1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
2913
2914
2915
2916
2917
2918
2919
2920
2921
2922
2923
2924
2925
2926
2927
2928
2929
2930
2931
2932
2933
2934
2935
2936
2937
2938
2939
2940
2941
2942
2943
2944
2945
2946
2947
2948
2949
2950
2951
2952
2953
2954
2955
2956
2957
2958
2959
2960
2961
2962
2963
2964
2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
2989
2990
2991
2992
2993
2994
2995
2996
2997
2998
2999
3000
3001
3002
3003
3004
3005
3006
3007
3008
3009
3010
3011
3012
3013
3014
3015
3016
3017
3018
3019
3020
3021
3022
3023
3024
3025
3026
3027
3028
3029
3030
3031
3032
3033
3034
3035
3036
3037
3038
3039
3040
3041
3042
3043
3044
3045
3046
3047
3048
3049
3050
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
3066
3067
3068
3069
3070
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
3081
3082
3083
3084
3085
3086
3087
3088
3089
3090
3091
3092
3093
3094
3095
3096
3097
3098
3099
3100
3101
3102
3103
3104
3105
3106
3107
3108
3109
3110
3111
3112
3113
3114
3115
3116
3117
3118
3119
3120
3121
3122
3123
3124
3125
3126
3127
3128
3129
3130
3131
3132
3133
3134
3135
3136
3137
3138
3139
3140
3141
3142
3143
3144
3145
3146
3147
3148
3149
3150
3151
3152
3153
3154
3155
3156
3157
3158
3159
3160
3161
3162
3163
3164
3165
3166
3167
3168
3169
3170
3171
3172
3173
3174
3175
3176
3177
3178
3179
3180
3181
3182
3183
3184
3185
3186
3187
3188
3189
3190
3191
3192
3193
3194
3195
3196
3197
3198
3199
3200
3201
3202
3203
3204
3205
3206
3207
3208
3209
3210
3211
3212
3213
3214
3215
3216
3217
3218
3219
3220
3221
3222
3223
3224
3225
3226
3227
3228
3229
3230
3231
3232
3233
3234
3235
3236
3237
3238
3239
3240
3241
3242
3243
3244
3245
3246
3247
3248
3249
3250
3251
3252
3253
3254
3255
3256
3257
3258
3259
3260
3261
3262
3263
3264
3265
3266
3267
3268
3269
3270
3271
3272
3273
3274
3275
3276
3277
3278
3279
3280
3281
3282
3283
3284
3285
3286
3287
3288
3289
3290
3291
3292
3293
3294
3295
3296
3297
3298
3299
3300
3301
3302
3303
3304
3305
3306
3307
3308
3309
3310
3311
3312
3313
3314
3315
3316
3317
3318
3319
3320
3321
3322
3323
3324
3325
3326
3327
3328
3329
3330
3331
3332
3333
3334
3335
3336
3337
3338
3339
3340
3341
3342
3343
3344
3345
3346
3347
3348
3349
3350
3351
3352
3353
3354
3355
3356
3357
3358
3359
3360
3361
3362
3363
3364
3365
3366
3367
3368
3369
3370
3371
3372
3373
3374
3375
3376
3377
3378
3379
3380
3381
3382
3383
3384
3385
3386
3387
3388
3389
3390
3391
3392
3393
3394
3395
3396
3397
3398
3399
3400
3401
3402
3403
3404
3405
3406
3407
3408
3409
3410
3411
3412
3413
3414
3415
3416
3417
3418
3419
3420
3421
3422
3423
3424
3425
3426
3427
3428
3429
3430
3431
3432
3433
3434
3435
3436
3437
3438
3439
3440
3441
3442
3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455
3456
3457
3458
3459
3460
3461
3462
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
3494
3495
3496
3497
3498
3499
3500
3501
3502
3503
3504
3505
3506
3507
3508
3509
3510
3511
3512
3513
3514
3515
3516
3517
3518
3519
3520
3521
3522
3523
3524
3525
3526
3527
3528
3529
3530
3531
3532
3533
3534
3535
3536
3537
3538
3539
3540
3541
3542
3543
3544
3545
3546
3547
3548
3549
3550
3551
3552
3553
3554
3555
3556
3557
3558
3559
3560
3561
3562
3563
3564
3565
3566
3567
3568
3569
3570
3571
3572
3573
3574
3575
3576
3577
3578
3579
3580
3581
3582
3583
3584
3585
3586
3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
3610
3611
3612
3613
3614
3615
3616
3617
3618
3619
3620
3621
3622
3623
3624
3625
3626
3627
3628
3629
3630
3631
3632
3633
3634
3635
3636
3637
3638
3639
3640
3641
3642
3643
3644
3645
3646
3647
3648
3649
3650
3651
3652
3653
3654
3655
3656
3657
3658
3659
3660
3661
3662
3663
3664
3665
3666
3667
3668
3669
3670
3671
3672
3673
3674
3675
3676
3677
3678
3679
3680
3681
3682
3683
3684
3685
3686
3687
3688
3689
3690
3691
3692
3693
3694
3695
3696
3697
3698
3699
3700
3701
3702
3703
3704
3705
3706
3707
3708
3709
3710
3711
3712
3713
3714
3715
3716
3717
3718
3719
3720
3721
3722
3723
3724
3725
3726
3727
3728
3729
3730
3731
3732
3733
3734
3735
3736
3737
3738
3739
3740
3741
3742
3743
3744
3745
3746
3747
3748
3749
3750
3751
3752
3753
3754
3755
3756
3757
3758
3759
3760
3761
3762
3763
3764
3765
3766
3767
3768
3769
3770
3771
3772
3773
3774
3775
3776
3777
3778
3779
3780
3781
3782
3783
3784
3785
3786
3787
3788
3789
3790
3791
3792
3793
3794
3795
3796
3797
3798
3799
3800
3801
3802
3803
3804
3805
3806
3807
3808
3809
3810
3811
3812
3813
3814
3815
3816
3817
3818
3819
3820
3821
3822
3823
3824
3825
3826
3827
3828
3829
3830
3831
3832
3833
3834
3835
3836
3837
3838
3839
3840
3841
3842
3843
3844
3845
3846
3847
3848
3849
3850
3851
3852
3853
3854
3855
3856
3857
3858
3859
3860
3861
3862
3863
3864
3865
3866
3867
3868
3869
3870
3871
3872
3873
3874
3875
3876
3877
3878
3879
3880
3881
3882
3883
3884
3885
3886
3887
3888
3889
3890
3891
3892
3893
3894
3895
3896
3897
3898
3899
3900
3901
3902
3903
3904
3905
3906
3907
3908
3909
3910
3911
3912
3913
3914
3915
3916
3917
3918
3919
3920
3921
3922
3923
3924
3925
3926
3927
3928
3929
3930
3931
3932
3933
3934
3935
3936
3937
3938
3939
3940
3941
3942
3943
3944
3945
3946
3947
3948
3949
3950
3951
3952
3953
3954
3955
3956
3957
3958
3959
3960
3961
3962
3963
3964
3965
3966
3967
3968
3969
3970
3971
3972
3973
3974
3975
3976
3977
3978
3979
3980
3981
3982
3983
3984
3985
3986
3987
3988
3989
3990
3991
3992
3993
3994
3995
3996
3997
3998
3999
4000
4001
4002
4003
4004
4005
4006
4007
4008
4009
4010
4011
4012
4013
4014
4015
4016
4017
4018
4019
4020
4021
4022
4023
4024
4025
4026
4027
4028
4029
4030
4031
4032
4033
4034
4035
4036
4037
4038
4039
4040
4041
4042
4043
4044
4045
4046
4047
4048
4049
4050
4051
4052
4053
4054
4055
4056
4057
4058
4059
4060
4061
4062
4063
4064
4065
4066
4067
4068
4069
4070
4071
4072
4073
4074
4075
4076
4077
4078
4079
4080
4081
4082
4083
4084
4085
4086
4087
4088
4089
4090
4091
4092
4093
4094
4095
4096
4097
4098
4099
4100
4101
4102
4103
4104
4105
4106
4107
4108
4109
4110
4111
4112
4113
4114
4115
4116
4117
4118
4119
4120
4121
4122
4123
4124
4125
4126
4127
4128
4129
4130
4131
4132
4133
4134
4135
4136
4137
4138
4139
4140
4141
4142
4143
4144
4145
4146
4147
4148
4149
4150
4151
4152
4153
4154
4155
4156
4157
4158
4159
4160
4161
4162
4163
4164
4165
4166
4167
4168
4169
4170
4171
4172
4173
4174
4175
4176
4177
4178
4179
4180
4181
4182
4183
4184
4185
4186
4187
4188
4189
4190
4191
4192
4193
4194
4195
4196
4197
4198
4199
4200
4201
4202
4203
4204
4205
4206
4207
4208
4209
4210
4211
4212
4213
4214
4215
4216
4217
4218
4219
4220
4221
4222
4223
4224
4225
4226
4227
4228
4229
4230
4231
4232
4233
4234
4235
4236
4237
4238
4239
4240
4241
4242
4243
4244
4245
4246
4247
4248
4249
4250
4251
4252
4253
4254
4255
4256
4257
4258
4259
4260
4261
4262
4263
4264
4265
4266
4267
4268
4269
4270
4271
4272
4273
4274
4275
4276
4277
4278
4279
4280
4281
4282
4283
4284
4285
4286
4287
4288
4289
4290
4291
4292
4293
4294
4295
4296
4297
4298
4299
4300
4301
4302
4303
4304
4305
4306
4307
4308
4309
4310
4311
4312
4313
4314
4315
4316
4317
4318
4319
4320
4321
4322
4323
4324
4325
4326
4327
4328
4329
4330
4331
4332
4333
4334
4335
4336
4337
4338
4339
4340
4341
4342
4343
4344
4345
4346
4347
4348
4349
4350
4351
4352
4353
4354
4355
4356
4357
4358
4359
4360
4361
4362
4363
4364
4365
4366
4367
4368
4369
4370
4371
4372
4373
4374
4375
4376
4377
4378
4379
4380
4381
4382
4383
4384
4385
4386
4387
4388
4389
4390
4391
4392
4393
4394
4395
4396
4397
4398
4399
4400
4401
4402
4403
4404
4405
4406
4407
4408
4409
4410
4411
4412
4413
4414
4415
4416
4417
4418
4419
4420
4421
4422
4423
4424
4425
4426
4427
4428
4429
4430
4431
4432
4433
4434
4435
4436
4437
4438
4439
4440
4441
4442
4443
4444
4445
4446
4447
4448
4449
4450
4451
4452
4453
4454
4455
4456
4457
4458
4459
4460
4461
4462
4463
4464
4465
4466
4467
4468
4469
4470
4471
4472
4473
4474
4475
4476
4477
4478
4479
4480
4481
4482
4483
4484
4485
4486
4487
4488
4489
4490
4491
4492
4493
4494
4495
4496
4497
4498
4499
4500
4501
4502
4503
4504
4505
4506
4507
4508
4509
4510
4511
4512
4513
4514
4515
4516
4517
4518
4519
4520
4521
4522
4523
4524
4525
4526
4527
4528
4529
4530
4531
4532
4533
4534
4535
4536
4537
4538
4539
4540
4541
4542
4543
4544
4545
4546
4547
4548
4549
4550
4551
4552
4553
4554
4555
4556
4557
4558
4559
4560
4561
4562
4563
4564
4565
4566
4567
4568
4569
4570
4571
4572
4573
4574
4575
4576
4577
4578
4579
4580
4581
4582
4583
4584
4585
4586
4587
4588
4589
4590
4591
4592
4593
4594
4595
4596
4597
4598
4599
4600
4601
4602
4603
4604
4605
4606
4607
4608
4609
4610
4611
4612
4613
4614
4615
4616
4617
4618
4619
4620
4621
4622
4623
4624
4625
4626
4627
4628
4629
4630
4631
4632
4633
4634
4635
4636
4637
4638
4639
4640
4641
4642
4643
4644
4645
4646
4647
4648
4649
4650
4651
4652
4653
4654
4655
4656
4657
4658
4659
4660
4661
4662
4663
4664
4665
4666
4667
4668
4669
4670
4671
4672
4673
4674
4675
4676
4677
4678
4679
4680
4681
4682
4683
4684
4685
4686
4687
4688
4689
4690
4691
4692
4693
4694
4695
4696
4697
4698
4699
4700
4701
4702
4703
4704
4705
4706
4707
4708
4709
4710
4711
4712
4713
4714
4715
4716
4717
4718
4719
4720
4721
4722
4723
4724
4725
4726
4727
4728
4729
4730
4731
4732
4733
4734
4735
4736
4737
4738
4739
4740
4741
4742
4743
4744
4745
4746
4747
4748
4749
4750
4751
4752
4753
4754
4755
4756
4757
4758
4759
4760
4761
4762
4763
4764
4765
4766
4767
4768
4769
4770
4771
4772
4773
4774
4775
4776
4777
4778
4779
4780
4781
4782
4783
4784
4785
4786
4787
4788
4789
4790
4791
4792
4793
4794
4795
4796
4797
4798
4799
4800
4801
4802
4803
4804
4805
4806
4807
4808
4809
4810
4811
4812
4813
4814
4815
4816
4817
4818
4819
4820
4821
4822
4823
4824
4825
4826
4827
4828
4829
4830
4831
4832
4833
4834
4835
4836
4837
4838
4839
4840
4841
4842
4843
4844
4845
4846
4847
4848
4849
4850
4851
4852
4853
4854
4855
4856
4857
4858
4859
4860
4861
4862
4863
4864
4865
4866
4867
4868
4869
4870
4871
4872
4873
4874
4875
4876
4877
4878
4879
4880
4881
4882
4883
4884
4885
4886
4887
4888
4889
4890
4891
4892
4893
4894
4895
4896
4897
4898
4899
4900
4901
4902
4903
4904
4905
4906
4907
4908
4909
4910
4911
4912
4913
4914
4915
4916
4917
4918
4919
4920
4921
4922
4923
4924
4925
4926
4927
4928
4929
4930
4931
4932
4933
4934
4935
4936
4937
4938
4939
4940
4941
4942
4943
4944
4945
4946
4947
4948
4949
4950
4951
4952
4953
4954
4955
4956
4957
4958
4959
4960
4961
4962
4963
4964
4965
4966
4967
4968
4969
4970
4971
4972
4973
4974
4975
4976
4977
4978
4979
4980
4981
4982
4983
4984
4985
4986
4987
4988
4989
4990
4991
4992
4993
4994
4995
4996
4997
4998
4999
5000
5001
5002
5003
5004
5005
5006
5007
5008
5009
5010
5011
5012
5013
5014
5015
5016
5017
5018
5019
5020
5021
5022
5023
5024
5025
5026
5027
5028
5029
5030
5031
5032
5033
5034
5035
5036
5037
5038
5039
5040
5041
5042
5043
5044
5045
5046
5047
5048
5049
5050
5051
5052
5053
5054
5055
5056
5057
5058
5059
5060
5061
5062
5063
5064
5065
5066
5067
5068
5069
5070
5071
5072
5073
5074
5075
5076
5077
5078
5079
5080
5081
5082
5083
5084
5085
5086
5087
5088
5089
5090
5091
5092
5093
5094
5095
5096
5097
5098
5099
5100
5101
5102
5103
5104
5105
5106
5107
5108
5109
5110
5111
5112
5113
5114
5115
5116
5117
5118
5119
5120
5121
5122
5123
5124
5125
5126
5127
5128
5129
5130
5131
5132
5133
5134
5135
5136
5137
5138
5139
5140
5141
5142
5143
5144
5145
5146
5147
5148
5149
5150
5151
5152
5153
5154
5155
5156
5157
5158
5159
5160
5161
5162
5163
5164
5165
5166
5167
5168
5169
5170
5171
5172
5173
5174
5175
5176
5177
5178
5179
5180
5181
5182
5183
5184
5185
5186
5187
5188
5189
5190
5191
5192
5193
5194
5195
5196
5197
5198
5199
5200
5201
5202
5203
5204
5205
5206
5207
5208
5209
5210
5211
5212
5213
5214
5215
5216
5217
5218
5219
5220
5221
5222
5223
5224
5225
5226
5227
5228
5229
5230
5231
5232
5233
5234
5235
5236
5237
5238
5239
5240
5241
5242
5243
5244
5245
5246
5247
5248
5249
5250
5251
5252
5253
5254
5255
5256
5257
5258
5259
5260
5261
5262
5263
5264
5265
5266
5267
5268
5269
5270
5271
5272
5273
5274
5275
5276
5277
5278
5279
5280
5281
5282
5283
5284
5285
5286
5287
5288
5289
5290
5291
5292
5293
5294
5295
5296
5297
5298
5299
5300
5301
5302
5303
5304
5305
5306
5307
5308
5309
5310
5311
5312
5313
5314
5315
5316
5317
5318
5319
5320
5321
5322
5323
5324
5325
5326
5327
5328
5329
5330
5331
5332
5333
5334
5335
5336
5337
5338
5339
5340
5341
5342
5343
5344
5345
5346
5347
5348
5349
5350
5351
5352
5353
5354
5355
5356
5357
5358
5359
5360
5361
5362
5363
5364
5365
5366
5367
5368
5369
5370
5371
5372
5373
5374
5375
5376
5377
5378
5379
5380
5381
5382
5383
5384
5385
5386
5387
5388
5389
5390
5391
5392
5393
5394
5395
5396
5397
5398
5399
5400
5401
5402
5403
5404
5405
5406
5407
5408
5409
5410
5411
5412
5413
5414
5415
5416
5417
5418
5419
5420
5421
5422
5423
5424
5425
5426
5427
5428
5429
5430
5431
5432
5433
5434
5435
5436
5437
5438
5439
5440
5441
5442
5443
5444
5445
5446
5447
5448
5449
5450
5451
5452
5453
5454
5455
5456
5457
5458
5459
5460
5461
5462
5463
5464
5465
5466
5467
5468
5469
5470
5471
5472
5473
5474
5475
5476
5477
5478
5479
5480
5481
5482
5483
5484
5485
5486
5487
5488
5489
5490
5491
5492
5493
5494
5495
5496
5497
5498
5499
5500
5501
5502
5503
5504
5505
5506
5507
5508
5509
5510
5511
5512
5513
5514
5515
5516
5517
5518
5519
5520
5521
5522
5523
5524
5525
5526
5527
5528
5529
5530
5531
5532
5533
5534
5535
5536
5537
5538
5539
5540
5541
5542
5543
5544
5545
5546
5547
5548
5549
5550
5551
5552
5553
5554
5555
5556
5557
5558
5559
5560
5561
5562
5563
5564
5565
5566
5567
5568
5569
5570
5571
5572
5573
5574
5575
5576
5577
5578
5579
5580
5581
5582
5583
5584
5585
5586
5587
5588
5589
5590
5591
5592
5593
5594
5595
5596
5597
5598
5599
5600
5601
5602
5603
5604
5605
5606
5607
5608
5609
5610
5611
5612
5613
5614
5615
5616
5617
5618
5619
5620
5621
5622
5623
5624
5625
5626
5627
5628
5629
5630
5631
5632
5633
5634
5635
5636
5637
5638
5639
5640
5641
5642
5643
5644
5645
5646
5647
5648
5649
5650
5651
5652
5653
5654
5655
5656
5657
5658
5659
5660
5661
5662
5663
5664
5665
5666
5667
5668
5669
5670
5671
5672
5673
5674
5675
5676
5677
5678
5679
5680
5681
5682
5683
5684
5685
5686
5687
5688
5689
5690
5691
5692
5693
5694
5695
5696
5697
5698
5699
5700
5701
5702
5703
5704
5705
5706
5707
5708
5709
5710
5711
5712
5713
5714
5715
5716
5717
5718
5719
5720
5721
5722
5723
5724
5725
5726
5727
5728
5729
5730
5731
5732
5733
5734
5735
5736
5737
5738
5739
5740
5741
5742
5743
5744
5745
5746
5747
5748
5749
5750
5751
5752
5753
5754
5755
5756
5757
5758
5759
5760
5761
5762
5763
5764
5765
5766
5767
5768
5769
5770
5771
5772
5773
5774
5775
5776
5777
5778
5779
5780
5781
5782
5783
5784
5785
5786
5787
5788
5789
5790
5791
5792
5793
5794
5795
5796
5797
5798
5799
5800
5801
5802
5803
5804
5805
5806
5807
5808
5809
5810
5811
5812
5813
5814
5815
5816
5817
5818
5819
5820
5821
5822
5823
5824
5825
5826
5827
5828
5829
5830
5831
5832
5833
5834
5835
5836
5837
5838
5839
5840
5841
5842
5843
5844
5845
5846
5847
5848
5849
5850
5851
5852
5853
5854
5855
5856
5857
5858
5859
5860
5861
5862
5863
5864
5865
5866
5867
5868
5869
5870
5871
5872
5873
5874
5875
5876
5877
5878
5879
5880
5881
5882
5883
5884
5885
5886
5887
5888
5889
5890
5891
5892
5893
5894
5895
5896
5897
5898
5899
5900
5901
5902
5903
5904
5905
5906
5907
5908
5909
5910
5911
5912
5913
5914
5915
5916
5917
5918
5919
5920
5921
5922
5923
5924
5925
5926
5927
5928
5929
5930
5931
5932
5933
5934
5935
5936
5937
5938
5939
5940
5941
5942
5943
5944
5945
5946
5947
5948
5949
5950
5951
5952
5953
5954
5955
5956
5957
5958
5959
5960
5961
5962
5963
5964
5965
5966
5967
5968
5969
5970
5971
5972
5973
5974
5975
5976
5977
5978
5979
5980
5981
5982
5983
5984
5985
5986
5987
5988
5989
5990
5991
5992
5993
5994
5995
5996
5997
5998
5999
6000
6001
6002
6003
6004
6005
6006
6007
6008
6009
6010
6011
6012
6013
6014
6015
6016
6017
6018
6019
6020
6021
6022
6023
6024
6025
6026
6027
6028
6029
6030
6031
6032
6033
6034
6035
6036
6037
6038
6039
6040
6041
6042
6043
6044
6045
6046
6047
6048
6049
6050
6051
6052
6053
6054
6055
6056
6057
6058
6059
6060
6061
6062
6063
6064
6065
6066
6067
6068
6069
6070
6071
6072
6073
6074
6075
6076
6077
6078
6079
6080
6081
6082
6083
6084
6085
6086
6087
6088
6089
6090
6091
6092
6093
6094
6095
6096
6097
6098
6099
6100
6101
6102
6103
6104
6105
6106
6107
6108
6109
6110
6111
6112
6113
6114
6115
6116
6117
6118
6119
6120
6121
6122
6123
6124
6125
6126
6127
6128
6129
6130
6131
6132
6133
6134
6135
6136
6137
6138
6139
6140
6141
6142
6143
6144
6145
6146
6147
6148
6149
6150
6151
6152
6153
6154
6155
6156
6157
6158
6159
6160
6161
6162
6163
6164
6165
6166
6167
6168
6169
6170
6171
6172
6173
6174
6175
6176
6177
6178
6179
6180
6181
6182
6183
6184
6185
6186
6187
6188
6189
6190
6191
6192
6193
6194
6195
6196
6197
6198
6199
6200
6201
6202
6203
6204
6205
6206
6207
6208
6209
6210
6211
6212
6213
6214
6215
6216
6217
6218
6219
6220
6221
6222
6223
6224
6225
6226
6227
6228
6229
6230
6231
6232
6233
6234
6235
6236
6237
6238
6239
6240
6241
6242
6243
6244
6245
6246
6247
6248
6249
6250
6251
6252
6253
6254
6255
6256
6257
6258
6259
6260
6261
6262
6263
6264
6265
6266
6267
6268
6269
6270
6271
6272
6273
6274
6275
6276
6277
6278
6279
6280
6281
6282
6283
6284
6285
6286
6287
6288
6289
6290
6291
6292
6293
6294
6295
6296
6297
6298
6299
6300
6301
6302
6303
6304
6305
6306
6307
6308
6309
6310
6311
6312
6313
6314
6315
6316
6317
6318
6319
6320
6321
6322
6323
6324
6325
6326
6327
6328
6329
6330
6331
6332
6333
6334
6335
6336
6337
6338
6339
6340
6341
6342
6343
6344
6345
6346
6347
6348
6349
6350
6351
6352
6353
6354
6355
6356
6357
6358
6359
6360
6361
6362
6363
6364
6365
6366
6367
6368
6369
6370
6371
6372
6373
6374
6375
6376
6377
6378
6379
6380
6381
6382
6383
6384
6385
6386
6387
6388
6389
6390
6391
6392
6393
6394
6395
6396
6397
6398
6399
6400
6401
6402
6403
6404
6405
6406
6407
6408
6409
6410
6411
6412
6413
6414
6415
6416
6417
6418
6419
6420
6421
6422
6423
6424
6425
6426
6427
6428
6429
6430
6431
6432
6433
6434
6435
6436
6437
6438
6439
6440
6441
6442
6443
6444
6445
6446
6447
6448
6449
6450
6451
6452
6453
6454
6455
6456
6457
6458
6459
6460
6461
6462
6463
6464
6465
6466
6467
6468
6469
6470
6471
6472
6473
6474
6475
6476
6477
6478
6479
6480
6481
6482
6483
6484
6485
6486
6487
6488
6489
6490
6491
6492
6493
6494
6495
6496
6497
6498
6499
6500
6501
6502
6503
6504
6505
6506
6507
6508
6509
6510
6511
6512
6513
6514
6515
6516
6517
6518
6519
6520
6521
6522
6523
6524
6525
6526
6527
6528
6529
6530
6531
6532
6533
6534
6535
6536
6537
6538
6539
6540
6541
6542
6543
6544
6545
6546
6547
6548
6549
6550
6551
6552
6553
6554
6555
6556
6557
6558
6559
6560
6561
6562
6563
6564
6565
6566
6567
6568
6569
6570
6571
6572
6573
6574
6575
6576
6577
6578
6579
6580
6581
6582
6583
6584
6585
6586
6587
6588
6589
6590
6591
6592
6593
6594
6595
6596
6597
6598
6599
6600
6601
6602
6603
6604
6605
6606
6607
6608
6609
6610
6611
6612
6613
6614
6615
6616
6617
6618
6619
6620
6621
6622
6623
6624
6625
6626
6627
6628
6629
6630
6631
6632
6633
6634
6635
6636
6637
6638
6639
6640
6641
6642
6643
6644
| | /* Extended regular expression matching and search library, version
0.12. (Implements POSIX draft P1003.2/D11.2, except for some of the
internationalization features.)
Copyright (C) 1993-2016 Free Software Foundation, Inc.
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 3, or (at your option)
any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <http://www.gnu.org/licenses/>. */
/* TODO:
- structure the opcode space into opcode+flag.
- merge with glibc's regex.[ch].
- replace (succeed_n + jump_n + set_number_at) with something that doesn't
need to modify the compiled regexp so that re_match can be reentrant.
- get rid of on_failure_jump_smart by doing the optimization in re_comp
rather than at run-time, so that re_match can be reentrant.
*/
/* AIX requires this to be the first thing in the file. */
#if defined _AIX && !defined REGEX_MALLOC
#pragma alloca
#endif
/* Ignore some GCC warnings for now. This section should go away
once the Emacs and Gnulib regex code is merged. */
#if 4 < __GNUC__ + (5 <= __GNUC_MINOR__) || defined __clang__
# pragma GCC diagnostic ignored "-Wstrict-overflow"
# ifndef emacs
# pragma GCC diagnostic ignored "-Wunused-function"
# pragma GCC diagnostic ignored "-Wunused-macros"
# pragma GCC diagnostic ignored "-Wunused-result"
# pragma GCC diagnostic ignored "-Wunused-variable"
# endif
#endif
#if 4 < __GNUC__ + (6 <= __GNUC_MINOR__) && ! defined __clang__
# pragma GCC diagnostic ignored "-Wunused-but-set-variable"
#endif
#include <config.h>
#include <stddef.h>
#ifdef emacs
/* We need this for `regex.h', and perhaps for the Emacs include files. */
# include <sys/types.h>
#endif
/* Whether to use ISO C Amendment 1 wide char functions.
Those should not be used for Emacs since it uses its own. */
#if defined _LIBC
#define WIDE_CHAR_SUPPORT 1
#else
#define WIDE_CHAR_SUPPORT \
(HAVE_WCTYPE_H && HAVE_WCHAR_H && HAVE_BTOWC && !emacs)
#endif
/* For platform which support the ISO C amendment 1 functionality we
support user defined character classes. */
#if WIDE_CHAR_SUPPORT
/* Solaris 2.5 has a bug: <wchar.h> must be included before <wctype.h>. */
# include <wchar.h>
# include <wctype.h>
#endif
#ifdef _LIBC
/* We have to keep the namespace clean. */
# define regfree(preg) __regfree (preg)
# define regexec(pr, st, nm, pm, ef) __regexec (pr, st, nm, pm, ef)
# define regcomp(preg, pattern, cflags) __regcomp (preg, pattern, cflags)
# define regerror(err_code, preg, errbuf, errbuf_size) \
__regerror (err_code, preg, errbuf, errbuf_size)
# define re_set_registers(bu, re, nu, st, en) \
__re_set_registers (bu, re, nu, st, en)
# define re_match_2(bufp, string1, size1, string2, size2, pos, regs, stop) \
__re_match_2 (bufp, string1, size1, string2, size2, pos, regs, stop)
# define re_match(bufp, string, size, pos, regs) \
__re_match (bufp, string, size, pos, regs)
# define re_search(bufp, string, size, startpos, range, regs) \
__re_search (bufp, string, size, startpos, range, regs)
# define re_compile_pattern(pattern, length, bufp) \
__re_compile_pattern (pattern, length, bufp)
# define re_set_syntax(syntax) __re_set_syntax (syntax)
# define re_search_2(bufp, st1, s1, st2, s2, startpos, range, regs, stop) \
__re_search_2 (bufp, st1, s1, st2, s2, startpos, range, regs, stop)
# define re_compile_fastmap(bufp) __re_compile_fastmap (bufp)
/* Make sure we call libc's function even if the user overrides them. */
# define btowc __btowc
# define iswctype __iswctype
# define wctype __wctype
# define WEAK_ALIAS(a,b) weak_alias (a, b)
/* We are also using some library internals. */
# include <locale/localeinfo.h>
# include <locale/elem-hash.h>
# include <langinfo.h>
#else
# define WEAK_ALIAS(a,b)
#endif
/* This is for other GNU distributions with internationalized messages. */
#if HAVE_LIBINTL_H || defined _LIBC
# include <libintl.h>
#else
# define gettext(msgid) (msgid)
#endif
#ifndef gettext_noop
/* This define is so xgettext can find the internationalizable
strings. */
# define gettext_noop(String) String
#endif
/* The `emacs' switch turns on certain matching commands
that make sense only in Emacs. */
#ifdef emacs
# include "lisp.h"
# include "character.h"
# include "buffer.h"
# include "syntax.h"
# include "category.h"
/* Make syntax table lookup grant data in gl_state. */
# define SYNTAX(c) syntax_property (c, 1)
# ifdef malloc
# undef malloc
# endif
# define malloc xmalloc
# ifdef realloc
# undef realloc
# endif
# define realloc xrealloc
# ifdef free
# undef free
# endif
# define free xfree
/* Converts the pointer to the char to BEG-based offset from the start. */
# define PTR_TO_OFFSET(d) POS_AS_IN_BUFFER (POINTER_TO_OFFSET (d))
# define POS_AS_IN_BUFFER(p) ((p) + (NILP (re_match_object) || BUFFERP (re_match_object)))
# define RE_MULTIBYTE_P(bufp) ((bufp)->multibyte)
# define RE_TARGET_MULTIBYTE_P(bufp) ((bufp)->target_multibyte)
# define RE_STRING_CHAR(p, multibyte) \
(multibyte ? (STRING_CHAR (p)) : (*(p)))
# define RE_STRING_CHAR_AND_LENGTH(p, len, multibyte) \
(multibyte ? (STRING_CHAR_AND_LENGTH (p, len)) : ((len) = 1, *(p)))
# define RE_CHAR_TO_MULTIBYTE(c) UNIBYTE_TO_CHAR (c)
# define RE_CHAR_TO_UNIBYTE(c) CHAR_TO_BYTE_SAFE (c)
/* Set C a (possibly converted to multibyte) character before P. P
points into a string which is the virtual concatenation of STR1
(which ends at END1) or STR2 (which ends at END2). */
# define GET_CHAR_BEFORE_2(c, p, str1, end1, str2, end2) \
do { \
if (target_multibyte) \
{ \
re_char *dtemp = (p) == (str2) ? (end1) : (p); \
re_char *dlimit = ((p) > (str2) && (p) <= (end2)) ? (str2) : (str1); \
while (dtemp-- > dlimit && !CHAR_HEAD_P (*dtemp)); \
c = STRING_CHAR (dtemp); \
} \
else \
{ \
(c = ((p) == (str2) ? (end1) : (p))[-1]); \
(c) = RE_CHAR_TO_MULTIBYTE (c); \
} \
} while (0)
/* Set C a (possibly converted to multibyte) character at P, and set
LEN to the byte length of that character. */
# define GET_CHAR_AFTER(c, p, len) \
do { \
if (target_multibyte) \
(c) = STRING_CHAR_AND_LENGTH (p, len); \
else \
{ \
(c) = *p; \
len = 1; \
(c) = RE_CHAR_TO_MULTIBYTE (c); \
} \
} while (0)
#else /* not emacs */
/* If we are not linking with Emacs proper,
we can't use the relocating allocator
even if config.h says that we can. */
# undef REL_ALLOC
# include <unistd.h>
/* When used in Emacs's lib-src, we need xmalloc and xrealloc. */
static void *
xmalloc (size_t size)
{
void *val = malloc (size);
if (!val && size)
{
write (STDERR_FILENO, "virtual memory exhausted\n", 25);
exit (1);
}
return val;
}
static void *
xrealloc (void *block, size_t size)
{
void *val;
/* We must call malloc explicitly when BLOCK is 0, since some
reallocs don't do this. */
if (! block)
val = malloc (size);
else
val = realloc (block, size);
if (!val && size)
{
write (STDERR_FILENO, "virtual memory exhausted\n", 25);
exit (1);
}
return val;
}
# ifdef malloc
# undef malloc
# endif
# define malloc xmalloc
# ifdef realloc
# undef realloc
# endif
# define realloc xrealloc
# include <stdbool.h>
# include <string.h>
/* Define the syntax stuff for \<, \>, etc. */
/* Sword must be nonzero for the wordchar pattern commands in re_match_2. */
enum syntaxcode { Swhitespace = 0, Sword = 1, Ssymbol = 2 };
/* Dummy macros for non-Emacs environments. */
# define MAX_MULTIBYTE_LENGTH 1
# define RE_MULTIBYTE_P(x) 0
# define RE_TARGET_MULTIBYTE_P(x) 0
# define WORD_BOUNDARY_P(c1, c2) (0)
# define BYTES_BY_CHAR_HEAD(p) (1)
# define PREV_CHAR_BOUNDARY(p, limit) ((p)--)
# define STRING_CHAR(p) (*(p))
# define RE_STRING_CHAR(p, multibyte) STRING_CHAR (p)
# define CHAR_STRING(c, s) (*(s) = (c), 1)
# define STRING_CHAR_AND_LENGTH(p, actual_len) ((actual_len) = 1, *(p))
# define RE_STRING_CHAR_AND_LENGTH(p, len, multibyte) STRING_CHAR_AND_LENGTH (p, len)
# define RE_CHAR_TO_MULTIBYTE(c) (c)
# define RE_CHAR_TO_UNIBYTE(c) (c)
# define GET_CHAR_BEFORE_2(c, p, str1, end1, str2, end2) \
(c = ((p) == (str2) ? *((end1) - 1) : *((p) - 1)))
# define GET_CHAR_AFTER(c, p, len) \
(c = *p, len = 1)
# define CHAR_BYTE8_P(c) (0)
# define CHAR_LEADING_CODE(c) (c)
#endif /* not emacs */
#ifndef RE_TRANSLATE
# define RE_TRANSLATE(TBL, C) ((unsigned char)(TBL)[C])
# define RE_TRANSLATE_P(TBL) (TBL)
#endif
\f
/* Get the interface, including the syntax bits. */
#include "regex.h"
/* isalpha etc. are used for the character classes. */
#include <ctype.h>
#ifdef emacs
/* 1 if C is an ASCII character. */
# define IS_REAL_ASCII(c) ((c) < 0200)
/* 1 if C is a unibyte character. */
# define ISUNIBYTE(c) (SINGLE_BYTE_CHAR_P ((c)))
/* The Emacs definitions should not be directly affected by locales. */
/* In Emacs, these are only used for single-byte characters. */
# define ISDIGIT(c) ((c) >= '0' && (c) <= '9')
# define ISCNTRL(c) ((c) < ' ')
# define ISXDIGIT(c) (((c) >= '0' && (c) <= '9') \
|| ((c) >= 'a' && (c) <= 'f') \
|| ((c) >= 'A' && (c) <= 'F'))
/* This is only used for single-byte characters. */
# define ISBLANK(c) ((c) == ' ' || (c) == '\t')
/* The rest must handle multibyte characters. */
# define ISGRAPH(c) (SINGLE_BYTE_CHAR_P (c) \
? (c) > ' ' && !((c) >= 0177 && (c) <= 0240) \
: graphicp (c))
# define ISPRINT(c) (SINGLE_BYTE_CHAR_P (c) \
? (c) >= ' ' && !((c) >= 0177 && (c) <= 0237) \
: printablep (c))
# define ISALNUM(c) (IS_REAL_ASCII (c) \
? (((c) >= 'a' && (c) <= 'z') \
|| ((c) >= 'A' && (c) <= 'Z') \
|| ((c) >= '0' && (c) <= '9')) \
: alphanumericp (c))
# define ISALPHA(c) (IS_REAL_ASCII (c) \
? (((c) >= 'a' && (c) <= 'z') \
|| ((c) >= 'A' && (c) <= 'Z')) \
: alphabeticp (c))
# define ISLOWER(c) lowercasep (c)
# define ISPUNCT(c) (IS_REAL_ASCII (c) \
? ((c) > ' ' && (c) < 0177 \
&& !(((c) >= 'a' && (c) <= 'z') \
|| ((c) >= 'A' && (c) <= 'Z') \
|| ((c) >= '0' && (c) <= '9'))) \
: SYNTAX (c) != Sword)
# define ISSPACE(c) (SYNTAX (c) == Swhitespace)
# define ISUPPER(c) uppercasep (c)
# define ISWORD(c) (SYNTAX (c) == Sword)
#else /* not emacs */
/* 1 if C is an ASCII character. */
# define IS_REAL_ASCII(c) ((c) < 0200)
/* This distinction is not meaningful, except in Emacs. */
# define ISUNIBYTE(c) 1
# ifdef isblank
# define ISBLANK(c) isblank (c)
# else
# define ISBLANK(c) ((c) == ' ' || (c) == '\t')
# endif
# ifdef isgraph
# define ISGRAPH(c) isgraph (c)
# else
# define ISGRAPH(c) (isprint (c) && !isspace (c))
# endif
/* Solaris defines ISPRINT so we must undefine it first. */
# undef ISPRINT
# define ISPRINT(c) isprint (c)
# define ISDIGIT(c) isdigit (c)
# define ISALNUM(c) isalnum (c)
# define ISALPHA(c) isalpha (c)
# define ISCNTRL(c) iscntrl (c)
# define ISLOWER(c) islower (c)
# define ISPUNCT(c) ispunct (c)
# define ISSPACE(c) isspace (c)
# define ISUPPER(c) isupper (c)
# define ISXDIGIT(c) isxdigit (c)
# define ISWORD(c) ISALPHA (c)
# ifdef _tolower
# define TOLOWER(c) _tolower (c)
# else
# define TOLOWER(c) tolower (c)
# endif
/* How many characters in the character set. */
# define CHAR_SET_SIZE 256
# ifdef SYNTAX_TABLE
extern char *re_syntax_table;
# else /* not SYNTAX_TABLE */
static char re_syntax_table[CHAR_SET_SIZE];
static void
init_syntax_once (void)
{
register int c;
static int done = 0;
if (done)
return;
memset (re_syntax_table, 0, sizeof re_syntax_table);
for (c = 0; c < CHAR_SET_SIZE; ++c)
if (ISALNUM (c))
re_syntax_table[c] = Sword;
re_syntax_table['_'] = Ssymbol;
done = 1;
}
# endif /* not SYNTAX_TABLE */
# define SYNTAX(c) re_syntax_table[(c)]
#endif /* not emacs */
\f
#define SIGN_EXTEND_CHAR(c) ((signed char) (c))
\f
/* Should we use malloc or alloca? If REGEX_MALLOC is not defined, we
use `alloca' instead of `malloc'. This is because using malloc in
re_search* or re_match* could cause memory leaks when C-g is used in
Emacs; also, malloc is slower and causes storage fragmentation. On
the other hand, malloc is more portable, and easier to debug.
Because we sometimes use alloca, some routines have to be macros,
not functions -- `alloca'-allocated space disappears at the end of the
function it is called in. */
#ifdef REGEX_MALLOC
# define REGEX_ALLOCATE malloc
# define REGEX_REALLOCATE(source, osize, nsize) realloc (source, nsize)
# define REGEX_FREE free
#else /* not REGEX_MALLOC */
/* Emacs already defines alloca, sometimes. */
# ifndef alloca
/* Make alloca work the best possible way. */
# ifdef __GNUC__
# define alloca __builtin_alloca
# else /* not __GNUC__ */
# ifdef HAVE_ALLOCA_H
# include <alloca.h>
# endif /* HAVE_ALLOCA_H */
# endif /* not __GNUC__ */
# endif /* not alloca */
# ifdef emacs
# define REGEX_USE_SAFE_ALLOCA USE_SAFE_ALLOCA
# define REGEX_SAFE_FREE() SAFE_FREE ()
# define REGEX_ALLOCATE SAFE_ALLOCA
# else
# define REGEX_ALLOCATE alloca
# endif
/* Assumes a `char *destination' variable. */
# define REGEX_REALLOCATE(source, osize, nsize) \
(destination = REGEX_ALLOCATE (nsize), \
memcpy (destination, source, osize))
/* No need to do anything to free, after alloca. */
# define REGEX_FREE(arg) ((void)0) /* Do nothing! But inhibit gcc warning. */
#endif /* not REGEX_MALLOC */
#ifndef REGEX_USE_SAFE_ALLOCA
# define REGEX_USE_SAFE_ALLOCA ((void) 0)
# define REGEX_SAFE_FREE() ((void) 0)
#endif
/* Define how to allocate the failure stack. */
#if defined REL_ALLOC && defined REGEX_MALLOC
# define REGEX_ALLOCATE_STACK(size) \
r_alloc (&failure_stack_ptr, (size))
# define REGEX_REALLOCATE_STACK(source, osize, nsize) \
r_re_alloc (&failure_stack_ptr, (nsize))
# define REGEX_FREE_STACK(ptr) \
r_alloc_free (&failure_stack_ptr)
#else /* not using relocating allocator */
# define REGEX_ALLOCATE_STACK(size) REGEX_ALLOCATE (size)
# define REGEX_REALLOCATE_STACK(source, o, n) REGEX_REALLOCATE (source, o, n)
# define REGEX_FREE_STACK(ptr) REGEX_FREE (ptr)
#endif /* not using relocating allocator */
/* True if `size1' is non-NULL and PTR is pointing anywhere inside
`string1' or just past its end. This works if PTR is NULL, which is
a good thing. */
#define FIRST_STRING_P(ptr) \
(size1 && string1 <= (ptr) && (ptr) <= string1 + size1)
/* (Re)Allocate N items of type T using malloc, or fail. */
#define TALLOC(n, t) ((t *) malloc ((n) * sizeof (t)))
#define RETALLOC(addr, n, t) ((addr) = (t *) realloc (addr, (n) * sizeof (t)))
#define REGEX_TALLOC(n, t) ((t *) REGEX_ALLOCATE ((n) * sizeof (t)))
#define BYTEWIDTH 8 /* In bits. */
#ifndef emacs
# undef max
# undef min
# define max(a, b) ((a) > (b) ? (a) : (b))
# define min(a, b) ((a) < (b) ? (a) : (b))
#endif
/* Type of source-pattern and string chars. */
#ifdef _MSC_VER
typedef unsigned char re_char;
typedef const re_char const_re_char;
#else
typedef const unsigned char re_char;
typedef re_char const_re_char;
#endif
typedef char boolean;
static regoff_t re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1, size_t size1,
re_char *string2, size_t size2,
ssize_t pos,
struct re_registers *regs,
ssize_t stop);
\f
/* These are the command codes that appear in compiled regular
expressions. Some opcodes are followed by argument bytes. A
command code can specify any interpretation whatsoever for its
arguments. Zero bytes may appear in the compiled regular expression. */
typedef enum
{
no_op = 0,
/* Succeed right away--no more backtracking. */
succeed,
/* Followed by one byte giving n, then by n literal bytes. */
exactn,
/* Matches any (more or less) character. */
anychar,
/* Matches any one char belonging to specified set. First
following byte is number of bitmap bytes. Then come bytes
for a bitmap saying which chars are in. Bits in each byte
are ordered low-bit-first. A character is in the set if its
bit is 1. A character too large to have a bit in the map is
automatically not in the set.
If the length byte has the 0x80 bit set, then that stuff
is followed by a range table:
2 bytes of flags for character sets (low 8 bits, high 8 bits)
See RANGE_TABLE_WORK_BITS below.
2 bytes, the number of pairs that follow (upto 32767)
pairs, each 2 multibyte characters,
each multibyte character represented as 3 bytes. */
charset,
/* Same parameters as charset, but match any character that is
not one of those specified. */
charset_not,
/* Start remembering the text that is matched, for storing in a
register. Followed by one byte with the register number, in
the range 0 to one less than the pattern buffer's re_nsub
field. */
start_memory,
/* Stop remembering the text that is matched and store it in a
memory register. Followed by one byte with the register
number, in the range 0 to one less than `re_nsub' in the
pattern buffer. */
stop_memory,
/* Match a duplicate of something remembered. Followed by one
byte containing the register number. */
duplicate,
/* Fail unless at beginning of line. */
begline,
/* Fail unless at end of line. */
endline,
/* Succeeds if at beginning of buffer (if emacs) or at beginning
of string to be matched (if not). */
begbuf,
/* Analogously, for end of buffer/string. */
endbuf,
/* Followed by two byte relative address to which to jump. */
jump,
/* Followed by two-byte relative address of place to resume at
in case of failure. */
on_failure_jump,
/* Like on_failure_jump, but pushes a placeholder instead of the
current string position when executed. */
on_failure_keep_string_jump,
/* Just like `on_failure_jump', except that it checks that we
don't get stuck in an infinite loop (matching an empty string
indefinitely). */
on_failure_jump_loop,
/* Just like `on_failure_jump_loop', except that it checks for
a different kind of loop (the kind that shows up with non-greedy
operators). This operation has to be immediately preceded
by a `no_op'. */
on_failure_jump_nastyloop,
/* A smart `on_failure_jump' used for greedy * and + operators.
It analyzes the loop before which it is put and if the
loop does not require backtracking, it changes itself to
`on_failure_keep_string_jump' and short-circuits the loop,
else it just defaults to changing itself into `on_failure_jump'.
It assumes that it is pointing to just past a `jump'. */
on_failure_jump_smart,
/* Followed by two-byte relative address and two-byte number n.
After matching N times, jump to the address upon failure.
Does not work if N starts at 0: use on_failure_jump_loop
instead. */
succeed_n,
/* Followed by two-byte relative address, and two-byte number n.
Jump to the address N times, then fail. */
jump_n,
/* Set the following two-byte relative address to the
subsequent two-byte number. The address *includes* the two
bytes of number. */
set_number_at,
wordbeg, /* Succeeds if at word beginning. */
wordend, /* Succeeds if at word end. */
wordbound, /* Succeeds if at a word boundary. */
notwordbound, /* Succeeds if not at a word boundary. */
symbeg, /* Succeeds if at symbol beginning. */
symend, /* Succeeds if at symbol end. */
/* Matches any character whose syntax is specified. Followed by
a byte which contains a syntax code, e.g., Sword. */
syntaxspec,
/* Matches any character whose syntax is not that specified. */
notsyntaxspec
#ifdef emacs
, at_dot, /* Succeeds if at point. */
/* Matches any character whose category-set contains the specified
category. The operator is followed by a byte which contains a
category code (mnemonic ASCII character). */
categoryspec,
/* Matches any character whose category-set does not contain the
specified category. The operator is followed by a byte which
contains the category code (mnemonic ASCII character). */
notcategoryspec
#endif /* emacs */
} re_opcode_t;
\f
/* Common operations on the compiled pattern. */
/* Store NUMBER in two contiguous bytes starting at DESTINATION. */
#define STORE_NUMBER(destination, number) \
do { \
(destination)[0] = (number) & 0377; \
(destination)[1] = (number) >> 8; \
} while (0)
/* Same as STORE_NUMBER, except increment DESTINATION to
the byte after where the number is stored. Therefore, DESTINATION
must be an lvalue. */
#define STORE_NUMBER_AND_INCR(destination, number) \
do { \
STORE_NUMBER (destination, number); \
(destination) += 2; \
} while (0)
/* Put into DESTINATION a number stored in two contiguous bytes starting
at SOURCE. */
#define EXTRACT_NUMBER(destination, source) \
((destination) = extract_number (source))
static int
extract_number (re_char *source)
{
unsigned leading_byte = SIGN_EXTEND_CHAR (source[1]);
return (leading_byte << 8) + source[0];
}
/* Same as EXTRACT_NUMBER, except increment SOURCE to after the number.
SOURCE must be an lvalue. */
#define EXTRACT_NUMBER_AND_INCR(destination, source) \
((destination) = extract_number_and_incr (&source))
static int
extract_number_and_incr (re_char **source)
{
int num = extract_number (*source);
*source += 2;
return num;
}
\f
/* Store a multibyte character in three contiguous bytes starting
DESTINATION, and increment DESTINATION to the byte after where the
character is stored. Therefore, DESTINATION must be an lvalue. */
#define STORE_CHARACTER_AND_INCR(destination, character) \
do { \
(destination)[0] = (character) & 0377; \
(destination)[1] = ((character) >> 8) & 0377; \
(destination)[2] = (character) >> 16; \
(destination) += 3; \
} while (0)
/* Put into DESTINATION a character stored in three contiguous bytes
starting at SOURCE. */
#define EXTRACT_CHARACTER(destination, source) \
do { \
(destination) = ((source)[0] \
| ((source)[1] << 8) \
| ((source)[2] << 16)); \
} while (0)
/* Macros for charset. */
/* Size of bitmap of charset P in bytes. P is a start of charset,
i.e. *P is (re_opcode_t) charset or (re_opcode_t) charset_not. */
#define CHARSET_BITMAP_SIZE(p) ((p)[1] & 0x7F)
/* Nonzero if charset P has range table. */
#define CHARSET_RANGE_TABLE_EXISTS_P(p) ((p)[1] & 0x80)
/* Return the address of range table of charset P. But not the start
of table itself, but the before where the number of ranges is
stored. `2 +' means to skip re_opcode_t and size of bitmap,
and the 2 bytes of flags at the start of the range table. */
#define CHARSET_RANGE_TABLE(p) (&(p)[4 + CHARSET_BITMAP_SIZE (p)])
#ifdef emacs
/* Extract the bit flags that start a range table. */
#define CHARSET_RANGE_TABLE_BITS(p) \
((p)[2 + CHARSET_BITMAP_SIZE (p)] \
+ (p)[3 + CHARSET_BITMAP_SIZE (p)] * 0x100)
#endif
/* Return the address of end of RANGE_TABLE. COUNT is number of
ranges (which is a pair of (start, end)) in the RANGE_TABLE. `* 2'
is start of range and end of range. `* 3' is size of each start
and end. */
#define CHARSET_RANGE_TABLE_END(range_table, count) \
((range_table) + (count) * 2 * 3)
\f
/* If DEBUG is defined, Regex prints many voluminous messages about what
it is doing (if the variable `debug' is nonzero). If linked with the
main program in `iregex.c', you can enter patterns and strings
interactively. And if linked with the main program in `main.c' and
the other test files, you can run the already-written tests. */
#ifdef DEBUG
/* We use standard I/O for debugging. */
# include <stdio.h>
/* It is useful to test things that ``must'' be true when debugging. */
# include <assert.h>
static int debug = -100000;
# define DEBUG_STATEMENT(e) e
# define DEBUG_PRINT(...) if (debug > 0) printf (__VA_ARGS__)
# define DEBUG_COMPILES_ARGUMENTS
# define DEBUG_PRINT_COMPILED_PATTERN(p, s, e) \
if (debug > 0) print_partial_compiled_pattern (s, e)
# define DEBUG_PRINT_DOUBLE_STRING(w, s1, sz1, s2, sz2) \
if (debug > 0) print_double_string (w, s1, sz1, s2, sz2)
/* Print the fastmap in human-readable form. */
static void
print_fastmap (char *fastmap)
{
unsigned was_a_range = 0;
unsigned i = 0;
while (i < (1 << BYTEWIDTH))
{
if (fastmap[i++])
{
was_a_range = 0;
putchar (i - 1);
while (i < (1 << BYTEWIDTH) && fastmap[i])
{
was_a_range = 1;
i++;
}
if (was_a_range)
{
printf ("-");
putchar (i - 1);
}
}
}
putchar ('\n');
}
/* Print a compiled pattern string in human-readable form, starting at
the START pointer into it and ending just before the pointer END. */
static void
print_partial_compiled_pattern (re_char *start, re_char *end)
{
int mcnt, mcnt2;
re_char *p = start;
re_char *pend = end;
if (start == NULL)
{
fprintf (stderr, "(null)\n");
return;
}
/* Loop over pattern commands. */
while (p < pend)
{
fprintf (stderr, "%td:\t", p - start);
switch ((re_opcode_t) *p++)
{
case no_op:
fprintf (stderr, "/no_op");
break;
case succeed:
fprintf (stderr, "/succeed");
break;
case exactn:
mcnt = *p++;
fprintf (stderr, "/exactn/%d", mcnt);
do
{
fprintf (stderr, "/%c", *p++);
}
while (--mcnt);
break;
case start_memory:
fprintf (stderr, "/start_memory/%d", *p++);
break;
case stop_memory:
fprintf (stderr, "/stop_memory/%d", *p++);
break;
case duplicate:
fprintf (stderr, "/duplicate/%d", *p++);
break;
case anychar:
fprintf (stderr, "/anychar");
break;
case charset:
case charset_not:
{
register int c, last = -100;
register int in_range = 0;
int length = CHARSET_BITMAP_SIZE (p - 1);
int has_range_table = CHARSET_RANGE_TABLE_EXISTS_P (p - 1);
fprintf (stderr, "/charset [%s",
(re_opcode_t) *(p - 1) == charset_not ? "^" : "");
if (p + *p >= pend)
fprintf (stderr, " !extends past end of pattern! ");
for (c = 0; c < 256; c++)
if (c / 8 < length
&& (p[1 + (c/8)] & (1 << (c % 8))))
{
/* Are we starting a range? */
if (last + 1 == c && ! in_range)
{
fprintf (stderr, "-");
in_range = 1;
}
/* Have we broken a range? */
else if (last + 1 != c && in_range)
{
fprintf (stderr, "%c", last);
in_range = 0;
}
if (! in_range)
fprintf (stderr, "%c", c);
last = c;
}
if (in_range)
fprintf (stderr, "%c", last);
fprintf (stderr, "]");
p += 1 + length;
if (has_range_table)
{
int count;
fprintf (stderr, "has-range-table");
/* ??? Should print the range table; for now, just skip it. */
p += 2; /* skip range table bits */
EXTRACT_NUMBER_AND_INCR (count, p);
p = CHARSET_RANGE_TABLE_END (p, count);
}
}
break;
case begline:
fprintf (stderr, "/begline");
break;
case endline:
fprintf (stderr, "/endline");
break;
case on_failure_jump:
EXTRACT_NUMBER_AND_INCR (mcnt, p);
fprintf (stderr, "/on_failure_jump to %td", p + mcnt - start);
break;
case on_failure_keep_string_jump:
EXTRACT_NUMBER_AND_INCR (mcnt, p);
fprintf (stderr, "/on_failure_keep_string_jump to %td",
p + mcnt - start);
break;
case on_failure_jump_nastyloop:
EXTRACT_NUMBER_AND_INCR (mcnt, p);
fprintf (stderr, "/on_failure_jump_nastyloop to %td",
p + mcnt - start);
break;
case on_failure_jump_loop:
EXTRACT_NUMBER_AND_INCR (mcnt, p);
fprintf (stderr, "/on_failure_jump_loop to %td",
p + mcnt - start);
break;
case on_failure_jump_smart:
EXTRACT_NUMBER_AND_INCR (mcnt, p);
fprintf (stderr, "/on_failure_jump_smart to %td",
p + mcnt - start);
break;
case jump:
EXTRACT_NUMBER_AND_INCR (mcnt, p);
fprintf (stderr, "/jump to %td", p + mcnt - start);
break;
case succeed_n:
EXTRACT_NUMBER_AND_INCR (mcnt, p);
EXTRACT_NUMBER_AND_INCR (mcnt2, p);
fprintf (stderr, "/succeed_n to %td, %d times",
p - 2 + mcnt - start, mcnt2);
break;
case jump_n:
EXTRACT_NUMBER_AND_INCR (mcnt, p);
EXTRACT_NUMBER_AND_INCR (mcnt2, p);
fprintf (stderr, "/jump_n to %td, %d times",
p - 2 + mcnt - start, mcnt2);
break;
case set_number_at:
EXTRACT_NUMBER_AND_INCR (mcnt, p);
EXTRACT_NUMBER_AND_INCR (mcnt2, p);
fprintf (stderr, "/set_number_at location %td to %d",
p - 2 + mcnt - start, mcnt2);
break;
case wordbound:
fprintf (stderr, "/wordbound");
break;
case notwordbound:
fprintf (stderr, "/notwordbound");
break;
case wordbeg:
fprintf (stderr, "/wordbeg");
break;
case wordend:
fprintf (stderr, "/wordend");
break;
case symbeg:
fprintf (stderr, "/symbeg");
break;
case symend:
fprintf (stderr, "/symend");
break;
case syntaxspec:
fprintf (stderr, "/syntaxspec");
mcnt = *p++;
fprintf (stderr, "/%d", mcnt);
break;
case notsyntaxspec:
fprintf (stderr, "/notsyntaxspec");
mcnt = *p++;
fprintf (stderr, "/%d", mcnt);
break;
# ifdef emacs
case at_dot:
fprintf (stderr, "/at_dot");
break;
case categoryspec:
fprintf (stderr, "/categoryspec");
mcnt = *p++;
fprintf (stderr, "/%d", mcnt);
break;
case notcategoryspec:
fprintf (stderr, "/notcategoryspec");
mcnt = *p++;
fprintf (stderr, "/%d", mcnt);
break;
# endif /* emacs */
case begbuf:
fprintf (stderr, "/begbuf");
break;
case endbuf:
fprintf (stderr, "/endbuf");
break;
default:
fprintf (stderr, "?%d", *(p-1));
}
fprintf (stderr, "\n");
}
fprintf (stderr, "%td:\tend of pattern.\n", p - start);
}
static void
print_compiled_pattern (struct re_pattern_buffer *bufp)
{
re_char *buffer = bufp->buffer;
print_partial_compiled_pattern (buffer, buffer + bufp->used);
printf ("%ld bytes used/%ld bytes allocated.\n",
bufp->used, bufp->allocated);
if (bufp->fastmap_accurate && bufp->fastmap)
{
printf ("fastmap: ");
print_fastmap (bufp->fastmap);
}
printf ("re_nsub: %zu\t", bufp->re_nsub);
printf ("regs_alloc: %d\t", bufp->regs_allocated);
printf ("can_be_null: %d\t", bufp->can_be_null);
printf ("no_sub: %d\t", bufp->no_sub);
printf ("not_bol: %d\t", bufp->not_bol);
printf ("not_eol: %d\t", bufp->not_eol);
#ifndef emacs
printf ("syntax: %lx\n", bufp->syntax);
#endif
fflush (stdout);
/* Perhaps we should print the translate table? */
}
static void
print_double_string (re_char *where, re_char *string1, ssize_t size1,
re_char *string2, ssize_t size2)
{
ssize_t this_char;
if (where == NULL)
printf ("(null)");
else
{
if (FIRST_STRING_P (where))
{
for (this_char = where - string1; this_char < size1; this_char++)
putchar (string1[this_char]);
where = string2;
}
for (this_char = where - string2; this_char < size2; this_char++)
putchar (string2[this_char]);
}
}
#else /* not DEBUG */
# undef assert
# define assert(e)
# define DEBUG_STATEMENT(e)
# define DEBUG_PRINT(...)
# define DEBUG_PRINT_COMPILED_PATTERN(p, s, e)
# define DEBUG_PRINT_DOUBLE_STRING(w, s1, sz1, s2, sz2)
#endif /* not DEBUG */
\f
#ifndef emacs
/* Set by `re_set_syntax' to the current regexp syntax to recognize. Can
also be assigned to arbitrarily: each pattern buffer stores its own
syntax, so it can be changed between regex compilations. */
/* This has no initializer because initialized variables in Emacs
become read-only after dumping. */
reg_syntax_t re_syntax_options;
/* Specify the precise syntax of regexps for compilation. This provides
for compatibility for various utilities which historically have
different, incompatible syntaxes.
The argument SYNTAX is a bit mask comprised of the various bits
defined in regex.h. We return the old syntax. */
reg_syntax_t
re_set_syntax (reg_syntax_t syntax)
{
reg_syntax_t ret = re_syntax_options;
re_syntax_options = syntax;
return ret;
}
WEAK_ALIAS (__re_set_syntax, re_set_syntax)
#endif
\f
/* This table gives an error message for each of the error codes listed
in regex.h. Obviously the order here has to be same as there.
POSIX doesn't require that we do anything for REG_NOERROR,
but why not be nice? */
static const char *re_error_msgid[] =
{
gettext_noop ("Success"), /* REG_NOERROR */
gettext_noop ("No match"), /* REG_NOMATCH */
gettext_noop ("Invalid regular expression"), /* REG_BADPAT */
gettext_noop ("Invalid collation character"), /* REG_ECOLLATE */
gettext_noop ("Invalid character class name"), /* REG_ECTYPE */
gettext_noop ("Trailing backslash"), /* REG_EESCAPE */
gettext_noop ("Invalid back reference"), /* REG_ESUBREG */
gettext_noop ("Unmatched [ or [^"), /* REG_EBRACK */
gettext_noop ("Unmatched ( or \\("), /* REG_EPAREN */
gettext_noop ("Unmatched \\{"), /* REG_EBRACE */
gettext_noop ("Invalid content of \\{\\}"), /* REG_BADBR */
gettext_noop ("Invalid range end"), /* REG_ERANGE */
gettext_noop ("Memory exhausted"), /* REG_ESPACE */
gettext_noop ("Invalid preceding regular expression"), /* REG_BADRPT */
gettext_noop ("Premature end of regular expression"), /* REG_EEND */
gettext_noop ("Regular expression too big"), /* REG_ESIZE */
gettext_noop ("Unmatched ) or \\)"), /* REG_ERPAREN */
gettext_noop ("Range striding over charsets") /* REG_ERANGEX */
};
\f
/* Avoiding alloca during matching, to placate r_alloc. */
/* Define MATCH_MAY_ALLOCATE unless we need to make sure that the
searching and matching functions should not call alloca. On some
systems, alloca is implemented in terms of malloc, and if we're
using the relocating allocator routines, then malloc could cause a
relocation, which might (if the strings being searched are in the
ralloc heap) shift the data out from underneath the regexp
routines.
Here's another reason to avoid allocation: Emacs
processes input from X in a signal handler; processing X input may
call malloc; if input arrives while a matching routine is calling
malloc, then we're scrod. But Emacs can't just block input while
calling matching routines; then we don't notice interrupts when
they come in. So, Emacs blocks input around all regexp calls
except the matching calls, which it leaves unprotected, in the
faith that they will not malloc. */
/* Normally, this is fine. */
#define MATCH_MAY_ALLOCATE
/* The match routines may not allocate if (1) they would do it with malloc
and (2) it's not safe for them to use malloc.
Note that if REL_ALLOC is defined, matching would not use malloc for the
failure stack, but we would still use it for the register vectors;
so REL_ALLOC should not affect this. */
#if defined REGEX_MALLOC && defined emacs
# undef MATCH_MAY_ALLOCATE
#endif
\f
/* Failure stack declarations and macros; both re_compile_fastmap and
re_match_2 use a failure stack. These have to be macros because of
REGEX_ALLOCATE_STACK. */
/* Approximate number of failure points for which to initially allocate space
when matching. If this number is exceeded, we allocate more
space, so it is not a hard limit. */
#ifndef INIT_FAILURE_ALLOC
# define INIT_FAILURE_ALLOC 20
#endif
/* Roughly the maximum number of failure points on the stack. Would be
exactly that if always used TYPICAL_FAILURE_SIZE items each time we failed.
This is a variable only so users of regex can assign to it; we never
change it ourselves. We always multiply it by TYPICAL_FAILURE_SIZE
before using it, so it should probably be a byte-count instead. */
# if defined MATCH_MAY_ALLOCATE
/* Note that 4400 was enough to cause a crash on Alpha OSF/1,
whose default stack limit is 2mb. In order for a larger
value to work reliably, you have to try to make it accord
with the process stack limit. */
size_t re_max_failures = 40000;
# else
size_t re_max_failures = 4000;
# endif
union fail_stack_elt
{
re_char *pointer;
/* This should be the biggest `int' that's no bigger than a pointer. */
long integer;
};
typedef union fail_stack_elt fail_stack_elt_t;
typedef struct
{
fail_stack_elt_t *stack;
size_t size;
size_t avail; /* Offset of next open position. */
size_t frame; /* Offset of the cur constructed frame. */
} fail_stack_type;
#define FAIL_STACK_EMPTY() (fail_stack.frame == 0)
/* Define macros to initialize and free the failure stack.
Do `return -2' if the alloc fails. */
#ifdef MATCH_MAY_ALLOCATE
# define INIT_FAIL_STACK() \
do { \
fail_stack.stack = \
REGEX_ALLOCATE_STACK (INIT_FAILURE_ALLOC * TYPICAL_FAILURE_SIZE \
* sizeof (fail_stack_elt_t)); \
\
if (fail_stack.stack == NULL) \
return -2; \
\
fail_stack.size = INIT_FAILURE_ALLOC; \
fail_stack.avail = 0; \
fail_stack.frame = 0; \
} while (0)
#else
# define INIT_FAIL_STACK() \
do { \
fail_stack.avail = 0; \
fail_stack.frame = 0; \
} while (0)
# define RETALLOC_IF(addr, n, t) \
if (addr) RETALLOC((addr), (n), t); else (addr) = TALLOC ((n), t)
#endif
/* Double the size of FAIL_STACK, up to a limit
which allows approximately `re_max_failures' items.
Return 1 if succeeds, and 0 if either ran out of memory
allocating space for it or it was already too large.
REGEX_REALLOCATE_STACK requires `destination' be declared. */
/* Factor to increase the failure stack size by
when we increase it.
This used to be 2, but 2 was too wasteful
because the old discarded stacks added up to as much space
were as ultimate, maximum-size stack. */
#define FAIL_STACK_GROWTH_FACTOR 4
#define GROW_FAIL_STACK(fail_stack) \
(((fail_stack).size * sizeof (fail_stack_elt_t) \
>= re_max_failures * TYPICAL_FAILURE_SIZE) \
? 0 \
: ((fail_stack).stack \
= REGEX_REALLOCATE_STACK ((fail_stack).stack, \
(fail_stack).size * sizeof (fail_stack_elt_t), \
min (re_max_failures * TYPICAL_FAILURE_SIZE, \
((fail_stack).size * sizeof (fail_stack_elt_t) \
* FAIL_STACK_GROWTH_FACTOR))), \
\
(fail_stack).stack == NULL \
? 0 \
: ((fail_stack).size \
= (min (re_max_failures * TYPICAL_FAILURE_SIZE, \
((fail_stack).size * sizeof (fail_stack_elt_t) \
* FAIL_STACK_GROWTH_FACTOR)) \
/ sizeof (fail_stack_elt_t)), \
1)))
/* Push a pointer value onto the failure stack.
Assumes the variable `fail_stack'. Probably should only
be called from within `PUSH_FAILURE_POINT'. */
#define PUSH_FAILURE_POINTER(item) \
fail_stack.stack[fail_stack.avail++].pointer = (item)
/* This pushes an integer-valued item onto the failure stack.
Assumes the variable `fail_stack'. Probably should only
be called from within `PUSH_FAILURE_POINT'. */
#define PUSH_FAILURE_INT(item) \
fail_stack.stack[fail_stack.avail++].integer = (item)
/* These POP... operations complement the PUSH... operations.
All assume that `fail_stack' is nonempty. */
#define POP_FAILURE_POINTER() fail_stack.stack[--fail_stack.avail].pointer
#define POP_FAILURE_INT() fail_stack.stack[--fail_stack.avail].integer
/* Individual items aside from the registers. */
#define NUM_NONREG_ITEMS 3
/* Used to examine the stack (to detect infinite loops). */
#define FAILURE_PAT(h) fail_stack.stack[(h) - 1].pointer
#define FAILURE_STR(h) (fail_stack.stack[(h) - 2].pointer)
#define NEXT_FAILURE_HANDLE(h) fail_stack.stack[(h) - 3].integer
#define TOP_FAILURE_HANDLE() fail_stack.frame
#define ENSURE_FAIL_STACK(space) \
while (REMAINING_AVAIL_SLOTS <= space) { \
if (!GROW_FAIL_STACK (fail_stack)) \
return -2; \
DEBUG_PRINT ("\n Doubled stack; size now: %zd\n", (fail_stack).size);\
DEBUG_PRINT (" slots available: %zd\n", REMAINING_AVAIL_SLOTS);\
}
/* Push register NUM onto the stack. */
#define PUSH_FAILURE_REG(num) \
do { \
char *destination; \
long n = num; \
ENSURE_FAIL_STACK(3); \
DEBUG_PRINT (" Push reg %ld (spanning %p -> %p)\n", \
n, regstart[n], regend[n]); \
PUSH_FAILURE_POINTER (regstart[n]); \
PUSH_FAILURE_POINTER (regend[n]); \
PUSH_FAILURE_INT (n); \
} while (0)
/* Change the counter's value to VAL, but make sure that it will
be reset when backtracking. */
#define PUSH_NUMBER(ptr,val) \
do { \
char *destination; \
int c; \
ENSURE_FAIL_STACK(3); \
EXTRACT_NUMBER (c, ptr); \
DEBUG_PRINT (" Push number %p = %d -> %d\n", ptr, c, val); \
PUSH_FAILURE_INT (c); \
PUSH_FAILURE_POINTER (ptr); \
PUSH_FAILURE_INT (-1); \
STORE_NUMBER (ptr, val); \
} while (0)
/* Pop a saved register off the stack. */
#define POP_FAILURE_REG_OR_COUNT() \
do { \
long pfreg = POP_FAILURE_INT (); \
if (pfreg == -1) \
{ \
/* It's a counter. */ \
/* Here, we discard `const', making re_match non-reentrant. */ \
unsigned char *ptr = (unsigned char*) POP_FAILURE_POINTER (); \
pfreg = POP_FAILURE_INT (); \
STORE_NUMBER (ptr, pfreg); \
DEBUG_PRINT (" Pop counter %p = %ld\n", ptr, pfreg); \
} \
else \
{ \
regend[pfreg] = POP_FAILURE_POINTER (); \
regstart[pfreg] = POP_FAILURE_POINTER (); \
DEBUG_PRINT (" Pop reg %ld (spanning %p -> %p)\n", \
pfreg, regstart[pfreg], regend[pfreg]); \
} \
} while (0)
/* Check that we are not stuck in an infinite loop. */
#define CHECK_INFINITE_LOOP(pat_cur, string_place) \
do { \
ssize_t failure = TOP_FAILURE_HANDLE (); \
/* Check for infinite matching loops */ \
while (failure > 0 \
&& (FAILURE_STR (failure) == string_place \
|| FAILURE_STR (failure) == NULL)) \
{ \
assert (FAILURE_PAT (failure) >= bufp->buffer \
&& FAILURE_PAT (failure) <= bufp->buffer + bufp->used); \
if (FAILURE_PAT (failure) == pat_cur) \
{ \
cycle = 1; \
break; \
} \
DEBUG_PRINT (" Other pattern: %p\n", FAILURE_PAT (failure)); \
failure = NEXT_FAILURE_HANDLE(failure); \
} \
DEBUG_PRINT (" Other string: %p\n", FAILURE_STR (failure)); \
} while (0)
/* Push the information about the state we will need
if we ever fail back to it.
Requires variables fail_stack, regstart, regend and
num_regs be declared. GROW_FAIL_STACK requires `destination' be
declared.
Does `return FAILURE_CODE' if runs out of memory. */
#define PUSH_FAILURE_POINT(pattern, string_place) \
do { \
char *destination; \
/* Must be int, so when we don't save any registers, the arithmetic \
of 0 + -1 isn't done as unsigned. */ \
\
DEBUG_STATEMENT (nfailure_points_pushed++); \
DEBUG_PRINT ("\nPUSH_FAILURE_POINT:\n"); \
DEBUG_PRINT (" Before push, next avail: %zd\n", (fail_stack).avail); \
DEBUG_PRINT (" size: %zd\n", (fail_stack).size);\
\
ENSURE_FAIL_STACK (NUM_NONREG_ITEMS); \
\
DEBUG_PRINT ("\n"); \
\
DEBUG_PRINT (" Push frame index: %zd\n", fail_stack.frame); \
PUSH_FAILURE_INT (fail_stack.frame); \
\
DEBUG_PRINT (" Push string %p: \"", string_place); \
DEBUG_PRINT_DOUBLE_STRING (string_place, string1, size1, string2, size2);\
DEBUG_PRINT ("\"\n"); \
PUSH_FAILURE_POINTER (string_place); \
\
DEBUG_PRINT (" Push pattern %p: ", pattern); \
DEBUG_PRINT_COMPILED_PATTERN (bufp, pattern, pend); \
PUSH_FAILURE_POINTER (pattern); \
\
/* Close the frame by moving the frame pointer past it. */ \
fail_stack.frame = fail_stack.avail; \
} while (0)
/* Estimate the size of data pushed by a typical failure stack entry.
An estimate is all we need, because all we use this for
is to choose a limit for how big to make the failure stack. */
/* BEWARE, the value `20' is hard-coded in emacs.c:main(). */
#define TYPICAL_FAILURE_SIZE 20
/* How many items can still be added to the stack without overflowing it. */
#define REMAINING_AVAIL_SLOTS ((fail_stack).size - (fail_stack).avail)
/* Pops what PUSH_FAIL_STACK pushes.
We restore into the parameters, all of which should be lvalues:
STR -- the saved data position.
PAT -- the saved pattern position.
REGSTART, REGEND -- arrays of string positions.
Also assumes the variables `fail_stack' and (if debugging), `bufp',
`pend', `string1', `size1', `string2', and `size2'. */
#define POP_FAILURE_POINT(str, pat) \
do { \
assert (!FAIL_STACK_EMPTY ()); \
\
/* Remove failure points and point to how many regs pushed. */ \
DEBUG_PRINT ("POP_FAILURE_POINT:\n"); \
DEBUG_PRINT (" Before pop, next avail: %zd\n", fail_stack.avail); \
DEBUG_PRINT (" size: %zd\n", fail_stack.size); \
\
/* Pop the saved registers. */ \
while (fail_stack.frame < fail_stack.avail) \
POP_FAILURE_REG_OR_COUNT (); \
\
pat = POP_FAILURE_POINTER (); \
DEBUG_PRINT (" Popping pattern %p: ", pat); \
DEBUG_PRINT_COMPILED_PATTERN (bufp, pat, pend); \
\
/* If the saved string location is NULL, it came from an \
on_failure_keep_string_jump opcode, and we want to throw away the \
saved NULL, thus retaining our current position in the string. */ \
str = POP_FAILURE_POINTER (); \
DEBUG_PRINT (" Popping string %p: \"", str); \
DEBUG_PRINT_DOUBLE_STRING (str, string1, size1, string2, size2); \
DEBUG_PRINT ("\"\n"); \
\
fail_stack.frame = POP_FAILURE_INT (); \
DEBUG_PRINT (" Popping frame index: %zd\n", fail_stack.frame); \
\
assert (fail_stack.avail >= 0); \
assert (fail_stack.frame <= fail_stack.avail); \
\
DEBUG_STATEMENT (nfailure_points_popped++); \
} while (0) /* POP_FAILURE_POINT */
\f
/* Registers are set to a sentinel when they haven't yet matched. */
#define REG_UNSET(e) ((e) == NULL)
\f
/* Subroutine declarations and macros for regex_compile. */
static reg_errcode_t regex_compile (re_char *pattern, size_t size,
#ifdef emacs
bool posix_backtracking,
const char *whitespace_regexp,
#else
reg_syntax_t syntax,
#endif
struct re_pattern_buffer *bufp);
static void store_op1 (re_opcode_t op, unsigned char *loc, int arg);
static void store_op2 (re_opcode_t op, unsigned char *loc, int arg1, int arg2);
static void insert_op1 (re_opcode_t op, unsigned char *loc,
int arg, unsigned char *end);
static void insert_op2 (re_opcode_t op, unsigned char *loc,
int arg1, int arg2, unsigned char *end);
static boolean at_begline_loc_p (re_char *pattern, re_char *p,
reg_syntax_t syntax);
static boolean at_endline_loc_p (re_char *p, re_char *pend,
reg_syntax_t syntax);
static re_char *skip_one_char (re_char *p);
static int analyze_first (re_char *p, re_char *pend,
char *fastmap, const int multibyte);
/* Fetch the next character in the uncompiled pattern, with no
translation. */
#define PATFETCH(c) \
do { \
int len; \
if (p == pend) return REG_EEND; \
c = RE_STRING_CHAR_AND_LENGTH (p, len, multibyte); \
p += len; \
} while (0)
/* If `translate' is non-null, return translate[D], else just D. We
cast the subscript to translate because some data is declared as
`char *', to avoid warnings when a string constant is passed. But
when we use a character as a subscript we must make it unsigned. */
#ifndef TRANSLATE
# define TRANSLATE(d) \
(RE_TRANSLATE_P (translate) ? RE_TRANSLATE (translate, (d)) : (d))
#endif
/* Macros for outputting the compiled pattern into `buffer'. */
/* If the buffer isn't allocated when it comes in, use this. */
#define INIT_BUF_SIZE 32
/* Make sure we have at least N more bytes of space in buffer. */
#define GET_BUFFER_SPACE(n) \
while ((size_t) (b - bufp->buffer + (n)) > bufp->allocated) \
EXTEND_BUFFER ()
/* Make sure we have one more byte of buffer space and then add C to it. */
#define BUF_PUSH(c) \
do { \
GET_BUFFER_SPACE (1); \
*b++ = (unsigned char) (c); \
} while (0)
/* Ensure we have two more bytes of buffer space and then append C1 and C2. */
#define BUF_PUSH_2(c1, c2) \
do { \
GET_BUFFER_SPACE (2); \
*b++ = (unsigned char) (c1); \
*b++ = (unsigned char) (c2); \
} while (0)
/* Store a jump with opcode OP at LOC to location TO. We store a
relative address offset by the three bytes the jump itself occupies. */
#define STORE_JUMP(op, loc, to) \
store_op1 (op, loc, (to) - (loc) - 3)
/* Likewise, for a two-argument jump. */
#define STORE_JUMP2(op, loc, to, arg) \
store_op2 (op, loc, (to) - (loc) - 3, arg)
/* Like `STORE_JUMP', but for inserting. Assume `b' is the buffer end. */
#define INSERT_JUMP(op, loc, to) \
insert_op1 (op, loc, (to) - (loc) - 3, b)
/* Like `STORE_JUMP2', but for inserting. Assume `b' is the buffer end. */
#define INSERT_JUMP2(op, loc, to, arg) \
insert_op2 (op, loc, (to) - (loc) - 3, arg, b)
/* This is not an arbitrary limit: the arguments which represent offsets
into the pattern are two bytes long. So if 2^15 bytes turns out to
be too small, many things would have to change. */
# define MAX_BUF_SIZE (1L << 15)
/* Extend the buffer by twice its current size via realloc and
reset the pointers that pointed into the old block to point to the
correct places in the new one. If extending the buffer results in it
being larger than MAX_BUF_SIZE, then flag memory exhausted. */
#if __BOUNDED_POINTERS__
# define SET_HIGH_BOUND(P) (__ptrhigh (P) = __ptrlow (P) + bufp->allocated)
# define MOVE_BUFFER_POINTER(P) \
(__ptrlow (P) = new_buffer + (__ptrlow (P) - old_buffer), \
SET_HIGH_BOUND (P), \
__ptrvalue (P) = new_buffer + (__ptrvalue (P) - old_buffer))
# define ELSE_EXTEND_BUFFER_HIGH_BOUND \
else \
{ \
SET_HIGH_BOUND (b); \
SET_HIGH_BOUND (begalt); \
if (fixup_alt_jump) \
SET_HIGH_BOUND (fixup_alt_jump); \
if (laststart) \
SET_HIGH_BOUND (laststart); \
if (pending_exact) \
SET_HIGH_BOUND (pending_exact); \
}
#else
# define MOVE_BUFFER_POINTER(P) ((P) = new_buffer + ((P) - old_buffer))
# define ELSE_EXTEND_BUFFER_HIGH_BOUND
#endif
#define EXTEND_BUFFER() \
do { \
unsigned char *old_buffer = bufp->buffer; \
if (bufp->allocated == MAX_BUF_SIZE) \
return REG_ESIZE; \
bufp->allocated <<= 1; \
if (bufp->allocated > MAX_BUF_SIZE) \
bufp->allocated = MAX_BUF_SIZE; \
RETALLOC (bufp->buffer, bufp->allocated, unsigned char); \
if (bufp->buffer == NULL) \
return REG_ESPACE; \
/* If the buffer moved, move all the pointers into it. */ \
if (old_buffer != bufp->buffer) \
{ \
unsigned char *new_buffer = bufp->buffer; \
MOVE_BUFFER_POINTER (b); \
MOVE_BUFFER_POINTER (begalt); \
if (fixup_alt_jump) \
MOVE_BUFFER_POINTER (fixup_alt_jump); \
if (laststart) \
MOVE_BUFFER_POINTER (laststart); \
if (pending_exact) \
MOVE_BUFFER_POINTER (pending_exact); \
} \
ELSE_EXTEND_BUFFER_HIGH_BOUND \
} while (0)
/* Since we have one byte reserved for the register number argument to
{start,stop}_memory, the maximum number of groups we can report
things about is what fits in that byte. */
#define MAX_REGNUM 255
/* But patterns can have more than `MAX_REGNUM' registers. We just
ignore the excess. */
typedef int regnum_t;
/* Macros for the compile stack. */
/* Since offsets can go either forwards or backwards, this type needs to
be able to hold values from -(MAX_BUF_SIZE - 1) to MAX_BUF_SIZE - 1. */
/* int may be not enough when sizeof(int) == 2. */
typedef long pattern_offset_t;
typedef struct
{
pattern_offset_t begalt_offset;
pattern_offset_t fixup_alt_jump;
pattern_offset_t laststart_offset;
regnum_t regnum;
} compile_stack_elt_t;
typedef struct
{
compile_stack_elt_t *stack;
size_t size;
size_t avail; /* Offset of next open position. */
} compile_stack_type;
#define INIT_COMPILE_STACK_SIZE 32
#define COMPILE_STACK_EMPTY (compile_stack.avail == 0)
#define COMPILE_STACK_FULL (compile_stack.avail == compile_stack.size)
/* The next available element. */
#define COMPILE_STACK_TOP (compile_stack.stack[compile_stack.avail])
/* Explicit quit checking is needed for Emacs, which uses polling to
process input events. */
#ifdef emacs
# define IMMEDIATE_QUIT_CHECK \
do { \
if (immediate_quit) QUIT; \
} while (0)
#else
# define IMMEDIATE_QUIT_CHECK ((void)0)
#endif
\f
/* Structure to manage work area for range table. */
struct range_table_work_area
{
int *table; /* actual work area. */
int allocated; /* allocated size for work area in bytes. */
int used; /* actually used size in words. */
int bits; /* flag to record character classes */
};
#ifdef emacs
/* Make sure that WORK_AREA can hold more N multibyte characters.
This is used only in set_image_of_range and set_image_of_range_1.
It expects WORK_AREA to be a pointer.
If it can't get the space, it returns from the surrounding function. */
#define EXTEND_RANGE_TABLE(work_area, n) \
do { \
if (((work_area).used + (n)) * sizeof (int) > (work_area).allocated) \
{ \
extend_range_table_work_area (&work_area); \
if ((work_area).table == 0) \
return (REG_ESPACE); \
} \
} while (0)
#define SET_RANGE_TABLE_WORK_AREA_BIT(work_area, bit) \
(work_area).bits |= (bit)
/* Set a range (RANGE_START, RANGE_END) to WORK_AREA. */
#define SET_RANGE_TABLE_WORK_AREA(work_area, range_start, range_end) \
do { \
EXTEND_RANGE_TABLE ((work_area), 2); \
(work_area).table[(work_area).used++] = (range_start); \
(work_area).table[(work_area).used++] = (range_end); \
} while (0)
#endif /* emacs */
/* Free allocated memory for WORK_AREA. */
#define FREE_RANGE_TABLE_WORK_AREA(work_area) \
do { \
if ((work_area).table) \
free ((work_area).table); \
} while (0)
#define CLEAR_RANGE_TABLE_WORK_USED(work_area) ((work_area).used = 0, (work_area).bits = 0)
#define RANGE_TABLE_WORK_USED(work_area) ((work_area).used)
#define RANGE_TABLE_WORK_BITS(work_area) ((work_area).bits)
#define RANGE_TABLE_WORK_ELT(work_area, i) ((work_area).table[i])
/* Bits used to implement the multibyte-part of the various character classes
such as [:alnum:] in a charset's range table. The code currently assumes
that only the low 16 bits are used. */
#define BIT_WORD 0x1
#define BIT_LOWER 0x2
#define BIT_PUNCT 0x4
#define BIT_SPACE 0x8
#define BIT_UPPER 0x10
#define BIT_MULTIBYTE 0x20
#define BIT_ALPHA 0x40
#define BIT_ALNUM 0x80
#define BIT_GRAPH 0x100
#define BIT_PRINT 0x200
\f
/* Set the bit for character C in a list. */
#define SET_LIST_BIT(c) (b[((c)) / BYTEWIDTH] |= 1 << ((c) % BYTEWIDTH))
#ifdef emacs
/* Store characters in the range FROM to TO in the bitmap at B (for
ASCII and unibyte characters) and WORK_AREA (for multibyte
characters) while translating them and paying attention to the
continuity of translated characters.
Implementation note: It is better to implement these fairly big
macros by a function, but it's not that easy because macros called
in this macro assume various local variables already declared. */
/* Both FROM and TO are ASCII characters. */
#define SETUP_ASCII_RANGE(work_area, FROM, TO) \
do { \
int C0, C1; \
\
for (C0 = (FROM); C0 <= (TO); C0++) \
{ \
C1 = TRANSLATE (C0); \
if (! ASCII_CHAR_P (C1)) \
{ \
SET_RANGE_TABLE_WORK_AREA ((work_area), C1, C1); \
if ((C1 = RE_CHAR_TO_UNIBYTE (C1)) < 0) \
C1 = C0; \
} \
SET_LIST_BIT (C1); \
} \
} while (0)
/* Both FROM and TO are unibyte characters (0x80..0xFF). */
#define SETUP_UNIBYTE_RANGE(work_area, FROM, TO) \
do { \
int C0, C1, C2, I; \
int USED = RANGE_TABLE_WORK_USED (work_area); \
\
for (C0 = (FROM); C0 <= (TO); C0++) \
{ \
C1 = RE_CHAR_TO_MULTIBYTE (C0); \
if (CHAR_BYTE8_P (C1)) \
SET_LIST_BIT (C0); \
else \
{ \
C2 = TRANSLATE (C1); \
if (C2 == C1 \
|| (C1 = RE_CHAR_TO_UNIBYTE (C2)) < 0) \
C1 = C0; \
SET_LIST_BIT (C1); \
for (I = RANGE_TABLE_WORK_USED (work_area) - 2; I >= USED; I -= 2) \
{ \
int from = RANGE_TABLE_WORK_ELT (work_area, I); \
int to = RANGE_TABLE_WORK_ELT (work_area, I + 1); \
\
if (C2 >= from - 1 && C2 <= to + 1) \
{ \
if (C2 == from - 1) \
RANGE_TABLE_WORK_ELT (work_area, I)--; \
else if (C2 == to + 1) \
RANGE_TABLE_WORK_ELT (work_area, I + 1)++; \
break; \
} \
} \
if (I < USED) \
SET_RANGE_TABLE_WORK_AREA ((work_area), C2, C2); \
} \
} \
} while (0)
/* Both FROM and TO are multibyte characters. */
#define SETUP_MULTIBYTE_RANGE(work_area, FROM, TO) \
do { \
int C0, C1, C2, I, USED = RANGE_TABLE_WORK_USED (work_area); \
\
SET_RANGE_TABLE_WORK_AREA ((work_area), (FROM), (TO)); \
for (C0 = (FROM); C0 <= (TO); C0++) \
{ \
C1 = TRANSLATE (C0); \
if ((C2 = RE_CHAR_TO_UNIBYTE (C1)) >= 0 \
|| (C1 != C0 && (C2 = RE_CHAR_TO_UNIBYTE (C0)) >= 0)) \
SET_LIST_BIT (C2); \
if (C1 >= (FROM) && C1 <= (TO)) \
continue; \
for (I = RANGE_TABLE_WORK_USED (work_area) - 2; I >= USED; I -= 2) \
{ \
int from = RANGE_TABLE_WORK_ELT (work_area, I); \
int to = RANGE_TABLE_WORK_ELT (work_area, I + 1); \
\
if (C1 >= from - 1 && C1 <= to + 1) \
{ \
if (C1 == from - 1) \
RANGE_TABLE_WORK_ELT (work_area, I)--; \
else if (C1 == to + 1) \
RANGE_TABLE_WORK_ELT (work_area, I + 1)++; \
break; \
} \
} \
if (I < USED) \
SET_RANGE_TABLE_WORK_AREA ((work_area), C1, C1); \
} \
} while (0)
#endif /* emacs */
/* Get the next unsigned number in the uncompiled pattern. */
#define GET_INTERVAL_COUNT(num) \
do { \
if (p == pend) \
FREE_STACK_RETURN (REG_EBRACE); \
else \
{ \
PATFETCH (c); \
while ('0' <= c && c <= '9') \
{ \
if (num < 0) \
num = 0; \
if (RE_DUP_MAX / 10 - (RE_DUP_MAX % 10 < c - '0') < num) \
FREE_STACK_RETURN (REG_BADBR); \
num = num * 10 + c - '0'; \
if (p == pend) \
FREE_STACK_RETURN (REG_EBRACE); \
PATFETCH (c); \
} \
} \
} while (0)
\f
#if ! WIDE_CHAR_SUPPORT
/* Parse a character class, i.e. string such as "[:name:]". *strp
points to the string to be parsed and limit is length, in bytes, of
that string.
If *strp point to a string that begins with "[:name:]", where name is
a non-empty sequence of lower case letters, *strp will be advanced past the
closing square bracket and RECC_* constant which maps to the name will be
returned. If name is not a valid character class name zero, or RECC_ERROR,
is returned.
Otherwise, if *strp doesn’t begin with "[:name:]", -1 is returned.
The function can be used on ASCII and multibyte (UTF-8-encoded) strings.
*/
re_wctype_t
re_wctype_parse (const unsigned char **strp, unsigned limit)
{
const char *beg = (const char *)*strp, *it;
if (limit < 4 || beg[0] != '[' || beg[1] != ':')
return -1;
beg += 2; /* skip opening ‘[:’ */
limit -= 3; /* opening ‘[:’ and half of closing ‘:]’; --limit handles rest */
for (it = beg; it[0] != ':' || it[1] != ']'; ++it)
if (!--limit)
return -1;
*strp = (const unsigned char *)(it + 2);
/* Sort tests in the length=five case by frequency the classes to minimize
number of times we fail the comparison. The frequencies of character class
names used in Emacs sources as of 2016-07-27:
$ find \( -name \*.c -o -name \*.el \) -exec grep -h '\[:[a-z]*:]' {} + |
sed 's/]/]\n/g' |grep -o '\[:[a-z]*:]' |sort |uniq -c |sort -nr
213 [:alnum:]
104 [:alpha:]
62 [:space:]
39 [:digit:]
36 [:blank:]
26 [:word:]
26 [:upper:]
21 [:lower:]
10 [:xdigit:]
10 [:punct:]
10 [:ascii:]
4 [:nonascii:]
4 [:graph:]
2 [:print:]
2 [:cntrl:]
1 [:ff:]
If you update this list, consider also updating chain of or’ed conditions
in execute_charset function.
*/
switch (it - beg) {
case 4:
if (!memcmp (beg, "word", 4)) return RECC_WORD;
break;
case 5:
if (!memcmp (beg, "alnum", 5)) return RECC_ALNUM;
if (!memcmp (beg, "alpha", 5)) return RECC_ALPHA;
if (!memcmp (beg, "space", 5)) return RECC_SPACE;
if (!memcmp (beg, "digit", 5)) return RECC_DIGIT;
if (!memcmp (beg, "blank", 5)) return RECC_BLANK;
if (!memcmp (beg, "upper", 5)) return RECC_UPPER;
if (!memcmp (beg, "lower", 5)) return RECC_LOWER;
if (!memcmp (beg, "punct", 5)) return RECC_PUNCT;
if (!memcmp (beg, "ascii", 5)) return RECC_ASCII;
if (!memcmp (beg, "graph", 5)) return RECC_GRAPH;
if (!memcmp (beg, "print", 5)) return RECC_PRINT;
if (!memcmp (beg, "cntrl", 5)) return RECC_CNTRL;
break;
case 6:
if (!memcmp (beg, "xdigit", 6)) return RECC_XDIGIT;
break;
case 7:
if (!memcmp (beg, "unibyte", 7)) return RECC_UNIBYTE;
break;
case 8:
if (!memcmp (beg, "nonascii", 8)) return RECC_NONASCII;
break;
case 9:
if (!memcmp (beg, "multibyte", 9)) return RECC_MULTIBYTE;
break;
}
return RECC_ERROR;
}
/* True if CH is in the char class CC. */
boolean
re_iswctype (int ch, re_wctype_t cc)
{
switch (cc)
{
case RECC_ALNUM: return ISALNUM (ch) != 0;
case RECC_ALPHA: return ISALPHA (ch) != 0;
case RECC_BLANK: return ISBLANK (ch) != 0;
case RECC_CNTRL: return ISCNTRL (ch) != 0;
case RECC_DIGIT: return ISDIGIT (ch) != 0;
case RECC_GRAPH: return ISGRAPH (ch) != 0;
case RECC_LOWER: return ISLOWER (ch) != 0;
case RECC_PRINT: return ISPRINT (ch) != 0;
case RECC_PUNCT: return ISPUNCT (ch) != 0;
case RECC_SPACE: return ISSPACE (ch) != 0;
case RECC_UPPER: return ISUPPER (ch) != 0;
case RECC_XDIGIT: return ISXDIGIT (ch) != 0;
case RECC_ASCII: return IS_REAL_ASCII (ch) != 0;
case RECC_NONASCII: return !IS_REAL_ASCII (ch);
case RECC_UNIBYTE: return ISUNIBYTE (ch) != 0;
case RECC_MULTIBYTE: return !ISUNIBYTE (ch);
case RECC_WORD: return ISWORD (ch) != 0;
case RECC_ERROR: return false;
default:
abort ();
}
}
/* Return a bit-pattern to use in the range-table bits to match multibyte
chars of class CC. */
static int
re_wctype_to_bit (re_wctype_t cc)
{
switch (cc)
{
case RECC_NONASCII:
case RECC_MULTIBYTE: return BIT_MULTIBYTE;
case RECC_ALPHA: return BIT_ALPHA;
case RECC_ALNUM: return BIT_ALNUM;
case RECC_WORD: return BIT_WORD;
case RECC_LOWER: return BIT_LOWER;
case RECC_UPPER: return BIT_UPPER;
case RECC_PUNCT: return BIT_PUNCT;
case RECC_SPACE: return BIT_SPACE;
case RECC_GRAPH: return BIT_GRAPH;
case RECC_PRINT: return BIT_PRINT;
case RECC_ASCII: case RECC_DIGIT: case RECC_XDIGIT: case RECC_CNTRL:
case RECC_BLANK: case RECC_UNIBYTE: case RECC_ERROR: return 0;
default:
abort ();
}
}
#endif
\f
/* Filling in the work area of a range. */
/* Actually extend the space in WORK_AREA. */
static void
extend_range_table_work_area (struct range_table_work_area *work_area)
{
work_area->allocated += 16 * sizeof (int);
work_area->table = realloc (work_area->table, work_area->allocated);
}
#if 0
#ifdef emacs
/* Carefully find the ranges of codes that are equivalent
under case conversion to the range start..end when passed through
TRANSLATE. Handle the case where non-letters can come in between
two upper-case letters (which happens in Latin-1).
Also handle the case of groups of more than 2 case-equivalent chars.
The basic method is to look at consecutive characters and see
if they can form a run that can be handled as one.
Returns -1 if successful, REG_ESPACE if ran out of space. */
static int
set_image_of_range_1 (struct range_table_work_area *work_area,
re_wchar_t start, re_wchar_t end,
RE_TRANSLATE_TYPE translate)
{
/* `one_case' indicates a character, or a run of characters,
each of which is an isolate (no case-equivalents).
This includes all ASCII non-letters.
`two_case' indicates a character, or a run of characters,
each of which has two case-equivalent forms.
This includes all ASCII letters.
`strange' indicates a character that has more than one
case-equivalent. */
enum case_type {one_case, two_case, strange};
/* Describe the run that is in progress,
which the next character can try to extend.
If run_type is strange, that means there really is no run.
If run_type is one_case, then run_start...run_end is the run.
If run_type is two_case, then the run is run_start...run_end,
and the case-equivalents end at run_eqv_end. */
enum case_type run_type = strange;
int run_start, run_end, run_eqv_end;
Lisp_Object eqv_table;
if (!RE_TRANSLATE_P (translate))
{
EXTEND_RANGE_TABLE (work_area, 2);
work_area->table[work_area->used++] = (start);
work_area->table[work_area->used++] = (end);
return -1;
}
eqv_table = XCHAR_TABLE (translate)->extras[2];
for (; start <= end; start++)
{
enum case_type this_type;
int eqv = RE_TRANSLATE (eqv_table, start);
int minchar, maxchar;
/* Classify this character */
if (eqv == start)
this_type = one_case;
else if (RE_TRANSLATE (eqv_table, eqv) == start)
this_type = two_case;
else
this_type = strange;
if (start < eqv)
minchar = start, maxchar = eqv;
else
minchar = eqv, maxchar = start;
/* Can this character extend the run in progress? */
if (this_type == strange || this_type != run_type
|| !(minchar == run_end + 1
&& (run_type == two_case
? maxchar == run_eqv_end + 1 : 1)))
{
/* No, end the run.
Record each of its equivalent ranges. */
if (run_type == one_case)
{
EXTEND_RANGE_TABLE (work_area, 2);
work_area->table[work_area->used++] = run_start;
work_area->table[work_area->used++] = run_end;
}
else if (run_type == two_case)
{
EXTEND_RANGE_TABLE (work_area, 4);
work_area->table[work_area->used++] = run_start;
work_area->table[work_area->used++] = run_end;
work_area->table[work_area->used++]
= RE_TRANSLATE (eqv_table, run_start);
work_area->table[work_area->used++]
= RE_TRANSLATE (eqv_table, run_end);
}
run_type = strange;
}
if (this_type == strange)
{
/* For a strange character, add each of its equivalents, one
by one. Don't start a range. */
do
{
EXTEND_RANGE_TABLE (work_area, 2);
work_area->table[work_area->used++] = eqv;
work_area->table[work_area->used++] = eqv;
eqv = RE_TRANSLATE (eqv_table, eqv);
}
while (eqv != start);
}
/* Add this char to the run, or start a new run. */
else if (run_type == strange)
{
/* Initialize a new range. */
run_type = this_type;
run_start = start;
run_end = start;
run_eqv_end = RE_TRANSLATE (eqv_table, run_end);
}
else
{
/* Extend a running range. */
run_end = minchar;
run_eqv_end = RE_TRANSLATE (eqv_table, run_end);
}
}
/* If a run is still in progress at the end, finish it now
by recording its equivalent ranges. */
if (run_type == one_case)
{
EXTEND_RANGE_TABLE (work_area, 2);
work_area->table[work_area->used++] = run_start;
work_area->table[work_area->used++] = run_end;
}
else if (run_type == two_case)
{
EXTEND_RANGE_TABLE (work_area, 4);
work_area->table[work_area->used++] = run_start;
work_area->table[work_area->used++] = run_end;
work_area->table[work_area->used++]
= RE_TRANSLATE (eqv_table, run_start);
work_area->table[work_area->used++]
= RE_TRANSLATE (eqv_table, run_end);
}
return -1;
}
#endif /* emacs */
/* Record the image of the range start..end when passed through
TRANSLATE. This is not necessarily TRANSLATE(start)..TRANSLATE(end)
and is not even necessarily contiguous.
Normally we approximate it with the smallest contiguous range that contains
all the chars we need. However, for Latin-1 we go to extra effort
to do a better job.
This function is not called for ASCII ranges.
Returns -1 if successful, REG_ESPACE if ran out of space. */
static int
set_image_of_range (struct range_table_work_area *work_area,
re_wchar_t start, re_wchar_t end,
RE_TRANSLATE_TYPE translate)
{
re_wchar_t cmin, cmax;
#ifdef emacs
/* For Latin-1 ranges, use set_image_of_range_1
to get proper handling of ranges that include letters and nonletters.
For a range that includes the whole of Latin-1, this is not necessary.
For other character sets, we don't bother to get this right. */
if (RE_TRANSLATE_P (translate) && start < 04400
&& !(start < 04200 && end >= 04377))
{
int newend;
int tem;
newend = end;
if (newend > 04377)
newend = 04377;
tem = set_image_of_range_1 (work_area, start, newend, translate);
if (tem > 0)
return tem;
start = 04400;
if (end < 04400)
return -1;
}
#endif
EXTEND_RANGE_TABLE (work_area, 2);
work_area->table[work_area->used++] = (start);
work_area->table[work_area->used++] = (end);
cmin = -1, cmax = -1;
if (RE_TRANSLATE_P (translate))
{
int ch;
for (ch = start; ch <= end; ch++)
{
re_wchar_t c = TRANSLATE (ch);
if (! (start <= c && c <= end))
{
if (cmin == -1)
cmin = c, cmax = c;
else
{
cmin = min (cmin, c);
cmax = max (cmax, c);
}
}
}
if (cmin != -1)
{
EXTEND_RANGE_TABLE (work_area, 2);
work_area->table[work_area->used++] = (cmin);
work_area->table[work_area->used++] = (cmax);
}
}
return -1;
}
#endif /* 0 */
\f
#ifndef MATCH_MAY_ALLOCATE
/* If we cannot allocate large objects within re_match_2_internal,
we make the fail stack and register vectors global.
The fail stack, we grow to the maximum size when a regexp
is compiled.
The register vectors, we adjust in size each time we
compile a regexp, according to the number of registers it needs. */
static fail_stack_type fail_stack;
/* Size with which the following vectors are currently allocated.
That is so we can make them bigger as needed,
but never make them smaller. */
static int regs_allocated_size;
static re_char ** regstart, ** regend;
static re_char **best_regstart, **best_regend;
/* Make the register vectors big enough for NUM_REGS registers,
but don't make them smaller. */
static
regex_grow_registers (int num_regs)
{
if (num_regs > regs_allocated_size)
{
RETALLOC_IF (regstart, num_regs, re_char *);
RETALLOC_IF (regend, num_regs, re_char *);
RETALLOC_IF (best_regstart, num_regs, re_char *);
RETALLOC_IF (best_regend, num_regs, re_char *);
regs_allocated_size = num_regs;
}
}
#endif /* not MATCH_MAY_ALLOCATE */
\f
static boolean group_in_compile_stack (compile_stack_type compile_stack,
regnum_t regnum);
/* `regex_compile' compiles PATTERN (of length SIZE) according to SYNTAX.
Returns one of error codes defined in `regex.h', or zero for success.
If WHITESPACE_REGEXP is given (only #ifdef emacs), it is used instead of
a space character in PATTERN.
Assumes the `allocated' (and perhaps `buffer') and `translate'
fields are set in BUFP on entry.
If it succeeds, results are put in BUFP (if it returns an error, the
contents of BUFP are undefined):
`buffer' is the compiled pattern;
`syntax' is set to SYNTAX;
`used' is set to the length of the compiled pattern;
`fastmap_accurate' is zero;
`re_nsub' is the number of subexpressions in PATTERN;
`not_bol' and `not_eol' are zero;
The `fastmap' field is neither examined nor set. */
/* Insert the `jump' from the end of last alternative to "here".
The space for the jump has already been allocated. */
#define FIXUP_ALT_JUMP() \
do { \
if (fixup_alt_jump) \
STORE_JUMP (jump, fixup_alt_jump, b); \
} while (0)
/* Return, freeing storage we allocated. */
#define FREE_STACK_RETURN(value) \
do { \
FREE_RANGE_TABLE_WORK_AREA (range_table_work); \
free (compile_stack.stack); \
return value; \
} while (0)
static reg_errcode_t
regex_compile (const_re_char *pattern, size_t size,
#ifdef emacs
# define syntax RE_SYNTAX_EMACS
bool posix_backtracking,
const char *whitespace_regexp,
#else
reg_syntax_t syntax,
# define posix_backtracking (!(syntax & RE_NO_POSIX_BACKTRACKING))
#endif
struct re_pattern_buffer *bufp)
{
/* We fetch characters from PATTERN here. */
register re_wchar_t c, c1;
/* Points to the end of the buffer, where we should append. */
register unsigned char *b;
/* Keeps track of unclosed groups. */
compile_stack_type compile_stack;
/* Points to the current (ending) position in the pattern. */
#ifdef AIX
/* `const' makes AIX compiler fail. */
unsigned char *p = pattern;
#else
re_char *p = pattern;
#endif
re_char *pend = pattern + size;
/* How to translate the characters in the pattern. */
RE_TRANSLATE_TYPE translate = bufp->translate;
/* Address of the count-byte of the most recently inserted `exactn'
command. This makes it possible to tell if a new exact-match
character can be added to that command or if the character requires
a new `exactn' command. */
unsigned char *pending_exact = 0;
/* Address of start of the most recently finished expression.
This tells, e.g., postfix * where to find the start of its
operand. Reset at the beginning of groups and alternatives. */
unsigned char *laststart = 0;
/* Address of beginning of regexp, or inside of last group. */
unsigned char *begalt;
/* Place in the uncompiled pattern (i.e., the {) to
which to go back if the interval is invalid. */
re_char *beg_interval;
/* Address of the place where a forward jump should go to the end of
the containing expression. Each alternative of an `or' -- except the
last -- ends with a forward jump of this sort. */
unsigned char *fixup_alt_jump = 0;
/* Work area for range table of charset. */
struct range_table_work_area range_table_work;
/* If the object matched can contain multibyte characters. */
const boolean multibyte = RE_MULTIBYTE_P (bufp);
#ifdef emacs
/* Nonzero if we have pushed down into a subpattern. */
int in_subpattern = 0;
/* These hold the values of p, pattern, and pend from the main
pattern when we have pushed into a subpattern. */
re_char *main_p;
re_char *main_pattern;
re_char *main_pend;
#endif
#ifdef DEBUG
debug++;
DEBUG_PRINT ("\nCompiling pattern: ");
if (debug > 0)
{
unsigned debug_count;
for (debug_count = 0; debug_count < size; debug_count++)
putchar (pattern[debug_count]);
putchar ('\n');
}
#endif /* DEBUG */
/* Initialize the compile stack. */
compile_stack.stack = TALLOC (INIT_COMPILE_STACK_SIZE, compile_stack_elt_t);
if (compile_stack.stack == NULL)
return REG_ESPACE;
compile_stack.size = INIT_COMPILE_STACK_SIZE;
compile_stack.avail = 0;
range_table_work.table = 0;
range_table_work.allocated = 0;
/* Initialize the pattern buffer. */
#ifndef emacs
bufp->syntax = syntax;
#endif
bufp->fastmap_accurate = 0;
bufp->not_bol = bufp->not_eol = 0;
bufp->used_syntax = 0;
/* Set `used' to zero, so that if we return an error, the pattern
printer (for debugging) will think there's no pattern. We reset it
at the end. */
bufp->used = 0;
/* Always count groups, whether or not bufp->no_sub is set. */
bufp->re_nsub = 0;
#if !defined emacs && !defined SYNTAX_TABLE
/* Initialize the syntax table. */
init_syntax_once ();
#endif
if (bufp->allocated == 0)
{
if (bufp->buffer)
{ /* If zero allocated, but buffer is non-null, try to realloc
enough space. This loses if buffer's address is bogus, but
that is the user's responsibility. */
RETALLOC (bufp->buffer, INIT_BUF_SIZE, unsigned char);
}
else
{ /* Caller did not allocate a buffer. Do it for them. */
bufp->buffer = TALLOC (INIT_BUF_SIZE, unsigned char);
}
if (!bufp->buffer) FREE_STACK_RETURN (REG_ESPACE);
bufp->allocated = INIT_BUF_SIZE;
}
begalt = b = bufp->buffer;
/* Loop through the uncompiled pattern until we're at the end. */
while (1)
{
if (p == pend)
{
#ifdef emacs
/* If this is the end of an included regexp,
pop back to the main regexp and try again. */
if (in_subpattern)
{
in_subpattern = 0;
pattern = main_pattern;
p = main_p;
pend = main_pend;
continue;
}
#endif
/* If this is the end of the main regexp, we are done. */
break;
}
PATFETCH (c);
switch (c)
{
#ifdef emacs
case ' ':
{
re_char *p1 = p;
/* If there's no special whitespace regexp, treat
spaces normally. And don't try to do this recursively. */
if (!whitespace_regexp || in_subpattern)
goto normal_char;
/* Peek past following spaces. */
while (p1 != pend)
{
if (*p1 != ' ')
break;
p1++;
}
/* If the spaces are followed by a repetition op,
treat them normally. */
if (p1 != pend
&& (*p1 == '*' || *p1 == '+' || *p1 == '?'
|| (*p1 == '\\' && p1 + 1 != pend && p1[1] == '{')))
goto normal_char;
/* Replace the spaces with the whitespace regexp. */
in_subpattern = 1;
main_p = p1;
main_pend = pend;
main_pattern = pattern;
p = pattern = (re_char *) whitespace_regexp;
pend = p + strlen (whitespace_regexp);
break;
}
#endif
case '^':
{
if ( /* If at start of pattern, it's an operator. */
p == pattern + 1
/* If context independent, it's an operator. */
|| syntax & RE_CONTEXT_INDEP_ANCHORS
/* Otherwise, depends on what's come before. */
|| at_begline_loc_p (pattern, p, syntax))
BUF_PUSH ((syntax & RE_NO_NEWLINE_ANCHOR) ? begbuf : begline);
else
goto normal_char;
}
break;
case '$':
{
if ( /* If at end of pattern, it's an operator. */
p == pend
/* If context independent, it's an operator. */
|| syntax & RE_CONTEXT_INDEP_ANCHORS
/* Otherwise, depends on what's next. */
|| at_endline_loc_p (p, pend, syntax))
BUF_PUSH ((syntax & RE_NO_NEWLINE_ANCHOR) ? endbuf : endline);
else
goto normal_char;
}
break;
case '+':
case '?':
if ((syntax & RE_BK_PLUS_QM)
|| (syntax & RE_LIMITED_OPS))
goto normal_char;
handle_plus:
case '*':
/* If there is no previous pattern... */
if (!laststart)
{
if (syntax & RE_CONTEXT_INVALID_OPS)
FREE_STACK_RETURN (REG_BADRPT);
else if (!(syntax & RE_CONTEXT_INDEP_OPS))
goto normal_char;
}
{
/* 1 means zero (many) matches is allowed. */
boolean zero_times_ok = 0, many_times_ok = 0;
boolean greedy = 1;
/* If there is a sequence of repetition chars, collapse it
down to just one (the right one). We can't combine
interval operators with these because of, e.g., `a{2}*',
which should only match an even number of `a's. */
for (;;)
{
if ((syntax & RE_FRUGAL)
&& c == '?' && (zero_times_ok || many_times_ok))
greedy = 0;
else
{
zero_times_ok |= c != '+';
many_times_ok |= c != '?';
}
if (p == pend)
break;
else if (*p == '*'
|| (!(syntax & RE_BK_PLUS_QM)
&& (*p == '+' || *p == '?')))
;
else if (syntax & RE_BK_PLUS_QM && *p == '\\')
{
if (p+1 == pend)
FREE_STACK_RETURN (REG_EESCAPE);
if (p[1] == '+' || p[1] == '?')
PATFETCH (c); /* Gobble up the backslash. */
else
break;
}
else
break;
/* If we get here, we found another repeat character. */
PATFETCH (c);
}
/* Star, etc. applied to an empty pattern is equivalent
to an empty pattern. */
if (!laststart || laststart == b)
break;
/* Now we know whether or not zero matches is allowed
and also whether or not two or more matches is allowed. */
if (greedy)
{
if (many_times_ok)
{
boolean simple = skip_one_char (laststart) == b;
size_t startoffset = 0;
re_opcode_t ofj =
/* Check if the loop can match the empty string. */
(simple || !analyze_first (laststart, b, NULL, 0))
? on_failure_jump : on_failure_jump_loop;
assert (skip_one_char (laststart) <= b);
if (!zero_times_ok && simple)
{ /* Since simple * loops can be made faster by using
on_failure_keep_string_jump, we turn simple P+
into PP* if P is simple. */
unsigned char *p1, *p2;
startoffset = b - laststart;
GET_BUFFER_SPACE (startoffset);
p1 = b; p2 = laststart;
while (p2 < p1)
*b++ = *p2++;
zero_times_ok = 1;
}
GET_BUFFER_SPACE (6);
if (!zero_times_ok)
/* A + loop. */
STORE_JUMP (ofj, b, b + 6);
else
/* Simple * loops can use on_failure_keep_string_jump
depending on what follows. But since we don't know
that yet, we leave the decision up to
on_failure_jump_smart. */
INSERT_JUMP (simple ? on_failure_jump_smart : ofj,
laststart + startoffset, b + 6);
b += 3;
STORE_JUMP (jump, b, laststart + startoffset);
b += 3;
}
else
{
/* A simple ? pattern. */
assert (zero_times_ok);
GET_BUFFER_SPACE (3);
INSERT_JUMP (on_failure_jump, laststart, b + 3);
b += 3;
}
}
else /* not greedy */
{ /* I wish the greedy and non-greedy cases could be merged. */
GET_BUFFER_SPACE (7); /* We might use less. */
if (many_times_ok)
{
boolean emptyp = analyze_first (laststart, b, NULL, 0);
/* The non-greedy multiple match looks like
a repeat..until: we only need a conditional jump
at the end of the loop. */
if (emptyp) BUF_PUSH (no_op);
STORE_JUMP (emptyp ? on_failure_jump_nastyloop
: on_failure_jump, b, laststart);
b += 3;
if (zero_times_ok)
{
/* The repeat...until naturally matches one or more.
To also match zero times, we need to first jump to
the end of the loop (its conditional jump). */
INSERT_JUMP (jump, laststart, b);
b += 3;
}
}
else
{
/* non-greedy a?? */
INSERT_JUMP (jump, laststart, b + 3);
b += 3;
INSERT_JUMP (on_failure_jump, laststart, laststart + 6);
b += 3;
}
}
}
pending_exact = 0;
break;
case '.':
laststart = b;
BUF_PUSH (anychar);
break;
case '[':
{
re_char *p1;
CLEAR_RANGE_TABLE_WORK_USED (range_table_work);
if (p == pend) FREE_STACK_RETURN (REG_EBRACK);
/* Ensure that we have enough space to push a charset: the
opcode, the length count, and the bitset; 34 bytes in all. */
GET_BUFFER_SPACE (34);
laststart = b;
/* We test `*p == '^' twice, instead of using an if
statement, so we only need one BUF_PUSH. */
BUF_PUSH (*p == '^' ? charset_not : charset);
if (*p == '^')
p++;
/* Remember the first position in the bracket expression. */
p1 = p;
/* Push the number of bytes in the bitmap. */
BUF_PUSH ((1 << BYTEWIDTH) / BYTEWIDTH);
/* Clear the whole map. */
memset (b, 0, (1 << BYTEWIDTH) / BYTEWIDTH);
/* charset_not matches newline according to a syntax bit. */
if ((re_opcode_t) b[-2] == charset_not
&& (syntax & RE_HAT_LISTS_NOT_NEWLINE))
SET_LIST_BIT ('\n');
/* Read in characters and ranges, setting map bits. */
for (;;)
{
boolean escaped_char = false;
const unsigned char *p2 = p;
re_wctype_t cc;
re_wchar_t ch;
if (p == pend) FREE_STACK_RETURN (REG_EBRACK);
/* See if we're at the beginning of a possible character
class. */
if (syntax & RE_CHAR_CLASSES &&
(cc = re_wctype_parse(&p, pend - p)) != -1)
{
if (cc == 0)
FREE_STACK_RETURN (REG_ECTYPE);
if (p == pend)
FREE_STACK_RETURN (REG_EBRACK);
#ifndef emacs
for (ch = 0; ch < (1 << BYTEWIDTH); ++ch)
if (re_iswctype (btowc (ch), cc))
{
c = TRANSLATE (ch);
if (c < (1 << BYTEWIDTH))
SET_LIST_BIT (c);
}
#else /* emacs */
/* Most character classes in a multibyte match just set
a flag. Exceptions are is_blank, is_digit, is_cntrl, and
is_xdigit, since they can only match ASCII characters.
We don't need to handle them for multibyte. */
/* Setup the gl_state object to its buffer-defined value.
This hardcodes the buffer-global syntax-table for ASCII
chars, while the other chars will obey syntax-table
properties. It's not ideal, but it's the way it's been
done until now. */
SETUP_BUFFER_SYNTAX_TABLE ();
for (ch = 0; ch < 256; ++ch)
{
c = RE_CHAR_TO_MULTIBYTE (ch);
if (! CHAR_BYTE8_P (c)
&& re_iswctype (c, cc))
{
SET_LIST_BIT (ch);
c1 = TRANSLATE (c);
if (c1 == c)
continue;
if (ASCII_CHAR_P (c1))
SET_LIST_BIT (c1);
else if ((c1 = RE_CHAR_TO_UNIBYTE (c1)) >= 0)
SET_LIST_BIT (c1);
}
}
SET_RANGE_TABLE_WORK_AREA_BIT
(range_table_work, re_wctype_to_bit (cc));
#endif /* emacs */
/* In most cases the matching rule for char classes only
uses the syntax table for multibyte chars, so that the
content of the syntax-table is not hardcoded in the
range_table. SPACE and WORD are the two exceptions. */
if ((1 << cc) & ((1 << RECC_SPACE) | (1 << RECC_WORD)))
bufp->used_syntax = 1;
/* Repeat the loop. */
continue;
}
/* Don't translate yet. The range TRANSLATE(X..Y) cannot
always be determined from TRANSLATE(X) and TRANSLATE(Y)
So the translation is done later in a loop. Example:
(let ((case-fold-search t)) (string-match "[A-_]" "A")) */
PATFETCH (c);
/* \ might escape characters inside [...] and [^...]. */
if ((syntax & RE_BACKSLASH_ESCAPE_IN_LISTS) && c == '\\')
{
if (p == pend) FREE_STACK_RETURN (REG_EESCAPE);
PATFETCH (c);
escaped_char = true;
}
else
{
/* Could be the end of the bracket expression. If it's
not (i.e., when the bracket expression is `[]' so
far), the ']' character bit gets set way below. */
if (c == ']' && p2 != p1)
break;
}
if (p < pend && p[0] == '-' && p[1] != ']')
{
/* Discard the `-'. */
PATFETCH (c1);
/* Fetch the character which ends the range. */
PATFETCH (c1);
#ifdef emacs
if (CHAR_BYTE8_P (c1)
&& ! ASCII_CHAR_P (c) && ! CHAR_BYTE8_P (c))
/* Treat the range from a multibyte character to
raw-byte character as empty. */
c = c1 + 1;
#endif /* emacs */
}
else
/* Range from C to C. */
c1 = c;
if (c > c1)
{
if (syntax & RE_NO_EMPTY_RANGES)
FREE_STACK_RETURN (REG_ERANGEX);
/* Else, repeat the loop. */
}
else
{
#ifndef emacs
/* Set the range into bitmap */
for (; c <= c1; c++)
{
ch = TRANSLATE (c);
if (ch < (1 << BYTEWIDTH))
SET_LIST_BIT (ch);
}
#else /* emacs */
if (c < 128)
{
ch = min (127, c1);
SETUP_ASCII_RANGE (range_table_work, c, ch);
c = ch + 1;
if (CHAR_BYTE8_P (c1))
c = BYTE8_TO_CHAR (128);
}
if (c <= c1)
{
if (CHAR_BYTE8_P (c))
{
c = CHAR_TO_BYTE8 (c);
c1 = CHAR_TO_BYTE8 (c1);
for (; c <= c1; c++)
SET_LIST_BIT (c);
}
else if (multibyte)
{
SETUP_MULTIBYTE_RANGE (range_table_work, c, c1);
}
else
{
SETUP_UNIBYTE_RANGE (range_table_work, c, c1);
}
}
#endif /* emacs */
}
}
/* Discard any (non)matching list bytes that are all 0 at the
end of the map. Decrease the map-length byte too. */
while ((int) b[-1] > 0 && b[b[-1] - 1] == 0)
b[-1]--;
b += b[-1];
/* Build real range table from work area. */
if (RANGE_TABLE_WORK_USED (range_table_work)
|| RANGE_TABLE_WORK_BITS (range_table_work))
{
int i;
int used = RANGE_TABLE_WORK_USED (range_table_work);
/* Allocate space for COUNT + RANGE_TABLE. Needs two
bytes for flags, two for COUNT, and three bytes for
each character. */
GET_BUFFER_SPACE (4 + used * 3);
/* Indicate the existence of range table. */
laststart[1] |= 0x80;
/* Store the character class flag bits into the range table.
If not in emacs, these flag bits are always 0. */
*b++ = RANGE_TABLE_WORK_BITS (range_table_work) & 0xff;
*b++ = RANGE_TABLE_WORK_BITS (range_table_work) >> 8;
STORE_NUMBER_AND_INCR (b, used / 2);
for (i = 0; i < used; i++)
STORE_CHARACTER_AND_INCR
(b, RANGE_TABLE_WORK_ELT (range_table_work, i));
}
}
break;
case '(':
if (syntax & RE_NO_BK_PARENS)
goto handle_open;
else
goto normal_char;
case ')':
if (syntax & RE_NO_BK_PARENS)
goto handle_close;
else
goto normal_char;
case '\n':
if (syntax & RE_NEWLINE_ALT)
goto handle_alt;
else
goto normal_char;
case '|':
if (syntax & RE_NO_BK_VBAR)
goto handle_alt;
else
goto normal_char;
case '{':
if (syntax & RE_INTERVALS && syntax & RE_NO_BK_BRACES)
goto handle_interval;
else
goto normal_char;
case '\\':
if (p == pend) FREE_STACK_RETURN (REG_EESCAPE);
/* Do not translate the character after the \, so that we can
distinguish, e.g., \B from \b, even if we normally would
translate, e.g., B to b. */
PATFETCH (c);
switch (c)
{
case '(':
if (syntax & RE_NO_BK_PARENS)
goto normal_backslash;
handle_open:
{
int shy = 0;
regnum_t regnum = 0;
if (p+1 < pend)
{
/* Look for a special (?...) construct */
if ((syntax & RE_SHY_GROUPS) && *p == '?')
{
PATFETCH (c); /* Gobble up the '?'. */
while (!shy)
{
PATFETCH (c);
switch (c)
{
case ':': shy = 1; break;
case '0':
/* An explicitly specified regnum must start
with non-0. */
if (regnum == 0)
FREE_STACK_RETURN (REG_BADPAT);
case '1': case '2': case '3': case '4':
case '5': case '6': case '7': case '8': case '9':
regnum = 10*regnum + (c - '0'); break;
default:
/* Only (?:...) is supported right now. */
FREE_STACK_RETURN (REG_BADPAT);
}
}
}
}
if (!shy)
regnum = ++bufp->re_nsub;
else if (regnum)
{ /* It's actually not shy, but explicitly numbered. */
shy = 0;
if (regnum > bufp->re_nsub)
bufp->re_nsub = regnum;
else if (regnum > bufp->re_nsub
/* Ideally, we'd want to check that the specified
group can't have matched (i.e. all subgroups
using the same regnum are in other branches of
OR patterns), but we don't currently keep track
of enough info to do that easily. */
|| group_in_compile_stack (compile_stack, regnum))
FREE_STACK_RETURN (REG_BADPAT);
}
else
/* It's really shy. */
regnum = - bufp->re_nsub;
if (COMPILE_STACK_FULL)
{
RETALLOC (compile_stack.stack, compile_stack.size << 1,
compile_stack_elt_t);
if (compile_stack.stack == NULL) return REG_ESPACE;
compile_stack.size <<= 1;
}
/* These are the values to restore when we hit end of this
group. They are all relative offsets, so that if the
whole pattern moves because of realloc, they will still
be valid. */
COMPILE_STACK_TOP.begalt_offset = begalt - bufp->buffer;
COMPILE_STACK_TOP.fixup_alt_jump
= fixup_alt_jump ? fixup_alt_jump - bufp->buffer + 1 : 0;
COMPILE_STACK_TOP.laststart_offset = b - bufp->buffer;
COMPILE_STACK_TOP.regnum = regnum;
/* Do not push a start_memory for groups beyond the last one
we can represent in the compiled pattern. */
if (regnum <= MAX_REGNUM && regnum > 0)
BUF_PUSH_2 (start_memory, regnum);
compile_stack.avail++;
fixup_alt_jump = 0;
laststart = 0;
begalt = b;
/* If we've reached MAX_REGNUM groups, then this open
won't actually generate any code, so we'll have to
clear pending_exact explicitly. */
pending_exact = 0;
break;
}
case ')':
if (syntax & RE_NO_BK_PARENS) goto normal_backslash;
if (COMPILE_STACK_EMPTY)
{
if (syntax & RE_UNMATCHED_RIGHT_PAREN_ORD)
goto normal_backslash;
else
FREE_STACK_RETURN (REG_ERPAREN);
}
handle_close:
FIXUP_ALT_JUMP ();
/* See similar code for backslashed left paren above. */
if (COMPILE_STACK_EMPTY)
{
if (syntax & RE_UNMATCHED_RIGHT_PAREN_ORD)
goto normal_char;
else
FREE_STACK_RETURN (REG_ERPAREN);
}
/* Since we just checked for an empty stack above, this
``can't happen''. */
assert (compile_stack.avail != 0);
{
/* We don't just want to restore into `regnum', because
later groups should continue to be numbered higher,
as in `(ab)c(de)' -- the second group is #2. */
regnum_t regnum;
compile_stack.avail--;
begalt = bufp->buffer + COMPILE_STACK_TOP.begalt_offset;
fixup_alt_jump
= COMPILE_STACK_TOP.fixup_alt_jump
? bufp->buffer + COMPILE_STACK_TOP.fixup_alt_jump - 1
: 0;
laststart = bufp->buffer + COMPILE_STACK_TOP.laststart_offset;
regnum = COMPILE_STACK_TOP.regnum;
/* If we've reached MAX_REGNUM groups, then this open
won't actually generate any code, so we'll have to
clear pending_exact explicitly. */
pending_exact = 0;
/* We're at the end of the group, so now we know how many
groups were inside this one. */
if (regnum <= MAX_REGNUM && regnum > 0)
BUF_PUSH_2 (stop_memory, regnum);
}
break;
case '|': /* `\|'. */
if (syntax & RE_LIMITED_OPS || syntax & RE_NO_BK_VBAR)
goto normal_backslash;
handle_alt:
if (syntax & RE_LIMITED_OPS)
goto normal_char;
/* Insert before the previous alternative a jump which
jumps to this alternative if the former fails. */
GET_BUFFER_SPACE (3);
INSERT_JUMP (on_failure_jump, begalt, b + 6);
pending_exact = 0;
b += 3;
/* The alternative before this one has a jump after it
which gets executed if it gets matched. Adjust that
jump so it will jump to this alternative's analogous
jump (put in below, which in turn will jump to the next
(if any) alternative's such jump, etc.). The last such
jump jumps to the correct final destination. A picture:
_____ _____
| | | |
| v | v
a | b | c
If we are at `b', then fixup_alt_jump right now points to a
three-byte space after `a'. We'll put in the jump, set
fixup_alt_jump to right after `b', and leave behind three
bytes which we'll fill in when we get to after `c'. */
FIXUP_ALT_JUMP ();
/* Mark and leave space for a jump after this alternative,
to be filled in later either by next alternative or
when know we're at the end of a series of alternatives. */
fixup_alt_jump = b;
GET_BUFFER_SPACE (3);
b += 3;
laststart = 0;
begalt = b;
break;
case '{':
/* If \{ is a literal. */
if (!(syntax & RE_INTERVALS)
/* If we're at `\{' and it's not the open-interval
operator. */
|| (syntax & RE_NO_BK_BRACES))
goto normal_backslash;
handle_interval:
{
/* If got here, then the syntax allows intervals. */
/* At least (most) this many matches must be made. */
int lower_bound = 0, upper_bound = -1;
beg_interval = p;
GET_INTERVAL_COUNT (lower_bound);
if (c == ',')
GET_INTERVAL_COUNT (upper_bound);
else
/* Interval such as `{1}' => match exactly once. */
upper_bound = lower_bound;
if (lower_bound < 0
|| (0 <= upper_bound && upper_bound < lower_bound))
FREE_STACK_RETURN (REG_BADBR);
if (!(syntax & RE_NO_BK_BRACES))
{
if (c != '\\')
FREE_STACK_RETURN (REG_BADBR);
if (p == pend)
FREE_STACK_RETURN (REG_EESCAPE);
PATFETCH (c);
}
if (c != '}')
FREE_STACK_RETURN (REG_BADBR);
/* We just parsed a valid interval. */
/* If it's invalid to have no preceding re. */
if (!laststart)
{
if (syntax & RE_CONTEXT_INVALID_OPS)
FREE_STACK_RETURN (REG_BADRPT);
else if (syntax & RE_CONTEXT_INDEP_OPS)
laststart = b;
else
goto unfetch_interval;
}
if (upper_bound == 0)
/* If the upper bound is zero, just drop the sub pattern
altogether. */
b = laststart;
else if (lower_bound == 1 && upper_bound == 1)
/* Just match it once: nothing to do here. */
;
/* Otherwise, we have a nontrivial interval. When
we're all done, the pattern will look like:
set_number_at <jump count> <upper bound>
set_number_at <succeed_n count> <lower bound>
succeed_n <after jump addr> <succeed_n count>
<body of loop>
jump_n <succeed_n addr> <jump count>
(The upper bound and `jump_n' are omitted if
`upper_bound' is 1, though.) */
else
{ /* If the upper bound is > 1, we need to insert
more at the end of the loop. */
unsigned int nbytes = (upper_bound < 0 ? 3
: upper_bound > 1 ? 5 : 0);
unsigned int startoffset = 0;
GET_BUFFER_SPACE (20); /* We might use less. */
if (lower_bound == 0)
{
/* A succeed_n that starts with 0 is really a
a simple on_failure_jump_loop. */
INSERT_JUMP (on_failure_jump_loop, laststart,
b + 3 + nbytes);
b += 3;
}
else
{
/* Initialize lower bound of the `succeed_n', even
though it will be set during matching by its
attendant `set_number_at' (inserted next),
because `re_compile_fastmap' needs to know.
Jump to the `jump_n' we might insert below. */
INSERT_JUMP2 (succeed_n, laststart,
b + 5 + nbytes,
lower_bound);
b += 5;
/* Code to initialize the lower bound. Insert
before the `succeed_n'. The `5' is the last two
bytes of this `set_number_at', plus 3 bytes of
the following `succeed_n'. */
insert_op2 (set_number_at, laststart, 5, lower_bound, b);
b += 5;
startoffset += 5;
}
if (upper_bound < 0)
{
/* A negative upper bound stands for infinity,
in which case it degenerates to a plain jump. */
STORE_JUMP (jump, b, laststart + startoffset);
b += 3;
}
else if (upper_bound > 1)
{ /* More than one repetition is allowed, so
append a backward jump to the `succeed_n'
that starts this interval.
When we've reached this during matching,
we'll have matched the interval once, so
jump back only `upper_bound - 1' times. */
STORE_JUMP2 (jump_n, b, laststart + startoffset,
upper_bound - 1);
b += 5;
/* The location we want to set is the second
parameter of the `jump_n'; that is `b-2' as
an absolute address. `laststart' will be
the `set_number_at' we're about to insert;
`laststart+3' the number to set, the source
for the relative address. But we are
inserting into the middle of the pattern --
so everything is getting moved up by 5.
Conclusion: (b - 2) - (laststart + 3) + 5,
i.e., b - laststart.
We insert this at the beginning of the loop
so that if we fail during matching, we'll
reinitialize the bounds. */
insert_op2 (set_number_at, laststart, b - laststart,
upper_bound - 1, b);
b += 5;
}
}
pending_exact = 0;
beg_interval = NULL;
}
break;
unfetch_interval:
/* If an invalid interval, match the characters as literals. */
assert (beg_interval);
p = beg_interval;
beg_interval = NULL;
/* normal_char and normal_backslash need `c'. */
c = '{';
if (!(syntax & RE_NO_BK_BRACES))
{
assert (p > pattern && p[-1] == '\\');
goto normal_backslash;
}
else
goto normal_char;
#ifdef emacs
case '=':
laststart = b;
BUF_PUSH (at_dot);
break;
case 's':
laststart = b;
PATFETCH (c);
BUF_PUSH_2 (syntaxspec, syntax_spec_code[c]);
break;
case 'S':
laststart = b;
PATFETCH (c);
BUF_PUSH_2 (notsyntaxspec, syntax_spec_code[c]);
break;
case 'c':
laststart = b;
PATFETCH (c);
BUF_PUSH_2 (categoryspec, c);
break;
case 'C':
laststart = b;
PATFETCH (c);
BUF_PUSH_2 (notcategoryspec, c);
break;
#endif /* emacs */
case 'w':
if (syntax & RE_NO_GNU_OPS)
goto normal_char;
laststart = b;
BUF_PUSH_2 (syntaxspec, Sword);
break;
case 'W':
if (syntax & RE_NO_GNU_OPS)
goto normal_char;
laststart = b;
BUF_PUSH_2 (notsyntaxspec, Sword);
break;
case '<':
if (syntax & RE_NO_GNU_OPS)
goto normal_char;
laststart = b;
BUF_PUSH (wordbeg);
break;
case '>':
if (syntax & RE_NO_GNU_OPS)
goto normal_char;
laststart = b;
BUF_PUSH (wordend);
break;
case '_':
if (syntax & RE_NO_GNU_OPS)
goto normal_char;
laststart = b;
PATFETCH (c);
if (c == '<')
BUF_PUSH (symbeg);
else if (c == '>')
BUF_PUSH (symend);
else
FREE_STACK_RETURN (REG_BADPAT);
break;
case 'b':
if (syntax & RE_NO_GNU_OPS)
goto normal_char;
BUF_PUSH (wordbound);
break;
case 'B':
if (syntax & RE_NO_GNU_OPS)
goto normal_char;
BUF_PUSH (notwordbound);
break;
case '`':
if (syntax & RE_NO_GNU_OPS)
goto normal_char;
BUF_PUSH (begbuf);
break;
case '\'':
if (syntax & RE_NO_GNU_OPS)
goto normal_char;
BUF_PUSH (endbuf);
break;
case '1': case '2': case '3': case '4': case '5':
case '6': case '7': case '8': case '9':
{
regnum_t reg;
if (syntax & RE_NO_BK_REFS)
goto normal_backslash;
reg = c - '0';
if (reg > bufp->re_nsub || reg < 1
/* Can't back reference to a subexp before its end. */
|| group_in_compile_stack (compile_stack, reg))
FREE_STACK_RETURN (REG_ESUBREG);
laststart = b;
BUF_PUSH_2 (duplicate, reg);
}
break;
case '+':
case '?':
if (syntax & RE_BK_PLUS_QM)
goto handle_plus;
else
goto normal_backslash;
default:
normal_backslash:
/* You might think it would be useful for \ to mean
not to translate; but if we don't translate it
it will never match anything. */
goto normal_char;
}
break;
default:
/* Expects the character in `c'. */
normal_char:
/* If no exactn currently being built. */
if (!pending_exact
/* If last exactn not at current position. */
|| pending_exact + *pending_exact + 1 != b
/* We have only one byte following the exactn for the count. */
|| *pending_exact >= (1 << BYTEWIDTH) - MAX_MULTIBYTE_LENGTH
/* If followed by a repetition operator. */
|| (p != pend && (*p == '*' || *p == '^'))
|| ((syntax & RE_BK_PLUS_QM)
? p + 1 < pend && *p == '\\' && (p[1] == '+' || p[1] == '?')
: p != pend && (*p == '+' || *p == '?'))
|| ((syntax & RE_INTERVALS)
&& ((syntax & RE_NO_BK_BRACES)
? p != pend && *p == '{'
: p + 1 < pend && p[0] == '\\' && p[1] == '{')))
{
/* Start building a new exactn. */
laststart = b;
BUF_PUSH_2 (exactn, 0);
pending_exact = b - 1;
}
GET_BUFFER_SPACE (MAX_MULTIBYTE_LENGTH);
{
int len;
if (multibyte)
{
c = TRANSLATE (c);
len = CHAR_STRING (c, b);
b += len;
}
else
{
c1 = RE_CHAR_TO_MULTIBYTE (c);
if (! CHAR_BYTE8_P (c1))
{
re_wchar_t c2 = TRANSLATE (c1);
if (c1 != c2 && (c1 = RE_CHAR_TO_UNIBYTE (c2)) >= 0)
c = c1;
}
*b++ = c;
len = 1;
}
(*pending_exact) += len;
}
break;
} /* switch (c) */
} /* while p != pend */
/* Through the pattern now. */
FIXUP_ALT_JUMP ();
if (!COMPILE_STACK_EMPTY)
FREE_STACK_RETURN (REG_EPAREN);
/* If we don't want backtracking, force success
the first time we reach the end of the compiled pattern. */
if (!posix_backtracking)
BUF_PUSH (succeed);
/* We have succeeded; set the length of the buffer. */
bufp->used = b - bufp->buffer;
#ifdef DEBUG
if (debug > 0)
{
re_compile_fastmap (bufp);
DEBUG_PRINT ("\nCompiled pattern: \n");
print_compiled_pattern (bufp);
}
debug--;
#endif /* DEBUG */
#ifndef MATCH_MAY_ALLOCATE
/* Initialize the failure stack to the largest possible stack. This
isn't necessary unless we're trying to avoid calling alloca in
the search and match routines. */
{
int num_regs = bufp->re_nsub + 1;
if (fail_stack.size < re_max_failures * TYPICAL_FAILURE_SIZE)
{
fail_stack.size = re_max_failures * TYPICAL_FAILURE_SIZE;
falk_stack.stack = realloc (fail_stack.stack,
fail_stack.size * sizeof *falk_stack.stack);
}
regex_grow_registers (num_regs);
}
#endif /* not MATCH_MAY_ALLOCATE */
FREE_STACK_RETURN (REG_NOERROR);
#ifdef emacs
# undef syntax
#else
# undef posix_backtracking
#endif
} /* regex_compile */
\f
/* Subroutines for `regex_compile'. */
/* Store OP at LOC followed by two-byte integer parameter ARG. */
static void
store_op1 (re_opcode_t op, unsigned char *loc, int arg)
{
*loc = (unsigned char) op;
STORE_NUMBER (loc + 1, arg);
}
/* Like `store_op1', but for two two-byte parameters ARG1 and ARG2. */
static void
store_op2 (re_opcode_t op, unsigned char *loc, int arg1, int arg2)
{
*loc = (unsigned char) op;
STORE_NUMBER (loc + 1, arg1);
STORE_NUMBER (loc + 3, arg2);
}
/* Copy the bytes from LOC to END to open up three bytes of space at LOC
for OP followed by two-byte integer parameter ARG. */
static void
insert_op1 (re_opcode_t op, unsigned char *loc, int arg, unsigned char *end)
{
register unsigned char *pfrom = end;
register unsigned char *pto = end + 3;
while (pfrom != loc)
*--pto = *--pfrom;
store_op1 (op, loc, arg);
}
/* Like `insert_op1', but for two two-byte parameters ARG1 and ARG2. */
static void
insert_op2 (re_opcode_t op, unsigned char *loc, int arg1, int arg2, unsigned char *end)
{
register unsigned char *pfrom = end;
register unsigned char *pto = end + 5;
while (pfrom != loc)
*--pto = *--pfrom;
store_op2 (op, loc, arg1, arg2);
}
/* P points to just after a ^ in PATTERN. Return true if that ^ comes
after an alternative or a begin-subexpression. We assume there is at
least one character before the ^. */
static boolean
at_begline_loc_p (const_re_char *pattern, const_re_char *p, reg_syntax_t syntax)
{
re_char *prev = p - 2;
boolean odd_backslashes;
/* After a subexpression? */
if (*prev == '(')
odd_backslashes = (syntax & RE_NO_BK_PARENS) == 0;
/* After an alternative? */
else if (*prev == '|')
odd_backslashes = (syntax & RE_NO_BK_VBAR) == 0;
/* After a shy subexpression? */
else if (*prev == ':' && (syntax & RE_SHY_GROUPS))
{
/* Skip over optional regnum. */
while (prev - 1 >= pattern && prev[-1] >= '0' && prev[-1] <= '9')
--prev;
if (!(prev - 2 >= pattern
&& prev[-1] == '?' && prev[-2] == '('))
return false;
prev -= 2;
odd_backslashes = (syntax & RE_NO_BK_PARENS) == 0;
}
else
return false;
/* Count the number of preceding backslashes. */
p = prev;
while (prev - 1 >= pattern && prev[-1] == '\\')
--prev;
return (p - prev) & odd_backslashes;
}
/* The dual of at_begline_loc_p. This one is for $. We assume there is
at least one character after the $, i.e., `P < PEND'. */
static boolean
at_endline_loc_p (const_re_char *p, const_re_char *pend, reg_syntax_t syntax)
{
re_char *next = p;
boolean next_backslash = *next == '\\';
re_char *next_next = p + 1 < pend ? p + 1 : 0;
return
/* Before a subexpression? */
(syntax & RE_NO_BK_PARENS ? *next == ')'
: next_backslash && next_next && *next_next == ')')
/* Before an alternative? */
|| (syntax & RE_NO_BK_VBAR ? *next == '|'
: next_backslash && next_next && *next_next == '|');
}
/* Returns true if REGNUM is in one of COMPILE_STACK's elements and
false if it's not. */
static boolean
group_in_compile_stack (compile_stack_type compile_stack, regnum_t regnum)
{
ssize_t this_element;
for (this_element = compile_stack.avail - 1;
this_element >= 0;
this_element--)
if (compile_stack.stack[this_element].regnum == regnum)
return true;
return false;
}
\f
/* analyze_first.
If fastmap is non-NULL, go through the pattern and fill fastmap
with all the possible leading chars. If fastmap is NULL, don't
bother filling it up (obviously) and only return whether the
pattern could potentially match the empty string.
Return 1 if p..pend might match the empty string.
Return 0 if p..pend matches at least one char.
Return -1 if fastmap was not updated accurately. */
static int
analyze_first (const_re_char *p, const_re_char *pend, char *fastmap,
const int multibyte)
{
int j, k;
boolean not;
/* If all elements for base leading-codes in fastmap is set, this
flag is set true. */
boolean match_any_multibyte_characters = false;
assert (p);
/* The loop below works as follows:
- It has a working-list kept in the PATTERN_STACK and which basically
starts by only containing a pointer to the first operation.
- If the opcode we're looking at is a match against some set of
chars, then we add those chars to the fastmap and go on to the
next work element from the worklist (done via `break').
- If the opcode is a control operator on the other hand, we either
ignore it (if it's meaningless at this point, such as `start_memory')
or execute it (if it's a jump). If the jump has several destinations
(i.e. `on_failure_jump'), then we push the other destination onto the
worklist.
We guarantee termination by ignoring backward jumps (more or less),
so that `p' is monotonically increasing. More to the point, we
never set `p' (or push) anything `<= p1'. */
while (p < pend)
{
/* `p1' is used as a marker of how far back a `on_failure_jump'
can go without being ignored. It is normally equal to `p'
(which prevents any backward `on_failure_jump') except right
after a plain `jump', to allow patterns such as:
0: jump 10
3..9: <body>
10: on_failure_jump 3
as used for the *? operator. */
re_char *p1 = p;
switch (*p++)
{
case succeed:
return 1;
case duplicate:
/* If the first character has to match a backreference, that means
that the group was empty (since it already matched). Since this
is the only case that interests us here, we can assume that the
backreference must match the empty string. */
p++;
continue;
/* Following are the cases which match a character. These end
with `break'. */
case exactn:
if (fastmap)
{
/* If multibyte is nonzero, the first byte of each
character is an ASCII or a leading code. Otherwise,
each byte is a character. Thus, this works in both
cases. */
fastmap[p[1]] = 1;
if (! multibyte)
{
/* For the case of matching this unibyte regex
against multibyte, we must set a leading code of
the corresponding multibyte character. */
int c = RE_CHAR_TO_MULTIBYTE (p[1]);
fastmap[CHAR_LEADING_CODE (c)] = 1;
}
}
break;
case anychar:
/* We could put all the chars except for \n (and maybe \0)
but we don't bother since it is generally not worth it. */
if (!fastmap) break;
return -1;
case charset_not:
if (!fastmap) break;
{
/* Chars beyond end of bitmap are possible matches. */
for (j = CHARSET_BITMAP_SIZE (&p[-1]) * BYTEWIDTH;
j < (1 << BYTEWIDTH); j++)
fastmap[j] = 1;
}
/* Fallthrough */
case charset:
if (!fastmap) break;
not = (re_opcode_t) *(p - 1) == charset_not;
for (j = CHARSET_BITMAP_SIZE (&p[-1]) * BYTEWIDTH - 1, p++;
j >= 0; j--)
if (!!(p[j / BYTEWIDTH] & (1 << (j % BYTEWIDTH))) ^ not)
fastmap[j] = 1;
#ifdef emacs
if (/* Any leading code can possibly start a character
which doesn't match the specified set of characters. */
not
||
/* If we can match a character class, we can match any
multibyte characters. */
(CHARSET_RANGE_TABLE_EXISTS_P (&p[-2])
&& CHARSET_RANGE_TABLE_BITS (&p[-2]) != 0))
{
if (match_any_multibyte_characters == false)
{
for (j = MIN_MULTIBYTE_LEADING_CODE;
j <= MAX_MULTIBYTE_LEADING_CODE; j++)
fastmap[j] = 1;
match_any_multibyte_characters = true;
}
}
else if (!not && CHARSET_RANGE_TABLE_EXISTS_P (&p[-2])
&& match_any_multibyte_characters == false)
{
/* Set fastmap[I] to 1 where I is a leading code of each
multibyte character in the range table. */
int c, count;
unsigned char lc1, lc2;
/* Make P points the range table. `+ 2' is to skip flag
bits for a character class. */
p += CHARSET_BITMAP_SIZE (&p[-2]) + 2;
/* Extract the number of ranges in range table into COUNT. */
EXTRACT_NUMBER_AND_INCR (count, p);
for (; count > 0; count--, p += 3)
{
/* Extract the start and end of each range. */
EXTRACT_CHARACTER (c, p);
lc1 = CHAR_LEADING_CODE (c);
p += 3;
EXTRACT_CHARACTER (c, p);
lc2 = CHAR_LEADING_CODE (c);
for (j = lc1; j <= lc2; j++)
fastmap[j] = 1;
}
}
#endif
break;
case syntaxspec:
case notsyntaxspec:
if (!fastmap) break;
#ifndef emacs
not = (re_opcode_t)p[-1] == notsyntaxspec;
k = *p++;
for (j = 0; j < (1 << BYTEWIDTH); j++)
if ((SYNTAX (j) == (enum syntaxcode) k) ^ not)
fastmap[j] = 1;
break;
#else /* emacs */
/* This match depends on text properties. These end with
aborting optimizations. */
return -1;
case categoryspec:
case notcategoryspec:
if (!fastmap) break;
not = (re_opcode_t)p[-1] == notcategoryspec;
k = *p++;
for (j = (1 << BYTEWIDTH); j >= 0; j--)
if ((CHAR_HAS_CATEGORY (j, k)) ^ not)
fastmap[j] = 1;
/* Any leading code can possibly start a character which
has or doesn't has the specified category. */
if (match_any_multibyte_characters == false)
{
for (j = MIN_MULTIBYTE_LEADING_CODE;
j <= MAX_MULTIBYTE_LEADING_CODE; j++)
fastmap[j] = 1;
match_any_multibyte_characters = true;
}
break;
/* All cases after this match the empty string. These end with
`continue'. */
case at_dot:
#endif /* !emacs */
case no_op:
case begline:
case endline:
case begbuf:
case endbuf:
case wordbound:
case notwordbound:
case wordbeg:
case wordend:
case symbeg:
case symend:
continue;
case jump:
EXTRACT_NUMBER_AND_INCR (j, p);
if (j < 0)
/* Backward jumps can only go back to code that we've already
visited. `re_compile' should make sure this is true. */
break;
p += j;
switch (*p)
{
case on_failure_jump:
case on_failure_keep_string_jump:
case on_failure_jump_loop:
case on_failure_jump_nastyloop:
case on_failure_jump_smart:
p++;
break;
default:
continue;
};
/* Keep `p1' to allow the `on_failure_jump' we are jumping to
to jump back to "just after here". */
/* Fallthrough */
case on_failure_jump:
case on_failure_keep_string_jump:
case on_failure_jump_nastyloop:
case on_failure_jump_loop:
case on_failure_jump_smart:
EXTRACT_NUMBER_AND_INCR (j, p);
if (p + j <= p1)
; /* Backward jump to be ignored. */
else
{ /* We have to look down both arms.
We first go down the "straight" path so as to minimize
stack usage when going through alternatives. */
int r = analyze_first (p, pend, fastmap, multibyte);
if (r) return r;
p += j;
}
continue;
case jump_n:
/* This code simply does not properly handle forward jump_n. */
DEBUG_STATEMENT (EXTRACT_NUMBER (j, p); assert (j < 0));
p += 4;
/* jump_n can either jump or fall through. The (backward) jump
case has already been handled, so we only need to look at the
fallthrough case. */
continue;
case succeed_n:
/* If N == 0, it should be an on_failure_jump_loop instead. */
DEBUG_STATEMENT (EXTRACT_NUMBER (j, p + 2); assert (j > 0));
p += 4;
/* We only care about one iteration of the loop, so we don't
need to consider the case where this behaves like an
on_failure_jump. */
continue;
case set_number_at:
p += 4;
continue;
case start_memory:
case stop_memory:
p += 1;
continue;
default:
abort (); /* We have listed all the cases. */
} /* switch *p++ */
/* Getting here means we have found the possible starting
characters for one path of the pattern -- and that the empty
string does not match. We need not follow this path further. */
return 0;
} /* while p */
/* We reached the end without matching anything. */
return 1;
} /* analyze_first */
\f
/* re_compile_fastmap computes a ``fastmap'' for the compiled pattern in
BUFP. A fastmap records which of the (1 << BYTEWIDTH) possible
characters can start a string that matches the pattern. This fastmap
is used by re_search to skip quickly over impossible starting points.
Character codes above (1 << BYTEWIDTH) are not represented in the
fastmap, but the leading codes are represented. Thus, the fastmap
indicates which character sets could start a match.
The caller must supply the address of a (1 << BYTEWIDTH)-byte data
area as BUFP->fastmap.
We set the `fastmap', `fastmap_accurate', and `can_be_null' fields in
the pattern buffer.
Returns 0 if we succeed, -2 if an internal error. */
int
re_compile_fastmap (struct re_pattern_buffer *bufp)
{
char *fastmap = bufp->fastmap;
int analysis;
assert (fastmap && bufp->buffer);
memset (fastmap, 0, 1 << BYTEWIDTH); /* Assume nothing's valid. */
bufp->fastmap_accurate = 1; /* It will be when we're done. */
analysis = analyze_first (bufp->buffer, bufp->buffer + bufp->used,
fastmap, RE_MULTIBYTE_P (bufp));
bufp->can_be_null = (analysis != 0);
return 0;
} /* re_compile_fastmap */
\f
/* Set REGS to hold NUM_REGS registers, storing them in STARTS and
ENDS. Subsequent matches using PATTERN_BUFFER and REGS will use
this memory for recording register information. STARTS and ENDS
must be allocated using the malloc library routine, and must each
be at least NUM_REGS * sizeof (regoff_t) bytes long.
If NUM_REGS == 0, then subsequent matches should allocate their own
register data.
Unless this function is called, the first search or match using
PATTERN_BUFFER will allocate its own register data, without
freeing the old data. */
void
re_set_registers (struct re_pattern_buffer *bufp, struct re_registers *regs, unsigned int num_regs, regoff_t *starts, regoff_t *ends)
{
if (num_regs)
{
bufp->regs_allocated = REGS_REALLOCATE;
regs->num_regs = num_regs;
regs->start = starts;
regs->end = ends;
}
else
{
bufp->regs_allocated = REGS_UNALLOCATED;
regs->num_regs = 0;
regs->start = regs->end = 0;
}
}
WEAK_ALIAS (__re_set_registers, re_set_registers)
\f
/* Searching routines. */
/* Like re_search_2, below, but only one string is specified, and
doesn't let you say where to stop matching. */
regoff_t
re_search (struct re_pattern_buffer *bufp, const char *string, size_t size,
ssize_t startpos, ssize_t range, struct re_registers *regs)
{
return re_search_2 (bufp, NULL, 0, string, size, startpos, range,
regs, size);
}
WEAK_ALIAS (__re_search, re_search)
/* Head address of virtual concatenation of string. */
#define HEAD_ADDR_VSTRING(P) \
(((P) >= size1 ? string2 : string1))
/* Address of POS in the concatenation of virtual string. */
#define POS_ADDR_VSTRING(POS) \
(((POS) >= size1 ? string2 - size1 : string1) + (POS))
/* Using the compiled pattern in BUFP->buffer, first tries to match the
virtual concatenation of STRING1 and STRING2, starting first at index
STARTPOS, then at STARTPOS + 1, and so on.
STRING1 and STRING2 have length SIZE1 and SIZE2, respectively.
RANGE is how far to scan while trying to match. RANGE = 0 means try
only at STARTPOS; in general, the last start tried is STARTPOS +
RANGE.
In REGS, return the indices of the virtual concatenation of STRING1
and STRING2 that matched the entire BUFP->buffer and its contained
subexpressions.
Do not consider matching one past the index STOP in the virtual
concatenation of STRING1 and STRING2.
We return either the position in the strings at which the match was
found, -1 if no match, or -2 if error (such as failure
stack overflow). */
regoff_t
re_search_2 (struct re_pattern_buffer *bufp, const char *str1, size_t size1,
const char *str2, size_t size2, ssize_t startpos, ssize_t range,
struct re_registers *regs, ssize_t stop)
{
regoff_t val;
re_char *string1 = (re_char*) str1;
re_char *string2 = (re_char*) str2;
register char *fastmap = bufp->fastmap;
register RE_TRANSLATE_TYPE translate = bufp->translate;
size_t total_size = size1 + size2;
ssize_t endpos = startpos + range;
boolean anchored_start;
/* Nonzero if we are searching multibyte string. */
const boolean multibyte = RE_TARGET_MULTIBYTE_P (bufp);
/* Check for out-of-range STARTPOS. */
if (startpos < 0 || startpos > total_size)
return -1;
/* Fix up RANGE if it might eventually take us outside
the virtual concatenation of STRING1 and STRING2.
Make sure we won't move STARTPOS below 0 or above TOTAL_SIZE. */
if (endpos < 0)
range = 0 - startpos;
else if (endpos > total_size)
range = total_size - startpos;
/* If the search isn't to be a backwards one, don't waste time in a
search for a pattern anchored at beginning of buffer. */
if (bufp->used > 0 && (re_opcode_t) bufp->buffer[0] == begbuf && range > 0)
{
if (startpos > 0)
return -1;
else
range = 0;
}
#ifdef emacs
/* In a forward search for something that starts with \=.
don't keep searching past point. */
if (bufp->used > 0 && (re_opcode_t) bufp->buffer[0] == at_dot && range > 0)
{
range = PT_BYTE - BEGV_BYTE - startpos;
if (range < 0)
return -1;
}
#endif /* emacs */
/* Update the fastmap now if not correct already. */
if (fastmap && !bufp->fastmap_accurate)
re_compile_fastmap (bufp);
/* See whether the pattern is anchored. */
anchored_start = (bufp->buffer[0] == begline);
#ifdef emacs
gl_state.object = re_match_object; /* Used by SYNTAX_TABLE_BYTE_TO_CHAR. */
{
ssize_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (POS_AS_IN_BUFFER (startpos));
SETUP_SYNTAX_TABLE_FOR_OBJECT (re_match_object, charpos, 1);
}
#endif
/* Loop through the string, looking for a place to start matching. */
for (;;)
{
/* If the pattern is anchored,
skip quickly past places we cannot match.
We don't bother to treat startpos == 0 specially
because that case doesn't repeat. */
if (anchored_start && startpos > 0)
{
if (! ((startpos <= size1 ? string1[startpos - 1]
: string2[startpos - size1 - 1])
== '\n'))
goto advance;
}
/* If a fastmap is supplied, skip quickly over characters that
cannot be the start of a match. If the pattern can match the
null string, however, we don't need to skip characters; we want
the first null string. */
if (fastmap && startpos < total_size && !bufp->can_be_null)
{
register re_char *d;
register re_wchar_t buf_ch;
d = POS_ADDR_VSTRING (startpos);
if (range > 0) /* Searching forwards. */
{
ssize_t irange = range, lim = 0;
if (startpos < size1 && startpos + range >= size1)
lim = range - (size1 - startpos);
/* Written out as an if-else to avoid testing `translate'
inside the loop. */
if (RE_TRANSLATE_P (translate))
{
if (multibyte)
while (range > lim)
{
int buf_charlen;
buf_ch = STRING_CHAR_AND_LENGTH (d, buf_charlen);
buf_ch = RE_TRANSLATE (translate, buf_ch);
if (fastmap[CHAR_LEADING_CODE (buf_ch)])
break;
range -= buf_charlen;
d += buf_charlen;
}
else
while (range > lim)
{
register re_wchar_t ch, translated;
buf_ch = *d;
ch = RE_CHAR_TO_MULTIBYTE (buf_ch);
translated = RE_TRANSLATE (translate, ch);
if (translated != ch
&& (ch = RE_CHAR_TO_UNIBYTE (translated)) >= 0)
buf_ch = ch;
if (fastmap[buf_ch])
break;
d++;
range--;
}
}
else
{
if (multibyte)
while (range > lim)
{
int buf_charlen;
buf_ch = STRING_CHAR_AND_LENGTH (d, buf_charlen);
if (fastmap[CHAR_LEADING_CODE (buf_ch)])
break;
range -= buf_charlen;
d += buf_charlen;
}
else
while (range > lim && !fastmap[*d])
{
d++;
range--;
}
}
startpos += irange - range;
}
else /* Searching backwards. */
{
if (multibyte)
{
buf_ch = STRING_CHAR (d);
buf_ch = TRANSLATE (buf_ch);
if (! fastmap[CHAR_LEADING_CODE (buf_ch)])
goto advance;
}
else
{
register re_wchar_t ch, translated;
buf_ch = *d;
ch = RE_CHAR_TO_MULTIBYTE (buf_ch);
translated = TRANSLATE (ch);
if (translated != ch
&& (ch = RE_CHAR_TO_UNIBYTE (translated)) >= 0)
buf_ch = ch;
if (! fastmap[TRANSLATE (buf_ch)])
goto advance;
}
}
}
/* If can't match the null string, and that's all we have left, fail. */
if (range >= 0 && startpos == total_size && fastmap
&& !bufp->can_be_null)
return -1;
val = re_match_2_internal (bufp, string1, size1, string2, size2,
startpos, regs, stop);
if (val >= 0)
return startpos;
if (val == -2)
return -2;
advance:
if (!range)
break;
else if (range > 0)
{
/* Update STARTPOS to the next character boundary. */
if (multibyte)
{
re_char *p = POS_ADDR_VSTRING (startpos);
int len = BYTES_BY_CHAR_HEAD (*p);
range -= len;
if (range < 0)
break;
startpos += len;
}
else
{
range--;
startpos++;
}
}
else
{
range++;
startpos--;
/* Update STARTPOS to the previous character boundary. */
if (multibyte)
{
re_char *p = POS_ADDR_VSTRING (startpos) + 1;
re_char *p0 = p;
re_char *phead = HEAD_ADDR_VSTRING (startpos);
/* Find the head of multibyte form. */
PREV_CHAR_BOUNDARY (p, phead);
range += p0 - 1 - p;
if (range > 0)
break;
startpos -= p0 - 1 - p;
}
}
}
return -1;
} /* re_search_2 */
WEAK_ALIAS (__re_search_2, re_search_2)
\f
/* Declarations and macros for re_match_2. */
static int bcmp_translate (re_char *s1, re_char *s2,
register ssize_t len,
RE_TRANSLATE_TYPE translate,
const int multibyte);
/* This converts PTR, a pointer into one of the search strings `string1'
and `string2' into an offset from the beginning of that string. */
#define POINTER_TO_OFFSET(ptr) \
(FIRST_STRING_P (ptr) \
? (ptr) - string1 \
: (ptr) - string2 + (ptrdiff_t) size1)
/* Call before fetching a character with *d. This switches over to
string2 if necessary.
Check re_match_2_internal for a discussion of why end_match_2 might
not be within string2 (but be equal to end_match_1 instead). */
#define PREFETCH() \
while (d == dend) \
{ \
/* End of string2 => fail. */ \
if (dend == end_match_2) \
goto fail; \
/* End of string1 => advance to string2. */ \
d = string2; \
dend = end_match_2; \
}
/* Call before fetching a char with *d if you already checked other limits.
This is meant for use in lookahead operations like wordend, etc..
where we might need to look at parts of the string that might be
outside of the LIMITs (i.e past `stop'). */
#define PREFETCH_NOLIMIT() \
if (d == end1) \
{ \
d = string2; \
dend = end_match_2; \
} \
/* Test if at very beginning or at very end of the virtual concatenation
of `string1' and `string2'. If only one string, it's `string2'. */
#define AT_STRINGS_BEG(d) ((d) == (size1 ? string1 : string2) || !size2)
#define AT_STRINGS_END(d) ((d) == end2)
/* Disabled due to a compiler bug -- see comment at case wordbound */
/* The comment at case wordbound is following one, but we don't use
AT_WORD_BOUNDARY anymore to support multibyte form.
The DEC Alpha C compiler 3.x generates incorrect code for the
test WORDCHAR_P (d - 1) != WORDCHAR_P (d) in the expansion of
AT_WORD_BOUNDARY, so this code is disabled. Expanding the
macro and introducing temporary variables works around the bug. */
#if 0
/* Test if D points to a character which is word-constituent. We have
two special cases to check for: if past the end of string1, look at
the first character in string2; and if before the beginning of
string2, look at the last character in string1. */
#define WORDCHAR_P(d) \
(SYNTAX ((d) == end1 ? *string2 \
: (d) == string2 - 1 ? *(end1 - 1) : *(d)) \
== Sword)
/* Test if the character before D and the one at D differ with respect
to being word-constituent. */
#define AT_WORD_BOUNDARY(d) \
(AT_STRINGS_BEG (d) || AT_STRINGS_END (d) \
|| WORDCHAR_P (d - 1) != WORDCHAR_P (d))
#endif
/* Free everything we malloc. */
#ifdef MATCH_MAY_ALLOCATE
# define FREE_VAR(var) \
do { \
if (var) \
{ \
REGEX_FREE (var); \
var = NULL; \
} \
} while (0)
# define FREE_VARIABLES() \
do { \
REGEX_FREE_STACK (fail_stack.stack); \
FREE_VAR (regstart); \
FREE_VAR (regend); \
FREE_VAR (best_regstart); \
FREE_VAR (best_regend); \
REGEX_SAFE_FREE (); \
} while (0)
#else
# define FREE_VARIABLES() ((void)0) /* Do nothing! But inhibit gcc warning. */
#endif /* not MATCH_MAY_ALLOCATE */
\f
/* Optimization routines. */
/* If the operation is a match against one or more chars,
return a pointer to the next operation, else return NULL. */
static re_char *
skip_one_char (const_re_char *p)
{
switch (*p++)
{
case anychar:
break;
case exactn:
p += *p + 1;
break;
case charset_not:
case charset:
if (CHARSET_RANGE_TABLE_EXISTS_P (p - 1))
{
int mcnt;
p = CHARSET_RANGE_TABLE (p - 1);
EXTRACT_NUMBER_AND_INCR (mcnt, p);
p = CHARSET_RANGE_TABLE_END (p, mcnt);
}
else
p += 1 + CHARSET_BITMAP_SIZE (p - 1);
break;
case syntaxspec:
case notsyntaxspec:
#ifdef emacs
case categoryspec:
case notcategoryspec:
#endif /* emacs */
p++;
break;
default:
p = NULL;
}
return p;
}
/* Jump over non-matching operations. */
static re_char *
skip_noops (const_re_char *p, const_re_char *pend)
{
int mcnt;
while (p < pend)
{
switch (*p)
{
case start_memory:
case stop_memory:
p += 2; break;
case no_op:
p += 1; break;
case jump:
p += 1;
EXTRACT_NUMBER_AND_INCR (mcnt, p);
p += mcnt;
break;
default:
return p;
}
}
assert (p == pend);
return p;
}
/* Test if C matches charset op. *PP points to the charset or charset_not
opcode. When the function finishes, *PP will be advanced past that opcode.
C is character to test (possibly after translations) and CORIG is original
character (i.e. without any translations). UNIBYTE denotes whether c is
unibyte or multibyte character. */
static bool
execute_charset (const_re_char **pp, unsigned c, unsigned corig, bool unibyte)
{
re_char *p = *pp, *rtp = NULL;
bool not = (re_opcode_t) *p == charset_not;
if (CHARSET_RANGE_TABLE_EXISTS_P (p))
{
int count;
rtp = CHARSET_RANGE_TABLE (p);
EXTRACT_NUMBER_AND_INCR (count, rtp);
*pp = CHARSET_RANGE_TABLE_END ((rtp), (count));
}
else
*pp += 2 + CHARSET_BITMAP_SIZE (p);
if (unibyte && c < (1 << BYTEWIDTH))
{ /* Lookup bitmap. */
/* Cast to `unsigned' instead of `unsigned char' in
case the bit list is a full 32 bytes long. */
if (c < (unsigned) (CHARSET_BITMAP_SIZE (p) * BYTEWIDTH)
&& p[2 + c / BYTEWIDTH] & (1 << (c % BYTEWIDTH)))
return !not;
}
#ifdef emacs
else if (rtp)
{
int class_bits = CHARSET_RANGE_TABLE_BITS (p);
re_wchar_t range_start, range_end;
/* Sort tests by the most commonly used classes with some adjustment to which
tests are easiest to perform. Take a look at comment in re_wctype_parse
for table with frequencies of character class names. */
if ((class_bits & BIT_MULTIBYTE) ||
(class_bits & BIT_ALNUM && ISALNUM (c)) ||
(class_bits & BIT_ALPHA && ISALPHA (c)) ||
(class_bits & BIT_SPACE && ISSPACE (c)) ||
(class_bits & BIT_WORD && ISWORD (c)) ||
((class_bits & BIT_UPPER) &&
(ISUPPER (c) || (corig != c &&
c == downcase (corig) && ISLOWER (c)))) ||
((class_bits & BIT_LOWER) &&
(ISLOWER (c) || (corig != c &&
c == upcase (corig) && ISUPPER(c)))) ||
(class_bits & BIT_PUNCT && ISPUNCT (c)) ||
(class_bits & BIT_GRAPH && ISGRAPH (c)) ||
(class_bits & BIT_PRINT && ISPRINT (c)))
return !not;
for (p = *pp; rtp < p; rtp += 2 * 3)
{
EXTRACT_CHARACTER (range_start, rtp);
EXTRACT_CHARACTER (range_end, rtp + 3);
if (range_start <= c && c <= range_end)
return !not;
}
}
#endif /* emacs */
return not;
}
/* Non-zero if "p1 matches something" implies "p2 fails". */
static int
mutually_exclusive_p (struct re_pattern_buffer *bufp, const_re_char *p1,
const_re_char *p2)
{
re_opcode_t op2;
const boolean multibyte = RE_MULTIBYTE_P (bufp);
unsigned char *pend = bufp->buffer + bufp->used;
assert (p1 >= bufp->buffer && p1 < pend
&& p2 >= bufp->buffer && p2 <= pend);
/* Skip over open/close-group commands.
If what follows this loop is a ...+ construct,
look at what begins its body, since we will have to
match at least one of that. */
p2 = skip_noops (p2, pend);
/* The same skip can be done for p1, except that this function
is only used in the case where p1 is a simple match operator. */
/* p1 = skip_noops (p1, pend); */
assert (p1 >= bufp->buffer && p1 < pend
&& p2 >= bufp->buffer && p2 <= pend);
op2 = p2 == pend ? succeed : *p2;
switch (op2)
{
case succeed:
case endbuf:
/* If we're at the end of the pattern, we can change. */
if (skip_one_char (p1))
{
DEBUG_PRINT (" End of pattern: fast loop.\n");
return 1;
}
break;
case endline:
case exactn:
{
register re_wchar_t c
= (re_opcode_t) *p2 == endline ? '\n'
: RE_STRING_CHAR (p2 + 2, multibyte);
if ((re_opcode_t) *p1 == exactn)
{
if (c != RE_STRING_CHAR (p1 + 2, multibyte))
{
DEBUG_PRINT (" '%c' != '%c' => fast loop.\n", c, p1[2]);
return 1;
}
}
else if ((re_opcode_t) *p1 == charset
|| (re_opcode_t) *p1 == charset_not)
{
if (!execute_charset (&p1, c, c, !multibyte || IS_REAL_ASCII (c)))
{
DEBUG_PRINT (" No match => fast loop.\n");
return 1;
}
}
else if ((re_opcode_t) *p1 == anychar
&& c == '\n')
{
DEBUG_PRINT (" . != \\n => fast loop.\n");
return 1;
}
}
break;
case charset:
{
if ((re_opcode_t) *p1 == exactn)
/* Reuse the code above. */
return mutually_exclusive_p (bufp, p2, p1);
/* It is hard to list up all the character in charset
P2 if it includes multibyte character. Give up in
such case. */
else if (!multibyte || !CHARSET_RANGE_TABLE_EXISTS_P (p2))
{
/* Now, we are sure that P2 has no range table.
So, for the size of bitmap in P2, `p2[1]' is
enough. But P1 may have range table, so the
size of bitmap table of P1 is extracted by
using macro `CHARSET_BITMAP_SIZE'.
In a multibyte case, we know that all the character
listed in P2 is ASCII. In a unibyte case, P1 has only a
bitmap table. So, in both cases, it is enough to test
only the bitmap table of P1. */
if ((re_opcode_t) *p1 == charset)
{
int idx;
/* We win if the charset inside the loop
has no overlap with the one after the loop. */
for (idx = 0;
(idx < (int) p2[1]
&& idx < CHARSET_BITMAP_SIZE (p1));
idx++)
if ((p2[2 + idx] & p1[2 + idx]) != 0)
break;
if (idx == p2[1]
|| idx == CHARSET_BITMAP_SIZE (p1))
{
DEBUG_PRINT (" No match => fast loop.\n");
return 1;
}
}
else if ((re_opcode_t) *p1 == charset_not)
{
int idx;
/* We win if the charset_not inside the loop lists
every character listed in the charset after. */
for (idx = 0; idx < (int) p2[1]; idx++)
if (! (p2[2 + idx] == 0
|| (idx < CHARSET_BITMAP_SIZE (p1)
&& ((p2[2 + idx] & ~ p1[2 + idx]) == 0))))
break;
if (idx == p2[1])
{
DEBUG_PRINT (" No match => fast loop.\n");
return 1;
}
}
}
}
break;
case charset_not:
switch (*p1)
{
case exactn:
case charset:
/* Reuse the code above. */
return mutually_exclusive_p (bufp, p2, p1);
case charset_not:
/* When we have two charset_not, it's very unlikely that
they don't overlap. The union of the two sets of excluded
chars should cover all possible chars, which, as a matter of
fact, is virtually impossible in multibyte buffers. */
break;
}
break;
case wordend:
return ((re_opcode_t) *p1 == syntaxspec && p1[1] == Sword);
case symend:
return ((re_opcode_t) *p1 == syntaxspec
&& (p1[1] == Ssymbol || p1[1] == Sword));
case notsyntaxspec:
return ((re_opcode_t) *p1 == syntaxspec && p1[1] == p2[1]);
case wordbeg:
return ((re_opcode_t) *p1 == notsyntaxspec && p1[1] == Sword);
case symbeg:
return ((re_opcode_t) *p1 == notsyntaxspec
&& (p1[1] == Ssymbol || p1[1] == Sword));
case syntaxspec:
return ((re_opcode_t) *p1 == notsyntaxspec && p1[1] == p2[1]);
case wordbound:
return (((re_opcode_t) *p1 == notsyntaxspec
|| (re_opcode_t) *p1 == syntaxspec)
&& p1[1] == Sword);
#ifdef emacs
case categoryspec:
return ((re_opcode_t) *p1 == notcategoryspec && p1[1] == p2[1]);
case notcategoryspec:
return ((re_opcode_t) *p1 == categoryspec && p1[1] == p2[1]);
#endif /* emacs */
default:
;
}
/* Safe default. */
return 0;
}
\f
/* Matching routines. */
#ifndef emacs /* Emacs never uses this. */
/* re_match is like re_match_2 except it takes only a single string. */
regoff_t
re_match (struct re_pattern_buffer *bufp, const char *string,
size_t size, ssize_t pos, struct re_registers *regs)
{
regoff_t result = re_match_2_internal (bufp, NULL, 0, (re_char*) string,
size, pos, regs, size);
return result;
}
WEAK_ALIAS (__re_match, re_match)
#endif /* not emacs */
#ifdef emacs
/* In Emacs, this is the string or buffer in which we
are matching. It is used for looking up syntax properties. */
Lisp_Object re_match_object;
#endif
/* re_match_2 matches the compiled pattern in BUFP against the
the (virtual) concatenation of STRING1 and STRING2 (of length SIZE1
and SIZE2, respectively). We start matching at POS, and stop
matching at STOP.
If REGS is non-null and the `no_sub' field of BUFP is nonzero, we
store offsets for the substring each group matched in REGS. See the
documentation for exactly how many groups we fill.
We return -1 if no match, -2 if an internal error (such as the
failure stack overflowing). Otherwise, we return the length of the
matched substring. */
regoff_t
re_match_2 (struct re_pattern_buffer *bufp, const char *string1,
size_t size1, const char *string2, size_t size2, ssize_t pos,
struct re_registers *regs, ssize_t stop)
{
regoff_t result;
#ifdef emacs
ssize_t charpos;
gl_state.object = re_match_object; /* Used by SYNTAX_TABLE_BYTE_TO_CHAR. */
charpos = SYNTAX_TABLE_BYTE_TO_CHAR (POS_AS_IN_BUFFER (pos));
SETUP_SYNTAX_TABLE_FOR_OBJECT (re_match_object, charpos, 1);
#endif
result = re_match_2_internal (bufp, (re_char*) string1, size1,
(re_char*) string2, size2,
pos, regs, stop);
return result;
}
WEAK_ALIAS (__re_match_2, re_match_2)
/* This is a separate function so that we can force an alloca cleanup
afterwards. */
static regoff_t
re_match_2_internal (struct re_pattern_buffer *bufp, const_re_char *string1,
size_t size1, const_re_char *string2, size_t size2,
ssize_t pos, struct re_registers *regs, ssize_t stop)
{
/* General temporaries. */
int mcnt;
size_t reg;
/* Just past the end of the corresponding string. */
re_char *end1, *end2;
/* Pointers into string1 and string2, just past the last characters in
each to consider matching. */
re_char *end_match_1, *end_match_2;
/* Where we are in the data, and the end of the current string. */
re_char *d, *dend;
/* Used sometimes to remember where we were before starting matching
an operator so that we can go back in case of failure. This "atomic"
behavior of matching opcodes is indispensable to the correctness
of the on_failure_keep_string_jump optimization. */
re_char *dfail;
/* Where we are in the pattern, and the end of the pattern. */
re_char *p = bufp->buffer;
re_char *pend = p + bufp->used;
/* We use this to map every character in the string. */
RE_TRANSLATE_TYPE translate = bufp->translate;
/* Nonzero if BUFP is setup from a multibyte regex. */
const boolean multibyte = RE_MULTIBYTE_P (bufp);
/* Nonzero if STRING1/STRING2 are multibyte. */
const boolean target_multibyte = RE_TARGET_MULTIBYTE_P (bufp);
/* Failure point stack. Each place that can handle a failure further
down the line pushes a failure point on this stack. It consists of
regstart, and regend for all registers corresponding to
the subexpressions we're currently inside, plus the number of such
registers, and, finally, two char *'s. The first char * is where
to resume scanning the pattern; the second one is where to resume
scanning the strings. */
#ifdef MATCH_MAY_ALLOCATE /* otherwise, this is global. */
fail_stack_type fail_stack;
#endif
#ifdef DEBUG_COMPILES_ARGUMENTS
unsigned nfailure_points_pushed = 0, nfailure_points_popped = 0;
#endif
#if defined REL_ALLOC && defined REGEX_MALLOC
/* This holds the pointer to the failure stack, when
it is allocated relocatably. */
fail_stack_elt_t *failure_stack_ptr;
#endif
/* We fill all the registers internally, independent of what we
return, for use in backreferences. The number here includes
an element for register zero. */
size_t num_regs = bufp->re_nsub + 1;
/* Information on the contents of registers. These are pointers into
the input strings; they record just what was matched (on this
attempt) by a subexpression part of the pattern, that is, the
regnum-th regstart pointer points to where in the pattern we began
matching and the regnum-th regend points to right after where we
stopped matching the regnum-th subexpression. (The zeroth register
keeps track of what the whole pattern matches.) */
#ifdef MATCH_MAY_ALLOCATE /* otherwise, these are global. */
re_char **regstart, **regend;
#endif
/* The following record the register info as found in the above
variables when we find a match better than any we've seen before.
This happens as we backtrack through the failure points, which in
turn happens only if we have not yet matched the entire string. */
unsigned best_regs_set = false;
#ifdef MATCH_MAY_ALLOCATE /* otherwise, these are global. */
re_char **best_regstart, **best_regend;
#endif
/* Logically, this is `best_regend[0]'. But we don't want to have to
allocate space for that if we're not allocating space for anything
else (see below). Also, we never need info about register 0 for
any of the other register vectors, and it seems rather a kludge to
treat `best_regend' differently than the rest. So we keep track of
the end of the best match so far in a separate variable. We
initialize this to NULL so that when we backtrack the first time
and need to test it, it's not garbage. */
re_char *match_end = NULL;
#ifdef DEBUG_COMPILES_ARGUMENTS
/* Counts the total number of registers pushed. */
unsigned num_regs_pushed = 0;
#endif
DEBUG_PRINT ("\n\nEntering re_match_2.\n");
REGEX_USE_SAFE_ALLOCA;
INIT_FAIL_STACK ();
#ifdef MATCH_MAY_ALLOCATE
/* Do not bother to initialize all the register variables if there are
no groups in the pattern, as it takes a fair amount of time. If
there are groups, we include space for register 0 (the whole
pattern), even though we never use it, since it simplifies the
array indexing. We should fix this. */
if (bufp->re_nsub)
{
regstart = REGEX_TALLOC (num_regs, re_char *);
regend = REGEX_TALLOC (num_regs, re_char *);
best_regstart = REGEX_TALLOC (num_regs, re_char *);
best_regend = REGEX_TALLOC (num_regs, re_char *);
if (!(regstart && regend && best_regstart && best_regend))
{
FREE_VARIABLES ();
return -2;
}
}
else
{
/* We must initialize all our variables to NULL, so that
`FREE_VARIABLES' doesn't try to free them. */
regstart = regend = best_regstart = best_regend = NULL;
}
#endif /* MATCH_MAY_ALLOCATE */
/* The starting position is bogus. */
if (pos < 0 || pos > size1 + size2)
{
FREE_VARIABLES ();
return -1;
}
/* Initialize subexpression text positions to -1 to mark ones that no
start_memory/stop_memory has been seen for. Also initialize the
register information struct. */
for (reg = 1; reg < num_regs; reg++)
regstart[reg] = regend[reg] = NULL;
/* We move `string1' into `string2' if the latter's empty -- but not if
`string1' is null. */
if (size2 == 0 && string1 != NULL)
{
string2 = string1;
size2 = size1;
string1 = 0;
size1 = 0;
}
end1 = string1 + size1;
end2 = string2 + size2;
/* `p' scans through the pattern as `d' scans through the data.
`dend' is the end of the input string that `d' points within. `d'
is advanced into the following input string whenever necessary, but
this happens before fetching; therefore, at the beginning of the
loop, `d' can be pointing at the end of a string, but it cannot
equal `string2'. */
if (pos >= size1)
{
/* Only match within string2. */
d = string2 + pos - size1;
dend = end_match_2 = string2 + stop - size1;
end_match_1 = end1; /* Just to give it a value. */
}
else
{
if (stop < size1)
{
/* Only match within string1. */
end_match_1 = string1 + stop;
/* BEWARE!
When we reach end_match_1, PREFETCH normally switches to string2.
But in the present case, this means that just doing a PREFETCH
makes us jump from `stop' to `gap' within the string.
What we really want here is for the search to stop as
soon as we hit end_match_1. That's why we set end_match_2
to end_match_1 (since PREFETCH fails as soon as we hit
end_match_2). */
end_match_2 = end_match_1;
}
else
{ /* It's important to use this code when stop == size so that
moving `d' from end1 to string2 will not prevent the d == dend
check from catching the end of string. */
end_match_1 = end1;
end_match_2 = string2 + stop - size1;
}
d = string1 + pos;
dend = end_match_1;
}
DEBUG_PRINT ("The compiled pattern is: ");
DEBUG_PRINT_COMPILED_PATTERN (bufp, p, pend);
DEBUG_PRINT ("The string to match is: \"");
DEBUG_PRINT_DOUBLE_STRING (d, string1, size1, string2, size2);
DEBUG_PRINT ("\"\n");
/* This loops over pattern commands. It exits by returning from the
function if the match is complete, or it drops through if the match
fails at this starting point in the input data. */
for (;;)
{
DEBUG_PRINT ("\n%p: ", p);
if (p == pend)
{
/* End of pattern means we might have succeeded. */
DEBUG_PRINT ("end of pattern ... ");
/* If we haven't matched the entire string, and we want the
longest match, try backtracking. */
if (d != end_match_2)
{
/* True if this match is the best seen so far. */
bool best_match_p;
{
/* True if this match ends in the same string (string1
or string2) as the best previous match. */
bool same_str_p = (FIRST_STRING_P (match_end)
== FIRST_STRING_P (d));
/* AIX compiler got confused when this was combined
with the previous declaration. */
if (same_str_p)
best_match_p = d > match_end;
else
best_match_p = !FIRST_STRING_P (d);
}
DEBUG_PRINT ("backtracking.\n");
if (!FAIL_STACK_EMPTY ())
{ /* More failure points to try. */
/* If exceeds best match so far, save it. */
if (!best_regs_set || best_match_p)
{
best_regs_set = true;
match_end = d;
DEBUG_PRINT ("\nSAVING match as best so far.\n");
for (reg = 1; reg < num_regs; reg++)
{
best_regstart[reg] = regstart[reg];
best_regend[reg] = regend[reg];
}
}
goto fail;
}
/* If no failure points, don't restore garbage. And if
last match is real best match, don't restore second
best one. */
else if (best_regs_set && !best_match_p)
{
restore_best_regs:
/* Restore best match. It may happen that `dend ==
end_match_1' while the restored d is in string2.
For example, the pattern `x.*y.*z' against the
strings `x-' and `y-z-', if the two strings are
not consecutive in memory. */
DEBUG_PRINT ("Restoring best registers.\n");
d = match_end;
dend = ((d >= string1 && d <= end1)
? end_match_1 : end_match_2);
for (reg = 1; reg < num_regs; reg++)
{
regstart[reg] = best_regstart[reg];
regend[reg] = best_regend[reg];
}
}
} /* d != end_match_2 */
succeed_label:
DEBUG_PRINT ("Accepting match.\n");
/* If caller wants register contents data back, do it. */
if (regs && !bufp->no_sub)
{
/* Have the register data arrays been allocated? */
if (bufp->regs_allocated == REGS_UNALLOCATED)
{ /* No. So allocate them with malloc. We need one
extra element beyond `num_regs' for the `-1' marker
GNU code uses. */
regs->num_regs = max (RE_NREGS, num_regs + 1);
regs->start = TALLOC (regs->num_regs, regoff_t);
regs->end = TALLOC (regs->num_regs, regoff_t);
if (regs->start == NULL || regs->end == NULL)
{
FREE_VARIABLES ();
return -2;
}
bufp->regs_allocated = REGS_REALLOCATE;
}
else if (bufp->regs_allocated == REGS_REALLOCATE)
{ /* Yes. If we need more elements than were already
allocated, reallocate them. If we need fewer, just
leave it alone. */
if (regs->num_regs < num_regs + 1)
{
regs->num_regs = num_regs + 1;
RETALLOC (regs->start, regs->num_regs, regoff_t);
RETALLOC (regs->end, regs->num_regs, regoff_t);
if (regs->start == NULL || regs->end == NULL)
{
FREE_VARIABLES ();
return -2;
}
}
}
else
{
/* These braces fend off a "empty body in an else-statement"
warning under GCC when assert expands to nothing. */
assert (bufp->regs_allocated == REGS_FIXED);
}
/* Convert the pointer data in `regstart' and `regend' to
indices. Register zero has to be set differently,
since we haven't kept track of any info for it. */
if (regs->num_regs > 0)
{
regs->start[0] = pos;
regs->end[0] = POINTER_TO_OFFSET (d);
}
/* Go through the first `min (num_regs, regs->num_regs)'
registers, since that is all we initialized. */
for (reg = 1; reg < min (num_regs, regs->num_regs); reg++)
{
if (REG_UNSET (regstart[reg]) || REG_UNSET (regend[reg]))
regs->start[reg] = regs->end[reg] = -1;
else
{
regs->start[reg] = POINTER_TO_OFFSET (regstart[reg]);
regs->end[reg] = POINTER_TO_OFFSET (regend[reg]);
}
}
/* If the regs structure we return has more elements than
were in the pattern, set the extra elements to -1. If
we (re)allocated the registers, this is the case,
because we always allocate enough to have at least one
-1 at the end. */
for (reg = num_regs; reg < regs->num_regs; reg++)
regs->start[reg] = regs->end[reg] = -1;
} /* regs && !bufp->no_sub */
DEBUG_PRINT ("%u failure points pushed, %u popped (%u remain).\n",
nfailure_points_pushed, nfailure_points_popped,
nfailure_points_pushed - nfailure_points_popped);
DEBUG_PRINT ("%u registers pushed.\n", num_regs_pushed);
ptrdiff_t dcnt = POINTER_TO_OFFSET (d) - pos;
DEBUG_PRINT ("Returning %td from re_match_2.\n", dcnt);
FREE_VARIABLES ();
return dcnt;
}
/* Otherwise match next pattern command. */
switch (*p++)
{
/* Ignore these. Used to ignore the n of succeed_n's which
currently have n == 0. */
case no_op:
DEBUG_PRINT ("EXECUTING no_op.\n");
break;
case succeed:
DEBUG_PRINT ("EXECUTING succeed.\n");
goto succeed_label;
/* Match the next n pattern characters exactly. The following
byte in the pattern defines n, and the n bytes after that
are the characters to match. */
case exactn:
mcnt = *p++;
DEBUG_PRINT ("EXECUTING exactn %d.\n", mcnt);
/* Remember the start point to rollback upon failure. */
dfail = d;
#ifndef emacs
/* This is written out as an if-else so we don't waste time
testing `translate' inside the loop. */
if (RE_TRANSLATE_P (translate))
do
{
PREFETCH ();
if (RE_TRANSLATE (translate, *d) != *p++)
{
d = dfail;
goto fail;
}
d++;
}
while (--mcnt);
else
do
{
PREFETCH ();
if (*d++ != *p++)
{
d = dfail;
goto fail;
}
}
while (--mcnt);
#else /* emacs */
/* The cost of testing `translate' is comparatively small. */
if (target_multibyte)
do
{
int pat_charlen, buf_charlen;
int pat_ch, buf_ch;
PREFETCH ();
if (multibyte)
pat_ch = STRING_CHAR_AND_LENGTH (p, pat_charlen);
else
{
pat_ch = RE_CHAR_TO_MULTIBYTE (*p);
pat_charlen = 1;
}
buf_ch = STRING_CHAR_AND_LENGTH (d, buf_charlen);
if (TRANSLATE (buf_ch) != pat_ch)
{
d = dfail;
goto fail;
}
p += pat_charlen;
d += buf_charlen;
mcnt -= pat_charlen;
}
while (mcnt > 0);
else
do
{
int pat_charlen;
int pat_ch, buf_ch;
PREFETCH ();
if (multibyte)
{
pat_ch = STRING_CHAR_AND_LENGTH (p, pat_charlen);
pat_ch = RE_CHAR_TO_UNIBYTE (pat_ch);
}
else
{
pat_ch = *p;
pat_charlen = 1;
}
buf_ch = RE_CHAR_TO_MULTIBYTE (*d);
if (! CHAR_BYTE8_P (buf_ch))
{
buf_ch = TRANSLATE (buf_ch);
buf_ch = RE_CHAR_TO_UNIBYTE (buf_ch);
if (buf_ch < 0)
buf_ch = *d;
}
else
buf_ch = *d;
if (buf_ch != pat_ch)
{
d = dfail;
goto fail;
}
p += pat_charlen;
d++;
}
while (--mcnt);
#endif
break;
/* Match any character except possibly a newline or a null. */
case anychar:
{
int buf_charlen;
re_wchar_t buf_ch;
reg_syntax_t syntax;
DEBUG_PRINT ("EXECUTING anychar.\n");
PREFETCH ();
buf_ch = RE_STRING_CHAR_AND_LENGTH (d, buf_charlen,
target_multibyte);
buf_ch = TRANSLATE (buf_ch);
#ifdef emacs
syntax = RE_SYNTAX_EMACS;
#else
syntax = bufp->syntax;
#endif
if ((!(syntax & RE_DOT_NEWLINE) && buf_ch == '\n')
|| ((syntax & RE_DOT_NOT_NULL) && buf_ch == '\000'))
goto fail;
DEBUG_PRINT (" Matched \"%d\".\n", *d);
d += buf_charlen;
}
break;
case charset:
case charset_not:
{
register unsigned int c, corig;
int len;
/* Whether matching against a unibyte character. */
boolean unibyte_char = false;
DEBUG_PRINT ("EXECUTING charset%s.\n",
(re_opcode_t) *(p - 1) == charset_not ? "_not" : "");
PREFETCH ();
corig = c = RE_STRING_CHAR_AND_LENGTH (d, len, target_multibyte);
if (target_multibyte)
{
int c1;
c = TRANSLATE (c);
c1 = RE_CHAR_TO_UNIBYTE (c);
if (c1 >= 0)
{
unibyte_char = true;
c = c1;
}
}
else
{
int c1 = RE_CHAR_TO_MULTIBYTE (c);
if (! CHAR_BYTE8_P (c1))
{
c1 = TRANSLATE (c1);
c1 = RE_CHAR_TO_UNIBYTE (c1);
if (c1 >= 0)
{
unibyte_char = true;
c = c1;
}
}
else
unibyte_char = true;
}
p -= 1;
if (!execute_charset (&p, c, corig, unibyte_char))
goto fail;
d += len;
}
break;
/* The beginning of a group is represented by start_memory.
The argument is the register number. The text
matched within the group is recorded (in the internal
registers data structure) under the register number. */
case start_memory:
DEBUG_PRINT ("EXECUTING start_memory %d:\n", *p);
/* In case we need to undo this operation (via backtracking). */
PUSH_FAILURE_REG (*p);
regstart[*p] = d;
regend[*p] = NULL; /* probably unnecessary. -sm */
DEBUG_PRINT (" regstart: %td\n", POINTER_TO_OFFSET (regstart[*p]));
/* Move past the register number and inner group count. */
p += 1;
break;
/* The stop_memory opcode represents the end of a group. Its
argument is the same as start_memory's: the register number. */
case stop_memory:
DEBUG_PRINT ("EXECUTING stop_memory %d:\n", *p);
assert (!REG_UNSET (regstart[*p]));
/* Strictly speaking, there should be code such as:
assert (REG_UNSET (regend[*p]));
PUSH_FAILURE_REGSTOP ((unsigned int)*p);
But the only info to be pushed is regend[*p] and it is known to
be UNSET, so there really isn't anything to push.
Not pushing anything, on the other hand deprives us from the
guarantee that regend[*p] is UNSET since undoing this operation
will not reset its value properly. This is not important since
the value will only be read on the next start_memory or at
the very end and both events can only happen if this stop_memory
is *not* undone. */
regend[*p] = d;
DEBUG_PRINT (" regend: %td\n", POINTER_TO_OFFSET (regend[*p]));
/* Move past the register number and the inner group count. */
p += 1;
break;
/* \<digit> has been turned into a `duplicate' command which is
followed by the numeric value of <digit> as the register number. */
case duplicate:
{
register re_char *d2, *dend2;
int regno = *p++; /* Get which register to match against. */
DEBUG_PRINT ("EXECUTING duplicate %d.\n", regno);
/* Can't back reference a group which we've never matched. */
if (REG_UNSET (regstart[regno]) || REG_UNSET (regend[regno]))
goto fail;
/* Where in input to try to start matching. */
d2 = regstart[regno];
/* Remember the start point to rollback upon failure. */
dfail = d;
/* Where to stop matching; if both the place to start and
the place to stop matching are in the same string, then
set to the place to stop, otherwise, for now have to use
the end of the first string. */
dend2 = ((FIRST_STRING_P (regstart[regno])
== FIRST_STRING_P (regend[regno]))
? regend[regno] : end_match_1);
for (;;)
{
ptrdiff_t dcnt;
/* If necessary, advance to next segment in register
contents. */
while (d2 == dend2)
{
if (dend2 == end_match_2) break;
if (dend2 == regend[regno]) break;
/* End of string1 => advance to string2. */
d2 = string2;
dend2 = regend[regno];
}
/* At end of register contents => success */
if (d2 == dend2) break;
/* If necessary, advance to next segment in data. */
PREFETCH ();
/* How many characters left in this segment to match. */
dcnt = dend - d;
/* Want how many consecutive characters we can match in
one shot, so, if necessary, adjust the count. */
if (dcnt > dend2 - d2)
dcnt = dend2 - d2;
/* Compare that many; failure if mismatch, else move
past them. */
if (RE_TRANSLATE_P (translate)
? bcmp_translate (d, d2, dcnt, translate, target_multibyte)
: memcmp (d, d2, dcnt))
{
d = dfail;
goto fail;
}
d += dcnt, d2 += dcnt;
}
}
break;
/* begline matches the empty string at the beginning of the string
(unless `not_bol' is set in `bufp'), and after newlines. */
case begline:
DEBUG_PRINT ("EXECUTING begline.\n");
if (AT_STRINGS_BEG (d))
{
if (!bufp->not_bol) break;
}
else
{
unsigned c;
GET_CHAR_BEFORE_2 (c, d, string1, end1, string2, end2);
if (c == '\n')
break;
}
/* In all other cases, we fail. */
goto fail;
/* endline is the dual of begline. */
case endline:
DEBUG_PRINT ("EXECUTING endline.\n");
if (AT_STRINGS_END (d))
{
if (!bufp->not_eol) break;
}
else
{
PREFETCH_NOLIMIT ();
if (*d == '\n')
break;
}
goto fail;
/* Match at the very beginning of the data. */
case begbuf:
DEBUG_PRINT ("EXECUTING begbuf.\n");
if (AT_STRINGS_BEG (d))
break;
goto fail;
/* Match at the very end of the data. */
case endbuf:
DEBUG_PRINT ("EXECUTING endbuf.\n");
if (AT_STRINGS_END (d))
break;
goto fail;
/* on_failure_keep_string_jump is used to optimize `.*\n'. It
pushes NULL as the value for the string on the stack. Then
`POP_FAILURE_POINT' will keep the current value for the
string, instead of restoring it. To see why, consider
matching `foo\nbar' against `.*\n'. The .* matches the foo;
then the . fails against the \n. But the next thing we want
to do is match the \n against the \n; if we restored the
string value, we would be back at the foo.
Because this is used only in specific cases, we don't need to
check all the things that `on_failure_jump' does, to make
sure the right things get saved on the stack. Hence we don't
share its code. The only reason to push anything on the
stack at all is that otherwise we would have to change
`anychar's code to do something besides goto fail in this
case; that seems worse than this. */
case on_failure_keep_string_jump:
EXTRACT_NUMBER_AND_INCR (mcnt, p);
DEBUG_PRINT ("EXECUTING on_failure_keep_string_jump %d (to %p):\n",
mcnt, p + mcnt);
PUSH_FAILURE_POINT (p - 3, NULL);
break;
/* A nasty loop is introduced by the non-greedy *? and +?.
With such loops, the stack only ever contains one failure point
at a time, so that a plain on_failure_jump_loop kind of
cycle detection cannot work. Worse yet, such a detection
can not only fail to detect a cycle, but it can also wrongly
detect a cycle (between different instantiations of the same
loop).
So the method used for those nasty loops is a little different:
We use a special cycle-detection-stack-frame which is pushed
when the on_failure_jump_nastyloop failure-point is *popped*.
This special frame thus marks the beginning of one iteration
through the loop and we can hence easily check right here
whether something matched between the beginning and the end of
the loop. */
case on_failure_jump_nastyloop:
EXTRACT_NUMBER_AND_INCR (mcnt, p);
DEBUG_PRINT ("EXECUTING on_failure_jump_nastyloop %d (to %p):\n",
mcnt, p + mcnt);
assert ((re_opcode_t)p[-4] == no_op);
{
int cycle = 0;
CHECK_INFINITE_LOOP (p - 4, d);
if (!cycle)
/* If there's a cycle, just continue without pushing
this failure point. The failure point is the "try again"
option, which shouldn't be tried.
We want (x?)*?y\1z to match both xxyz and xxyxz. */
PUSH_FAILURE_POINT (p - 3, d);
}
break;
/* Simple loop detecting on_failure_jump: just check on the
failure stack if the same spot was already hit earlier. */
case on_failure_jump_loop:
on_failure:
EXTRACT_NUMBER_AND_INCR (mcnt, p);
DEBUG_PRINT ("EXECUTING on_failure_jump_loop %d (to %p):\n",
mcnt, p + mcnt);
{
int cycle = 0;
CHECK_INFINITE_LOOP (p - 3, d);
if (cycle)
/* If there's a cycle, get out of the loop, as if the matching
had failed. We used to just `goto fail' here, but that was
aborting the search a bit too early: we want to keep the
empty-loop-match and keep matching after the loop.
We want (x?)*y\1z to match both xxyz and xxyxz. */
p += mcnt;
else
PUSH_FAILURE_POINT (p - 3, d);
}
break;
/* Uses of on_failure_jump:
Each alternative starts with an on_failure_jump that points
to the beginning of the next alternative. Each alternative
except the last ends with a jump that in effect jumps past
the rest of the alternatives. (They really jump to the
ending jump of the following alternative, because tensioning
these jumps is a hassle.)
Repeats start with an on_failure_jump that points past both
the repetition text and either the following jump or
pop_failure_jump back to this on_failure_jump. */
case on_failure_jump:
EXTRACT_NUMBER_AND_INCR (mcnt, p);
DEBUG_PRINT ("EXECUTING on_failure_jump %d (to %p):\n",
mcnt, p + mcnt);
PUSH_FAILURE_POINT (p -3, d);
break;
/* This operation is used for greedy *.
Compare the beginning of the repeat with what in the
pattern follows its end. If we can establish that there
is nothing that they would both match, i.e., that we
would have to backtrack because of (as in, e.g., `a*a')
then we can use a non-backtracking loop based on
on_failure_keep_string_jump instead of on_failure_jump. */
case on_failure_jump_smart:
EXTRACT_NUMBER_AND_INCR (mcnt, p);
DEBUG_PRINT ("EXECUTING on_failure_jump_smart %d (to %p).\n",
mcnt, p + mcnt);
{
re_char *p1 = p; /* Next operation. */
/* Here, we discard `const', making re_match non-reentrant. */
unsigned char *p2 = (unsigned char*) p + mcnt; /* Jump dest. */
unsigned char *p3 = (unsigned char*) p - 3; /* opcode location. */
p -= 3; /* Reset so that we will re-execute the
instruction once it's been changed. */
EXTRACT_NUMBER (mcnt, p2 - 2);
/* Ensure this is a indeed the trivial kind of loop
we are expecting. */
assert (skip_one_char (p1) == p2 - 3);
assert ((re_opcode_t) p2[-3] == jump && p2 + mcnt == p);
DEBUG_STATEMENT (debug += 2);
if (mutually_exclusive_p (bufp, p1, p2))
{
/* Use a fast `on_failure_keep_string_jump' loop. */
DEBUG_PRINT (" smart exclusive => fast loop.\n");
*p3 = (unsigned char) on_failure_keep_string_jump;
STORE_NUMBER (p2 - 2, mcnt + 3);
}
else
{
/* Default to a safe `on_failure_jump' loop. */
DEBUG_PRINT (" smart default => slow loop.\n");
*p3 = (unsigned char) on_failure_jump;
}
DEBUG_STATEMENT (debug -= 2);
}
break;
/* Unconditionally jump (without popping any failure points). */
case jump:
unconditional_jump:
IMMEDIATE_QUIT_CHECK;
EXTRACT_NUMBER_AND_INCR (mcnt, p); /* Get the amount to jump. */
DEBUG_PRINT ("EXECUTING jump %d ", mcnt);
p += mcnt; /* Do the jump. */
DEBUG_PRINT ("(to %p).\n", p);
break;
/* Have to succeed matching what follows at least n times.
After that, handle like `on_failure_jump'. */
case succeed_n:
/* Signedness doesn't matter since we only compare MCNT to 0. */
EXTRACT_NUMBER (mcnt, p + 2);
DEBUG_PRINT ("EXECUTING succeed_n %d.\n", mcnt);
/* Originally, mcnt is how many times we HAVE to succeed. */
if (mcnt != 0)
{
/* Here, we discard `const', making re_match non-reentrant. */
unsigned char *p2 = (unsigned char*) p + 2; /* counter loc. */
mcnt--;
p += 4;
PUSH_NUMBER (p2, mcnt);
}
else
/* The two bytes encoding mcnt == 0 are two no_op opcodes. */
goto on_failure;
break;
case jump_n:
/* Signedness doesn't matter since we only compare MCNT to 0. */
EXTRACT_NUMBER (mcnt, p + 2);
DEBUG_PRINT ("EXECUTING jump_n %d.\n", mcnt);
/* Originally, this is how many times we CAN jump. */
if (mcnt != 0)
{
/* Here, we discard `const', making re_match non-reentrant. */
unsigned char *p2 = (unsigned char*) p + 2; /* counter loc. */
mcnt--;
PUSH_NUMBER (p2, mcnt);
goto unconditional_jump;
}
/* If don't have to jump any more, skip over the rest of command. */
else
p += 4;
break;
case set_number_at:
{
unsigned char *p2; /* Location of the counter. */
DEBUG_PRINT ("EXECUTING set_number_at.\n");
EXTRACT_NUMBER_AND_INCR (mcnt, p);
/* Here, we discard `const', making re_match non-reentrant. */
p2 = (unsigned char*) p + mcnt;
/* Signedness doesn't matter since we only copy MCNT's bits. */
EXTRACT_NUMBER_AND_INCR (mcnt, p);
DEBUG_PRINT (" Setting %p to %d.\n", p2, mcnt);
PUSH_NUMBER (p2, mcnt);
break;
}
case wordbound:
case notwordbound:
{
boolean not = (re_opcode_t) *(p - 1) == notwordbound;
DEBUG_PRINT ("EXECUTING %swordbound.\n", not ? "not" : "");
/* We SUCCEED (or FAIL) in one of the following cases: */
/* Case 1: D is at the beginning or the end of string. */
if (AT_STRINGS_BEG (d) || AT_STRINGS_END (d))
not = !not;
else
{
/* C1 is the character before D, S1 is the syntax of C1, C2
is the character at D, and S2 is the syntax of C2. */
re_wchar_t c1, c2;
int s1, s2;
int dummy;
#ifdef emacs
ssize_t offset = PTR_TO_OFFSET (d - 1);
ssize_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
UPDATE_SYNTAX_TABLE_FAST (charpos);
#endif
GET_CHAR_BEFORE_2 (c1, d, string1, end1, string2, end2);
s1 = SYNTAX (c1);
#ifdef emacs
UPDATE_SYNTAX_TABLE_FORWARD_FAST (charpos + 1);
#endif
PREFETCH_NOLIMIT ();
GET_CHAR_AFTER (c2, d, dummy);
s2 = SYNTAX (c2);
if (/* Case 2: Only one of S1 and S2 is Sword. */
((s1 == Sword) != (s2 == Sword))
/* Case 3: Both of S1 and S2 are Sword, and macro
WORD_BOUNDARY_P (C1, C2) returns nonzero. */
|| ((s1 == Sword) && WORD_BOUNDARY_P (c1, c2)))
not = !not;
}
if (not)
break;
else
goto fail;
}
case wordbeg:
DEBUG_PRINT ("EXECUTING wordbeg.\n");
/* We FAIL in one of the following cases: */
/* Case 1: D is at the end of string. */
if (AT_STRINGS_END (d))
goto fail;
else
{
/* C1 is the character before D, S1 is the syntax of C1, C2
is the character at D, and S2 is the syntax of C2. */
re_wchar_t c1, c2;
int s1, s2;
int dummy;
#ifdef emacs
ssize_t offset = PTR_TO_OFFSET (d);
ssize_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
UPDATE_SYNTAX_TABLE_FAST (charpos);
#endif
PREFETCH ();
GET_CHAR_AFTER (c2, d, dummy);
s2 = SYNTAX (c2);
/* Case 2: S2 is not Sword. */
if (s2 != Sword)
goto fail;
/* Case 3: D is not at the beginning of string ... */
if (!AT_STRINGS_BEG (d))
{
GET_CHAR_BEFORE_2 (c1, d, string1, end1, string2, end2);
#ifdef emacs
UPDATE_SYNTAX_TABLE_BACKWARD (charpos - 1);
#endif
s1 = SYNTAX (c1);
/* ... and S1 is Sword, and WORD_BOUNDARY_P (C1, C2)
returns 0. */
if ((s1 == Sword) && !WORD_BOUNDARY_P (c1, c2))
goto fail;
}
}
break;
case wordend:
DEBUG_PRINT ("EXECUTING wordend.\n");
/* We FAIL in one of the following cases: */
/* Case 1: D is at the beginning of string. */
if (AT_STRINGS_BEG (d))
goto fail;
else
{
/* C1 is the character before D, S1 is the syntax of C1, C2
is the character at D, and S2 is the syntax of C2. */
re_wchar_t c1, c2;
int s1, s2;
int dummy;
#ifdef emacs
ssize_t offset = PTR_TO_OFFSET (d) - 1;
ssize_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
UPDATE_SYNTAX_TABLE_FAST (charpos);
#endif
GET_CHAR_BEFORE_2 (c1, d, string1, end1, string2, end2);
s1 = SYNTAX (c1);
/* Case 2: S1 is not Sword. */
if (s1 != Sword)
goto fail;
/* Case 3: D is not at the end of string ... */
if (!AT_STRINGS_END (d))
{
PREFETCH_NOLIMIT ();
GET_CHAR_AFTER (c2, d, dummy);
#ifdef emacs
UPDATE_SYNTAX_TABLE_FORWARD_FAST (charpos);
#endif
s2 = SYNTAX (c2);
/* ... and S2 is Sword, and WORD_BOUNDARY_P (C1, C2)
returns 0. */
if ((s2 == Sword) && !WORD_BOUNDARY_P (c1, c2))
goto fail;
}
}
break;
case symbeg:
DEBUG_PRINT ("EXECUTING symbeg.\n");
/* We FAIL in one of the following cases: */
/* Case 1: D is at the end of string. */
if (AT_STRINGS_END (d))
goto fail;
else
{
/* C1 is the character before D, S1 is the syntax of C1, C2
is the character at D, and S2 is the syntax of C2. */
re_wchar_t c1, c2;
int s1, s2;
#ifdef emacs
ssize_t offset = PTR_TO_OFFSET (d);
ssize_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
UPDATE_SYNTAX_TABLE_FAST (charpos);
#endif
PREFETCH ();
c2 = RE_STRING_CHAR (d, target_multibyte);
s2 = SYNTAX (c2);
/* Case 2: S2 is neither Sword nor Ssymbol. */
if (s2 != Sword && s2 != Ssymbol)
goto fail;
/* Case 3: D is not at the beginning of string ... */
if (!AT_STRINGS_BEG (d))
{
GET_CHAR_BEFORE_2 (c1, d, string1, end1, string2, end2);
#ifdef emacs
UPDATE_SYNTAX_TABLE_BACKWARD (charpos - 1);
#endif
s1 = SYNTAX (c1);
/* ... and S1 is Sword or Ssymbol. */
if (s1 == Sword || s1 == Ssymbol)
goto fail;
}
}
break;
case symend:
DEBUG_PRINT ("EXECUTING symend.\n");
/* We FAIL in one of the following cases: */
/* Case 1: D is at the beginning of string. */
if (AT_STRINGS_BEG (d))
goto fail;
else
{
/* C1 is the character before D, S1 is the syntax of C1, C2
is the character at D, and S2 is the syntax of C2. */
re_wchar_t c1, c2;
int s1, s2;
#ifdef emacs
ssize_t offset = PTR_TO_OFFSET (d) - 1;
ssize_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
UPDATE_SYNTAX_TABLE_FAST (charpos);
#endif
GET_CHAR_BEFORE_2 (c1, d, string1, end1, string2, end2);
s1 = SYNTAX (c1);
/* Case 2: S1 is neither Ssymbol nor Sword. */
if (s1 != Sword && s1 != Ssymbol)
goto fail;
/* Case 3: D is not at the end of string ... */
if (!AT_STRINGS_END (d))
{
PREFETCH_NOLIMIT ();
c2 = RE_STRING_CHAR (d, target_multibyte);
#ifdef emacs
UPDATE_SYNTAX_TABLE_FORWARD_FAST (charpos + 1);
#endif
s2 = SYNTAX (c2);
/* ... and S2 is Sword or Ssymbol. */
if (s2 == Sword || s2 == Ssymbol)
goto fail;
}
}
break;
case syntaxspec:
case notsyntaxspec:
{
boolean not = (re_opcode_t) *(p - 1) == notsyntaxspec;
mcnt = *p++;
DEBUG_PRINT ("EXECUTING %ssyntaxspec %d.\n", not ? "not" : "",
mcnt);
PREFETCH ();
#ifdef emacs
{
ssize_t offset = PTR_TO_OFFSET (d);
ssize_t pos1 = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
UPDATE_SYNTAX_TABLE_FAST (pos1);
}
#endif
{
int len;
re_wchar_t c;
GET_CHAR_AFTER (c, d, len);
if ((SYNTAX (c) != (enum syntaxcode) mcnt) ^ not)
goto fail;
d += len;
}
}
break;
#ifdef emacs
case at_dot:
DEBUG_PRINT ("EXECUTING at_dot.\n");
if (PTR_BYTE_POS (d) != PT_BYTE)
goto fail;
break;
case categoryspec:
case notcategoryspec:
{
boolean not = (re_opcode_t) *(p - 1) == notcategoryspec;
mcnt = *p++;
DEBUG_PRINT ("EXECUTING %scategoryspec %d.\n",
not ? "not" : "", mcnt);
PREFETCH ();
{
int len;
re_wchar_t c;
GET_CHAR_AFTER (c, d, len);
if ((!CHAR_HAS_CATEGORY (c, mcnt)) ^ not)
goto fail;
d += len;
}
}
break;
#endif /* emacs */
default:
abort ();
}
continue; /* Successfully executed one pattern command; keep going. */
/* We goto here if a matching operation fails. */
fail:
IMMEDIATE_QUIT_CHECK;
if (!FAIL_STACK_EMPTY ())
{
re_char *str, *pat;
/* A restart point is known. Restore to that state. */
DEBUG_PRINT ("\nFAIL:\n");
POP_FAILURE_POINT (str, pat);
switch (*pat++)
{
case on_failure_keep_string_jump:
assert (str == NULL);
goto continue_failure_jump;
case on_failure_jump_nastyloop:
assert ((re_opcode_t)pat[-2] == no_op);
PUSH_FAILURE_POINT (pat - 2, str);
/* Fallthrough */
case on_failure_jump_loop:
case on_failure_jump:
case succeed_n:
d = str;
continue_failure_jump:
EXTRACT_NUMBER_AND_INCR (mcnt, pat);
p = pat + mcnt;
break;
case no_op:
/* A special frame used for nastyloops. */
goto fail;
default:
abort ();
}
assert (p >= bufp->buffer && p <= pend);
if (d >= string1 && d <= end1)
dend = end_match_1;
}
else
break; /* Matching at this starting point really fails. */
} /* for (;;) */
if (best_regs_set)
goto restore_best_regs;
FREE_VARIABLES ();
return -1; /* Failure to match. */
}
\f
/* Subroutine definitions for re_match_2. */
/* Return zero if TRANSLATE[S1] and TRANSLATE[S2] are identical for LEN
bytes; nonzero otherwise. */
static int
bcmp_translate (const_re_char *s1, const_re_char *s2, register ssize_t len,
RE_TRANSLATE_TYPE translate, const int target_multibyte)
{
register re_char *p1 = s1, *p2 = s2;
re_char *p1_end = s1 + len;
re_char *p2_end = s2 + len;
/* FIXME: Checking both p1 and p2 presumes that the two strings might have
different lengths, but relying on a single `len' would break this. -sm */
while (p1 < p1_end && p2 < p2_end)
{
int p1_charlen, p2_charlen;
re_wchar_t p1_ch, p2_ch;
GET_CHAR_AFTER (p1_ch, p1, p1_charlen);
GET_CHAR_AFTER (p2_ch, p2, p2_charlen);
if (RE_TRANSLATE (translate, p1_ch)
!= RE_TRANSLATE (translate, p2_ch))
return 1;
p1 += p1_charlen, p2 += p2_charlen;
}
if (p1 != p1_end || p2 != p2_end)
return 1;
return 0;
}
\f
/* Entry points for GNU code. */
/* re_compile_pattern is the GNU regular expression compiler: it
compiles PATTERN (of length SIZE) and puts the result in BUFP.
Returns 0 if the pattern was valid, otherwise an error string.
Assumes the `allocated' (and perhaps `buffer') and `translate' fields
are set in BUFP on entry.
We call regex_compile to do the actual compilation. */
const char *
re_compile_pattern (const char *pattern, size_t length,
#ifdef emacs
bool posix_backtracking, const char *whitespace_regexp,
#endif
struct re_pattern_buffer *bufp)
{
reg_errcode_t ret;
/* GNU code is written to assume at least RE_NREGS registers will be set
(and at least one extra will be -1). */
bufp->regs_allocated = REGS_UNALLOCATED;
/* And GNU code determines whether or not to get register information
by passing null for the REGS argument to re_match, etc., not by
setting no_sub. */
bufp->no_sub = 0;
ret = regex_compile ((re_char*) pattern, length,
#ifdef emacs
posix_backtracking,
whitespace_regexp,
#else
re_syntax_options,
#endif
bufp);
if (!ret)
return NULL;
return gettext (re_error_msgid[(int) ret]);
}
WEAK_ALIAS (__re_compile_pattern, re_compile_pattern)
\f
/* Entry points compatible with 4.2 BSD regex library. We don't define
them unless specifically requested. */
#if defined _REGEX_RE_COMP || defined _LIBC
/* BSD has one and only one pattern buffer. */
static struct re_pattern_buffer re_comp_buf;
char *
# ifdef _LIBC
/* Make these definitions weak in libc, so POSIX programs can redefine
these names if they don't use our functions, and still use
regcomp/regexec below without link errors. */
weak_function
# endif
re_comp (const char *s)
{
reg_errcode_t ret;
if (!s)
{
if (!re_comp_buf.buffer)
/* Yes, we're discarding `const' here if !HAVE_LIBINTL. */
return (char *) gettext ("No previous regular expression");
return 0;
}
if (!re_comp_buf.buffer)
{
re_comp_buf.buffer = malloc (200);
if (re_comp_buf.buffer == NULL)
/* Yes, we're discarding `const' here if !HAVE_LIBINTL. */
return (char *) gettext (re_error_msgid[(int) REG_ESPACE]);
re_comp_buf.allocated = 200;
re_comp_buf.fastmap = malloc (1 << BYTEWIDTH);
if (re_comp_buf.fastmap == NULL)
/* Yes, we're discarding `const' here if !HAVE_LIBINTL. */
return (char *) gettext (re_error_msgid[(int) REG_ESPACE]);
}
/* Since `re_exec' always passes NULL for the `regs' argument, we
don't need to initialize the pattern buffer fields which affect it. */
ret = regex_compile (s, strlen (s), re_syntax_options, &re_comp_buf);
if (!ret)
return NULL;
/* Yes, we're discarding `const' here if !HAVE_LIBINTL. */
return (char *) gettext (re_error_msgid[(int) ret]);
}
int
# ifdef _LIBC
weak_function
# endif
re_exec (const char *s)
{
const size_t len = strlen (s);
return re_search (&re_comp_buf, s, len, 0, len, 0) >= 0;
}
#endif /* _REGEX_RE_COMP */
\f
/* POSIX.2 functions. Don't define these for Emacs. */
#ifndef emacs
/* regcomp takes a regular expression as a string and compiles it.
PREG is a regex_t *. We do not expect any fields to be initialized,
since POSIX says we shouldn't. Thus, we set
`buffer' to the compiled pattern;
`used' to the length of the compiled pattern;
`syntax' to RE_SYNTAX_POSIX_EXTENDED if the
REG_EXTENDED bit in CFLAGS is set; otherwise, to
RE_SYNTAX_POSIX_BASIC;
`fastmap' to an allocated space for the fastmap;
`fastmap_accurate' to zero;
`re_nsub' to the number of subexpressions in PATTERN.
PATTERN is the address of the pattern string.
CFLAGS is a series of bits which affect compilation.
If REG_EXTENDED is set, we use POSIX extended syntax; otherwise, we
use POSIX basic syntax.
If REG_NEWLINE is set, then . and [^...] don't match newline.
Also, regexec will try a match beginning after every newline.
If REG_ICASE is set, then we considers upper- and lowercase
versions of letters to be equivalent when matching.
If REG_NOSUB is set, then when PREG is passed to regexec, that
routine will report only success or failure, and nothing about the
registers.
It returns 0 if it succeeds, nonzero if it doesn't. (See regex.h for
the return codes and their meanings.) */
reg_errcode_t
regcomp (regex_t *_Restrict_ preg, const char *_Restrict_ pattern,
int cflags)
{
reg_errcode_t ret;
reg_syntax_t syntax
= (cflags & REG_EXTENDED) ?
RE_SYNTAX_POSIX_EXTENDED : RE_SYNTAX_POSIX_BASIC;
/* regex_compile will allocate the space for the compiled pattern. */
preg->buffer = 0;
preg->allocated = 0;
preg->used = 0;
/* Try to allocate space for the fastmap. */
preg->fastmap = malloc (1 << BYTEWIDTH);
if (cflags & REG_ICASE)
{
unsigned i;
preg->translate = malloc (CHAR_SET_SIZE * sizeof *preg->translate);
if (preg->translate == NULL)
return (int) REG_ESPACE;
/* Map uppercase characters to corresponding lowercase ones. */
for (i = 0; i < CHAR_SET_SIZE; i++)
preg->translate[i] = ISUPPER (i) ? TOLOWER (i) : i;
}
else
preg->translate = NULL;
/* If REG_NEWLINE is set, newlines are treated differently. */
if (cflags & REG_NEWLINE)
{ /* REG_NEWLINE implies neither . nor [^...] match newline. */
syntax &= ~RE_DOT_NEWLINE;
syntax |= RE_HAT_LISTS_NOT_NEWLINE;
}
else
syntax |= RE_NO_NEWLINE_ANCHOR;
preg->no_sub = !!(cflags & REG_NOSUB);
/* POSIX says a null character in the pattern terminates it, so we
can use strlen here in compiling the pattern. */
ret = regex_compile ((re_char*) pattern, strlen (pattern), syntax, preg);
/* POSIX doesn't distinguish between an unmatched open-group and an
unmatched close-group: both are REG_EPAREN. */
if (ret == REG_ERPAREN)
ret = REG_EPAREN;
if (ret == REG_NOERROR && preg->fastmap)
{ /* Compute the fastmap now, since regexec cannot modify the pattern
buffer. */
re_compile_fastmap (preg);
if (preg->can_be_null)
{ /* The fastmap can't be used anyway. */
free (preg->fastmap);
preg->fastmap = NULL;
}
}
return ret;
}
WEAK_ALIAS (__regcomp, regcomp)
/* regexec searches for a given pattern, specified by PREG, in the
string STRING.
If NMATCH is zero or REG_NOSUB was set in the cflags argument to
`regcomp', we ignore PMATCH. Otherwise, we assume PMATCH has at
least NMATCH elements, and we set them to the offsets of the
corresponding matched substrings.
EFLAGS specifies `execution flags' which affect matching: if
REG_NOTBOL is set, then ^ does not match at the beginning of the
string; if REG_NOTEOL is set, then $ does not match at the end.
We return 0 if we find a match and REG_NOMATCH if not. */
reg_errcode_t
regexec (const regex_t *_Restrict_ preg, const char *_Restrict_ string,
size_t nmatch, regmatch_t pmatch[_Restrict_arr_], int eflags)
{
regoff_t ret;
struct re_registers regs;
regex_t private_preg;
size_t len = strlen (string);
boolean want_reg_info = !preg->no_sub && nmatch > 0 && pmatch;
private_preg = *preg;
private_preg.not_bol = !!(eflags & REG_NOTBOL);
private_preg.not_eol = !!(eflags & REG_NOTEOL);
/* The user has told us exactly how many registers to return
information about, via `nmatch'. We have to pass that on to the
matching routines. */
private_preg.regs_allocated = REGS_FIXED;
if (want_reg_info)
{
regs.num_regs = nmatch;
regs.start = TALLOC (nmatch * 2, regoff_t);
if (regs.start == NULL)
return REG_NOMATCH;
regs.end = regs.start + nmatch;
}
/* Instead of using not_eol to implement REG_NOTEOL, we could simply
pass (&private_preg, string, len + 1, 0, len, ...) pretending the string
was a little bit longer but still only matching the real part.
This works because the `endline' will check for a '\n' and will find a
'\0', correctly deciding that this is not the end of a line.
But it doesn't work out so nicely for REG_NOTBOL, since we don't have
a convenient '\0' there. For all we know, the string could be preceded
by '\n' which would throw things off. */
/* Perform the searching operation. */
ret = re_search (&private_preg, string, len,
/* start: */ 0, /* range: */ len,
want_reg_info ? ®s : 0);
/* Copy the register information to the POSIX structure. */
if (want_reg_info)
{
if (ret >= 0)
{
unsigned r;
for (r = 0; r < nmatch; r++)
{
pmatch[r].rm_so = regs.start[r];
pmatch[r].rm_eo = regs.end[r];
}
}
/* If we needed the temporary register info, free the space now. */
free (regs.start);
}
/* We want zero return to mean success, unlike `re_search'. */
return ret >= 0 ? REG_NOERROR : REG_NOMATCH;
}
WEAK_ALIAS (__regexec, regexec)
/* Returns a message corresponding to an error code, ERR_CODE, returned
from either regcomp or regexec. We don't use PREG here.
ERR_CODE was previously called ERRCODE, but that name causes an
error with msvc8 compiler. */
size_t
regerror (int err_code, const regex_t *preg, char *errbuf, size_t errbuf_size)
{
const char *msg;
size_t msg_size;
if (err_code < 0
|| err_code >= (sizeof (re_error_msgid) / sizeof (re_error_msgid[0])))
/* Only error codes returned by the rest of the code should be passed
to this routine. If we are given anything else, or if other regex
code generates an invalid error code, then the program has a bug.
Dump core so we can fix it. */
abort ();
msg = gettext (re_error_msgid[err_code]);
msg_size = strlen (msg) + 1; /* Includes the null. */
if (errbuf_size != 0)
{
if (msg_size > errbuf_size)
{
memcpy (errbuf, msg, errbuf_size - 1);
errbuf[errbuf_size - 1] = 0;
}
else
strcpy (errbuf, msg);
}
return msg_size;
}
WEAK_ALIAS (__regerror, regerror)
/* Free dynamically allocated space used by PREG. */
void
regfree (regex_t *preg)
{
free (preg->buffer);
preg->buffer = NULL;
preg->allocated = 0;
preg->used = 0;
free (preg->fastmap);
preg->fastmap = NULL;
preg->fastmap_accurate = 0;
free (preg->translate);
preg->translate = NULL;
}
WEAK_ALIAS (__regfree, regfree)
#endif /* not emacs */
|