summaryrefslogtreecommitdiffstats
path: root/src/corelib/tools/qregexp.cpp
blob: 8c3662a1fddf7e7f8bf80d492504b2923a990069 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
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
/****************************************************************************
**
** Copyright (C) 2009 Nokia Corporation and/or its subsidiary(-ies).
** Contact: Nokia Corporation (qt-info@nokia.com)
**
** This file is part of the QtCore module of the Qt Toolkit.
**
** $QT_BEGIN_LICENSE:LGPL$
** No Commercial Usage
** This file contains pre-release code and may not be distributed.
** You may use this file in accordance with the terms and conditions
** contained in the either Technology Preview License Agreement or the
** Beta Release License Agreement.
**
** GNU Lesser General Public License Usage
** Alternatively, this file may be used under the terms of the GNU Lesser
** General Public License version 2.1 as published by the Free Software
** Foundation and appearing in the file LICENSE.LGPL included in the
** packaging of this file.  Please review the following information to
** ensure the GNU Lesser General Public License version 2.1 requirements
** will be met: http://www.gnu.org/licenses/old-licenses/lgpl-2.1.html.
**
** In addition, as a special exception, Nokia gives you certain
** additional rights. These rights are described in the Nokia Qt LGPL
** Exception version 1.0, included in the file LGPL_EXCEPTION.txt in this
** package.
**
** GNU General Public License Usage
** Alternatively, this file may be used under the terms of the GNU
** General Public License version 3.0 as published by the Free Software
** Foundation and appearing in the file LICENSE.GPL included in the
** packaging of this file.  Please review the following information to
** ensure the GNU General Public License version 3.0 requirements will be
** met: http://www.gnu.org/copyleft/gpl.html.
**
** If you are unsure which license is appropriate for your use, please
** contact the sales department at http://qt.nokia.com/contact.
** $QT_END_LICENSE$
**
****************************************************************************/

#include "qregexp.h"

#include "qalgorithms.h"
#include "qbitarray.h"
#include "qcache.h"
#include "qdatastream.h"
#include "qlist.h"
#include "qmap.h"
#include "qmutex.h"
#include "qstring.h"
#include "qstringlist.h"
#include "qstringmatcher.h"
#include "qvector.h"

#include <limits.h>

QT_BEGIN_NAMESPACE

int qFindString(const QChar *haystack, int haystackLen, int from,
    const QChar *needle, int needleLen, Qt::CaseSensitivity cs);

// error strings for the regexp parser
#define RXERR_OK         QT_TRANSLATE_NOOP("QRegExp", "no error occurred")
#define RXERR_DISABLED   QT_TRANSLATE_NOOP("QRegExp", "disabled feature used")
#define RXERR_CHARCLASS  QT_TRANSLATE_NOOP("QRegExp", "bad char class syntax")
#define RXERR_LOOKAHEAD  QT_TRANSLATE_NOOP("QRegExp", "bad lookahead syntax")
#define RXERR_REPETITION QT_TRANSLATE_NOOP("QRegExp", "bad repetition syntax")
#define RXERR_OCTAL      QT_TRANSLATE_NOOP("QRegExp", "invalid octal value")
#define RXERR_LEFTDELIM  QT_TRANSLATE_NOOP("QRegExp", "missing left delim")
#define RXERR_END        QT_TRANSLATE_NOOP("QRegExp", "unexpected end")
#define RXERR_LIMIT      QT_TRANSLATE_NOOP("QRegExp", "met internal limit")
#define RXERR_INTERVAL   QT_TRANSLATE_NOOP("QRegExp", "invalid interval")
#define RXERR_CATEGORY   QT_TRANSLATE_NOOP("QRegExp", "invalid category")

/*
  WARNING! Be sure to read qregexp.tex before modifying this file.
*/

/*!
    \class QRegExp
    \reentrant
    \brief The QRegExp class provides pattern matching using regular expressions.

    \ingroup tools
    \ingroup misc
    \ingroup shared
    \mainclass
    \keyword regular expression

    A regular expression, or "regexp", is a pattern for matching
    substrings in a text. This is useful in many contexts, e.g.,

    \table
    \row \i Validation
         \i A regexp can test whether a substring meets some criteria,
         e.g. is an integer or contains no whitespace.
    \row \i Searching
         \i A regexp provides more powerful pattern matching than
         simple substring matching, e.g., match one of the words
         \e{mail}, \e{letter} or \e{correspondence}, but none of the
         words \e{email}, \e{mailman}, \e{mailer}, \e{letterbox}, etc.
     \row \i Search and Replace
         \i A regexp can replace all occurrences of a substring with a
         different substring, e.g., replace all occurrences of \e{&}
         with \e{\&amp;} except where the \e{&} is already followed by
         an \e{amp;}.
    \row \i String Splitting
         \i A regexp can be used to identify where a string should be
         split apart, e.g. splitting tab-delimited strings.
    \endtable

    A brief introduction to regexps is presented, a description of
    Qt's regexp language, some examples, and the function
    documentation itself. QRegExp is modeled on Perl's regexp
    language. It fully supports Unicode. QRegExp can also be used in a
    simpler, \e{wildcard mode} that is similar to the functionality
    found in command shells. The syntax rules used by QRegExp can be
    changed with setPatternSyntax(). In particular, the pattern syntax
    can be set to QRegExp::FixedString, which means the pattern to be
    matched is interpreted as a plain string, i.e., special characters
    (e.g., backslash) are not escaped.

    A good text on regexps is \e {Mastering Regular Expressions}
    (Third Edition) by Jeffrey E. F.  Friedl, ISBN 0-596-52812-4.

    \tableofcontents

    \section1 Introduction

    Regexps are built up from expressions, quantifiers, and
    assertions. The simplest expression is a character, e.g. \bold{x}
    or \bold{5}. An expression can also be a set of characters
    enclosed in square brackets. \bold{[ABCD]} will match an \bold{A}
    or a \bold{B} or a \bold{C} or a \bold{D}. We can write this same
    expression as \bold{[A-D]}, and an experession to match any
    captital letter in the English alphabet is written as
    \bold{[A-Z]}.

    A quantifier specifies the number of occurrences of an expression
    that must be matched. \bold{x{1,1}} means match one and only one
    \bold{x}. \bold{x{1,5}} means match a sequence of \bold{x}
    characters that contains at least one \bold{x} but no more than
    five.

    Note that in general regexps cannot be used to check for balanced
    brackets or tags. For example, a regexp can be written to match an
    opening html \c{<b>} and its closing \c{</b>}, if the \c{<b>} tags
    are not nested, but if the \c{<b>} tags are nested, that same
    regexp will match an opening \c{<b>} tag with the wrong closing
    \c{</b>}.  For the fragment \c{<b>bold <b>bolder</b></b>}, the
    first \c{<b>} would be matched with the first \c{</b>}, which is
    not correct. However, it is possible to write a regexp that will
    match nested brackets or tags correctly, but only if the number of
    nesting levels is fixed and known. If the number of nesting levels
    is not fixed and known, it is impossible to write a regexp that
    will not fail.

    Suppose we want a regexp to match integers in the range 0 to 99.
    At least one digit is required, so we start with the expression
    \bold{[0-9]{1,1}}, which matches a single digit exactly once. This
    regexp matches integers in the range 0 to 9. To match integers up
    to 99, increase the maximum number of occurrences to 2, so the
    regexp becomes \bold{[0-9]{1,2}}. This regexp satisfies the
    original requirement to match integers from 0 to 99, but it will
    also match integers that occur in the middle of strings. If we
    want the matched integer to be the whole string, we must use the
    anchor assertions, \bold{^} (caret) and \bold{$} (dollar). When
    \bold{^} is the first character in a regexp, it means the regexp
    must match from the beginning of the string. When \bold{$} is the
    last character of the regexp, it means the regexp must match to
    the end of the string. The regexp becomes \bold{^[0-9]{1,2}$}.
    Note that assertions, e.g. \bold{^} and \bold{$}, do not match
    characters but locations in the string.

    If you have seen regexps described elsewhere, they may have looked
    different from the ones shown here. This is because some sets of
    characters and some quantifiers are so common that they have been
    given special symbols to represent them. \bold{[0-9]} can be
    replaced with the symbol \bold{\\d}. The quantifier to match
    exactly one occurrence, \bold{{1,1}}, can be replaced with the
    expression itself, i.e. \bold{x{1,1}} is the same as \bold{x}. So
    our 0 to 99 matcher could be written as \bold{^\\d{1,2}$}. It can
    also be written \bold{^\\d\\d{0,1}$}, i.e. \e{From the start of
    the string, match a digit, followed immediately by 0 or 1 digits}.
    In practice, it would be written as \bold{^\\d\\d?$}. The \bold{?}
    is shorthand for the quantifier \bold{{0,1}}, i.e. 0 or 1
    occurrences. \bold{?} makes an expression optional. The regexp
    \bold{^\\d\\d?$} means \e{From the beginning of the string, match
    one digit, followed immediately by 0 or 1 more digit, followed
    immediately by end of string}.

    To write a regexp that matches one of the words 'mail' \e or
    'letter' \e or 'correspondence' but does not match words that
    contain these words, e.g., 'email', 'mailman', 'mailer', and
    'letterbox', start with a regexp that matches 'mail'. Expressed
    fully, the regexp is \bold{m{1,1}a{1,1}i{1,1}l{1,1}}, but because
    a character expression is automatically quantified by
    \bold{{1,1}}, we can simplify the regexp to \bold{mail}, i.e., an
    'm' followed by an 'a' followed by an 'i' followed by an 'l'. Now
    we can use the vertical bar \bold{|}, which means \bold{or}, to
    include the other two words, so our regexp for matching any of the
    three words becomes \bold{mail|letter|correspondence}. Match
    'mail' \bold{or} 'letter' \bold{or} 'correspondence'. While this
    regexp will match one of the three words we want to match, it will
    also match words we don't want to match, e.g., 'email'.  To
    prevent the regexp from matching unwanted words, we must tell it
    to begin and end the match at word boundaries. First we enclose
    our regexp in parentheses, \bold{(mail|letter|correspondence)}.
    Parentheses group expressions together, and they identify a part
    of the regexp that we wish to \l{capturing text}{capture}.
    Enclosing the expression in parentheses allows us to use it as a
    component in more complex regexps. It also allows us to examine
    which of the three words was actually matched. To force the match
    to begin and end on word boundaries, we enclose the regexp in
    \bold{\\b} \e{word boundary} assertions:
    \bold{\\b(mail|letter|correspondence)\\b}.  Now the regexp means:
    \e{Match a word boundary, followed by the regexp in parentheses,
    followed by a word boundary}. The \bold{\\b} assertion matches a
    \e position in the regexp, not a \e character. A word boundary is
    any non-word character, e.g., a space, newline, or the beginning
    or ending of a string.

    If we want to replace ampersand characters with the HTML entity
    \bold{\&amp;}, the regexp to match is simply \bold{\&}. But this
    regexp will also match ampersands that have already been converted
    to HTML entities. We want to replace only ampersands that are not
    already followed by \bold{amp;}. For this, we need the negative
    lookahead assertion, \bold{(?!}__\bold{)}. The regexp can then be
    written as \bold{\&(?!amp;)}, i.e. \e{Match an ampersand that is}
    \bold{not} \e{followed by} \bold{amp;}.

    If we want to count all the occurrences of 'Eric' and 'Eirik' in a
    string, two valid solutions are \bold{\\b(Eric|Eirik)\\b} and
    \bold{\\bEi?ri[ck]\\b}. The word boundary assertion '\\b' is
    required to avoid matching words that contain either name,
    e.g. 'Ericsson'. Note that the second regexp matches more
    spellings than we want: 'Eric', 'Erik', 'Eiric' and 'Eirik'.

    Some of the examples discussed above are implemented in the
    \link #code-examples code examples \endlink section.

    \target characters-and-abbreviations-for-sets-of-characters
    \section1 Characters and Abbreviations for Sets of Characters

    \table
    \header \i Element \i Meaning
    \row \i \bold{c}
         \i A character represents itself unless it has a special
         regexp meaning. e.g. \bold{c} matches the character \e c.
    \row \i \bold{\\c}
         \i A character that follows a backslash matches the character
         itself, except as specified below. e.g., To match a literal
         caret at the beginning of a string, write \bold{\\^}.
    \row \i \bold{\\a}
         \i Matches the ASCII bell (BEL, 0x07).
    \row \i \bold{\\f}
         \i Matches the ASCII form feed (FF, 0x0C).
    \row \i \bold{\\n}
         \i Matches the ASCII line feed (LF, 0x0A, Unix newline).
    \row \i \bold{\\r}
         \i Matches the ASCII carriage return (CR, 0x0D).
    \row \i \bold{\\t}
         \i Matches the ASCII horizontal tab (HT, 0x09).
    \row \i \bold{\\v}
         \i Matches the ASCII vertical tab (VT, 0x0B).
    \row \i \bold{\\x\e{hhhh}}
         \i Matches the Unicode character corresponding to the
         hexadecimal number \e{hhhh} (between 0x0000 and 0xFFFF).
    \row \i \bold{\\0\e{ooo}} (i.e., \\zero \e{ooo})
         \i matches the ASCII/Latin1 character for the octal number
         \e{ooo} (between 0 and 0377).
    \row \i \bold{. (dot)}
         \i Matches any character (including newline).
    \row \i \bold{\\d}
         \i Matches a digit (QChar::isDigit()).
    \row \i \bold{\\D}
         \i Matches a non-digit.
    \row \i \bold{\\s}
         \i Matches a whitespace character (QChar::isSpace()).
    \row \i \bold{\\S}
         \i Matches a non-whitespace character.
    \row \i \bold{\\w}
         \i Matches a word character (QChar::isLetterOrNumber(), QChar::isMark(), or '_').
    \row \i \bold{\\W}
         \i Matches a non-word character.
    \row \i \bold{\\\e{n}}
         \i The \e{n}-th \l backreference, e.g. \\1, \\2, etc.
    \endtable

    \bold{Note:} The C++ compiler transforms backslashes in strings.
    To include a \bold{\\} in a regexp, enter it twice, i.e. \c{\\}.
    To match the backslash character itself, enter it four times, i.e.
    \c{\\\\}.

    \target sets-of-characters
    \section1 Sets of Characters

    Square brackets mean match any character contained in the square
    brackets. The character set abbreviations described above can
    appear in a character set in square brackets. Except for the
    character set abbreviations and the following two exceptions, 
    characters do not have special meanings in square brackets.

    \table
    \row \i \bold{^}

         \i The caret negates the character set if it occurs as the
         first character (i.e. immediately after the opening square
         bracket). \bold{[abc]} matches 'a' or 'b' or 'c', but
         \bold{[^abc]} matches anything \e but 'a' or 'b' or 'c'.

    \row \i \bold{-}

         \i The dash indicates a range of characters. \bold{[W-Z]}
         matches 'W' or 'X' or 'Y' or 'Z'.

    \endtable

    Using the predefined character set abbreviations is more portable
    than using character ranges across platforms and languages. For
    example, \bold{[0-9]} matches a digit in Western alphabets but
    \bold{\\d} matches a digit in \e any alphabet.

    Note: In other regexp documentation, sets of characters are often
    called "character classes".

    \target quantifiers
    \section1 Quantifiers

    By default, an expression is automatically quantified by
    \bold{{1,1}}, i.e. it should occur exactly once. In the following
    list, \bold{\e {E}} stands for expression. An expression is a
    character, or an abbreviation for a set of characters, or a set of
    characters in square brackets, or an expression in parentheses.

    \table
    \row \i \bold{\e {E}?}

         \i Matches zero or one occurrences of \e E. This quantifier
         means \e{The previous expression is optional}, because it
         will match whether or not the expression is found. \bold{\e
         {E}?} is the same as \bold{\e {E}{0,1}}. e.g., \bold{dents?}
         matches 'dent' or 'dents'.

    \row \i \bold{\e {E}+}

         \i Matches one or more occurrences of \e E. \bold{\e {E}+} is
         the same as \bold{\e {E}{1,}}. e.g., \bold{0+} matches '0',
         '00', '000', etc.

    \row \i \bold{\e {E}*}

         \i Matches zero or more occurrences of \e E. It is the same
         as \bold{\e {E}{0,}}. The \bold{*} quantifier is often used
         in error where \bold{+} should be used. For example, if
         \bold{\\s*$} is used in an expression to match strings that
         end in whitespace, it will match every string because
         \bold{\\s*$} means \e{Match zero or more whitespaces followed
         by end of string}. The correct regexp to match strings that
         have at least one trailing whitespace character is
         \bold{\\s+$}.

    \row \i \bold{\e {E}{n}}

         \i Matches exactly \e n occurrences of \e E. \bold{\e {E}{n}}
         is the same as repeating \e E \e n times. For example,
         \bold{x{5}} is the same as \bold{xxxxx}. It is also the same
         as \bold{\e {E}{n,n}}, e.g. \bold{x{5,5}}.

    \row \i \bold{\e {E}{n,}}
         \i Matches at least \e n occurrences of \e E.

    \row \i \bold{\e {E}{,m}}
         \i Matches at most \e m occurrences of \e E. \bold{\e {E}{,m}}
         is the same as \bold{\e {E}{0,m}}.

    \row \i \bold{\e {E}{n,m}}
         \i Matches at least \e n and at most \e m occurrences of \e E.
    \endtable

    To apply a quantifier to more than just the preceding character,
    use parentheses to group characters together in an expression. For
    example, \bold{tag+} matches a 't' followed by an 'a' followed by
    at least one 'g', whereas \bold{(tag)+} matches at least one
    occurrence of 'tag'.

    Note: Quantifiers are normally "greedy". They always match as much
    text as they can. For example, \bold{0+} matches the first zero it
    finds and all the consecutive zeros after the first zero. Applied
    to '20005', it matches'2\underline{000}5'. Quantifiers can be made
    non-greedy, see setMinimal().

    \target capturing parentheses
    \target backreferences
    \section1 Capturing Text

    Parentheses allow us to group elements together so that we can
    quantify and capture them. For example if we have the expression
    \bold{mail|letter|correspondence} that matches a string we know
    that \e one of the words matched but not which one. Using
    parentheses allows us to "capture" whatever is matched within
    their bounds, so if we used \bold{(mail|letter|correspondence)}
    and matched this regexp against the string "I sent you some email"
    we can use the cap() or capturedTexts() functions to extract the
    matched characters, in this case 'mail'.

    We can use captured text within the regexp itself. To refer to the
    captured text we use \e backreferences which are indexed from 1,
    the same as for cap(). For example we could search for duplicate
    words in a string using \bold{\\b(\\w+)\\W+\\1\\b} which means match a
    word boundary followed by one or more word characters followed by
    one or more non-word characters followed by the same text as the
    first parenthesized expression followed by a word boundary.

    If we want to use parentheses purely for grouping and not for
    capturing we can use the non-capturing syntax, e.g.
    \bold{(?:green|blue)}. Non-capturing parentheses begin '(?:' and
    end ')'. In this example we match either 'green' or 'blue' but we
    do not capture the match so we only know whether or not we matched
    but not which color we actually found. Using non-capturing
    parentheses is more efficient than using capturing parentheses
    since the regexp engine has to do less book-keeping.

    Both capturing and non-capturing parentheses may be nested.

    \target greedy quantifiers

    For historical reasons, quantifiers (e.g. \bold{*}) that apply to
    capturing parentheses are more "greedy" than other quantifiers.
    For example, \bold{a*(a)*} will match "aaa" with cap(1) == "aaa".
    This behavior is different from what other regexp engines do
    (notably, Perl). To obtain a more intuitive capturing behavior,
    specify QRegExp::RegExp2 to the QRegExp constructor or call
    setPatternSyntax(QRegExp::RegExp2).

    \target cap_in_a_loop

    When the number of matches cannot be determined in advance, a
    common idiom is to use cap() in a loop. For example:

    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 0

    \target assertions
    \section1 Assertions

    Assertions make some statement about the text at the point where
    they occur in the regexp but they do not match any characters. In
    the following list \bold{\e {E}} stands for any expression.

    \table
    \row \i \bold{^}
         \i The caret signifies the beginning of the string. If you
         wish to match a literal \c{^} you must escape it by
         writing \c{\\^}. For example, \bold{^#include} will only
         match strings which \e begin with the characters '#include'.
         (When the caret is the first character of a character set it
         has a special meaning, see \link #sets-of-characters Sets of
         Characters \endlink.)

    \row \i \bold{$}
         \i The dollar signifies the end of the string. For example
         \bold{\\d\\s*$} will match strings which end with a digit
         optionally followed by whitespace. If you wish to match a
         literal \c{$} you must escape it by writing
         \c{\\$}.

    \row \i \bold{\\b}
         \i A word boundary. For example the regexp
         \bold{\\bOK\\b} means match immediately after a word
         boundary (e.g. start of string or whitespace) the letter 'O'
         then the letter 'K' immediately before another word boundary
         (e.g. end of string or whitespace). But note that the
         assertion does not actually match any whitespace so if we
         write \bold{(\\bOK\\b)} and we have a match it will only
         contain 'OK' even if the string is "It's \underline{OK} now".

    \row \i \bold{\\B}
         \i A non-word boundary. This assertion is true wherever
         \bold{\\b} is false. For example if we searched for
         \bold{\\Bon\\B} in "Left on" the match would fail (space
         and end of string aren't non-word boundaries), but it would
         match in "t\underline{on}ne".

    \row \i \bold{(?=\e E)}
         \i Positive lookahead. This assertion is true if the
         expression matches at this point in the regexp. For example,
         \bold{const(?=\\s+char)} matches 'const' whenever it is
         followed by 'char', as in 'static \underline{const} char *'.
         (Compare with \bold{const\\s+char}, which matches 'static
         \underline{const char} *'.)

    \row \i \bold{(?!\e E)}
         \i Negative lookahead. This assertion is true if the
         expression does not match at this point in the regexp. For
         example, \bold{const(?!\\s+char)} matches 'const' \e except
         when it is followed by 'char'.
    \endtable

    \keyword QRegExp wildcard matching
    \section1 Wildcard Matching

    Most command shells such as \e bash or \e cmd.exe support "file
    globbing", the ability to identify a group of files by using
    wildcards. The setPatternSyntax() function is used to switch
    between regexp and wildcard mode. Wildcard matching is much
    simpler than full regexps and has only four features:

    \table
    \row \i \bold{c}
         \i Any character represents itself apart from those mentioned
         below. Thus \bold{c} matches the character \e c.
    \row \i \bold{?}
         \i Matches any single character. It is the same as
         \bold{.} in full regexps.
    \row \i \bold{*}
         \i Matches zero or more of any characters. It is the
         same as \bold{.*} in full regexps.
    \row \i \bold{[...]}
         \i Sets of characters can be represented in square brackets,
         similar to full regexps. Within the character class, like
         outside, backslash has no special meaning.
    \endtable

    For example if we are in wildcard mode and have strings which
    contain filenames we could identify HTML files with \bold{*.html}.
    This will match zero or more characters followed by a dot followed
    by 'h', 't', 'm' and 'l'.

    To test a string against a wildcard expression, use exactMatch().
    For example:

    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 1

    \target perl-users
    \section1 Notes for Perl Users

    Most of the character class abbreviations supported by Perl are
    supported by QRegExp, see \link
    #characters-and-abbreviations-for-sets-of-characters characters
    and abbreviations for sets of characters \endlink.

    In QRegExp, apart from within character classes, \c{^} always
    signifies the start of the string, so carets must always be
    escaped unless used for that purpose. In Perl the meaning of caret
    varies automagically depending on where it occurs so escaping it
    is rarely necessary. The same applies to \c{$} which in
    QRegExp always signifies the end of the string.

    QRegExp's quantifiers are the same as Perl's greedy quantifiers
    (but see the \l{greedy quantifiers}{note above}). Non-greedy
    matching cannot be applied to individual quantifiers, but can be
    applied to all the quantifiers in the pattern. For example, to
    match the Perl regexp \bold{ro+?m} requires:

    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 2

    The equivalent of Perl's \c{/i} option is
    setCaseSensitivity(Qt::CaseInsensitive).

    Perl's \c{/g} option can be emulated using a \l{#cap_in_a_loop}{loop}.

    In QRegExp \bold{.} matches any character, therefore all QRegExp
    regexps have the equivalent of Perl's \c{/s} option. QRegExp
    does not have an equivalent to Perl's \c{/m} option, but this
    can be emulated in various ways for example by splitting the input
    into lines or by looping with a regexp that searches for newlines.

    Because QRegExp is string oriented, there are no \\A, \\Z, or \\z
    assertions. The \\G assertion is not supported but can be emulated
    in a loop.

    Perl's $& is cap(0) or capturedTexts()[0]. There are no QRegExp
    equivalents for $`, $' or $+. Perl's capturing variables, $1, $2,
    ... correspond to cap(1) or capturedTexts()[1], cap(2) or
    capturedTexts()[2], etc.

    To substitute a pattern use QString::replace().

    Perl's extended \c{/x} syntax is not supported, nor are
    directives, e.g. (?i), or regexp comments, e.g. (?#comment). On
    the other hand, C++'s rules for literal strings can be used to
    achieve the same:

    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 3

    Both zero-width positive and zero-width negative lookahead
    assertions (?=pattern) and (?!pattern) are supported with the same
    syntax as Perl. Perl's lookbehind assertions, "independent"
    subexpressions and conditional expressions are not supported.

    Non-capturing parentheses are also supported, with the same
    (?:pattern) syntax.

    See QString::split() and QStringList::join() for equivalents
    to Perl's split and join functions.

    Note: because C++ transforms \\'s they must be written \e twice in
    code, e.g. \bold{\\b} must be written \bold{\\\\b}.

    \target code-examples
    \section1 Code Examples

    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 4

    The third string matches '\underline{6}'. This is a simple validation
    regexp for integers in the range 0 to 99.

    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 5

    The second string matches '\underline{This_is-OK}'. We've used the
    character set abbreviation '\\S' (non-whitespace) and the anchors
    to match strings which contain no whitespace.

    In the following example we match strings containing 'mail' or
    'letter' or 'correspondence' but only match whole words i.e. not
    'email'

    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 6

    The second string matches "Please write the \underline{letter}". The
    word 'letter' is also captured (because of the parentheses). We
    can see what text we've captured like this:

    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 7

    This will capture the text from the first set of capturing
    parentheses (counting capturing left parentheses from left to
    right). The parentheses are counted from 1 since cap(0) is the
    whole matched regexp (equivalent to '&' in most regexp engines).

    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 8

    Here we've passed the QRegExp to QString's replace() function to
    replace the matched text with new text.

    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 9

    We've used the indexIn() function to repeatedly match the regexp in
    the string. Note that instead of moving forward by one character
    at a time \c pos++ we could have written \c {pos +=
    rx.matchedLength()} to skip over the already matched string. The
    count will equal 3, matching 'One \underline{Eric} another
    \underline{Eirik}, and an Ericsson. How many Eiriks, \underline{Eric}?'; it
    doesn't match 'Ericsson' or 'Eiriks' because they are not bounded
    by non-word boundaries.

    One common use of regexps is to split lines of delimited data into
    their component fields.

    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 10

    In this example our input lines have the format company name, web
    address and country. Unfortunately the regexp is rather long and
    not very versatile -- the code will break if we add any more
    fields. A simpler and better solution is to look for the
    separator, '\\t' in this case, and take the surrounding text. The
    QString::split() function can take a separator string or regexp
    as an argument and split a string accordingly.

    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 11

    Here field[0] is the company, field[1] the web address and so on.

    To imitate the matching of a shell we can use wildcard mode.

    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 12

    Wildcard matching can be convenient because of its simplicity, but
    any wildcard regexp can be defined using full regexps, e.g.
    \bold{.*\.html$}. Notice that we can't match both \c .html and \c
    .htm files with a wildcard unless we use \bold{*.htm*} which will
    also match 'test.html.bak'. A full regexp gives us the precision
    we need, \bold{.*\\.html?$}.

    QRegExp can match case insensitively using setCaseSensitivity(),
    and can use non-greedy matching, see setMinimal(). By
    default QRegExp uses full regexps but this can be changed with
    setWildcard(). Searching can be forward with indexIn() or backward
    with lastIndexIn(). Captured text can be accessed using
    capturedTexts() which returns a string list of all captured
    strings, or using cap() which returns the captured string for the
    given index. The pos() function takes a match index and returns
    the position in the string where the match was made (or -1 if
    there was no match).

    \sa QString, QStringList, QRegExpValidator, QSortFilterProxyModel,
        {tools/regexp}{Regular Expression Example}
*/

#if defined(Q_OS_VXWORKS) && defined(EOS)
#  undef EOS
#endif

const int NumBadChars = 64;
#define BadChar(ch) ((ch).unicode() % NumBadChars)

const int NoOccurrence = INT_MAX;
const int EmptyCapture = INT_MAX;
const int InftyLen = INT_MAX;
const int InftyRep = 1025;
const int EOS = -1;

static bool isWord(QChar ch)
{
    return ch.isLetterOrNumber() || ch.isMark() || ch == QLatin1Char('_');
}

/*
  Merges two vectors of ints and puts the result into the first
  one.
*/
static void mergeInto(QVector<int> *a, const QVector<int> &b)
{
    int asize = a->size();
    int bsize = b.size();
    if (asize == 0) {
        *a = b;
#ifndef QT_NO_REGEXP_OPTIM
    } else if (bsize == 1 && a->at(asize - 1) < b.at(0)) {
        a->resize(asize + 1);
        (*a)[asize] = b.at(0);
#endif
    } else if (bsize >= 1) {
        int csize = asize + bsize;
        QVector<int> c(csize);
        int i = 0, j = 0, k = 0;
        while (i < asize) {
            if (j < bsize) {
                if (a->at(i) == b.at(j)) {
                    ++i;
                    --csize;
                } else if (a->at(i) < b.at(j)) {
                    c[k++] = a->at(i++);
                } else {
                    c[k++] = b.at(j++);
                }
            } else {
                memcpy(c.data() + k, a->constData() + i, (asize - i) * sizeof(int));
                break;
            }
        }
        c.resize(csize);
        if (j < bsize)
            memcpy(c.data() + k, b.constData() + j, (bsize - j) * sizeof(int));
        *a = c;
    }
}

#ifndef QT_NO_REGEXP_WILDCARD
/*
  Translates a wildcard pattern to an equivalent regular expression
  pattern (e.g., *.cpp to .*\.cpp).
*/
static QString wc2rx(const QString &wc_str)
{
    int wclen = wc_str.length();
    QString rx;
    int i = 0;
    const QChar *wc = wc_str.unicode();
    while (i < wclen) {
        QChar c = wc[i++];
        switch (c.unicode()) {
        case '*':
            rx += QLatin1String(".*");
            break;
        case '?':
            rx += QLatin1Char('.');
            break;
        case '$':
        case '(':
        case ')':
        case '+':
        case '.':
        case '\\':
        case '^':
        case '{':
        case '|':
        case '}':
            rx += QLatin1Char('\\');
            rx += c;
            break;
        case '[':
            rx += c;
            if (wc[i] == QLatin1Char('^'))
                rx += wc[i++];
            if (i < wclen) {
                if (rx[i] == QLatin1Char(']'))
                    rx += wc[i++];
                while (i < wclen && wc[i] != QLatin1Char(']')) {
                    if (wc[i] == QLatin1Char('\\'))
                        rx += QLatin1Char('\\');
                    rx += wc[i++];
                }
            }
            break;
        default:
            rx += c;
        }
    }
    return rx;
}
#endif

static int caretIndex(int offset, QRegExp::CaretMode caretMode)
{
    if (caretMode == QRegExp::CaretAtZero) {
        return 0;
    } else if (caretMode == QRegExp::CaretAtOffset) {
        return offset;
    } else { // QRegExp::CaretWontMatch
        return -1;
    }
}

/*
    The QRegExpEngineKey struct uniquely identifies an engine.
*/
struct QRegExpEngineKey
{
    QString pattern;
    QRegExp::PatternSyntax patternSyntax;
    Qt::CaseSensitivity cs;

    inline QRegExpEngineKey(const QString &pattern, QRegExp::PatternSyntax patternSyntax,
                            Qt::CaseSensitivity cs)
        : pattern(pattern), patternSyntax(patternSyntax), cs(cs) {}

    inline void clear() {
        pattern.clear();
        patternSyntax = QRegExp::RegExp;
        cs = Qt::CaseSensitive;
    }
};

bool operator==(const QRegExpEngineKey &key1, const QRegExpEngineKey &key2)
{
    return key1.pattern == key2.pattern && key1.patternSyntax == key2.patternSyntax
           && key1.cs == key2.cs;
}

class QRegExpEngine;

//Q_DECLARE_TYPEINFO(QVector<int>, Q_MOVABLE_TYPE);

/*
  This is the engine state during matching.
*/
struct QRegExpMatchState
{
    const QChar *in; // a pointer to the input string data
    int pos; // the current position in the string
    int caretPos;
    int len; // the length of the input string
    bool minimal; // minimal matching?
    int *bigArray; // big array holding the data for the next pointers
    int *inNextStack; // is state is nextStack?
    int *curStack; // stack of current states
    int *nextStack; // stack of next states
    int *curCapBegin; // start of current states' captures
    int *nextCapBegin; // start of next states' captures
    int *curCapEnd; // end of current states' captures
    int *nextCapEnd; // end of next states' captures
    int *tempCapBegin; // start of temporary captures
    int *tempCapEnd; // end of temporary captures
    int *capBegin; // start of captures for a next state
    int *capEnd; // end of captures for a next state
    int *slideTab; // bump-along slide table for bad-character heuristic
    int *captured; // what match() returned last
    int slideTabSize; // size of slide table
    int capturedSize;
#ifndef QT_NO_REGEXP_BACKREF
    QList<QVector<int> > sleeping; // list of back-reference sleepers
#endif
    int matchLen; // length of match
    int oneTestMatchedLen; // length of partial match

    const QRegExpEngine *eng;

    inline QRegExpMatchState() : bigArray(0), captured(0) {}
    inline ~QRegExpMatchState() { free(bigArray); }

    void drain() { free(bigArray); bigArray = 0; captured = 0; } // to save memory
    void prepareForMatch(QRegExpEngine *eng);
    void match(const QChar *str, int len, int pos, bool minimal,
        bool oneTest, int caretIndex);
    bool matchHere();
    bool testAnchor(int i, int a, const int *capBegin);
};

/*
  The struct QRegExpAutomatonState represents one state in a modified NFA. The
  input characters matched are stored in the state instead of on
  the transitions, something possible for an automaton
  constructed from a regular expression.
*/
struct QRegExpAutomatonState
{
#ifndef QT_NO_REGEXP_CAPTURE
    int atom; // which atom does this state belong to?
#endif
    int match; // what does it match? (see CharClassBit and BackRefBit)
    QVector<int> outs; // out-transitions
    QMap<int, int> reenter; // atoms reentered when transiting out
    QMap<int, int> anchors; // anchors met when transiting out

    inline QRegExpAutomatonState() { }
#ifndef QT_NO_REGEXP_CAPTURE
    inline QRegExpAutomatonState(int a, int m)
        : atom(a), match(m) { }
#else
    inline QRegExpAutomatonState(int m)
        : match(m) { }
#endif
};

Q_DECLARE_TYPEINFO(QRegExpAutomatonState, Q_MOVABLE_TYPE);

/*
  The struct QRegExpCharClassRange represents a range of characters (e.g.,
  [0-9] denotes range 48 to 57).
*/
struct QRegExpCharClassRange
{
    ushort from; // 48
    ushort len; // 10
};

Q_DECLARE_TYPEINFO(QRegExpCharClassRange, Q_PRIMITIVE_TYPE);

#ifndef QT_NO_REGEXP_CAPTURE
/*
  The struct QRegExpAtom represents one node in the hierarchy of regular
  expression atoms.
*/
struct QRegExpAtom
{
    enum { NoCapture = -1, OfficialCapture = -2, UnofficialCapture = -3 };

    int parent; // index of parent in array of atoms
    int capture; // index of capture, from 1 to ncap - 1
};

Q_DECLARE_TYPEINFO(QRegExpAtom, Q_PRIMITIVE_TYPE);
#endif

struct QRegExpLookahead;

#ifndef QT_NO_REGEXP_ANCHOR_ALT
/*
  The struct QRegExpAnchorAlternation represents a pair of anchors with
  OR semantics.
*/
struct QRegExpAnchorAlternation
{
    int a; // this anchor...
    int b; // ...or this one
};

Q_DECLARE_TYPEINFO(QRegExpAnchorAlternation, Q_PRIMITIVE_TYPE);
#endif

#ifndef QT_NO_REGEXP_CCLASS
/*
  The class QRegExpCharClass represents a set of characters, such as can
  be found in regular expressions (e.g., [a-z] denotes the set
  {a, b, ..., z}).
*/
class QRegExpCharClass
{
public:
    QRegExpCharClass();
    inline QRegExpCharClass(const QRegExpCharClass &cc) { operator=(cc); }

    QRegExpCharClass &operator=(const QRegExpCharClass &cc);

    void clear();
    bool negative() const { return n; }
    void setNegative(bool negative);
    void addCategories(int cats);
    void addRange(ushort from, ushort to);
    void addSingleton(ushort ch) { addRange(ch, ch); }

    bool in(QChar ch) const;
#ifndef QT_NO_REGEXP_OPTIM
    const QVector<int> &firstOccurrence() const { return occ1; }
#endif

#if defined(QT_DEBUG)
    void dump() const;
#endif

private:
    int c; // character classes
    QVector<QRegExpCharClassRange> r; // character ranges
    bool n; // negative?
#ifndef QT_NO_REGEXP_OPTIM
    QVector<int> occ1; // first-occurrence array
#endif
};
#else
struct QRegExpCharClass
{
    int dummy;

#ifndef QT_NO_REGEXP_OPTIM
    QRegExpCharClass() { occ1.fill(0, NumBadChars); }

    const QVector<int> &firstOccurrence() const { return occ1; }
    QVector<int> occ1;
#endif
};
#endif

Q_DECLARE_TYPEINFO(QRegExpCharClass, Q_MOVABLE_TYPE);

/*
  The QRegExpEngine class encapsulates a modified nondeterministic
  finite automaton (NFA).
*/
class QRegExpEngine
{
public:
    QRegExpEngine(Qt::CaseSensitivity cs, bool greedyQuantifiers)
        : cs(cs), greedyQuantifiers(greedyQuantifiers) { setup(); }

    QRegExpEngine(const QRegExpEngineKey &key);
    ~QRegExpEngine();

    bool isValid() const { return valid; }
    const QString &errorString() const { return yyError; }
    int numCaptures() const { return officialncap; }

    int createState(QChar ch);
    int createState(const QRegExpCharClass &cc);
#ifndef QT_NO_REGEXP_BACKREF
    int createState(int bref);
#endif

    void addCatTransitions(const QVector<int> &from, const QVector<int> &to);
#ifndef QT_NO_REGEXP_CAPTURE
    void addPlusTransitions(const QVector<int> &from, const QVector<int> &to, int atom);
#endif

#ifndef QT_NO_REGEXP_ANCHOR_ALT
    int anchorAlternation(int a, int b);
    int anchorConcatenation(int a, int b);
#else
    int anchorAlternation(int a, int b) { return a & b; }
    int anchorConcatenation(int a, int b) { return a | b; }
#endif
    void addAnchors(int from, int to, int a);

#ifndef QT_NO_REGEXP_OPTIM
    void heuristicallyChooseHeuristic();
#endif

#if defined(QT_DEBUG)
    void dump() const;
#endif

    QAtomicInt ref;

private:
    enum { CharClassBit = 0x10000, BackRefBit = 0x20000 };
    enum { InitialState = 0, FinalState = 1 };

    void setup();
    int setupState(int match);

    /*
      Let's hope that 13 lookaheads and 14 back-references are
      enough.
     */
    enum { MaxLookaheads = 13, MaxBackRefs = 14 };
    enum { Anchor_Dollar = 0x00000001, Anchor_Caret = 0x00000002, Anchor_Word = 0x00000004,
           Anchor_NonWord = 0x00000008, Anchor_FirstLookahead = 0x00000010,
           Anchor_BackRef1Empty = Anchor_FirstLookahead << MaxLookaheads,
           Anchor_BackRef0Empty = Anchor_BackRef1Empty >> 1,
           Anchor_Alternation = unsigned(Anchor_BackRef1Empty) << MaxBackRefs,

           Anchor_LookaheadMask = (Anchor_FirstLookahead - 1) ^
                   ((Anchor_FirstLookahead << MaxLookaheads) - 1) };
#ifndef QT_NO_REGEXP_CAPTURE
    int startAtom(bool officialCapture);
    void finishAtom(int atom, bool needCapture);
#endif

#ifndef QT_NO_REGEXP_LOOKAHEAD
    int addLookahead(QRegExpEngine *eng, bool negative);
#endif

#ifndef QT_NO_REGEXP_OPTIM
    bool goodStringMatch(QRegExpMatchState &matchState) const;
    bool badCharMatch(QRegExpMatchState &matchState) const;
#else
    bool bruteMatch(QRegExpMatchState &matchState) const;
#endif

    QVector<QRegExpAutomatonState> s; // array of states
#ifndef QT_NO_REGEXP_CAPTURE
    QVector<QRegExpAtom> f; // atom hierarchy
    int nf; // number of atoms
    int cf; // current atom
    QVector<int> captureForOfficialCapture;
#endif
    int officialncap; // number of captures, seen from the outside
    int ncap; // number of captures, seen from the inside
#ifndef QT_NO_REGEXP_CCLASS
    QVector<QRegExpCharClass> cl; // array of character classes
#endif
#ifndef QT_NO_REGEXP_LOOKAHEAD
    QVector<QRegExpLookahead *> ahead; // array of lookaheads
#endif
#ifndef QT_NO_REGEXP_ANCHOR_ALT
    QVector<QRegExpAnchorAlternation> aa; // array of (a, b) pairs of anchors
#endif
#ifndef QT_NO_REGEXP_OPTIM
    bool caretAnchored; // does the regexp start with ^?
    bool trivial; // is the good-string all that needs to match?
#endif
    bool valid; // is the regular expression valid?
    Qt::CaseSensitivity cs; // case sensitive?
    bool greedyQuantifiers; // RegExp2?
    bool xmlSchemaExtensions;
#ifndef QT_NO_REGEXP_BACKREF
    int nbrefs; // number of back-references
#endif

#ifndef QT_NO_REGEXP_OPTIM
    bool useGoodStringHeuristic; // use goodStringMatch? otherwise badCharMatch

    int goodEarlyStart; // the index where goodStr can first occur in a match
    int goodLateStart; // the index where goodStr can last occur in a match
    QString goodStr; // the string that any match has to contain

    int minl; // the minimum length of a match
    QVector<int> occ1; // first-occurrence array
#endif

    /*
      The class Box is an abstraction for a regular expression
      fragment. It can also be seen as one node in the syntax tree of
      a regular expression with synthetized attributes.

      Its interface is ugly for performance reasons.
    */
    class Box
    {
    public:
        Box(QRegExpEngine *engine);
        Box(const Box &b) { operator=(b); }

        Box &operator=(const Box &b);

        void clear() { operator=(Box(eng)); }
        void set(QChar ch);
        void set(const QRegExpCharClass &cc);
#ifndef QT_NO_REGEXP_BACKREF
        void set(int bref);
#endif

        void cat(const Box &b);
        void orx(const Box &b);
        void plus(int atom);
        void opt();
        void catAnchor(int a);
#ifndef QT_NO_REGEXP_OPTIM
        void setupHeuristics();
#endif

#if defined(QT_DEBUG)
        void dump() const;
#endif

    private:
        void addAnchorsToEngine(const Box &to) const;

        QRegExpEngine *eng; // the automaton under construction
        QVector<int> ls; // the left states (firstpos)
        QVector<int> rs; // the right states (lastpos)
        QMap<int, int> lanchors; // the left anchors
        QMap<int, int> ranchors; // the right anchors
        int skipanchors; // the anchors to match if the box is skipped

#ifndef QT_NO_REGEXP_OPTIM
        int earlyStart; // the index where str can first occur
        int lateStart; // the index where str can last occur
        QString str; // a string that has to occur in any match
        QString leftStr; // a string occurring at the left of this box
        QString rightStr; // a string occurring at the right of this box
        int maxl; // the maximum length of this box (possibly InftyLen)
#endif

        int minl; // the minimum length of this box
#ifndef QT_NO_REGEXP_OPTIM
        QVector<int> occ1; // first-occurrence array
#endif
    };

    friend class Box;

    void setupCategoriesRangeMap();

    /*
      This is the lexical analyzer for regular expressions.
    */
    enum { Tok_Eos, Tok_Dollar, Tok_LeftParen, Tok_MagicLeftParen, Tok_PosLookahead,
           Tok_NegLookahead, Tok_RightParen, Tok_CharClass, Tok_Caret, Tok_Quantifier, Tok_Bar,
           Tok_Word, Tok_NonWord, Tok_Char = 0x10000, Tok_BackRef = 0x20000 };
    int getChar();
    int getEscape();
#ifndef QT_NO_REGEXP_INTERVAL
    int getRep(int def);
#endif
#ifndef QT_NO_REGEXP_LOOKAHEAD
    void skipChars(int n);
#endif
    void error(const char *msg);
    void startTokenizer(const QChar *rx, int len);
    int getToken();

    const QChar *yyIn; // a pointer to the input regular expression pattern
    int yyPos0; // the position of yyTok in the input pattern
    int yyPos; // the position of the next character to read
    int yyLen; // the length of yyIn
    int yyCh; // the last character read
    QRegExpCharClass *yyCharClass; // attribute for Tok_CharClass tokens
    int yyMinRep; // attribute for Tok_Quantifier
    int yyMaxRep; // ditto
    QString yyError; // syntax error or overflow during parsing?

    /*
      This is the syntactic analyzer for regular expressions.
    */
    int parse(const QChar *rx, int len);
    void parseAtom(Box *box);
    void parseFactor(Box *box);
    void parseTerm(Box *box);
    void parseExpression(Box *box);

    int yyTok; // the last token read
    bool yyMayCapture; // set this to false to disable capturing
    QHash<QByteArray, QPair<int, int> > categoriesRangeMap; // fast lookup hash for xml schema extensions

    friend struct QRegExpMatchState;
};

#ifndef QT_NO_REGEXP_LOOKAHEAD
/*
  The struct QRegExpLookahead represents a lookahead a la Perl (e.g.,
  (?=foo) and (?!bar)).
*/
struct QRegExpLookahead
{
    QRegExpEngine *eng; // NFA representing the embedded regular expression
    bool neg; // negative lookahead?

    inline QRegExpLookahead(QRegExpEngine *eng0, bool neg0)
        : eng(eng0), neg(neg0) { }
    inline ~QRegExpLookahead() { delete eng; }
};
#endif

QRegExpEngine::QRegExpEngine(const QRegExpEngineKey &key)
    : cs(key.cs), greedyQuantifiers(key.patternSyntax == QRegExp::RegExp2),
      xmlSchemaExtensions(false)
{
    setup();

    QString rx;

    switch (key.patternSyntax) {
    case QRegExp::Wildcard:
#ifndef QT_NO_REGEXP_WILDCARD
        rx = wc2rx(key.pattern);
#endif
        break;
    case QRegExp::FixedString:
        rx = QRegExp::escape(key.pattern);
        break;
    case QRegExp::W3CXmlSchema11:
        xmlSchemaExtensions = true;
    default:
        rx = key.pattern;
    }

    valid = (parse(rx.unicode(), rx.length()) == rx.length());
    if (!valid) {
#ifndef QT_NO_REGEXP_OPTIM
        trivial = false;
#endif
        error(RXERR_LEFTDELIM);
    }
}

QRegExpEngine::~QRegExpEngine()
{
#ifndef QT_NO_REGEXP_LOOKAHEAD
    qDeleteAll(ahead);
#endif
}

void QRegExpMatchState::prepareForMatch(QRegExpEngine *eng)
{
    /*
      We use one QVector<int> for all the big data used a lot in
      matchHere() and friends.
    */
    int ns = eng->s.size(); // number of states
    int ncap = eng->ncap;
#ifndef QT_NO_REGEXP_OPTIM
    slideTabSize = qMax(eng->minl + 1, 16);
#else
    slideTabSize = 0;
#endif
    int numCaptures = eng->numCaptures();
    capturedSize = 2 + 2 * numCaptures;
    bigArray = (int *)realloc(bigArray, ((3 + 4 * ncap) * ns + 4 * ncap + slideTabSize + capturedSize)*sizeof(int));

    inNextStack = bigArray;
    memset(inNextStack, -1, ns * sizeof(int));
    curStack = inNextStack + ns;
    nextStack = inNextStack + 2 * ns;

    curCapBegin = inNextStack + 3 * ns;
    nextCapBegin = curCapBegin + ncap * ns;
    curCapEnd = curCapBegin + 2 * ncap * ns;
    nextCapEnd = curCapBegin + 3 * ncap * ns;

    tempCapBegin = curCapBegin + 4 * ncap * ns;
    tempCapEnd = tempCapBegin + ncap;
    capBegin = tempCapBegin + 2 * ncap;
    capEnd = tempCapBegin + 3 * ncap;

    slideTab = tempCapBegin + 4 * ncap;
    captured = slideTab + slideTabSize;
    memset(captured, -1, capturedSize*sizeof(int));
    this->eng = eng;
}

/*
  Tries to match in str and returns an array of (begin, length) pairs
  for captured text. If there is no match, all pairs are (-1, -1).
*/
void QRegExpMatchState::match(const QChar *str0, int len0, int pos0,
    bool minimal0, bool oneTest, int caretIndex)
{
    bool matched = false;
    QChar char_null;

#ifndef QT_NO_REGEXP_OPTIM
    if (eng->trivial && !oneTest) {
        pos = qFindString(str0, len0, pos0, eng->goodStr.unicode(), eng->goodStr.length(), eng->cs);
        matchLen = eng->goodStr.length();
        matched = (pos != -1);
    } else
#endif
    {
        in = str0;
        if (in == 0)
            in = &char_null;
        pos = pos0;
        caretPos = caretIndex;
        len = len0;
        minimal = minimal0;
        matchLen = 0;
        oneTestMatchedLen = 0;

        if (eng->valid && pos >= 0 && pos <= len) {
#ifndef QT_NO_REGEXP_OPTIM
            if (oneTest) {
                matched = matchHere();
            } else {
                if (pos <= len - eng->minl) {
                    if (eng->caretAnchored) {
                        matched = matchHere();
                    } else if (eng->useGoodStringHeuristic) {
                        matched = eng->goodStringMatch(*this);
                    } else {
                        matched = eng->badCharMatch(*this);
                    }
                }
            }
#else
            matched = oneTest ? matchHere() : eng->bruteMatch(*this);
#endif
        }
    }

    if (matched) {
        int *c = captured;
        *c++ = pos;
        *c++ = matchLen;

        int numCaptures = (capturedSize - 2) >> 1;
#ifndef QT_NO_REGEXP_CAPTURE
        for (int i = 0; i < numCaptures; ++i) {
            int j = eng->captureForOfficialCapture.at(i);
            int len = capEnd[j] - capBegin[j];
            *c++ = (len > 0) ? pos + capBegin[j] : 0;
            *c++ = len;
        }
#endif
    } else {
        // we rely on 2's complement here
        memset(captured, -1, capturedSize * sizeof(int));
    }
}

/*
  The three following functions add one state to the automaton and
  return the number of the state.
*/

int QRegExpEngine::createState(QChar ch)
{
    return setupState(ch.unicode());
}

int QRegExpEngine::createState(const QRegExpCharClass &cc)
{
#ifndef QT_NO_REGEXP_CCLASS
    int n = cl.size();
    cl += QRegExpCharClass(cc);
    return setupState(CharClassBit | n);
#else
    Q_UNUSED(cc);
    return setupState(CharClassBit);
#endif
}

#ifndef QT_NO_REGEXP_BACKREF
int QRegExpEngine::createState(int bref)
{
    if (bref > nbrefs) {
        nbrefs = bref;
        if (nbrefs > MaxBackRefs) {
            error(RXERR_LIMIT);
            return 0;
        }
    }
    return setupState(BackRefBit | bref);
}
#endif

/*
  The two following functions add a transition between all pairs of
  states (i, j) where i is found in from, and j is found in to.

  Cat-transitions are distinguished from plus-transitions for
  capturing.
*/

void QRegExpEngine::addCatTransitions(const QVector<int> &from, const QVector<int> &to)
{
    for (int i = 0; i < from.size(); i++)
        mergeInto(&s[from.at(i)].outs, to);
}

#ifndef QT_NO_REGEXP_CAPTURE
void QRegExpEngine::addPlusTransitions(const QVector<int> &from, const QVector<int> &to, int atom)
{
    for (int i = 0; i < from.size(); i++) {
        QRegExpAutomatonState &st = s[from.at(i)];
        const QVector<int> oldOuts = st.outs;
        mergeInto(&st.outs, to);
        if (f.at(atom).capture != QRegExpAtom::NoCapture) {
            for (int j = 0; j < to.size(); j++) {
                // ### st.reenter.contains(to.at(j)) check looks suspicious
                if (!st.reenter.contains(to.at(j)) &&
                     qBinaryFind(oldOuts.constBegin(), oldOuts.constEnd(), to.at(j)) == oldOuts.end())
                    st.reenter.insert(to.at(j), atom);
            }
        }
    }
}
#endif

#ifndef QT_NO_REGEXP_ANCHOR_ALT
/*
  Returns an anchor that means a OR b.
*/
int QRegExpEngine::anchorAlternation(int a, int b)
{
    if (((a & b) == a || (a & b) == b) && ((a | b) & Anchor_Alternation) == 0)
        return a & b;

    int n = aa.size();
#ifndef QT_NO_REGEXP_OPTIM
    if (n > 0 && aa.at(n - 1).a == a && aa.at(n - 1).b == b)
        return Anchor_Alternation | (n - 1);
#endif

    QRegExpAnchorAlternation element = {a, b};
    aa.append(element);
    return Anchor_Alternation | n;
}

/*
  Returns an anchor that means a AND b.
*/
int QRegExpEngine::anchorConcatenation(int a, int b)
{
    if (((a | b) & Anchor_Alternation) == 0)
        return a | b;
    if ((b & Anchor_Alternation) != 0)
        qSwap(a, b);

    int aprime = anchorConcatenation(aa.at(a ^ Anchor_Alternation).a, b);
    int bprime = anchorConcatenation(aa.at(a ^ Anchor_Alternation).b, b);
    return anchorAlternation(aprime, bprime);
}
#endif

/*
  Adds anchor a on a transition caracterised by its from state and
  its to state.
*/
void QRegExpEngine::addAnchors(int from, int to, int a)
{
    QRegExpAutomatonState &st = s[from];
    if (st.anchors.contains(to))
        a = anchorAlternation(st.anchors.value(to), a);
    st.anchors.insert(to, a);
}

#ifndef QT_NO_REGEXP_OPTIM
/*
  This function chooses between the good-string and the bad-character
  heuristics. It computes two scores and chooses the heuristic with
  the highest score.

  Here are some common-sense constraints on the scores that should be
  respected if the formulas are ever modified: (1) If goodStr is
  empty, the good-string heuristic scores 0. (2) If the regular
  expression is trivial, the good-string heuristic should be used.
  (3) If the search is case insensitive, the good-string heuristic
  should be used, unless it scores 0. (Case insensitivity turns all
  entries of occ1 to 0.) (4) If (goodLateStart - goodEarlyStart) is
  big, the good-string heuristic should score less.
*/
void QRegExpEngine::heuristicallyChooseHeuristic()
{
    if (minl == 0) {
        useGoodStringHeuristic = false;
    } else if (trivial) {
        useGoodStringHeuristic = true;
    } else {
        /*
          Magic formula: The good string has to constitute a good
          proportion of the minimum-length string, and appear at a
          more-or-less known index.
        */
        int goodStringScore = (64 * goodStr.length() / minl) -
                              (goodLateStart - goodEarlyStart);
        /*
          Less magic formula: We pick some characters at random, and
          check whether they are good or bad.
        */
        int badCharScore = 0;
        int step = qMax(1, NumBadChars / 32);
        for (int i = 1; i < NumBadChars; i += step) {
            if (occ1.at(i) == NoOccurrence)
                badCharScore += minl;
            else
                badCharScore += occ1.at(i);
        }
        badCharScore /= minl;
        useGoodStringHeuristic = (goodStringScore > badCharScore);
    }
}
#endif

#if defined(QT_DEBUG)
void QRegExpEngine::dump() const
{
    int i, j;
    qDebug("Case %ssensitive engine", cs ? "" : "in");
    qDebug("  States");
    for (i = 0; i < s.size(); i++) {
        qDebug("  %d%s", i, i == InitialState ? " (initial)" : i == FinalState ? " (final)" : "");
#ifndef QT_NO_REGEXP_CAPTURE
        if (nf > 0)
            qDebug("    in atom %d", s[i].atom);
#endif
        int m = s[i].match;
        if ((m & CharClassBit) != 0) {
            qDebug("    match character class %d", m ^ CharClassBit);
#ifndef QT_NO_REGEXP_CCLASS
            cl[m ^ CharClassBit].dump();
#else
            qDebug("    negative character class");
#endif
        } else if ((m & BackRefBit) != 0) {
            qDebug("    match back-reference %d", m ^ BackRefBit);
        } else if (m >= 0x20 && m <= 0x7e) {
            qDebug("    match 0x%.4x (%c)", m, m);
        } else {
            qDebug("    match 0x%.4x", m);
        }
        for (j = 0; j < s[i].outs.size(); j++) {
            int next = s[i].outs[j];
            qDebug("    -> %d", next);
            if (s[i].reenter.contains(next))
                qDebug("       [reenter %d]", s[i].reenter[next]);
            if (s[i].anchors.value(next) != 0)
                qDebug("       [anchors 0x%.8x]", s[i].anchors[next]);
        }
    }
#ifndef QT_NO_REGEXP_CAPTURE
    if (nf > 0) {
        qDebug("  Atom    Parent  Capture");
        for (i = 0; i < nf; i++) {
            if (f[i].capture == QRegExpAtom::NoCapture) {
                qDebug("  %6d  %6d     nil", i, f[i].parent);
            } else {
                int cap = f[i].capture;
                bool official = captureForOfficialCapture.contains(cap);
                qDebug("  %6d  %6d  %6d  %s", i, f[i].parent, f[i].capture,
                       official ? "official" : "");
            }
        }
    }
#endif
#ifndef QT_NO_REGEXP_ANCHOR_ALT
    for (i = 0; i < aa.size(); i++)
        qDebug("  Anchor alternation 0x%.8x: 0x%.8x 0x%.9x", i, aa[i].a, aa[i].b);
#endif
}
#endif

void QRegExpEngine::setup()
{
    ref = 1;
#ifndef QT_NO_REGEXP_CAPTURE
    f.resize(32);
    nf = 0;
    cf = -1;
#endif
    officialncap = 0;
    ncap = 0;
#ifndef QT_NO_REGEXP_OPTIM
    caretAnchored = true;
    trivial = true;
#endif
    valid = false;
#ifndef QT_NO_REGEXP_BACKREF
    nbrefs = 0;
#endif
#ifndef QT_NO_REGEXP_OPTIM
    useGoodStringHeuristic = true;
    minl = 0;
    occ1.fill(0, NumBadChars);
#endif
}

int QRegExpEngine::setupState(int match)
{
#ifndef QT_NO_REGEXP_CAPTURE
    s += QRegExpAutomatonState(cf, match);
#else
    s += QRegExpAutomatonState(match);
#endif
    return s.size() - 1;
}

#ifndef QT_NO_REGEXP_CAPTURE
/*
  Functions startAtom() and finishAtom() should be called to delimit
  atoms. When a state is created, it is assigned to the current atom.
  The information is later used for capturing.
*/
int QRegExpEngine::startAtom(bool officialCapture)
{
    if ((nf & (nf + 1)) == 0 && nf + 1 >= f.size())
        f.resize((nf + 1) << 1);
    f[nf].parent = cf;
    cf = nf++;
    f[cf].capture = officialCapture ? QRegExpAtom::OfficialCapture : QRegExpAtom::NoCapture;
    return cf;
}

void QRegExpEngine::finishAtom(int atom, bool needCapture)
{
    if (greedyQuantifiers && needCapture && f[atom].capture == QRegExpAtom::NoCapture)
        f[atom].capture = QRegExpAtom::UnofficialCapture;
    cf = f.at(atom).parent;
}
#endif

#ifndef QT_NO_REGEXP_LOOKAHEAD
/*
  Creates a lookahead anchor.
*/
int QRegExpEngine::addLookahead(QRegExpEngine *eng, bool negative)
{
    int n = ahead.size();
    if (n == MaxLookaheads) {
        error(RXERR_LIMIT);
        return 0;
    }
    ahead += new QRegExpLookahead(eng, negative);
    return Anchor_FirstLookahead << n;
}
#endif

#ifndef QT_NO_REGEXP_CAPTURE
/*
  We want the longest leftmost captures.
*/
static bool isBetterCapture(int ncap, const int *begin1, const int *end1, const int *begin2,
                            const int *end2)
{
    for (int i = 0; i < ncap; i++) {
        int delta = begin2[i] - begin1[i]; // it has to start early...
        if (delta == 0)
            delta = end1[i] - end2[i]; // ...and end late

        if (delta != 0)
            return delta > 0;
    }
    return false;
}
#endif

/*
  Returns true if anchor a matches at position pos + i in the input
  string, otherwise false.
*/
bool QRegExpMatchState::testAnchor(int i, int a, const int *capBegin)
{
    int j;

#ifndef QT_NO_REGEXP_ANCHOR_ALT
    if ((a & QRegExpEngine::Anchor_Alternation) != 0)
        return testAnchor(i, eng->aa.at(a ^ QRegExpEngine::Anchor_Alternation).a, capBegin)
               || testAnchor(i, eng->aa.at(a ^ QRegExpEngine::Anchor_Alternation).b, capBegin);
#endif

    if ((a & QRegExpEngine::Anchor_Caret) != 0) {
        if (pos + i != caretPos)
            return false;
    }
    if ((a & QRegExpEngine::Anchor_Dollar) != 0) {
        if (pos + i != len)
            return false;
    }
#ifndef QT_NO_REGEXP_ESCAPE
    if ((a & (QRegExpEngine::Anchor_Word | QRegExpEngine::Anchor_NonWord)) != 0) {
        bool before = false;
        bool after = false;
        if (pos + i != 0)
            before = isWord(in[pos + i - 1]);
        if (pos + i != len)
            after = isWord(in[pos + i]);
        if ((a & QRegExpEngine::Anchor_Word) != 0 && (before == after))
            return false;
        if ((a & QRegExpEngine::Anchor_NonWord) != 0 && (before != after))
            return false;
    }
#endif
#ifndef QT_NO_REGEXP_LOOKAHEAD
    if ((a & QRegExpEngine::Anchor_LookaheadMask) != 0) {
        const QVector<QRegExpLookahead *> &ahead = eng->ahead;
        for (j = 0; j < ahead.size(); j++) {
            if ((a & (QRegExpEngine::Anchor_FirstLookahead << j)) != 0) {
                QRegExpMatchState matchState;
                matchState.prepareForMatch(ahead[j]->eng);
                matchState.match(in + pos + i, len - pos - i, 0,
                    true, true, matchState.caretPos - matchState.pos - i);
                if ((matchState.captured[0] == 0) == ahead[j]->neg)
                    return false;
            }
        }
    }
#endif
#ifndef QT_NO_REGEXP_CAPTURE
#ifndef QT_NO_REGEXP_BACKREF
    for (j = 0; j < eng->nbrefs; j++) {
        if ((a & (QRegExpEngine::Anchor_BackRef1Empty << j)) != 0) {
            int i = eng->captureForOfficialCapture.at(j);
            if (capBegin[i] != EmptyCapture)
                return false;
        }
    }
#endif
#endif
    return true;
}

#ifndef QT_NO_REGEXP_OPTIM
/*
  The three following functions are what Jeffrey Friedl would call
  transmissions (or bump-alongs). Using one or the other should make
  no difference except in performance.
*/

bool QRegExpEngine::goodStringMatch(QRegExpMatchState &matchState) const
{
    int k = matchState.pos + goodEarlyStart;
    QStringMatcher matcher(goodStr.unicode(), goodStr.length(), cs);
    while ((k = matcher.indexIn(matchState.in, matchState.len, k)) != -1) {
        int from = k - goodLateStart;
        int to = k - goodEarlyStart;
        if (from > matchState.pos)
            matchState.pos = from;

        while (matchState.pos <= to) {
            if (matchState.matchHere())
                return true;
            ++matchState.pos;
        }
        ++k;
    }
    return false;
}

bool QRegExpEngine::badCharMatch(QRegExpMatchState &matchState) const
{
    int slideHead = 0;
    int slideNext = 0;
    int i;
    int lastPos = matchState.len - minl;
    memset(matchState.slideTab, 0, matchState.slideTabSize * sizeof(int));

    /*
      Set up the slide table, used for the bad-character heuristic,
      using the table of first occurrence of each character.
    */
    for (i = 0; i < minl; i++) {
        int sk = occ1[BadChar(matchState.in[matchState.pos + i])];
        if (sk == NoOccurrence)
            sk = i + 1;
        if (sk > 0) {
            int k = i + 1 - sk;
            if (k < 0) {
                sk = i + 1;
                k = 0;
            }
            if (sk > matchState.slideTab[k])
                matchState.slideTab[k] = sk;
        }
    }

    if (matchState.pos > lastPos)
        return false;

    for (;;) {
        if (++slideNext >= matchState.slideTabSize)
            slideNext = 0;
        if (matchState.slideTab[slideHead] > 0) {
            if (matchState.slideTab[slideHead] - 1 > matchState.slideTab[slideNext])
                matchState.slideTab[slideNext] = matchState.slideTab[slideHead] - 1;
            matchState.slideTab[slideHead] = 0;
        } else {
            if (matchState.matchHere())
                return true;
        }

        if (matchState.pos == lastPos)
            break;

        /*
          Update the slide table. This code has much in common with
          the initialization code.
        */
        int sk = occ1[BadChar(matchState.in[matchState.pos + minl])];
        if (sk == NoOccurrence) {
            matchState.slideTab[slideNext] = minl;
        } else if (sk > 0) {
            int k = slideNext + minl - sk;
            if (k >= matchState.slideTabSize)
                k -= matchState.slideTabSize;
            if (sk > matchState.slideTab[k])
                matchState.slideTab[k] = sk;
        }
        slideHead = slideNext;
        ++matchState.pos;
    }
    return false;
}
#else
bool QRegExpEngine::bruteMatch(QRegExpMatchState &matchState) const
{
    while (matchState.pos <= matchState.len) {
        if (matchState.matchHere())
            return true;
        ++matchState.pos;
    }
    return false;
}
#endif

/*
  Here's the core of the engine. It tries to do a match here and now.
*/
bool QRegExpMatchState::matchHere()
{
    int ncur = 1, nnext = 0;
    int i = 0, j, k, m;
    bool stop = false;

    matchLen = -1;
    oneTestMatchedLen = -1;
    curStack[0] = QRegExpEngine::InitialState;

    int ncap = eng->ncap;
#ifndef QT_NO_REGEXP_CAPTURE
    if (ncap > 0) {
        for (j = 0; j < ncap; j++) {
            curCapBegin[j] = EmptyCapture;
            curCapEnd[j] = EmptyCapture;
        }
    }
#endif

#ifndef QT_NO_REGEXP_BACKREF
    while ((ncur > 0 || !sleeping.isEmpty()) && i <= len - pos && !stop)
#else
    while (ncur > 0 && i <= len - pos && !stop)
#endif
    {
        int ch = (i < len - pos) ? in[pos + i].unicode() : 0;
        for (j = 0; j < ncur; j++) {
            int cur = curStack[j];
            const QRegExpAutomatonState &scur = eng->s.at(cur);
            const QVector<int> &outs = scur.outs;
            for (k = 0; k < outs.size(); k++) {
                int next = outs.at(k);
                const QRegExpAutomatonState &snext = eng->s.at(next);
                bool inside = true;
#if !defined(QT_NO_REGEXP_BACKREF) && !defined(QT_NO_REGEXP_CAPTURE)
                int needSomeSleep = 0;
#endif

                /*
                  First, check if the anchors are anchored properly.
                */
                int a = scur.anchors.value(next);
                if (a != 0 && !testAnchor(i, a, curCapBegin + j * ncap))
                    inside = false;

                /*
                  If indeed they are, check if the input character is
                  correct for this transition.
                */
                if (inside) {
                    m = snext.match;
                    if ((m & (QRegExpEngine::CharClassBit | QRegExpEngine::BackRefBit)) == 0) {
                        if (eng->cs)
                            inside = (m == ch);
                        else
                            inside = (QChar(m).toLower() == QChar(ch).toLower());
                    } else if (next == QRegExpEngine::FinalState) {
                        matchLen = i;
                        stop = minimal;
                        inside = true;
                    } else if ((m & QRegExpEngine::CharClassBit) != 0) {
#ifndef QT_NO_REGEXP_CCLASS
                        const QRegExpCharClass &cc = eng->cl.at(m ^ QRegExpEngine::CharClassBit);
                        if (eng->cs)
                            inside = cc.in(ch);
                        else if (cc.negative())
                            inside = cc.in(QChar(ch).toLower()) &&
                                     cc.in(QChar(ch).toUpper());
                        else
                            inside = cc.in(QChar(ch).toLower()) ||
                                     cc.in(QChar(ch).toUpper());
#endif
#if !defined(QT_NO_REGEXP_BACKREF) && !defined(QT_NO_REGEXP_CAPTURE)
                    } else { /* ((m & QRegExpEngine::BackRefBit) != 0) */
                        int bref = m ^ QRegExpEngine::BackRefBit;
                        int ell = j * ncap + eng->captureForOfficialCapture.at(bref - 1);

                        inside = bref <= ncap && curCapBegin[ell] != EmptyCapture;
                        if (inside) {
                            if (eng->cs)
                                inside = (in[pos + curCapBegin[ell]] == QChar(ch));
                            else
                                inside = (in[pos + curCapBegin[ell]].toLower()
                                       == QChar(ch).toLower());
                        }

                        if (inside) {
                            int delta;
                            if (curCapEnd[ell] == EmptyCapture)
                                delta = i - curCapBegin[ell];
                            else
                                delta = curCapEnd[ell] - curCapBegin[ell];

                            inside = (delta <= len - (pos + i));
                            if (inside && delta > 1) {
                                int n = 1;
                                if (eng->cs) {
                                    while (n < delta) {
                                        if (in[pos + curCapBegin[ell] + n]
                                            != in[pos + i + n])
                                            break;
                                        ++n;
                                    }
                                } else {
                                    while (n < delta) {
                                        QChar a = in[pos + curCapBegin[ell] + n];
                                        QChar b = in[pos + i + n];
                                        if (a.toLower() != b.toLower())
                                            break;
                                        ++n;
                                    }
                                }
                                inside = (n == delta);
                                if (inside)
                                    needSomeSleep = delta - 1;
                            }
                        }
#endif
                    }
                }

                /*
                  We must now update our data structures.
                */
                if (inside) {
#ifndef QT_NO_REGEXP_CAPTURE
                    int *capBegin, *capEnd;
#endif
                    /*
                      If the next state was not encountered yet, all
                      is fine.
                    */
                    if ((m = inNextStack[next]) == -1) {
                        m = nnext++;
                        nextStack[m] = next;
                        inNextStack[next] = m;
#ifndef QT_NO_REGEXP_CAPTURE
                        capBegin = nextCapBegin + m * ncap;
                        capEnd = nextCapEnd + m * ncap;

                    /*
                      Otherwise, we'll first maintain captures in
                      temporary arrays, and decide at the end whether
                      it's best to keep the previous capture zones or
                      the new ones.
                    */
                    } else {
                        capBegin = tempCapBegin;
                        capEnd = tempCapEnd;
#endif
                    }

#ifndef QT_NO_REGEXP_CAPTURE
                    /*
                      Updating the capture zones is much of a task.
                    */
                    if (ncap > 0) {
                        memcpy(capBegin, curCapBegin + j * ncap, ncap * sizeof(int));
                        memcpy(capEnd, curCapEnd + j * ncap, ncap * sizeof(int));
                        int c = scur.atom, n = snext.atom;
                        int p = -1, q = -1;
                        int cap;

                        /*
                          Lemma 1. For any x in the range [0..nf), we
                          have f[x].parent < x.

                          Proof. By looking at startAtom(), it is
                          clear that cf < nf holds all the time, and
                          thus that f[nf].parent < nf.
                        */

                        /*
                          If we are reentering an atom, we empty all
                          capture zones inside it.
                        */
                        if ((q = scur.reenter.value(next)) != 0) {
                            QBitArray b(eng->nf, false);
                            b.setBit(q, true);
                            for (int ell = q + 1; ell < eng->nf; ell++) {
                                if (b.testBit(eng->f.at(ell).parent)) {
                                    b.setBit(ell, true);
                                    cap = eng->f.at(ell).capture;
                                    if (cap >= 0) {
                                        capBegin[cap] = EmptyCapture;
                                        capEnd[cap] = EmptyCapture;
                                    }
                                }
                            }
                            p = eng->f.at(q).parent;

                        /*
                          Otherwise, close the capture zones we are
                          leaving. We are leaving f[c].capture,
                          f[f[c].parent].capture,
                          f[f[f[c].parent].parent].capture, ...,
                          until f[x].capture, with x such that
                          f[x].parent is the youngest common ancestor
                          for c and n.

                          We go up along c's and n's ancestry until
                          we find x.
                        */
                        } else {
                            p = c;
                            q = n;
                            while (p != q) {
                                if (p > q) {
                                    cap = eng->f.at(p).capture;
                                    if (cap >= 0) {
                                        if (capBegin[cap] == i) {
                                            capBegin[cap] = EmptyCapture;
                                            capEnd[cap] = EmptyCapture;
                                        } else {
                                            capEnd[cap] = i;
                                        }
                                    }
                                    p = eng->f.at(p).parent;
                                } else {
                                    q = eng->f.at(q).parent;
                                }
                            }
                        }

                        /*
                          In any case, we now open the capture zones
                          we are entering. We work upwards from n
                          until we reach p (the parent of the atom we
                          reenter or the youngest common ancestor).
                        */
                        while (n > p) {
                            cap = eng->f.at(n).capture;
                            if (cap >= 0) {
                                capBegin[cap] = i;
                                capEnd[cap] = EmptyCapture;
                            }
                            n = eng->f.at(n).parent;
                        }
                        /*
                          If the next state was already in
                          nextStack, we must choose carefully which
                          capture zones we want to keep.
                        */
                        if (capBegin == tempCapBegin &&
                                isBetterCapture(ncap, capBegin, capEnd, nextCapBegin + m * ncap,
                                                nextCapEnd + m * ncap)) {
                            memcpy(nextCapBegin + m * ncap, capBegin, ncap * sizeof(int));
                            memcpy(nextCapEnd + m * ncap, capEnd, ncap * sizeof(int));
                        }
                    }
#ifndef QT_NO_REGEXP_BACKREF
                    /*
                      We are done with updating the capture zones.
                      It's now time to put the next state to sleep,
                      if it needs to, and to remove it from
                      nextStack.
                    */
                    if (needSomeSleep > 0) {
                        QVector<int> zzZ(2 + 2 * ncap);
                        zzZ[0] = i + needSomeSleep;
                        zzZ[1] = next;
                        if (ncap > 0) {
                            memcpy(zzZ.data() + 2, capBegin, ncap * sizeof(int));
                            memcpy(zzZ.data() + 2 + ncap, capEnd, ncap * sizeof(int));
                        }
                        inNextStack[nextStack[--nnext]] = -1;
                        sleeping.append(zzZ);
                    }
#endif
#endif
                }
            }
        }
#ifndef QT_NO_REGEXP_CAPTURE
        /*
          If we reached the final state, hurray! Copy the captured
          zone.
        */
        if (ncap > 0 && (m = inNextStack[QRegExpEngine::FinalState]) != -1) {
            memcpy(capBegin, nextCapBegin + m * ncap, ncap * sizeof(int));
            memcpy(capEnd, nextCapEnd + m * ncap, ncap * sizeof(int));
        }
#ifndef QT_NO_REGEXP_BACKREF
        /*
          It's time to wake up the sleepers.
        */
        j = 0;
        while (j < sleeping.count()) {
            if (sleeping.at(j)[0] == i) {
                const QVector<int> &zzZ = sleeping.at(j);
                int next = zzZ[1];
                const int *capBegin = zzZ.data() + 2;
                const int *capEnd = zzZ.data() + 2 + ncap;
                bool copyOver = true;

                if ((m = inNextStack[next]) == -1) {
                    m = nnext++;
                    nextStack[m] = next;
                    inNextStack[next] = m;
                } else {
                    copyOver = isBetterCapture(ncap, nextCapBegin + m * ncap, nextCapEnd + m * ncap,
                                               capBegin, capEnd);
                }
                if (copyOver) {
                    memcpy(nextCapBegin + m * ncap, capBegin, ncap * sizeof(int));
                    memcpy(nextCapEnd + m * ncap, capEnd, ncap * sizeof(int));
                }

                sleeping.removeAt(j);
            } else {
                ++j;
            }
        }
#endif
#endif
        for (j = 0; j < nnext; j++)
            inNextStack[nextStack[j]] = -1;

        // avoid needless iteration that confuses oneTestMatchedLen
        if (nnext == 1 && nextStack[0] == QRegExpEngine::FinalState
#ifndef QT_NO_REGEXP_BACKREF
             && sleeping.isEmpty()
#endif
           )
            stop = true;

        qSwap(curStack, nextStack);
#ifndef QT_NO_REGEXP_CAPTURE
        qSwap(curCapBegin, nextCapBegin);
        qSwap(curCapEnd, nextCapEnd);
#endif
        ncur = nnext;
        nnext = 0;
        ++i;
    }

#ifndef QT_NO_REGEXP_BACKREF
    /*
      If minimal matching is enabled, we might have some sleepers
      left.
    */
    if (!sleeping.isEmpty())
        sleeping.clear();
#endif

    oneTestMatchedLen = i - 1;
    return (matchLen >= 0);
}

#ifndef QT_NO_REGEXP_CCLASS

QRegExpCharClass::QRegExpCharClass()
    : c(0), n(false)
{
#ifndef QT_NO_REGEXP_OPTIM
    occ1.fill(NoOccurrence, NumBadChars);
#endif
}

QRegExpCharClass &QRegExpCharClass::operator=(const QRegExpCharClass &cc)
{
    c = cc.c;
    r = cc.r;
    n = cc.n;
#ifndef QT_NO_REGEXP_OPTIM
    occ1 = cc.occ1;
#endif
    return *this;
}

void QRegExpCharClass::clear()
{
    c = 0;
    r.resize(0);
    n = false;
}

void QRegExpCharClass::setNegative(bool negative)
{
    n = negative;
#ifndef QT_NO_REGEXP_OPTIM
    occ1.fill(0, NumBadChars);
#endif
}

void QRegExpCharClass::addCategories(int cats)
{
    c |= cats;
#ifndef QT_NO_REGEXP_OPTIM
    occ1.fill(0, NumBadChars);
#endif
}

void QRegExpCharClass::addRange(ushort from, ushort to)
{
    if (from > to)
        qSwap(from, to);
    int m = r.size();
    r.resize(m + 1);
    r[m].from = from;
    r[m].len = to - from + 1;

#ifndef QT_NO_REGEXP_OPTIM
    int i;

    if (to - from < NumBadChars) {
        if (from % NumBadChars <= to % NumBadChars) {
            for (i = from % NumBadChars; i <= to % NumBadChars; i++)
                occ1[i] = 0;
        } else {
            for (i = 0; i <= to % NumBadChars; i++)
                occ1[i] = 0;
            for (i = from % NumBadChars; i < NumBadChars; i++)
                occ1[i] = 0;
        }
    } else {
        occ1.fill(0, NumBadChars);
    }
#endif
}

bool QRegExpCharClass::in(QChar ch) const
{
#ifndef QT_NO_REGEXP_OPTIM
    if (occ1.at(BadChar(ch)) == NoOccurrence)
        return n;
#endif

    if (c != 0 && (c & (1 << (int)ch.category())) != 0)
        return !n;

    const int uc = ch.unicode();
    int size = r.size();

    for (int i = 0; i < size; ++i) {
        const QRegExpCharClassRange &range = r.at(i);
        if (uint(uc - range.from) < uint(r.at(i).len))
            return !n;
    }
    return n;
}

#if defined(QT_DEBUG)
void QRegExpCharClass::dump() const
{
    int i;
    qDebug("    %stive character class", n ? "nega" : "posi");
#ifndef QT_NO_REGEXP_CCLASS
    if (c != 0)
        qDebug("      categories 0x%.8x", c);
#endif
    for (i = 0; i < r.size(); i++)
        qDebug("      0x%.4x through 0x%.4x", r[i].from, r[i].from + r[i].len - 1);
}
#endif
#endif

QRegExpEngine::Box::Box(QRegExpEngine *engine)
    : eng(engine), skipanchors(0)
#ifndef QT_NO_REGEXP_OPTIM
      , earlyStart(0), lateStart(0), maxl(0)
#endif
{
#ifndef QT_NO_REGEXP_OPTIM
    occ1.fill(NoOccurrence, NumBadChars);
#endif
    minl = 0;
}

QRegExpEngine::Box &QRegExpEngine::Box::operator=(const Box &b)
{
    eng = b.eng;
    ls = b.ls;
    rs = b.rs;
    lanchors = b.lanchors;
    ranchors = b.ranchors;
    skipanchors = b.skipanchors;
#ifndef QT_NO_REGEXP_OPTIM
    earlyStart = b.earlyStart;
    lateStart = b.lateStart;
    str = b.str;
    leftStr = b.leftStr;
    rightStr = b.rightStr;
    maxl = b.maxl;
    occ1 = b.occ1;
#endif
    minl = b.minl;
    return *this;
}

void QRegExpEngine::Box::set(QChar ch)
{
    ls.resize(1);
    ls[0] = eng->createState(ch);
    rs = ls;
#ifndef QT_NO_REGEXP_OPTIM
    str = ch;
    leftStr = ch;
    rightStr = ch;
    maxl = 1;
    occ1[BadChar(ch)] = 0;
#endif
    minl = 1;
}

void QRegExpEngine::Box::set(const QRegExpCharClass &cc)
{
    ls.resize(1);
    ls[0] = eng->createState(cc);
    rs = ls;
#ifndef QT_NO_REGEXP_OPTIM
    maxl = 1;
    occ1 = cc.firstOccurrence();
#endif
    minl = 1;
}

#ifndef QT_NO_REGEXP_BACKREF
void QRegExpEngine::Box::set(int bref)
{
    ls.resize(1);
    ls[0] = eng->createState(bref);
    rs = ls;
    if (bref >= 1 && bref <= MaxBackRefs)
        skipanchors = Anchor_BackRef0Empty << bref;
#ifndef QT_NO_REGEXP_OPTIM
    maxl = InftyLen;
#endif
    minl = 0;
}
#endif

void QRegExpEngine::Box::cat(const Box &b)
{
    eng->addCatTransitions(rs, b.ls);
    addAnchorsToEngine(b);
    if (minl == 0) {
        lanchors.unite(b.lanchors);
        if (skipanchors != 0) {
            for (int i = 0; i < b.ls.size(); i++) {
                int a = eng->anchorConcatenation(lanchors.value(b.ls.at(i), 0), skipanchors);
                lanchors.insert(b.ls.at(i), a);
            }
        }
        mergeInto(&ls, b.ls);
    }
    if (b.minl == 0) {
        ranchors.unite(b.ranchors);
        if (b.skipanchors != 0) {
            for (int i = 0; i < rs.size(); i++) {
                int a = eng->anchorConcatenation(ranchors.value(rs.at(i), 0), b.skipanchors);
                ranchors.insert(rs.at(i), a);
            }
        }
        mergeInto(&rs, b.rs);
    } else {
        ranchors = b.ranchors;
        rs = b.rs;
    }

#ifndef QT_NO_REGEXP_OPTIM
    if (maxl != InftyLen) {
        if (rightStr.length() + b.leftStr.length() >
             qMax(str.length(), b.str.length())) {
            earlyStart = minl - rightStr.length();
            lateStart = maxl - rightStr.length();
            str = rightStr + b.leftStr;
        } else if (b.str.length() > str.length()) {
            earlyStart = minl + b.earlyStart;
            lateStart = maxl + b.lateStart;
            str = b.str;
        }
    }

    if (leftStr.length() == maxl)
        leftStr += b.leftStr;

    if (b.rightStr.length() == b.maxl) {
        rightStr += b.rightStr;
    } else {
        rightStr = b.rightStr;
    }

    if (maxl == InftyLen || b.maxl == InftyLen) {
        maxl = InftyLen;
    } else {
        maxl += b.maxl;
    }

    for (int i = 0; i < NumBadChars; i++) {
        if (b.occ1.at(i) != NoOccurrence && minl + b.occ1.at(i) < occ1.at(i))
            occ1[i] = minl + b.occ1.at(i);
    }
#endif

    minl += b.minl;
    if (minl == 0)
        skipanchors = eng->anchorConcatenation(skipanchors, b.skipanchors);
    else
        skipanchors = 0;
}

void QRegExpEngine::Box::orx(const Box &b)
{
    mergeInto(&ls, b.ls);
    lanchors.unite(b.lanchors);
    mergeInto(&rs, b.rs);
    ranchors.unite(b.ranchors);

    if (b.minl == 0) {
        if (minl == 0)
            skipanchors = eng->anchorAlternation(skipanchors, b.skipanchors);
        else
            skipanchors = b.skipanchors;
    }

#ifndef QT_NO_REGEXP_OPTIM
    for (int i = 0; i < NumBadChars; i++) {
        if (occ1.at(i) > b.occ1.at(i))
            occ1[i] = b.occ1.at(i);
    }
    earlyStart = 0;
    lateStart = 0;
    str = QString();
    leftStr = QString();
    rightStr = QString();
    if (b.maxl > maxl)
        maxl = b.maxl;
#endif
    if (b.minl < minl)
        minl = b.minl;
}

void QRegExpEngine::Box::plus(int atom)
{
#ifndef QT_NO_REGEXP_CAPTURE
    eng->addPlusTransitions(rs, ls, atom);
#else
    Q_UNUSED(atom);
    eng->addCatTransitions(rs, ls);
#endif
    addAnchorsToEngine(*this);
#ifndef QT_NO_REGEXP_OPTIM
    maxl = InftyLen;
#endif
}

void QRegExpEngine::Box::opt()
{
#ifndef QT_NO_REGEXP_OPTIM
    earlyStart = 0;
    lateStart = 0;
    str = QString();
    leftStr = QString();
    rightStr = QString();
#endif
    skipanchors = 0;
    minl = 0;
}

void QRegExpEngine::Box::catAnchor(int a)
{
    if (a != 0) {
        for (int i = 0; i < rs.size(); i++) {
            a = eng->anchorConcatenation(ranchors.value(rs.at(i), 0), a);
            ranchors.insert(rs.at(i), a);
        }
        if (minl == 0)
            skipanchors = eng->anchorConcatenation(skipanchors, a);
    }
}

#ifndef QT_NO_REGEXP_OPTIM
void QRegExpEngine::Box::setupHeuristics()
{
    eng->goodEarlyStart = earlyStart;
    eng->goodLateStart = lateStart;
    eng->goodStr = eng->cs ? str : str.toLower();

    eng->minl = minl;
    if (eng->cs) {
        /*
          A regular expression such as 112|1 has occ1['2'] = 2 and minl =
          1 at this point. An entry of occ1 has to be at most minl or
          infinity for the rest of the algorithm to go well.

          We waited until here before normalizing these cases (instead of
          doing it in Box::orx()) because sometimes things improve by
          themselves. Consider for example (112|1)34.
        */
        for (int i = 0; i < NumBadChars; i++) {
            if (occ1.at(i) != NoOccurrence && occ1.at(i) >= minl)
                occ1[i] = minl;
        }
        eng->occ1 = occ1;
    } else {
        eng->occ1.fill(0, NumBadChars);
    }

    eng->heuristicallyChooseHeuristic();
}
#endif

#if defined(QT_DEBUG)
void QRegExpEngine::Box::dump() const
{
    int i;
    qDebug("Box of at least %d character%s", minl, minl == 1 ? "" : "s");
    qDebug("  Left states:");
    for (i = 0; i < ls.size(); i++) {
        if (lanchors.value(ls[i], 0) == 0)
            qDebug("    %d", ls[i]);
        else
            qDebug("    %d [anchors 0x%.8x]", ls[i], lanchors[ls[i]]);
    }
    qDebug("  Right states:");
    for (i = 0; i < rs.size(); i++) {
        if (ranchors.value(rs[i], 0) == 0)
            qDebug("    %d", rs[i]);
        else
            qDebug("    %d [anchors 0x%.8x]", rs[i], ranchors[rs[i]]);
    }
    qDebug("  Skip anchors: 0x%.8x", skipanchors);
}
#endif

void QRegExpEngine::Box::addAnchorsToEngine(const Box &to) const
{
    for (int i = 0; i < to.ls.size(); i++) {
        for (int j = 0; j < rs.size(); j++) {
            int a = eng->anchorConcatenation(ranchors.value(rs.at(j), 0),
                                             to.lanchors.value(to.ls.at(i), 0));
            eng->addAnchors(rs[j], to.ls[i], a);
        }
    }
}

void QRegExpEngine::setupCategoriesRangeMap()
{
   categoriesRangeMap.insert("IsBasicLatin",                           qMakePair(0x0000, 0x007F));
   categoriesRangeMap.insert("IsLatin-1Supplement",                    qMakePair(0x0080, 0x00FF));
   categoriesRangeMap.insert("IsLatinExtended-A",                      qMakePair(0x0100, 0x017F));
   categoriesRangeMap.insert("IsLatinExtended-B",                      qMakePair(0x0180, 0x024F));
   categoriesRangeMap.insert("IsIPAExtensions",                        qMakePair(0x0250, 0x02AF));
   categoriesRangeMap.insert("IsSpacingModifierLetters",               qMakePair(0x02B0, 0x02FF));
   categoriesRangeMap.insert("IsCombiningDiacriticalMarks",            qMakePair(0x0300, 0x036F));
   categoriesRangeMap.insert("IsGreek",                                qMakePair(0x0370, 0x03FF));
   categoriesRangeMap.insert("IsCyrillic",                             qMakePair(0x0400, 0x04FF));
   categoriesRangeMap.insert("IsCyrillicSupplement",                   qMakePair(0x0500, 0x052F));
   categoriesRangeMap.insert("IsArmenian",                             qMakePair(0x0530, 0x058F));
   categoriesRangeMap.insert("IsHebrew",                               qMakePair(0x0590, 0x05FF));
   categoriesRangeMap.insert("IsArabic",                               qMakePair(0x0600, 0x06FF));
   categoriesRangeMap.insert("IsSyriac",                               qMakePair(0x0700, 0x074F));
   categoriesRangeMap.insert("IsArabicSupplement",                     qMakePair(0x0750, 0x077F));
   categoriesRangeMap.insert("IsThaana",                               qMakePair(0x0780, 0x07BF));
   categoriesRangeMap.insert("IsDevanagari",                           qMakePair(0x0900, 0x097F));
   categoriesRangeMap.insert("IsBengali",                              qMakePair(0x0980, 0x09FF));
   categoriesRangeMap.insert("IsGurmukhi",                             qMakePair(0x0A00, 0x0A7F));
   categoriesRangeMap.insert("IsGujarati",                             qMakePair(0x0A80, 0x0AFF));
   categoriesRangeMap.insert("IsOriya",                                qMakePair(0x0B00, 0x0B7F));
   categoriesRangeMap.insert("IsTamil",                                qMakePair(0x0B80, 0x0BFF));
   categoriesRangeMap.insert("IsTelugu",                               qMakePair(0x0C00, 0x0C7F));
   categoriesRangeMap.insert("IsKannada",                              qMakePair(0x0C80, 0x0CFF));
   categoriesRangeMap.insert("IsMalayalam",                            qMakePair(0x0D00, 0x0D7F));
   categoriesRangeMap.insert("IsSinhala",                              qMakePair(0x0D80, 0x0DFF));
   categoriesRangeMap.insert("IsThai",                                 qMakePair(0x0E00, 0x0E7F));
   categoriesRangeMap.insert("IsLao",                                  qMakePair(0x0E80, 0x0EFF));
   categoriesRangeMap.insert("IsTibetan",                              qMakePair(0x0F00, 0x0FFF));
   categoriesRangeMap.insert("IsMyanmar",                              qMakePair(0x1000, 0x109F));
   categoriesRangeMap.insert("IsGeorgian",                             qMakePair(0x10A0, 0x10FF));
   categoriesRangeMap.insert("IsHangulJamo",                           qMakePair(0x1100, 0x11FF));
   categoriesRangeMap.insert("IsEthiopic",                             qMakePair(0x1200, 0x137F));
   categoriesRangeMap.insert("IsEthiopicSupplement",                   qMakePair(0x1380, 0x139F));
   categoriesRangeMap.insert("IsCherokee",                             qMakePair(0x13A0, 0x13FF));
   categoriesRangeMap.insert("IsUnifiedCanadianAboriginalSyllabics",   qMakePair(0x1400, 0x167F));
   categoriesRangeMap.insert("IsOgham",                                qMakePair(0x1680, 0x169F));
   categoriesRangeMap.insert("IsRunic",                                qMakePair(0x16A0, 0x16FF));
   categoriesRangeMap.insert("IsTagalog",                              qMakePair(0x1700, 0x171F));
   categoriesRangeMap.insert("IsHanunoo",                              qMakePair(0x1720, 0x173F));
   categoriesRangeMap.insert("IsBuhid",                                qMakePair(0x1740, 0x175F));
   categoriesRangeMap.insert("IsTagbanwa",                             qMakePair(0x1760, 0x177F));
   categoriesRangeMap.insert("IsKhmer",                                qMakePair(0x1780, 0x17FF));
   categoriesRangeMap.insert("IsMongolian",                            qMakePair(0x1800, 0x18AF));
   categoriesRangeMap.insert("IsLimbu",                                qMakePair(0x1900, 0x194F));
   categoriesRangeMap.insert("IsTaiLe",                                qMakePair(0x1950, 0x197F));
   categoriesRangeMap.insert("IsNewTaiLue",                            qMakePair(0x1980, 0x19DF));
   categoriesRangeMap.insert("IsKhmerSymbols",                         qMakePair(0x19E0, 0x19FF));
   categoriesRangeMap.insert("IsBuginese",                             qMakePair(0x1A00, 0x1A1F));
   categoriesRangeMap.insert("IsPhoneticExtensions",                   qMakePair(0x1D00, 0x1D7F));
   categoriesRangeMap.insert("IsPhoneticExtensionsSupplement",         qMakePair(0x1D80, 0x1DBF));
   categoriesRangeMap.insert("IsCombiningDiacriticalMarksSupplement",  qMakePair(0x1DC0, 0x1DFF));
   categoriesRangeMap.insert("IsLatinExtendedAdditional",              qMakePair(0x1E00, 0x1EFF));
   categoriesRangeMap.insert("IsGreekExtended",                        qMakePair(0x1F00, 0x1FFF));
   categoriesRangeMap.insert("IsGeneralPunctuation",                   qMakePair(0x2000, 0x206F));
   categoriesRangeMap.insert("IsSuperscriptsandSubscripts",            qMakePair(0x2070, 0x209F));
   categoriesRangeMap.insert("IsCurrencySymbols",                      qMakePair(0x20A0, 0x20CF));
   categoriesRangeMap.insert("IsCombiningMarksforSymbols",             qMakePair(0x20D0, 0x20FF));
   categoriesRangeMap.insert("IsLetterlikeSymbols",                    qMakePair(0x2100, 0x214F));
   categoriesRangeMap.insert("IsNumberForms",                          qMakePair(0x2150, 0x218F));
   categoriesRangeMap.insert("IsArrows",                               qMakePair(0x2190, 0x21FF));
   categoriesRangeMap.insert("IsMathematicalOperators",                qMakePair(0x2200, 0x22FF));
   categoriesRangeMap.insert("IsMiscellaneousTechnical",               qMakePair(0x2300, 0x23FF));
   categoriesRangeMap.insert("IsControlPictures",                      qMakePair(0x2400, 0x243F));
   categoriesRangeMap.insert("IsOpticalCharacterRecognition",          qMakePair(0x2440, 0x245F));
   categoriesRangeMap.insert("IsEnclosedAlphanumerics",                qMakePair(0x2460, 0x24FF));
   categoriesRangeMap.insert("IsBoxDrawing",                           qMakePair(0x2500, 0x257F));
   categoriesRangeMap.insert("IsBlockElements",                        qMakePair(0x2580, 0x259F));
   categoriesRangeMap.insert("IsGeometricShapes",                      qMakePair(0x25A0, 0x25FF));
   categoriesRangeMap.insert("IsMiscellaneousSymbols",                 qMakePair(0x2600, 0x26FF));
   categoriesRangeMap.insert("IsDingbats",                             qMakePair(0x2700, 0x27BF));
   categoriesRangeMap.insert("IsMiscellaneousMathematicalSymbols-A",   qMakePair(0x27C0, 0x27EF));
   categoriesRangeMap.insert("IsSupplementalArrows-A",                 qMakePair(0x27F0, 0x27FF));
   categoriesRangeMap.insert("IsBraillePatterns",                      qMakePair(0x2800, 0x28FF));
   categoriesRangeMap.insert("IsSupplementalArrows-B",                 qMakePair(0x2900, 0x297F));
   categoriesRangeMap.insert("IsMiscellaneousMathematicalSymbols-B",   qMakePair(0x2980, 0x29FF));
   categoriesRangeMap.insert("IsSupplementalMathematicalOperators",    qMakePair(0x2A00, 0x2AFF));
   categoriesRangeMap.insert("IsMiscellaneousSymbolsandArrows",        qMakePair(0x2B00, 0x2BFF));
   categoriesRangeMap.insert("IsGlagolitic",                           qMakePair(0x2C00, 0x2C5F));
   categoriesRangeMap.insert("IsCoptic",                               qMakePair(0x2C80, 0x2CFF));
   categoriesRangeMap.insert("IsGeorgianSupplement",                   qMakePair(0x2D00, 0x2D2F));
   categoriesRangeMap.insert("IsTifinagh",                             qMakePair(0x2D30, 0x2D7F));
   categoriesRangeMap.insert("IsEthiopicExtended",                     qMakePair(0x2D80, 0x2DDF));
   categoriesRangeMap.insert("IsSupplementalPunctuation",              qMakePair(0x2E00, 0x2E7F));
   categoriesRangeMap.insert("IsCJKRadicalsSupplement",                qMakePair(0x2E80, 0x2EFF));
   categoriesRangeMap.insert("IsKangxiRadicals",                       qMakePair(0x2F00, 0x2FDF));
   categoriesRangeMap.insert("IsIdeographicDescriptionCharacters",     qMakePair(0x2FF0, 0x2FFF));
   categoriesRangeMap.insert("IsCJKSymbolsandPunctuation",             qMakePair(0x3000, 0x303F));
   categoriesRangeMap.insert("IsHiragana",                             qMakePair(0x3040, 0x309F));
   categoriesRangeMap.insert("IsKatakana",                             qMakePair(0x30A0, 0x30FF));
   categoriesRangeMap.insert("IsBopomofo",                             qMakePair(0x3100, 0x312F));
   categoriesRangeMap.insert("IsHangulCompatibilityJamo",              qMakePair(0x3130, 0x318F));
   categoriesRangeMap.insert("IsKanbun",                               qMakePair(0x3190, 0x319F));
   categoriesRangeMap.insert("IsBopomofoExtended",                     qMakePair(0x31A0, 0x31BF));
   categoriesRangeMap.insert("IsCJKStrokes",                           qMakePair(0x31C0, 0x31EF));
   categoriesRangeMap.insert("IsKatakanaPhoneticExtensions",           qMakePair(0x31F0, 0x31FF));
   categoriesRangeMap.insert("IsEnclosedCJKLettersandMonths",          qMakePair(0x3200, 0x32FF));
   categoriesRangeMap.insert("IsCJKCompatibility",                     qMakePair(0x3300, 0x33FF));
   categoriesRangeMap.insert("IsCJKUnifiedIdeographsExtensionA",       qMakePair(0x3400, 0x4DB5));
   categoriesRangeMap.insert("IsYijingHexagramSymbols",                qMakePair(0x4DC0, 0x4DFF));
   categoriesRangeMap.insert("IsCJKUnifiedIdeographs",                 qMakePair(0x4E00, 0x9FFF));
   categoriesRangeMap.insert("IsYiSyllables",                          qMakePair(0xA000, 0xA48F));
   categoriesRangeMap.insert("IsYiRadicals",                           qMakePair(0xA490, 0xA4CF));
   categoriesRangeMap.insert("IsModifierToneLetters",                  qMakePair(0xA700, 0xA71F));
   categoriesRangeMap.insert("IsSylotiNagri",                          qMakePair(0xA800, 0xA82F));
   categoriesRangeMap.insert("IsHangulSyllables",                      qMakePair(0xAC00, 0xD7A3));
   categoriesRangeMap.insert("IsPrivateUse",                           qMakePair(0xE000, 0xF8FF));
   categoriesRangeMap.insert("IsCJKCompatibilityIdeographs",           qMakePair(0xF900, 0xFAFF));
   categoriesRangeMap.insert("IsAlphabeticPresentationForms",          qMakePair(0xFB00, 0xFB4F));
   categoriesRangeMap.insert("IsArabicPresentationForms-A",            qMakePair(0xFB50, 0xFDFF));
   categoriesRangeMap.insert("IsVariationSelectors",                   qMakePair(0xFE00, 0xFE0F));
   categoriesRangeMap.insert("IsVerticalForms",                        qMakePair(0xFE10, 0xFE1F));
   categoriesRangeMap.insert("IsCombiningHalfMarks",                   qMakePair(0xFE20, 0xFE2F));
   categoriesRangeMap.insert("IsCJKCompatibilityForms",                qMakePair(0xFE30, 0xFE4F));
   categoriesRangeMap.insert("IsSmallFormVariants",                    qMakePair(0xFE50, 0xFE6F));
   categoriesRangeMap.insert("IsArabicPresentationForms-B",            qMakePair(0xFE70, 0xFEFF));
   categoriesRangeMap.insert("IsHalfwidthandFullwidthForms",           qMakePair(0xFF00, 0xFFEF));
   categoriesRangeMap.insert("IsSpecials",                             qMakePair(0xFFF0, 0xFFFF));
   categoriesRangeMap.insert("IsLinearBSyllabary",                     qMakePair(0x10000, 0x1007F));
   categoriesRangeMap.insert("IsLinearBIdeograms",                     qMakePair(0x10080, 0x100FF));
   categoriesRangeMap.insert("IsAegeanNumbers",                        qMakePair(0x10100, 0x1013F));
   categoriesRangeMap.insert("IsAncientGreekNumbers",                  qMakePair(0x10140, 0x1018F));
   categoriesRangeMap.insert("IsOldItalic",                            qMakePair(0x10300, 0x1032F));
   categoriesRangeMap.insert("IsGothic",                               qMakePair(0x10330, 0x1034F));
   categoriesRangeMap.insert("IsUgaritic",                             qMakePair(0x10380, 0x1039F));
   categoriesRangeMap.insert("IsOldPersian",                           qMakePair(0x103A0, 0x103DF));
   categoriesRangeMap.insert("IsDeseret",                              qMakePair(0x10400, 0x1044F));
   categoriesRangeMap.insert("IsShavian",                              qMakePair(0x10450, 0x1047F));
   categoriesRangeMap.insert("IsOsmanya",                              qMakePair(0x10480, 0x104AF));
   categoriesRangeMap.insert("IsCypriotSyllabary",                     qMakePair(0x10800, 0x1083F));
   categoriesRangeMap.insert("IsKharoshthi",                           qMakePair(0x10A00, 0x10A5F));
   categoriesRangeMap.insert("IsByzantineMusicalSymbols",              qMakePair(0x1D000, 0x1D0FF));
   categoriesRangeMap.insert("IsMusicalSymbols",                       qMakePair(0x1D100, 0x1D1FF));
   categoriesRangeMap.insert("IsAncientGreekMusicalNotation",          qMakePair(0x1D200, 0x1D24F));
   categoriesRangeMap.insert("IsTaiXuanJingSymbols",                   qMakePair(0x1D300, 0x1D35F));
   categoriesRangeMap.insert("IsMathematicalAlphanumericSymbols",      qMakePair(0x1D400, 0x1D7FF));
   categoriesRangeMap.insert("IsCJKUnifiedIdeographsExtensionB",       qMakePair(0x20000, 0x2A6DF));
   categoriesRangeMap.insert("IsCJKCompatibilityIdeographsSupplement", qMakePair(0x2F800, 0x2FA1F));
   categoriesRangeMap.insert("IsTags",                                 qMakePair(0xE0000, 0xE007F));
   categoriesRangeMap.insert("IsVariationSelectorsSupplement",         qMakePair(0xE0100, 0xE01EF));
   categoriesRangeMap.insert("IsSupplementaryPrivateUseArea-A",        qMakePair(0xF0000, 0xFFFFF));
   categoriesRangeMap.insert("IsSupplementaryPrivateUseArea-B",        qMakePair(0x100000, 0x10FFFF));
}

int QRegExpEngine::getChar()
{
    return (yyPos == yyLen) ? EOS : yyIn[yyPos++].unicode();
}

int QRegExpEngine::getEscape()
{
#ifndef QT_NO_REGEXP_ESCAPE
    const char tab[] = "afnrtv"; // no b, as \b means word boundary
    const char backTab[] = "\a\f\n\r\t\v";
    ushort low;
    int i;
#endif
    ushort val;
    int prevCh = yyCh;

    if (prevCh == EOS) {
        error(RXERR_END);
        return Tok_Char | '\\';
    }
    yyCh = getChar();
#ifndef QT_NO_REGEXP_ESCAPE
    if ((prevCh & ~0xff) == 0) {
        const char *p = strchr(tab, prevCh);
        if (p != 0)
            return Tok_Char | backTab[p - tab];
    }
#endif

    switch (prevCh) {
#ifndef QT_NO_REGEXP_ESCAPE
    case '0':
        val = 0;
        for (i = 0; i < 3; i++) {
            if (yyCh >= '0' && yyCh <= '7')
                val = (val << 3) | (yyCh - '0');
            else
                break;
            yyCh = getChar();
        }
        if ((val & ~0377) != 0)
            error(RXERR_OCTAL);
        return Tok_Char | val;
#endif
#ifndef QT_NO_REGEXP_ESCAPE
    case 'B':
        return Tok_NonWord;
#endif
#ifndef QT_NO_REGEXP_CCLASS
    case 'D':
        // see QChar::isDigit()
        yyCharClass->addCategories(0x7fffffef);
        return Tok_CharClass;
    case 'S':
        // see QChar::isSpace()
        yyCharClass->addCategories(0x7ffff87f);
        yyCharClass->addRange(0x0000, 0x0008);
        yyCharClass->addRange(0x000e, 0x001f);
        yyCharClass->addRange(0x007f, 0x009f);
        return Tok_CharClass;
    case 'W':
        // see QChar::isLetterOrNumber() and QChar::isMark()
        yyCharClass->addCategories(0x7fe07f81);
        yyCharClass->addRange(0x203f, 0x2040);
        yyCharClass->addSingleton(0x2040);
        yyCharClass->addSingleton(0x2054);
        yyCharClass->addSingleton(0x30fb);
        yyCharClass->addRange(0xfe33, 0xfe34);
        yyCharClass->addRange(0xfe4d, 0xfe4f);
        yyCharClass->addSingleton(0xff3f);
        yyCharClass->addSingleton(0xff65);
        return Tok_CharClass;
#endif
#ifndef QT_NO_REGEXP_ESCAPE
    case 'b':
        return Tok_Word;
#endif
#ifndef QT_NO_REGEXP_CCLASS
    case 'd':
        // see QChar::isDigit()
        yyCharClass->addCategories(0x00000010);
        return Tok_CharClass;
    case 's':
        // see QChar::isSpace()
        yyCharClass->addCategories(0x00000380);
        yyCharClass->addRange(0x0009, 0x000d);
        return Tok_CharClass;
    case 'w':
        // see QChar::isLetterOrNumber() and QChar::isMark()
        yyCharClass->addCategories(0x000f807e);
        yyCharClass->addSingleton(0x005f); // '_'
        return Tok_CharClass;
    case 'I':
        if (xmlSchemaExtensions) {
            yyCharClass->setNegative(!yyCharClass->negative());
            // fall through
        }
    case 'i':
        if (xmlSchemaExtensions) {
            yyCharClass->addCategories(0x000f807e);
            yyCharClass->addSingleton(0x003a); // ':'
            yyCharClass->addSingleton(0x005f); // '_'
            yyCharClass->addRange(0x0041, 0x005a); // [A-Z]
            yyCharClass->addRange(0x0061, 0x007a); // [a-z]
            yyCharClass->addRange(0xc0, 0xd6);
            yyCharClass->addRange(0xd8, 0xf6);
            yyCharClass->addRange(0xf8, 0x2ff);
            yyCharClass->addRange(0x370, 0x37d);
            yyCharClass->addRange(0x37f, 0x1fff);
            yyCharClass->addRange(0x200c, 0x200d);
            yyCharClass->addRange(0x2070, 0x218f);
            yyCharClass->addRange(0x2c00, 0x2fef);
            yyCharClass->addRange(0x3001, 0xd7ff);
            yyCharClass->addRange(0xf900, 0xfdcf);
            yyCharClass->addRange(0xfdf0, 0xfffd);
            yyCharClass->addRange((ushort)0x10000, (ushort)0xeffff);
        }
        return Tok_CharClass;
    case 'C':
        if (xmlSchemaExtensions) {
            yyCharClass->setNegative(!yyCharClass->negative());
            // fall through
        }
    case 'c':
        if (xmlSchemaExtensions) {
            yyCharClass->addCategories(0x000f807e);
            yyCharClass->addSingleton(0x002d); // '-'
            yyCharClass->addSingleton(0x002e); // '.'
            yyCharClass->addSingleton(0x003a); // ':'
            yyCharClass->addSingleton(0x005f); // '_'
            yyCharClass->addSingleton(0xb7);
            yyCharClass->addRange(0x0030, 0x0039); // [0-9]
            yyCharClass->addRange(0x0041, 0x005a); // [A-Z]
            yyCharClass->addRange(0x0061, 0x007a); // [a-z]
            yyCharClass->addRange(0xc0, 0xd6);
            yyCharClass->addRange(0xd8, 0xf6);
            yyCharClass->addRange(0xf8, 0x2ff);
            yyCharClass->addRange(0x370, 0x37d);
            yyCharClass->addRange(0x37f, 0x1fff);
            yyCharClass->addRange(0x200c, 0x200d);
            yyCharClass->addRange(0x2070, 0x218f);
            yyCharClass->addRange(0x2c00, 0x2fef);
            yyCharClass->addRange(0x3001, 0xd7ff);
            yyCharClass->addRange(0xf900, 0xfdcf);
            yyCharClass->addRange(0xfdf0, 0xfffd);
            yyCharClass->addRange((ushort)0x10000, (ushort)0xeffff);
            yyCharClass->addRange(0x0300, 0x036f);
            yyCharClass->addRange(0x203f, 0x2040);
        }
        return Tok_CharClass;
    case 'P':
        if (xmlSchemaExtensions) {
            yyCharClass->setNegative(!yyCharClass->negative());
            // fall through
        }
    case 'p':
        if (xmlSchemaExtensions) {
            if (yyCh != '{') {
                error(RXERR_CHARCLASS);
                return Tok_CharClass;
            }

            QByteArray category;
            yyCh = getChar();
            while (yyCh != '}') {
                if (yyCh == EOS) {
                    error(RXERR_END);
                    return Tok_CharClass;
                }
                category.append(yyCh);
                yyCh = getChar();
            }
            yyCh = getChar(); // skip closing '}'

            if (category == "M") {
                yyCharClass->addCategories(0x0000000e);
            } else if (category == "Mn") {
                yyCharClass->addCategories(0x00000002);
            } else if (category == "Mc") {
                yyCharClass->addCategories(0x00000004);
            } else if (category == "Me") {
                yyCharClass->addCategories(0x00000008);
            } else if (category == "N") {
                yyCharClass->addCategories(0x00000070);
            } else if (category == "Nd") {
                yyCharClass->addCategories(0x00000010);
            } else if (category == "Nl") {
                yyCharClass->addCategories(0x00000020);
            } else if (category == "No") {
                yyCharClass->addCategories(0x00000040);
            } else if (category == "Z") {
                yyCharClass->addCategories(0x00000380);
            } else if (category == "Zs") {
                yyCharClass->addCategories(0x00000080);
            } else if (category == "Zl") {
                yyCharClass->addCategories(0x00000100);
            } else if (category == "Zp") {
                yyCharClass->addCategories(0x00000200);
            } else if (category == "C") {
                yyCharClass->addCategories(0x00006c00);
            } else if (category == "Cc") {
                yyCharClass->addCategories(0x00000400);
            } else if (category == "Cf") {
                yyCharClass->addCategories(0x00000800);
            } else if (category == "Cs") {
                yyCharClass->addCategories(0x00001000);
            } else if (category == "Co") {
                yyCharClass->addCategories(0x00002000);
            } else if (category == "Cn") {
                yyCharClass->addCategories(0x00004000);
            } else if (category == "L") {
                yyCharClass->addCategories(0x000f8000);
            } else if (category == "Lu") {
                yyCharClass->addCategories(0x00008000);
            } else if (category == "Ll") {
                yyCharClass->addCategories(0x00010000);
            } else if (category == "Lt") {
                yyCharClass->addCategories(0x00020000);
            } else if (category == "Lm") {
                yyCharClass->addCategories(0x00040000);
            } else if (category == "Lo") {
                yyCharClass->addCategories(0x00080000);
            } else if (category == "P") {
                yyCharClass->addCategories(0x4f580780);
            } else if (category == "Pc") {
                yyCharClass->addCategories(0x00100000);
            } else if (category == "Pd") {
                yyCharClass->addCategories(0x00200000);
            } else if (category == "Ps") {
                yyCharClass->addCategories(0x00400000);
            } else if (category == "Pe") {
                yyCharClass->addCategories(0x00800000);
            } else if (category == "Pi") {
                yyCharClass->addCategories(0x01000000);
            } else if (category == "Pf") {
                yyCharClass->addCategories(0x02000000);
            } else if (category == "Po") {
                yyCharClass->addCategories(0x04000000);
            } else if (category == "S") {
                yyCharClass->addCategories(0x78000000);
            } else if (category == "Sm") {
                yyCharClass->addCategories(0x08000000);
            } else if (category == "Sc") {
                yyCharClass->addCategories(0x10000000);
            } else if (category == "Sk") {
                yyCharClass->addCategories(0x20000000);
            } else if (category == "So") {
                yyCharClass->addCategories(0x40000000);
            } else if (category.startsWith("Is")) {
                if (categoriesRangeMap.isEmpty())
                    setupCategoriesRangeMap();

                if (categoriesRangeMap.contains(category)) {
                    const QPair<int, int> range = categoriesRangeMap.value(category);
                    yyCharClass->addRange(range.first, range.second);
                } else {
                    error(RXERR_CATEGORY);
                }
            } else {
                error(RXERR_CATEGORY);
            }
        }
        return Tok_CharClass;
#endif
#ifndef QT_NO_REGEXP_ESCAPE
    case 'x':
        val = 0;
        for (i = 0; i < 4; i++) {
            low = QChar(yyCh).toLower().unicode();
            if (low >= '0' && low <= '9')
                val = (val << 4) | (low - '0');
            else if (low >= 'a' && low <= 'f')
                val = (val << 4) | (low - 'a' + 10);
            else
                break;
            yyCh = getChar();
        }
        return Tok_Char | val;
#endif
    default:
        if (prevCh >= '1' && prevCh <= '9') {
#ifndef QT_NO_REGEXP_BACKREF
            val = prevCh - '0';
            while (yyCh >= '0' && yyCh <= '9') {
                val = (val * 10) + (yyCh - '0');
                yyCh = getChar();
            }
            return Tok_BackRef | val;
#else
            error(RXERR_DISABLED);
#endif
        }
        return Tok_Char | prevCh;
    }
}

#ifndef QT_NO_REGEXP_INTERVAL
int QRegExpEngine::getRep(int def)
{
    if (yyCh >= '0' && yyCh <= '9') {
        int rep = 0;
        do {
            rep = 10 * rep + yyCh - '0';
            if (rep >= InftyRep) {
                error(RXERR_REPETITION);
                rep = def;
            }
            yyCh = getChar();
        } while (yyCh >= '0' && yyCh <= '9');
        return rep;
    } else {
        return def;
    }
}
#endif

#ifndef QT_NO_REGEXP_LOOKAHEAD
void QRegExpEngine::skipChars(int n)
{
    if (n > 0) {
        yyPos += n - 1;
        yyCh = getChar();
    }
}
#endif

void QRegExpEngine::error(const char *msg)
{
    if (yyError.isEmpty())
        yyError = QLatin1String(msg);
}

void QRegExpEngine::startTokenizer(const QChar *rx, int len)
{
    yyIn = rx;
    yyPos0 = 0;
    yyPos = 0;
    yyLen = len;
    yyCh = getChar();
    yyCharClass = new QRegExpCharClass;
    yyMinRep = 0;
    yyMaxRep = 0;
    yyError = QString();
}

int QRegExpEngine::getToken()
{
#ifndef QT_NO_REGEXP_CCLASS
    ushort pendingCh = 0;
    bool charPending;
    bool rangePending;
    int tok;
#endif
    int prevCh = yyCh;

    yyPos0 = yyPos - 1;
#ifndef QT_NO_REGEXP_CCLASS
    yyCharClass->clear();
#endif
    yyMinRep = 0;
    yyMaxRep = 0;
    yyCh = getChar();

    switch (prevCh) {
    case EOS:
        yyPos0 = yyPos;
        return Tok_Eos;
    case '$':
        return Tok_Dollar;
    case '(':
        if (yyCh == '?') {
            prevCh = getChar();
            yyCh = getChar();
            switch (prevCh) {
#ifndef QT_NO_REGEXP_LOOKAHEAD
            case '!':
                return Tok_NegLookahead;
            case '=':
                return Tok_PosLookahead;
#endif
            case ':':
                return Tok_MagicLeftParen;
            default:
                error(RXERR_LOOKAHEAD);
                return Tok_MagicLeftParen;
            }
        } else {
            return Tok_LeftParen;
        }
    case ')':
        return Tok_RightParen;
    case '*':
        yyMinRep = 0;
        yyMaxRep = InftyRep;
        return Tok_Quantifier;
    case '+':
        yyMinRep = 1;
        yyMaxRep = InftyRep;
        return Tok_Quantifier;
    case '.':
#ifndef QT_NO_REGEXP_CCLASS
        yyCharClass->setNegative(true);
#endif
        return Tok_CharClass;
    case '?':
        yyMinRep = 0;
        yyMaxRep = 1;
        return Tok_Quantifier;
    case '[':
#ifndef QT_NO_REGEXP_CCLASS
        if (yyCh == '^') {
            yyCharClass->setNegative(true);
            yyCh = getChar();
        }
        charPending = false;
        rangePending = false;
        do {
            if (yyCh == '-' && charPending && !rangePending) {
                rangePending = true;
                yyCh = getChar();
            } else {
                if (charPending && !rangePending) {
                    yyCharClass->addSingleton(pendingCh);
                    charPending = false;
                }
                if (yyCh == '\\') {
                    yyCh = getChar();
                    tok = getEscape();
                    if (tok == Tok_Word)
                        tok = '\b';
                } else {
                    tok = Tok_Char | yyCh;
                    yyCh = getChar();
                }
                if (tok == Tok_CharClass) {
                    if (rangePending) {
                        yyCharClass->addSingleton('-');
                        yyCharClass->addSingleton(pendingCh);
                        charPending = false;
                        rangePending = false;
                    }
                } else if ((tok & Tok_Char) != 0) {
                    if (rangePending) {
                        yyCharClass->addRange(pendingCh, tok ^ Tok_Char);
                        charPending = false;
                        rangePending = false;
                    } else {
                        pendingCh = tok ^ Tok_Char;
                        charPending = true;
                    }
                } else {
                    error(RXERR_CHARCLASS);
                }
            }
        }  while (yyCh != ']' && yyCh != EOS);
        if (rangePending)
            yyCharClass->addSingleton('-');
        if (charPending)
            yyCharClass->addSingleton(pendingCh);
        if (yyCh == EOS)
            error(RXERR_END);
        else
            yyCh = getChar();
        return Tok_CharClass;
#else
        error(RXERR_END);
        return Tok_Char | '[';
#endif
    case '\\':
        return getEscape();
    case ']':
        error(RXERR_LEFTDELIM);
        return Tok_Char | ']';
    case '^':
        return Tok_Caret;
    case '{':
#ifndef QT_NO_REGEXP_INTERVAL
        yyMinRep = getRep(0);
        yyMaxRep = yyMinRep;
        if (yyCh == ',') {
            yyCh = getChar();
            yyMaxRep = getRep(InftyRep);
        }
        if (yyMaxRep < yyMinRep)
            error(RXERR_INTERVAL);
        if (yyCh != '}')
            error(RXERR_REPETITION);
        yyCh = getChar();
        return Tok_Quantifier;
#else
        error(RXERR_DISABLED);
        return Tok_Char | '{';
#endif
    case '|':
        return Tok_Bar;
    case '}':
        error(RXERR_LEFTDELIM);
        return Tok_Char | '}';
    default:
        return Tok_Char | prevCh;
    }
}

int QRegExpEngine::parse(const QChar *pattern, int len)
{
    valid = true;
    startTokenizer(pattern, len);
    yyTok = getToken();
#ifndef QT_NO_REGEXP_CAPTURE
    yyMayCapture = true;
#else
    yyMayCapture = false;
#endif

#ifndef QT_NO_REGEXP_CAPTURE
    int atom = startAtom(false);
#endif
    QRegExpCharClass anything;
    Box box(this); // create InitialState
    box.set(anything);
    Box rightBox(this); // create FinalState
    rightBox.set(anything);

    Box middleBox(this);
    parseExpression(&middleBox);
#ifndef QT_NO_REGEXP_CAPTURE
    finishAtom(atom, false);
#endif
#ifndef QT_NO_REGEXP_OPTIM
    middleBox.setupHeuristics();
#endif
    box.cat(middleBox);
    box.cat(rightBox);
    delete yyCharClass;
    yyCharClass = 0;

#ifndef QT_NO_REGEXP_CAPTURE
    for (int i = 0; i < nf; ++i) {
        switch (f[i].capture) {
        case QRegExpAtom::NoCapture:
            break;
        case QRegExpAtom::OfficialCapture:
            f[i].capture = ncap;
            captureForOfficialCapture.append(ncap);
            ++ncap;
            ++officialncap;
            break;
        case QRegExpAtom::UnofficialCapture:
            f[i].capture = greedyQuantifiers ? ncap++ : QRegExpAtom::NoCapture;
        }
    }

#ifndef QT_NO_REGEXP_BACKREF
#ifndef QT_NO_REGEXP_OPTIM
    if (officialncap == 0 && nbrefs == 0) {
        ncap = nf = 0;
        f.clear();
    }
#endif
    // handle the case where there's a \5 with no corresponding capture
    // (captureForOfficialCapture.size() != officialncap)
    for (int i = 0; i < nbrefs - officialncap; ++i) {
        captureForOfficialCapture.append(ncap);
        ++ncap;
    }
#endif
#endif

    if (!yyError.isEmpty())
        return -1;

#ifndef QT_NO_REGEXP_OPTIM
    const QRegExpAutomatonState &sinit = s.at(InitialState);
    caretAnchored = !sinit.anchors.isEmpty();
    if (caretAnchored) {
        const QMap<int, int> &anchors = sinit.anchors;
        QMap<int, int>::const_iterator a;
        for (a = anchors.constBegin(); a != anchors.constEnd(); ++a) {
            if (
#ifndef QT_NO_REGEXP_ANCHOR_ALT
                (*a & Anchor_Alternation) != 0 ||
#endif
                (*a & Anchor_Caret) == 0)
            {
                caretAnchored = false;
                break;
            }
        }
    }
#endif

    // cleanup anchors
    int numStates = s.count();
    for (int i = 0; i < numStates; ++i) {
        QRegExpAutomatonState &state = s[i];
        if (!state.anchors.isEmpty()) {
            QMap<int, int>::iterator a = state.anchors.begin();
            while (a != state.anchors.end()) {
                if (a.value() == 0)
                    a = state.anchors.erase(a);
                else
                    ++a;
            }
        }
    }

    return yyPos0;
}

void QRegExpEngine::parseAtom(Box *box)
{
#ifndef QT_NO_REGEXP_LOOKAHEAD
    QRegExpEngine *eng = 0;
    bool neg;
    int len;
#endif

    if ((yyTok & Tok_Char) != 0) {
        box->set(QChar(yyTok ^ Tok_Char));
    } else {
#ifndef QT_NO_REGEXP_OPTIM
        trivial = false;
#endif
        switch (yyTok) {
        case Tok_Dollar:
            box->catAnchor(Anchor_Dollar);
            break;
        case Tok_Caret:
            box->catAnchor(Anchor_Caret);
            break;
#ifndef QT_NO_REGEXP_LOOKAHEAD
        case Tok_PosLookahead:
        case Tok_NegLookahead:
            neg = (yyTok == Tok_NegLookahead);
            eng = new QRegExpEngine(cs, greedyQuantifiers);
            len = eng->parse(yyIn + yyPos - 1, yyLen - yyPos + 1);
            if (len >= 0)
                skipChars(len);
            else
                error(RXERR_LOOKAHEAD);
            box->catAnchor(addLookahead(eng, neg));
            yyTok = getToken();
            if (yyTok != Tok_RightParen)
                error(RXERR_LOOKAHEAD);
            break;
#endif
#ifndef QT_NO_REGEXP_ESCAPE
        case Tok_Word:
            box->catAnchor(Anchor_Word);
            break;
        case Tok_NonWord:
            box->catAnchor(Anchor_NonWord);
            break;
#endif
        case Tok_LeftParen:
        case Tok_MagicLeftParen:
            yyTok = getToken();
            parseExpression(box);
            if (yyTok != Tok_RightParen)
                error(RXERR_END);
            break;
        case Tok_CharClass:
            box->set(*yyCharClass);
            break;
        case Tok_Quantifier:
            error(RXERR_REPETITION);
            break;
        default:
#ifndef QT_NO_REGEXP_BACKREF
            if ((yyTok & Tok_BackRef) != 0)
                box->set(yyTok ^ Tok_BackRef);
            else
#endif
                error(RXERR_DISABLED);
        }
    }
    yyTok = getToken();
}

void QRegExpEngine::parseFactor(Box *box)
{
#ifndef QT_NO_REGEXP_CAPTURE
    int outerAtom = greedyQuantifiers ? startAtom(false) : -1;
    int innerAtom = startAtom(yyMayCapture && yyTok == Tok_LeftParen);
    bool magicLeftParen = (yyTok == Tok_MagicLeftParen);
#else
    const int innerAtom = -1;
#endif

#ifndef QT_NO_REGEXP_INTERVAL
#define YYREDO() \
        yyIn = in, yyPos0 = pos0, yyPos = pos, yyLen = len, yyCh = ch, \
        *yyCharClass = charClass, yyMinRep = 0, yyMaxRep = 0, yyTok = tok

    const QChar *in = yyIn;
    int pos0 = yyPos0;
    int pos = yyPos;
    int len = yyLen;
    int ch = yyCh;
    QRegExpCharClass charClass;
    if (yyTok == Tok_CharClass)
        charClass = *yyCharClass;
    int tok = yyTok;
    bool mayCapture = yyMayCapture;
#endif

    parseAtom(box);
#ifndef QT_NO_REGEXP_CAPTURE
    finishAtom(innerAtom, magicLeftParen);
#endif

    bool hasQuantifier = (yyTok == Tok_Quantifier);
    if (hasQuantifier) {
#ifndef QT_NO_REGEXP_OPTIM
        trivial = false;
#endif
        if (yyMaxRep == InftyRep) {
            box->plus(innerAtom);
#ifndef QT_NO_REGEXP_INTERVAL
        } else if (yyMaxRep == 0) {
            box->clear();
#endif
        }
        if (yyMinRep == 0)
            box->opt();

#ifndef QT_NO_REGEXP_INTERVAL
        yyMayCapture = false;
        int alpha = (yyMinRep == 0) ? 0 : yyMinRep - 1;
        int beta = (yyMaxRep == InftyRep) ? 0 : yyMaxRep - (alpha + 1);

        Box rightBox(this);
        int i;

        for (i = 0; i < beta; i++) {
            YYREDO();
            Box leftBox(this);
            parseAtom(&leftBox);
            leftBox.cat(rightBox);
            leftBox.opt();
            rightBox = leftBox;
        }
        for (i = 0; i < alpha; i++) {
            YYREDO();
            Box leftBox(this);
            parseAtom(&leftBox);
            leftBox.cat(rightBox);
            rightBox = leftBox;
        }
        rightBox.cat(*box);
        *box = rightBox;
#endif
        yyTok = getToken();
#ifndef QT_NO_REGEXP_INTERVAL
        yyMayCapture = mayCapture;
#endif
    }
#undef YYREDO
#ifndef QT_NO_REGEXP_CAPTURE
    if (greedyQuantifiers)
        finishAtom(outerAtom, hasQuantifier);
#endif
}

void QRegExpEngine::parseTerm(Box *box)
{
#ifndef QT_NO_REGEXP_OPTIM
    if (yyTok != Tok_Eos && yyTok != Tok_RightParen && yyTok != Tok_Bar)
        parseFactor(box);
#endif
    while (yyTok != Tok_Eos && yyTok != Tok_RightParen && yyTok != Tok_Bar) {
        Box rightBox(this);
        parseFactor(&rightBox);
        box->cat(rightBox);
    }
}

void QRegExpEngine::parseExpression(Box *box)
{
    parseTerm(box);
    while (yyTok == Tok_Bar) {
#ifndef QT_NO_REGEXP_OPTIM
        trivial = false;
#endif
        Box rightBox(this);
        yyTok = getToken();
        parseTerm(&rightBox);
        box->orx(rightBox);
    }
}

/*
  The struct QRegExpPrivate contains the private data of a regular
  expression other than the automaton. It makes it possible for many
  QRegExp objects to use the same QRegExpEngine object with different
  QRegExpPrivate objects.
*/
struct QRegExpPrivate
{
    QRegExpEngine *eng;
    QRegExpEngineKey engineKey;
    bool minimal;
#ifndef QT_NO_REGEXP_CAPTURE
    QString t; // last string passed to QRegExp::indexIn() or lastIndexIn()
    QStringList capturedCache; // what QRegExp::capturedTexts() returned last
#endif
    QRegExpMatchState matchState;

    inline QRegExpPrivate()
        : eng(0), engineKey(QString(), QRegExp::RegExp, Qt::CaseSensitive), minimal(false) { }
    inline QRegExpPrivate(const QRegExpEngineKey &key)
        : eng(0), engineKey(key), minimal(false) {}
};

#if !defined(QT_NO_REGEXP_OPTIM)
uint qHash(const QRegExpEngineKey &key)
{
    return qHash(key.pattern);
}

typedef QCache<QRegExpEngineKey, QRegExpEngine> EngineCache;
Q_GLOBAL_STATIC(EngineCache, globalEngineCache)
Q_GLOBAL_STATIC(QMutex, mutex)
#endif // QT_NO_REGEXP_OPTIM

static void derefEngine(QRegExpEngine *eng, const QRegExpEngineKey &key)
{
    if (!eng->ref.deref()) {
#if !defined(QT_NO_REGEXP_OPTIM)
        if (globalEngineCache()) {
            QMutexLocker locker(mutex());
            globalEngineCache()->insert(key, eng, 4 + key.pattern.length() / 4);
        }
        else
            delete eng;
#else
        Q_UNUSED(key);
        delete eng;
#endif
    }
}

static void prepareEngine_helper(QRegExpPrivate *priv)
{
    bool initMatchState = !priv->eng;
#if !defined(QT_NO_REGEXP_OPTIM)
    if (!priv->eng && globalEngineCache()) {
        QMutexLocker locker(mutex());
        priv->eng = globalEngineCache()->take(priv->engineKey);
        if (priv->eng != 0)
            priv->eng->ref.ref();
    }
#endif // QT_NO_REGEXP_OPTIM

    if (!priv->eng)
        priv->eng = new QRegExpEngine(priv->engineKey);

    if (initMatchState)
        priv->matchState.prepareForMatch(priv->eng);
}

inline static void prepareEngine(QRegExpPrivate *priv)
{
    if (priv->eng)
        return;
    prepareEngine_helper(priv);
}

static void prepareEngineForMatch(QRegExpPrivate *priv, const QString &str)
{
    prepareEngine(priv);
    priv->matchState.prepareForMatch(priv->eng);
#ifndef QT_NO_REGEXP_CAPTURE
    priv->t = str;
    priv->capturedCache.clear();
#else
    Q_UNUSED(str);
#endif
}

static void invalidateEngine(QRegExpPrivate *priv)
{
    if (priv->eng != 0) {
        derefEngine(priv->eng, priv->engineKey);
        priv->eng = 0;
        priv->matchState.drain();
    }
}

/*!
    \enum QRegExp::CaretMode

    The CaretMode enum defines the different meanings of the caret
    (\bold{^}) in a regular expression. The possible values are:

    \value CaretAtZero
           The caret corresponds to index 0 in the searched string.

    \value CaretAtOffset
           The caret corresponds to the start offset of the search.

    \value CaretWontMatch
           The caret never matches.
*/

/*!
    \enum QRegExp::PatternSyntax

    The syntax used to interpret the meaning of the pattern.

    \value RegExp A rich Perl-like pattern matching syntax. This is
    the default.

    \value RegExp2 Like RegExp, but with \l{greedy quantifiers}. This
    will be the default in Qt 5. (Introduced in Qt 4.2.)

    \value Wildcard This provides a simple pattern matching syntax
    similar to that used by shells (command interpreters) for "file
    globbing". See \l{Wildcard Matching}.

    \value FixedString The pattern is a fixed string. This is
    equivalent to using the RegExp pattern on a string in
    which all metacharacters are escaped using escape().

    \value W3CXmlSchema11 The pattern is a regular expression as
    defined by the W3C XML Schema 1.1 specification.

    \sa setPatternSyntax()
*/

/*!
    Constructs an empty regexp.

    \sa isValid(), errorString()
*/
QRegExp::QRegExp()
{
    priv = new QRegExpPrivate;
}

/*!
    Constructs a regular expression object for the given \a pattern
    string. The pattern must be given using wildcard notation if \a
    syntax is \l Wildcard; the default is \l RegExp. The pattern is
    case sensitive, unless \a cs is Qt::CaseInsensitive. Matching is
    greedy (maximal), but can be changed by calling
    setMinimal().

    \sa setPattern(), setCaseSensitivity(), setPatternSyntax()
*/
QRegExp::QRegExp(const QString &pattern, Qt::CaseSensitivity cs, PatternSyntax syntax)
{
    priv = new QRegExpPrivate(QRegExpEngineKey(pattern, syntax, cs));
}

/*!
    Constructs a regular expression as a copy of \a rx.

    \sa operator=()
*/
QRegExp::QRegExp(const QRegExp &rx)
{
    priv = new QRegExpPrivate;
    operator=(rx);
}

/*!
    Destroys the regular expression and cleans up its internal data.
*/
QRegExp::~QRegExp()
{
    invalidateEngine(priv);
    delete priv;
}

/*!
    Copies the regular expression \a rx and returns a reference to the
    copy. The case sensitivity, wildcard, and minimal matching options
    are also copied.
*/
QRegExp &QRegExp::operator=(const QRegExp &rx)
{
    prepareEngine(rx.priv); // to allow sharing
    QRegExpEngine *otherEng = rx.priv->eng;
    if (otherEng)
        otherEng->ref.ref();
    invalidateEngine(priv);
    priv->eng = otherEng;
    priv->engineKey = rx.priv->engineKey;
    priv->minimal = rx.priv->minimal;
#ifndef QT_NO_REGEXP_CAPTURE
    priv->t = rx.priv->t;
    priv->capturedCache = rx.priv->capturedCache;
#endif
    if (priv->eng)
        priv->matchState.prepareForMatch(priv->eng);
    priv->matchState.captured = rx.priv->matchState.captured;
    return *this;
}

/*!
    Returns true if this regular expression is equal to \a rx;
    otherwise returns false.

    Two QRegExp objects are equal if they have the same pattern
    strings and the same settings for case sensitivity, wildcard and
    minimal matching.
*/
bool QRegExp::operator==(const QRegExp &rx) const
{
    return priv->engineKey == rx.priv->engineKey && priv->minimal == rx.priv->minimal;
}

/*!
    \fn bool QRegExp::operator!=(const QRegExp &rx) const

    Returns true if this regular expression is not equal to \a rx;
    otherwise returns false.

    \sa operator==()
*/

/*!
    Returns true if the pattern string is empty; otherwise returns
    false.

    If you call exactMatch() with an empty pattern on an empty string
    it will return true; otherwise it returns false since it operates
    over the whole string. If you call indexIn() with an empty pattern
    on \e any string it will return the start offset (0 by default)
    because the empty pattern matches the 'emptiness' at the start of
    the string. In this case the length of the match returned by
    matchedLength() will be 0.

    See QString::isEmpty().
*/

bool QRegExp::isEmpty() const
{
    return priv->engineKey.pattern.isEmpty();
}

/*!
    Returns true if the regular expression is valid; otherwise returns
    false. An invalid regular expression never matches.

    The pattern \bold{[a-z} is an example of an invalid pattern, since
    it lacks a closing square bracket.

    Note that the validity of a regexp may also depend on the setting
    of the wildcard flag, for example \bold{*.html} is a valid
    wildcard regexp but an invalid full regexp.

    \sa errorString()
*/
bool QRegExp::isValid() const
{
    if (priv->engineKey.pattern.isEmpty()) {
        return true;
    } else {
        prepareEngine(priv);
        return priv->eng->isValid();
    }
}

/*!
    Returns the pattern string of the regular expression. The pattern
    has either regular expression syntax or wildcard syntax, depending
    on patternSyntax().

    \sa patternSyntax(), caseSensitivity()
*/
QString QRegExp::pattern() const
{
    return priv->engineKey.pattern;
}

/*!
    Sets the pattern string to \a pattern. The case sensitivity,
    wildcard, and minimal matching options are not changed.

    \sa setPatternSyntax(), setCaseSensitivity()
*/
void QRegExp::setPattern(const QString &pattern)
{
    if (priv->engineKey.pattern != pattern) {
        invalidateEngine(priv);
        priv->engineKey.pattern = pattern;
    }
}

/*!
    Returns Qt::CaseSensitive if the regexp is matched case
    sensitively; otherwise returns Qt::CaseInsensitive.

    \sa patternSyntax(), pattern(), isMinimal()
*/
Qt::CaseSensitivity QRegExp::caseSensitivity() const
{
    return priv->engineKey.cs;
}

/*!
    Sets case sensitive matching to \a cs.

    If \a cs is Qt::CaseSensitive, \bold{\\.txt$} matches
    \c{readme.txt} but not \c{README.TXT}.

    \sa setPatternSyntax(), setPattern(), setMinimal()
*/
void QRegExp::setCaseSensitivity(Qt::CaseSensitivity cs)
{
    if ((bool)cs != (bool)priv->engineKey.cs) {
        invalidateEngine(priv);
        priv->engineKey.cs = cs;
    }
}

/*!
    Returns the syntax used by the regular expression. The default is
    QRegExp::RegExp.

    \sa pattern(), caseSensitivity()
*/
QRegExp::PatternSyntax QRegExp::patternSyntax() const
{
    return priv->engineKey.patternSyntax;
}

/*!
    Sets the syntax mode for the regular expression. The default is
    QRegExp::RegExp.

    Setting \a syntax to QRegExp::Wildcard enables simple shell-like
    \l{wildcard matching}. For example, \bold{r*.txt} matches the
    string \c{readme.txt} in wildcard mode, but does not match
    \c{readme}.

    Setting \a syntax to QRegExp::FixedString means that the pattern
    is interpreted as a plain string. Special characters (e.g.,
    backslash) don't need to be escaped then.

    \sa setPattern(), setCaseSensitivity(), escape()
*/
void QRegExp::setPatternSyntax(PatternSyntax syntax)
{
    if (syntax != priv->engineKey.patternSyntax) {
        invalidateEngine(priv);
        priv->engineKey.patternSyntax = syntax;
    }
}

/*!
    Returns true if minimal (non-greedy) matching is enabled;
    otherwise returns false.

    \sa caseSensitivity(), setMinimal()
*/
bool QRegExp::isMinimal() const
{
    return priv->minimal;
}

/*!
    Enables or disables minimal matching. If \a minimal is false,
    matching is greedy (maximal) which is the default.

    For example, suppose we have the input string "We must be
    <b>bold</b>, very <b>bold</b>!" and the pattern
    \bold{<b>.*</b>}. With the default greedy (maximal) matching,
    the match is "We must be \underline{<b>bold</b>, very
    <b>bold</b>}!". But with minimal (non-greedy) matching, the
    first match is: "We must be \underline{<b>bold</b>}, very
    <b>bold</b>!" and the second match is "We must be <b>bold</b>,
    very \underline{<b>bold</b>}!". In practice we might use the pattern
    \bold{<b>[^<]*\</b>} instead, although this will still fail for
    nested tags.

    \sa setCaseSensitivity()
*/
void QRegExp::setMinimal(bool minimal)
{
    priv->minimal = minimal;
}

// ### Qt 5: make non-const
/*!
    Returns true if \a str is matched exactly by this regular
    expression; otherwise returns false. You can determine how much of
    the string was matched by calling matchedLength().

    For a given regexp string R, exactMatch("R") is the equivalent of
    indexIn("^R$") since exactMatch() effectively encloses the regexp
    in the start of string and end of string anchors, except that it
    sets matchedLength() differently.

    For example, if the regular expression is \bold{blue}, then
    exactMatch() returns true only for input \c blue. For inputs \c
    bluebell, \c blutak and \c lightblue, exactMatch() returns false
    and matchedLength() will return 4, 3 and 0 respectively.

    Although const, this function sets matchedLength(),
    capturedTexts(), and pos().

    \sa indexIn(), lastIndexIn()
*/
bool QRegExp::exactMatch(const QString &str) const
{
    prepareEngineForMatch(priv, str);
    priv->matchState.match(str.unicode(), str.length(), 0, priv->minimal, true, 0);
    if (priv->matchState.captured[1] == str.length()) {
        return true;
    } else {
        priv->matchState.captured[0] = 0;
        priv->matchState.captured[1] = priv->matchState.oneTestMatchedLen;
        return false;
    }
}

// ### Qt 5: make non-const
/*!
    Attempts to find a match in \a str from position \a offset (0 by
    default). If \a offset is -1, the search starts at the last
    character; if -2, at the next to last character; etc.

    Returns the position of the first match, or -1 if there was no
    match.

    The \a caretMode parameter can be used to instruct whether \bold{^}
    should match at index 0 or at \a offset.

    You might prefer to use QString::indexOf(), QString::contains(),
    or even QStringList::filter(). To replace matches use
    QString::replace().

    Example:
    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 13

    Although const, this function sets matchedLength(),
    capturedTexts() and pos().

    If the QRegExp is a wildcard expression (see setPatternSyntax())
    and want to test a string against the whole wildcard expression,
    use exactMatch() instead of this function.

    \sa lastIndexIn(), exactMatch()
*/

int QRegExp::indexIn(const QString &str, int offset, CaretMode caretMode) const
{
    prepareEngineForMatch(priv, str);
    if (offset < 0)
        offset += str.length();
    priv->matchState.match(str.unicode(), str.length(), offset,
        priv->minimal, false, caretIndex(offset, caretMode));
    return priv->matchState.captured[0];
}

// ### Qt 5: make non-const
/*!
    Attempts to find a match backwards in \a str from position \a
    offset. If \a offset is -1 (the default), the search starts at the
    last character; if -2, at the next to last character; etc.

    Returns the position of the first match, or -1 if there was no
    match.

    The \a caretMode parameter can be used to instruct whether \bold{^}
    should match at index 0 or at \a offset.

    Although const, this function sets matchedLength(),
    capturedTexts() and pos().

    \warning Searching backwards is much slower than searching
    forwards.

    \sa indexIn(), exactMatch()
*/

int QRegExp::lastIndexIn(const QString &str, int offset, CaretMode caretMode) const
{
    prepareEngineForMatch(priv, str);
    if (offset < 0)
        offset += str.length();
    if (offset < 0 || offset > str.length()) {
        memset(priv->matchState.captured, -1, priv->matchState.capturedSize*sizeof(int));
        return -1;
    }

    while (offset >= 0) {
        priv->matchState.match(str.unicode(), str.length(), offset,
            priv->minimal, true, caretIndex(offset, caretMode));
        if (priv->matchState.captured[0] == offset)
            return offset;
        --offset;
    }
    return -1;
}

/*!
    Returns the length of the last matched string, or -1 if there was
    no match.

    \sa exactMatch(), indexIn(), lastIndexIn()
*/
int QRegExp::matchedLength() const
{
    return priv->matchState.captured[1];
}

#ifndef QT_NO_REGEXP_CAPTURE
/*!
  Returns the number of captures contained in the regular expression.
 */
int QRegExp::numCaptures() const
{
    prepareEngine(priv);
    return priv->eng->numCaptures();
}

/*!
    Returns a list of the captured text strings.

    The first string in the list is the entire matched string. Each
    subsequent list element contains a string that matched a
    (capturing) subexpression of the regexp.

    For example:
    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 14

    The above example also captures elements that may be present but
    which we have no interest in. This problem can be solved by using
    non-capturing parentheses:

    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 15

    Note that if you want to iterate over the list, you should iterate
    over a copy, e.g.
    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 16

    Some regexps can match an indeterminate number of times. For
    example if the input string is "Offsets: 12 14 99 231 7" and the
    regexp, \c{rx}, is \bold{(\\d+)+}, we would hope to get a list of
    all the numbers matched. However, after calling
    \c{rx.indexIn(str)}, capturedTexts() will return the list ("12",
    "12"), i.e. the entire match was "12" and the first subexpression
    matched was "12". The correct approach is to use cap() in a
    \l{QRegExp#cap_in_a_loop}{loop}.

    The order of elements in the string list is as follows. The first
    element is the entire matching string. Each subsequent element
    corresponds to the next capturing open left parentheses. Thus
    capturedTexts()[1] is the text of the first capturing parentheses,
    capturedTexts()[2] is the text of the second and so on
    (corresponding to $1, $2, etc., in some other regexp languages).

    \sa cap(), pos()
*/
QStringList QRegExp::capturedTexts() const
{
    if (priv->capturedCache.isEmpty()) {
        prepareEngine(priv);
        const int *captured = priv->matchState.captured;
        int n = priv->matchState.capturedSize;

        for (int i = 0; i < n; i += 2) {
            QString m;
            if (captured[i + 1] == 0)
                m = QLatin1String(""); // ### Qt 5: don't distinguish between null and empty
            else if (captured[i] >= 0)
                m = priv->t.mid(captured[i], captured[i + 1]);
            priv->capturedCache.append(m);
        }
        priv->t.clear();
    }
    return priv->capturedCache;
}

/*!
    \internal
*/
QStringList QRegExp::capturedTexts()
{
    return const_cast<const QRegExp *>(this)->capturedTexts();
}

/*!
    Returns the text captured by the \a nth subexpression. The entire
    match has index 0 and the parenthesized subexpressions have
    indexes starting from 1 (excluding non-capturing parentheses).

    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 17

    The order of elements matched by cap() is as follows. The first
    element, cap(0), is the entire matching string. Each subsequent
    element corresponds to the next capturing open left parentheses.
    Thus cap(1) is the text of the first capturing parentheses, cap(2)
    is the text of the second, and so on.

    \sa capturedTexts(), pos()
*/
QString QRegExp::cap(int nth) const
{
    return capturedTexts().value(nth);
}

/*!
    \internal
*/
QString QRegExp::cap(int nth)
{
    return const_cast<const QRegExp *>(this)->cap(nth);
}

/*!
    Returns the position of the \a nth captured text in the searched
    string. If \a nth is 0 (the default), pos() returns the position
    of the whole match.

    Example:
    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 18

    For zero-length matches, pos() always returns -1. (For example, if
    cap(4) would return an empty string, pos(4) returns -1.) This is
    a feature of the implementation.

    \sa cap(), capturedTexts()
*/
int QRegExp::pos(int nth) const
{
    if (nth < 0 || nth >= priv->matchState.capturedSize / 2)
        return -1;
    else
        return priv->matchState.captured[2 * nth];
}

/*!
    \internal
*/
int QRegExp::pos(int nth)
{
    return const_cast<const QRegExp *>(this)->pos(nth);
}

/*!
  Returns a text string that explains why a regexp pattern is
  invalid the case being; otherwise returns "no error occurred".

  \sa isValid()
*/
QString QRegExp::errorString() const
{
    if (isValid()) {
        return QString::fromLatin1(RXERR_OK);
    } else {
        return priv->eng->errorString();
    }
}

/*!
    \internal
*/
QString QRegExp::errorString()
{
    return const_cast<const QRegExp *>(this)->errorString();
}
#endif

/*!
    Returns the string \a str with every regexp special character
    escaped with a backslash. The special characters are $, (,), *, +,
    ., ?, [, \,], ^, {, | and }.

    Example:

    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 19

    This function is useful to construct regexp patterns dynamically:

    \snippet doc/src/snippets/code/src_corelib_tools_qregexp.cpp 20

    \sa setPatternSyntax()
*/
QString QRegExp::escape(const QString &str)
{
    QString quoted;
    const int count = str.count();
    quoted.reserve(count * 2);
    const QLatin1Char backslash('\\');
    for (int i = 0; i < count; i++) {
        switch (str.at(i).toLatin1()) {
        case '$':
        case '(':
        case ')':
        case '*':
        case '+':
        case '.':
        case '?':
        case '[':
        case '\\':
        case ']':
        case '^':
        case '{':
        case '|':
        case '}':
            quoted.append(backslash);
        }
        quoted.append(str.at(i));
    }
    return quoted;
}

/*!
    \fn bool QRegExp::caseSensitive() const

    Use \l caseSensitivity() instead.
*/

/*!
    \fn void QRegExp::setCaseSensitive(bool sensitive)

    Use \l setCaseSensitivity() instead.
*/

/*!
    \fn bool QRegExp::wildcard() const

    Use \l patternSyntax() instead.

    \oldcode
        bool wc = rx.wildcard();
    \newcode
        bool wc = (rx.patternSyntax() == QRegExp::Wildcard);
    \endcode
*/

/*!
    \fn void QRegExp::setWildcard(bool wildcard)

    Use \l setPatternSyntax() instead.

    \oldcode
        rx.setWildcard(wc);
    \newcode
        rx.setPatternSyntax(wc ? QRegExp::Wildcard : QRegExp::RegExp);
    \endcode
*/

/*!
    \fn bool QRegExp::minimal() const

    Use \l isMinimal() instead.
*/

/*!
    \fn int QRegExp::search(const QString &str, int from = 0,
                            CaretMode caretMode = CaretAtZero) const

    Use \l indexIn() instead.
*/

/*!
    \fn int QRegExp::searchRev(const QString &str, int from = -1, \
                               CaretMode caretMode = CaretAtZero) const

    Use \l lastIndexIn() instead.
*/

/*!
    \fn QRegExp::QRegExp(const QString &pattern, bool cs, bool wildcard = false)

    Use another constructor instead.

    \oldcode
        QRegExp rx("*.txt", false, true);
    \newcode
        QRegExp rx("*.txt", Qt::CaseInsensitive, QRegExp::Wildcard);
    \endcode
*/

#ifndef QT_NO_DATASTREAM
/*!
    \relates QRegExp

    Writes the regular expression \a regExp to stream \a out.

    \sa {Format of the QDataStream Operators}
*/
QDataStream &operator<<(QDataStream &out, const QRegExp &regExp)
{
    return out << regExp.pattern() << (quint8)regExp.caseSensitivity()
               << (quint8)regExp.patternSyntax()
               << (quint8)!!regExp.isMinimal();
}

/*!
    \relates QRegExp

    Reads a regular expression from stream \a in into \a regExp.

    \sa {Format of the QDataStream Operators}
*/
QDataStream &operator>>(QDataStream &in, QRegExp &regExp)
{
    QString pattern;
    quint8 cs;
    quint8 patternSyntax;
    quint8 isMinimal;

    in >> pattern >> cs >> patternSyntax >> isMinimal;

    QRegExp newRegExp(pattern, Qt::CaseSensitivity(cs),
                      QRegExp::PatternSyntax(patternSyntax));

    newRegExp.setMinimal(isMinimal);
    regExp = newRegExp;
    return in;
}
#endif // QT_NO_DATASTREAM

QT_END_NAMESPACE