListToLaTeX lst =
  vsep `fmap` mapM (\b -> setEmptyLine True >> blockToLaTeX b) lst

tableRowToLaTeX :: PandocMonad m
                => Bool
                -> [Alignment]
                -> [Double]
                -> [[Block]]
                -> LW m Doc
tableRowToLaTeX header aligns widths cols = do
  -- scale factor compensates for extra space between columns
  -- so the whole table isn't larger than columnwidth
  let scaleFactor = 0.97 ** fromIntegral (length aligns)
  let isSimple [Plain _] = True
      isSimple [Para  _] = True
      isSimple []        = True
      isSimple _         = False
  -- simple tables have to have simple cells:
  let widths' = if all (== 0) widths && not (all isSimple cols)
                   then replicate (length aligns)
                          (scaleFactor / fromIntegral (length aligns))
                   else map (scaleFactor *) widths
  cells <- mapM (tableCellToLaTeX header) $ zip3 widths' aligns cols
  return $ hsep (intersperse "&" cells) <> "\\tabularnewline"

-- For simple latex tables (without minipages or parboxes),
-- we need to go to some lengths to get line breaks working:
-- as LineBreak bs = \vtop{\hbox{\strut as}\hbox{\strut bs}}.
fixLineBreaks :: Block -> Block
fixLineBreaks (Para ils)  = Para $ fixLineBreaks' ils
fixLineBreaks (Plain ils) = Plain $ fixLineBreaks' ils
fixLineBreaks x           = x

fixLineBreaks' :: [Inline] -> [Inline]
fixLineBreaks' ils = case splitBy (== LineBreak) ils of
                       []     -> []
                       [xs]   -> xs
                       chunks -> RawInline "tex" "\\vtop{" :
                                 concatMap tohbox chunks ++
                                 [RawInline "tex" "}"]
  where tohbox ys = RawInline "tex" "\\hbox{\\strut " : ys ++
                    [RawInline "tex" "}"]

-- We also change display math to inline math, since display
-- math breaks in simple tables.
displayMathToInline :: Inline -> Inline
displayMathToInline (Math DisplayMath x) = Math InlineMath x
displayMathToInline x                    = x

tableCellToLaTeX :: PandocMonad m => Bool -> (Double, Alignment, [Block])
                 -> LW m Doc
tableCellToLaTeX _      (0,     _,     blocks) =
  blockListToLaTeX $ walk fixLineBreaks $ walk displayMathToInline blocks
tableCellToLaTeX header (width, align, blocks) = do
  cellContents <- blockListToLaTeX blocks
  let valign = text $ if header then "[b]" else "[t]"
  let halign = case align of
               AlignLeft    -> "\\raggedright"
               AlignRight   -> "\\raggedleft"
               AlignCenter  -> "\\centering"
               AlignDefault -> "\\raggedright"
  return $ ("\\begin{minipage}" <> valign <>
            braces (text (printf "%.2f\\columnwidth" width)) <>
            (halign <> cr <> cellContents <> "\\strut" <> cr) <>
            "\\end{minipage}")

notesToLaTeX :: [Doc] -> Doc
notesToLaTeX [] = empty
notesToLaTeX ns = (case length ns of
                              n | n > 1 -> "\\addtocounter" <>
                                           braces "footnote" <>
                                           braces (text $ show $ 1 - n)
                                | otherwise -> empty)
                   $$
                   vcat (intersperse
                     ("\\addtocounter" <> braces "footnote" <> braces "1")
                     $ map (\x -> "\\footnotetext" <> braces x)
                     $ reverse ns)

listItemToLaTeX :: PandocMonad m => [Block] -> LW m Doc
listItemToLaTeX lst
  -- we need to put some text before a header if it's the first
  -- element in an item. This will look ugly in LaTeX regardless, but
  -- this will keep the typesetter from throwing an error.
  | (Header{} :_) <- lst =
    blockListToLaTeX lst >>= return . (text "\\item ~" $$) . nest 2
  | Plain (Str "☐":Space:is) : bs <- lst = taskListItem False is bs
  | Plain (Str "☒":Space:is) : bs <- lst = taskListItem True  is bs
  | Para  (Str "☐":Space:is) : bs <- lst = taskListItem False is bs
  | Para  (Str "☒":Space:is) : bs <- lst = taskListItem True  is bs
  | otherwise = blockListToLaTeX lst >>= return . (text "\\item" $$) . nest 2
  where
    taskListItem checked is bs = do
      let checkbox  = if checked
                      then "$\\boxtimes$"
                      else "$\\square$"
      isContents <- inlineListToLaTeX is
      bsContents <- blockListToLaTeX bs
      return $ "\\item" <> brackets checkbox
        $$ nest 2 (isContents $+$ bsContents)

defListItemToLaTeX :: PandocMonad m => ([Inline], [[Block]]) -> LW m Doc
defListItemToLaTeX (term, defs) = do
    -- needed to turn off 'listings' because it breaks inside \item[...]:
    modify $ \s -> s{stInItem = True}
    term' <- inlineListToLaTeX term
    modify $ \s -> s{stInItem = False}
    -- put braces around term if it contains an internal link,
    -- since otherwise we get bad bracket interactions: \item[\hyperref[..]
    let isInternalLink (Link _ _ ('#':_,_)) = True
        isInternalLink _                    = False
    let term'' = if any isInternalLink term
                    then braces term'
                    else term'
    def'  <- liftM vsep $ mapM blockListToLaTeX defs
    return $ case defs of
     ((Header{} : _) : _)    ->
       "\\item" <> brackets term'' <> " ~ " $$ def'
     ((CodeBlock{} : _) : _) -> -- see #4662
       "\\item" <> brackets term'' <> " ~ " $$ def'
     _                       ->
       "\\item" <> brackets term'' $$ def'

-- | Craft the section header, inserting the secton reference, if supplied.
sectionHeader :: PandocMonad m
              => Bool    -- True for unnumbered
              -> [Char]
              -> Int
              -> [Inline]
              -> LW m Doc
sectionHeader unnumbered ident level lst = do
  txt <- inlineListToLaTeX lst
  plain <- stringToLaTeX TextString $ concatMap stringify lst
  let removeInvalidInline (Note _)             = []
      removeInvalidInline (Span (id', _, _) _) | not (null id') = []
      removeInvalidInline Image{}            = []
      removeInvalidInline x                    = [x]
  let lstNoNotes = foldr (mappend . (\x -> walkM removeInvalidInline x)) mempty lst
  txtNoNotes <- inlineListToLaTeX lstNoNotes
  -- footnotes in sections don't work (except for starred variants)
  -- unless you specify an optional argument:
  -- \section[mysec]{mysec\footnote{blah}}
  optional <- if unnumbered || lstNoNotes == lst || null lstNoNotes
                 then return empty
                 else
                   return $ brackets txtNoNotes
  let contents = if render Nothing txt == plain
                    then braces txt
                    else braces (text "\\texorpdfstring"
                         <> braces txt
                         <> braces (text plain))
  book <- gets stHasChapters
  opts <- gets stOptions
  let topLevelDivision = if book && writerTopLevelDivision opts == TopLevelDefault
                         then TopLevelChapter
                         else writerTopLevelDivision opts
  beamer <- gets stBeamer
  let level' = if beamer &&
                  topLevelDivision `elem` [TopLevelPart, TopLevelChapter]
               -- beamer has parts but no chapters
               then if level == 1 then -1 else level - 1
               else case topLevelDivision of
                      TopLevelPart    -> level - 2
                      TopLevelChapter -> level - 1
                      TopLevelSection -> level
                      TopLevelDefault -> level
  let sectionType = case level' of
                          -1 -> "part"
                          0  -> "chapter"
                          1  -> "section"
                          2  -> "subsection"
                          3  -> "subsubsection"
                          4  -> "paragraph"
                          5  -> "subparagraph"
                          _  -> ""
  inQuote <- gets stInQuote
  let prefix = if inQuote && level' >= 4
                  then text "\\mbox{}%"
                  -- needed for \paragraph, \subparagraph in quote environment
                  -- see http://tex.stackexchange.com/questions/169830/
                  else empty
  lab <- labelFor ident
  let star = if unnumbered && level' < 4 then text "*" else empty
  let stuffing = star <> optional <> contents
  stuffing' <- hypertarget True ident $
                  text ('\\':sectionType) <> stuffing <> lab
  return $ if level' > 5
              then txt
              else prefix $$ stuffing'
                   $$ if unnumbered
                         then "\\addcontentsline{toc}" <>
                                braces (text sectionType) <>
                                braces txtNoNotes
                         else empty

hypertarget :: PandocMonad m => Bool -> String -> Doc -> LW m Doc
hypertarget _ "" x    = return x
hypertarget addnewline ident x = do
  ref <- text `fmap` toLabel ident
  return $ text "\\hypertarget"
              <> braces ref
              <> braces ((if addnewline && not (isEmpty x)
                             then ("%" <> cr)
                             else empty) <> x)

labelFor :: PandocMonad m => String -> LW m Doc
labelFor ""    = return empty
labelFor ident = do
  ref <- text `fmap` toLabel ident
  return $ text "\\label" <> braces ref

-- | Convert list of inline elements to LaTeX.
inlineListToLaTeX :: PandocMonad m
                  => [Inline]  -- ^ Inlines to convert
                  -> LW m Doc
inlineListToLaTeX lst =
  mapM inlineToLaTeX (fixLineInitialSpaces lst)
    >>= return . hcat
    -- nonbreaking spaces (~) in LaTeX don't work after line breaks,
    -- so we turn nbsps after hard breaks to \hspace commands.
    -- this is mostly used in verse.
 where fixLineInitialSpaces [] = []
       fixLineInitialSpaces (LineBreak : Str s@('\160':_) : xs) =
         LineBreak : fixNbsps s ++ fixLineInitialSpaces xs
       fixLineInitialSpaces (x:xs) = x : fixLineInitialSpaces xs
       fixNbsps s = let (ys,zs) = span (=='\160') s
                    in  replicate (length ys) hspace ++ [Str zs]
       hspace = RawInline "latex" "\\hspace*{0.333em}"

isQuoted :: Inline -> Bool
isQuoted (Quoted _ _) = True
isQuoted _            = False

-- | Convert inline element to LaTeX
inlineToLaTeX :: PandocMonad m
              => Inline    -- ^ Inline to convert
              -> LW m Doc
inlineToLaTeX (Span (id',classes,kvs) ils) = do
  linkAnchor <- hypertarget False id' empty
  lang <- toLang $ lookup "lang" kvs
  let cmds = ["textup" | "csl-no-emph" `elem` classes] ++
             ["textnormal" | "csl-no-strong" `elem` classes ||
                             "csl-no-smallcaps" `elem` classes] ++
             ["RL" | ("dir", "rtl") `elem` kvs] ++
             ["LR" | ("dir", "ltr") `elem` kvs] ++
             (case lang of
                Just lng -> let (l, o) = toPolyglossia lng
                                ops = if null o then "" else ("[" ++ o ++ "]")
                            in  ["text" ++ l ++ ops]
                Nothing  -> [])
  contents <- inlineListToLaTeX ils
  return $ (if null id'
               then empty
               else "\\protect" <> linkAnchor) <>
           (if null cmds
               then braces contents
               else foldr inCmd contents cmds)
inlineToLaTeX (Emph lst) =
  inlineListToLaTeX lst >>= return . inCmd "emph"
inlineToLaTeX (Strong lst) =
  inlineListToLaTeX lst >>= return . inCmd "textbf"
inlineToLaTeX (Strikeout lst) = do
  -- we need to protect VERB in an mbox or we get an error
  -- see #1294
  contents <- inlineListToLaTeX $ protectCode lst
  modify $ \s -> s{ stStrikeout = True }
  return $ inCmd "sout" contents
inlineToLaTeX (Superscript lst) =
  inlineListToLaTeX lst >>= return . inCmd "textsuperscript"
inlineToLaTeX (Subscript lst) =
  inlineListToLaTeX lst >>= return . inCmd "textsubscript"
inlineToLaTeX (SmallCaps lst) =
  inlineListToLaTeX lst >>= return . inCmd "textsc"
inlineToLaTeX (Cite cits lst) = do
  st <- get
  let opts = stOptions st
  case writerCiteMethod opts of
     Natbib   -> citationsToNatbib cits
     Biblatex -> citationsToBiblatex cits
     _        -> inlineListToLaTeX lst

inlineToLaTeX (Code (_,classes,_) str) = do
  opts <- gets stOptions
  inHeading <- gets stInHeading
  inItem <- gets stInItem
  let listingsCode = do
        let listingsopt = case getListingsLanguage classes of
                               Just l  -> "[language=" ++ mbBraced l ++ "]"
                               Nothing -> ""
        inNote <- gets stInNote
        when inNote $ modify $ \s -> s{ stVerbInNote = True }
        let chr = case "!\"'()*,-./:;?@" \\ str of
                       (c:_) -> c
                       []    -> '!'
        let str' = escapeStringUsing (backslashEscapes "\\{}%~_&#") str
        -- we always put lstinline in a dummy 'passthrough' command
        -- (defined in the default template) so that we don't have
        -- to change the way we escape characters depending on whether
        -- the lstinline is inside another command.  See #1629:
        return $ text $ "\\passthrough{\\lstinline" ++ listingsopt ++ [chr] ++ str' ++ [chr] ++ "}"
  let rawCode = liftM (text . (\s -> "\\texttt{" ++ escapeSpaces s ++ "}"))
                 $ stringToLaTeX CodeString str
                where escapeSpaces =  concatMap
                         (\c -> if c == ' ' then "\\ " else [c])
  let highlightCode =
        case highlight (writerSyntaxMap opts)
                 formatLaTeXInline ("",classes,[]) str of
               Left msg -> do
                 unless (null msg) $ report $ CouldNotHighlight msg
                 rawCode
               Right h -> modify (\st -> st{ stHighlighting = True }) >>
                          return (text (T.unpack h))
  case () of
     _ | writerListings opts  && not (inHeading || inItem) -> listingsCode
       | isJust (writerHighlightStyle opts) && not (null classes)
                                                    -> highlightCode
       | otherwise                                  -> rawCode
inlineToLaTeX (Quoted qt lst) = do
  contents <- inlineListToLaTeX lst
  csquotes <- liftM stCsquotes get
  opts <- gets stOptions
  if csquotes
     then return $ "\\enquote" <> braces contents
     else do
       let s1 = if not (null lst) && isQuoted (head lst)
                   then "\\,"
                   else empty
       let s2 = if not (null lst) && isQuoted (last lst)
                   then "\\,"
                   else empty
       let inner = s1 <> contents <> s2
       return $ case qt of
                DoubleQuote ->
                   if isEnabled Ext_smart opts
                      then text "``" <> inner <> text "''"
                      else char '\x201C' <> inner <> char '\x201D'
                SingleQuote ->
                   if isEnabled Ext_smart opts
                      then char '`' <> inner <> char '\''
                      else char '\x2018' <> inner <> char '\x2019'
inlineToLaTeX (Str str) = do
  setEmptyLine False
  liftM text $ stringToLaTeX TextString str
inlineToLaTeX (Math InlineMath str) = do
  setEmptyLine False
  return $ "\\(" <> text (handleMathComment str) <> "\\)"
inlineToLaTeX (Math DisplayMath str) = do
  setEmptyLine False
  return $ "\\[" <> text (handleMathComment str) <> "\\]"
inlineToLaTeX il@(RawInline f str) = do
  beamer <- gets stBeamer
  if (f == Format "latex" || f == Format "tex" ||
        (f == Format "beamer" && beamer))
     then do
       setEmptyLine False
       return $ text str
     else do
       report $ InlineNotRendered il
       return empty
inlineToLaTeX LineBreak = do
  emptyLine <- gets stEmptyLine
  setEmptyLine True
  return $ (if emptyLine then "~" else "") <> "\\\\" <> cr
inlineToLaTeX SoftBreak = do
  wrapText <- gets (writerWrapText . stOptions)
  case wrapText of
       WrapAuto     -> return space
       WrapNone     -> return space
       WrapPreserve -> return cr
inlineToLaTeX Space = return space
inlineToLaTeX (Link _ txt ('#':ident, _)) = do
  contents <- inlineListToLaTeX txt
  lab <- toLabel ident
  return $ text "\\protect\\hyperlink" <> braces (text lab) <> braces contents
inlineToLaTeX (Link _ txt (src, _)) =
  case txt of
        [Str x] | unEscapeString x == unEscapeString src ->  -- autolink
             do modify $ \s -> s{ stUrl = True }
                src' <- stringToLaTeX URLString (escapeURI src)
                return $ text $ "\\url{" ++ src' ++ "}"
        [Str x] | Just rest <- stripPrefix "mailto:" src,
                  unEscapeString x == unEscapeString rest -> -- email autolink
             do modify $ \s -> s{ stUrl = True }
                src' <- stringToLaTeX URLString (escapeURI src)
                contents <- inlineListToLaTeX txt
                return $ "\\href" <> braces (text src') <>
                   braces ("\\nolinkurl" <> braces contents)
        _ -> do contents <- inlineListToLaTeX txt
                src' <- stringToLaTeX URLString (escapeURI src)
                return $ text ("\\href{" ++ src' ++ "}{") <>
                         contents <> char '}'
inlineToLaTeX il@(Image _ _ ('d':'a':'t':'a':':':_, _)) = do
  report $ InlineNotRendered il
  return empty
inlineToLaTeX (Image attr _ (source, _)) = do
  setEmptyLine False
  modify $ \s -> s{ stGraphics = True }
  opts <- gets stOptions
  let showDim dir = let d = text (show dir) <> "="
                    in case dimension dir attr of
                         Just (Pixel a)   ->
                           [d <> text (showInInch opts (Pixel a)) <> "in"]
                         Just (Percent a) ->
                           [d <> text (showFl (a / 100)) <>
                             case dir of
                                Width  -> "\\textwidth"
                                Height -> "\\textheight"
                           ]
                         Just dim         ->
                           [d <> text (show dim)]
                         Nothing          ->
                           case dir of
                                Width | isJust (dimension Height attr) ->
                                  [d <> "\\textwidth"]
                                Height | isJust (dimension Width attr) ->
                                  [d <> "\\textheight"]
                                _ -> []
      dimList = showDim Width ++ showDim Height
      dims = if null dimList
                then empty
                else brackets $ cat (intersperse "," dimList)
      source' = if isURI source
                   then source
                   else unEscapeString source
  source'' <- stringToLaTeX URLString source'
  inHeading <- gets stInHeading
  return $
    (if inHeading then "\\protect\\includegraphics" else "\\includegraphics") <>
    dims <> braces (text source'')
inlineToLaTeX (Note contents) = do
  setEmptyLine False
  inMinipage <- gets stInMinipage
  modify (\s -> s{stInNote = True})
  contents' <- blockListToLaTeX contents
  modify (\s -> s {stInNote = False})
  let optnl = case reverse contents of
                   (CodeBlock _ _ : _) -> cr
                   _                   -> empty
  let noteContents = nest 2 contents' <> optnl
  beamer <- gets stBeamer
  -- in beamer slides, display footnote from current overlay forward
  let beamerMark = if beamer
                      then text "<.->"
                      else empty
  modify $ \st -> st{ stNotes = noteContents : stNotes st }
  return $
    if inMinipage
       then "\\footnotemark{}"
       -- note: a \n before } needed when note ends with a Verbatim environment
       else "\\footnote" <> beamerMark <> braces noteContents

-- A comment at the end of math needs to be followed by a newline,
-- or the closing delimiter gets swallowed.
handleMathComment :: String -> String
handleMathComment s =
  let (_, ys) = break (\c -> c == '\n' || c == '%') $ reverse s
  in  case ys of
         '%':'\\':_ -> s
         '%':_      -> s ++ "\n"
         _          -> s

protectCode :: [Inline] -> [Inline]
protectCode [] = []
protectCode (x@(Code ("",[],[]) _) : xs) = x : protectCode xs
protectCode (x@(Code _ _) : xs) = ltx "\\mbox{" : x : ltx "}" : xs
  where ltx = RawInline (Format "latex")
protectCode (x : xs) = x : protectCode xs

setEmptyLine :: PandocMonad m => Bool -> LW m ()
setEmptyLine b = modify $ \st -> st{ stEmptyLine = b }

citationsToNatbib :: PandocMonad m => [Citation] -> LW m Doc
citationsToNatbib
            [one]
  = citeCommand c p s k
  where
    Citation { citationId = k
             , citationPrefix = p
             , citationSuffix = s
             , citationMode = m
             }
      = one
    c = case m of
             AuthorInText   -> "citet"
             SuppressAuthor -> "citeyearpar"
             NormalCitation -> "citep"

citationsToNatbib cits
  | noPrefix (tail cits) && noSuffix (init cits) && ismode NormalCitation cits
  = citeCommand "citep" p s ks
  where
     noPrefix  = all (null . citationPrefix)
     noSuffix  = all (null . citationSuffix)
     ismode m  = all ((==) m  . citationMode)
     p         = citationPrefix  $
                 head cits
     s         = citationSuffix  $
                 last cits
     ks        = intercalate ", " $ map citationId cits

citationsToNatbib (c:cs) | citationMode c == AuthorInText = do
     author <- citeCommand "citeauthor" [] [] (citationId c)
     cits   <- citationsToNatbib (c { citationMode = SuppressAuthor } : cs)
     return $ author <+> cits

citationsToNatbib cits = do
  cits' <- mapM convertOne cits
  return $ text "\\citetext{" <> foldl' combineTwo empty cits' <> text "}"
  where
    combineTwo a b | isEmpty a = b
                   | otherwise = a <> text "; " <> b
    convertOne Citation { citationId = k
                        , citationPrefix = p
                        , citationSuffix = s
                        , citationMode = m
                        }
        = case m of
               AuthorInText   -> citeCommand "citealt"  p s k
               SuppressAuthor -> citeCommand "citeyear" p s k
               NormalCitation -> citeCommand "citealp"  p s k

citeCommand :: PandocMonad m
            => String -> [Inline] -> [Inline] -> String -> LW m Doc
citeCommand c p s k = do
  args <- citeArguments p s k
  return $ text ("\\" ++ c) <> args

citeArguments :: PandocMonad m
              => [Inline] -> [Inline] -> String -> LW m Doc
citeArguments p s k = do
  let s' = case s of
        (Str
                    [x] : r) | isPunctuation x -> dropWhile (== Space) r
        (Str (x:xs) : r) | isPunctuation x -> Str xs : r
        _                -> s
  pdoc <- inlineListToLaTeX p
  sdoc <- inlineListToLaTeX s'
  let optargs = case (isEmpty pdoc, isEmpty sdoc) of
                     (True, True ) -> empty
                     (True, False) -> brackets sdoc
                     (_   , _    ) -> brackets pdoc <> brackets sdoc
  return $ optargs <> braces (text k)

citationsToBiblatex :: PandocMonad m => [Citation] -> LW m Doc
citationsToBiblatex
            [one]
  = citeCommand cmd p s k
    where
       Citation { citationId = k
                , citationPrefix = p
                , citationSuffix = s
                , citationMode = m
                } = one
       cmd = case m of
                  SuppressAuthor -> "autocite*"
                  AuthorInText   -> "textcite"
                  NormalCitation -> "autocite"

citationsToBiblatex (c:cs)
  | all (\cit -> null (citationPrefix cit) && null (citationSuffix cit)) (c:cs)
    = do
      let cmd = case citationMode c of
                    SuppressAuthor -> "\\autocite*"
                    AuthorInText   -> "\\textcite"
                    NormalCitation -> "\\autocite"
      return $ text cmd <>
               braces (text (intercalate "," (map citationId (c:cs))))
  | otherwise = do
    let cmd = case citationMode c of
                    SuppressAuthor -> "\\autocites*"
                    AuthorInText   -> "\\textcites"
                    NormalCitation -> "\\autocites"
    let convertOne Citation { citationId = k
                            , citationPrefix = p
                            , citationSuffix = s
                            }
                = citeArguments p s k
    args <- mapM convertOne (c:cs)
    return $ text cmd <> foldl' (<>) empty args

citationsToBiblatex _ = return empty

-- Determine listings language from list of class attributes.
getListingsLanguage :: [String] -> Maybe String
getListingsLanguage xs
  = foldr ((<|>) . toListingsLanguage) Nothing xs

mbBraced :: String -> String
mbBraced x = if not (all isAlphaNum x)
                then "{" <> x <> "}"
                else x

-- Extract a key from divs and spans
extract :: String -> Block -> [String]
extract key (Div attr _)     = lookKey key attr
extract key (Plain ils)      = query (extractInline key) ils
extract key (Para ils)       = query (extractInline key) ils
extract key (Header _ _ ils) = query (extractInline key) ils
extract _ _                  = []

-- Extract a key from spans
extractInline :: String -> Inline -> [String]
extractInline key (Span attr _) = lookKey key attr
extractInline _ _               = []

-- Look up a key in an attribute and give a list of its values
lookKey :: String -> Attr -> [String]
lookKey key (_,_,kvs) =  maybe [] words $ lookup key kvs

-- In environments \Arabic instead of \arabic is used
toPolyglossiaEnv :: Lang -> (String, String)
toPolyglossiaEnv l =
  case toPolyglossia l of
    ("arabic", o) -> ("Arabic", o)
    x             -> x

-- Takes a list of the constituents of a BCP 47 language code and
-- converts it to a Polyglossia (language, options) tuple
-- http://mirrors.ctan.org/macros/latex/contrib/polyglossia/polyglossia.pdf
toPolyglossia :: Lang -> (String, String)
toPolyglossia (Lang "ar" _ "DZ" _)        = ("arabic", "locale=algeria")
toPolyglossia (Lang "ar" _ "IQ" _)        = ("arabic", "locale=mashriq")
toPolyglossia (Lang "ar" _ "JO" _)        = ("arabic", "locale=mashriq")
toPolyglossia (Lang "ar" _ "LB" _)        = ("arabic", "locale=mashriq")
toPolyglossia (Lang "ar" _ "LY" _)        = ("arabic", "locale=libya")
toPolyglossia (Lang "ar" _ "MA" _)        = ("arabic", "locale=morocco")
toPolyglossia (Lang "ar" _ "MR" _)        = ("arabic", "locale=mauritania")
toPolyglossia (Lang "ar" _ "PS" _)        = ("arabic", "locale=mashriq")
toPolyglossia (Lang "ar" _ "SY" _)        = ("arabic", "locale=mashriq")
toPolyglossia (Lang "ar" _ "TN" _)        = ("arabic", "locale=tunisia")
toPolyglossia (Lang "de" _ _ vars)
  | "1901" `elem` vars                    = ("german", "spelling=old")
toPolyglossia (Lang "de" _ "AT" vars)
  | "1901" `elem` vars                    = ("german", "variant=austrian, spelling=old")
toPolyglossia (Lang "de" _ "AT" _)        = ("german", "variant=austrian")
toPolyglossia (Lang "de" _ "CH" vars)
  | "1901" `elem` vars                    = ("german", "variant=swiss, spelling=old")
toPolyglossia (Lang "de" _ "CH" _)        = ("german", "variant=swiss")
toPolyglossia (Lang "de" _ _ _)           = ("german", "")
toPolyglossia (Lang "dsb" _ _ _)          = ("lsorbian", "")
toPolyglossia (Lang "el" _ "polyton" _)   = ("greek",   "variant=poly")
toPolyglossia (Lang "en" _ "AU" _)        = ("english", "variant=australian")
toPolyglossia (Lang "en" _ "CA" _)        = ("english", "variant=canadian")
toPolyglossia (Lang "en" _ "GB" _)        = ("english", "variant=british")
toPolyglossia (Lang "en" _ "NZ" _)        = ("english", "variant=newzealand")
toPolyglossia (Lang "en" _ "UK" _)        = ("english", "variant=british")
toPolyglossia (Lang "en" _ "US" _)        = ("english", "variant=american")
toPolyglossia (Lang "grc" _ _ _)          = ("greek",   "variant=ancient")
toPolyglossia (Lang "hsb" _ _  _)         = ("usorbian", "")
toPolyglossia (Lang "la" _ _ vars)
  | "x-classic" `elem` vars               = ("latin", "variant=classic")
toPolyglossia (Lang "sl" _ _ _)           = ("slovenian", "")
toPolyglossia x                           = (commonFromBcp47 x, "")

-- Takes a list of the constituents of a BCP 47 language code and
-- converts it to a Babel language string.
-- http://mirrors.ctan.org/macros/latex/required/babel/base/babel.pdf
-- List of supported languages (slightly outdated):
-- http://tug.ctan.org/language/hyph-utf8/doc/generic/hyph-utf8/hyphenation.pdf
toBabel :: Lang -> String
toBabel (Lang "de" _ "AT" vars)
  | "1901" `elem` vars                  = "austrian"
  | otherwise                           = "naustrian"
toBabel (Lang "de" _ "CH" vars)
  | "1901" `elem` vars                  = "swissgerman"
  | otherwise                           = "nswissgerman"
toBabel (Lang "de" _ _ vars)
  | "1901" `elem` vars                  = "german"
  | otherwise                           = "ngerman"
toBabel (Lang "dsb" _ _ _)              = "lowersorbian"
toBabel (Lang "el" _ _ vars)
  | "polyton" `elem` vars               = "polutonikogreek"
toBabel (Lang "en" _ "AU" _)            = "australian"
toBabel (Lang "en" _ "CA" _)            = "canadian"
toBabel (Lang "en" _ "GB" _)            = "british"
toBabel (Lang "en" _ "NZ" _)            = "newzealand"
toBabel (Lang "en" _ "UK" _)            = "british"
toBabel (Lang "en" _ "US" _)            = "american"
toBabel (Lang "fr" _ "CA" _)            = "canadien"
toBabel (Lang "fra" _ _ vars)
  | "aca" `elem` vars                   = "acadian"
toBabel (Lang "grc" _ _ _)              = "polutonikogreek"
toBabel (Lang "hsb" _ _ _)              = "uppersorbian"
toBabel (Lang "la" _ _ vars)
  | "x-classic" `elem` vars             = "classiclatin"
toBabel (Lang "sl" _ _ _)               = "slovene"
toBabel x                               = commonFromBcp47 x

-- Takes a list of the constituents of a BCP 47 language code
-- and converts it to a string shared by Babel and Polyglossia.
-- https://tools.ietf.org/html/bcp47#section-2.1
commonFromBcp47 :: Lang -> String
commonFromBcp47 (Lang "pt" _ "BR" _)            = "brazil"
-- Note: documentation says "brazilian" works too, but it doesn't seem to work
-- on some systems.  See #2953.
commonFromBcp47 (Lang "sr" "Cyrl" _ _)          = "serbianc"
commonFromBcp47 (Lang "zh" "Latn" _ vars)
  | "pinyin" `elem` vars                        = "pinyin"
commonFromBcp47 (Lang l _ _ _) = fromIso l
  where
    fromIso "af"  = "afrikaans"
    fromIso "am"  = "amharic"
    fromIso "ar"  = "arabic"
    fromIso "as"  = "assamese"
    fromIso "ast" = "asturian"
    fromIso "bg"  = "bulgarian"
    fromIso "bn"  = "bengali"
    fromIso "bo"  = "tibetan"
    fromIso "br"  = "breton"
    fromIso "ca"  = "catalan"
    fromIso "cy"  = "welsh"
    fromIso "cs"  = "czech"
    fromIso "cop" = "coptic"
    fromIso "da"  = "danish"
    fromIso "dv"  = "divehi"
    fromIso "el"  = "greek"
    fromIso "en"  = "english"
    fromIso "eo"  = "esperanto"
    fromIso "es"  = "spanish"
    fromIso "et"  = "estonian"
    fromIso "eu"  = "basque"
    fromIso "fa"  = "farsi"
    fromIso "fi"  = "finnish"
    fromIso "fr"  = "french"
    fromIso "fur" = "friulan"
    fromIso "ga"  = "irish"
    fromIso "gd"  = "scottish"
    fromIso "gez" = "ethiopic"
    fromIso "gl"  = "galician"
    fromIso "he"  = "hebrew"
    fromIso "hi"  = "hindi"
    fromIso "hr"  = "croatian"
    fromIso "hu"  = "magyar"
    fromIso "hy"  = "armenian"
    fromIso "ia"  = "interlingua"
    fromIso "id"  = "indonesian"
    fromIso "ie"  = "interlingua"
    fromIso "is"  = "icelandic"
    fromIso "it"  = "italian"
    fromIso "jp"  = "japanese"
    fromIso "km"  = "khmer"
    fromIso "kmr" = "kurmanji"
    fromIso "kn"  = "kannada"
    fromIso "ko"  = "korean"
    fromIso "la"  = "latin"
    fromIso "lo"  = "lao"
    fromIso "lt"  = "lithuanian"
    fromIso "lv"  = "latvian"
    fromIso "ml"  = "malayalam"
    fromIso "mn"  = "mongolian"
    fromIso "mr"  = "marathi"
    fromIso "nb"  = "norsk"
    fromIso "nl"  = "dutch"
    fromIso "nn"  = "nynorsk"
    fromIso "no"  = "norsk"
    fromIso "nqo" = "nko"
    fromIso "oc"  = "occitan"
    fromIso "pa"  = "panjabi"
    fromIso "pl"  = "polish"
    fromIso "pms" = "piedmontese"
    fromIso "pt"  = "portuguese"
    fromIso "rm"  = "romansh"
    fromIso "ro"  = "romanian"
    fromIso "ru"  = "russian"
    fromIso "sa"  = "sanskrit"
    fromIso "se"  = "samin"
    fromIso "sk"  = "slovak"
    fromIso "sq"  = "albanian"
    fromIso "sr"  = "serbian"
    fromIso "sv"  = "swedish"
    fromIso "syr" = "syriac"
    fromIso "ta"  = "tamil"
    fromIso "te"  = "telugu"
    fromIso "th"  = "thai"
    fromIso "ti"  = "ethiopic"
    fromIso "tk"  = "turkmen"
    fromIso "tr"  = "turkish"
    fromIso "uk"  = "ukrainian"
    fromIso "ur"  = "urdu"
    fromIso "vi"  = "vietnamese"
    fromIso _     = ""

pDocumentOptions :: P.Parsec String () [String]
pDocumentOptions = do
  P.char '['
  opts <- P.sepBy
    (P.many $ P.spaces *> P.noneOf (" ,]" :: String) <* P.spaces)
    (P.char ',')
  P.char ']'
  return opts

pDocumentClass :: P.Parsec String () String
pDocumentClass =
  do P.skipMany (P.satisfy (/='\\'))
     P.string "\\documentclass"
     classOptions <- pDocumentOptions <|> return []
     if ("article" :: String) `elem` classOptions
       then return "article"
       else do P.skipMany (P.satisfy (/='{'))
               P.char '{'
               P.manyTill P.letter (P.char '}')