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
|
if not modules then modules = { } end modules ['font-odv'] = {
version = 1.001,
comment = "companion to font-ini.mkiv",
author = "Kai Eigner, TAT Zetwerk / Hans Hagen, PRAGMA ADE",
copyright = "TAT Zetwerk / PRAGMA ADE / ConTeXt Development Team",
license = "see context related readme files"
}
-- A few remarks:
--
-- This code is a partial rewrite of the code that deals with devanagari. The data and logic
-- is by Kai Eigner and based based on Microsoft's OpenType specifications for specific
-- scripts, but with a few improvements. More information can be found at:
--
-- deva: http://www.microsoft.com/typography/OpenType%20Dev/devanagari/introO.mspx
-- dev2: http://www.microsoft.com/typography/OpenType%20Dev/devanagari/intro.mspx
--
-- As I touched nearly all code, reshuffled it, optimized a lot, etc. etc. (imagine how
-- much can get messed up in over a week work) it could be that I introduced bugs. There
-- is more to gain (esp in the functions applied to a range) but I'll do that when
-- everything works as expected. Kai's original code is kept in font-odk.lua as a reference
-- so blame me (HH) for bugs.
--
-- Interesting is that Kai managed to write this on top of the existing otf handler. Only a
-- few extensions were needed, like a few more analyzing states and dealing with changed
-- head nodes in the core scanner as that only happens here. There's a lot going on here
-- and it's only because I touched nearly all code that I got a bit of a picture of what
-- happens. For in-depth knowledge one needs to consult Kai.
--
-- The rewrite mostly deals with efficiency, both in terms of speed and code. We also made
-- sure that it suits generic use as well as use in ConTeXt. I removed some buglets but can
-- as well have messed up the logic by doing this. For this we keep the original around
-- as that serves as reference. Due to the lots of reshuffling glyphs quite some leaks
-- occur(red) but once I'm satisfied with the rewrite I'll weed them. I also integrated
-- initialization etc into the regular mechanisms.
--
-- In the meantime, we're down from 25.5-3.5=22 seconds to 17.7-3.5=14.2 seconds for a 100
-- page sample (mid 2012) with both variants so it's worth the effort. Some more speedup is
-- to be expected. Due to the method chosen it will never be real fast. If I ever become a
-- power user I'll have a go at some further speed up. I will rename some functions (and
-- features) once we don't need to check the original code. We now use a special subset
-- sequence for use inside the analyzer (after all we could can store this in the dataset
-- and save redundant analysis).
--
-- I might go for an array approach with respect to attributes (and reshuffling). Easier.
--
-- Some data will move to char-def.lua (some day).
--
-- Hans Hagen, PRAGMA-ADE, Hasselt NL
-- Matras: according to Microsoft typography specifications "up to one of each type:
-- pre-, above-, below- or post- base", but that does not seem to be right. It could
-- become an option.
--
-- The next code looks weird anyway: the "and boolean" should move inside the if
-- or we should check differently (case vs successive).
--
-- local function ms_matra(c)
-- local prebase, abovebase, belowbase, postbase = true, true, true, true
-- local n = c.next
-- while n and n.id == glyph_code and n.subtype<256 and n.font == font do
-- local char = n.char
-- if not dependent_vowel[char] then
-- break
-- elseif pre_mark[char] and prebase then
-- prebase = false
-- elseif above_mark[char] and abovebase then
-- abovebase = false
-- elseif below_mark[char] and belowbase then
-- belowbase = false
-- elseif post_mark[char] and postbase then
-- postbase = false
-- else
-- return c
-- end
-- c = c.next
-- end
-- return c
-- end
-- todo: first test for font then for subtype
local insert, imerge = table.insert, table.imerge
local next = next
local trace_analyzing = false trackers.register("otf.analyzing", function(v) trace_analyzing = v end)
local report_devanagari = logs.reporter("otf","devanagari")
fonts = fonts or { }
fonts.analyzers = fonts.analyzers or { }
fonts.analyzers.methods = fonts.analyzers.methods or { node = { otf = { } } }
local otf = fonts.handlers.otf
local nodecodes = nodes.nodecodes
local glyph_code = nodecodes.glyph
local handlers = otf.handlers
local methods = fonts.analyzers.methods
local otffeatures = fonts.constructors.newfeatures("otf")
local registerotffeature = otffeatures.register
local processcharacters = nodes.handlers.characters
local insert_node_after = node.insert_after
local copy_node = node.copy
local free_node = node.free
local remove_node = node.remove
local flush_list = node.flush_list
local unsetvalue = attributes.unsetvalue
local fontdata = fonts.hashes.identifiers
local a_state = attributes.private('state')
local a_syllabe = attributes.private('syllabe')
local dotted_circle = 0x25CC
local states = fonts.analyzers.states -- not features
local s_rphf = states.rphf
local s_half = states.half
local s_pref = states.pref
local s_blwf = states.blwf
local s_pstf = states.pstf
-- In due time there will be entries here for scripts like Bengali, Gujarati,
-- Gurmukhi, Kannada, Malayalam, Oriya, Tamil, Telugu. Feel free to provide the
-- code points.
local consonant = {
[0x0915] = true, [0x0916] = true, [0x0917] = true, [0x0918] = true,
[0x0919] = true, [0x091A] = true, [0x091B] = true, [0x091C] = true,
[0x091D] = true, [0x091E] = true, [0x091F] = true, [0x0920] = true,
[0x0921] = true, [0x0922] = true, [0x0923] = true, [0x0924] = true,
[0x0925] = true, [0x0926] = true, [0x0927] = true, [0x0928] = true,
[0x0929] = true, [0x092A] = true, [0x092B] = true, [0x092C] = true,
[0x092D] = true, [0x092E] = true, [0x092F] = true, [0x0930] = true,
[0x0931] = true, [0x0932] = true, [0x0933] = true, [0x0934] = true,
[0x0935] = true, [0x0936] = true, [0x0937] = true, [0x0938] = true,
[0x0939] = true, [0x0958] = true, [0x0959] = true, [0x095A] = true,
[0x095B] = true, [0x095C] = true, [0x095D] = true, [0x095E] = true,
[0x095F] = true, [0x0979] = true, [0x097A] = true,
}
local independent_vowel = {
[0x0904] = true, [0x0905] = true, [0x0906] = true, [0x0907] = true,
[0x0908] = true, [0x0909] = true, [0x090A] = true, [0x090B] = true,
[0x090C] = true, [0x090D] = true, [0x090E] = true, [0x090F] = true,
[0x0910] = true, [0x0911] = true, [0x0912] = true, [0x0913] = true,
[0x0914] = true, [0x0960] = true, [0x0961] = true, [0x0972] = true,
[0x0973] = true, [0x0974] = true, [0x0975] = true, [0x0976] = true,
[0x0977] = true,
}
local dependent_vowel = { -- matra
[0x093A] = true, [0x093B] = true, [0x093E] = true, [0x093F] = true,
[0x0940] = true, [0x0941] = true, [0x0942] = true, [0x0943] = true,
[0x0944] = true, [0x0945] = true, [0x0946] = true, [0x0947] = true,
[0x0948] = true, [0x0949] = true, [0x094A] = true, [0x094B] = true,
[0x094C] = true, [0x094E] = true, [0x094F] = true, [0x0955] = true,
[0x0956] = true, [0x0957] = true, [0x0962] = true, [0x0963] = true,
}
local vowel_modifier = {
[0x0900] = true, [0x0901] = true, [0x0902] = true, [0x0903] = true,
-- A8E0 - A8F1 are cantillation marks for the Samaveda and may not belong here.
[0xA8E0] = true, [0xA8E1] = true, [0xA8E2] = true, [0xA8E3] = true,
[0xA8E4] = true, [0xA8E5] = true, [0xA8E6] = true, [0xA8E7] = true,
[0xA8E8] = true, [0xA8E9] = true, [0xA8EA] = true, [0xA8EB] = true,
[0xA8EC] = true, [0xA8ED] = true, [0xA8EE] = true, [0xA8EF] = true,
[0xA8F0] = true, [0xA8F1] = true,
}
local stress_tone_mark = {
[0x0951] = true, [0x0952] = true, [0x0953] = true, [0x0954] = true,
}
local c_nukta = 0x093C -- used to be tables
local c_halant = 0x094D -- used to be tables
local c_ra = 0x0930 -- used to be tables
local c_anudatta = 0x0952 -- used to be tables
local c_nbsp = 0x00A0 -- used to be tables
local c_zwnj = 0x200C -- used to be tables
local c_zwj = 0x200D -- used to be tables
local zw_char = { -- could also be inlined
[0x200C] = true,
[0x200D] = true,
}
local pre_mark = {
[0x093F] = true, [0x094E] = true,
}
local above_mark = {
[0x0900] = true, [0x0901] = true, [0x0902] = true, [0x093A] = true,
[0x0945] = true, [0x0946] = true, [0x0947] = true, [0x0948] = true,
[0x0951] = true, [0x0953] = true, [0x0954] = true, [0x0955] = true,
[0xA8E0] = true, [0xA8E1] = true, [0xA8E2] = true, [0xA8E3] = true,
[0xA8E4] = true, [0xA8E5] = true, [0xA8E6] = true, [0xA8E7] = true,
[0xA8E8] = true, [0xA8E9] = true, [0xA8EA] = true, [0xA8EB] = true,
[0xA8EC] = true, [0xA8ED] = true, [0xA8EE] = true, [0xA8EF] = true,
[0xA8F0] = true, [0xA8F1] = true,
}
local below_mark = {
[0x093C] = true, [0x0941] = true, [0x0942] = true, [0x0943] = true,
[0x0944] = true, [0x094D] = true, [0x0952] = true, [0x0956] = true,
[0x0957] = true, [0x0962] = true, [0x0963] = true,
}
local post_mark = {
[0x0903] = true, [0x093B] = true, [0x093E] = true, [0x0940] = true,
[0x0949] = true, [0x094A] = true, [0x094B] = true, [0x094C] = true,
[0x094F] = true,
}
local mark_four = { } -- As we access these frequently an extra hash is used.
for k, v in next, pre_mark do mark_four[k] = pre_mark end
for k, v in next, above_mark do mark_four[k] = above_mark end
for k, v in next, below_mark do mark_four[k] = below_mark end
for k, v in next, post_mark do mark_four[k] = post_mark end
local mark_above_below_post = { }
for k, v in next, above_mark do mark_above_below_post[k] = above_mark end
for k, v in next, below_mark do mark_above_below_post[k] = below_mark end
for k, v in next, post_mark do mark_above_below_post[k] = post_mark end
-- Again, this table can be extended for other scripts than devanagari. Actually,
-- for ConTeXt this kind of dat is kept elsewhere so eventually we might move
-- tables to someplace else.
local reorder_class = {
[0x0930] = "before postscript",
[0x093F] = "before half",
[0x0940] = "after subscript",
[0x0941] = "after subscript",
[0x0942] = "after subscript",
[0x0943] = "after subscript",
[0x0944] = "after subscript",
[0x0945] = "after subscript",
[0x0946] = "after subscript",
[0x0947] = "after subscript",
[0x0948] = "after subscript",
[0x0949] = "after subscript",
[0x094A] = "after subscript",
[0x094B] = "after subscript",
[0x094C] = "after subscript",
[0x0962] = "after subscript",
[0x0963] = "after subscript",
[0x093E] = "after subscript",
}
-- We use some pseudo features as we need to manipulate the nodelist based
-- on information in the font as well as already applied features.
local dflt_true = {
dflt = true
}
local dev2_defaults = {
dev2 = dflt_true,
}
local deva_defaults = {
dev2 = dflt_true,
deva = dflt_true,
}
local false_flags = { false, false, false, false }
local both_joiners_true = {
[0x200C] = true,
[0x200D] = true,
}
local sequence_reorder_matras = {
chain = 0,
features = { dv01 = dev2_defaults },
flags = false_flags,
name = "dv01_reorder_matras",
subtables = { "dv01_reorder_matras" },
type = "devanagari_reorder_matras",
}
local sequence_reorder_reph = {
chain = 0,
features = { dv02 = dev2_defaults },
flags = false_flags,
name = "dv02_reorder_reph",
subtables = { "dv02_reorder_reph" },
type = "devanagari_reorder_reph",
}
local sequence_reorder_pre_base_reordering_consonants = {
chain = 0,
features = { dv03 = dev2_defaults },
flags = false_flags,
name = "dv03_reorder_pre_base_reordering_consonants",
subtables = { "dv03_reorder_pre_base_reordering_consonants" },
type = "devanagari_reorder_pre_base_reordering_consonants",
}
local sequence_remove_joiners = {
chain = 0,
features = { dv04 = deva_defaults },
flags = false_flags,
name = "dv04_remove_joiners",
subtables = { "dv04_remove_joiners" },
type = "devanagari_remove_joiners",
}
-- Looping over feature twice as efficient as looping over basic forms (some
-- 350 checks instead of 750 for one font). This is something to keep an eye on
-- as it might depends on the font. Not that it's a bottleneck.
local basic_shaping_forms = {
nukt = true,
akhn = true,
rphf = true,
pref = true,
rkrf = true,
blwf = true,
half = true,
pstf = true,
vatu = true,
cjct = true,
}
local function initializedevanagi(tfmdata)
local script, language = otf.scriptandlanguage(tfmdata,attr) -- todo: take fast variant
if script == "deva" or script == "dev2" then
local resources = tfmdata.resources
local lookuphash = resources.lookuphash
if not lookuphash["dv01"] then
report_devanagari("adding devanagari features to font")
--
local features = resources.features
local gsubfeatures = features.gsub
local sequences = resources.sequences
local sharedfeatures = tfmdata.shared.features
--
local lastmatch = 0
for s=1,#sequences do -- classify chars
local features = sequences[s].features
if features then
for k, v in next, features do
if basic_shaping_forms[k] then
lastmatch = s
end
end
end
end
local insertindex = lastmatch + 1
--
lookuphash["dv04_remove_joiners"] = both_joiners_true
--
gsubfeatures["dv01"] = dev2_defaults -- reorder matras
gsubfeatures["dv02"] = dev2_defaults -- reorder reph
gsubfeatures["dv03"] = dev2_defaults -- reorder pre base reordering consonants
gsubfeatures["dv04"] = deva_defaults -- remove joiners
--
insert(sequences,insertindex,sequence_reorder_pre_base_reordering_consonants)
insert(sequences,insertindex,sequence_reorder_reph)
insert(sequences,insertindex,sequence_reorder_matras)
insert(sequences,insertindex,sequence_remove_joiners)
--
if script == "deva" then
sharedfeatures["dv04"] = true -- dv04_remove_joiners
end
--
if script == "dev2" then
sharedfeatures["dv01"] = true -- dv01_reorder_matras
sharedfeatures["dv02"] = true -- dv02_reorder_reph
sharedfeatures["dv03"] = true -- dv03_reorder_pre_base_reordering_consonants
sharedfeatures["dv04"] = true -- dv04_remove_joiners
end
--
end
end
end
registerotffeature {
name = "devanagari",
description = "inject additional features",
default = true,
initializers = {
node = initializedevanagi,
},
}
-- hm, this is applied to one character:
local function deva_initialize(font,attr)
local tfmdata = fontdata[font]
local resources = tfmdata.resources
local lookuphash = resources.lookuphash
local datasets = otf.dataset(tfmdata,font,attr)
local devanagaridata = datasets.devanagari
if devanagaridata then -- maybe also check for e.g. reph
return lookuphash, devanagaridata.reph, devanagaridata.vattu, devanagaridata.blwfcache
else
devanagaridata = { }
datasets.devanagari = devanagaridata
local reph = false
local vattu = false
local blwfcache = { }
local sequences = resources.sequences
for s=1,#sequences do -- triggers creation of dataset
-- local sequence = sequences[s]
local dataset = datasets[s]
if dataset and dataset[1] then -- value
local kind = dataset[4]
if kind == "rphf" then
-- deva
reph = true
elseif kind == "blwf" then
-- deva
vattu = true
-- dev2
-- local subtables = sequence.subtables -- dataset[5].subtables
local subtables = dataset[5].subtables
for i=1,#subtables do
local lookupname = subtables[i]
local lookupcache = lookuphash[lookupname]
if lookupcache then
for k, v in next, lookupcache do
blwfcache[k] = blwfcache[k] or v
end
end
end
end
end
end
devanagaridata.reph = reph
devanagaridata.vattu = vattu
devanagaridata.blwfcache = blwfcache
return lookuphash, reph, vattu, blwfcache
end
end
local function deva_reorder(head,start,stop,font,attr)
local lookuphash, reph, vattu, blwfcache = deva_initialize(font,attr) -- could be inlines but ugly
local current = start
local n = start.next
local base = nil
local firstcons = nil
local lastcons = nil
local basefound = false
if start.char == c_ra and n.char == c_halant and reph then
-- if syllable starts with Ra + H and script has 'Reph' then exclude Reph
-- from candidates for base consonants
if n == stop then
return head, stop
end
if n.next.char == c_zwj then
current = start
else
current = n.next
start[a_state] = s_rphf
end
end
if current.char == c_nbsp then
-- Stand Alone cluster
if current == stop then
stop = stop.prev
head = remove_node(head,current)
free_node(current)
return head, stop
else
base, firstcons, lastcons = current, current, current
current = current.next
if current ~= stop then
if current.char == c_nukta then
current = current.next
end
if current.char == c_zwj then
if current ~= stop then
local next = current.next
if next ~= stop and next.char == c_halant then
current = next
next = current.next
local tmp = next.next
local changestop = next == stop
local tempcurrent = copy_node(next)
local nextcurrent = copy_node(current)
tempcurrent.next = nextcurrent
nextcurrent.prev = tempcurrent
tempcurrent[a_state] = s_blwf
tempcurrent = processcharacters(tempcurrent)
tempcurrent[a_state] = unsetvalue
if next.char == tempcurrent.char then
flush_list(tempcurrent)
local n = copy_node(current)
current.char = dotted_circle
head = insert_node_after(head, current, n)
else
current.char = tempcurrent.char -- (assumes that result of blwf consists of one node)
local freenode = current.next
current.next = tmp
tmp.prev = current
free_node(freenode)
flush_list(tempcurrent)
if changestop then
stop = current
end
end
end
end
end
end
end
end
while not basefound do
-- find base consonant
if consonant[current.char] then
current[a_state] = s_half
if not firstcons then
firstcons = current
end
lastcons = current
if not base then
base = current
elseif blwfcache[current.char] then
-- consonant has below-base (or post-base) form
current[a_state] = s_blwf
else
base = current
end
end
basefound = current == stop
current = current.next
end
if base ~= lastcons then
-- if base consonant is not last one then move halant from base consonant to last one
local np = base
local n = base.next
if n.char == c_nukta then
np = n
n = n.next
end
if n.char == c_halant then
if lastcons ~= stop then
local ln = lastcons.next
if ln.char == c_nukta then
lastcons = ln
end
end
-- local np = n.prev
local nn = n.next
local ln = lastcons.next -- what if lastcons is nn ?
np.next = nn
nn.prev = np
lastcons.next = n
if ln then
ln.prev = n
end
n.next = ln
n.prev = lastcons
if lastcons == stop then
stop = n
end
end
end
n = start.next
if start.char == c_ra and n.char == c_halant and not (n ~= stop and zw_char[n.next.char]) then
-- if syllable starts with Ra + H then move this combination so that it follows either:
-- the post-base 'matra' (if any) or the base consonant
local matra = base
if base ~= stop then
local next = base.next
if dependent_vowel[next.char] then
matra = next
end
end
-- [sp][start][n][nn] [matra|base][?]
-- [matra|base][start] [n][?] [sp][nn]
local sp = start.prev
local nn = n.next
local mn = matra.next
if sp then
sp.next = nn
end
nn.prev = sp
matra.next = start
start.prev = matra
n.next = mn
if mn then
mn.prev = n
end
if head == start then
head = nn
end
start = nn
if matra == stop then
stop = n
end
end
local current = start
while current ~= stop do
local next = current.next
if next ~= stop and next.char == c_halant and next.next.char == c_zwnj then
current[a_state] = unsetvalue
end
current = next
end
if base ~= stop and base[a_state] then
local next = base.next
if next.char == c_halant and not (next ~= stop and next.next.char == c_zwj) then
base[a_state] = unsetvalue
end
end
-- ToDo: split two- or three-part matras into their parts. Then, move the left 'matra' part to the beginning of the syllable.
-- Not necessary for Devanagari. However it is necessay for other scripts, such as Tamil (e.g. TAMIL VOWEL SIGN O - 0BCA)
-- classify consonants and 'matra' parts as pre-base, above-base (Reph), below-base or post-base, and group elements of the syllable (consonants and 'matras') according to this classification
local current, allreordered, moved = start, false, { [base] = true }
local a, b, p, bn = base, base, base, base.next
if base ~= stop and bn.char == c_nukta then
a, b, p = bn, bn, bn
end
while not allreordered do
-- current is always consonant
local c = current
local n = current.next
local l = nil -- used ?
if c ~= stop then
if n.char == c_nukta then
c = n
n = n.next
end
if c ~= stop then
if n.char == c_halant then
c = n
n = n.next
end
while c ~= stop and dependent_vowel[n.char] do
c = n
n = n.next
end
if c ~= stop then
if vowel_modifier[n.char] then
c = n
n = n.next
end
if c ~= stop and stress_tone_mark[n.char] then
c = n
n = n.next
end
end
end
end
local bp = firstcons.prev
local cn = current.next
local last = c.next
while cn ~= last do
-- move pre-base matras...
if pre_mark[cn.char] then
if bp then
bp.next = cn
end
local next = cn.next
local prev = cn.prev
if next then
next.prev = prev
end
prev.next = next
if cn == stop then
stop = prev
end
cn.prev = bp
cn.next = firstcons
firstcons.prev = cn
if firstcons == start then
if head == start then
head = cn
end
start = cn
end
break
end
cn = cn.next
end
allreordered = c == stop
current = c.next
end
if reph or vattu then
local current, cns = start, nil
while current ~= stop do
local c = current
local n = current.next
if current.char == c_ra and n.char == c_halant then
c = n
n = n.next
local b, bn = base, base
while bn ~= stop do
local next = bn.next
if dependent_vowel[next.char] then
b = next
end
bn = next
end
if current[a_state] == s_rphf then
-- position Reph (Ra + H) after post-base 'matra' (if any) since these
-- become marks on the 'matra', not on the base glyph
if b ~= current then
if current == start then
if head == start then
head = n
end
start = n
end
if b == stop then
stop = c
end
local prev = current.prev
if prev then
prev.next = n
end
if n then
n.prev = prev
end
local next = b.next
c.next = next
if next then
next.prev = c
end
c.next = next
b.next = current
current.prev = b
end
elseif cns and cns.next ~= current then
-- position below-base Ra (vattu) following the consonants on which it is placed (either the base consonant or one of the pre-base consonants)
local cp, cnsn = current.prev, cns.next
if cp then
cp.next = n
end
if n then
n.prev = cp
end
cns.next = current
current.prev = cns
c.next = cnsn
if cnsn then
cnsn.prev = c
end
if c == stop then
stop = cp
break
end
current = n.prev
end
else
local char = current.char
if consonant[char] or char == c_nbsp then -- maybe combined hash
cns = current
local next = cns.next
if next.char == c_halant then
cns = next
end
end
end
current = current.next
end
end
if base.char == c_nbsp then
head = remove_node(head,base)
free_node(base)
end
return head, stop
end
-- If a pre-base matra character had been reordered before applying basic features,
-- the glyph can be moved closer to the main consonant based on whether half-forms had been formed.
-- Actual position for the matra is defined as “after last standalone halant glyph,
-- after initial matra position and before the main consonant”.
-- If ZWJ or ZWNJ follow this halant, position is moved after it.
-- so we break out ... this is only done for the first 'word' (if we feed words we can as
-- well test for non glyph.
function handlers.devanagari_reorder_matras(head,start,kind,lookupname,replacement) -- no leak
local current = start -- we could cache attributes here
local startfont = start.font
local startattr = start[a_syllabe]
-- can be fast loop
while current and current.id == glyph_code and current.subtype<256 and current.font == font and current[a_syllabe] == startattr do
local next = current.next
if current.char == c_halant and not current[a_state] then
if next and next.id == glyph_code and next.subtype<256 and next.font == font and next[a_syllabe] == startattr and zw_char[next.char] then
current = next
end
local startnext = start.next
head = remove_node(head,start)
local next = current.next
if next then
next.prev = start
end
start.next = next
current.next = start
start.prev = current
start = startnext
break
end
current = next
end
return head, start, true
end
-- todo: way more caching of attributes and font
-- Reph’s original position is always at the beginning of the syllable, (i.e. it is not reordered at the character reordering stage).
-- However, it will be reordered according to the basic-forms shaping results.
-- Possible positions for reph, depending on the script, are; after main, before post-base consonant forms,
-- and after post-base consonant forms.
-- 1 If reph should be positioned after post-base consonant forms, proceed to step 5.
-- 2 If the reph repositioning class is not after post-base: target position is after the first explicit halant glyph between
-- the first post-reph consonant and last main consonant. If ZWJ or ZWNJ are following this halant, position is moved after it.
-- If such position is found, this is the target position. Otherwise, proceed to the next step.
-- Note: in old-implementation fonts, where classifications were fixed in shaping engine,
-- there was no case where reph position will be found on this step.
-- 3 If reph should be repositioned after the main consonant: from the first consonant not ligated with main,
-- or find the first consonant that is not a potential pre-base reordering Ra.
-- 4 If reph should be positioned before post-base consonant, find first post-base classified consonant not ligated with main.
-- If no consonant is found, the target position should be before the first matra, syllable modifier sign or vedic sign.
-- 5 If no consonant is found in steps 3 or 4, move reph to a position immediately before the first post-base matra,
-- syllable modifier sign or vedic sign that has a reordering class after the intended reph position.
-- For example, if the reordering position for reph is post-main, it will skip above-base matras that also have a post-main position.
-- 6 Otherwise, reorder reph to the end of the syllable.
-- hm, this only looks at the start of a nodelist ... is this supposed to be line based?
function handlers.devanagari_reorder_reph(head,start,kind,lookupname,replacement)
-- since in Devanagari reph has reordering position 'before postscript' dev2 only follows step 2, 4, and 6,
-- the other steps are still ToDo (required for scripts other than dev2)
local current = start.next
local startnext = nil
local startprev = nil
local startfont = start.font
local startattr = start[a_syllabe]
while current and current.id == glyph_code and current.subtype<256 and current.font == startfont and current[a_syllabe] == startattr do --step 2
if current.char == c_halant and not current[a_state] then
local next = current.next
if next and next.id == glyph_code and next.subtype<256 and next.font == startfont and next[a_syllabe] == startattr and zw_char[next.char] then
current = next
end
startnext = start.next
head = remove_node(head,start)
local next = current.next
if next then
next.prev = start
end
start.next = next
current.next = start
start.prev = current
start = startnext
startattr = start[a_syllabe]
break
end
current = current.next
end
if not startnext then
current = start.next
while current and current.id == glyph_code and current.subtype<256 and current.font == startfont and current[a_syllabe] == startattr do --step 4
if current[a_state] == s_pstf then --post-base
startnext = start.next
head = remove_node(head,start)
local prev = current.prev
start.prev = prev
prev.next = start
start.next = current
current.prev = start
start = startnext
startattr = start[a_syllabe]
break
end
current = current.next
end
end
-- ToDo: determine position for reph with reordering position other than 'before postscript'
-- (required for scripts other than dev2)
-- leaks
if not startnext then
current = start.next
local c = nil
while current and current.id == glyph_code and current.subtype<256 and current.font == startfont and current[a_syllabe] == startattr do --step 5
if not c then
local char = current.char
-- todo: combine in one
if mark_above_below_post[char] and reorder_class[char] ~= "after subscript" then
c = current
end
end
current = current.next
end
-- here we can loose the old start node: maybe best split cases
if c then
startnext = start.next
head = remove_node(head,start)
local prev = c.prev
start.prev = prev
prev.next = start
start.next = c
c.prev = start
-- end
start = startnext
startattr = start[a_syllabe]
end
end
-- leaks
if not startnext then
current = start
local next = current.next
while next and next.id == glyph_code and next.subtype<256 and next.font == startfont and next[a_syllabe] == startattr do --step 6
current = next
next = current.next
end
if start ~= current then
startnext = start.next
head = remove_node(head,start)
local next = current.next
if next then
next.prev = start
end
start.next = next
current.next = start
start.prev = current
start = startnext
end
end
--
return head, start, true
end
-- we can cache some checking (v)
-- If a pre-base reordering consonant is found, reorder it according to the following rules:
--
-- 1 Only reorder a glyph produced by substitution during application of the feature.
-- (Note that a font may shape a Ra consonant with the feature generally but block it in certain contexts.)
-- 2 Try to find a target position the same way as for pre-base matra. If it is found, reorder pre-base consonant glyph.
-- 3 If position is not found, reorder immediately before main consonant.
-- UNTESTED: NOT CALLED IN EXAMPLE
function handlers.devanagari_reorder_pre_base_reordering_consonants(head,start,kind,lookupname,replacement)
local current = start
local startnext = nil
local startprev = nil
local startfont = start.font
local startattr = start[a_syllabe]
-- can be fast for loop + caching state
while current and current.id == glyph_code and current.subtype<256 and current.font == startfont and current[a_syllabe] == startattr do
local next = current.next
if current.char == c_halant and not current[a_state] then
if next and next.id == glyph_code and next.subtype<256 and next.font == font and next[a_syllabe] == startattr then
local char = next.char
if char == c_zwnj or char == c_zwj then
current = next
end
end
startnext = start.next
removenode(start,start)
local next = current.next
if next then
next.prev = start
end
start.next = next
current.next = start
start.prev = current
start = startnext
break
end
current = next
end
if not startnext then
current = start.next
startattr = start[a_syllabe]
while current and current.id == glyph_code and current.subtype<256 and current.font == startfont and current[a_syllabe] == startattr do
if not consonant[current.char] and current[a_state] then --main
startnext = start.next
removenode(start,start)
local prev = current.prev
start.prev = prev
prev.next = start
start.next = current
current.prev = start
start = startnext
break
end
current = current.next
end
end
return head, start, true
end
function handlers.devanagari_remove_joiners(head,start,kind,lookupname,replacement)
local stop = start.next
local startfont = start.font
while stop and stop.id == glyph_code and stop.subtype<256 and stop.font == startfont do
local char = stop.char
if char == c_zwnj or char == c_zwj then
stop = stop.next
else
break
end
end
if stop then
stop.prev.next = nil
stop.prev = start.prev
end
local prev = start.prev
if prev then
prev.next = stop
end
flush_list(start)
return head, stop, true
end
local valid = {
rphf = true,
pref = true,
half = true,
blwf = true,
pstf = true,
}
local function dev2_initialize(font,attr)
local tfmdata = fontdata[font]
local resources = tfmdata.resources
local lookuphash = resources.lookuphash
local datasets = otf.dataset(tfmdata,font,attr)
local devanagaridata = datasets.devanagari
if devanagaridata then -- maybe also check for e.g. seqsubset
return lookuphash, devanagaridata.seqsubset
else
devanagaridata = { }
datasets.devanagari = devanagaridata
local seqsubset = { }
devanagaridata.seqsubset = seqsubset
local sequences = resources.sequences
for s=1,#sequences do
-- local sequence = sequences[s]
local dataset = datasets[s]
if dataset and dataset[1] then -- featurevalue
local kind = dataset[4]
if kind and valid[kind] then
-- could become a function call
-- local subtables = sequence.subtables
local subtables = dataset[5].subtables
for i=1,#subtables do
local lookupname = subtables[i]
local lookupcache = lookuphash[lookupname]
if lookupcache then
local reph = false
local chain = dataset[3]
if chain ~= 0 then --rphf is result of of chain
--ToDo: rphf might be result of other handler/chainproc
else
reph = lookupcache[0x0930]
if reph then
reph = reph[0x094D]
if reph then
reph = reph["ligature"]
end
end
--ToDo: rphf actualy acts on consonant + halant. This consonant might not necesseraly be 0x0930 ... (but fot dev2 it is)
end
seqsubset[#seqsubset+1] = { kind, lookupcache, reph }
end
end
end
end
end
lookuphash["dv01_reorder_matras"] = pre_mark -- move to initializer ?
return lookuphash, seqsubset
end
end
-- this one will be merged into the caller: it saves a call, but we will then make function
-- of the actions
local function dev2_reorder(head,start,stop,font,attr) -- maybe do a pass over (determine stop in sweep)
local lookuphash, seqsubset = dev2_initialize(font,attr)
local reph, pre_base_reordering_consonants = false, { } -- was nil ... probably went unnoticed because never assigned
local halfpos, basepos, subpos, postpos = nil, nil, nil, nil
local locl = { }
for i=1,#seqsubset do
-- maybe quit if start == stop
local subset = seqsubset[i]
local kind = subset[1]
local lookupcache = subset[2]
if kind == "rphf" then
if subset[3] then
reph = true
end
local current = start
local last = stop.next
while current ~= last do
if current ~= stop then
local c = locl[current] or current.char
local found = lookupcache[c]
if found then
local next = current.next
local n = locl[next] or next.char
if found[n] then --above-base: rphf Consonant + Halant
local afternext = next ~= stop and next.next
if afternext and zw_char[afternext.char] then -- ZWJ and ZWNJ prevent creation of reph
current = next
current = current.next
elseif current == start then
current[a_state] = s_rphf
current = next
else
current = next
end
end
end
end
current = current.next
end
elseif kind == "pref" then
-- why not global? pretty ineffient this way
-- this will move to the initializer and we will store the hash in dataset
for k, v in lookupcache[0x094D], next do
pre_base_reordering_consonants[k] = v and v["ligature"] --ToDo: reph might also be result of chain
end
--
local current = start
local last = stop.next
while current ~= last do
if current ~= stop then
local c = locl[current] or current.char
local found = lookupcache[c]
if found then
local next = current.next
local n = locl[next] or next.char
if found[n] then
current[a_state] = s_pref
next[a_state] = s_pref
current = next
end
end
end
current = current.next
end
elseif kind == "half" then -- half forms: half / Consonant + Halant
local current = start
local last = stop.next
while current ~= last do
if current ~= stop then
local c = locl[current] or current.char
local found = lookupcache[c]
if found then
local next = current.next
local n = locl[next] or next.char
if found[n] then
if next ~= stop and next.next.char == c_zwnj then --ZWNJ prevent creation of half
current = current.next
else
current[a_state] = s_half
if not halfpos then
halfpos = current
end
end
current = next
end
end
end
current = current.next
end
elseif kind == "blwf" then -- below-base: blwf / Halant + Consonant
local current = start
local last = stop.next
while current ~= last do
if current ~= stop then
local c = locl[current] or current.char
local found = lookupcache[c]
if found then
local next = current.next
local n = locl[next] or next.char
if found[n] then
current[a_state] = s_blwf
next[a_state] = s_blwf
current = next
subpos = current
end
end
end
current = current.next
end
elseif kind == "pstf" then -- post-base: pstf / Halant + Consonant
local current = start
local last = stop.next
while current ~= last do
if current ~= stop then
local c = locl[current] or current.char
local found = lookupcache[c]
if found then
local next = current.next
local n = locl[next] or next.char
if found[n] then
current[a_state] = s_pstf
next[a_state] = s_pstf
current = next
postpos = current
end
end
end
current = current.next
end
end
end
-- this one changes per word
lookuphash["dv02_reorder_reph"] = { [reph] = true }
lookuphash["dv03_reorder_pre_base_reordering_consonants"] = pre_base_reordering_consonants
local current, base, firstcons = start, nil, nil
if start[a_state] == s_rphf then
-- if syllable starts with Ra + H and script has 'Reph' then exclude Reph from candidates for base consonants
current = start.next.next
end
if current ~= stop.next and current.char == c_nbsp then
-- Stand Alone cluster
if current == stop then
stop = stop.prev
head = remove_node(head,current)
free_node(current)
return head, stop
else
base = current
current = current.next
if current ~= stop then
local char = current.char
if char == c_nukta then
current = current.next
char = current.char
end
if char == c_zwj then
local next = current.next
if current ~= stop and next ~= stop and next.char == c_halant then
current = next
next = current.next
local tmp = next.next
local changestop = next == stop
next.next = nil
current[a_state] = s_pref
current = processcharacters(current)
current[a_state] = s_blwf
current = processcharacters(current)
current[a_state] = s_pstf
current = processcharacters(current)
current[a_state] = unsetvalue
if current.char == c_halant then
current.next.next = tmp
local nc = copy_node(current)
current.char = dotted_circle
head = insert_node_after(head,current,nc)
else
current.next = tmp -- assumes that result of pref, blwf, or pstf consists of one node
if changestop then
stop = current
end
end
end
end
end
end
else -- not Stand Alone cluster
local last = stop.next
while current ~= last do -- find base consonant
local next = current.next
if consonant[current.char] then
if not (current ~= stop and next ~= stop and next.char == c_halant and next.next.char == c_zwj) then
if not firstcons then
firstcons = current
end
-- check whether consonant has below-base or post-base form or is pre-base reordering Ra
local a = current[a_state]
if not (a == s_pref or a == s_blwf or a == pstf) then
base = current
end
end
end
current = next
end
if not base then
base = firstcons
end
end
if not base then
if start[a_state] == s_rphf then
start[a_state] = unsetvalue
end
return head, stop
else
if base[a_state] then
base[a_state] = unsetvalue
end
basepos = base
end
if not halfpos then
halfpos = base
end
if not subpos then
subpos = base
end
if not postpos then
postpos = subpos or base
end
-- Matra characters are classified and reordered by which consonant in a conjunct they have affinity for
local moved = { }
local current = start
local last = stop.next
while current ~= last do
local char, target, cn = locl[current] or current.char, nil, current.next
if not moved[current] and dependent_vowel[char] then
if pre_mark[char] then -- Before first half form in the syllable
moved[current] = true
local prev = current.prev
local next = current.next
if prev then
prev.next = next
end
if next then
next.prev = prev
end
if current == stop then
stop = current.prev
end
if halfpos == start then
if head == start then
head = current
end
start = current
end
local prev = halfpos.prev
if prev then
prev.next = current
end
current.prev = prev
halfpos.prev = current
current.next = halfpos
halfpos = current
elseif above_mark[char] then -- After main consonant
target = basepos
if subpos == basepos then
subpos = current
end
if postpos == basepos then
postpos = current
end
basepos = current
elseif below_mark[char] then -- After subjoined consonants
target = subpos
if postpos == subpos then
postpos = current
end
subpos = current
elseif post_mark[char] then -- After post-form consonant
target = postpos
postpos = current
end
if mark_above_below_post[char] then
local prev = current.prev
if prev ~= target then
local next = current.next
if prev then -- not needed, already tested with target
prev.next = next
end
if next then
next.prev = prev
end
if current == stop then
stop = prev
end
local next = target.next
if next then
next.prev = current
end
current.next = next
target.next = current
current.prev = target
end
end
end
current = cn
end
-- Reorder marks to canonical order: Adjacent nukta and halant or nukta and vedic sign are always repositioned if necessary, so that the nukta is first.
local current, c = start, nil
while current ~= stop do
local char = current.char
if char == c_halant or stress_tone_mark[char] then
if not c then
c = current
end
else
c = nil
end
local next = current.next
if c and next.char == c_nukta then
if head == c then
head = next
end
if stop == next then
stop = current
end
local prev = c.prev
if prev then
prev.next = next
end
next.prev = prev
local nextnext = next.next
current.next = nextnext
local nextnextnext = nextnext.next
if nextnextnext then
nextnextnext.prev = current
end
c.prev = nextnext
nextnext.next = c
end
if stop == current then break end
current = current.next
end
if base.char == c_nbsp then
head = remove_node(head, base)
free_node(base)
end
return head, stop
end
-- cleaned up and optimized ... needs checking (local, check order, fixes, extra hash, etc)
local separator = { }
imerge(separator,consonant)
imerge(separator,independent_vowel)
imerge(separator,dependent_vowel)
imerge(separator,vowel_modifier)
imerge(separator,stress_tone_mark)
separator[0x093C] = true -- nukta
separator[0x094D] = true -- halant
local function analyze_next_chars_one(c,font,variant) -- skip one dependent vowel
-- why two variants ... the comment suggests that it's the same ruleset
local n = c.next
if not n then
return c
end
if variant == 1 then
local v = n.id == glyph_code and n.subtype<256 and n.font == font
if v and n.char == c_nukta then
n = n.next
if n then
v = n.id == glyph_code and n.subtype<256 and n.font == font
end
end
if n and v then
local nn = n.next
if nn and nn.id == glyph_code and nn.subtype<256 and nn.font == font then
local nnn = nn.next
if nnn and nnn.id == glyph_code and nnn.subtype<256 and nnn.font == font then
local nnc = nn.char
local nnnc = nnn.char
if nnc == c_zwj and consonant[nnnc] then
c = nnn
elseif (nnc == c_zwnj or nnc == c_zwj) and nnnc == c_halant then
local nnnn = nnn.next
if nnnn and nnnn.id == glyph_code and consonant[nnnn.char] and nnnn.subtype<256 and nnnn.font == font then
c = nnnn
end
end
end
end
end
elseif variant == 2 then
if n.id == glyph_code and n.char == c_nukta and n.subtype<256 and n.font == font then
c = n
end
n = c.next
if n and n.id == glyph_code and n.subtype<256 and n.font == font then
local nn = n.next
if nn then
local nv = nn.id == glyph_code and nn.subtype<256 and nn.font == font
if nv and zw_char[n.char] then
n = nn
nn = nn.next
nv = nn.id == glyph_code and nn.subtype<256 and nn.font == font
end
if nn and nv and n.char == c_halant and consonant[nn.char] then
c = nn
end
end
end
end
-- c = ms_matra(c)
local n = c.next
if not n then
return c
end
local v = n.id == glyph_code and n.subtype<256 and n.font == font
if not v then
return c
end
local char = n.char
if dependent_vowel[char] then
c = c.next
n = c.next
if not n then
return c
end
v = n.id == glyph_code and n.subtype<256 and n.font == font
if not v then
return c
end
char = n.char
end
if char == c_nukta then
c = c.next
n = c.next
if not n then
return c
end
v = n.id == glyph_code and n.subtype<256 and n.font == font
if not v then
return c
end
char = n.char
end
if char == c_halant then
c = c.next
n = c.next
if not n then
return c
end
v = n.id == glyph_code and n.subtype<256 and n.font == font
if not v then
return c
end
char = n.char
end
if vowel_modifier[char] then
c = c.next
n = c.next
if not n then
return c
end
v = n.id == glyph_code and n.subtype<256 and n.font == font
if not v then
return c
end
char = n.char
end
if stress_tone_mark[char] then
c = c.next
n = c.next
if not n then
return c
end
v = n.id == glyph_code and n.subtype<256 and n.font == font
if not v then
return c
end
char = n.char
end
if stress_tone_mark[char] then
return n
else
return c
end
end
local function analyze_next_chars_two(c,font)
local n = c.next
if not n then
return c
end
if n.id == glyph_code and n.char == c_nukta and n.subtype<256 and n.font == font then
c = n
end
n = c
while true do
local nn = n.next
if nn and nn.id == glyph_code and nn.subtype<256 and nn.font == font then
local char = nn.char
if char == c_halant then
n = nn
local nnn = nn.next
if nnn and nnn.id == glyph_code and zw_char[nnn.char] and nnn.subtype<256 and nnn.font == font then
n = nnn
end
elseif char == c_zwnj or char == c_zwj then
-- n = nn -- not here (?)
local nnn = nn.next
if nnn and nnn.id == glyph_code and nnn.char == c_halant and nnn.subtype<256 and nnn.font == font then
n = nnn
end
else
break
end
local nn = n.next
if nn and nn.id == glyph_code and consonant[nn.char] and nn.subtype<256 and nn.font == font then
n = nn
local nnn = nn.next
if nnn and nnn.id == glyph_code and nnn.char == c_nukta and nnn.subtype<256 and nnn.font == font then
n = nnn
end
c = n
else
break
end
else
break
end
end
--
if not c then
-- This shouldn't happen I guess.
return
end
local n = c.next
if not n then
return c
end
local v = n.id == glyph_code and n.subtype<256 and n.font == font
if not v then
return c
end
local char = n.char
if char == c_anudatta then
c = n
n = c.next
if not n then
return c
end
v = n.id == glyph_code and n.subtype<256 and n.font == font
if not v then
return c
end
char = n.char
end
if char == c_halant then
c = c.next
n = c.next
if not n then
return c
end
v = n.id == glyph_code and n.subtype<256 and n.font == font
if not v then
return c
end
char = n.char
if char == c_zwnj or char == c_zwj then
c = c.next
n = c.next
if not n then
return c
end
v = n.id == glyph_code and n.subtype<256 and n.font == font
if not v then
return c
end
char = n.char
end
else
-- c = ms_matra(c)
-- same as one
if dependent_vowel[char] then
c = c.next
n = c.next
if not n then
return c
end
v = n.id == glyph_code and n.subtype<256 and n.font == font
if not v then
return c
end
char = n.char
end
if char == c_nukta then
c = c.next
n = c.next
if not n then
return c
end
v = n.id == glyph_code and n.subtype<256 and n.font == font
if not v then
return c
end
char = n.char
end
if char == c_halant then
c = c.next
n = c.next
if not n then
return c
end
v = n.id == glyph_code and n.subtype<256 and n.font == font
if not v then
return c
end
char = n.char
end
end
-- same as one
if vowel_modifier[char] then
c = c.next
n = c.next
if not n then
return c
end
v = n.id == glyph_code and n.subtype<256 and n.font == font
if not v then
return c
end
char = n.char
end
if stress_tone_mark[char] then
c = c.next
n = c.next
if not n then
return c
end
v = n.id == glyph_code and n.subtype<256 and n.font == font
if not v then
return c
end
char = n.char
end
if stress_tone_mark[char] then
return n
else
return c
end
end
local function inject_syntax_error(head,current,mark)
local signal = copy_node(current)
if mark == pre_mark then
signal.char = dotted_circle
else
current.char = dotted_circle
end
return insert_node_after(head,current,signal)
end
-- It looks like these two analyzers were written independently but they share
-- a lot. Common code has been synced.
function methods.deva(head,font,attr)
local current, start, done = head, true, false
while current do
if current.id == glyph_code and current.subtype<256 and current.font == font then
done = true
local syllablestart = current
local syllableend = nil
local c = current
local n = c.next
if n and c.char == c_ra and n.id == glyph_code and n.char == c_halant and n.subtype<256 and n.font == font then
local n = n.next
if n and n.id == glyph_code and n.subtype<256 and n.font == font then
c = n
end
end
local standalone = c.char == c_nbsp
if standalone then
local prev = current.prev
if not prev then
-- begin of paragraph or box
elseif prev.id ~= glyph_code or prev.subtype>=256 or prev.font ~= font then
-- different font or language so quite certainly a different word
elseif not separator[prev.char] then
-- something that separates words
else
standalone = false
end
end
if standalone then
-- stand alone cluster (at the start of the word only): #[Ra+H]+NBSP+[N]+[<[<ZWJ|ZWNJ>]+H+C>]+[{M}+[N]+[H]]+[SM]+[(VD)]
local syllabeend, current = analyze_next_chars_one(c,font,2) -- watch out, here we set current to next
if syllablestart ~= syllableend then
head, current = deva_reorder(head,syllablestart,syllableend,font,attr)
current = current.next
end
else
-- we can delay the n.subtype and n.font and test for say halant first
-- as an table access is faster than two function calls (subtype and font are
-- pseudo fields) but the code becomes messy (unless we make it a function)
local char = current.char
if consonant[char] then
-- syllable containing consonant
local prevc = true
while prevc do
prevc = false
local n = current.next
if not n then
break
end
local v = n.id == glyph_code and n.subtype<256 and n.font == font
if not v then
break
end
local c = n.char
if c == c_nukta then
n = n.next
if not n then
break
end
v = n.id == glyph_code and n.subtype<256 and n.font == font
if not v then
break
end
c = n.char
end
if c == c_halant then
n = n.next
if not n then
break
end
v = n.id == glyph_code and n.subtype<256 and n.font == font
if not v then
break
end
c = n.char
if c == c_zwnj or c == c_zwj then
n = n.next
if not n then
break
end
v = n.id == glyph_code and n.subtype<256 and n.font == font
if not v then
break
end
c = n.char
end
if consonant[c] then
prevc = true
current = n
end
end
end
local n = current.next
if n and n.id == glyph_code and n.char == c_nukta and n.subtype<256 and n.font == font then
-- nukta (not specified in Microsft Devanagari OpenType specification)
current = n
n = current.next
end
syllableend = current
current = n
if current then
local v = current.id == glyph_code and current.subtype<256 and current.font == font
if v then
if current.char == c_halant then
-- syllable containing consonant without vowels: {C + [Nukta] + H} + C + H
local n = current.next
if n and n.id == glyph_code and zw_char[n.char] and n.subtype<256 and n.font == font then
-- code collapsed, probably needs checking with intention
syllableend = n
current = n.next
else
syllableend = current
current = n
end
else
-- syllable containing consonant with vowels: {C + [Nukta] + H} + C + [M] + [VM] + [SM]
local c = current.char
if dependent_vowel[c] then
syllableend = current
current = current.next
v = current and current.id == glyph_code and current.subtype<256 and current.font == font
if v then
c = current.char
end
end
if v and vowel_modifier[c] then
syllableend = current
current = current.next
v = current and current.id == glyph_code and current.subtype<256 and current.font == font
if v then
c = current.char
end
end
if v and stress_tone_mark[c] then
syllableend = current
current = current.next
end
end
end
end
if syllablestart ~= syllableend then
head, current = deva_reorder(head,syllablestart,syllableend,font,attr)
current = current.next
end
elseif independent_vowel[char] then
-- syllable without consonants: VO + [VM] + [SM]
syllableend = current
current = current.next
if current then
local v = current.id == glyph_code and current.subtype<256 and current.font == font
if v then
local c = current.char
if vowel_modifier[c] then
syllableend = current
current = current.next
v = current and current.id == glyph_code and current.subtype<256 and current.font == font
if v then
c = current.char
end
end
if v and stress_tone_mark[c] then
syllableend = current
current = current.next
end
end
end
else
local mark = mark_four[char]
if mark then
head, current = inject_syntax_error(head,current,mark)
end
current = current.next
end
end
else
current = current.next
end
start = false
end
return head, done
end
-- there is a good change that when we run into one with subtype < 256 that the rest is also done
-- so maybe we can omit this check (it's pretty hard to get glyphs in the stream out of the blue)
-- handler(head,start,kind,lookupname,lookupmatch,sequence,lookuphash,1)
function methods.dev2(head,font,attr)
local current = head
local start = true
local done = false
local syllabe = 0
while current do
local syllablestart, syllableend = nil, nil
if current.id == glyph_code and current.subtype<256 and current.font == font then
done = true
syllablestart = current
local c = current
local n = current.next
if n and c.char == c_ra and n.id == glyph_code and n.char == c_halant and n.subtype<256 and n.font == font then
local n = n.next
if n and n.id == glyph_code and n.subtype<256 and n.font == font then
c = n
end
end
local char = c.char
if independent_vowel[char] then
-- vowel-based syllable: [Ra+H]+V+[N]+[<[<ZWJ|ZWNJ>]+H+C|ZWJ+C>]+[{M}+[N]+[H]]+[SM]+[(VD)]
current = analyze_next_chars_one(c,font,1)
syllableend = current
else
local standalone = char == c_nbsp
if standalone then
local p = current.prev
if not p then
-- begin of paragraph or box
elseif p.id ~= glyph_code or p.subtype>=256 or p.font ~= font then
-- different font or language so quite certainly a different word
elseif not separator[p.char] then
-- something that separates words
else
standalone = false
end
end
if standalone then
-- Stand Alone cluster (at the start of the word only): #[Ra+H]+NBSP+[N]+[<[<ZWJ|ZWNJ>]+H+C>]+[{M}+[N]+[H]]+[SM]+[(VD)]
current = analyze_next_chars_one(c,font,2)
syllableend = current
elseif consonant[current.char] then
-- WHY current INSTEAD OF c ?
-- Consonant syllable: {C+[N]+<H+[<ZWNJ|ZWJ>]|<ZWNJ|ZWJ>+H>} + C+[N]+[A] + [< H+[<ZWNJ|ZWJ>] | {M}+[N]+[H]>]+[SM]+[(VD)]
current = analyze_next_chars_two(current,font) -- not c !
syllableend = current
end
end
end
if syllableend then
syllabe = syllabe + 1
local c = syllablestart
local n = syllableend.next
while c ~= n do
c[a_syllabe] = syllabe
c = c.next
end
end
if syllableend and syllablestart ~= syllableend then
head, current = dev2_reorder(head,syllablestart,syllableend,font,attr)
end
if not syllableend and current.id == glyph_code and current.subtype<256 and current.font == font and not current[a_state] then
local mark = mark_four[current.char]
if mark then
head, current = inject_syntax_error(head,current,mark)
end
end
start = false
current = current.next
end
return head, done
end
|