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
|
/*
* vi:se ts=8:
*
* regcomp and regexec -- regsub and regerror are elsewhere
*
* Copyright (c) 1986 by University of Toronto.
* Written by Henry Spencer. Not derived from licensed software.
*
* Permission is granted to anyone to use this software for any
* purpose on any computer system, and to redistribute it freely,
* subject to the following restrictions:
*
* 1. The author is not responsible for the consequences of use of
* this software, no matter how awful, even if they arise
* from defects in it.
*
* 2. The origin of this software must not be misrepresented, either
* by explicit claim or by omission.
*
* 3. Altered versions must be plainly marked as such, and must not
* be misrepresented as being the original software.
*** THIS IS AN ALTERED VERSION. It was altered by John Gilmore,
*** hoptoad!gnu, on 27 Dec 1986, to add \n as an alternative to |
*** to assist in implementing egrep.
*** THIS IS AN ALTERED VERSION. It was altered by John Gilmore,
*** hoptoad!gnu, on 27 Dec 1986, to add \< and \> for word-matching
*** as in BSD grep and ex.
*** THIS IS AN ALTERED VERSION. It was altered by John Gilmore,
*** hoptoad!gnu, on 28 Dec 1986, to optimize characters quoted with \.
*** THIS IS AN ALTERED VERSION. It was altered by James A. Woods,
*** ames!jaw, on 19 June 1987, to quash a regcomp() redundancy.
*** THIS IS AN ALTERED VERSION. It was altered by Christopher Seiwald
*** seiwald@vix.com, on 28 August 1993, for use in jam. Regmagic.h
*** was moved into regexp.h, and the include of regexp.h now uses "'s
*** to avoid conflicting with the system regexp.h. Const, bless its
*** soul, was removed so it can compile everywhere. The declaration
*** of strchr() was in conflict on AIX, so it was removed (as it is
*** happily defined in string.h).
*** THIS IS AN ALTERED VERSION. It was altered by Christopher Seiwald
*** seiwald@perforce.com, on 20 January 2000, to use function prototypes.
*** THIS IS AN ALTERED VERSION. It was altered by Christopher Seiwald
*** seiwald@perforce.com, on 05 November 2002, to const string literals.
*
* THIS IS AN ALTERED VERSION. It was altered by Steve Bennett <steveb@workware.net.au>
* on 16 October 2010, to remove static state and add better Tcl ARE compatibility.
* This includes counted repetitions, UTF-8 support, character classes,
* shorthand character classes, increased number of parentheses to 100,
* backslash escape sequences. It also removes \n as an alternative to |.
*
* Beware that some of this code is subtly aware of the way operator
* precedence is structured in regular expressions. Serious changes in
* regular-expression syntax might require a total rethink.
*/
#include "jimautoconf.h"
#if defined(JIM_REGEXP)
#include <stdio.h>
#include <ctype.h>
#include <stdlib.h>
#include <string.h>
#include "jim.h"
#include "jimregexp.h"
#include "utf8.h"
/* An arbitrary limit, but this seems enough. Must be less than 1000. */
#define REG_MAX_PAREN 100
/*
* Structure for regexp "program". This is essentially a linear encoding
* of a nondeterministic finite-state machine (aka syntax charts or
* "railroad normal form" in parsing technology). Each node is an opcode
* plus a "next" pointer, possibly plus an operand. "Next" pointers of
* all nodes except BRANCH implement concatenation; a "next" pointer with
* a BRANCH on both ends of it is connecting two alternatives. (Here we
* have one of the subtle syntax dependencies: an individual BRANCH (as
* opposed to a collection of them) is never concatenated with anything
* because of operator precedence.) The operand of some types of node is
* a literal string; for others, it is a node leading into a sub-FSM. In
* particular, the operand of a BRANCH node is the first node of the branch.
* (NB this is *not* a tree structure: the tail of the branch connects
* to the thing following the set of BRANCHes.) The opcodes are:
*/
/* definition number opnd? meaning */
#define END 0 /* no End of program. */
#define BOL 1 /* no Match "" at beginning of line. */
#define EOL 2 /* no Match "" at end of line. */
#define ANY 3 /* no Match any one character. */
#define ANYOF 4 /* str Match any character in this string. */
#define ANYBUT 5 /* str Match any character not in this string. */
#define BRANCH 6 /* node Match this alternative, or the next... */
#define BACK 7 /* no Match "", "next" ptr points backward. */
#define EXACTLY 8 /* str Match this string. */
#define NOTHING 9 /* no Match empty string. */
#define REP 10 /* max,min Match this (simple) thing [min,max] times. */
#define REPMIN 11 /* max,min Match this (simple) thing [min,max] times, minimal match. */
#define REPX 12 /* max,min Match this (complex) thing [min,max] times. */
#define REPXMIN 13 /* max,min Match this (complex) thing [min,max] times, minimal match. */
#define BOLX 14 /* no Match "" at beginning of input. */
#define EOLX 15 /* no Match "" at end of input. */
#define WORDA 16 /* no Match "" at wordchar, where prev is nonword */
#define WORDZ 17 /* no Match "" at nonwordchar, where prev is word */
#define OPENNC 1000 /* no Non-capturing parentheses - must be OPEN-1 */
#define OPEN 1001 /* no Mark this point in input as start of #n. */
/* OPEN+1 is number 1, etc. */
/* must not be any other opts between OPEN and CLOSE */
#define CLOSENC 2000 /* no Non-capturing parentheses - must be CLOSE-1 */
#define CLOSE 2001 /* no Analogous to OPEN. */
#define CLOSE_END (CLOSE+REG_MAX_PAREN)
/*
* The first word of the regexp internal "program" is actually this magic
* number; the start node begins in the second word.
*/
#define REG_MAGIC 0xFADED00D
/*
* Opcode notes:
*
* BRANCH The set of branches constituting a single choice are hooked
* together with their "next" pointers, since precedence prevents
* anything being concatenated to any individual branch. The
* "next" pointer of the last BRANCH in a choice points to the
* thing following the whole choice. This is also where the
* final "next" pointer of each individual branch points; each
* branch starts with the operand node of a BRANCH node.
*
* BACK Normal "next" pointers all implicitly point forward; BACK
* exists to make loop structures possible.
*
* REP,REPX Repeated matches ('?', '*', '+' and {min,max}) are implemented
* as either simple repeats (REP) or complex repeats (REPX).
* These opcodes include a "min" and "max" count after the opcode.
* This is followed by a fourth "current count" word that is
* only used by REPX, as it implements a recursive match.
* REPMIN and REPXMIN are identical except they implement minimal repeats.
*
* OPEN,CLOSE ...are numbered at compile time.
*/
/*
* A node is one word of opcode followed by one word of "next" pointer.
* The "next" pointer value is a positive offset from the opcode of the node
* containing it.
* An operand, if any, simply follows the node. (Note that much of the
* code generation knows about this implicit relationship.)
*/
#define OP(preg, p) (preg->program[p])
#define NEXT(preg, p) (preg->program[p + 1])
#define OPERAND(p) ((p) + 2)
/*
* See regmagic.h for one further detail of program structure.
*/
/*
* Utility definitions.
*/
#define FAIL(R,M) { (R)->err = (M); return (M); }
#define ISMULT(c) ((c) == '*' || (c) == '+' || (c) == '?' || (c) == '{')
#define META "^$.[()|?{+*"
/*
* Flags to be passed up and down.
*/
#define HASWIDTH 1 /* Known never to match null string. */
#define SIMPLE 2 /* Simple enough to be STAR/PLUS operand. */
#define SPSTART 4 /* Starts with * or +. */
#define WORST 0 /* Worst case. */
#define MAX_REP_COUNT 1000000
/*
* Forward declarations for regcomp()'s friends.
*/
static int reg(regex_t *preg, int paren /* Parenthesized? */, int *flagp );
static int regpiece(regex_t *preg, int *flagp );
static int regbranch(regex_t *preg, int *flagp );
static int regatom(regex_t *preg, int *flagp );
static int regnode(regex_t *preg, int op );
static int regnext(regex_t *preg, int p );
static void regc(regex_t *preg, int b );
static int reginsert(regex_t *preg, int op, int size, int opnd );
static void regtail(regex_t *preg, int p, int val);
static void regoptail(regex_t *preg, int p, int val );
static int regopsize(regex_t *preg, int p );
static int reg_range_find(const int *string, int c);
static const char *str_find(const char *string, int c, int nocase);
static int prefix_cmp(const int *prog, int proglen, const char *string, int nocase);
/*#define DEBUG*/
#ifdef DEBUG
static int regnarrate = 0;
static void regdump(regex_t *preg);
static const char *regprop( int op );
#endif
/**
* Returns the length of the null-terminated integer sequence.
*/
static int str_int_len(const int *seq)
{
int n = 0;
while (*seq++) {
n++;
}
return n;
}
/*
- regcomp - compile a regular expression into internal code
*
* We can't allocate space until we know how big the compiled form will be,
* but we can't compile it (and thus know how big it is) until we've got a
* place to put the code. So we cheat: we compile it twice, once with code
* generation turned off and size counting turned on, and once "for real".
* This also means that we don't allocate space until we are sure that the
* thing really will compile successfully, and we never have to move the
* code and thus invalidate pointers into it. (Note that it has to be in
* one piece because free() must be able to free it all.)
*
* Beware that the optimization-preparation code in here knows about some
* of the structure of the compiled regexp.
*/
int jim_regcomp(regex_t *preg, const char *exp, int cflags)
{
int scan;
int longest;
unsigned len;
int flags;
#ifdef DEBUG
fprintf(stderr, "Compiling: '%s'\n", exp);
#endif
memset(preg, 0, sizeof(*preg));
if (exp == NULL)
FAIL(preg, REG_ERR_NULL_ARGUMENT);
/* First pass: determine size, legality. */
preg->cflags = cflags;
preg->regparse = exp;
/* Allocate space. */
preg->proglen = (strlen(exp) + 1) * 5;
preg->program = malloc(preg->proglen * sizeof(int));
if (preg->program == NULL)
FAIL(preg, REG_ERR_NOMEM);
/* Note that since we store a magic value as the first item in the program,
* program offsets will never be 0
*/
regc(preg, REG_MAGIC);
if (reg(preg, 0, &flags) == 0) {
return preg->err;
}
/* Small enough for pointer-storage convention? */
if (preg->re_nsub >= REG_MAX_PAREN) /* Probably could be 65535L. */
FAIL(preg,REG_ERR_TOO_BIG);
/* Dig out information for optimizations. */
preg->regstart = 0; /* Worst-case defaults. */
preg->reganch = 0;
preg->regmust = 0;
preg->regmlen = 0;
scan = 1; /* First BRANCH. */
if (OP(preg, regnext(preg, scan)) == END) { /* Only one top-level choice. */
scan = OPERAND(scan);
/* Starting-point info. */
if (OP(preg, scan) == EXACTLY) {
preg->regstart = preg->program[OPERAND(scan)];
}
else if (OP(preg, scan) == BOL)
preg->reganch++;
/*
* If there's something expensive in the r.e., find the
* longest literal string that must appear and make it the
* regmust. Resolve ties in favor of later strings, since
* the regstart check works with the beginning of the r.e.
* and avoiding duplication strengthens checking. Not a
* strong reason, but sufficient in the absence of others.
*/
if (flags&SPSTART) {
longest = 0;
len = 0;
for (; scan != 0; scan = regnext(preg, scan)) {
if (OP(preg, scan) == EXACTLY) {
int plen = str_int_len(preg->program + OPERAND(scan));
if (plen >= len) {
longest = OPERAND(scan);
len = plen;
}
}
}
preg->regmust = longest;
preg->regmlen = len;
}
}
#ifdef DEBUG
regdump(preg);
#endif
return 0;
}
/*
- reg - regular expression, i.e. main body or parenthesized thing
*
* Caller must absorb opening parenthesis.
*
* Combining parenthesis handling with the base level of regular expression
* is a trifle forced, but the need to tie the tails of the branches to what
* follows makes it hard to avoid.
*/
static int reg(regex_t *preg, int paren /* Parenthesized? */, int *flagp )
{
int ret;
int br;
int ender;
int parno = 0;
int flags;
*flagp = HASWIDTH; /* Tentatively. */
/* Make an OPEN node, if parenthesized. */
if (paren) {
if (preg->regparse[0] == '?' && preg->regparse[1] == ':') {
/* non-capturing paren */
preg->regparse += 2;
parno = -1;
}
else {
parno = ++preg->re_nsub;
}
ret = regnode(preg, OPEN+parno);
} else
ret = 0;
/* Pick up the branches, linking them together. */
br = regbranch(preg, &flags);
if (br == 0)
return 0;
if (ret != 0)
regtail(preg, ret, br); /* OPEN -> first. */
else
ret = br;
if (!(flags&HASWIDTH))
*flagp &= ~HASWIDTH;
*flagp |= flags&SPSTART;
while (*preg->regparse == '|') {
preg->regparse++;
br = regbranch(preg, &flags);
if (br == 0)
return 0;
regtail(preg, ret, br); /* BRANCH -> BRANCH. */
if (!(flags&HASWIDTH))
*flagp &= ~HASWIDTH;
*flagp |= flags&SPSTART;
}
/* Make a closing node, and hook it on the end. */
ender = regnode(preg, (paren) ? CLOSE+parno : END);
regtail(preg, ret, ender);
/* Hook the tails of the branches to the closing node. */
for (br = ret; br != 0; br = regnext(preg, br))
regoptail(preg, br, ender);
/* Check for proper termination. */
if (paren && *preg->regparse++ != ')') {
preg->err = REG_ERR_UNMATCHED_PAREN;
return 0;
} else if (!paren && *preg->regparse != '\0') {
if (*preg->regparse == ')') {
preg->err = REG_ERR_UNMATCHED_PAREN;
return 0;
} else {
preg->err = REG_ERR_JUNK_ON_END;
return 0;
}
}
return(ret);
}
/*
- regbranch - one alternative of an | operator
*
* Implements the concatenation operator.
*/
static int regbranch(regex_t *preg, int *flagp )
{
int ret;
int chain;
int latest;
int flags;
*flagp = WORST; /* Tentatively. */
ret = regnode(preg, BRANCH);
chain = 0;
while (*preg->regparse != '\0' && *preg->regparse != ')' &&
*preg->regparse != '|') {
latest = regpiece(preg, &flags);
if (latest == 0)
return 0;
*flagp |= flags&HASWIDTH;
if (chain == 0) {/* First piece. */
*flagp |= flags&SPSTART;
}
else {
regtail(preg, chain, latest);
}
chain = latest;
}
if (chain == 0) /* Loop ran zero times. */
(void) regnode(preg, NOTHING);
return(ret);
}
/*
- regpiece - something followed by possible [*+?]
*
* Note that the branching code sequences used for ? and the general cases
* of * and + are somewhat optimized: they use the same NOTHING node as
* both the endmarker for their branch list and the body of the last branch.
* It might seem that this node could be dispensed with entirely, but the
* endmarker role is not redundant.
*/
static int regpiece(regex_t *preg, int *flagp)
{
int ret;
char op;
int next;
int flags;
int min;
int max;
ret = regatom(preg, &flags);
if (ret == 0)
return 0;
op = *preg->regparse;
if (!ISMULT(op)) {
*flagp = flags;
return(ret);
}
if (!(flags&HASWIDTH) && op != '?') {
preg->err = REG_ERR_OPERAND_COULD_BE_EMPTY;
return 0;
}
/* Handle braces (counted repetition) by expansion */
if (op == '{') {
char *end;
min = strtoul(preg->regparse + 1, &end, 10);
if (end == preg->regparse + 1) {
preg->err = REG_ERR_BAD_COUNT;
return 0;
}
if (*end == '}') {
max = min;
}
else if (*end == '\0') {
preg->err = REG_ERR_UNMATCHED_BRACES;
return 0;
}
else {
preg->regparse = end;
max = strtoul(preg->regparse + 1, &end, 10);
if (*end != '}') {
preg->err = REG_ERR_UNMATCHED_BRACES;
return 0;
}
}
if (end == preg->regparse + 1) {
max = MAX_REP_COUNT;
}
else if (max < min || max >= 100) {
preg->err = REG_ERR_BAD_COUNT;
return 0;
}
if (min >= 100) {
preg->err = REG_ERR_BAD_COUNT;
return 0;
}
preg->regparse = strchr(preg->regparse, '}');
}
else {
min = (op == '+');
max = (op == '?' ? 1 : MAX_REP_COUNT);
}
if (preg->regparse[1] == '?') {
preg->regparse++;
next = reginsert(preg, flags & SIMPLE ? REPMIN : REPXMIN, 5, ret);
}
else {
next = reginsert(preg, flags & SIMPLE ? REP: REPX, 5, ret);
}
preg->program[ret + 2] = max;
preg->program[ret + 3] = min;
preg->program[ret + 4] = 0;
*flagp = (min) ? (WORST|HASWIDTH) : (WORST|SPSTART);
if (!(flags & SIMPLE)) {
int back = regnode(preg, BACK);
regtail(preg, back, ret);
regtail(preg, next, back);
}
preg->regparse++;
if (ISMULT(*preg->regparse)) {
preg->err = REG_ERR_NESTED_COUNT;
return 0;
}
return ret;
}
/**
* Add all characters in the inclusive range between lower and upper.
*
* Handles a swapped range (upper < lower).
*/
static void reg_addrange(regex_t *preg, int lower, int upper)
{
if (lower > upper) {
reg_addrange(preg, upper, lower);
}
/* Add a range as length, start */
regc(preg, upper - lower + 1);
regc(preg, lower);
}
/**
* Add a null-terminated literal string as a set of ranges.
*/
static void reg_addrange_str(regex_t *preg, const char *str)
{
while (*str) {
reg_addrange(preg, *str, *str);
str++;
}
}
/**
* Extracts the next unicode char from utf8.
*
* If 'upper' is set, converts the char to uppercase.
*/
static int reg_utf8_tounicode_case(const char *s, int *uc, int upper)
{
int l = utf8_tounicode(s, uc);
if (upper) {
*uc = utf8_upper(*uc);
}
return l;
}
/**
* Converts a hex digit to decimal.
*
* Returns -1 for an invalid hex digit.
*/
static int hexdigitval(int c)
{
if (c >= '0' && c <= '9')
return c - '0';
if (c >= 'a' && c <= 'f')
return c - 'a' + 10;
if (c >= 'A' && c <= 'F')
return c - 'A' + 10;
return -1;
}
/**
* Parses up to 'n' hex digits at 's' and stores the result in *uc.
*
* Returns the number of hex digits parsed.
* If there are no hex digits, returns 0 and stores nothing.
*/
static int parse_hex(const char *s, int n, int *uc)
{
int val = 0;
int k;
for (k = 0; k < n; k++) {
int c = hexdigitval(*s++);
if (c == -1) {
break;
}
val = (val << 4) | c;
}
if (k) {
*uc = val;
}
return k;
}
/**
* Call for chars after a backlash to decode the escape sequence.
*
* Stores the result in *ch.
*
* Returns the number of bytes consumed.
*/
static int reg_decode_escape(const char *s, int *ch)
{
int n;
const char *s0 = s;
*ch = *s++;
switch (*ch) {
case 'b': *ch = '\b'; break;
case 'e': *ch = 27; break;
case 'f': *ch = '\f'; break;
case 'n': *ch = '\n'; break;
case 'r': *ch = '\r'; break;
case 't': *ch = '\t'; break;
case 'v': *ch = '\v'; break;
case 'u':
if (*s == '{') {
/* Expect \u{NNNN} */
n = parse_hex(s + 1, 6, ch);
if (n > 0 && s[n + 1] == '}' && *ch >= 0 && *ch <= 0x1fffff) {
s += n + 2;
}
else {
/* Invalid, so just treat as an escaped 'u' */
*ch = 'u';
}
}
else if ((n = parse_hex(s, 4, ch)) > 0) {
s += n;
}
break;
case 'U':
if ((n = parse_hex(s, 8, ch)) > 0) {
s += n;
}
break;
case 'x':
if ((n = parse_hex(s, 2, ch)) > 0) {
s += n;
}
break;
case '\0':
s--;
*ch = '\\';
break;
}
return s - s0;
}
/*
- regatom - the lowest level
*
* Optimization: gobbles an entire sequence of ordinary characters so that
* it can turn them into a single node, which is smaller to store and
* faster to run. Backslashed characters are exceptions, each becoming a
* separate node; the code is simpler that way and it's not worth fixing.
*/
static int regatom(regex_t *preg, int *flagp)
{
int ret;
int flags;
int nocase = (preg->cflags & REG_ICASE);
int ch;
int n = reg_utf8_tounicode_case(preg->regparse, &ch, nocase);
*flagp = WORST; /* Tentatively. */
preg->regparse += n;
switch (ch) {
/* FIXME: these chars only have meaning at beg/end of pat? */
case '^':
ret = regnode(preg, BOL);
break;
case '$':
ret = regnode(preg, EOL);
break;
case '.':
ret = regnode(preg, ANY);
*flagp |= HASWIDTH|SIMPLE;
break;
case '[': {
const char *pattern = preg->regparse;
if (*pattern == '^') { /* Complement of range. */
ret = regnode(preg, ANYBUT);
pattern++;
} else
ret = regnode(preg, ANYOF);
/* Special case. If the first char is ']' or '-', it is part of the set */
if (*pattern == ']' || *pattern == '-') {
reg_addrange(preg, *pattern, *pattern);
pattern++;
}
while (*pattern != ']') {
/* Is this a range? a-z */
int start;
int end;
enum {
CC_ALPHA, CC_ALNUM, CC_SPACE, CC_BLANK, CC_UPPER, CC_LOWER,
CC_DIGIT, CC_XDIGIT, CC_CNTRL, CC_GRAPH, CC_PRINT, CC_PUNCT,
CC_NUM
};
int cc;
if (!*pattern) {
preg->err = REG_ERR_UNMATCHED_BRACKET;
return 0;
}
pattern += reg_utf8_tounicode_case(pattern, &start, nocase);
if (start == '\\') {
/* First check for class shorthand escapes */
switch (*pattern) {
case 's':
pattern++;
cc = CC_SPACE;
goto cc_switch;
case 'd':
pattern++;
cc = CC_DIGIT;
goto cc_switch;
case 'w':
pattern++;
reg_addrange(preg, '_', '_');
cc = CC_ALNUM;
goto cc_switch;
}
pattern += reg_decode_escape(pattern, &start);
if (start == 0) {
preg->err = REG_ERR_NULL_CHAR;
return 0;
}
if (start == '\\' && *pattern == 0) {
preg->err = REG_ERR_INVALID_ESCAPE;
return 0;
}
}
if (pattern[0] == '-' && pattern[1] && pattern[1] != ']') {
/* skip '-' */
pattern += utf8_tounicode(pattern, &end);
pattern += reg_utf8_tounicode_case(pattern, &end, nocase);
if (end == '\\') {
pattern += reg_decode_escape(pattern, &end);
if (end == 0) {
preg->err = REG_ERR_NULL_CHAR;
return 0;
}
if (start == '\\' && *pattern == 0) {
preg->err = REG_ERR_INVALID_ESCAPE;
return 0;
}
}
reg_addrange(preg, start, end);
continue;
}
if (start == '[' && pattern[0] == ':') {
static const char *character_class[] = {
":alpha:", ":alnum:", ":space:", ":blank:", ":upper:", ":lower:",
":digit:", ":xdigit:", ":cntrl:", ":graph:", ":print:", ":punct:",
};
for (cc = 0; cc < CC_NUM; cc++) {
n = strlen(character_class[cc]);
if (strncmp(pattern, character_class[cc], n) == 0) {
if (pattern[n] != ']') {
preg->err = REG_ERR_UNMATCHED_BRACKET;
return 0;
}
/* Found a character class */
pattern += n + 1;
break;
}
}
if (cc != CC_NUM) {
cc_switch:
switch (cc) {
case CC_ALNUM:
reg_addrange(preg, '0', '9');
/* Fall through */
case CC_ALPHA:
if ((preg->cflags & REG_ICASE) == 0) {
reg_addrange(preg, 'a', 'z');
}
reg_addrange(preg, 'A', 'Z');
break;
case CC_SPACE:
reg_addrange_str(preg, " \t\r\n\f\v");
break;
case CC_BLANK:
reg_addrange_str(preg, " \t");
break;
case CC_UPPER:
reg_addrange(preg, 'A', 'Z');
break;
case CC_LOWER:
reg_addrange(preg, 'a', 'z');
break;
case CC_XDIGIT:
reg_addrange(preg, 'a', 'f');
reg_addrange(preg, 'A', 'F');
/* Fall through */
case CC_DIGIT:
reg_addrange(preg, '0', '9');
break;
case CC_CNTRL:
reg_addrange(preg, 0, 31);
reg_addrange(preg, 127, 127);
break;
case CC_PRINT:
reg_addrange(preg, ' ', '~');
break;
case CC_GRAPH:
reg_addrange(preg, '!', '~');
break;
case CC_PUNCT:
reg_addrange(preg, '!', '/');
reg_addrange(preg, ':', '@');
reg_addrange(preg, '[', '`');
reg_addrange(preg, '{', '~');
break;
}
continue;
}
}
/* Not a range, so just add the char */
reg_addrange(preg, start, start);
}
regc(preg, '\0');
if (*pattern) {
pattern++;
}
preg->regparse = pattern;
*flagp |= HASWIDTH|SIMPLE;
}
break;
case '(':
ret = reg(preg, 1, &flags);
if (ret == 0)
return 0;
*flagp |= flags&(HASWIDTH|SPSTART);
break;
case '\0':
case '|':
case ')':
preg->err = REG_ERR_INTERNAL;
return 0; /* Supposed to be caught earlier. */
case '?':
case '+':
case '*':
case '{':
preg->err = REG_ERR_COUNT_FOLLOWS_NOTHING;
return 0;
case '\\':
ch = *preg->regparse++;
switch (ch) {
case '\0':
preg->err = REG_ERR_INVALID_ESCAPE;
return 0;
case 'A':
ret = regnode(preg, BOLX);
break;
case 'Z':
ret = regnode(preg, EOLX);
break;
case '<':
case 'm':
ret = regnode(preg, WORDA);
break;
case '>':
case 'M':
ret = regnode(preg, WORDZ);
break;
case 'd':
case 'D':
ret = regnode(preg, ch == 'd' ? ANYOF : ANYBUT);
reg_addrange(preg, '0', '9');
regc(preg, '\0');
*flagp |= HASWIDTH|SIMPLE;
break;
case 'w':
case 'W':
ret = regnode(preg, ch == 'w' ? ANYOF : ANYBUT);
if ((preg->cflags & REG_ICASE) == 0) {
reg_addrange(preg, 'a', 'z');
}
reg_addrange(preg, 'A', 'Z');
reg_addrange(preg, '0', '9');
reg_addrange(preg, '_', '_');
regc(preg, '\0');
*flagp |= HASWIDTH|SIMPLE;
break;
case 's':
case 'S':
ret = regnode(preg, ch == 's' ? ANYOF : ANYBUT);
reg_addrange_str(preg," \t\r\n\f\v");
regc(preg, '\0');
*flagp |= HASWIDTH|SIMPLE;
break;
/* FIXME: Someday handle \1, \2, ... */
default:
/* Handle general quoted chars in exact-match routine */
/* Back up to include the backslash */
preg->regparse--;
goto de_fault;
}
break;
de_fault:
default: {
/*
* Encode a string of characters to be matched exactly.
*/
int added = 0;
/* Back up to pick up the first char of interest */
preg->regparse -= n;
ret = regnode(preg, EXACTLY);
/* Note that a META operator such as ? or * consumes the
* preceding char.
* Thus we must be careful to look ahead by 2 and add the
* last char as it's own EXACTLY if necessary
*/
/* Until end of string or a META char is reached */
while (*preg->regparse && strchr(META, *preg->regparse) == NULL) {
n = reg_utf8_tounicode_case(preg->regparse, &ch, (preg->cflags & REG_ICASE));
if (ch == '\\' && preg->regparse[n]) {
/* Non-trailing backslash.
* Is this a special escape, or a regular escape?
*/
if (strchr("<>mMwWdDsSAZ", preg->regparse[n])) {
/* A special escape. All done with EXACTLY */
break;
}
/* Decode it. Note that we add the length for the escape
* sequence to the length for the backlash so we can skip
* the entire sequence, or not as required.
*/
n += reg_decode_escape(preg->regparse + n, &ch);
if (ch == 0) {
preg->err = REG_ERR_NULL_CHAR;
return 0;
}
}
/* Now we have one char 'ch' of length 'n'.
* Check to see if the following char is a MULT
*/
if (ISMULT(preg->regparse[n])) {
/* Yes. But do we already have some EXACTLY chars? */
if (added) {
/* Yes, so return what we have and pick up the current char next time around */
break;
}
/* No, so add this single char and finish */
regc(preg, ch);
added++;
preg->regparse += n;
break;
}
/* No, so just add this char normally */
regc(preg, ch);
added++;
preg->regparse += n;
}
regc(preg, '\0');
*flagp |= HASWIDTH;
if (added == 1)
*flagp |= SIMPLE;
break;
}
break;
}
return(ret);
}
static void reg_grow(regex_t *preg, int n)
{
if (preg->p + n >= preg->proglen) {
preg->proglen = (preg->p + n) * 2;
preg->program = realloc(preg->program, preg->proglen * sizeof(int));
}
}
/*
- regnode - emit a node
*/
/* Location. */
static int regnode(regex_t *preg, int op)
{
reg_grow(preg, 2);
/* The OP followed by a next pointer */
preg->program[preg->p++] = op;
preg->program[preg->p++] = 0;
/* Return the start of the node */
return preg->p - 2;
}
/*
- regc - emit (if appropriate) a byte of code
*/
static void regc(regex_t *preg, int b )
{
reg_grow(preg, 1);
preg->program[preg->p++] = b;
}
/*
- reginsert - insert an operator in front of already-emitted operand
*
* Means relocating the operand.
* Returns the new location of the original operand.
*/
static int reginsert(regex_t *preg, int op, int size, int opnd )
{
reg_grow(preg, size);
/* Move everything from opnd up */
memmove(preg->program + opnd + size, preg->program + opnd, sizeof(int) * (preg->p - opnd));
/* Zero out the new space */
memset(preg->program + opnd, 0, sizeof(int) * size);
preg->program[opnd] = op;
preg->p += size;
return opnd + size;
}
/*
- regtail - set the next-pointer at the end of a node chain
*/
static void regtail(regex_t *preg, int p, int val)
{
int scan;
int temp;
int offset;
/* Find last node. */
scan = p;
for (;;) {
temp = regnext(preg, scan);
if (temp == 0)
break;
scan = temp;
}
if (OP(preg, scan) == BACK)
offset = scan - val;
else
offset = val - scan;
preg->program[scan + 1] = offset;
}
/*
- regoptail - regtail on operand of first argument; nop if operandless
*/
static void regoptail(regex_t *preg, int p, int val )
{
/* "Operandless" and "op != BRANCH" are synonymous in practice. */
if (p != 0 && OP(preg, p) == BRANCH) {
regtail(preg, OPERAND(p), val);
}
}
/*
* regexec and friends
*/
/*
* Forwards.
*/
static int regtry(regex_t *preg, const char *string );
static int regmatch(regex_t *preg, int prog);
static int regrepeat(regex_t *preg, int p, int max);
/*
- regexec - match a regexp against a string
*/
int jim_regexec(regex_t *preg, const char *string, size_t nmatch, regmatch_t pmatch[], int eflags)
{
const char *s;
int scan;
/* Be paranoid... */
if (preg == NULL || preg->program == NULL || string == NULL) {
return REG_ERR_NULL_ARGUMENT;
}
/* Check validity of program. */
if (*preg->program != REG_MAGIC) {
return REG_ERR_CORRUPTED;
}
#ifdef DEBUG
fprintf(stderr, "regexec: %s\n", string);
regdump(preg);
#endif
preg->eflags = eflags;
preg->pmatch = pmatch;
preg->nmatch = nmatch;
preg->start = string; /* All offsets are computed from here */
/* Must clear out the embedded repeat counts of REPX and REPXMIN opcodes */
for (scan = OPERAND(1); scan != 0; scan += regopsize(preg, scan)) {
int op = OP(preg, scan);
if (op == END)
break;
if (op == REPX || op == REPXMIN)
preg->program[scan + 4] = 0;
}
/* If there is a "must appear" string, look for it. */
if (preg->regmust != 0) {
s = string;
while ((s = str_find(s, preg->program[preg->regmust], preg->cflags & REG_ICASE)) != NULL) {
if (prefix_cmp(preg->program + preg->regmust, preg->regmlen, s, preg->cflags & REG_ICASE) >= 0) {
break;
}
s++;
}
if (s == NULL) /* Not present. */
return REG_NOMATCH;
}
/* Mark beginning of line for ^ . */
preg->regbol = string;
/* Simplest case: anchored match need be tried only once (maybe per line). */
if (preg->reganch) {
if (eflags & REG_NOTBOL) {
/* This is an anchored search, but not an BOL, so possibly skip to the next line */
goto nextline;
}
while (1) {
if (regtry(preg, string)) {
return REG_NOERROR;
}
if (*string) {
nextline:
if (preg->cflags & REG_NEWLINE) {
/* Try the next anchor? */
string = strchr(string, '\n');
if (string) {
preg->regbol = ++string;
continue;
}
}
}
return REG_NOMATCH;
}
}
/* Messy cases: unanchored match. */
s = string;
if (preg->regstart != '\0') {
/* We know what char it must start with. */
while ((s = str_find(s, preg->regstart, preg->cflags & REG_ICASE)) != NULL) {
if (regtry(preg, s))
return REG_NOERROR;
s++;
}
}
else
/* We don't -- general case. */
while (1) {
if (regtry(preg, s))
return REG_NOERROR;
if (*s == '\0') {
break;
}
else {
int c;
s += utf8_tounicode(s, &c);
}
}
/* Failure. */
return REG_NOMATCH;
}
/*
- regtry - try match at specific point
*/
/* 0 failure, 1 success */
static int regtry( regex_t *preg, const char *string )
{
int i;
preg->reginput = string;
for (i = 0; i < preg->nmatch; i++) {
preg->pmatch[i].rm_so = -1;
preg->pmatch[i].rm_eo = -1;
}
if (regmatch(preg, 1)) {
preg->pmatch[0].rm_so = string - preg->start;
preg->pmatch[0].rm_eo = preg->reginput - preg->start;
return(1);
} else
return(0);
}
/**
* Returns bytes matched if 'pattern' is a prefix of 'string'.
*
* If 'nocase' is non-zero, does a case-insensitive match.
*
* Returns -1 on not found.
*/
static int prefix_cmp(const int *prog, int proglen, const char *string, int nocase)
{
const char *s = string;
while (proglen && *s) {
int ch;
int n = reg_utf8_tounicode_case(s, &ch, nocase);
if (ch != *prog) {
return -1;
}
prog++;
s += n;
proglen--;
}
if (proglen == 0) {
return s - string;
}
return -1;
}
/**
* Searchs for 'c' in the range 'range'.
*
* Returns 1 if found, or 0 if not.
*/
static int reg_range_find(const int *range, int c)
{
while (*range) {
/*printf("Checking %d in range [%d,%d]\n", c, range[1], (range[0] + range[1] - 1));*/
if (c >= range[1] && c <= (range[0] + range[1] - 1)) {
return 1;
}
range += 2;
}
return 0;
}
/**
* Search for the character 'c' in the utf-8 string 'string'.
*
* If 'nocase' is set, the 'string' is assumed to be uppercase
* and 'c' is converted to uppercase before matching.
*
* Returns the byte position in the string where the 'c' was found, or
* NULL if not found.
*/
static const char *str_find(const char *string, int c, int nocase)
{
if (nocase) {
/* The "string" should already be converted to uppercase */
c = utf8_upper(c);
}
while (*string) {
int ch;
int n = reg_utf8_tounicode_case(string, &ch, nocase);
if (c == ch) {
return string;
}
string += n;
}
return NULL;
}
/**
* Returns true if 'ch' is an end-of-line char.
*
* In REG_NEWLINE mode, \n is considered EOL in
* addition to \0
*/
static int reg_iseol(regex_t *preg, int ch)
{
if (preg->cflags & REG_NEWLINE) {
return ch == '\0' || ch == '\n';
}
else {
return ch == '\0';
}
}
static int regmatchsimplerepeat(regex_t *preg, int scan, int matchmin)
{
int nextch = '\0';
const char *save;
int no;
int c;
int max = preg->program[scan + 2];
int min = preg->program[scan + 3];
int next = regnext(preg, scan);
/*
* Lookahead to avoid useless match attempts
* when we know what character comes next.
*/
if (OP(preg, next) == EXACTLY) {
nextch = preg->program[OPERAND(next)];
}
save = preg->reginput;
no = regrepeat(preg, scan + 5, max);
if (no < min) {
return 0;
}
if (matchmin) {
/* from min up to no */
max = no;
no = min;
}
/* else from no down to min */
while (1) {
if (matchmin) {
if (no > max) {
break;
}
}
else {
if (no < min) {
break;
}
}
preg->reginput = save + utf8_index(save, no);
reg_utf8_tounicode_case(preg->reginput, &c, (preg->cflags & REG_ICASE));
/* If it could work, try it. */
if (reg_iseol(preg, nextch) || c == nextch) {
if (regmatch(preg, next)) {
return(1);
}
}
if (matchmin) {
/* Couldn't or didn't, add one more */
no++;
}
else {
/* Couldn't or didn't -- back up. */
no--;
}
}
return(0);
}
static int regmatchrepeat(regex_t *preg, int scan, int matchmin)
{
int *scanpt = preg->program + scan;
int max = scanpt[2];
int min = scanpt[3];
/* Have we reached min? */
if (scanpt[4] < min) {
/* No, so get another one */
scanpt[4]++;
if (regmatch(preg, scan + 5)) {
return 1;
}
scanpt[4]--;
return 0;
}
if (scanpt[4] > max) {
return 0;
}
if (matchmin) {
/* minimal, so try other branch first */
if (regmatch(preg, regnext(preg, scan))) {
return 1;
}
/* No, so try one more */
scanpt[4]++;
if (regmatch(preg, scan + 5)) {
return 1;
}
scanpt[4]--;
return 0;
}
/* maximal, so try this branch again */
if (scanpt[4] < max) {
scanpt[4]++;
if (regmatch(preg, scan + 5)) {
return 1;
}
scanpt[4]--;
}
/* At this point we are at max with no match. Try the other branch */
return regmatch(preg, regnext(preg, scan));
}
/*
- regmatch - main matching routine
*
* Conceptually the strategy is simple: check to see whether the current
* node matches, call self recursively to see whether the rest matches,
* and then act accordingly. In practice we make some effort to avoid
* recursion, in particular by going through "ordinary" nodes (that don't
* need to know whether the rest of the match failed) by a loop instead of
* by recursion.
*/
/* 0 failure, 1 success */
static int regmatch(regex_t *preg, int prog)
{
int scan; /* Current node. */
int next; /* Next node. */
const char *save;
scan = prog;
#ifdef DEBUG
if (scan != 0 && regnarrate)
fprintf(stderr, "%s(\n", regprop(scan));
#endif
while (scan != 0) {
int n;
int c;
#ifdef DEBUG
if (regnarrate) {
fprintf(stderr, "%3d: %s...\n", scan, regprop(OP(preg, scan))); /* Where, what. */
}
#endif
next = regnext(preg, scan);
n = reg_utf8_tounicode_case(preg->reginput, &c, (preg->cflags & REG_ICASE));
switch (OP(preg, scan)) {
case BOLX:
if ((preg->eflags & REG_NOTBOL)) {
return(0);
}
/* Fall through */
case BOL:
if (preg->reginput != preg->regbol) {
return(0);
}
break;
case EOLX:
if (c != 0) {
/* For EOLX, only match real end of line, not newline */
return 0;
}
break;
case EOL:
if (!reg_iseol(preg, c)) {
return(0);
}
break;
case WORDA:
/* Must be looking at a letter, digit, or _ */
if ((!isalnum(UCHAR(c))) && c != '_')
return(0);
/* Prev must be BOL or nonword */
if (preg->reginput > preg->regbol &&
(isalnum(UCHAR(preg->reginput[-1])) || preg->reginput[-1] == '_'))
return(0);
break;
case WORDZ:
/* Can't match at BOL */
if (preg->reginput > preg->regbol) {
/* Current must be EOL or nonword */
if (reg_iseol(preg, c) || !(isalnum(UCHAR(c)) || c == '_')) {
c = preg->reginput[-1];
/* Previous must be word */
if (isalnum(UCHAR(c)) || c == '_') {
break;
}
}
}
/* No */
return(0);
case ANY:
if (reg_iseol(preg, c))
return 0;
preg->reginput += n;
break;
case EXACTLY: {
int opnd;
int len;
int slen;
opnd = OPERAND(scan);
len = str_int_len(preg->program + opnd);
slen = prefix_cmp(preg->program + opnd, len, preg->reginput, preg->cflags & REG_ICASE);
if (slen < 0) {
return(0);
}
preg->reginput += slen;
}
break;
case ANYOF:
if (reg_iseol(preg, c) || reg_range_find(preg->program + OPERAND(scan), c) == 0) {
return(0);
}
preg->reginput += n;
break;
case ANYBUT:
if (reg_iseol(preg, c) || reg_range_find(preg->program + OPERAND(scan), c) != 0) {
return(0);
}
preg->reginput += n;
break;
case NOTHING:
break;
case BACK:
break;
case BRANCH:
if (OP(preg, next) != BRANCH) /* No choice. */
next = OPERAND(scan); /* Avoid recursion. */
else {
do {
save = preg->reginput;
if (regmatch(preg, OPERAND(scan))) {
return(1);
}
preg->reginput = save;
scan = regnext(preg, scan);
} while (scan != 0 && OP(preg, scan) == BRANCH);
return(0);
/* NOTREACHED */
}
break;
case REP:
case REPMIN:
return regmatchsimplerepeat(preg, scan, OP(preg, scan) == REPMIN);
case REPX:
case REPXMIN:
return regmatchrepeat(preg, scan, OP(preg, scan) == REPXMIN);
case END:
return 1; /* Success! */
case OPENNC:
case CLOSENC:
return regmatch(preg, next);
default:
if (OP(preg, scan) >= OPEN+1 && OP(preg, scan) < CLOSE_END) {
save = preg->reginput;
if (regmatch(preg, next)) {
if (OP(preg, scan) < CLOSE) {
int no = OP(preg, scan) - OPEN;
if (no < preg->nmatch && preg->pmatch[no].rm_so == -1) {
preg->pmatch[no].rm_so = save - preg->start;
}
}
else {
int no = OP(preg, scan) - CLOSE;
if (no < preg->nmatch && preg->pmatch[no].rm_eo == -1) {
preg->pmatch[no].rm_eo = save - preg->start;
}
}
return(1);
}
/* Restore input position after failure */
preg->reginput = save;
return(0);
}
return REG_ERR_INTERNAL;
}
scan = next;
}
/*
* We get here only if there's trouble -- normally "case END" is
* the terminating point.
*/
return REG_ERR_INTERNAL;
}
/*
- regrepeat - repeatedly match something simple, report how many
*/
static int regrepeat(regex_t *preg, int p, int max)
{
int count = 0;
const char *scan;
int opnd;
int ch;
int n;
scan = preg->reginput;
opnd = OPERAND(p);
switch (OP(preg, p)) {
case ANY:
while (!reg_iseol(preg, *scan) && count < max) {
count++;
scan += utf8_charlen(*scan);
}
break;
case EXACTLY:
while (count < max) {
n = reg_utf8_tounicode_case(scan, &ch, preg->cflags & REG_ICASE);
if (preg->program[opnd] != ch) {
break;
}
count++;
scan += n;
}
break;
case ANYOF:
while (count < max) {
n = reg_utf8_tounicode_case(scan, &ch, preg->cflags & REG_ICASE);
if (reg_iseol(preg, ch) || reg_range_find(preg->program + opnd, ch) == 0) {
break;
}
count++;
scan += n;
}
break;
case ANYBUT:
while (count < max) {
n = reg_utf8_tounicode_case(scan, &ch, preg->cflags & REG_ICASE);
if (reg_iseol(preg, ch) || reg_range_find(preg->program + opnd, ch) != 0) {
break;
}
count++;
scan += n;
}
break;
default: /* Oh dear. Called inappropriately. */
preg->err = REG_ERR_INTERNAL;
count = 0; /* Best compromise. */
break;
}
preg->reginput = scan;
return(count);
}
/*
- regnext - dig the "next" pointer out of a node
*/
static int regnext(regex_t *preg, int p )
{
int offset;
offset = NEXT(preg, p);
if (offset == 0)
return 0;
if (OP(preg, p) == BACK)
return(p-offset);
else
return(p+offset);
}
/*
- regopsize - returns the size of opcode + operands at 'p' in words
*/
static int regopsize(regex_t *preg, int p )
{
/* Almost all opcodes are 2 words, but some are more */
switch (OP(preg, p)) {
case REP:
case REPMIN:
case REPX:
case REPXMIN:
return 5;
case ANYOF:
case ANYBUT:
case EXACTLY: {
int s = p + 2;
while (preg->program[s++]) {
}
return s - p;
}
}
return 2;
}
#if defined(DEBUG) && !defined(JIM_BOOTSTRAP)
/*
- regdump - dump a regexp onto stdout in vaguely comprehensible form
*/
static void regdump(regex_t *preg)
{
int s;
int op = EXACTLY; /* Arbitrary non-END op. */
int next;
char buf[MAX_UTF8_LEN + 1];
int i;
for (i = 1; i < preg->p; i++) {
printf("%02x ", (unsigned char)preg->program[i]);
if (i % 16 == 0) {
printf("\n");
}
}
printf("\n");
s = 1;
while (op != END && s < preg->p) { /* While that wasn't END last time... */
op = OP(preg, s);
printf("%3d: %s", s, regprop(op)); /* Where, what. */
next = regnext(preg, s);
if (next == 0) /* Next ptr. */
printf("(0)");
else
printf("(%d)", next);
s += 2;
if (op == REP || op == REPMIN || op == REPX || op == REPXMIN) {
int max = preg->program[s];
int min = preg->program[s + 1];
if (max == 65535) {
printf("{%d,*}", min);
}
else {
printf("{%d,%d}", min, max);
}
printf(" %d", preg->program[s + 2]);
s += 3;
}
else if (op == ANYOF || op == ANYBUT) {
/* set of ranges */
while (preg->program[s]) {
int len = preg->program[s++];
int first = preg->program[s++];
buf[utf8_getchars(buf, first)] = 0;
printf("%s", buf);
if (len > 1) {
buf[utf8_getchars(buf, first + len - 1)] = 0;
printf("-%s", buf);
}
}
s++;
}
else if (op == EXACTLY) {
/* Literal string, where present. */
while (preg->program[s]) {
buf[utf8_getchars(buf, preg->program[s])] = 0;
printf("%s", buf);
s++;
}
s++;
}
putchar('\n');
}
if (op == END) {
/* Header fields of interest. */
if (preg->regstart) {
buf[utf8_getchars(buf, preg->regstart)] = 0;
printf("start '%s' ", buf);
}
if (preg->reganch)
printf("anchored ");
if (preg->regmust != 0) {
int i;
printf("must have:");
for (i = 0; i < preg->regmlen; i++) {
putchar(preg->program[preg->regmust + i]);
}
putchar('\n');
}
}
printf("\n");
}
/*
- regprop - printable representation of opcode
*/
static const char *regprop( int op )
{
static char buf[50];
switch (op) {
case BOL:
return "BOL";
case EOL:
return "EOL";
case BOLX:
return "BOLX";
case EOLX:
return "EOLX";
case ANY:
return "ANY";
case ANYOF:
return "ANYOF";
case ANYBUT:
return "ANYBUT";
case BRANCH:
return "BRANCH";
case EXACTLY:
return "EXACTLY";
case NOTHING:
return "NOTHING";
case BACK:
return "BACK";
case END:
return "END";
case REP:
return "REP";
case REPMIN:
return "REPMIN";
case REPX:
return "REPX";
case REPXMIN:
return "REPXMIN";
case WORDA:
return "WORDA";
case WORDZ:
return "WORDZ";
case OPENNC:
return "OPEN";
case CLOSENC:
return "CLOSE";
default:
if (op >= OPEN && op < CLOSE) {
snprintf(buf, sizeof(buf), "OPEN%d", op-OPEN);
}
else if (op >= CLOSE && op < CLOSE_END) {
snprintf(buf, sizeof(buf), "CLOSE%d", op-CLOSE);
}
else {
snprintf(buf, sizeof(buf), "?%d?\n", op);
}
return(buf);
}
}
#endif /* JIM_BOOTSTRAP */
size_t jim_regerror(int errcode, const regex_t *preg, char *errbuf, size_t errbuf_size)
{
static const char *error_strings[] = {
"success",
"no match",
"bad pattern",
"null argument",
"unknown error",
"too big",
"out of memory",
"too many ()",
"parentheses () not balanced",
"braces {} not balanced",
"invalid repetition count(s)",
"extra characters",
"*+ of empty atom",
"nested count",
"internal error",
"count follows nothing",
"invalid escape \\ sequence",
"corrupted program",
"contains null char",
"brackets [] not balanced",
};
const char *err;
if (errcode < 0 || errcode >= REG_ERR_NUM) {
err = "Bad error code";
}
else {
err = error_strings[errcode];
}
return snprintf(errbuf, errbuf_size, "%s", err);
}
void jim_regfree(regex_t *preg)
{
free(preg->program);
}
#endif
|