1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298 3299 3300 3301 3302 3303 3304 3305 3306 3307 3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319 3320 3321 3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333 3334 3335 3336 3337 3338 3339 3340 3341 3342 3343 3344 3345 3346 3347 3348 3349 3350 3351 3352 3353 3354 3355 3356 3357 3358 3359 3360 3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397 3398 3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467 3468 3469 3470 3471 3472 3473 3474 3475 3476 3477 3478 3479 3480 3481 3482 3483 3484 3485 3486 3487 3488 3489 3490 3491 3492 3493 3494 3495 3496 3497 3498 3499 3500 3501 3502 3503 3504 3505 3506 3507 3508 3509 3510 3511 3512 3513 3514 3515 3516 3517 3518 3519 3520 3521 3522 3523 3524 3525 3526 3527 3528 3529 3530 3531 3532 3533 3534 3535 3536 3537 3538 3539 3540 3541 3542 3543 3544 3545 3546 3547 3548 3549 3550 3551 3552 3553 3554 3555 3556 3557 3558 3559 3560 3561 3562 3563 3564 3565 3566 3567 3568 3569 3570 3571 3572 3573 3574 3575 3576 3577 3578 3579 3580 3581 3582 3583 3584 3585 3586 3587 3588 3589 3590 3591 3592 3593 3594 3595 3596 3597 3598 3599 3600 3601 3602 3603 3604 3605 3606 3607 3608 3609 3610 3611 3612 3613 3614 3615 3616 3617 3618 3619 3620 3621 3622 3623 3624 3625 3626 3627 3628 3629 3630 3631 3632 3633 3634 3635 3636 3637 3638 3639 3640 3641 3642 3643 3644 3645 3646 3647 3648 3649 3650 3651 3652 3653 3654 3655 3656 3657 3658 3659 3660 3661 3662 3663 3664 3665 3666 3667 3668 3669 3670 3671 3672 3673 3674 3675 3676 3677 3678 3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690 3691 3692 3693 3694 3695 3696 3697 3698 3699 3700 3701 3702 3703 3704 3705 3706 3707 3708 3709 3710 3711 3712 3713 3714 3715 3716 3717 3718 3719 3720 3721 3722 3723 3724 3725 3726 3727 3728 3729 3730 3731 3732 3733 3734 3735 3736 3737 3738 3739 3740 3741 3742 3743 3744 3745 3746 3747 3748 3749 3750 3751 3752 3753 3754 3755 3756 3757 3758 3759 3760 3761 3762 3763 3764 3765 3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799 3800 3801 3802 3803 3804 3805 3806 3807 3808 3809 3810 3811 3812 3813 3814 3815 3816 3817 3818 3819 3820 3821 3822 3823 3824 3825 3826 3827 3828 3829 3830 3831 3832 3833 3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854 3855 3856 3857 3858 3859 3860 3861 3862 3863 3864 3865 3866 3867 3868 3869 3870 3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885 3886 3887 3888 3889 3890 3891 3892 3893 3894 3895 3896 3897 3898 3899 3900 3901 3902 3903 3904 3905 3906 3907 3908 3909 3910 3911 3912 3913 3914 3915 3916 3917 3918 3919 3920 3921 3922 3923 3924 3925 3926 3927 3928 3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947 3948 3949 3950 3951 3952 3953 3954 3955 3956 3957 3958 3959 3960 3961 3962 3963 3964 3965 3966 3967 3968 3969 3970 3971 3972 3973 3974 3975 3976 3977 3978 3979 3980 3981 3982 3983 3984 3985 3986 3987 3988 3989 3990 3991 3992 3993 3994 3995 3996 3997 3998 3999 4000 4001 4002 4003 4004 4005 4006 4007 4008 4009 4010 4011 4012 4013 4014 4015 4016 4017 4018 4019 4020 4021 4022 4023 4024 4025 4026 4027 4028 4029 4030 4031 4032 4033 4034 4035 4036 4037 4038 4039 4040 4041 4042 4043 4044 4045 4046 4047 4048 4049 4050 4051 4052 4053 4054 4055 4056 4057 4058 4059 4060 4061 4062 4063 4064 4065 4066 4067 4068 4069 4070 4071 4072 4073 4074 4075 4076 4077 4078 4079 4080 4081 4082 4083 4084 4085 4086 4087 4088 4089 4090 4091 4092 4093 4094 4095 4096 4097 4098 4099 4100 4101 4102 4103 4104 4105 4106 4107 4108 4109 4110 4111 4112 4113 4114 4115 4116 4117 4118 4119 4120 4121 4122 4123 4124 4125 4126 4127 4128 4129 4130 4131 4132 4133 4134 4135 4136 4137 4138 4139 4140 4141 4142 4143 4144 4145 4146 4147 4148 4149 4150 4151 4152 4153 4154 4155 4156 4157 4158 4159 4160 4161 4162 4163 4164 4165 4166 4167 4168 4169 4170 4171 4172 4173 4174 4175 4176 4177 4178 4179 4180 4181 4182 4183 4184 4185 4186 4187 4188 4189 4190 4191 4192 4193 4194 4195 4196 4197 4198 4199 4200 4201 4202 4203 4204 4205 4206 4207 4208 4209 4210 4211 4212 4213 4214 4215 4216 4217 4218 4219 4220 4221 4222 4223 4224 4225 4226 4227 4228 4229 4230 4231 4232 4233 4234 4235 4236 4237 4238 4239 4240 4241 4242 4243 4244 4245 4246 4247 4248 4249 4250 4251 4252 4253 4254 4255 4256 4257 4258 4259 4260 4261 4262 4263 4264 4265 4266 4267 4268 4269 4270 4271 4272 4273 4274 4275 4276 4277 4278 4279 4280 4281 4282 4283 4284 4285 4286 4287 4288 4289 4290 4291 4292 4293 4294 4295 4296 4297 4298 4299 4300 4301 4302 4303 4304 4305 4306 4307 4308 4309 4310 4311 4312 4313 4314 4315 4316 4317 4318 4319 4320 4321 4322 4323 4324 4325 4326 4327 4328 4329 4330 4331 4332 4333 4334 4335 4336 4337 4338 4339 4340 4341 4342 4343 4344 4345 4346 4347 4348 4349 4350 4351 4352 4353 4354 4355 4356 4357 4358 4359 4360 4361 4362 4363 4364 4365 4366 4367 4368 4369 4370 4371 4372 4373 4374 4375 4376 4377 4378 4379 4380 4381 4382 4383 4384 4385 4386 4387 4388 4389 4390 4391 4392 4393 4394 4395 4396 4397 4398 4399 4400 4401 4402 4403 4404 4405 4406 4407 4408 4409 4410 4411 4412 4413 4414 4415 4416 4417 4418 4419 4420 4421 4422 4423 4424 4425 4426 4427 4428 4429 4430 4431 4432 4433 4434 4435 4436 4437 4438 4439 4440 4441 4442 4443 4444 4445 4446 4447 4448 4449 4450 4451 4452 4453 4454 4455 4456 4457 4458 4459 4460 4461 4462 4463 4464 4465 4466 4467 4468 4469 4470 4471 4472 4473 4474 4475 4476 4477 4478 4479 4480 4481 4482 4483 4484 4485 4486 4487 4488 4489 4490 4491 4492 4493 4494 4495 4496 4497 4498 4499 4500 4501 4502 4503 4504 4505 4506 4507 4508 4509 4510 4511 4512 4513 4514 4515 4516 4517 4518 4519 4520 4521 4522 4523 4524 4525 4526 4527 4528 4529 4530 4531 4532 4533 4534 4535 4536 4537 4538 4539 4540 4541 4542 4543 4544 4545 4546 4547 4548 4549 4550 4551 4552 4553 4554 4555 4556 4557 4558 4559 4560 4561 4562 4563 4564 4565 4566 4567 4568 4569 4570 4571 4572 4573 4574 4575 4576 4577 4578 4579 4580 4581 4582 4583 4584 4585 4586 4587 4588 4589 4590 4591 4592 4593 4594 4595 4596 4597 4598 4599 4600 4601 4602 4603 4604 4605 4606 4607 4608 4609 4610 4611 4612 4613 4614 4615 4616 4617 4618 4619 4620 4621 4622 4623 4624 4625 4626 4627 4628 4629 4630 4631 4632 4633 4634 4635 4636 4637 4638 4639 4640 4641 4642 4643 4644 4645 4646 4647 4648 4649 4650 4651 4652 4653 4654 4655 4656 4657 4658 4659 4660 4661 4662 4663 4664 4665 4666 4667 4668 4669 4670 4671 4672 4673 4674 4675 4676 4677 4678 4679 4680 4681 4682 4683 4684 4685 4686 4687 4688 4689 4690 4691 4692 4693 4694 4695 4696 4697 4698 4699 4700 4701 4702 4703 4704 4705 4706 4707 4708 4709 4710 4711 4712 4713 4714 4715 4716 4717 4718 4719 4720 4721 4722 4723 4724 4725 4726 4727 4728 4729 4730 4731 4732 4733 4734 4735 4736 4737 4738 4739 4740 4741 4742 4743 4744 4745 4746 4747 4748 4749 4750 4751 4752 4753 4754 4755 4756 4757 4758 4759 4760 4761 4762 4763 4764 4765 4766 4767 4768 4769 4770 4771 4772 4773 4774 4775 4776 4777 4778 4779 4780 4781 4782 4783 4784 4785 4786 4787 4788 4789 4790 4791 4792 4793 4794 4795 4796 4797 4798 4799 4800 4801 4802 4803 4804 4805 4806 4807 4808 4809 4810 4811 4812 4813 4814 4815 4816 4817 4818 4819 4820 4821 4822 4823 4824 4825 4826 4827 4828 4829 4830 4831 4832 4833 4834 4835 4836 4837 4838 4839 4840 4841 4842 4843 4844 4845 4846 4847 4848 4849 4850 4851 4852 4853 4854 4855 4856 4857 4858 4859 4860 4861 4862 4863 4864 4865 4866 4867 4868 4869 4870 4871 4872 4873 4874 4875 4876 4877 4878 4879 4880 4881 4882 4883 4884 4885 4886 4887 4888 4889 4890 4891 4892 4893 4894 4895 4896 4897 4898 4899 4900 4901 4902 4903 4904 4905 4906 4907 4908 4909 4910 4911 4912 4913 4914 4915 4916 4917 4918 4919 4920 4921 4922 4923 4924 4925 4926 4927 4928 4929 4930 4931 4932 4933 4934 4935 4936 4937 4938 4939 4940 4941 4942 4943 4944 4945 4946 4947 4948 4949 4950 4951 4952 4953 4954 4955 4956 4957 4958 4959 4960 4961 4962 4963 4964 4965 4966 4967 4968 4969 4970 4971 4972 4973 4974 4975 4976 4977 4978 4979 4980 4981 4982 4983 4984 4985 4986 4987 4988 4989 4990 4991 4992 4993 4994 4995 4996 4997 4998 4999 5000 5001 5002 5003 5004 5005 5006 5007 5008 5009 5010 5011 5012 5013 5014 5015 5016 5017 5018 5019 5020 5021 5022 5023 5024 5025 5026 5027 5028 5029 5030 5031 5032 5033 5034 5035 5036 5037 5038 5039 5040 5041 5042 5043 5044 5045 5046 5047 5048 5049 5050 5051 5052 5053 5054 5055 5056 5057 5058 5059 5060 5061 5062 5063 5064 5065 5066 5067 5068 5069 5070 5071 5072 5073 5074 5075 5076 5077 5078 5079 5080 5081 5082 5083 5084 5085 5086 5087 5088 5089 5090 5091 5092 5093 5094 5095 5096 5097 5098 5099 5100 5101 5102 5103 5104 5105 5106 5107 5108 5109 5110 5111 5112 5113 5114 5115 5116 5117 5118 5119 5120 5121 5122 5123 5124 5125 5126 5127 5128 5129 5130 5131 5132 5133 5134 5135 5136 5137 5138 5139 5140 5141 5142 5143 5144 5145 5146 5147 5148 5149 5150 5151 5152 5153 5154 5155 5156 5157 5158 5159 5160 5161 5162 5163 5164 5165 5166 5167 5168 5169 5170 5171 5172 5173 5174 5175 5176 5177 5178 5179 5180 5181 5182 5183 5184 5185 5186 5187 5188 5189 5190 5191 5192 5193 5194 5195 5196 5197 5198 5199 5200 5201 5202 5203 5204 5205 5206 5207 5208 5209 5210 5211 5212 5213 5214 5215 5216 5217 5218 5219 5220 5221 5222 5223 5224 5225 5226 5227 5228 5229 5230 5231 5232 5233 5234 5235 5236 5237 5238 5239 5240 5241 5242 5243 5244 5245 5246 5247 5248 5249 5250 5251 5252 5253 5254 5255 5256 5257 5258 5259 5260 5261 5262 5263 5264 5265 5266 5267 5268 5269 5270 5271 5272 5273 5274 5275 5276 5277 5278 5279 5280 5281 5282 5283 5284 5285 5286 5287 5288 5289 5290 5291 5292 5293 5294 5295 5296 5297 5298 5299 5300 5301 5302 5303 5304 5305 5306 5307 5308 5309 5310 5311 5312 5313 5314 5315 5316 5317 5318 5319 5320 5321 5322 5323 5324 5325 5326 5327 5328 5329 5330 5331 5332 5333 5334 5335 5336 5337 5338 5339 5340 5341 5342 5343 5344 5345 5346 5347 5348 5349 5350 5351 5352 5353 5354 5355 5356 5357 5358 5359 5360 5361 5362 5363 5364 5365 5366 5367 5368 5369 5370 5371 5372 5373 5374 5375 5376 5377 5378 5379 5380 5381 5382 5383 5384 5385 5386 5387 5388 5389 5390 5391 5392 5393 5394 5395 5396 5397 5398 5399 5400 5401 5402 5403 5404 5405 5406 5407 5408 5409 5410 5411 5412 5413 5414 5415 5416 5417 5418 5419 5420 5421 5422 5423 5424 5425 5426 5427 5428 5429 5430 5431 5432 5433 5434 5435 5436 5437 5438 5439 5440 5441 5442 5443 5444 5445 5446 5447 5448 5449 5450 5451 5452 5453 5454 5455 5456 5457 5458 5459 5460 5461 5462 5463 5464 5465 5466 5467 5468 5469 5470 5471 5472 5473 5474 5475 5476 5477 5478 5479 5480 5481 5482 5483 5484 5485 5486 5487 5488 5489 5490 5491 5492 5493 5494 5495 5496 5497 5498 5499 5500 5501 5502 5503 5504 5505 5506 5507 5508 5509 5510 5511 5512 5513 5514 5515 5516 5517 5518 5519 5520 5521 5522 5523 5524 5525 5526 5527 5528 5529 5530 5531 5532 5533 5534 5535 5536 5537 5538 5539 5540 5541 5542 5543 5544 5545 5546 5547 5548 5549 5550 5551 5552 5553 5554 5555 5556 5557 5558 5559 5560 5561 5562 5563 5564 5565 5566 5567 5568 5569 5570 5571 5572 5573 5574 5575 5576 5577 5578 5579 5580 5581 5582 5583 5584 5585 5586 5587 5588 5589 5590 5591 5592 5593 5594 5595 5596 5597 5598 5599 5600 5601 5602 5603 5604 5605 5606 5607 5608 5609 5610 5611 5612 5613 5614 5615 5616 5617 5618 5619 5620 5621 5622 5623 5624 5625 5626 5627 5628 5629 5630 5631 5632 5633 5634 5635 5636 5637 5638 5639 5640 5641 5642 5643 5644 5645 5646 5647 5648 5649 5650 5651 5652 5653 5654 5655 5656 5657 5658 5659 5660 5661 5662 5663 5664 5665 5666 5667 5668 5669 5670 5671 5672 5673 5674 5675 5676 5677 5678 5679 5680 5681 5682 5683 5684 5685 5686 5687 5688 5689 5690 5691 5692 5693 5694 5695 5696 5697 5698 5699 5700 5701 5702 5703 5704 5705 5706 5707 5708 5709 5710 5711 5712 5713 5714 5715 5716 5717 5718 5719 5720 5721 5722 5723 5724 5725 5726 5727 5728 5729 5730 5731 5732 5733 5734 5735 5736 5737 5738 5739 5740 5741 5742 5743 5744 5745 5746 5747 5748 5749 5750 5751 5752 5753 5754 5755 5756 5757 5758 5759 5760 5761 5762 5763 5764 5765 5766 5767 5768 5769 5770 5771 5772 5773 5774 5775 5776 5777 5778 5779 5780 5781 5782 5783 5784 5785 5786 5787 5788 5789 5790 5791 5792 5793 5794 5795 5796 5797 5798 5799 5800 5801 5802 5803 5804 5805 5806 5807 5808 5809 5810 5811 5812 5813 5814 5815 5816 5817 5818 5819 5820 5821 5822 5823 5824 5825 5826 5827 5828 5829 5830 5831 5832 5833 5834 5835 5836 5837 5838 5839 5840 5841 5842 5843 5844 5845 5846 5847 5848 5849 5850 5851 5852 5853 5854 5855 5856 5857 5858 5859 5860 5861 5862 5863 5864 5865 5866 5867 5868 5869 5870 5871 5872 5873 5874 5875 5876 5877 5878 5879 5880 5881 5882 5883 5884 5885 5886 5887 5888 5889 5890 5891 5892 5893 5894 5895 5896 5897 5898 5899 5900 5901 5902 5903 5904 5905 5906 5907 5908 5909 5910 5911 5912 5913 5914 5915 5916 5917 5918 5919 5920 5921 5922 5923 5924 5925 5926 5927 5928 5929 5930 5931 5932 5933 5934 5935 5936 5937 5938 5939 5940 5941 5942 5943 5944 5945 5946 5947 5948 5949 5950 5951 5952 5953 5954 5955 5956 5957 5958 5959 5960 5961 5962 5963 5964 5965 5966 5967 5968 5969 5970 5971 5972 5973 5974 5975 5976 5977 5978 5979 5980 5981 5982 5983 5984 5985 5986 5987 5988 5989 5990 5991 5992 5993 5994 5995 5996 5997 5998 5999 6000 6001 6002 6003 6004 6005 6006 6007 6008 6009 6010 6011 6012 6013 6014 6015 6016 6017 6018 6019 6020 6021 6022 6023 6024 6025 6026 6027 6028 6029 6030 6031 6032 6033 6034 6035 6036 6037 6038 6039 6040 6041 6042 6043 6044 6045 6046 6047 6048 6049 6050 6051 6052 6053 6054 6055 6056 6057 6058 6059 6060 6061 6062 6063 6064 6065 6066 6067 6068 6069 6070 6071 6072 6073 6074 6075 6076 6077 6078 6079 6080 6081 6082 6083 6084 6085 6086 6087 6088 6089 6090 6091 6092 6093 6094 6095 6096 6097 6098 6099 6100 6101 6102 6103 6104 6105 6106 6107 6108 6109 6110 6111 6112 6113 6114 6115 6116 6117 6118 6119 6120 6121 6122 6123 6124 6125 6126 6127 6128 6129 6130 6131 6132 6133 6134 6135 6136 6137 6138 6139 6140 6141 6142 6143 6144 6145 6146 6147 6148 6149 6150 6151 6152 6153 6154 6155 6156 6157 6158 6159 6160 6161 6162 6163 6164 6165 6166 6167 6168 6169 6170 6171 6172 6173 6174 6175 6176 6177 6178 6179 6180 6181 6182 6183 6184 6185 6186 6187 6188 6189 6190 6191 6192 6193 6194 6195 6196 6197 6198 6199 6200 6201 6202 6203 6204 6205 6206 6207 6208 6209 6210 6211 6212 6213 6214 6215 6216 6217 6218 6219 6220 6221 6222 6223 6224 6225 6226 6227 6228 6229 6230 6231 6232 6233 6234 6235 6236 6237 6238 6239 6240 6241 6242 6243 6244 6245 6246 6247 6248 6249 6250 6251 6252 6253 6254 6255 6256 6257 6258 6259 6260 6261 6262 6263 6264 6265 6266 6267 6268 6269 6270 6271 6272 6273 6274 6275 6276 6277 6278 6279 6280 6281 6282 6283 6284 6285 6286 6287 6288 6289 6290 6291 6292 6293 6294 6295 6296 6297 6298 6299 6300 6301 6302 6303 6304 6305 6306 6307 6308 6309 6310 6311 6312 6313 6314 6315 6316 6317 6318 6319 6320 6321 6322 6323 6324 6325 6326 6327 6328 6329 6330 6331 6332 6333 6334 6335 6336 6337 6338 6339 6340 6341 6342 6343 6344 6345 6346 6347 6348 6349 6350 6351 6352 6353 6354 6355 6356 6357 6358 6359 6360 6361 6362 6363 6364 6365 6366 6367 6368 6369 6370 6371 6372 6373 6374 6375 6376 6377 6378 6379 6380 6381 6382 6383 6384 6385 6386 6387 6388 6389 6390 6391 6392 6393 6394 6395 6396 6397 6398 6399 6400 6401 6402 6403 6404 6405 6406 6407 6408 6409 6410 6411 6412 6413 6414 6415 6416 6417 6418 6419 6420 6421 6422 6423 6424 6425 6426 6427 6428 6429 6430 6431 6432 6433 6434 6435 6436 6437 6438 6439 6440 6441 6442 6443 6444 6445 6446 6447 6448 6449 6450 6451 6452 6453 6454 6455 6456 6457 6458 6459 6460 6461 6462 6463 6464 6465 6466 6467 6468 6469 6470 6471 6472 6473 6474 6475 6476 6477 6478 6479 6480 6481 6482 6483 6484 6485 6486 6487 6488 6489 6490 6491 6492 6493 6494 6495 6496 6497 6498 6499 6500 6501 6502 6503 6504 6505 6506 6507 6508 6509 6510 6511 6512 6513 6514 6515 6516 6517 6518 6519 6520 6521 6522 6523 6524 6525 6526 6527 6528 6529 6530 6531 6532 6533 6534 6535 6536 6537 6538 6539 6540 6541 6542 6543 6544 6545 6546 6547 6548 6549 6550 6551 6552 6553 6554 6555 6556 6557 6558 6559 6560 6561 6562 6563 6564 6565 6566 6567 6568 6569 6570 6571 6572 6573 6574 6575 6576 6577 6578 6579 6580 6581 6582 6583 6584 6585 6586 6587 6588 6589 6590 6591 6592 6593 6594 6595 6596 6597 6598 6599 6600 6601 6602 6603 6604 6605 6606 6607 6608 6609 6610 6611 6612 6613 6614 6615 6616 6617 6618 6619 6620 6621 6622 6623 6624 6625 6626 6627 6628 6629 6630 6631 6632 6633 6634 6635 6636 6637 6638 6639 6640 6641 6642 6643 6644 6645 6646 6647 6648 6649 6650 6651 6652 6653 6654 6655 6656 6657 6658 6659 6660 6661 6662 6663 6664 6665 6666 6667 6668 6669 6670 6671 6672 6673 6674 6675 6676 6677 6678 6679 6680 6681 6682 6683 6684 6685 6686 6687 6688 6689 6690 6691 6692 6693 6694 6695 6696 6697 6698 6699 6700 6701 6702 6703 6704 6705 6706 6707 6708 6709 6710 6711 6712 6713 6714 6715 6716 6717 6718 6719 6720 6721 6722 6723 6724 6725 6726 6727 6728 6729 6730 6731 6732 6733 6734 6735 6736 6737 6738 6739 6740 6741 6742 6743 6744 6745 6746 6747 6748 6749 6750 6751 6752 6753 6754 6755 6756 6757 6758 6759 6760 6761 6762 6763 6764 6765 6766 6767 6768 6769 6770 6771 6772 6773 6774 6775 6776 6777 6778 6779 6780 6781 6782 6783 6784 6785 6786 6787 6788 6789 6790 6791 6792 6793 6794 6795 6796 6797 6798 6799 6800 6801 6802 6803 6804 6805 6806 6807 6808 6809 6810 6811 6812 6813 6814 6815 6816 6817 6818 6819 6820 6821 6822 6823 6824 6825 6826 6827 6828 6829 6830 6831 6832 6833 6834 6835 6836 6837 6838 6839 6840 6841 6842 6843 6844 6845 6846 6847 6848 6849 6850 6851 6852 6853 6854 6855 6856 6857 6858 6859 6860 6861 6862 6863 6864 6865 6866 6867 6868 6869 6870 6871 6872 6873 6874 6875 6876 6877 6878 6879 6880 6881 6882 6883 6884 6885 6886 6887 6888 6889 6890 6891 6892 6893 6894 6895 6896 6897 6898 6899 6900 6901 6902 6903 6904 6905 6906 6907 6908 6909 6910 6911 6912 6913 6914 6915 6916 6917 6918 6919 6920 6921 6922 6923 6924 6925 6926 6927 6928 6929 6930 6931 6932 6933 6934 6935 6936 6937 6938 6939 6940 6941 6942 6943 6944 6945 6946 6947 6948 6949 6950 6951 6952 6953 6954 6955 6956 6957 6958 6959 6960 6961 6962 6963 6964 6965 6966 6967 6968 6969 6970 6971 6972 6973 6974 6975 6976 6977 6978 6979 6980 6981 6982 6983 6984 6985 6986 6987 6988 6989 6990 6991 6992 6993 6994 6995 6996 6997 6998 6999 7000 7001 7002 7003 7004 7005 7006 7007 7008 7009 7010 7011 7012 7013 7014 7015 7016 7017 7018 7019 7020 7021 7022 7023 7024 7025 7026 7027 7028 7029 7030 7031 7032 7033 7034 7035 7036 7037 7038 7039 7040 7041 7042 7043 7044 7045 7046 7047 7048 7049 7050 7051 7052 7053 7054 7055 7056 7057 7058 7059 7060 7061 7062 7063 7064 7065 7066 7067 7068 7069 7070 7071 7072 7073 7074 7075 7076 7077 7078 7079 7080 7081 7082 7083 7084 7085 7086 7087 7088 7089 7090 7091 7092 7093 7094 7095 7096 7097 7098 7099 7100 7101 7102 7103 7104 7105 7106 7107 7108 7109 7110 7111 7112 7113 7114 7115 7116 7117 7118 7119 7120 7121 7122 7123 7124 7125 7126 7127 7128 7129 7130 7131 7132 7133 7134 7135 7136 7137 7138 7139 7140 7141 7142 7143 7144 7145 7146 7147 7148 7149 7150 7151 7152 7153 7154 7155 7156 7157 7158 7159 7160 7161 7162 7163 7164 7165 7166 7167 7168 7169 7170 7171 7172 7173 7174 7175 7176 7177 7178 7179 7180 7181 7182 7183 7184 7185 7186 7187 7188 7189 7190 7191 7192 7193 7194 7195 7196 7197 7198 7199 7200 7201 7202 7203 7204 7205 7206 7207 7208 7209 7210 7211 7212 7213 7214 7215 7216 7217 7218 7219 7220 7221 7222 7223 7224 7225 7226 7227 7228 7229 7230 7231 7232 7233 7234 7235 7236 7237 7238 7239 7240 7241 7242 7243 7244 7245 7246 7247 7248 7249 7250 7251 7252 7253 7254 7255 7256 7257 7258 7259 7260 7261 7262 7263 7264 7265 7266 7267 7268 7269 7270 7271 7272 7273 7274 7275 7276 7277 7278 7279 7280 7281 7282 7283 7284 7285 7286 7287 7288 7289 7290 7291 7292 7293 7294 7295 7296 7297 7298 7299 7300 7301 7302 7303 7304 7305 7306 7307 7308 7309 7310 7311 7312 7313 7314 7315 7316 7317 7318 7319 7320 7321 7322 7323 7324 7325 7326 7327 7328 7329 7330 7331 7332 7333 7334 7335 7336 7337 7338 7339 7340 7341 7342 7343 7344 7345 7346 7347 7348 7349 7350 7351 7352 7353 7354 7355 7356 7357 7358 7359 7360 7361 7362 7363 7364 7365 7366 7367 7368 7369 7370 7371 7372 7373 7374 7375 7376 7377 7378 7379 7380 7381 7382 7383 7384 7385 7386 7387 7388 7389 7390 7391 7392 7393 7394 7395 7396 7397 7398 7399 7400 7401 7402 7403 7404 7405 7406 7407 7408 7409 7410 7411 7412 7413 7414 7415 7416 7417 7418 7419 7420 7421 7422 7423 7424 7425 7426 7427 7428 7429 7430 7431 7432 7433 7434 7435 7436 7437 7438 7439 7440 7441 7442 7443 7444 7445 7446 7447 7448 7449 7450 7451 7452 7453 7454 7455 7456 7457 7458 7459 7460 7461 7462 7463 7464 7465 7466 7467 7468 7469 7470 7471 7472 7473 7474 7475 7476 7477 7478 7479 7480 7481 7482 7483 7484 7485 7486 7487 7488 7489 7490 7491 7492 7493 7494 7495 7496 7497 7498 7499 7500 7501 7502 7503 7504 7505 7506 7507 7508 7509 7510 7511 7512 7513 7514 7515 7516 7517 7518 7519 7520 7521 7522 7523 7524 7525 7526 7527 7528 7529 7530 7531 7532 7533 7534 7535 7536 7537 7538 7539 7540 7541 7542 7543 7544 7545 7546 7547 7548 7549 7550 7551 7552 7553 7554 7555 7556 7557 7558 7559 7560 7561 7562 7563 7564 7565 7566 7567 7568 7569 7570 7571 7572 7573 7574 7575 7576 7577 7578 7579 7580 7581 7582 7583 7584 7585 7586 7587 7588 7589 7590 7591 7592 7593 7594 7595 7596 7597 7598 7599 7600 7601 7602 7603 7604 7605 7606 7607 7608 7609 7610 7611 7612 7613 7614 7615 7616 7617 7618 7619 7620 7621 7622 7623 7624 7625 7626 7627 7628 7629 7630 7631 7632 7633 7634 7635 7636 7637 7638 7639 7640 7641 7642 7643 7644 7645 7646 7647 7648 7649 7650 7651 7652 7653 7654 7655 7656 7657 7658 7659 7660 7661 7662 7663 7664 7665 7666 7667 7668 7669 7670 7671 7672 7673 7674 7675 7676 7677 7678 7679 7680 7681 7682 7683 7684 7685 7686 7687 7688 7689 7690 7691 7692 7693 7694 7695 7696 7697 7698 7699 7700 7701 7702 7703 7704 7705 7706 7707 7708 7709 7710 7711 7712 7713 7714 7715 7716 7717 7718 7719 7720 7721 7722 7723 7724 7725 7726 7727 7728 7729 7730 7731 7732 7733 7734 7735 7736 7737 7738 7739 7740 7741 7742 7743 7744 7745 7746 7747 7748 7749 7750 7751 7752 7753 7754 7755 7756 7757 7758 7759 7760 7761 7762 7763 7764 7765 7766 7767 7768 7769 7770 7771 7772 7773 7774 7775 7776 7777
|
/*************************************************
* Perl-Compatible Regular Expressions *
*************************************************/
/* PCRE is a library of functions to support regular expressions whose syntax
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
Original API code Copyright (c) 1997-2012 University of Cambridge
New API code Copyright (c) 2015-2024 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:
* Redistributions of source code must retain the above copyright notice,
this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above copyright
notice, this list of conditions and the following disclaimer in the
documentation and/or other materials provided with the distribution.
* Neither the name of the University of Cambridge nor the names of its
contributors may be used to endorse or promote products derived from
this software without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
POSSIBILITY OF SUCH DAMAGE.
-----------------------------------------------------------------------------
*/
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include "pcre2_internal.h"
/* These defines enable debugging code */
/* #define DEBUG_FRAMES_DISPLAY */
/* #define DEBUG_SHOW_OPS */
/* #define DEBUG_SHOW_RMATCH */
#ifdef DEBUG_FRAMES_DISPLAY
#include <stdarg.h>
#endif
#ifdef DEBUG_SHOW_OPS
static const char *OP_names[] = { OP_NAME_LIST };
#endif
/* These defines identify the name of the block containing "static"
information, and fields within it. */
#define NLBLOCK mb /* Block containing newline information */
#define PSSTART start_subject /* Field containing processed string start */
#define PSEND end_subject /* Field containing processed string end */
#define RECURSE_UNSET 0xffffffffu /* Bigger than max group number */
/* Masks for identifying the public options that are permitted at match time. */
#define PUBLIC_MATCH_OPTIONS \
(PCRE2_ANCHORED|PCRE2_ENDANCHORED|PCRE2_NOTBOL|PCRE2_NOTEOL|PCRE2_NOTEMPTY| \
PCRE2_NOTEMPTY_ATSTART|PCRE2_NO_UTF_CHECK|PCRE2_PARTIAL_HARD| \
PCRE2_PARTIAL_SOFT|PCRE2_NO_JIT|PCRE2_COPY_MATCHED_SUBJECT| \
PCRE2_DISABLE_RECURSELOOP_CHECK)
#define PUBLIC_JIT_MATCH_OPTIONS \
(PCRE2_NO_UTF_CHECK|PCRE2_NOTBOL|PCRE2_NOTEOL|PCRE2_NOTEMPTY|\
PCRE2_NOTEMPTY_ATSTART|PCRE2_PARTIAL_SOFT|PCRE2_PARTIAL_HARD|\
PCRE2_COPY_MATCHED_SUBJECT)
/* Non-error returns from and within the match() function. Error returns are
externally defined PCRE2_ERROR_xxx codes, which are all negative. */
#define MATCH_MATCH 1
#define MATCH_NOMATCH 0
/* Special internal returns used in the match() function. Make them
sufficiently negative to avoid the external error codes. */
#define MATCH_ACCEPT (-999)
#define MATCH_KETRPOS (-998)
/* The next 5 must be kept together and in sequence so that a test that checks
for any one of them can use a range. */
#define MATCH_COMMIT (-997)
#define MATCH_PRUNE (-996)
#define MATCH_SKIP (-995)
#define MATCH_SKIP_ARG (-994)
#define MATCH_THEN (-993)
#define MATCH_BACKTRACK_MAX MATCH_THEN
#define MATCH_BACKTRACK_MIN MATCH_COMMIT
/* Group frame type values. Zero means the frame is not a group frame. The
lower 16 bits are used for data (e.g. the capture number). Group frames are
used for most groups so that information about the start is easily available at
the end without having to scan back through intermediate frames (backtrack
points). */
#define GF_CAPTURE 0x00010000u
#define GF_NOCAPTURE 0x00020000u
#define GF_CONDASSERT 0x00030000u
#define GF_RECURSE 0x00040000u
/* Masks for the identity and data parts of the group frame type. */
#define GF_IDMASK(a) ((a) & 0xffff0000u)
#define GF_DATAMASK(a) ((a) & 0x0000ffffu)
/* Repetition types */
enum { REPTYPE_MIN, REPTYPE_MAX, REPTYPE_POS };
/* Min and max values for the common repeats; a maximum of UINT32_MAX =>
infinity. */
static const uint32_t rep_min[] = {
0, 0, /* * and *? */
1, 1, /* + and +? */
0, 0, /* ? and ?? */
0, 0, /* dummy placefillers for OP_CR[MIN]RANGE */
0, 1, 0 }; /* OP_CRPOS{STAR, PLUS, QUERY} */
static const uint32_t rep_max[] = {
UINT32_MAX, UINT32_MAX, /* * and *? */
UINT32_MAX, UINT32_MAX, /* + and +? */
1, 1, /* ? and ?? */
0, 0, /* dummy placefillers for OP_CR[MIN]RANGE */
UINT32_MAX, UINT32_MAX, 1 }; /* OP_CRPOS{STAR, PLUS, QUERY} */
/* Repetition types - must include OP_CRPOSRANGE (not needed above) */
static const uint32_t rep_typ[] = {
REPTYPE_MAX, REPTYPE_MIN, /* * and *? */
REPTYPE_MAX, REPTYPE_MIN, /* + and +? */
REPTYPE_MAX, REPTYPE_MIN, /* ? and ?? */
REPTYPE_MAX, REPTYPE_MIN, /* OP_CRRANGE and OP_CRMINRANGE */
REPTYPE_POS, REPTYPE_POS, /* OP_CRPOSSTAR, OP_CRPOSPLUS */
REPTYPE_POS, REPTYPE_POS }; /* OP_CRPOSQUERY, OP_CRPOSRANGE */
/* Numbers for RMATCH calls at backtracking points. When these lists are
changed, the code at RETURN_SWITCH below must be updated in sync. */
enum { RM1=1, RM2, RM3, RM4, RM5, RM6, RM7, RM8, RM9, RM10,
RM11, RM12, RM13, RM14, RM15, RM16, RM17, RM18, RM19, RM20,
RM21, RM22, RM23, RM24, RM25, RM26, RM27, RM28, RM29, RM30,
RM31, RM32, RM33, RM34, RM35, RM36, RM37 };
#ifdef SUPPORT_WIDE_CHARS
enum { RM100=100, RM101 };
#endif
#ifdef SUPPORT_UNICODE
enum { RM200=200, RM201, RM202, RM203, RM204, RM205, RM206, RM207,
RM208, RM209, RM210, RM211, RM212, RM213, RM214, RM215,
RM216, RM217, RM218, RM219, RM220, RM221, RM222, RM223,
RM224, RM225 };
#endif
/* Define short names for general fields in the current backtrack frame, which
is always pointed to by the F variable. Occasional references to fields in
other frames are written out explicitly. There are also some fields in the
current frame whose names start with "temp" that are used for short-term,
localised backtracking memory. These are #defined with Lxxx names at the point
of use and undefined afterwards. */
#define Fback_frame F->back_frame
#define Fcapture_last F->capture_last
#define Fcurrent_recurse F->current_recurse
#define Fecode F->ecode
#define Feptr F->eptr
#define Fgroup_frame_type F->group_frame_type
#define Flast_group_offset F->last_group_offset
#define Flength F->length
#define Fmark F->mark
#define Frdepth F->rdepth
#define Fstart_match F->start_match
#define Foffset_top F->offset_top
#define Foccu F->occu
#define Fop F->op
#define Fovector F->ovector
#define Freturn_id F->return_id
#ifdef DEBUG_FRAMES_DISPLAY
/*************************************************
* Display current frames and contents *
*************************************************/
/* This debugging function displays the current set of frames and their
contents. It is not called automatically from anywhere, the intention being
that calls can be inserted where necessary when debugging frame-related
problems.
Arguments:
f the file to write to
F the current top frame
P a previous frame of interest
frame_size the frame size
mb points to the match block
match_data points to the match data block
s identification text
Returns: nothing
*/
static void
display_frames(FILE *f, heapframe *F, heapframe *P, PCRE2_SIZE frame_size,
match_block *mb, pcre2_match_data *match_data, const char *s, ...)
{
uint32_t i;
heapframe *Q;
va_list ap;
va_start(ap, s);
fprintf(f, "FRAMES ");
vfprintf(f, s, ap);
va_end(ap);
if (P != NULL) fprintf(f, " P=%lu",
((char *)P - (char *)(match_data->heapframes))/frame_size);
fprintf(f, "\n");
for (i = 0, Q = match_data->heapframes;
Q <= F;
i++, Q = (heapframe *)((char *)Q + frame_size))
{
fprintf(f, "Frame %d type=%x subj=%lu code=%d back=%lu id=%d",
i, Q->group_frame_type, Q->eptr - mb->start_subject, *(Q->ecode),
Q->back_frame, Q->return_id);
if (Q->last_group_offset == PCRE2_UNSET)
fprintf(f, " lgoffset=unset\n");
else
fprintf(f, " lgoffset=%lu\n", Q->last_group_offset/frame_size);
}
}
#endif
/*************************************************
* Process a callout *
*************************************************/
/* This function is called for all callouts, whether "standalone" or at the
start of a conditional group. Feptr will be pointing to either OP_CALLOUT or
OP_CALLOUT_STR. A callout block is allocated in pcre2_match() and initialized
with fixed values.
Arguments:
F points to the current backtracking frame
mb points to the match block
lengthptr where to return the length of the callout item
Returns: the return from the callout
or 0 if no callout function exists
*/
static int
do_callout(heapframe *F, match_block *mb, PCRE2_SIZE *lengthptr)
{
int rc;
PCRE2_SIZE save0, save1;
PCRE2_SIZE *callout_ovector;
pcre2_callout_block *cb;
*lengthptr = (*Fecode == OP_CALLOUT)?
PRIV(OP_lengths)[OP_CALLOUT] : GET(Fecode, 1 + 2*LINK_SIZE);
if (mb->callout == NULL) return 0; /* No callout function provided */
/* The original matching code (pre 10.30) worked directly with the ovector
passed by the user, and this was passed to callouts. Now that the working
ovector is in the backtracking frame, it no longer needs to reserve space for
the overall match offsets (which would waste space in the frame). For backward
compatibility, however, we pass capture_top and offset_vector to the callout as
if for the extended ovector, and we ensure that the first two slots are unset
by preserving and restoring their current contents. Picky compilers complain if
references such as Fovector[-2] are use directly, so we set up a separate
pointer. */
callout_ovector = (PCRE2_SIZE *)(Fovector) - 2;
/* The cb->version, cb->subject, cb->subject_length, and cb->start_match fields
are set externally. The first 3 never change; the last is updated for each
bumpalong. */
cb = mb->cb;
cb->capture_top = (uint32_t)Foffset_top/2 + 1;
cb->capture_last = Fcapture_last;
cb->offset_vector = callout_ovector;
cb->mark = mb->nomatch_mark;
cb->current_position = (PCRE2_SIZE)(Feptr - mb->start_subject);
cb->pattern_position = GET(Fecode, 1);
cb->next_item_length = GET(Fecode, 1 + LINK_SIZE);
if (*Fecode == OP_CALLOUT) /* Numerical callout */
{
cb->callout_number = Fecode[1 + 2*LINK_SIZE];
cb->callout_string_offset = 0;
cb->callout_string = NULL;
cb->callout_string_length = 0;
}
else /* String callout */
{
cb->callout_number = 0;
cb->callout_string_offset = GET(Fecode, 1 + 3*LINK_SIZE);
cb->callout_string = Fecode + (1 + 4*LINK_SIZE) + 1;
cb->callout_string_length =
*lengthptr - (1 + 4*LINK_SIZE) - 2;
}
save0 = callout_ovector[0];
save1 = callout_ovector[1];
callout_ovector[0] = callout_ovector[1] = PCRE2_UNSET;
rc = mb->callout(cb, mb->callout_data);
callout_ovector[0] = save0;
callout_ovector[1] = save1;
cb->callout_flags = 0;
return rc;
}
/*************************************************
* Match a back-reference *
*************************************************/
/* This function is called only when it is known that the offset lies within
the offsets that have so far been used in the match. Note that in caseless
UTF-8 mode, the number of subject bytes matched may be different to the number
of reference bytes. (In theory this could also happen in UTF-16 mode, but it
seems unlikely.)
Arguments:
offset index into the offset vector
caseless TRUE if caseless
F the current backtracking frame pointer
mb points to match block
lengthptr pointer for returning the length matched
Returns: = 0 sucessful match; number of code units matched is set
< 0 no match
> 0 partial match
*/
static int
match_ref(PCRE2_SIZE offset, BOOL caseless, heapframe *F, match_block *mb,
PCRE2_SIZE *lengthptr)
{
PCRE2_SPTR p;
PCRE2_SIZE length;
PCRE2_SPTR eptr;
PCRE2_SPTR eptr_start;
/* Deal with an unset group. The default is no match, but there is an option to
match an empty string. */
if (offset >= Foffset_top || Fovector[offset] == PCRE2_UNSET)
{
if ((mb->poptions & PCRE2_MATCH_UNSET_BACKREF) != 0)
{
*lengthptr = 0;
return 0; /* Match */
}
else return -1; /* No match */
}
/* Separate the caseless and UTF cases for speed. */
eptr = eptr_start = Feptr;
p = mb->start_subject + Fovector[offset];
length = Fovector[offset+1] - Fovector[offset];
if (caseless)
{
#if defined SUPPORT_UNICODE
BOOL utf = (mb->poptions & PCRE2_UTF) != 0;
if (utf || (mb->poptions & PCRE2_UCP) != 0)
{
PCRE2_SPTR endptr = p + length;
/* Match characters up to the end of the reference. NOTE: the number of
code units matched may differ, because in UTF-8 there are some characters
whose upper and lower case codes have different numbers of bytes. For
example, U+023A (2 bytes in UTF-8) is the upper case version of U+2C65 (3
bytes in UTF-8); a sequence of 3 of the former uses 6 bytes, as does a
sequence of two of the latter. It is important, therefore, to check the
length along the reference, not along the subject (earlier code did this
wrong). UCP without uses Unicode properties but without UTF encoding. */
while (p < endptr)
{
uint32_t c, d;
const ucd_record *ur;
if (eptr >= mb->end_subject) return 1; /* Partial match */
if (utf)
{
GETCHARINC(c, eptr);
GETCHARINC(d, p);
}
else
{
c = *eptr++;
d = *p++;
}
ur = GET_UCD(d);
if (c != d && c != (uint32_t)((int)d + ur->other_case))
{
const uint32_t *pp = PRIV(ucd_caseless_sets) + ur->caseset;
for (;;)
{
if (c < *pp) return -1; /* No match */
if (c == *pp++) break;
}
}
}
}
else
#endif
/* Not in UTF or UCP mode */
{
for (; length > 0; length--)
{
uint32_t cc, cp;
if (eptr >= mb->end_subject) return 1; /* Partial match */
cc = UCHAR21TEST(eptr);
cp = UCHAR21TEST(p);
if (TABLE_GET(cp, mb->lcc, cp) != TABLE_GET(cc, mb->lcc, cc))
return -1; /* No match */
p++;
eptr++;
}
}
}
/* In the caseful case, we can just compare the code units, whether or not we
are in UTF and/or UCP mode. When partial matching, we have to do this unit by
unit. */
else
{
if (mb->partial != 0)
{
for (; length > 0; length--)
{
if (eptr >= mb->end_subject) return 1; /* Partial match */
if (UCHAR21INCTEST(p) != UCHAR21INCTEST(eptr)) return -1; /* No match */
}
}
/* Not partial matching */
else
{
if ((PCRE2_SIZE)(mb->end_subject - eptr) < length) return 1; /* Partial */
if (memcmp(p, eptr, CU2BYTES(length)) != 0) return -1; /* No match */
eptr += length;
}
}
*lengthptr = eptr - eptr_start;
return 0; /* Match */
}
/******************************************************************************
*******************************************************************************
"Recursion" in the match() function
The original match() function was highly recursive, but this proved to be the
source of a number of problems over the years, mostly because of the relatively
small system stacks that are commonly found. As new features were added to
patterns, various kludges were invented to reduce the amount of stack used,
making the code hard to understand in places.
A version did exist that used individual frames on the heap instead of calling
match() recursively, but this ran substantially slower. The current version is
a refactoring that uses a vector of frames to remember backtracking points.
This runs no slower, and possibly even a bit faster than the original recursive
implementation.
At first, an initial vector of size START_FRAMES_SIZE (enough for maybe 50
frames) was allocated on the system stack. If this was not big enough, the heap
was used for a larger vector. However, it turns out that there are environments
where taking as little as 20KiB from the system stack is an embarrassment.
After another refactoring, the heap is used exclusively, but a pointer the
frames vector and its size are cached in the match_data block, so that there is
no new memory allocation if the same match_data block is used for multiple
matches (unless the frames vector has to be extended).
*******************************************************************************
******************************************************************************/
/*************************************************
* Macros for the match() function *
*************************************************/
/* These macros pack up tests that are used for partial matching several times
in the code. The second one is used when we already know we are past the end of
the subject. We set the "hit end" flag if the pointer is at the end of the
subject and either (a) the pointer is past the earliest inspected character
(i.e. something has been matched, even if not part of the actual matched
string), or (b) the pattern contains a lookbehind. These are the conditions for
which adding more characters may allow the current match to continue.
For hard partial matching, we immediately return a partial match. Otherwise,
carrying on means that a complete match on the current subject will be sought.
A partial match is returned only if no complete match can be found. */
#define CHECK_PARTIAL()\
if (Feptr >= mb->end_subject) \
{ \
SCHECK_PARTIAL(); \
}
#define SCHECK_PARTIAL()\
if (mb->partial != 0 && \
(Feptr > mb->start_used_ptr || mb->allowemptypartial)) \
{ \
mb->hitend = TRUE; \
if (mb->partial > 1) return PCRE2_ERROR_PARTIAL; \
}
/* These macros are used to implement backtracking. They simulate a recursive
call to the match() function by means of a local vector of frames which
remember the backtracking points. */
#define RMATCH(ra,rb)\
{\
start_ecode = ra;\
Freturn_id = rb;\
goto MATCH_RECURSE;\
L_##rb:;\
}
#define RRETURN(ra)\
{\
rrc = ra;\
goto RETURN_SWITCH;\
}
/*************************************************
* Match from current position *
*************************************************/
/* This function is called to run one match attempt at a single starting point
in the subject.
Performance note: It might be tempting to extract commonly used fields from the
mb structure (e.g. end_subject) into individual variables to improve
performance. Tests using gcc on a SPARC disproved this; in the first case, it
made performance worse.
Arguments:
start_eptr starting character in subject
start_ecode starting position in compiled code
top_bracket number of capturing parentheses in the pattern
frame_size size of each backtracking frame
match_data pointer to the match_data block
mb pointer to "static" variables block
Returns: MATCH_MATCH if matched ) these values are >= 0
MATCH_NOMATCH if failed to match )
negative MATCH_xxx value for PRUNE, SKIP, etc
negative PCRE2_ERROR_xxx value if aborted by an error condition
(e.g. stopped by repeated call or depth limit)
*/
static int
match(PCRE2_SPTR start_eptr, PCRE2_SPTR start_ecode, uint16_t top_bracket,
PCRE2_SIZE frame_size, pcre2_match_data *match_data, match_block *mb)
{
/* Frame-handling variables */
heapframe *F; /* Current frame pointer */
heapframe *N = NULL; /* Temporary frame pointers */
heapframe *P = NULL;
heapframe *frames_top; /* End of frames vector */
heapframe *assert_accept_frame = NULL; /* For passing back a frame with captures */
PCRE2_SIZE frame_copy_size; /* Amount to copy when creating a new frame */
/* Local variables that do not need to be preserved over calls to RRMATCH(). */
PCRE2_SPTR branch_end = NULL;
PCRE2_SPTR branch_start;
PCRE2_SPTR bracode; /* Temp pointer to start of group */
PCRE2_SIZE offset; /* Used for group offsets */
PCRE2_SIZE length; /* Used for various length calculations */
int rrc; /* Return from functions & backtracking "recursions" */
#ifdef SUPPORT_UNICODE
int proptype; /* Type of character property */
#endif
uint32_t i; /* Used for local loops */
uint32_t fc; /* Character values */
uint32_t number; /* Used for group and other numbers */
uint32_t reptype = 0; /* Type of repetition (0 to avoid compiler warning) */
uint32_t group_frame_type; /* Specifies type for new group frames */
BOOL condition; /* Used in conditional groups */
BOOL cur_is_word; /* Used in "word" tests */
BOOL prev_is_word; /* Used in "word" tests */
/* UTF and UCP flags */
#ifdef SUPPORT_UNICODE
BOOL utf = (mb->poptions & PCRE2_UTF) != 0;
BOOL ucp = (mb->poptions & PCRE2_UCP) != 0;
#else
BOOL utf = FALSE; /* Required for convenience even when no Unicode support */
#endif
/* This is the length of the last part of a backtracking frame that must be
copied when a new frame is created. */
frame_copy_size = frame_size - offsetof(heapframe, eptr);
/* Set up the first frame and the end of the frames vector. */
F = match_data->heapframes;
frames_top = (heapframe *)((char *)F + match_data->heapframes_size);
Frdepth = 0; /* "Recursion" depth */
Fcapture_last = 0; /* Number of most recent capture */
Fcurrent_recurse = RECURSE_UNSET; /* Not pattern recursing. */
Fstart_match = Feptr = start_eptr; /* Current data pointer and start match */
Fmark = NULL; /* Most recent mark */
Foffset_top = 0; /* End of captures within the frame */
Flast_group_offset = PCRE2_UNSET; /* Saved frame of most recent group */
group_frame_type = 0; /* Not a start of group frame */
goto NEW_FRAME; /* Start processing with this frame */
/* Come back here when we want to create a new frame for remembering a
backtracking point. */
MATCH_RECURSE:
/* Set up a new backtracking frame. If the vector is full, get a new one,
doubling the size, but constrained by the heap limit (which is in KiB). */
N = (heapframe *)((char *)F + frame_size);
if ((heapframe *)((char *)N + frame_size) >= frames_top)
{
heapframe *new;
PCRE2_SIZE newsize;
PCRE2_SIZE usedsize = (char *)N - (char *)(match_data->heapframes);
if (match_data->heapframes_size >= PCRE2_SIZE_MAX / 2)
{
if (match_data->heapframes_size == PCRE2_SIZE_MAX - 1)
return PCRE2_ERROR_NOMEMORY;
newsize = PCRE2_SIZE_MAX - 1;
}
else
newsize = match_data->heapframes_size * 2;
if (newsize / 1024 >= mb->heap_limit)
{
PCRE2_SIZE old_size = match_data->heapframes_size / 1024;
if (mb->heap_limit <= old_size)
return PCRE2_ERROR_HEAPLIMIT;
else
{
PCRE2_SIZE max_delta = 1024 * (mb->heap_limit - old_size);
int over_bytes = match_data->heapframes_size % 1024;
if (over_bytes) max_delta -= (1024 - over_bytes);
newsize = match_data->heapframes_size + max_delta;
}
}
/* With a heap limit set, the permitted additional size may not be enough for
another frame, so do a final check. */
if (newsize - usedsize < frame_size) return PCRE2_ERROR_HEAPLIMIT;
new = match_data->memctl.malloc(newsize, match_data->memctl.memory_data);
if (new == NULL) return PCRE2_ERROR_NOMEMORY;
memcpy(new, match_data->heapframes, usedsize);
N = (heapframe *)((char *)new + usedsize);
F = (heapframe *)((char *)N - frame_size);
match_data->memctl.free(match_data->heapframes, match_data->memctl.memory_data);
match_data->heapframes = new;
match_data->heapframes_size = newsize;
frames_top = (heapframe *)((char *)new + newsize);
}
#ifdef DEBUG_SHOW_RMATCH
fprintf(stderr, "++ RMATCH %d frame=%d", Freturn_id, Frdepth + 1);
if (group_frame_type != 0)
{
fprintf(stderr, " type=%x ", group_frame_type);
switch (GF_IDMASK(group_frame_type))
{
case GF_CAPTURE:
fprintf(stderr, "capture=%d", GF_DATAMASK(group_frame_type));
break;
case GF_NOCAPTURE:
fprintf(stderr, "nocapture op=%d", GF_DATAMASK(group_frame_type));
break;
case GF_CONDASSERT:
fprintf(stderr, "condassert op=%d", GF_DATAMASK(group_frame_type));
break;
case GF_RECURSE:
fprintf(stderr, "recurse=%d", GF_DATAMASK(group_frame_type));
break;
default:
fprintf(stderr, "*** unknown ***");
break;
}
}
fprintf(stderr, "\n");
#endif
/* Copy those fields that must be copied into the new frame, increase the
"recursion" depth (i.e. the new frame's index) and then make the new frame
current. */
memcpy((char *)N + offsetof(heapframe, eptr),
(char *)F + offsetof(heapframe, eptr),
frame_copy_size);
N->rdepth = Frdepth + 1;
F = N;
/* Carry on processing with a new frame. */
NEW_FRAME:
Fgroup_frame_type = group_frame_type;
Fecode = start_ecode; /* Starting code pointer */
Fback_frame = frame_size; /* Default is go back one frame */
/* If this is a special type of group frame, remember its offset for quick
access at the end of the group. If this is a recursion, set a new current
recursion value. */
if (group_frame_type != 0)
{
Flast_group_offset = (char *)F - (char *)match_data->heapframes;
if (GF_IDMASK(group_frame_type) == GF_RECURSE)
Fcurrent_recurse = GF_DATAMASK(group_frame_type);
group_frame_type = 0;
}
/* ========================================================================= */
/* This is the main processing loop. First check that we haven't recorded too
many backtracks (search tree is too large), or that we haven't exceeded the
recursive depth limit (used too many backtracking frames). If not, process the
opcodes. */
if (mb->match_call_count++ >= mb->match_limit) return PCRE2_ERROR_MATCHLIMIT;
if (Frdepth >= mb->match_limit_depth) return PCRE2_ERROR_DEPTHLIMIT;
#ifdef DEBUG_SHOW_OPS
fprintf(stderr, "\n++ New frame: type=0x%x subject offset %ld\n",
GF_IDMASK(Fgroup_frame_type), Feptr - mb->start_subject);
#endif
for (;;)
{
#ifdef DEBUG_SHOW_OPS
fprintf(stderr, "++ %2ld op=%3d %s\n", Fecode - mb->start_code, *Fecode,
OP_names[*Fecode]);
#endif
Fop = (uint8_t)(*Fecode); /* Cast needed for 16-bit and 32-bit modes */
switch(Fop)
{
/* ===================================================================== */
/* Before OP_ACCEPT there may be any number of OP_CLOSE opcodes, to close
any currently open capturing brackets. Unlike reaching the end of a group,
where we know the starting frame is at the top of the chained frames, in
this case we have to search back for the relevant frame in case other types
of group that use chained frames have intervened. Multiple OP_CLOSEs always
come innermost first, which matches the chain order. We can ignore this in
a recursion, because captures are not passed out of recursions. */
case OP_CLOSE:
if (Fcurrent_recurse == RECURSE_UNSET)
{
number = GET2(Fecode, 1);
offset = Flast_group_offset;
for(;;)
{
if (offset == PCRE2_UNSET) return PCRE2_ERROR_INTERNAL;
N = (heapframe *)((char *)match_data->heapframes + offset);
P = (heapframe *)((char *)N - frame_size);
if (N->group_frame_type == (GF_CAPTURE | number)) break;
offset = P->last_group_offset;
}
offset = (number << 1) - 2;
Fcapture_last = number;
Fovector[offset] = P->eptr - mb->start_subject;
Fovector[offset+1] = Feptr - mb->start_subject;
if (offset >= Foffset_top) Foffset_top = offset + 2;
}
Fecode += PRIV(OP_lengths)[*Fecode];
break;
/* ===================================================================== */
/* Real or forced end of the pattern, assertion, or recursion. In an
assertion ACCEPT, update the last used pointer and remember the current
frame so that the captures and mark can be fished out of it. */
case OP_ASSERT_ACCEPT:
if (Feptr > mb->last_used_ptr) mb->last_used_ptr = Feptr;
assert_accept_frame = F;
RRETURN(MATCH_ACCEPT);
/* For ACCEPT within a recursion, we have to find the most recent
recursion. If not in a recursion, fall through to code that is common with
OP_END. */
case OP_ACCEPT:
if (Fcurrent_recurse != RECURSE_UNSET)
{
#ifdef DEBUG_SHOW_OPS
fprintf(stderr, "++ Accept within recursion\n");
#endif
offset = Flast_group_offset;
for(;;)
{
if (offset == PCRE2_UNSET) return PCRE2_ERROR_INTERNAL;
N = (heapframe *)((char *)match_data->heapframes + offset);
P = (heapframe *)((char *)N - frame_size);
if (GF_IDMASK(N->group_frame_type) == GF_RECURSE) break;
offset = P->last_group_offset;
}
/* N is now the frame of the recursion; the previous frame is at the
OP_RECURSE position. Go back there, copying the current subject position
and mark, and the start_match position (\K might have changed it), and
then move on past the OP_RECURSE. */
P->eptr = Feptr;
P->mark = Fmark;
P->start_match = Fstart_match;
F = P;
Fecode += 1 + LINK_SIZE;
continue;
}
/* Fall through */
/* OP_END itself can never be reached within a recursion because that is
picked up when the OP_KET that always precedes OP_END is reached. */
case OP_END:
/* Fail for an empty string match if either PCRE2_NOTEMPTY is set, or if
PCRE2_NOTEMPTY_ATSTART is set and we have matched at the start of the
subject. In both cases, backtracking will then try other alternatives, if
any. */
if (Feptr == Fstart_match &&
((mb->moptions & PCRE2_NOTEMPTY) != 0 ||
((mb->moptions & PCRE2_NOTEMPTY_ATSTART) != 0 &&
Fstart_match == mb->start_subject + mb->start_offset)))
{
#ifdef DEBUG_SHOW_OPS
fprintf(stderr, "++ Backtrack because empty string\n");
#endif
RRETURN(MATCH_NOMATCH);
}
/* Fail if PCRE2_ENDANCHORED is set and the end of the match is not
the end of the subject. After (*ACCEPT) we fail the entire match (at this
position) but backtrack if we've reached the end of the pattern. This
applies whether or not we are in a recursion. */
if (Feptr < mb->end_subject &&
((mb->moptions | mb->poptions) & PCRE2_ENDANCHORED) != 0)
{
if (Fop == OP_END)
{
#ifdef DEBUG_SHOW_OPS
fprintf(stderr, "++ Backtrack because not at end (endanchored set)\n");
#endif
RRETURN(MATCH_NOMATCH);
}
#ifdef DEBUG_SHOW_OPS
fprintf(stderr, "++ Failed ACCEPT not at end (endanchnored set)\n");
#endif
return MATCH_NOMATCH; /* (*ACCEPT) */
}
/* We have a successful match of the whole pattern. Record the result and
then do a direct return from the function. If there is space in the offset
vector, set any pairs that follow the highest-numbered captured string but
are less than the number of capturing groups in the pattern to PCRE2_UNSET.
It is documented that this happens. "Gaps" are set to PCRE2_UNSET
dynamically. It is only those at the end that need setting here. */
mb->end_match_ptr = Feptr; /* Record where we ended */
mb->end_offset_top = Foffset_top; /* and how many extracts were taken */
mb->mark = Fmark; /* and the last success mark */
if (Feptr > mb->last_used_ptr) mb->last_used_ptr = Feptr;
match_data->ovector[0] = Fstart_match - mb->start_subject;
match_data->ovector[1] = Feptr - mb->start_subject;
/* Set i to the smaller of the sizes of the external and frame ovectors. */
i = 2 * ((top_bracket + 1 > match_data->oveccount)?
match_data->oveccount : top_bracket + 1);
memcpy(match_data->ovector + 2, Fovector, (i - 2) * sizeof(PCRE2_SIZE));
while (--i >= Foffset_top + 2) match_data->ovector[i] = PCRE2_UNSET;
return MATCH_MATCH; /* Note: NOT RRETURN */
/*===================================================================== */
/* Match any single character type except newline; have to take care with
CRLF newlines and partial matching. */
case OP_ANY:
if (IS_NEWLINE(Feptr)) RRETURN(MATCH_NOMATCH);
if (mb->partial != 0 &&
Feptr == mb->end_subject - 1 &&
NLBLOCK->nltype == NLTYPE_FIXED &&
NLBLOCK->nllen == 2 &&
UCHAR21TEST(Feptr) == NLBLOCK->nl[0])
{
mb->hitend = TRUE;
if (mb->partial > 1) return PCRE2_ERROR_PARTIAL;
}
/* Fall through */
/* Match any single character whatsoever. */
case OP_ALLANY:
if (Feptr >= mb->end_subject) /* DO NOT merge the Feptr++ here; it must */
{ /* not be updated before SCHECK_PARTIAL. */
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
Feptr++;
#ifdef SUPPORT_UNICODE
if (utf) ACROSSCHAR(Feptr < mb->end_subject, Feptr, Feptr++);
#endif
Fecode++;
break;
/* ===================================================================== */
/* Match a single code unit, even in UTF mode. This opcode really does
match any code unit, even newline. (It really should be called ANYCODEUNIT,
of course - the byte name is from pre-16 bit days.) */
case OP_ANYBYTE:
if (Feptr >= mb->end_subject) /* DO NOT merge the Feptr++ here; it must */
{ /* not be updated before SCHECK_PARTIAL. */
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
Feptr++;
Fecode++;
break;
/* ===================================================================== */
/* Match a single character, casefully */
case OP_CHAR:
#ifdef SUPPORT_UNICODE
if (utf)
{
Flength = 1;
Fecode++;
GETCHARLEN(fc, Fecode, Flength);
if (Flength > (PCRE2_SIZE)(mb->end_subject - Feptr))
{
CHECK_PARTIAL(); /* Not SCHECK_PARTIAL() */
RRETURN(MATCH_NOMATCH);
}
for (; Flength > 0; Flength--)
{
if (*Fecode++ != UCHAR21INC(Feptr)) RRETURN(MATCH_NOMATCH);
}
}
else
#endif
/* Not UTF mode */
{
if (mb->end_subject - Feptr < 1)
{
SCHECK_PARTIAL(); /* This one can use SCHECK_PARTIAL() */
RRETURN(MATCH_NOMATCH);
}
if (Fecode[1] != *Feptr++) RRETURN(MATCH_NOMATCH);
Fecode += 2;
}
break;
/* ===================================================================== */
/* Match a single character, caselessly. If we are at the end of the
subject, give up immediately. We get here only when the pattern character
has at most one other case. Characters with more than two cases are coded
as OP_PROP with the pseudo-property PT_CLIST. */
case OP_CHARI:
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
#ifdef SUPPORT_UNICODE
if (utf)
{
Flength = 1;
Fecode++;
GETCHARLEN(fc, Fecode, Flength);
/* If the pattern character's value is < 128, we know that its other case
(if any) is also < 128 (and therefore only one code unit long in all
code-unit widths), so we can use the fast lookup table. We checked above
that there is at least one character left in the subject. */
if (fc < 128)
{
uint32_t cc = UCHAR21(Feptr);
if (mb->lcc[fc] != TABLE_GET(cc, mb->lcc, cc)) RRETURN(MATCH_NOMATCH);
Fecode++;
Feptr++;
}
/* Otherwise we must pick up the subject character and use Unicode
property support to test its other case. Note that we cannot use the
value of "Flength" to check for sufficient bytes left, because the other
case of the character may have more or fewer code units. */
else
{
uint32_t dc;
GETCHARINC(dc, Feptr);
Fecode += Flength;
if (dc != fc && dc != UCD_OTHERCASE(fc)) RRETURN(MATCH_NOMATCH);
}
}
/* If UCP is set without UTF we must do the same as above, but with one
character per code unit. */
else if (ucp)
{
uint32_t cc = UCHAR21(Feptr);
fc = Fecode[1];
if (fc < 128)
{
if (mb->lcc[fc] != TABLE_GET(cc, mb->lcc, cc)) RRETURN(MATCH_NOMATCH);
}
else
{
if (cc != fc && cc != UCD_OTHERCASE(fc)) RRETURN(MATCH_NOMATCH);
}
Feptr++;
Fecode += 2;
}
else
#endif /* SUPPORT_UNICODE */
/* Not UTF or UCP mode; use the table for characters < 256. */
{
if (TABLE_GET(Fecode[1], mb->lcc, Fecode[1])
!= TABLE_GET(*Feptr, mb->lcc, *Feptr)) RRETURN(MATCH_NOMATCH);
Feptr++;
Fecode += 2;
}
break;
/* ===================================================================== */
/* Match not a single character. */
case OP_NOT:
case OP_NOTI:
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
#ifdef SUPPORT_UNICODE
if (utf)
{
uint32_t ch;
Fecode++;
GETCHARINC(ch, Fecode);
GETCHARINC(fc, Feptr);
if (ch == fc)
{
RRETURN(MATCH_NOMATCH); /* Caseful match */
}
else if (Fop == OP_NOTI) /* If caseless */
{
if (ch > 127)
ch = UCD_OTHERCASE(ch);
else
ch = (mb->fcc)[ch];
if (ch == fc) RRETURN(MATCH_NOMATCH);
}
}
/* UCP without UTF is as above, but with one character per code unit. */
else if (ucp)
{
uint32_t ch;
fc = UCHAR21INC(Feptr);
ch = Fecode[1];
Fecode += 2;
if (ch == fc)
{
RRETURN(MATCH_NOMATCH); /* Caseful match */
}
else if (Fop == OP_NOTI) /* If caseless */
{
if (ch > 127)
ch = UCD_OTHERCASE(ch);
else
ch = (mb->fcc)[ch];
if (ch == fc) RRETURN(MATCH_NOMATCH);
}
}
else
#endif /* SUPPORT_UNICODE */
/* Neither UTF nor UCP is set */
{
uint32_t ch = Fecode[1];
fc = UCHAR21INC(Feptr);
if (ch == fc || (Fop == OP_NOTI && TABLE_GET(ch, mb->fcc, ch) == fc))
RRETURN(MATCH_NOMATCH);
Fecode += 2;
}
break;
/* ===================================================================== */
/* Match a single character repeatedly. */
#define Loclength F->temp_size
#define Lstart_eptr F->temp_sptr[0]
#define Lcharptr F->temp_sptr[1]
#define Lmin F->temp_32[0]
#define Lmax F->temp_32[1]
#define Lc F->temp_32[2]
#define Loc F->temp_32[3]
case OP_EXACT:
case OP_EXACTI:
Lmin = Lmax = GET2(Fecode, 1);
Fecode += 1 + IMM2_SIZE;
goto REPEATCHAR;
case OP_POSUPTO:
case OP_POSUPTOI:
reptype = REPTYPE_POS;
Lmin = 0;
Lmax = GET2(Fecode, 1);
Fecode += 1 + IMM2_SIZE;
goto REPEATCHAR;
case OP_UPTO:
case OP_UPTOI:
reptype = REPTYPE_MAX;
Lmin = 0;
Lmax = GET2(Fecode, 1);
Fecode += 1 + IMM2_SIZE;
goto REPEATCHAR;
case OP_MINUPTO:
case OP_MINUPTOI:
reptype = REPTYPE_MIN;
Lmin = 0;
Lmax = GET2(Fecode, 1);
Fecode += 1 + IMM2_SIZE;
goto REPEATCHAR;
case OP_POSSTAR:
case OP_POSSTARI:
reptype = REPTYPE_POS;
Lmin = 0;
Lmax = UINT32_MAX;
Fecode++;
goto REPEATCHAR;
case OP_POSPLUS:
case OP_POSPLUSI:
reptype = REPTYPE_POS;
Lmin = 1;
Lmax = UINT32_MAX;
Fecode++;
goto REPEATCHAR;
case OP_POSQUERY:
case OP_POSQUERYI:
reptype = REPTYPE_POS;
Lmin = 0;
Lmax = 1;
Fecode++;
goto REPEATCHAR;
case OP_STAR:
case OP_STARI:
case OP_MINSTAR:
case OP_MINSTARI:
case OP_PLUS:
case OP_PLUSI:
case OP_MINPLUS:
case OP_MINPLUSI:
case OP_QUERY:
case OP_QUERYI:
case OP_MINQUERY:
case OP_MINQUERYI:
fc = *Fecode++ - ((Fop < OP_STARI)? OP_STAR : OP_STARI);
Lmin = rep_min[fc];
Lmax = rep_max[fc];
reptype = rep_typ[fc];
/* Common code for all repeated single-character matches. We first check
for the minimum number of characters. If the minimum equals the maximum, we
are done. Otherwise, if minimizing, check the rest of the pattern for a
match; if there isn't one, advance up to the maximum, one character at a
time.
If maximizing, advance up to the maximum number of matching characters,
until Feptr is past the end of the maximum run. If possessive, we are
then done (no backing up). Otherwise, match at this position; anything
other than no match is immediately returned. For nomatch, back up one
character, unless we are matching \R and the last thing matched was
\r\n, in which case, back up two code units until we reach the first
optional character position.
The various UTF/non-UTF and caseful/caseless cases are handled separately,
for speed. */
REPEATCHAR:
#ifdef SUPPORT_UNICODE
if (utf)
{
Flength = 1;
Lcharptr = Fecode;
GETCHARLEN(fc, Fecode, Flength);
Fecode += Flength;
/* Handle multi-code-unit character matching, caseful and caseless. */
if (Flength > 1)
{
uint32_t othercase;
if (Fop >= OP_STARI && /* Caseless */
(othercase = UCD_OTHERCASE(fc)) != fc)
Loclength = PRIV(ord2utf)(othercase, Foccu);
else Loclength = 0;
for (i = 1; i <= Lmin; i++)
{
if (Feptr <= mb->end_subject - Flength &&
memcmp(Feptr, Lcharptr, CU2BYTES(Flength)) == 0) Feptr += Flength;
else if (Loclength > 0 &&
Feptr <= mb->end_subject - Loclength &&
memcmp(Feptr, Foccu, CU2BYTES(Loclength)) == 0)
Feptr += Loclength;
else
{
CHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
}
if (Lmin == Lmax) continue;
if (reptype == REPTYPE_MIN)
{
for (;;)
{
RMATCH(Fecode, RM202);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr <= mb->end_subject - Flength &&
memcmp(Feptr, Lcharptr, CU2BYTES(Flength)) == 0) Feptr += Flength;
else if (Loclength > 0 &&
Feptr <= mb->end_subject - Loclength &&
memcmp(Feptr, Foccu, CU2BYTES(Loclength)) == 0)
Feptr += Loclength;
else
{
CHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
}
/* Control never gets here */
}
else /* Maximize */
{
Lstart_eptr = Feptr;
for (i = Lmin; i < Lmax; i++)
{
if (Feptr <= mb->end_subject - Flength &&
memcmp(Feptr, Lcharptr, CU2BYTES(Flength)) == 0)
Feptr += Flength;
else if (Loclength > 0 &&
Feptr <= mb->end_subject - Loclength &&
memcmp(Feptr, Foccu, CU2BYTES(Loclength)) == 0)
Feptr += Loclength;
else
{
CHECK_PARTIAL();
break;
}
}
/* After \C in UTF mode, Lstart_eptr might be in the middle of a
Unicode character. Use <= Lstart_eptr to ensure backtracking doesn't
go too far. */
if (reptype != REPTYPE_POS) for(;;)
{
if (Feptr <= Lstart_eptr) break;
RMATCH(Fecode, RM203);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
Feptr--;
BACKCHAR(Feptr);
}
}
break; /* End of repeated wide character handling */
}
/* Length of UTF character is 1. Put it into the preserved variable and
fall through to the non-UTF code. */
Lc = fc;
}
else
#endif /* SUPPORT_UNICODE */
/* When not in UTF mode, load a single-code-unit character. Then proceed as
above, using Unicode casing if either UTF or UCP is set. */
Lc = *Fecode++;
/* Caseless comparison */
if (Fop >= OP_STARI)
{
#if PCRE2_CODE_UNIT_WIDTH == 8
#ifdef SUPPORT_UNICODE
if (ucp && !utf && Lc > 127) Loc = UCD_OTHERCASE(Lc);
else
#endif /* SUPPORT_UNICODE */
/* Lc will be < 128 in UTF-8 mode. */
Loc = mb->fcc[Lc];
#else /* 16-bit & 32-bit */
#ifdef SUPPORT_UNICODE
if ((utf || ucp) && Lc > 127) Loc = UCD_OTHERCASE(Lc);
else
#endif /* SUPPORT_UNICODE */
Loc = TABLE_GET(Lc, mb->fcc, Lc);
#endif /* PCRE2_CODE_UNIT_WIDTH == 8 */
for (i = 1; i <= Lmin; i++)
{
uint32_t cc; /* Faster than PCRE2_UCHAR */
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
cc = UCHAR21TEST(Feptr);
if (Lc != cc && Loc != cc) RRETURN(MATCH_NOMATCH);
Feptr++;
}
if (Lmin == Lmax) continue;
if (reptype == REPTYPE_MIN)
{
for (;;)
{
uint32_t cc; /* Faster than PCRE2_UCHAR */
RMATCH(Fecode, RM25);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
cc = UCHAR21TEST(Feptr);
if (Lc != cc && Loc != cc) RRETURN(MATCH_NOMATCH);
Feptr++;
}
/* Control never gets here */
}
else /* Maximize */
{
Lstart_eptr = Feptr;
for (i = Lmin; i < Lmax; i++)
{
uint32_t cc; /* Faster than PCRE2_UCHAR */
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
cc = UCHAR21TEST(Feptr);
if (Lc != cc && Loc != cc) break;
Feptr++;
}
if (reptype != REPTYPE_POS) for (;;)
{
if (Feptr == Lstart_eptr) break;
RMATCH(Fecode, RM26);
Feptr--;
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
}
}
}
/* Caseful comparisons (includes all multi-byte characters) */
else
{
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
if (Lc != UCHAR21INCTEST(Feptr)) RRETURN(MATCH_NOMATCH);
}
if (Lmin == Lmax) continue;
if (reptype == REPTYPE_MIN)
{
for (;;)
{
RMATCH(Fecode, RM27);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
if (Lc != UCHAR21INCTEST(Feptr)) RRETURN(MATCH_NOMATCH);
}
/* Control never gets here */
}
else /* Maximize */
{
Lstart_eptr = Feptr;
for (i = Lmin; i < Lmax; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
if (Lc != UCHAR21TEST(Feptr)) break;
Feptr++;
}
if (reptype != REPTYPE_POS) for (;;)
{
if (Feptr <= Lstart_eptr) break;
RMATCH(Fecode, RM28);
Feptr--;
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
}
}
}
break;
#undef Loclength
#undef Lstart_eptr
#undef Lcharptr
#undef Lmin
#undef Lmax
#undef Lc
#undef Loc
/* ===================================================================== */
/* Match a negated single one-byte character repeatedly. This is almost a
repeat of the code for a repeated single character, but I haven't found a
nice way of commoning these up that doesn't require a test of the
positive/negative option for each character match. Maybe that wouldn't add
very much to the time taken, but character matching *is* what this is all
about... */
#define Lstart_eptr F->temp_sptr[0]
#define Lmin F->temp_32[0]
#define Lmax F->temp_32[1]
#define Lc F->temp_32[2]
#define Loc F->temp_32[3]
case OP_NOTEXACT:
case OP_NOTEXACTI:
Lmin = Lmax = GET2(Fecode, 1);
Fecode += 1 + IMM2_SIZE;
goto REPEATNOTCHAR;
case OP_NOTUPTO:
case OP_NOTUPTOI:
Lmin = 0;
Lmax = GET2(Fecode, 1);
reptype = REPTYPE_MAX;
Fecode += 1 + IMM2_SIZE;
goto REPEATNOTCHAR;
case OP_NOTMINUPTO:
case OP_NOTMINUPTOI:
Lmin = 0;
Lmax = GET2(Fecode, 1);
reptype = REPTYPE_MIN;
Fecode += 1 + IMM2_SIZE;
goto REPEATNOTCHAR;
case OP_NOTPOSSTAR:
case OP_NOTPOSSTARI:
reptype = REPTYPE_POS;
Lmin = 0;
Lmax = UINT32_MAX;
Fecode++;
goto REPEATNOTCHAR;
case OP_NOTPOSPLUS:
case OP_NOTPOSPLUSI:
reptype = REPTYPE_POS;
Lmin = 1;
Lmax = UINT32_MAX;
Fecode++;
goto REPEATNOTCHAR;
case OP_NOTPOSQUERY:
case OP_NOTPOSQUERYI:
reptype = REPTYPE_POS;
Lmin = 0;
Lmax = 1;
Fecode++;
goto REPEATNOTCHAR;
case OP_NOTPOSUPTO:
case OP_NOTPOSUPTOI:
reptype = REPTYPE_POS;
Lmin = 0;
Lmax = GET2(Fecode, 1);
Fecode += 1 + IMM2_SIZE;
goto REPEATNOTCHAR;
case OP_NOTSTAR:
case OP_NOTSTARI:
case OP_NOTMINSTAR:
case OP_NOTMINSTARI:
case OP_NOTPLUS:
case OP_NOTPLUSI:
case OP_NOTMINPLUS:
case OP_NOTMINPLUSI:
case OP_NOTQUERY:
case OP_NOTQUERYI:
case OP_NOTMINQUERY:
case OP_NOTMINQUERYI:
fc = *Fecode++ - ((Fop >= OP_NOTSTARI)? OP_NOTSTARI: OP_NOTSTAR);
Lmin = rep_min[fc];
Lmax = rep_max[fc];
reptype = rep_typ[fc];
/* Common code for all repeated single-character non-matches. */
REPEATNOTCHAR:
GETCHARINCTEST(Lc, Fecode);
/* The code is duplicated for the caseless and caseful cases, for speed,
since matching characters is likely to be quite common. First, ensure the
minimum number of matches are present. If Lmin = Lmax, we are done.
Otherwise, if minimizing, keep trying the rest of the expression and
advancing one matching character if failing, up to the maximum.
Alternatively, if maximizing, find the maximum number of characters and
work backwards. */
if (Fop >= OP_NOTSTARI) /* Caseless */
{
#ifdef SUPPORT_UNICODE
if ((utf || ucp) && Lc > 127)
Loc = UCD_OTHERCASE(Lc);
else
#endif /* SUPPORT_UNICODE */
Loc = TABLE_GET(Lc, mb->fcc, Lc); /* Other case from table */
#ifdef SUPPORT_UNICODE
if (utf)
{
uint32_t d;
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINC(d, Feptr);
if (Lc == d || Loc == d) RRETURN(MATCH_NOMATCH);
}
}
else
#endif /* SUPPORT_UNICODE */
/* Not UTF mode */
{
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
if (Lc == *Feptr || Loc == *Feptr) RRETURN(MATCH_NOMATCH);
Feptr++;
}
}
if (Lmin == Lmax) continue; /* Finished for exact count */
if (reptype == REPTYPE_MIN)
{
#ifdef SUPPORT_UNICODE
if (utf)
{
uint32_t d;
for (;;)
{
RMATCH(Fecode, RM204);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINC(d, Feptr);
if (Lc == d || Loc == d) RRETURN(MATCH_NOMATCH);
}
}
else
#endif /*SUPPORT_UNICODE */
/* Not UTF mode */
{
for (;;)
{
RMATCH(Fecode, RM29);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
if (Lc == *Feptr || Loc == *Feptr) RRETURN(MATCH_NOMATCH);
Feptr++;
}
}
/* Control never gets here */
}
/* Maximize case */
else
{
Lstart_eptr = Feptr;
#ifdef SUPPORT_UNICODE
if (utf)
{
uint32_t d;
for (i = Lmin; i < Lmax; i++)
{
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLEN(d, Feptr, len);
if (Lc == d || Loc == d) break;
Feptr += len;
}
/* After \C in UTF mode, Lstart_eptr might be in the middle of a
Unicode character. Use <= Lstart_eptr to ensure backtracking doesn't
go too far. */
if (reptype != REPTYPE_POS) for(;;)
{
if (Feptr <= Lstart_eptr) break;
RMATCH(Fecode, RM205);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
Feptr--;
BACKCHAR(Feptr);
}
}
else
#endif /* SUPPORT_UNICODE */
/* Not UTF mode */
{
for (i = Lmin; i < Lmax; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
if (Lc == *Feptr || Loc == *Feptr) break;
Feptr++;
}
if (reptype != REPTYPE_POS) for (;;)
{
if (Feptr == Lstart_eptr) break;
RMATCH(Fecode, RM30);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
Feptr--;
}
}
}
}
/* Caseful comparisons */
else
{
#ifdef SUPPORT_UNICODE
if (utf)
{
uint32_t d;
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINC(d, Feptr);
if (Lc == d) RRETURN(MATCH_NOMATCH);
}
}
else
#endif
/* Not UTF mode */
{
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
if (Lc == *Feptr++) RRETURN(MATCH_NOMATCH);
}
}
if (Lmin == Lmax) continue;
if (reptype == REPTYPE_MIN)
{
#ifdef SUPPORT_UNICODE
if (utf)
{
uint32_t d;
for (;;)
{
RMATCH(Fecode, RM206);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINC(d, Feptr);
if (Lc == d) RRETURN(MATCH_NOMATCH);
}
}
else
#endif
/* Not UTF mode */
{
for (;;)
{
RMATCH(Fecode, RM31);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
if (Lc == *Feptr++) RRETURN(MATCH_NOMATCH);
}
}
/* Control never gets here */
}
/* Maximize case */
else
{
Lstart_eptr = Feptr;
#ifdef SUPPORT_UNICODE
if (utf)
{
uint32_t d;
for (i = Lmin; i < Lmax; i++)
{
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLEN(d, Feptr, len);
if (Lc == d) break;
Feptr += len;
}
/* After \C in UTF mode, Lstart_eptr might be in the middle of a
Unicode character. Use <= Lstart_eptr to ensure backtracking doesn't
go too far. */
if (reptype != REPTYPE_POS) for(;;)
{
if (Feptr <= Lstart_eptr) break;
RMATCH(Fecode, RM207);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
Feptr--;
BACKCHAR(Feptr);
}
}
else
#endif
/* Not UTF mode */
{
for (i = Lmin; i < Lmax; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
if (Lc == *Feptr) break;
Feptr++;
}
if (reptype != REPTYPE_POS) for (;;)
{
if (Feptr == Lstart_eptr) break;
RMATCH(Fecode, RM32);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
Feptr--;
}
}
}
}
break;
#undef Lstart_eptr
#undef Lmin
#undef Lmax
#undef Lc
#undef Loc
/* ===================================================================== */
/* Match a bit-mapped character class, possibly repeatedly. These opcodes
are used when all the characters in the class have values in the range
0-255, and either the matching is caseful, or the characters are in the
range 0-127 when UTF processing is enabled. The only difference between
OP_CLASS and OP_NCLASS occurs when a data character outside the range is
encountered. */
#define Lmin F->temp_32[0]
#define Lmax F->temp_32[1]
#define Lstart_eptr F->temp_sptr[0]
#define Lbyte_map_address F->temp_sptr[1]
#define Lbyte_map ((unsigned char *)Lbyte_map_address)
case OP_NCLASS:
case OP_CLASS:
{
Lbyte_map_address = Fecode + 1; /* Save for matching */
Fecode += 1 + (32 / sizeof(PCRE2_UCHAR)); /* Advance past the item */
/* Look past the end of the item to see if there is repeat information
following. Then obey similar code to character type repeats. */
switch (*Fecode)
{
case OP_CRSTAR:
case OP_CRMINSTAR:
case OP_CRPLUS:
case OP_CRMINPLUS:
case OP_CRQUERY:
case OP_CRMINQUERY:
case OP_CRPOSSTAR:
case OP_CRPOSPLUS:
case OP_CRPOSQUERY:
fc = *Fecode++ - OP_CRSTAR;
Lmin = rep_min[fc];
Lmax = rep_max[fc];
reptype = rep_typ[fc];
break;
case OP_CRRANGE:
case OP_CRMINRANGE:
case OP_CRPOSRANGE:
Lmin = GET2(Fecode, 1);
Lmax = GET2(Fecode, 1 + IMM2_SIZE);
if (Lmax == 0) Lmax = UINT32_MAX; /* Max 0 => infinity */
reptype = rep_typ[*Fecode - OP_CRSTAR];
Fecode += 1 + 2 * IMM2_SIZE;
break;
default: /* No repeat follows */
Lmin = Lmax = 1;
break;
}
/* First, ensure the minimum number of matches are present. */
#ifdef SUPPORT_UNICODE
if (utf)
{
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINC(fc, Feptr);
if (fc > 255)
{
if (Fop == OP_CLASS) RRETURN(MATCH_NOMATCH);
}
else
if ((Lbyte_map[fc/8] & (1u << (fc&7))) == 0) RRETURN(MATCH_NOMATCH);
}
}
else
#endif
/* Not UTF mode */
{
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
fc = *Feptr++;
#if PCRE2_CODE_UNIT_WIDTH != 8
if (fc > 255)
{
if (Fop == OP_CLASS) RRETURN(MATCH_NOMATCH);
}
else
#endif
if ((Lbyte_map[fc/8] & (1u << (fc&7))) == 0) RRETURN(MATCH_NOMATCH);
}
}
/* If Lmax == Lmin we are done. Continue with main loop. */
if (Lmin == Lmax) continue;
/* If minimizing, keep testing the rest of the expression and advancing
the pointer while it matches the class. */
if (reptype == REPTYPE_MIN)
{
#ifdef SUPPORT_UNICODE
if (utf)
{
for (;;)
{
RMATCH(Fecode, RM200);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINC(fc, Feptr);
if (fc > 255)
{
if (Fop == OP_CLASS) RRETURN(MATCH_NOMATCH);
}
else
if ((Lbyte_map[fc/8] & (1u << (fc&7))) == 0) RRETURN(MATCH_NOMATCH);
}
}
else
#endif
/* Not UTF mode */
{
for (;;)
{
RMATCH(Fecode, RM23);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
fc = *Feptr++;
#if PCRE2_CODE_UNIT_WIDTH != 8
if (fc > 255)
{
if (Fop == OP_CLASS) RRETURN(MATCH_NOMATCH);
}
else
#endif
if ((Lbyte_map[fc/8] & (1u << (fc&7))) == 0) RRETURN(MATCH_NOMATCH);
}
}
/* Control never gets here */
}
/* If maximizing, find the longest possible run, then work backwards. */
else
{
Lstart_eptr = Feptr;
#ifdef SUPPORT_UNICODE
if (utf)
{
for (i = Lmin; i < Lmax; i++)
{
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLEN(fc, Feptr, len);
if (fc > 255)
{
if (Fop == OP_CLASS) break;
}
else
if ((Lbyte_map[fc/8] & (1u << (fc&7))) == 0) break;
Feptr += len;
}
if (reptype == REPTYPE_POS) continue; /* No backtracking */
/* After \C in UTF mode, Lstart_eptr might be in the middle of a
Unicode character. Use <= Lstart_eptr to ensure backtracking doesn't
go too far. */
for (;;)
{
RMATCH(Fecode, RM201);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Feptr-- <= Lstart_eptr) break; /* Tried at original position */
BACKCHAR(Feptr);
}
}
else
#endif
/* Not UTF mode */
{
for (i = Lmin; i < Lmax; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
fc = *Feptr;
#if PCRE2_CODE_UNIT_WIDTH != 8
if (fc > 255)
{
if (Fop == OP_CLASS) break;
}
else
#endif
if ((Lbyte_map[fc/8] & (1u << (fc&7))) == 0) break;
Feptr++;
}
if (reptype == REPTYPE_POS) continue; /* No backtracking */
while (Feptr >= Lstart_eptr)
{
RMATCH(Fecode, RM24);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
Feptr--;
}
}
RRETURN(MATCH_NOMATCH);
}
}
/* Control never gets here */
#undef Lbyte_map_address
#undef Lbyte_map
#undef Lstart_eptr
#undef Lmin
#undef Lmax
/* ===================================================================== */
/* Match an extended character class. In the 8-bit library, this opcode is
encountered only when UTF-8 mode mode is supported. In the 16-bit and
32-bit libraries, codepoints greater than 255 may be encountered even when
UTF is not supported. */
#define Lstart_eptr F->temp_sptr[0]
#define Lxclass_data F->temp_sptr[1]
#define Lmin F->temp_32[0]
#define Lmax F->temp_32[1]
#ifdef SUPPORT_WIDE_CHARS
case OP_XCLASS:
{
Lxclass_data = Fecode + 1 + LINK_SIZE; /* Save for matching */
Fecode += GET(Fecode, 1); /* Advance past the item */
switch (*Fecode)
{
case OP_CRSTAR:
case OP_CRMINSTAR:
case OP_CRPLUS:
case OP_CRMINPLUS:
case OP_CRQUERY:
case OP_CRMINQUERY:
case OP_CRPOSSTAR:
case OP_CRPOSPLUS:
case OP_CRPOSQUERY:
fc = *Fecode++ - OP_CRSTAR;
Lmin = rep_min[fc];
Lmax = rep_max[fc];
reptype = rep_typ[fc];
break;
case OP_CRRANGE:
case OP_CRMINRANGE:
case OP_CRPOSRANGE:
Lmin = GET2(Fecode, 1);
Lmax = GET2(Fecode, 1 + IMM2_SIZE);
if (Lmax == 0) Lmax = UINT32_MAX; /* Max 0 => infinity */
reptype = rep_typ[*Fecode - OP_CRSTAR];
Fecode += 1 + 2 * IMM2_SIZE;
break;
default: /* No repeat follows */
Lmin = Lmax = 1;
break;
}
/* First, ensure the minimum number of matches are present. */
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
if (!PRIV(xclass)(fc, Lxclass_data, utf)) RRETURN(MATCH_NOMATCH);
}
/* If Lmax == Lmin we can just continue with the main loop. */
if (Lmin == Lmax) continue;
/* If minimizing, keep testing the rest of the expression and advancing
the pointer while it matches the class. */
if (reptype == REPTYPE_MIN)
{
for (;;)
{
RMATCH(Fecode, RM100);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
if (!PRIV(xclass)(fc, Lxclass_data, utf)) RRETURN(MATCH_NOMATCH);
}
/* Control never gets here */
}
/* If maximizing, find the longest possible run, then work backwards. */
else
{
Lstart_eptr = Feptr;
for (i = Lmin; i < Lmax; i++)
{
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
#ifdef SUPPORT_UNICODE
GETCHARLENTEST(fc, Feptr, len);
#else
fc = *Feptr;
#endif
if (!PRIV(xclass)(fc, Lxclass_data, utf)) break;
Feptr += len;
}
if (reptype == REPTYPE_POS) continue; /* No backtracking */
/* After \C in UTF mode, Lstart_eptr might be in the middle of a
Unicode character. Use <= Lstart_eptr to ensure backtracking doesn't
go too far. */
for(;;)
{
RMATCH(Fecode, RM101);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Feptr-- <= Lstart_eptr) break; /* Tried at original position */
#ifdef SUPPORT_UNICODE
if (utf) BACKCHAR(Feptr);
#endif
}
RRETURN(MATCH_NOMATCH);
}
/* Control never gets here */
}
#endif /* SUPPORT_WIDE_CHARS: end of XCLASS */
#undef Lstart_eptr
#undef Lxclass_data
#undef Lmin
#undef Lmax
/* ===================================================================== */
/* Match various character types when PCRE2_UCP is not set. These opcodes
are not generated when PCRE2_UCP is set - instead appropriate property
tests are compiled. */
case OP_NOT_DIGIT:
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
if (CHMAX_255(fc) && (mb->ctypes[fc] & ctype_digit) != 0)
RRETURN(MATCH_NOMATCH);
Fecode++;
break;
case OP_DIGIT:
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
if (!CHMAX_255(fc) || (mb->ctypes[fc] & ctype_digit) == 0)
RRETURN(MATCH_NOMATCH);
Fecode++;
break;
case OP_NOT_WHITESPACE:
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
if (CHMAX_255(fc) && (mb->ctypes[fc] & ctype_space) != 0)
RRETURN(MATCH_NOMATCH);
Fecode++;
break;
case OP_WHITESPACE:
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
if (!CHMAX_255(fc) || (mb->ctypes[fc] & ctype_space) == 0)
RRETURN(MATCH_NOMATCH);
Fecode++;
break;
case OP_NOT_WORDCHAR:
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
if (CHMAX_255(fc) && (mb->ctypes[fc] & ctype_word) != 0)
RRETURN(MATCH_NOMATCH);
Fecode++;
break;
case OP_WORDCHAR:
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
if (!CHMAX_255(fc) || (mb->ctypes[fc] & ctype_word) == 0)
RRETURN(MATCH_NOMATCH);
Fecode++;
break;
case OP_ANYNL:
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
switch(fc)
{
default: RRETURN(MATCH_NOMATCH);
case CHAR_CR:
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
}
else if (UCHAR21TEST(Feptr) == CHAR_LF) Feptr++;
break;
case CHAR_LF:
break;
case CHAR_VT:
case CHAR_FF:
case CHAR_NEL:
#ifndef EBCDIC
case 0x2028:
case 0x2029:
#endif /* Not EBCDIC */
if (mb->bsr_convention == PCRE2_BSR_ANYCRLF) RRETURN(MATCH_NOMATCH);
break;
}
Fecode++;
break;
case OP_NOT_HSPACE:
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
switch(fc)
{
HSPACE_CASES: RRETURN(MATCH_NOMATCH); /* Byte and multibyte cases */
default: break;
}
Fecode++;
break;
case OP_HSPACE:
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
switch(fc)
{
HSPACE_CASES: break; /* Byte and multibyte cases */
default: RRETURN(MATCH_NOMATCH);
}
Fecode++;
break;
case OP_NOT_VSPACE:
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
switch(fc)
{
VSPACE_CASES: RRETURN(MATCH_NOMATCH);
default: break;
}
Fecode++;
break;
case OP_VSPACE:
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
switch(fc)
{
VSPACE_CASES: break;
default: RRETURN(MATCH_NOMATCH);
}
Fecode++;
break;
#ifdef SUPPORT_UNICODE
/* ===================================================================== */
/* Check the next character by Unicode property. We will get here only
if the support is in the binary; otherwise a compile-time error occurs. */
case OP_PROP:
case OP_NOTPROP:
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
{
const uint32_t *cp;
uint32_t chartype;
const ucd_record *prop = GET_UCD(fc);
BOOL notmatch = Fop == OP_NOTPROP;
switch(Fecode[1])
{
case PT_ANY:
if (notmatch) RRETURN(MATCH_NOMATCH);
break;
case PT_LAMP:
chartype = prop->chartype;
if ((chartype == ucp_Lu ||
chartype == ucp_Ll ||
chartype == ucp_Lt) == notmatch)
RRETURN(MATCH_NOMATCH);
break;
case PT_GC:
if ((Fecode[2] == PRIV(ucp_gentype)[prop->chartype]) == notmatch)
RRETURN(MATCH_NOMATCH);
break;
case PT_PC:
if ((Fecode[2] == prop->chartype) == notmatch)
RRETURN(MATCH_NOMATCH);
break;
case PT_SC:
if ((Fecode[2] == prop->script) == notmatch)
RRETURN(MATCH_NOMATCH);
break;
case PT_SCX:
{
BOOL ok = (Fecode[2] == prop->script ||
MAPBIT(PRIV(ucd_script_sets) + UCD_SCRIPTX_PROP(prop), Fecode[2]) != 0);
if (ok == notmatch) RRETURN(MATCH_NOMATCH);
}
break;
/* These are specials */
case PT_ALNUM:
chartype = prop->chartype;
if ((PRIV(ucp_gentype)[chartype] == ucp_L ||
PRIV(ucp_gentype)[chartype] == ucp_N) == notmatch)
RRETURN(MATCH_NOMATCH);
break;
/* Perl space used to exclude VT, but from Perl 5.18 it is included,
which means that Perl space and POSIX space are now identical. PCRE
was changed at release 8.34. */
case PT_SPACE: /* Perl space */
case PT_PXSPACE: /* POSIX space */
switch(fc)
{
HSPACE_CASES:
VSPACE_CASES:
if (notmatch) RRETURN(MATCH_NOMATCH);
break;
default:
if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z) == notmatch)
RRETURN(MATCH_NOMATCH);
break;
}
break;
case PT_WORD:
chartype = prop->chartype;
if ((PRIV(ucp_gentype)[chartype] == ucp_L ||
PRIV(ucp_gentype)[chartype] == ucp_N ||
chartype == ucp_Mn ||
chartype == ucp_Pc) == notmatch)
RRETURN(MATCH_NOMATCH);
break;
case PT_CLIST:
#if PCRE2_CODE_UNIT_WIDTH == 32
if (fc > MAX_UTF_CODE_POINT)
{
if (notmatch) break;;
RRETURN(MATCH_NOMATCH);
}
#endif
cp = PRIV(ucd_caseless_sets) + Fecode[2];
for (;;)
{
if (fc < *cp)
{ if (notmatch) break; else { RRETURN(MATCH_NOMATCH); } }
if (fc == *cp++)
{ if (notmatch) { RRETURN(MATCH_NOMATCH); } else break; }
}
break;
case PT_UCNC:
if ((fc == CHAR_DOLLAR_SIGN || fc == CHAR_COMMERCIAL_AT ||
fc == CHAR_GRAVE_ACCENT || (fc >= 0xa0 && fc <= 0xd7ff) ||
fc >= 0xe000) == notmatch)
RRETURN(MATCH_NOMATCH);
break;
case PT_BIDICL:
if ((UCD_BIDICLASS_PROP(prop) == Fecode[2]) == notmatch)
RRETURN(MATCH_NOMATCH);
break;
case PT_BOOL:
{
BOOL ok = MAPBIT(PRIV(ucd_boolprop_sets) +
UCD_BPROPS_PROP(prop), Fecode[2]) != 0;
if (ok == notmatch) RRETURN(MATCH_NOMATCH);
}
break;
/* This should never occur */
default:
return PCRE2_ERROR_INTERNAL;
}
Fecode += 3;
}
break;
/* ===================================================================== */
/* Match an extended Unicode sequence. We will get here only if the support
is in the binary; otherwise a compile-time error occurs. */
case OP_EXTUNI:
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
else
{
GETCHARINCTEST(fc, Feptr);
Feptr = PRIV(extuni)(fc, Feptr, mb->start_subject, mb->end_subject, utf,
NULL);
}
CHECK_PARTIAL();
Fecode++;
break;
#endif /* SUPPORT_UNICODE */
/* ===================================================================== */
/* Match a single character type repeatedly. Note that the property type
does not need to be in a stack frame as it is not used within an RMATCH()
loop. */
#define Lstart_eptr F->temp_sptr[0]
#define Lmin F->temp_32[0]
#define Lmax F->temp_32[1]
#define Lctype F->temp_32[2]
#define Lpropvalue F->temp_32[3]
case OP_TYPEEXACT:
Lmin = Lmax = GET2(Fecode, 1);
Fecode += 1 + IMM2_SIZE;
goto REPEATTYPE;
case OP_TYPEUPTO:
case OP_TYPEMINUPTO:
Lmin = 0;
Lmax = GET2(Fecode, 1);
reptype = (*Fecode == OP_TYPEMINUPTO)? REPTYPE_MIN : REPTYPE_MAX;
Fecode += 1 + IMM2_SIZE;
goto REPEATTYPE;
case OP_TYPEPOSSTAR:
reptype = REPTYPE_POS;
Lmin = 0;
Lmax = UINT32_MAX;
Fecode++;
goto REPEATTYPE;
case OP_TYPEPOSPLUS:
reptype = REPTYPE_POS;
Lmin = 1;
Lmax = UINT32_MAX;
Fecode++;
goto REPEATTYPE;
case OP_TYPEPOSQUERY:
reptype = REPTYPE_POS;
Lmin = 0;
Lmax = 1;
Fecode++;
goto REPEATTYPE;
case OP_TYPEPOSUPTO:
reptype = REPTYPE_POS;
Lmin = 0;
Lmax = GET2(Fecode, 1);
Fecode += 1 + IMM2_SIZE;
goto REPEATTYPE;
case OP_TYPESTAR:
case OP_TYPEMINSTAR:
case OP_TYPEPLUS:
case OP_TYPEMINPLUS:
case OP_TYPEQUERY:
case OP_TYPEMINQUERY:
fc = *Fecode++ - OP_TYPESTAR;
Lmin = rep_min[fc];
Lmax = rep_max[fc];
reptype = rep_typ[fc];
/* Common code for all repeated character type matches. */
REPEATTYPE:
Lctype = *Fecode++; /* Code for the character type */
#ifdef SUPPORT_UNICODE
if (Lctype == OP_PROP || Lctype == OP_NOTPROP)
{
proptype = *Fecode++;
Lpropvalue = *Fecode++;
}
else proptype = -1;
#endif
/* First, ensure the minimum number of matches are present. Use inline
code for maximizing the speed, and do the type test once at the start
(i.e. keep it out of the loops). As there are no calls to RMATCH in the
loops, we can use an ordinary variable for "notmatch". The code for UTF
mode is separated out for tidiness, except for Unicode property tests. */
if (Lmin > 0)
{
#ifdef SUPPORT_UNICODE
if (proptype >= 0) /* Property tests in all modes */
{
BOOL notmatch = Lctype == OP_NOTPROP;
switch(proptype)
{
case PT_ANY:
if (notmatch) RRETURN(MATCH_NOMATCH);
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
}
break;
case PT_LAMP:
for (i = 1; i <= Lmin; i++)
{
int chartype;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
chartype = UCD_CHARTYPE(fc);
if ((chartype == ucp_Lu ||
chartype == ucp_Ll ||
chartype == ucp_Lt) == notmatch)
RRETURN(MATCH_NOMATCH);
}
break;
case PT_GC:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
if ((UCD_CATEGORY(fc) == Lpropvalue) == notmatch)
RRETURN(MATCH_NOMATCH);
}
break;
case PT_PC:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
if ((UCD_CHARTYPE(fc) == Lpropvalue) == notmatch)
RRETURN(MATCH_NOMATCH);
}
break;
case PT_SC:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
if ((UCD_SCRIPT(fc) == Lpropvalue) == notmatch)
RRETURN(MATCH_NOMATCH);
}
break;
case PT_SCX:
for (i = 1; i <= Lmin; i++)
{
BOOL ok;
const ucd_record *prop;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
prop = GET_UCD(fc);
ok = (prop->script == Lpropvalue ||
MAPBIT(PRIV(ucd_script_sets) + UCD_SCRIPTX_PROP(prop), Lpropvalue) != 0);
if (ok == notmatch)
RRETURN(MATCH_NOMATCH);
}
break;
case PT_ALNUM:
for (i = 1; i <= Lmin; i++)
{
int category;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
category = UCD_CATEGORY(fc);
if ((category == ucp_L || category == ucp_N) == notmatch)
RRETURN(MATCH_NOMATCH);
}
break;
/* Perl space used to exclude VT, but from Perl 5.18 it is included,
which means that Perl space and POSIX space are now identical. PCRE
was changed at release 8.34. */
case PT_SPACE: /* Perl space */
case PT_PXSPACE: /* POSIX space */
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
switch(fc)
{
HSPACE_CASES:
VSPACE_CASES:
if (notmatch) RRETURN(MATCH_NOMATCH);
break;
default:
if ((UCD_CATEGORY(fc) == ucp_Z) == notmatch)
RRETURN(MATCH_NOMATCH);
break;
}
}
break;
case PT_WORD:
for (i = 1; i <= Lmin; i++)
{
int chartype, category;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
chartype = UCD_CHARTYPE(fc);
category = PRIV(ucp_gentype)[chartype];
if ((category == ucp_L || category == ucp_N ||
chartype == ucp_Mn || chartype == ucp_Pc) == notmatch)
RRETURN(MATCH_NOMATCH);
}
break;
case PT_CLIST:
for (i = 1; i <= Lmin; i++)
{
const uint32_t *cp;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
#if PCRE2_CODE_UNIT_WIDTH == 32
if (fc > MAX_UTF_CODE_POINT)
{
if (notmatch) continue;
RRETURN(MATCH_NOMATCH);
}
#endif
cp = PRIV(ucd_caseless_sets) + Lpropvalue;
for (;;)
{
if (fc < *cp)
{
if (notmatch) break;
RRETURN(MATCH_NOMATCH);
}
if (fc == *cp++)
{
if (notmatch) RRETURN(MATCH_NOMATCH);
break;
}
}
}
break;
case PT_UCNC:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
if ((fc == CHAR_DOLLAR_SIGN || fc == CHAR_COMMERCIAL_AT ||
fc == CHAR_GRAVE_ACCENT || (fc >= 0xa0 && fc <= 0xd7ff) ||
fc >= 0xe000) == notmatch)
RRETURN(MATCH_NOMATCH);
}
break;
case PT_BIDICL:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
if ((UCD_BIDICLASS(fc) == Lpropvalue) == notmatch)
RRETURN(MATCH_NOMATCH);
}
break;
case PT_BOOL:
for (i = 1; i <= Lmin; i++)
{
BOOL ok;
const ucd_record *prop;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
prop = GET_UCD(fc);
ok = MAPBIT(PRIV(ucd_boolprop_sets) +
UCD_BPROPS_PROP(prop), Lpropvalue) != 0;
if (ok == notmatch)
RRETURN(MATCH_NOMATCH);
}
break;
/* This should not occur */
default:
return PCRE2_ERROR_INTERNAL;
}
}
/* Match extended Unicode sequences. We will get here only if the
support is in the binary; otherwise a compile-time error occurs. */
else if (Lctype == OP_EXTUNI)
{
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
else
{
GETCHARINCTEST(fc, Feptr);
Feptr = PRIV(extuni)(fc, Feptr, mb->start_subject,
mb->end_subject, utf, NULL);
}
CHECK_PARTIAL();
}
}
else
#endif /* SUPPORT_UNICODE */
/* Handle all other cases in UTF mode */
#ifdef SUPPORT_UNICODE
if (utf) switch(Lctype)
{
case OP_ANY:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
if (IS_NEWLINE(Feptr)) RRETURN(MATCH_NOMATCH);
if (mb->partial != 0 &&
Feptr + 1 >= mb->end_subject &&
NLBLOCK->nltype == NLTYPE_FIXED &&
NLBLOCK->nllen == 2 &&
UCHAR21(Feptr) == NLBLOCK->nl[0])
{
mb->hitend = TRUE;
if (mb->partial > 1) return PCRE2_ERROR_PARTIAL;
}
Feptr++;
ACROSSCHAR(Feptr < mb->end_subject, Feptr, Feptr++);
}
break;
case OP_ALLANY:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
Feptr++;
ACROSSCHAR(Feptr < mb->end_subject, Feptr, Feptr++);
}
break;
case OP_ANYBYTE:
if (Feptr > mb->end_subject - Lmin) RRETURN(MATCH_NOMATCH);
Feptr += Lmin;
break;
case OP_ANYNL:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINC(fc, Feptr);
switch(fc)
{
default: RRETURN(MATCH_NOMATCH);
case CHAR_CR:
if (Feptr < mb->end_subject && UCHAR21(Feptr) == CHAR_LF) Feptr++;
break;
case CHAR_LF:
break;
case CHAR_VT:
case CHAR_FF:
case CHAR_NEL:
#ifndef EBCDIC
case 0x2028:
case 0x2029:
#endif /* Not EBCDIC */
if (mb->bsr_convention == PCRE2_BSR_ANYCRLF) RRETURN(MATCH_NOMATCH);
break;
}
}
break;
case OP_NOT_HSPACE:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINC(fc, Feptr);
switch(fc)
{
HSPACE_CASES: RRETURN(MATCH_NOMATCH);
default: break;
}
}
break;
case OP_HSPACE:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINC(fc, Feptr);
switch(fc)
{
HSPACE_CASES: break;
default: RRETURN(MATCH_NOMATCH);
}
}
break;
case OP_NOT_VSPACE:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINC(fc, Feptr);
switch(fc)
{
VSPACE_CASES: RRETURN(MATCH_NOMATCH);
default: break;
}
}
break;
case OP_VSPACE:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINC(fc, Feptr);
switch(fc)
{
VSPACE_CASES: break;
default: RRETURN(MATCH_NOMATCH);
}
}
break;
case OP_NOT_DIGIT:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINC(fc, Feptr);
if (fc < 128 && (mb->ctypes[fc] & ctype_digit) != 0)
RRETURN(MATCH_NOMATCH);
}
break;
case OP_DIGIT:
for (i = 1; i <= Lmin; i++)
{
uint32_t cc;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
cc = UCHAR21(Feptr);
if (cc >= 128 || (mb->ctypes[cc] & ctype_digit) == 0)
RRETURN(MATCH_NOMATCH);
Feptr++;
/* No need to skip more code units - we know it has only one. */
}
break;
case OP_NOT_WHITESPACE:
for (i = 1; i <= Lmin; i++)
{
uint32_t cc;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
cc = UCHAR21(Feptr);
if (cc < 128 && (mb->ctypes[cc] & ctype_space) != 0)
RRETURN(MATCH_NOMATCH);
Feptr++;
ACROSSCHAR(Feptr < mb->end_subject, Feptr, Feptr++);
}
break;
case OP_WHITESPACE:
for (i = 1; i <= Lmin; i++)
{
uint32_t cc;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
cc = UCHAR21(Feptr);
if (cc >= 128 || (mb->ctypes[cc] & ctype_space) == 0)
RRETURN(MATCH_NOMATCH);
Feptr++;
/* No need to skip more code units - we know it has only one. */
}
break;
case OP_NOT_WORDCHAR:
for (i = 1; i <= Lmin; i++)
{
uint32_t cc;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
cc = UCHAR21(Feptr);
if (cc < 128 && (mb->ctypes[cc] & ctype_word) != 0)
RRETURN(MATCH_NOMATCH);
Feptr++;
ACROSSCHAR(Feptr < mb->end_subject, Feptr, Feptr++);
}
break;
case OP_WORDCHAR:
for (i = 1; i <= Lmin; i++)
{
uint32_t cc;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
cc = UCHAR21(Feptr);
if (cc >= 128 || (mb->ctypes[cc] & ctype_word) == 0)
RRETURN(MATCH_NOMATCH);
Feptr++;
/* No need to skip more code units - we know it has only one. */
}
break;
default:
return PCRE2_ERROR_INTERNAL;
} /* End switch(Lctype) */
else
#endif /* SUPPORT_UNICODE */
/* Code for the non-UTF case for minimum matching of operators other
than OP_PROP and OP_NOTPROP. */
switch(Lctype)
{
case OP_ANY:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
if (IS_NEWLINE(Feptr)) RRETURN(MATCH_NOMATCH);
if (mb->partial != 0 &&
Feptr + 1 >= mb->end_subject &&
NLBLOCK->nltype == NLTYPE_FIXED &&
NLBLOCK->nllen == 2 &&
*Feptr == NLBLOCK->nl[0])
{
mb->hitend = TRUE;
if (mb->partial > 1) return PCRE2_ERROR_PARTIAL;
}
Feptr++;
}
break;
case OP_ALLANY:
if (Feptr > mb->end_subject - Lmin)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
Feptr += Lmin;
break;
/* This OP_ANYBYTE case will never be reached because \C gets turned
into OP_ALLANY in non-UTF mode. Cut out the code so that coverage
reports don't complain about it's never being used. */
/* case OP_ANYBYTE:
* if (Feptr > mb->end_subject - Lmin)
* {
* SCHECK_PARTIAL();
* RRETURN(MATCH_NOMATCH);
* }
* Feptr += Lmin;
* break;
*/
case OP_ANYNL:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
switch(*Feptr++)
{
default: RRETURN(MATCH_NOMATCH);
case CHAR_CR:
if (Feptr < mb->end_subject && *Feptr == CHAR_LF) Feptr++;
break;
case CHAR_LF:
break;
case CHAR_VT:
case CHAR_FF:
case CHAR_NEL:
#if PCRE2_CODE_UNIT_WIDTH != 8
case 0x2028:
case 0x2029:
#endif
if (mb->bsr_convention == PCRE2_BSR_ANYCRLF) RRETURN(MATCH_NOMATCH);
break;
}
}
break;
case OP_NOT_HSPACE:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
switch(*Feptr++)
{
default: break;
HSPACE_BYTE_CASES:
#if PCRE2_CODE_UNIT_WIDTH != 8
HSPACE_MULTIBYTE_CASES:
#endif
RRETURN(MATCH_NOMATCH);
}
}
break;
case OP_HSPACE:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
switch(*Feptr++)
{
default: RRETURN(MATCH_NOMATCH);
HSPACE_BYTE_CASES:
#if PCRE2_CODE_UNIT_WIDTH != 8
HSPACE_MULTIBYTE_CASES:
#endif
break;
}
}
break;
case OP_NOT_VSPACE:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
switch(*Feptr++)
{
VSPACE_BYTE_CASES:
#if PCRE2_CODE_UNIT_WIDTH != 8
VSPACE_MULTIBYTE_CASES:
#endif
RRETURN(MATCH_NOMATCH);
default: break;
}
}
break;
case OP_VSPACE:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
switch(*Feptr++)
{
default: RRETURN(MATCH_NOMATCH);
VSPACE_BYTE_CASES:
#if PCRE2_CODE_UNIT_WIDTH != 8
VSPACE_MULTIBYTE_CASES:
#endif
break;
}
}
break;
case OP_NOT_DIGIT:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
if (MAX_255(*Feptr) && (mb->ctypes[*Feptr] & ctype_digit) != 0)
RRETURN(MATCH_NOMATCH);
Feptr++;
}
break;
case OP_DIGIT:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
if (!MAX_255(*Feptr) || (mb->ctypes[*Feptr] & ctype_digit) == 0)
RRETURN(MATCH_NOMATCH);
Feptr++;
}
break;
case OP_NOT_WHITESPACE:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
if (MAX_255(*Feptr) && (mb->ctypes[*Feptr] & ctype_space) != 0)
RRETURN(MATCH_NOMATCH);
Feptr++;
}
break;
case OP_WHITESPACE:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
if (!MAX_255(*Feptr) || (mb->ctypes[*Feptr] & ctype_space) == 0)
RRETURN(MATCH_NOMATCH);
Feptr++;
}
break;
case OP_NOT_WORDCHAR:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
if (MAX_255(*Feptr) && (mb->ctypes[*Feptr] & ctype_word) != 0)
RRETURN(MATCH_NOMATCH);
Feptr++;
}
break;
case OP_WORDCHAR:
for (i = 1; i <= Lmin; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
if (!MAX_255(*Feptr) || (mb->ctypes[*Feptr] & ctype_word) == 0)
RRETURN(MATCH_NOMATCH);
Feptr++;
}
break;
default:
return PCRE2_ERROR_INTERNAL;
}
}
/* If Lmin = Lmax we are done. Continue with the main loop. */
if (Lmin == Lmax) continue;
/* If minimizing, we have to test the rest of the pattern before each
subsequent match. This means we cannot use a local "notmatch" variable as
in the other cases. As all 4 temporary 32-bit values in the frame are
already in use, just test the type each time. */
if (reptype == REPTYPE_MIN)
{
#ifdef SUPPORT_UNICODE
if (proptype >= 0)
{
switch(proptype)
{
case PT_ANY:
for (;;)
{
RMATCH(Fecode, RM208);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
if (Lctype == OP_NOTPROP) RRETURN(MATCH_NOMATCH);
}
/* Control never gets here */
case PT_LAMP:
for (;;)
{
int chartype;
RMATCH(Fecode, RM209);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
chartype = UCD_CHARTYPE(fc);
if ((chartype == ucp_Lu ||
chartype == ucp_Ll ||
chartype == ucp_Lt) == (Lctype == OP_NOTPROP))
RRETURN(MATCH_NOMATCH);
}
/* Control never gets here */
case PT_GC:
for (;;)
{
RMATCH(Fecode, RM210);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
if ((UCD_CATEGORY(fc) == Lpropvalue) == (Lctype == OP_NOTPROP))
RRETURN(MATCH_NOMATCH);
}
/* Control never gets here */
case PT_PC:
for (;;)
{
RMATCH(Fecode, RM211);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
if ((UCD_CHARTYPE(fc) == Lpropvalue) == (Lctype == OP_NOTPROP))
RRETURN(MATCH_NOMATCH);
}
/* Control never gets here */
case PT_SC:
for (;;)
{
RMATCH(Fecode, RM212);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
if ((UCD_SCRIPT(fc) == Lpropvalue) == (Lctype == OP_NOTPROP))
RRETURN(MATCH_NOMATCH);
}
/* Control never gets here */
case PT_SCX:
for (;;)
{
BOOL ok;
const ucd_record *prop;
RMATCH(Fecode, RM225);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
prop = GET_UCD(fc);
ok = (prop->script == Lpropvalue
|| MAPBIT(PRIV(ucd_script_sets) + UCD_SCRIPTX_PROP(prop), Lpropvalue) != 0);
if (ok == (Lctype == OP_NOTPROP))
RRETURN(MATCH_NOMATCH);
}
/* Control never gets here */
case PT_ALNUM:
for (;;)
{
int category;
RMATCH(Fecode, RM213);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
category = UCD_CATEGORY(fc);
if ((category == ucp_L || category == ucp_N) == (Lctype == OP_NOTPROP))
RRETURN(MATCH_NOMATCH);
}
/* Control never gets here */
/* Perl space used to exclude VT, but from Perl 5.18 it is included,
which means that Perl space and POSIX space are now identical. PCRE
was changed at release 8.34. */
case PT_SPACE: /* Perl space */
case PT_PXSPACE: /* POSIX space */
for (;;)
{
RMATCH(Fecode, RM214);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
switch(fc)
{
HSPACE_CASES:
VSPACE_CASES:
if (Lctype == OP_NOTPROP) RRETURN(MATCH_NOMATCH);
break;
default:
if ((UCD_CATEGORY(fc) == ucp_Z) == (Lctype == OP_NOTPROP))
RRETURN(MATCH_NOMATCH);
break;
}
}
/* Control never gets here */
case PT_WORD:
for (;;)
{
int chartype, category;
RMATCH(Fecode, RM215);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
chartype = UCD_CHARTYPE(fc);
category = PRIV(ucp_gentype)[chartype];
if ((category == ucp_L ||
category == ucp_N ||
chartype == ucp_Mn ||
chartype == ucp_Pc) == (Lctype == OP_NOTPROP))
RRETURN(MATCH_NOMATCH);
}
/* Control never gets here */
case PT_CLIST:
for (;;)
{
const uint32_t *cp;
RMATCH(Fecode, RM216);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
#if PCRE2_CODE_UNIT_WIDTH == 32
if (fc > MAX_UTF_CODE_POINT)
{
if (Lctype == OP_NOTPROP) continue;
RRETURN(MATCH_NOMATCH);
}
#endif
cp = PRIV(ucd_caseless_sets) + Lpropvalue;
for (;;)
{
if (fc < *cp)
{
if (Lctype == OP_NOTPROP) break;
RRETURN(MATCH_NOMATCH);
}
if (fc == *cp++)
{
if (Lctype == OP_NOTPROP) RRETURN(MATCH_NOMATCH);
break;
}
}
}
/* Control never gets here */
case PT_UCNC:
for (;;)
{
RMATCH(Fecode, RM217);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
if ((fc == CHAR_DOLLAR_SIGN || fc == CHAR_COMMERCIAL_AT ||
fc == CHAR_GRAVE_ACCENT || (fc >= 0xa0 && fc <= 0xd7ff) ||
fc >= 0xe000) == (Lctype == OP_NOTPROP))
RRETURN(MATCH_NOMATCH);
}
/* Control never gets here */
case PT_BIDICL:
for (;;)
{
RMATCH(Fecode, RM224);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
if ((UCD_BIDICLASS(fc) == Lpropvalue) == (Lctype == OP_NOTPROP))
RRETURN(MATCH_NOMATCH);
}
/* Control never gets here */
case PT_BOOL:
for (;;)
{
BOOL ok;
const ucd_record *prop;
RMATCH(Fecode, RM223);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
GETCHARINCTEST(fc, Feptr);
prop = GET_UCD(fc);
ok = MAPBIT(PRIV(ucd_boolprop_sets) +
UCD_BPROPS_PROP(prop), Lpropvalue) != 0;
if (ok == (Lctype == OP_NOTPROP))
RRETURN(MATCH_NOMATCH);
}
/* Control never gets here */
/* This should never occur */
default:
return PCRE2_ERROR_INTERNAL;
}
}
/* Match extended Unicode sequences. We will get here only if the
support is in the binary; otherwise a compile-time error occurs. */
else if (Lctype == OP_EXTUNI)
{
for (;;)
{
RMATCH(Fecode, RM218);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
else
{
GETCHARINCTEST(fc, Feptr);
Feptr = PRIV(extuni)(fc, Feptr, mb->start_subject, mb->end_subject,
utf, NULL);
}
CHECK_PARTIAL();
}
}
else
#endif /* SUPPORT_UNICODE */
/* UTF mode for non-property testing character types. */
#ifdef SUPPORT_UNICODE
if (utf)
{
for (;;)
{
RMATCH(Fecode, RM219);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
if (Lctype == OP_ANY && IS_NEWLINE(Feptr)) RRETURN(MATCH_NOMATCH);
GETCHARINC(fc, Feptr);
switch(Lctype)
{
case OP_ANY: /* This is the non-NL case */
if (mb->partial != 0 && /* Take care with CRLF partial */
Feptr >= mb->end_subject &&
NLBLOCK->nltype == NLTYPE_FIXED &&
NLBLOCK->nllen == 2 &&
fc == NLBLOCK->nl[0])
{
mb->hitend = TRUE;
if (mb->partial > 1) return PCRE2_ERROR_PARTIAL;
}
break;
case OP_ALLANY:
case OP_ANYBYTE:
break;
case OP_ANYNL:
switch(fc)
{
default: RRETURN(MATCH_NOMATCH);
case CHAR_CR:
if (Feptr < mb->end_subject && UCHAR21(Feptr) == CHAR_LF) Feptr++;
break;
case CHAR_LF:
break;
case CHAR_VT:
case CHAR_FF:
case CHAR_NEL:
#ifndef EBCDIC
case 0x2028:
case 0x2029:
#endif /* Not EBCDIC */
if (mb->bsr_convention == PCRE2_BSR_ANYCRLF)
RRETURN(MATCH_NOMATCH);
break;
}
break;
case OP_NOT_HSPACE:
switch(fc)
{
HSPACE_CASES: RRETURN(MATCH_NOMATCH);
default: break;
}
break;
case OP_HSPACE:
switch(fc)
{
HSPACE_CASES: break;
default: RRETURN(MATCH_NOMATCH);
}
break;
case OP_NOT_VSPACE:
switch(fc)
{
VSPACE_CASES: RRETURN(MATCH_NOMATCH);
default: break;
}
break;
case OP_VSPACE:
switch(fc)
{
VSPACE_CASES: break;
default: RRETURN(MATCH_NOMATCH);
}
break;
case OP_NOT_DIGIT:
if (fc < 256 && (mb->ctypes[fc] & ctype_digit) != 0)
RRETURN(MATCH_NOMATCH);
break;
case OP_DIGIT:
if (fc >= 256 || (mb->ctypes[fc] & ctype_digit) == 0)
RRETURN(MATCH_NOMATCH);
break;
case OP_NOT_WHITESPACE:
if (fc < 256 && (mb->ctypes[fc] & ctype_space) != 0)
RRETURN(MATCH_NOMATCH);
break;
case OP_WHITESPACE:
if (fc >= 256 || (mb->ctypes[fc] & ctype_space) == 0)
RRETURN(MATCH_NOMATCH);
break;
case OP_NOT_WORDCHAR:
if (fc < 256 && (mb->ctypes[fc] & ctype_word) != 0)
RRETURN(MATCH_NOMATCH);
break;
case OP_WORDCHAR:
if (fc >= 256 || (mb->ctypes[fc] & ctype_word) == 0)
RRETURN(MATCH_NOMATCH);
break;
default:
return PCRE2_ERROR_INTERNAL;
}
}
}
else
#endif /* SUPPORT_UNICODE */
/* Not UTF mode */
{
for (;;)
{
RMATCH(Fecode, RM33);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
if (Lctype == OP_ANY && IS_NEWLINE(Feptr))
RRETURN(MATCH_NOMATCH);
fc = *Feptr++;
switch(Lctype)
{
case OP_ANY: /* This is the non-NL case */
if (mb->partial != 0 && /* Take care with CRLF partial */
Feptr >= mb->end_subject &&
NLBLOCK->nltype == NLTYPE_FIXED &&
NLBLOCK->nllen == 2 &&
fc == NLBLOCK->nl[0])
{
mb->hitend = TRUE;
if (mb->partial > 1) return PCRE2_ERROR_PARTIAL;
}
break;
case OP_ALLANY:
case OP_ANYBYTE:
break;
case OP_ANYNL:
switch(fc)
{
default: RRETURN(MATCH_NOMATCH);
case CHAR_CR:
if (Feptr < mb->end_subject && *Feptr == CHAR_LF) Feptr++;
break;
case CHAR_LF:
break;
case CHAR_VT:
case CHAR_FF:
case CHAR_NEL:
#if PCRE2_CODE_UNIT_WIDTH != 8
case 0x2028:
case 0x2029:
#endif
if (mb->bsr_convention == PCRE2_BSR_ANYCRLF)
RRETURN(MATCH_NOMATCH);
break;
}
break;
case OP_NOT_HSPACE:
switch(fc)
{
default: break;
HSPACE_BYTE_CASES:
#if PCRE2_CODE_UNIT_WIDTH != 8
HSPACE_MULTIBYTE_CASES:
#endif
RRETURN(MATCH_NOMATCH);
}
break;
case OP_HSPACE:
switch(fc)
{
default: RRETURN(MATCH_NOMATCH);
HSPACE_BYTE_CASES:
#if PCRE2_CODE_UNIT_WIDTH != 8
HSPACE_MULTIBYTE_CASES:
#endif
break;
}
break;
case OP_NOT_VSPACE:
switch(fc)
{
default: break;
VSPACE_BYTE_CASES:
#if PCRE2_CODE_UNIT_WIDTH != 8
VSPACE_MULTIBYTE_CASES:
#endif
RRETURN(MATCH_NOMATCH);
}
break;
case OP_VSPACE:
switch(fc)
{
default: RRETURN(MATCH_NOMATCH);
VSPACE_BYTE_CASES:
#if PCRE2_CODE_UNIT_WIDTH != 8
VSPACE_MULTIBYTE_CASES:
#endif
break;
}
break;
case OP_NOT_DIGIT:
if (MAX_255(fc) && (mb->ctypes[fc] & ctype_digit) != 0)
RRETURN(MATCH_NOMATCH);
break;
case OP_DIGIT:
if (!MAX_255(fc) || (mb->ctypes[fc] & ctype_digit) == 0)
RRETURN(MATCH_NOMATCH);
break;
case OP_NOT_WHITESPACE:
if (MAX_255(fc) && (mb->ctypes[fc] & ctype_space) != 0)
RRETURN(MATCH_NOMATCH);
break;
case OP_WHITESPACE:
if (!MAX_255(fc) || (mb->ctypes[fc] & ctype_space) == 0)
RRETURN(MATCH_NOMATCH);
break;
case OP_NOT_WORDCHAR:
if (MAX_255(fc) && (mb->ctypes[fc] & ctype_word) != 0)
RRETURN(MATCH_NOMATCH);
break;
case OP_WORDCHAR:
if (!MAX_255(fc) || (mb->ctypes[fc] & ctype_word) == 0)
RRETURN(MATCH_NOMATCH);
break;
default:
return PCRE2_ERROR_INTERNAL;
}
}
}
/* Control never gets here */
}
/* If maximizing, it is worth using inline code for speed, doing the type
test once at the start (i.e. keep it out of the loops). Once again,
"notmatch" can be an ordinary local variable because the loops do not call
RMATCH. */
else
{
Lstart_eptr = Feptr; /* Remember where we started */
#ifdef SUPPORT_UNICODE
if (proptype >= 0)
{
BOOL notmatch = Lctype == OP_NOTPROP;
switch(proptype)
{
case PT_ANY:
for (i = Lmin; i < Lmax; i++)
{
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLENTEST(fc, Feptr, len);
if (notmatch) break;
Feptr+= len;
}
break;
case PT_LAMP:
for (i = Lmin; i < Lmax; i++)
{
int chartype;
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLENTEST(fc, Feptr, len);
chartype = UCD_CHARTYPE(fc);
if ((chartype == ucp_Lu ||
chartype == ucp_Ll ||
chartype == ucp_Lt) == notmatch)
break;
Feptr+= len;
}
break;
case PT_GC:
for (i = Lmin; i < Lmax; i++)
{
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLENTEST(fc, Feptr, len);
if ((UCD_CATEGORY(fc) == Lpropvalue) == notmatch) break;
Feptr+= len;
}
break;
case PT_PC:
for (i = Lmin; i < Lmax; i++)
{
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLENTEST(fc, Feptr, len);
if ((UCD_CHARTYPE(fc) == Lpropvalue) == notmatch) break;
Feptr+= len;
}
break;
case PT_SC:
for (i = Lmin; i < Lmax; i++)
{
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLENTEST(fc, Feptr, len);
if ((UCD_SCRIPT(fc) == Lpropvalue) == notmatch) break;
Feptr+= len;
}
break;
case PT_SCX:
for (i = Lmin; i < Lmax; i++)
{
BOOL ok;
const ucd_record *prop;
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLENTEST(fc, Feptr, len);
prop = GET_UCD(fc);
ok = (prop->script == Lpropvalue ||
MAPBIT(PRIV(ucd_script_sets) + UCD_SCRIPTX_PROP(prop), Lpropvalue) != 0);
if (ok == notmatch) break;
Feptr+= len;
}
break;
case PT_ALNUM:
for (i = Lmin; i < Lmax; i++)
{
int category;
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLENTEST(fc, Feptr, len);
category = UCD_CATEGORY(fc);
if ((category == ucp_L || category == ucp_N) == notmatch)
break;
Feptr+= len;
}
break;
/* Perl space used to exclude VT, but from Perl 5.18 it is included,
which means that Perl space and POSIX space are now identical. PCRE
was changed at release 8.34. */
case PT_SPACE: /* Perl space */
case PT_PXSPACE: /* POSIX space */
for (i = Lmin; i < Lmax; i++)
{
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLENTEST(fc, Feptr, len);
switch(fc)
{
HSPACE_CASES:
VSPACE_CASES:
if (notmatch) goto ENDLOOP99; /* Break the loop */
break;
default:
if ((UCD_CATEGORY(fc) == ucp_Z) == notmatch)
goto ENDLOOP99; /* Break the loop */
break;
}
Feptr+= len;
}
ENDLOOP99:
break;
case PT_WORD:
for (i = Lmin; i < Lmax; i++)
{
int chartype, category;
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLENTEST(fc, Feptr, len);
chartype = UCD_CHARTYPE(fc);
category = PRIV(ucp_gentype)[chartype];
if ((category == ucp_L ||
category == ucp_N ||
chartype == ucp_Mn ||
chartype == ucp_Pc) == notmatch)
break;
Feptr+= len;
}
break;
case PT_CLIST:
for (i = Lmin; i < Lmax; i++)
{
const uint32_t *cp;
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLENTEST(fc, Feptr, len);
#if PCRE2_CODE_UNIT_WIDTH == 32
if (fc > MAX_UTF_CODE_POINT)
{
if (!notmatch) goto GOT_MAX;
}
else
#endif
{
cp = PRIV(ucd_caseless_sets) + Lpropvalue;
for (;;)
{
if (fc < *cp)
{ if (notmatch) break; else goto GOT_MAX; }
if (fc == *cp++)
{ if (notmatch) goto GOT_MAX; else break; }
}
}
Feptr += len;
}
GOT_MAX:
break;
case PT_UCNC:
for (i = Lmin; i < Lmax; i++)
{
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLENTEST(fc, Feptr, len);
if ((fc == CHAR_DOLLAR_SIGN || fc == CHAR_COMMERCIAL_AT ||
fc == CHAR_GRAVE_ACCENT || (fc >= 0xa0 && fc <= 0xd7ff) ||
fc >= 0xe000) == notmatch)
break;
Feptr += len;
}
break;
case PT_BIDICL:
for (i = Lmin; i < Lmax; i++)
{
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLENTEST(fc, Feptr, len);
if ((UCD_BIDICLASS(fc) == Lpropvalue) == notmatch) break;
Feptr+= len;
}
break;
case PT_BOOL:
for (i = Lmin; i < Lmax; i++)
{
BOOL ok;
const ucd_record *prop;
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLENTEST(fc, Feptr, len);
prop = GET_UCD(fc);
ok = MAPBIT(PRIV(ucd_boolprop_sets) +
UCD_BPROPS_PROP(prop), Lpropvalue) != 0;
if (ok == notmatch) break;
Feptr+= len;
}
break;
default:
return PCRE2_ERROR_INTERNAL;
}
/* Feptr is now past the end of the maximum run */
if (reptype == REPTYPE_POS) continue; /* No backtracking */
/* After \C in UTF mode, Lstart_eptr might be in the middle of a
Unicode character. Use <= Lstart_eptr to ensure backtracking doesn't
go too far. */
for(;;)
{
if (Feptr <= Lstart_eptr) break;
RMATCH(Fecode, RM222);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
Feptr--;
if (utf) BACKCHAR(Feptr);
}
}
/* Match extended Unicode grapheme clusters. We will get here only if the
support is in the binary; otherwise a compile-time error occurs. */
else if (Lctype == OP_EXTUNI)
{
for (i = Lmin; i < Lmax; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
else
{
GETCHARINCTEST(fc, Feptr);
Feptr = PRIV(extuni)(fc, Feptr, mb->start_subject, mb->end_subject,
utf, NULL);
}
CHECK_PARTIAL();
}
/* Feptr is now past the end of the maximum run */
if (reptype == REPTYPE_POS) continue; /* No backtracking */
/* We use <= Lstart_eptr rather than == Lstart_eptr to detect the start
of the run while backtracking because the use of \C in UTF mode can
cause BACKCHAR to move back past Lstart_eptr. This is just palliative;
the use of \C in UTF mode is fraught with danger. */
for(;;)
{
int lgb, rgb;
PCRE2_SPTR fptr;
if (Feptr <= Lstart_eptr) break; /* At start of char run */
RMATCH(Fecode, RM220);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
/* Backtracking over an extended grapheme cluster involves inspecting
the previous two characters (if present) to see if a break is
permitted between them. */
Feptr--;
if (!utf) fc = *Feptr; else
{
BACKCHAR(Feptr);
GETCHAR(fc, Feptr);
}
rgb = UCD_GRAPHBREAK(fc);
for (;;)
{
if (Feptr <= Lstart_eptr) break; /* At start of char run */
fptr = Feptr - 1;
if (!utf) fc = *fptr; else
{
BACKCHAR(fptr);
GETCHAR(fc, fptr);
}
lgb = UCD_GRAPHBREAK(fc);
if ((PRIV(ucp_gbtable)[lgb] & (1u << rgb)) == 0) break;
Feptr = fptr;
rgb = lgb;
}
}
}
else
#endif /* SUPPORT_UNICODE */
#ifdef SUPPORT_UNICODE
if (utf)
{
switch(Lctype)
{
case OP_ANY:
for (i = Lmin; i < Lmax; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
if (IS_NEWLINE(Feptr)) break;
if (mb->partial != 0 && /* Take care with CRLF partial */
Feptr + 1 >= mb->end_subject &&
NLBLOCK->nltype == NLTYPE_FIXED &&
NLBLOCK->nllen == 2 &&
UCHAR21(Feptr) == NLBLOCK->nl[0])
{
mb->hitend = TRUE;
if (mb->partial > 1) return PCRE2_ERROR_PARTIAL;
}
Feptr++;
ACROSSCHAR(Feptr < mb->end_subject, Feptr, Feptr++);
}
break;
case OP_ALLANY:
if (Lmax < UINT32_MAX)
{
for (i = Lmin; i < Lmax; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
Feptr++;
ACROSSCHAR(Feptr < mb->end_subject, Feptr, Feptr++);
}
}
else
{
Feptr = mb->end_subject; /* Unlimited UTF-8 repeat */
SCHECK_PARTIAL();
}
break;
/* The "byte" (i.e. "code unit") case is the same as non-UTF */
case OP_ANYBYTE:
fc = Lmax - Lmin;
if (fc > (uint32_t)(mb->end_subject - Feptr))
{
Feptr = mb->end_subject;
SCHECK_PARTIAL();
}
else Feptr += fc;
break;
case OP_ANYNL:
for (i = Lmin; i < Lmax; i++)
{
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLEN(fc, Feptr, len);
if (fc == CHAR_CR)
{
if (++Feptr >= mb->end_subject) break;
if (UCHAR21(Feptr) == CHAR_LF) Feptr++;
}
else
{
if (fc != CHAR_LF &&
(mb->bsr_convention == PCRE2_BSR_ANYCRLF ||
(fc != CHAR_VT && fc != CHAR_FF && fc != CHAR_NEL
#ifndef EBCDIC
&& fc != 0x2028 && fc != 0x2029
#endif /* Not EBCDIC */
)))
break;
Feptr += len;
}
}
break;
case OP_NOT_HSPACE:
case OP_HSPACE:
for (i = Lmin; i < Lmax; i++)
{
BOOL gotspace;
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLEN(fc, Feptr, len);
switch(fc)
{
HSPACE_CASES: gotspace = TRUE; break;
default: gotspace = FALSE; break;
}
if (gotspace == (Lctype == OP_NOT_HSPACE)) break;
Feptr += len;
}
break;
case OP_NOT_VSPACE:
case OP_VSPACE:
for (i = Lmin; i < Lmax; i++)
{
BOOL gotspace;
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLEN(fc, Feptr, len);
switch(fc)
{
VSPACE_CASES: gotspace = TRUE; break;
default: gotspace = FALSE; break;
}
if (gotspace == (Lctype == OP_NOT_VSPACE)) break;
Feptr += len;
}
break;
case OP_NOT_DIGIT:
for (i = Lmin; i < Lmax; i++)
{
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLEN(fc, Feptr, len);
if (fc < 256 && (mb->ctypes[fc] & ctype_digit) != 0) break;
Feptr+= len;
}
break;
case OP_DIGIT:
for (i = Lmin; i < Lmax; i++)
{
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLEN(fc, Feptr, len);
if (fc >= 256 ||(mb->ctypes[fc] & ctype_digit) == 0) break;
Feptr+= len;
}
break;
case OP_NOT_WHITESPACE:
for (i = Lmin; i < Lmax; i++)
{
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLEN(fc, Feptr, len);
if (fc < 256 && (mb->ctypes[fc] & ctype_space) != 0) break;
Feptr+= len;
}
break;
case OP_WHITESPACE:
for (i = Lmin; i < Lmax; i++)
{
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLEN(fc, Feptr, len);
if (fc >= 256 ||(mb->ctypes[fc] & ctype_space) == 0) break;
Feptr+= len;
}
break;
case OP_NOT_WORDCHAR:
for (i = Lmin; i < Lmax; i++)
{
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLEN(fc, Feptr, len);
if (fc < 256 && (mb->ctypes[fc] & ctype_word) != 0) break;
Feptr+= len;
}
break;
case OP_WORDCHAR:
for (i = Lmin; i < Lmax; i++)
{
int len = 1;
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
GETCHARLEN(fc, Feptr, len);
if (fc >= 256 || (mb->ctypes[fc] & ctype_word) == 0) break;
Feptr+= len;
}
break;
default:
return PCRE2_ERROR_INTERNAL;
}
if (reptype == REPTYPE_POS) continue; /* No backtracking */
/* After \C in UTF mode, Lstart_eptr might be in the middle of a
Unicode character. Use <= Lstart_eptr to ensure backtracking doesn't go
too far. */
for(;;)
{
if (Feptr <= Lstart_eptr) break;
RMATCH(Fecode, RM221);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
Feptr--;
BACKCHAR(Feptr);
if (Lctype == OP_ANYNL && Feptr > Lstart_eptr &&
UCHAR21(Feptr) == CHAR_NL && UCHAR21(Feptr - 1) == CHAR_CR)
Feptr--;
}
}
else
#endif /* SUPPORT_UNICODE */
/* Not UTF mode */
{
switch(Lctype)
{
case OP_ANY:
for (i = Lmin; i < Lmax; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
if (IS_NEWLINE(Feptr)) break;
if (mb->partial != 0 && /* Take care with CRLF partial */
Feptr + 1 >= mb->end_subject &&
NLBLOCK->nltype == NLTYPE_FIXED &&
NLBLOCK->nllen == 2 &&
*Feptr == NLBLOCK->nl[0])
{
mb->hitend = TRUE;
if (mb->partial > 1) return PCRE2_ERROR_PARTIAL;
}
Feptr++;
}
break;
case OP_ALLANY:
case OP_ANYBYTE:
fc = Lmax - Lmin;
if (fc > (uint32_t)(mb->end_subject - Feptr))
{
Feptr = mb->end_subject;
SCHECK_PARTIAL();
}
else Feptr += fc;
break;
case OP_ANYNL:
for (i = Lmin; i < Lmax; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
fc = *Feptr;
if (fc == CHAR_CR)
{
if (++Feptr >= mb->end_subject) break;
if (*Feptr == CHAR_LF) Feptr++;
}
else
{
if (fc != CHAR_LF && (mb->bsr_convention == PCRE2_BSR_ANYCRLF ||
(fc != CHAR_VT && fc != CHAR_FF && fc != CHAR_NEL
#if PCRE2_CODE_UNIT_WIDTH != 8
&& fc != 0x2028 && fc != 0x2029
#endif
))) break;
Feptr++;
}
}
break;
case OP_NOT_HSPACE:
for (i = Lmin; i < Lmax; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
switch(*Feptr)
{
default: Feptr++; break;
HSPACE_BYTE_CASES:
#if PCRE2_CODE_UNIT_WIDTH != 8
HSPACE_MULTIBYTE_CASES:
#endif
goto ENDLOOP00;
}
}
ENDLOOP00:
break;
case OP_HSPACE:
for (i = Lmin; i < Lmax; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
switch(*Feptr)
{
default: goto ENDLOOP01;
HSPACE_BYTE_CASES:
#if PCRE2_CODE_UNIT_WIDTH != 8
HSPACE_MULTIBYTE_CASES:
#endif
Feptr++; break;
}
}
ENDLOOP01:
break;
case OP_NOT_VSPACE:
for (i = Lmin; i < Lmax; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
switch(*Feptr)
{
default: Feptr++; break;
VSPACE_BYTE_CASES:
#if PCRE2_CODE_UNIT_WIDTH != 8
VSPACE_MULTIBYTE_CASES:
#endif
goto ENDLOOP02;
}
}
ENDLOOP02:
break;
case OP_VSPACE:
for (i = Lmin; i < Lmax; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
switch(*Feptr)
{
default: goto ENDLOOP03;
VSPACE_BYTE_CASES:
#if PCRE2_CODE_UNIT_WIDTH != 8
VSPACE_MULTIBYTE_CASES:
#endif
Feptr++; break;
}
}
ENDLOOP03:
break;
case OP_NOT_DIGIT:
for (i = Lmin; i < Lmax; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
if (MAX_255(*Feptr) && (mb->ctypes[*Feptr] & ctype_digit) != 0)
break;
Feptr++;
}
break;
case OP_DIGIT:
for (i = Lmin; i < Lmax; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
if (!MAX_255(*Feptr) || (mb->ctypes[*Feptr] & ctype_digit) == 0)
break;
Feptr++;
}
break;
case OP_NOT_WHITESPACE:
for (i = Lmin; i < Lmax; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
if (MAX_255(*Feptr) && (mb->ctypes[*Feptr] & ctype_space) != 0)
break;
Feptr++;
}
break;
case OP_WHITESPACE:
for (i = Lmin; i < Lmax; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
if (!MAX_255(*Feptr) || (mb->ctypes[*Feptr] & ctype_space) == 0)
break;
Feptr++;
}
break;
case OP_NOT_WORDCHAR:
for (i = Lmin; i < Lmax; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
if (MAX_255(*Feptr) && (mb->ctypes[*Feptr] & ctype_word) != 0)
break;
Feptr++;
}
break;
case OP_WORDCHAR:
for (i = Lmin; i < Lmax; i++)
{
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
break;
}
if (!MAX_255(*Feptr) || (mb->ctypes[*Feptr] & ctype_word) == 0)
break;
Feptr++;
}
break;
default:
return PCRE2_ERROR_INTERNAL;
}
if (reptype == REPTYPE_POS) continue; /* No backtracking */
for (;;)
{
if (Feptr == Lstart_eptr) break;
RMATCH(Fecode, RM34);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
Feptr--;
if (Lctype == OP_ANYNL && Feptr > Lstart_eptr && *Feptr == CHAR_LF &&
Feptr[-1] == CHAR_CR) Feptr--;
}
}
}
break; /* End of repeat character type processing */
#undef Lstart_eptr
#undef Lmin
#undef Lmax
#undef Lctype
#undef Lpropvalue
/* ===================================================================== */
/* Match a back reference, possibly repeatedly. Look past the end of the
item to see if there is repeat information following. The OP_REF and
OP_REFI opcodes are used for a reference to a numbered group or to a
non-duplicated named group. For a duplicated named group, OP_DNREF and
OP_DNREFI are used. In this case we must scan the list of groups to which
the name refers, and use the first one that is set. */
#define Lmin F->temp_32[0]
#define Lmax F->temp_32[1]
#define Lcaseless F->temp_32[2]
#define Lstart F->temp_sptr[0]
#define Loffset F->temp_size
case OP_DNREF:
case OP_DNREFI:
Lcaseless = (Fop == OP_DNREFI);
{
int count = GET2(Fecode, 1+IMM2_SIZE);
PCRE2_SPTR slot = mb->name_table + GET2(Fecode, 1) * mb->name_entry_size;
Fecode += 1 + 2*IMM2_SIZE;
while (count-- > 0)
{
Loffset = (GET2(slot, 0) << 1) - 2;
if (Loffset < Foffset_top && Fovector[Loffset] != PCRE2_UNSET) break;
slot += mb->name_entry_size;
}
}
goto REF_REPEAT;
case OP_REF:
case OP_REFI:
Lcaseless = (Fop == OP_REFI);
Loffset = (GET2(Fecode, 1) << 1) - 2;
Fecode += 1 + IMM2_SIZE;
/* Set up for repetition, or handle the non-repeated case. The maximum and
minimum must be in the heap frame, but as they are short-term values, we
use temporary fields. */
REF_REPEAT:
switch (*Fecode)
{
case OP_CRSTAR:
case OP_CRMINSTAR:
case OP_CRPLUS:
case OP_CRMINPLUS:
case OP_CRQUERY:
case OP_CRMINQUERY:
fc = *Fecode++ - OP_CRSTAR;
Lmin = rep_min[fc];
Lmax = rep_max[fc];
reptype = rep_typ[fc];
break;
case OP_CRRANGE:
case OP_CRMINRANGE:
Lmin = GET2(Fecode, 1);
Lmax = GET2(Fecode, 1 + IMM2_SIZE);
reptype = rep_typ[*Fecode - OP_CRSTAR];
if (Lmax == 0) Lmax = UINT32_MAX; /* Max 0 => infinity */
Fecode += 1 + 2 * IMM2_SIZE;
break;
default: /* No repeat follows */
{
rrc = match_ref(Loffset, Lcaseless, F, mb, &length);
if (rrc != 0)
{
if (rrc > 0) Feptr = mb->end_subject; /* Partial match */
CHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
}
Feptr += length;
continue; /* With the main loop */
}
/* Handle repeated back references. If a set group has length zero, just
continue with the main loop, because it matches however many times. For an
unset reference, if the minimum is zero, we can also just continue. We can
also continue if PCRE2_MATCH_UNSET_BACKREF is set, because this makes unset
group behave as a zero-length group. For any other unset cases, carrying
on will result in NOMATCH. */
if (Loffset < Foffset_top && Fovector[Loffset] != PCRE2_UNSET)
{
if (Fovector[Loffset] == Fovector[Loffset + 1]) continue;
}
else /* Group is not set */
{
if (Lmin == 0 || (mb->poptions & PCRE2_MATCH_UNSET_BACKREF) != 0)
continue;
}
/* First, ensure the minimum number of matches are present. */
for (i = 1; i <= Lmin; i++)
{
PCRE2_SIZE slength;
rrc = match_ref(Loffset, Lcaseless, F, mb, &slength);
if (rrc != 0)
{
if (rrc > 0) Feptr = mb->end_subject; /* Partial match */
CHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
Feptr += slength;
}
/* If min = max, we are done. They are not both allowed to be zero. */
if (Lmin == Lmax) continue;
/* If minimizing, keep trying and advancing the pointer. */
if (reptype == REPTYPE_MIN)
{
for (;;)
{
PCRE2_SIZE slength;
RMATCH(Fecode, RM20);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmin++ >= Lmax) RRETURN(MATCH_NOMATCH);
rrc = match_ref(Loffset, Lcaseless, F, mb, &slength);
if (rrc != 0)
{
if (rrc > 0) Feptr = mb->end_subject; /* Partial match */
CHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
Feptr += slength;
}
/* Control never gets here */
}
/* If maximizing, find the longest string and work backwards, as long as
the matched lengths for each iteration are the same. */
else
{
BOOL samelengths = TRUE;
Lstart = Feptr; /* Starting position */
Flength = Fovector[Loffset+1] - Fovector[Loffset];
for (i = Lmin; i < Lmax; i++)
{
PCRE2_SIZE slength;
rrc = match_ref(Loffset, Lcaseless, F, mb, &slength);
if (rrc != 0)
{
/* Can't use CHECK_PARTIAL because we don't want to update Feptr in
the soft partial matching case. */
if (rrc > 0 && mb->partial != 0 &&
mb->end_subject > mb->start_used_ptr)
{
mb->hitend = TRUE;
if (mb->partial > 1) return PCRE2_ERROR_PARTIAL;
}
break;
}
if (slength != Flength) samelengths = FALSE;
Feptr += slength;
}
/* If the length matched for each repetition is the same as the length of
the captured group, we can easily work backwards. This is the normal
case. However, in caseless UTF-8 mode there are pairs of case-equivalent
characters whose lengths (in terms of code units) differ. However, this
is very rare, so we handle it by re-matching fewer and fewer times. */
if (samelengths)
{
while (Feptr >= Lstart)
{
RMATCH(Fecode, RM21);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
Feptr -= Flength;
}
}
/* The rare case of non-matching lengths. Re-scan the repetition for each
iteration. We know that match_ref() will succeed every time. */
else
{
Lmax = i;
for (;;)
{
RMATCH(Fecode, RM22);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Feptr == Lstart) break; /* Failed after minimal repetition */
Feptr = Lstart;
Lmax--;
for (i = Lmin; i < Lmax; i++)
{
PCRE2_SIZE slength;
(void)match_ref(Loffset, Lcaseless, F, mb, &slength);
Feptr += slength;
}
}
}
RRETURN(MATCH_NOMATCH);
}
/* Control never gets here */
#undef Lcaseless
#undef Lmin
#undef Lmax
#undef Lstart
#undef Loffset
/* ========================================================================= */
/* Opcodes for the start of various parenthesized items */
/* ========================================================================= */
/* In all cases, if the result of RMATCH() is MATCH_THEN, check whether the
(*THEN) is within the current branch by comparing the address of OP_THEN
that is passed back with the end of the branch. If (*THEN) is within the
current branch, and the branch is one of two or more alternatives (it
either starts or ends with OP_ALT), we have reached the limit of THEN's
action, so convert the return code to NOMATCH, which will cause normal
backtracking to happen from now on. Otherwise, THEN is passed back to an
outer alternative. This implements Perl's treatment of parenthesized
groups, where a group not containing | does not affect the current
alternative, that is, (X) is NOT the same as (X|(*F)). */
/* ===================================================================== */
/* BRAZERO, BRAMINZERO and SKIPZERO occur just before a non-possessive
bracket group, indicating that it may occur zero times. It may repeat
infinitely, or not at all - i.e. it could be ()* or ()? or even (){0} in
the pattern. Brackets with fixed upper repeat limits are compiled as a
number of copies, with the optional ones preceded by BRAZERO or BRAMINZERO.
Possessive groups with possible zero repeats are preceded by BRAPOSZERO. */
#define Lnext_ecode F->temp_sptr[0]
case OP_BRAZERO:
Lnext_ecode = Fecode + 1;
RMATCH(Lnext_ecode, RM9);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
do Lnext_ecode += GET(Lnext_ecode, 1); while (*Lnext_ecode == OP_ALT);
Fecode = Lnext_ecode + 1 + LINK_SIZE;
break;
case OP_BRAMINZERO:
Lnext_ecode = Fecode + 1;
do Lnext_ecode += GET(Lnext_ecode, 1); while (*Lnext_ecode == OP_ALT);
RMATCH(Lnext_ecode + 1 + LINK_SIZE, RM10);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
Fecode++;
break;
#undef Lnext_ecode
case OP_SKIPZERO:
Fecode++;
do Fecode += GET(Fecode,1); while (*Fecode == OP_ALT);
Fecode += 1 + LINK_SIZE;
break;
/* ===================================================================== */
/* Handle possessive brackets with an unlimited repeat. The end of these
brackets will always be OP_KETRPOS, which returns MATCH_KETRPOS without
going further in the pattern. */
#define Lframe_type F->temp_32[0]
#define Lmatched_once F->temp_32[1]
#define Lzero_allowed F->temp_32[2]
#define Lstart_eptr F->temp_sptr[0]
#define Lstart_group F->temp_sptr[1]
case OP_BRAPOSZERO:
Lzero_allowed = TRUE; /* Zero repeat is allowed */
Fecode += 1;
if (*Fecode == OP_CBRAPOS || *Fecode == OP_SCBRAPOS)
goto POSSESSIVE_CAPTURE;
goto POSSESSIVE_NON_CAPTURE;
case OP_BRAPOS:
case OP_SBRAPOS:
Lzero_allowed = FALSE; /* Zero repeat not allowed */
POSSESSIVE_NON_CAPTURE:
Lframe_type = GF_NOCAPTURE; /* Remembered frame type */
goto POSSESSIVE_GROUP;
case OP_CBRAPOS:
case OP_SCBRAPOS:
Lzero_allowed = FALSE; /* Zero repeat not allowed */
POSSESSIVE_CAPTURE:
number = GET2(Fecode, 1+LINK_SIZE);
Lframe_type = GF_CAPTURE | number; /* Remembered frame type */
POSSESSIVE_GROUP:
Lmatched_once = FALSE; /* Never matched */
Lstart_group = Fecode; /* Start of this group */
for (;;)
{
Lstart_eptr = Feptr; /* Position at group start */
group_frame_type = Lframe_type;
RMATCH(Fecode + PRIV(OP_lengths)[*Fecode], RM8);
if (rrc == MATCH_KETRPOS)
{
Lmatched_once = TRUE; /* Matched at least once */
if (Feptr == Lstart_eptr) /* Empty match; skip to end */
{
do Fecode += GET(Fecode, 1); while (*Fecode == OP_ALT);
break;
}
Fecode = Lstart_group;
continue;
}
/* See comment above about handling THEN. */
if (rrc == MATCH_THEN)
{
PCRE2_SPTR next_ecode = Fecode + GET(Fecode,1);
if (mb->verb_ecode_ptr < next_ecode &&
(*Fecode == OP_ALT || *next_ecode == OP_ALT))
rrc = MATCH_NOMATCH;
}
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
Fecode += GET(Fecode, 1);
if (*Fecode != OP_ALT) break;
}
/* Success if matched something or zero repeat allowed */
if (Lmatched_once || Lzero_allowed)
{
Fecode += 1 + LINK_SIZE;
break;
}
RRETURN(MATCH_NOMATCH);
#undef Lmatched_once
#undef Lzero_allowed
#undef Lframe_type
#undef Lstart_eptr
#undef Lstart_group
/* ===================================================================== */
/* Handle non-capturing brackets that cannot match an empty string. When we
get to the final alternative within the brackets, as long as there are no
THEN's in the pattern, we can optimize by not recording a new backtracking
point. (Ideally we should test for a THEN within this group, but we don't
have that information.) Don't do this if we are at the very top level,
however, because that would make handling assertions and once-only brackets
messier when there is nothing to go back to. */
#define Lframe_type F->temp_32[0] /* Set for all that use GROUPLOOP */
#define Lnext_branch F->temp_sptr[0] /* Used only in OP_BRA handling */
case OP_BRA:
if (mb->hasthen || Frdepth == 0)
{
Lframe_type = 0;
goto GROUPLOOP;
}
for (;;)
{
Lnext_branch = Fecode + GET(Fecode, 1);
if (*Lnext_branch != OP_ALT) break;
/* This is never the final branch. We do not need to test for MATCH_THEN
here because this code is not used when there is a THEN in the pattern. */
RMATCH(Fecode + PRIV(OP_lengths)[*Fecode], RM1);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
Fecode = Lnext_branch;
}
/* Hit the start of the final branch. Continue at this level. */
Fecode += PRIV(OP_lengths)[*Fecode];
break;
#undef Lnext_branch
/* ===================================================================== */
/* Handle a capturing bracket, other than those that are possessive with an
unlimited repeat. */
case OP_CBRA:
case OP_SCBRA:
Lframe_type = GF_CAPTURE | GET2(Fecode, 1+LINK_SIZE);
goto GROUPLOOP;
/* ===================================================================== */
/* Atomic groups and non-capturing brackets that can match an empty string
must record a backtracking point and also set up a chained frame. */
case OP_ONCE:
case OP_SCRIPT_RUN:
case OP_SBRA:
Lframe_type = GF_NOCAPTURE | Fop;
GROUPLOOP:
for (;;)
{
group_frame_type = Lframe_type;
RMATCH(Fecode + PRIV(OP_lengths)[*Fecode], RM2);
if (rrc == MATCH_THEN)
{
PCRE2_SPTR next_ecode = Fecode + GET(Fecode,1);
if (mb->verb_ecode_ptr < next_ecode &&
(*Fecode == OP_ALT || *next_ecode == OP_ALT))
rrc = MATCH_NOMATCH;
}
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
Fecode += GET(Fecode, 1);
if (*Fecode != OP_ALT) RRETURN(MATCH_NOMATCH);
}
/* Control never reaches here. */
#undef Lframe_type
/* ===================================================================== */
/* Pattern recursion either matches the current regex, or some
subexpression. The offset data is the offset to the starting bracket from
the start of the whole pattern. This is so that it works from duplicated
subpatterns. For a whole-pattern recursion, we have to infer the number
zero. */
#define Lframe_type F->temp_32[0]
#define Lstart_branch F->temp_sptr[0]
case OP_RECURSE:
bracode = mb->start_code + GET(Fecode, 1);
number = (bracode == mb->start_code)? 0 : GET2(bracode, 1 + LINK_SIZE);
/* If we are already in a pattern recursion, check for repeating the same
one without changing the subject pointer or the last referenced character
in the subject. This should catch convoluted mutual recursions; some
simple cases are caught at compile time. However, there are rare cases when
this check needs to be turned off. In this case, actual recursion loops
will be caught by the match or heap limits. */
if (Fcurrent_recurse != RECURSE_UNSET)
{
offset = Flast_group_offset;
while (offset != PCRE2_UNSET)
{
N = (heapframe *)((char *)match_data->heapframes + offset);
P = (heapframe *)((char *)N - frame_size);
if (N->group_frame_type == (GF_RECURSE | number))
{
if (Feptr == P->eptr && mb->last_used_ptr == P->recurse_last_used &&
(mb->moptions & PCRE2_DISABLE_RECURSELOOP_CHECK) == 0)
return PCRE2_ERROR_RECURSELOOP;
break;
}
offset = P->last_group_offset;
}
}
/* Remember the current last referenced character and then run the
recursion branch by branch. */
F->recurse_last_used = mb->last_used_ptr;
Lstart_branch = bracode;
Lframe_type = GF_RECURSE | number;
for (;;)
{
PCRE2_SPTR next_ecode;
group_frame_type = Lframe_type;
RMATCH(Lstart_branch + PRIV(OP_lengths)[*Lstart_branch], RM11);
next_ecode = Lstart_branch + GET(Lstart_branch,1);
/* Handle backtracking verbs, which are defined in a range that can
easily be tested for. PCRE does not allow THEN, SKIP, PRUNE or COMMIT to
escape beyond a recursion; they cause a NOMATCH for the entire recursion.
When one of these verbs triggers, the current recursion group number is
recorded. If it matches the recursion we are processing, the verb
happened within the recursion and we must deal with it. Otherwise it must
have happened after the recursion completed, and so has to be passed
back. See comment above about handling THEN. */
if (rrc >= MATCH_BACKTRACK_MIN && rrc <= MATCH_BACKTRACK_MAX &&
mb->verb_current_recurse == (Lframe_type ^ GF_RECURSE))
{
if (rrc == MATCH_THEN && mb->verb_ecode_ptr < next_ecode &&
(*Lstart_branch == OP_ALT || *next_ecode == OP_ALT))
rrc = MATCH_NOMATCH;
else RRETURN(MATCH_NOMATCH);
}
/* Note that carrying on after (*ACCEPT) in a recursion is handled in the
OP_ACCEPT code. Nothing needs to be done here. */
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
Lstart_branch = next_ecode;
if (*Lstart_branch != OP_ALT) RRETURN(MATCH_NOMATCH);
}
/* Control never reaches here. */
#undef Lframe_type
#undef Lstart_branch
/* ===================================================================== */
/* Positive assertions are like other groups except that PCRE doesn't allow
the effect of (*THEN) to escape beyond an assertion; it is therefore
treated as NOMATCH. (*ACCEPT) is treated as successful assertion, with its
captures and mark retained. Any other return is an error. */
#define Lframe_type F->temp_32[0]
case OP_ASSERT:
case OP_ASSERTBACK:
case OP_ASSERT_NA:
case OP_ASSERTBACK_NA:
Lframe_type = GF_NOCAPTURE | Fop;
for (;;)
{
group_frame_type = Lframe_type;
RMATCH(Fecode + PRIV(OP_lengths)[*Fecode], RM3);
if (rrc == MATCH_ACCEPT)
{
memcpy(Fovector,
(char *)assert_accept_frame + offsetof(heapframe, ovector),
assert_accept_frame->offset_top * sizeof(PCRE2_SIZE));
Foffset_top = assert_accept_frame->offset_top;
Fmark = assert_accept_frame->mark;
break;
}
if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
Fecode += GET(Fecode, 1);
if (*Fecode != OP_ALT) RRETURN(MATCH_NOMATCH);
}
do Fecode += GET(Fecode, 1); while (*Fecode == OP_ALT);
Fecode += 1 + LINK_SIZE;
break;
#undef Lframe_type
/* ===================================================================== */
/* Handle negative assertions. Loop for each non-matching branch as for
positive assertions. */
#define Lframe_type F->temp_32[0]
case OP_ASSERT_NOT:
case OP_ASSERTBACK_NOT:
Lframe_type = GF_NOCAPTURE | Fop;
for (;;)
{
group_frame_type = Lframe_type;
RMATCH(Fecode + PRIV(OP_lengths)[*Fecode], RM4);
switch(rrc)
{
case MATCH_ACCEPT: /* Assertion matched, therefore it fails. */
case MATCH_MATCH:
RRETURN (MATCH_NOMATCH);
case MATCH_NOMATCH: /* Branch failed, try next if present. */
case MATCH_THEN:
Fecode += GET(Fecode, 1);
if (*Fecode != OP_ALT) goto ASSERT_NOT_FAILED;
break;
case MATCH_COMMIT: /* Assertion forced to fail, therefore continue. */
case MATCH_SKIP:
case MATCH_PRUNE:
do Fecode += GET(Fecode, 1); while (*Fecode == OP_ALT);
goto ASSERT_NOT_FAILED;
default: /* Pass back any other return */
RRETURN(rrc);
}
}
/* None of the branches have matched or there was a backtrack to (*COMMIT),
(*SKIP), (*PRUNE), or (*THEN) in the last branch. This is success for a
negative assertion, so carry on. */
ASSERT_NOT_FAILED:
Fecode += 1 + LINK_SIZE;
break;
#undef Lframe_type
/* ===================================================================== */
/* The callout item calls an external function, if one is provided, passing
details of the match so far. This is mainly for debugging, though the
function is able to force a failure. */
case OP_CALLOUT:
case OP_CALLOUT_STR:
rrc = do_callout(F, mb, &length);
if (rrc > 0) RRETURN(MATCH_NOMATCH);
if (rrc < 0) RRETURN(rrc);
Fecode += length;
break;
/* ===================================================================== */
/* Conditional group: compilation checked that there are no more than two
branches. If the condition is false, skipping the first branch takes us
past the end of the item if there is only one branch, but that's exactly
what we want. */
case OP_COND:
case OP_SCOND:
/* The variable Flength will be added to Fecode when the condition is
false, to get to the second branch. Setting it to the offset to the ALT or
KET, then incrementing Fecode achieves this effect. However, if the second
branch is non-existent, we must point to the KET so that the end of the
group is correctly processed. We now have Fecode pointing to the condition
or callout. */
Flength = GET(Fecode, 1); /* Offset to the second branch */
if (Fecode[Flength] != OP_ALT) Flength -= 1 + LINK_SIZE;
Fecode += 1 + LINK_SIZE; /* From this opcode */
/* Because of the way auto-callout works during compile, a callout item is
inserted between OP_COND and an assertion condition. Such a callout can
also be inserted manually. */
if (*Fecode == OP_CALLOUT || *Fecode == OP_CALLOUT_STR)
{
rrc = do_callout(F, mb, &length);
if (rrc > 0) RRETURN(MATCH_NOMATCH);
if (rrc < 0) RRETURN(rrc);
/* Advance Fecode past the callout, so it now points to the condition. We
must adjust Flength so that the value of Fecode+Flength is unchanged. */
Fecode += length;
Flength -= length;
}
/* Test the various possible conditions */
condition = FALSE;
switch(*Fecode)
{
case OP_RREF: /* Group recursion test */
if (Fcurrent_recurse != RECURSE_UNSET)
{
number = GET2(Fecode, 1);
condition = (number == RREF_ANY || number == Fcurrent_recurse);
}
break;
case OP_DNRREF: /* Duplicate named group recursion test */
if (Fcurrent_recurse != RECURSE_UNSET)
{
int count = GET2(Fecode, 1 + IMM2_SIZE);
PCRE2_SPTR slot = mb->name_table + GET2(Fecode, 1) * mb->name_entry_size;
while (count-- > 0)
{
number = GET2(slot, 0);
condition = number == Fcurrent_recurse;
if (condition) break;
slot += mb->name_entry_size;
}
}
break;
case OP_CREF: /* Numbered group used test */
offset = (GET2(Fecode, 1) << 1) - 2; /* Doubled ref number */
condition = offset < Foffset_top && Fovector[offset] != PCRE2_UNSET;
break;
case OP_DNCREF: /* Duplicate named group used test */
{
int count = GET2(Fecode, 1 + IMM2_SIZE);
PCRE2_SPTR slot = mb->name_table + GET2(Fecode, 1) * mb->name_entry_size;
while (count-- > 0)
{
offset = (GET2(slot, 0) << 1) - 2;
condition = offset < Foffset_top && Fovector[offset] != PCRE2_UNSET;
if (condition) break;
slot += mb->name_entry_size;
}
}
break;
case OP_FALSE:
case OP_FAIL: /* The assertion (?!) becomes OP_FAIL */
break;
case OP_TRUE:
condition = TRUE;
break;
/* The condition is an assertion. Run code similar to the assertion code
above. */
#define Lpositive F->temp_32[0]
#define Lstart_branch F->temp_sptr[0]
default:
Lpositive = (*Fecode == OP_ASSERT || *Fecode == OP_ASSERTBACK);
Lstart_branch = Fecode;
for (;;)
{
group_frame_type = GF_CONDASSERT | *Fecode;
RMATCH(Lstart_branch + PRIV(OP_lengths)[*Lstart_branch], RM5);
switch(rrc)
{
case MATCH_ACCEPT: /* Save captures */
memcpy(Fovector,
(char *)assert_accept_frame + offsetof(heapframe, ovector),
assert_accept_frame->offset_top * sizeof(PCRE2_SIZE));
Foffset_top = assert_accept_frame->offset_top;
/* Fall through */
/* In the case of a match, the captures have already been put into
the current frame. */
case MATCH_MATCH:
condition = Lpositive; /* TRUE for positive assertion */
break;
/* PCRE doesn't allow the effect of (*THEN) to escape beyond an
assertion; it is therefore always treated as NOMATCH. */
case MATCH_NOMATCH:
case MATCH_THEN:
Lstart_branch += GET(Lstart_branch, 1);
if (*Lstart_branch == OP_ALT) continue; /* Try next branch */
condition = !Lpositive; /* TRUE for negative assertion */
break;
/* These force no match without checking other branches. */
case MATCH_COMMIT:
case MATCH_SKIP:
case MATCH_PRUNE:
condition = !Lpositive;
break;
default:
RRETURN(rrc);
}
break; /* Out of the branch loop */
}
/* If the condition is true, find the end of the assertion so that
advancing past it gets us to the start of the first branch. */
if (condition)
{
do Fecode += GET(Fecode, 1); while (*Fecode == OP_ALT);
}
break; /* End of assertion condition */
}
#undef Lpositive
#undef Lstart_branch
/* Choose branch according to the condition. */
Fecode += condition? PRIV(OP_lengths)[*Fecode] : Flength;
/* If the opcode is OP_SCOND it means we are at a repeated conditional
group that might match an empty string. We must therefore descend a level
so that the start is remembered for checking. For OP_COND we can just
continue at this level. */
if (Fop == OP_SCOND)
{
group_frame_type = GF_NOCAPTURE | Fop;
RMATCH(Fecode, RM35);
RRETURN(rrc);
}
break;
/* ========================================================================= */
/* End of start of parenthesis opcodes */
/* ========================================================================= */
/* ===================================================================== */
/* Move the subject pointer back by one fixed amount. This occurs at the
start of each branch that has a fixed length in a lookbehind assertion. If
we are too close to the start to move back, fail. When working with UTF-8
we move back a number of characters, not bytes. */
case OP_REVERSE:
number = GET2(Fecode, 1);
#ifdef SUPPORT_UNICODE
if (utf)
{
while (number-- > 0)
{
if (Feptr <= mb->check_subject) RRETURN(MATCH_NOMATCH);
Feptr--;
BACKCHAR(Feptr);
}
}
else
#endif
/* No UTF support, or not in UTF mode: count is code unit count */
{
if ((ptrdiff_t)number > Feptr - mb->start_subject) RRETURN(MATCH_NOMATCH);
Feptr -= number;
}
/* Save the earliest consulted character, then skip to next opcode */
if (Feptr < mb->start_used_ptr) mb->start_used_ptr = Feptr;
Fecode += 1 + IMM2_SIZE;
break;
/* ===================================================================== */
/* Move the subject pointer back by a variable amount. This occurs at the
start of each branch of a lookbehind assertion when the branch has a
variable, but limited, length. A loop is needed to try matching the branch
after moving back different numbers of characters. If we are too close to
the start to move back even the minimum amount, fail. When working with
UTF-8 we move back a number of characters, not bytes. */
#define Lmin F->temp_32[0]
#define Lmax F->temp_32[1]
#define Leptr F->temp_sptr[0]
case OP_VREVERSE:
Lmin = GET2(Fecode, 1);
Lmax = GET2(Fecode, 1 + IMM2_SIZE);
Leptr = Feptr;
/* Move back by the maximum branch length and then work forwards. This
ensures that items such as \d{3,5} get the maximum length, which is
relevant for captures, and makes for Perl compatibility. */
#ifdef SUPPORT_UNICODE
if (utf)
{
for (i = 0; i < Lmax; i++)
{
if (Feptr == mb->start_subject)
{
if (i < Lmin) RRETURN(MATCH_NOMATCH);
Lmax = i;
break;
}
Feptr--;
BACKCHAR(Feptr);
}
}
else
#endif
/* No UTF support or not in UTF mode */
{
ptrdiff_t diff = Feptr - mb->start_subject;
uint32_t available = (diff > 65535)? 65535 : ((diff > 0)? (int)diff : 0);
if (Lmin > available) RRETURN(MATCH_NOMATCH);
if (Lmax > available) Lmax = available;
Feptr -= Lmax;
}
/* Now try matching, moving forward one character on failure, until we
reach the mimimum back length. */
for (;;)
{
RMATCH(Fecode + 1 + 2 * IMM2_SIZE, RM37);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
if (Lmax-- <= Lmin) RRETURN(MATCH_NOMATCH);
Feptr++;
#ifdef SUPPORT_UNICODE
if (utf) { FORWARDCHARTEST(Feptr, mb->end_subject); }
#endif
}
/* Control never reaches here */
#undef Lmin
#undef Lmax
#undef Leptr
/* ===================================================================== */
/* An alternation is the end of a branch; scan along to find the end of the
bracketed group. */
case OP_ALT:
branch_end = Fecode;
do Fecode += GET(Fecode,1); while (*Fecode == OP_ALT);
break;
/* ===================================================================== */
/* The end of a parenthesized group. For all but OP_BRA and OP_COND, the
starting frame was added to the chained frames in order to remember the
starting subject position for the group. (Not true for OP_BRA when it's a
whole pattern recursion, but that is handled separately below.)*/
case OP_KET:
case OP_KETRMIN:
case OP_KETRMAX:
case OP_KETRPOS:
bracode = Fecode - GET(Fecode, 1);
if (branch_end == NULL) branch_end = Fecode;
branch_start = bracode;
while (branch_start + GET(branch_start, 1) != branch_end)
branch_start += GET(branch_start, 1);
branch_end = NULL;
/* Point N to the frame at the start of the most recent group, and P to its
predecessor. Remember the subject pointer at the start of the group. */
if (*bracode != OP_BRA && *bracode != OP_COND)
{
N = (heapframe *)((char *)match_data->heapframes + Flast_group_offset);
P = (heapframe *)((char *)N - frame_size);
Flast_group_offset = P->last_group_offset;
#ifdef DEBUG_SHOW_RMATCH
fprintf(stderr, "++ KET for frame=%d type=%x prev char offset=%lu\n",
N->rdepth, N->group_frame_type,
(char *)P->eptr - (char *)mb->start_subject);
#endif
/* If we are at the end of an assertion that is a condition, return a
match, discarding any intermediate backtracking points. Copy back the
mark setting and the captures into the frame before N so that they are
set on return. Doing this for all assertions, both positive and negative,
seems to match what Perl does. */
if (GF_IDMASK(N->group_frame_type) == GF_CONDASSERT)
{
memcpy((char *)P + offsetof(heapframe, ovector), Fovector,
Foffset_top * sizeof(PCRE2_SIZE));
P->offset_top = Foffset_top;
P->mark = Fmark;
Fback_frame = (char *)F - (char *)P;
RRETURN(MATCH_MATCH);
}
}
else P = NULL; /* Indicates starting frame not recorded */
/* The group was not a conditional assertion. */
switch (*bracode)
{
/* Whole pattern recursion is handled as a recursion into group 0, but
the entire pattern is wrapped in OP_BRA/OP_KET rather than a capturing
group - a design mistake: it should perhaps have been capture group 0.
Anyway, that means the end of such recursion must be handled here. It is
detected by checking for an immediately following OP_END when we are
recursing in group 0. If this is not the end of a whole-pattern
recursion, there is nothing to be done. */
case OP_BRA:
if (Fcurrent_recurse != 0 || Fecode[1+LINK_SIZE] != OP_END) break;
/* It is the end of whole-pattern recursion. */
offset = Flast_group_offset;
if (offset == PCRE2_UNSET) return PCRE2_ERROR_INTERNAL;
N = (heapframe *)((char *)match_data->heapframes + offset);
P = (heapframe *)((char *)N - frame_size);
Flast_group_offset = P->last_group_offset;
/* Reinstate the previous set of captures and then carry on after the
recursion call. */
memcpy((char *)F + offsetof(heapframe, ovector), P->ovector,
Foffset_top * sizeof(PCRE2_SIZE));
Foffset_top = P->offset_top;
Fcapture_last = P->capture_last;
Fcurrent_recurse = P->current_recurse;
Fecode = P->ecode + 1 + LINK_SIZE;
continue; /* With next opcode */
case OP_COND: /* No need to do anything for these */
case OP_SCOND:
break;
/* Non-atomic positive assertions are like OP_BRA, except that the
subject pointer must be put back to where it was at the start of the
assertion. For a variable lookbehind, check its end point. */
case OP_ASSERTBACK_NA:
if (branch_start[1 + LINK_SIZE] == OP_VREVERSE && Feptr != P->eptr)
RRETURN(MATCH_NOMATCH);
/* Fall through */
case OP_ASSERT_NA:
if (Feptr > mb->last_used_ptr) mb->last_used_ptr = Feptr;
Feptr = P->eptr;
break;
/* Atomic positive assertions are like OP_ONCE, except that in addition
the subject pointer must be put back to where it was at the start of the
assertion. For a variable lookbehind, check its end point. */
case OP_ASSERTBACK:
if (branch_start[1 + LINK_SIZE] == OP_VREVERSE && Feptr != P->eptr)
RRETURN(MATCH_NOMATCH);
/* Fall through */
case OP_ASSERT:
if (Feptr > mb->last_used_ptr) mb->last_used_ptr = Feptr;
Feptr = P->eptr;
/* Fall through */
/* For an atomic group, discard internal backtracking points. We must
also ensure that any remaining branches within the top-level of the group
are not tried. Do this by adjusting the code pointer within the backtrack
frame so that it points to the final branch. */
case OP_ONCE:
Fback_frame = ((char *)F - (char *)P);
for (;;)
{
uint32_t y = GET(P->ecode,1);
if ((P->ecode)[y] != OP_ALT) break;
P->ecode += y;
}
break;
/* A matching negative assertion returns MATCH, which is turned into
NOMATCH at the assertion level. For a variable lookbehind, check its end
point. */
case OP_ASSERTBACK_NOT:
if (branch_start[1 + LINK_SIZE] == OP_VREVERSE && Feptr != P->eptr)
RRETURN(MATCH_NOMATCH);
/* Fall through */
case OP_ASSERT_NOT:
RRETURN(MATCH_MATCH);
/* At the end of a script run, apply the script-checking rules. This code
will never by exercised if Unicode support it not compiled, because in
that environment script runs cause an error at compile time. */
case OP_SCRIPT_RUN:
if (!PRIV(script_run)(P->eptr, Feptr, utf)) RRETURN(MATCH_NOMATCH);
break;
/* Whole-pattern recursion is coded as a recurse into group 0, and is
handled with OP_BRA above. Other recursion is handled here. */
case OP_CBRA:
case OP_CBRAPOS:
case OP_SCBRA:
case OP_SCBRAPOS:
number = GET2(bracode, 1+LINK_SIZE);
/* Handle a recursively called group. We reinstate the previous set of
captures and then carry on after the recursion call. */
if (Fcurrent_recurse == number)
{
P = (heapframe *)((char *)N - frame_size);
memcpy((char *)F + offsetof(heapframe, ovector), P->ovector,
Foffset_top * sizeof(PCRE2_SIZE));
Foffset_top = P->offset_top;
Fcapture_last = P->capture_last;
Fcurrent_recurse = P->current_recurse;
Fecode = P->ecode + 1 + LINK_SIZE;
continue; /* With next opcode */
}
/* Deal with actual capturing. */
offset = (number << 1) - 2;
Fcapture_last = number;
Fovector[offset] = P->eptr - mb->start_subject;
Fovector[offset+1] = Feptr - mb->start_subject;
if (offset >= Foffset_top) Foffset_top = offset + 2;
break;
} /* End actions relating to the starting opcode */
/* OP_KETRPOS is a possessive repeating ket. Remember the current position,
and return the MATCH_KETRPOS. This makes it possible to do the repeats one
at a time from the outer level. This must precede the empty string test -
in this case that test is done at the outer level. */
if (*Fecode == OP_KETRPOS)
{
memcpy((char *)P + offsetof(heapframe, eptr),
(char *)F + offsetof(heapframe, eptr),
frame_copy_size);
RRETURN(MATCH_KETRPOS);
}
/* Handle the different kinds of closing brackets. A non-repeating ket
needs no special action, just continuing at this level. This also happens
for the repeating kets if the group matched no characters, in order to
forcibly break infinite loops. Otherwise, the repeating kets try the rest
of the pattern or restart from the preceding bracket, in the appropriate
order. */
if (Fop != OP_KET && (P == NULL || Feptr != P->eptr))
{
if (Fop == OP_KETRMIN)
{
RMATCH(Fecode + 1 + LINK_SIZE, RM6);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
Fecode -= GET(Fecode, 1);
break; /* End of ket processing */
}
/* Repeat the maximum number of times (KETRMAX) */
RMATCH(bracode, RM7);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
}
/* Carry on at this level for a non-repeating ket, or after matching an
empty string, or after repeating for a maximum number of times. */
Fecode += 1 + LINK_SIZE;
break;
/* ===================================================================== */
/* Start and end of line assertions, not multiline mode. */
case OP_CIRC: /* Start of line, unless PCRE2_NOTBOL is set. */
if (Feptr != mb->start_subject || (mb->moptions & PCRE2_NOTBOL) != 0)
RRETURN(MATCH_NOMATCH);
Fecode++;
break;
case OP_SOD: /* Unconditional start of subject */
if (Feptr != mb->start_subject) RRETURN(MATCH_NOMATCH);
Fecode++;
break;
/* When PCRE2_NOTEOL is unset, assert before the subject end, or a
terminating newline unless PCRE2_DOLLAR_ENDONLY is set. */
case OP_DOLL:
if ((mb->moptions & PCRE2_NOTEOL) != 0) RRETURN(MATCH_NOMATCH);
if ((mb->poptions & PCRE2_DOLLAR_ENDONLY) == 0) goto ASSERT_NL_OR_EOS;
/* Fall through */
/* Unconditional end of subject assertion (\z). */
case OP_EOD:
if (Feptr < mb->true_end_subject) RRETURN(MATCH_NOMATCH);
if (mb->partial != 0)
{
mb->hitend = TRUE;
if (mb->partial > 1) return PCRE2_ERROR_PARTIAL;
}
Fecode++;
break;
/* End of subject or ending \n assertion (\Z) */
case OP_EODN:
ASSERT_NL_OR_EOS:
if (Feptr < mb->end_subject &&
(!IS_NEWLINE(Feptr) || Feptr != mb->end_subject - mb->nllen))
{
if (mb->partial != 0 &&
Feptr + 1 >= mb->end_subject &&
NLBLOCK->nltype == NLTYPE_FIXED &&
NLBLOCK->nllen == 2 &&
UCHAR21TEST(Feptr) == NLBLOCK->nl[0])
{
mb->hitend = TRUE;
if (mb->partial > 1) return PCRE2_ERROR_PARTIAL;
}
RRETURN(MATCH_NOMATCH);
}
/* Either at end of string or \n before end. */
if (mb->partial != 0)
{
mb->hitend = TRUE;
if (mb->partial > 1) return PCRE2_ERROR_PARTIAL;
}
Fecode++;
break;
/* ===================================================================== */
/* Start and end of line assertions, multiline mode. */
/* Start of subject unless notbol, or after any newline except for one at
the very end, unless PCRE2_ALT_CIRCUMFLEX is set. */
case OP_CIRCM:
if ((mb->moptions & PCRE2_NOTBOL) != 0 && Feptr == mb->start_subject)
RRETURN(MATCH_NOMATCH);
if (Feptr != mb->start_subject &&
((Feptr == mb->end_subject &&
(mb->poptions & PCRE2_ALT_CIRCUMFLEX) == 0) ||
!WAS_NEWLINE(Feptr)))
RRETURN(MATCH_NOMATCH);
Fecode++;
break;
/* Assert before any newline, or before end of subject unless noteol is
set. */
case OP_DOLLM:
if (Feptr < mb->end_subject)
{
if (!IS_NEWLINE(Feptr))
{
if (mb->partial != 0 &&
Feptr + 1 >= mb->end_subject &&
NLBLOCK->nltype == NLTYPE_FIXED &&
NLBLOCK->nllen == 2 &&
UCHAR21TEST(Feptr) == NLBLOCK->nl[0])
{
mb->hitend = TRUE;
if (mb->partial > 1) return PCRE2_ERROR_PARTIAL;
}
RRETURN(MATCH_NOMATCH);
}
}
else
{
if ((mb->moptions & PCRE2_NOTEOL) != 0) RRETURN(MATCH_NOMATCH);
SCHECK_PARTIAL();
}
Fecode++;
break;
/* ===================================================================== */
/* Start of match assertion */
case OP_SOM:
if (Feptr != mb->start_subject + mb->start_offset) RRETURN(MATCH_NOMATCH);
Fecode++;
break;
/* ===================================================================== */
/* Reset the start of match point */
case OP_SET_SOM:
Fstart_match = Feptr;
Fecode++;
break;
/* ===================================================================== */
/* Word boundary assertions. Find out if the previous and current
characters are "word" characters. It takes a bit more work in UTF mode.
Characters > 255 are assumed to be "non-word" characters when PCRE2_UCP is
not set. When it is set, use Unicode properties if available, even when not
in UTF mode. Remember the earliest and latest consulted characters. */
case OP_NOT_WORD_BOUNDARY:
case OP_WORD_BOUNDARY:
case OP_NOT_UCP_WORD_BOUNDARY:
case OP_UCP_WORD_BOUNDARY:
if (Feptr == mb->check_subject) prev_is_word = FALSE; else
{
PCRE2_SPTR lastptr = Feptr - 1;
#ifdef SUPPORT_UNICODE
if (utf)
{
BACKCHAR(lastptr);
GETCHAR(fc, lastptr);
}
else
#endif /* SUPPORT_UNICODE */
fc = *lastptr;
if (lastptr < mb->start_used_ptr) mb->start_used_ptr = lastptr;
#ifdef SUPPORT_UNICODE
if (Fop == OP_UCP_WORD_BOUNDARY || Fop == OP_NOT_UCP_WORD_BOUNDARY)
{
int chartype = UCD_CHARTYPE(fc);
int category = PRIV(ucp_gentype)[chartype];
prev_is_word = (category == ucp_L || category == ucp_N ||
chartype == ucp_Mn || chartype == ucp_Pc);
}
else
#endif /* SUPPORT_UNICODE */
prev_is_word = CHMAX_255(fc) && (mb->ctypes[fc] & ctype_word) != 0;
}
/* Get status of next character */
if (Feptr >= mb->end_subject)
{
SCHECK_PARTIAL();
cur_is_word = FALSE;
}
else
{
PCRE2_SPTR nextptr = Feptr + 1;
#ifdef SUPPORT_UNICODE
if (utf)
{
FORWARDCHARTEST(nextptr, mb->end_subject);
GETCHAR(fc, Feptr);
}
else
#endif /* SUPPORT_UNICODE */
fc = *Feptr;
if (nextptr > mb->last_used_ptr) mb->last_used_ptr = nextptr;
#ifdef SUPPORT_UNICODE
if (Fop == OP_UCP_WORD_BOUNDARY || Fop == OP_NOT_UCP_WORD_BOUNDARY)
{
int chartype = UCD_CHARTYPE(fc);
int category = PRIV(ucp_gentype)[chartype];
cur_is_word = (category == ucp_L || category == ucp_N ||
chartype == ucp_Mn || chartype == ucp_Pc);
}
else
#endif /* SUPPORT_UNICODE */
cur_is_word = CHMAX_255(fc) && (mb->ctypes[fc] & ctype_word) != 0;
}
/* Now see if the situation is what we want */
if ((*Fecode++ == OP_WORD_BOUNDARY || Fop == OP_UCP_WORD_BOUNDARY)?
cur_is_word == prev_is_word : cur_is_word != prev_is_word)
RRETURN(MATCH_NOMATCH);
break;
/* ===================================================================== */
/* Backtracking (*VERB)s, with and without arguments. Note that if the
pattern is successfully matched, we do not come back from RMATCH. */
case OP_MARK:
Fmark = mb->nomatch_mark = Fecode + 2;
RMATCH(Fecode + PRIV(OP_lengths)[*Fecode] + Fecode[1], RM12);
/* A return of MATCH_SKIP_ARG means that matching failed at SKIP with an
argument, and we must check whether that argument matches this MARK's
argument. It is passed back in mb->verb_skip_ptr. If it does match, we
return MATCH_SKIP with mb->verb_skip_ptr now pointing to the subject
position that corresponds to this mark. Otherwise, pass back the return
code unaltered. */
if (rrc == MATCH_SKIP_ARG &&
PRIV(strcmp)(Fecode + 2, mb->verb_skip_ptr) == 0)
{
mb->verb_skip_ptr = Feptr; /* Pass back current position */
RRETURN(MATCH_SKIP);
}
RRETURN(rrc);
case OP_FAIL:
RRETURN(MATCH_NOMATCH);
/* Record the current recursing group number in mb->verb_current_recurse
when a backtracking return such as MATCH_COMMIT is given. This enables the
recurse processing to catch verbs from within the recursion. */
case OP_COMMIT:
RMATCH(Fecode + PRIV(OP_lengths)[*Fecode], RM13);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
mb->verb_current_recurse = Fcurrent_recurse;
RRETURN(MATCH_COMMIT);
case OP_COMMIT_ARG:
Fmark = mb->nomatch_mark = Fecode + 2;
RMATCH(Fecode + PRIV(OP_lengths)[*Fecode] + Fecode[1], RM36);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
mb->verb_current_recurse = Fcurrent_recurse;
RRETURN(MATCH_COMMIT);
case OP_PRUNE:
RMATCH(Fecode + PRIV(OP_lengths)[*Fecode], RM14);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
mb->verb_current_recurse = Fcurrent_recurse;
RRETURN(MATCH_PRUNE);
case OP_PRUNE_ARG:
Fmark = mb->nomatch_mark = Fecode + 2;
RMATCH(Fecode + PRIV(OP_lengths)[*Fecode] + Fecode[1], RM15);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
mb->verb_current_recurse = Fcurrent_recurse;
RRETURN(MATCH_PRUNE);
case OP_SKIP:
RMATCH(Fecode + PRIV(OP_lengths)[*Fecode], RM16);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
mb->verb_skip_ptr = Feptr; /* Pass back current position */
mb->verb_current_recurse = Fcurrent_recurse;
RRETURN(MATCH_SKIP);
/* Note that, for Perl compatibility, SKIP with an argument does NOT set
nomatch_mark. When a pattern match ends with a SKIP_ARG for which there was
not a matching mark, we have to re-run the match, ignoring the SKIP_ARG
that failed and any that precede it (either they also failed, or were not
triggered). To do this, we maintain a count of executed SKIP_ARGs. If a
SKIP_ARG gets to top level, the match is re-run with mb->ignore_skip_arg
set to the count of the one that failed. */
case OP_SKIP_ARG:
mb->skip_arg_count++;
if (mb->skip_arg_count <= mb->ignore_skip_arg)
{
Fecode += PRIV(OP_lengths)[*Fecode] + Fecode[1];
break;
}
RMATCH(Fecode + PRIV(OP_lengths)[*Fecode] + Fecode[1], RM17);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
/* Pass back the current skip name and return the special MATCH_SKIP_ARG
return code. This will either be caught by a matching MARK, or get to the
top, where it causes a rematch with mb->ignore_skip_arg set to the value of
mb->skip_arg_count. */
mb->verb_skip_ptr = Fecode + 2;
mb->verb_current_recurse = Fcurrent_recurse;
RRETURN(MATCH_SKIP_ARG);
/* For THEN (and THEN_ARG) we pass back the address of the opcode, so that
the branch in which it occurs can be determined. */
case OP_THEN:
RMATCH(Fecode + PRIV(OP_lengths)[*Fecode], RM18);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
mb->verb_ecode_ptr = Fecode;
mb->verb_current_recurse = Fcurrent_recurse;
RRETURN(MATCH_THEN);
case OP_THEN_ARG:
Fmark = mb->nomatch_mark = Fecode + 2;
RMATCH(Fecode + PRIV(OP_lengths)[*Fecode] + Fecode[1], RM19);
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
mb->verb_ecode_ptr = Fecode;
mb->verb_current_recurse = Fcurrent_recurse;
RRETURN(MATCH_THEN);
/* ===================================================================== */
/* There's been some horrible disaster. Arrival here can only mean there is
something seriously wrong in the code above or the OP_xxx definitions. */
default:
return PCRE2_ERROR_INTERNAL;
}
/* Do not insert any code in here without much thought; it is assumed
that "continue" in the code above comes out to here to repeat the main
loop. */
} /* End of main loop */
/* Control never reaches here */
/* ========================================================================= */
/* The RRETURN() macro jumps here. The number that is saved in Freturn_id
indicates which label we actually want to return to. The value in Frdepth is
the index number of the frame in the vector. The return value has been placed
in rrc. */
#define LBL(val) case val: goto L_RM##val;
RETURN_SWITCH:
if (Feptr > mb->last_used_ptr) mb->last_used_ptr = Feptr;
if (Frdepth == 0) return rrc; /* Exit from the top level */
F = (heapframe *)((char *)F - Fback_frame); /* Backtrack */
mb->cb->callout_flags |= PCRE2_CALLOUT_BACKTRACK; /* Note for callouts */
#ifdef DEBUG_SHOW_RMATCH
fprintf(stderr, "++ RETURN %d to RM%d\n", rrc, Freturn_id);
#endif
switch (Freturn_id)
{
LBL( 1) LBL( 2) LBL( 3) LBL( 4) LBL( 5) LBL( 6) LBL( 7) LBL( 8)
LBL( 9) LBL(10) LBL(11) LBL(12) LBL(13) LBL(14) LBL(15) LBL(16)
LBL(17) LBL(18) LBL(19) LBL(20) LBL(21) LBL(22) LBL(23) LBL(24)
LBL(25) LBL(26) LBL(27) LBL(28) LBL(29) LBL(30) LBL(31) LBL(32)
LBL(33) LBL(34) LBL(35) LBL(36) LBL(37)
#ifdef SUPPORT_WIDE_CHARS
LBL(100) LBL(101)
#endif
#ifdef SUPPORT_UNICODE
LBL(200) LBL(201) LBL(202) LBL(203) LBL(204) LBL(205) LBL(206)
LBL(207) LBL(208) LBL(209) LBL(210) LBL(211) LBL(212) LBL(213)
LBL(214) LBL(215) LBL(216) LBL(217) LBL(218) LBL(219) LBL(220)
LBL(221) LBL(222) LBL(223) LBL(224) LBL(225)
#endif
default:
return PCRE2_ERROR_INTERNAL;
}
#undef LBL
}
/*************************************************
* Match a Regular Expression *
*************************************************/
/* This function applies a compiled pattern to a subject string and picks out
portions of the string if it matches. Two elements in the vector are set for
each substring: the offsets to the start and end of the substring.
Arguments:
code points to the compiled expression
subject points to the subject string
length length of subject string (may contain binary zeros)
start_offset where to start in the subject string
options option bits
match_data points to a match_data block
mcontext points a PCRE2 context
Returns: > 0 => success; value is the number of ovector pairs filled
= 0 => success, but ovector is not big enough
= -1 => failed to match (PCRE2_ERROR_NOMATCH)
= -2 => partial match (PCRE2_ERROR_PARTIAL)
< -2 => some kind of unexpected problem
*/
PCRE2_EXP_DEFN int PCRE2_CALL_CONVENTION
pcre2_match(const pcre2_code *code, PCRE2_SPTR subject, PCRE2_SIZE length,
PCRE2_SIZE start_offset, uint32_t options, pcre2_match_data *match_data,
pcre2_match_context *mcontext)
{
int rc;
int was_zero_terminated = 0;
const uint8_t *start_bits = NULL;
const pcre2_real_code *re = (const pcre2_real_code *)code;
BOOL anchored;
BOOL firstline;
BOOL has_first_cu = FALSE;
BOOL has_req_cu = FALSE;
BOOL startline;
#if PCRE2_CODE_UNIT_WIDTH == 8
PCRE2_SPTR memchr_found_first_cu;
PCRE2_SPTR memchr_found_first_cu2;
#endif
PCRE2_UCHAR first_cu = 0;
PCRE2_UCHAR first_cu2 = 0;
PCRE2_UCHAR req_cu = 0;
PCRE2_UCHAR req_cu2 = 0;
PCRE2_SPTR bumpalong_limit;
PCRE2_SPTR end_subject;
PCRE2_SPTR true_end_subject;
PCRE2_SPTR start_match;
PCRE2_SPTR req_cu_ptr;
PCRE2_SPTR start_partial;
PCRE2_SPTR match_partial;
#ifdef SUPPORT_JIT
BOOL use_jit;
#endif
/* This flag is needed even when Unicode is not supported for convenience
(it is used by the IS_NEWLINE macro). */
BOOL utf = FALSE;
#ifdef SUPPORT_UNICODE
BOOL ucp = FALSE;
BOOL allow_invalid;
uint32_t fragment_options = 0;
#ifdef SUPPORT_JIT
BOOL jit_checked_utf = FALSE;
#endif
#endif /* SUPPORT_UNICODE */
PCRE2_SIZE frame_size;
PCRE2_SIZE heapframes_size;
/* We need to have mb as a pointer to a match block, because the IS_NEWLINE
macro is used below, and it expects NLBLOCK to be defined as a pointer. */
pcre2_callout_block cb;
match_block actual_match_block;
match_block *mb = &actual_match_block;
/* Recognize NULL, length 0 as an empty string. */
if (subject == NULL && length == 0) subject = (PCRE2_SPTR)"";
/* Plausibility checks */
if ((options & ~PUBLIC_MATCH_OPTIONS) != 0) return PCRE2_ERROR_BADOPTION;
if (code == NULL || subject == NULL || match_data == NULL)
return PCRE2_ERROR_NULL;
start_match = subject + start_offset;
req_cu_ptr = start_match - 1;
if (length == PCRE2_ZERO_TERMINATED)
{
length = PRIV(strlen)(subject);
was_zero_terminated = 1;
}
true_end_subject = end_subject = subject + length;
if (start_offset > length) return PCRE2_ERROR_BADOFFSET;
/* Check that the first field in the block is the magic number. */
if (re->magic_number != MAGIC_NUMBER) return PCRE2_ERROR_BADMAGIC;
/* Check the code unit width. */
if ((re->flags & PCRE2_MODE_MASK) != PCRE2_CODE_UNIT_WIDTH/8)
return PCRE2_ERROR_BADMODE;
/* PCRE2_NOTEMPTY and PCRE2_NOTEMPTY_ATSTART are match-time flags in the
options variable for this function. Users of PCRE2 who are not calling the
function directly would like to have a way of setting these flags, in the same
way that they can set pcre2_compile() flags like PCRE2_NO_AUTOPOSSESS with
constructions like (*NO_AUTOPOSSESS). To enable this, (*NOTEMPTY) and
(*NOTEMPTY_ATSTART) set bits in the pattern's "flag" function which we now
transfer to the options for this function. The bits are guaranteed to be
adjacent, but do not have the same values. This bit of Boolean trickery assumes
that the match-time bits are not more significant than the flag bits. If by
accident this is not the case, a compile-time division by zero error will
occur. */
#define FF (PCRE2_NOTEMPTY_SET|PCRE2_NE_ATST_SET)
#define OO (PCRE2_NOTEMPTY|PCRE2_NOTEMPTY_ATSTART)
options |= (re->flags & FF) / ((FF & (~FF+1)) / (OO & (~OO+1)));
#undef FF
#undef OO
/* If the pattern was successfully studied with JIT support, we will run the
JIT executable instead of the rest of this function. Most options must be set
at compile time for the JIT code to be usable. */
#ifdef SUPPORT_JIT
use_jit = (re->executable_jit != NULL &&
(options & ~PUBLIC_JIT_MATCH_OPTIONS) == 0);
#endif
/* Initialize UTF/UCP parameters. */
#ifdef SUPPORT_UNICODE
utf = (re->overall_options & PCRE2_UTF) != 0;
allow_invalid = (re->overall_options & PCRE2_MATCH_INVALID_UTF) != 0;
ucp = (re->overall_options & PCRE2_UCP) != 0;
#endif /* SUPPORT_UNICODE */
/* Convert the partial matching flags into an integer. */
mb->partial = ((options & PCRE2_PARTIAL_HARD) != 0)? 2 :
((options & PCRE2_PARTIAL_SOFT) != 0)? 1 : 0;
/* Partial matching and PCRE2_ENDANCHORED are currently not allowed at the same
time. */
if (mb->partial != 0 &&
((re->overall_options | options) & PCRE2_ENDANCHORED) != 0)
return PCRE2_ERROR_BADOPTION;
/* It is an error to set an offset limit without setting the flag at compile
time. */
if (mcontext != NULL && mcontext->offset_limit != PCRE2_UNSET &&
(re->overall_options & PCRE2_USE_OFFSET_LIMIT) == 0)
return PCRE2_ERROR_BADOFFSETLIMIT;
/* If the match data block was previously used with PCRE2_COPY_MATCHED_SUBJECT,
free the memory that was obtained. Set the field to NULL for no match cases. */
if ((match_data->flags & PCRE2_MD_COPIED_SUBJECT) != 0)
{
match_data->memctl.free((void *)match_data->subject,
match_data->memctl.memory_data);
match_data->flags &= ~PCRE2_MD_COPIED_SUBJECT;
}
match_data->subject = NULL;
/* Zero the error offset in case the first code unit is invalid UTF. */
match_data->startchar = 0;
/* ============================= JIT matching ============================== */
/* Prepare for JIT matching. Check a UTF string for validity unless no check is
requested or invalid UTF can be handled. We check only the portion of the
subject that might be be inspected during matching - from the offset minus the
maximum lookbehind to the given length. This saves time when a small part of a
large subject is being matched by the use of a starting offset. Note that the
maximum lookbehind is a number of characters, not code units. */
#ifdef SUPPORT_JIT
if (use_jit)
{
#ifdef SUPPORT_UNICODE
if (utf && (options & PCRE2_NO_UTF_CHECK) == 0 && !allow_invalid)
{
#if PCRE2_CODE_UNIT_WIDTH != 32
unsigned int i;
#endif
/* For 8-bit and 16-bit UTF, check that the first code unit is a valid
character start. */
#if PCRE2_CODE_UNIT_WIDTH != 32
if (start_match < end_subject && NOT_FIRSTCU(*start_match))
{
if (start_offset > 0) return PCRE2_ERROR_BADUTFOFFSET;
#if PCRE2_CODE_UNIT_WIDTH == 8
return PCRE2_ERROR_UTF8_ERR20; /* Isolated 0x80 byte */
#else
return PCRE2_ERROR_UTF16_ERR3; /* Isolated low surrogate */
#endif
}
#endif /* WIDTH != 32 */
/* Move back by the maximum lookbehind, just in case it happens at the very
start of matching. */
#if PCRE2_CODE_UNIT_WIDTH != 32
for (i = re->max_lookbehind; i > 0 && start_match > subject; i--)
{
start_match--;
while (start_match > subject &&
#if PCRE2_CODE_UNIT_WIDTH == 8
(*start_match & 0xc0) == 0x80)
#else /* 16-bit */
(*start_match & 0xfc00) == 0xdc00)
#endif
start_match--;
}
#else /* PCRE2_CODE_UNIT_WIDTH != 32 */
/* In the 32-bit library, one code unit equals one character. However,
we cannot just subtract the lookbehind and then compare pointers, because
a very large lookbehind could create an invalid pointer. */
if (start_offset >= re->max_lookbehind)
start_match -= re->max_lookbehind;
else
start_match = subject;
#endif /* PCRE2_CODE_UNIT_WIDTH != 32 */
/* Validate the relevant portion of the subject. Adjust the offset of an
invalid code point to be an absolute offset in the whole string. */
match_data->rc = PRIV(valid_utf)(start_match,
length - (start_match - subject), &(match_data->startchar));
if (match_data->rc != 0)
{
match_data->startchar += start_match - subject;
return match_data->rc;
}
jit_checked_utf = TRUE;
}
#endif /* SUPPORT_UNICODE */
/* If JIT returns BADOPTION, which means that the selected complete or
partial matching mode was not compiled, fall through to the interpreter. */
rc = pcre2_jit_match(code, subject, length, start_offset, options,
match_data, mcontext);
if (rc != PCRE2_ERROR_JIT_BADOPTION)
{
match_data->subject_length = length;
if (rc >= 0 && (options & PCRE2_COPY_MATCHED_SUBJECT) != 0)
{
length = CU2BYTES(length + was_zero_terminated);
match_data->subject = match_data->memctl.malloc(length,
match_data->memctl.memory_data);
if (match_data->subject == NULL) return PCRE2_ERROR_NOMEMORY;
memcpy((void *)match_data->subject, subject, length);
match_data->flags |= PCRE2_MD_COPIED_SUBJECT;
}
return rc;
}
}
#endif /* SUPPORT_JIT */
/* ========================= End of JIT matching ========================== */
/* Proceed with non-JIT matching. The default is to allow lookbehinds to the
start of the subject. A UTF check when there is a non-zero offset may change
this. */
mb->check_subject = subject;
/* If a UTF subject string was not checked for validity in the JIT code above,
check it here, and handle support for invalid UTF strings. The check above
happens only when invalid UTF is not supported and PCRE2_NO_CHECK_UTF is unset.
If we get here in those circumstances, it means the subject string is valid,
but for some reason JIT matching was not successful. There is no need to check
the subject again.
We check only the portion of the subject that might be be inspected during
matching - from the offset minus the maximum lookbehind to the given length.
This saves time when a small part of a large subject is being matched by the
use of a starting offset. Note that the maximum lookbehind is a number of
characters, not code units.
Note also that support for invalid UTF forces a check, overriding the setting
of PCRE2_NO_CHECK_UTF. */
#ifdef SUPPORT_UNICODE
if (utf &&
#ifdef SUPPORT_JIT
!jit_checked_utf &&
#endif
((options & PCRE2_NO_UTF_CHECK) == 0 || allow_invalid))
{
#if PCRE2_CODE_UNIT_WIDTH != 32
BOOL skipped_bad_start = FALSE;
#endif
/* For 8-bit and 16-bit UTF, check that the first code unit is a valid
character start. If we are handling invalid UTF, just skip over such code
units. Otherwise, give an appropriate error. */
#if PCRE2_CODE_UNIT_WIDTH != 32
if (allow_invalid)
{
while (start_match < end_subject && NOT_FIRSTCU(*start_match))
{
start_match++;
skipped_bad_start = TRUE;
}
}
else if (start_match < end_subject && NOT_FIRSTCU(*start_match))
{
if (start_offset > 0) return PCRE2_ERROR_BADUTFOFFSET;
#if PCRE2_CODE_UNIT_WIDTH == 8
return PCRE2_ERROR_UTF8_ERR20; /* Isolated 0x80 byte */
#else
return PCRE2_ERROR_UTF16_ERR3; /* Isolated low surrogate */
#endif
}
#endif /* WIDTH != 32 */
/* The mb->check_subject field points to the start of UTF checking;
lookbehinds can go back no further than this. */
mb->check_subject = start_match;
/* Move back by the maximum lookbehind, just in case it happens at the very
start of matching, but don't do this if we skipped bad 8-bit or 16-bit code
units above. */
#if PCRE2_CODE_UNIT_WIDTH != 32
if (!skipped_bad_start)
{
unsigned int i;
for (i = re->max_lookbehind; i > 0 && mb->check_subject > subject; i--)
{
mb->check_subject--;
while (mb->check_subject > subject &&
#if PCRE2_CODE_UNIT_WIDTH == 8
(*mb->check_subject & 0xc0) == 0x80)
#else /* 16-bit */
(*mb->check_subject & 0xfc00) == 0xdc00)
#endif
mb->check_subject--;
}
}
#else /* PCRE2_CODE_UNIT_WIDTH != 32 */
/* In the 32-bit library, one code unit equals one character. However,
we cannot just subtract the lookbehind and then compare pointers, because
a very large lookbehind could create an invalid pointer. */
if (start_offset >= re->max_lookbehind)
mb->check_subject -= re->max_lookbehind;
else
mb->check_subject = subject;
#endif /* PCRE2_CODE_UNIT_WIDTH != 32 */
/* Validate the relevant portion of the subject. There's a loop in case we
encounter bad UTF in the characters preceding start_match which we are
scanning because of a lookbehind. */
for (;;)
{
match_data->rc = PRIV(valid_utf)(mb->check_subject,
length - (mb->check_subject - subject), &(match_data->startchar));
if (match_data->rc == 0) break; /* Valid UTF string */
/* Invalid UTF string. Adjust the offset to be an absolute offset in the
whole string. If we are handling invalid UTF strings, set end_subject to
stop before the bad code unit, and set the options to "not end of line".
Otherwise return the error. */
match_data->startchar += mb->check_subject - subject;
if (!allow_invalid || match_data->rc > 0) return match_data->rc;
end_subject = subject + match_data->startchar;
/* If the end precedes start_match, it means there is invalid UTF in the
extra code units we reversed over because of a lookbehind. Advance past the
first bad code unit, and then skip invalid character starting code units in
8-bit and 16-bit modes, and try again with the original end point. */
if (end_subject < start_match)
{
mb->check_subject = end_subject + 1;
#if PCRE2_CODE_UNIT_WIDTH != 32
while (mb->check_subject < start_match && NOT_FIRSTCU(*mb->check_subject))
mb->check_subject++;
#endif
end_subject = true_end_subject;
}
/* Otherwise, set the not end of line option, and do the match. */
else
{
fragment_options = PCRE2_NOTEOL;
break;
}
}
}
#endif /* SUPPORT_UNICODE */
/* A NULL match context means "use a default context", but we take the memory
control functions from the pattern. */
if (mcontext == NULL)
{
mcontext = (pcre2_match_context *)(&PRIV(default_match_context));
mb->memctl = re->memctl;
}
else mb->memctl = mcontext->memctl;
anchored = ((re->overall_options | options) & PCRE2_ANCHORED) != 0;
firstline = !anchored && (re->overall_options & PCRE2_FIRSTLINE) != 0;
startline = (re->flags & PCRE2_STARTLINE) != 0;
bumpalong_limit = (mcontext->offset_limit == PCRE2_UNSET)?
true_end_subject : subject + mcontext->offset_limit;
/* Initialize and set up the fixed fields in the callout block, with a pointer
in the match block. */
mb->cb = &cb;
cb.version = 2;
cb.subject = subject;
cb.subject_length = (PCRE2_SIZE)(end_subject - subject);
cb.callout_flags = 0;
/* Fill in the remaining fields in the match block, except for moptions, which
gets set later. */
mb->callout = mcontext->callout;
mb->callout_data = mcontext->callout_data;
mb->start_subject = subject;
mb->start_offset = start_offset;
mb->end_subject = end_subject;
mb->true_end_subject = true_end_subject;
mb->hasthen = (re->flags & PCRE2_HASTHEN) != 0;
mb->allowemptypartial = (re->max_lookbehind > 0) ||
(re->flags & PCRE2_MATCH_EMPTY) != 0;
mb->poptions = re->overall_options; /* Pattern options */
mb->ignore_skip_arg = 0;
mb->mark = mb->nomatch_mark = NULL; /* In case never set */
/* The name table is needed for finding all the numbers associated with a
given name, for condition testing. The code follows the name table. */
mb->name_table = (PCRE2_UCHAR *)((uint8_t *)re + sizeof(pcre2_real_code));
mb->name_count = re->name_count;
mb->name_entry_size = re->name_entry_size;
mb->start_code = mb->name_table + re->name_count * re->name_entry_size;
/* Process the \R and newline settings. */
mb->bsr_convention = re->bsr_convention;
mb->nltype = NLTYPE_FIXED;
switch(re->newline_convention)
{
case PCRE2_NEWLINE_CR:
mb->nllen = 1;
mb->nl[0] = CHAR_CR;
break;
case PCRE2_NEWLINE_LF:
mb->nllen = 1;
mb->nl[0] = CHAR_NL;
break;
case PCRE2_NEWLINE_NUL:
mb->nllen = 1;
mb->nl[0] = CHAR_NUL;
break;
case PCRE2_NEWLINE_CRLF:
mb->nllen = 2;
mb->nl[0] = CHAR_CR;
mb->nl[1] = CHAR_NL;
break;
case PCRE2_NEWLINE_ANY:
mb->nltype = NLTYPE_ANY;
break;
case PCRE2_NEWLINE_ANYCRLF:
mb->nltype = NLTYPE_ANYCRLF;
break;
default: return PCRE2_ERROR_INTERNAL;
}
/* The backtracking frames have fixed data at the front, and a PCRE2_SIZE
vector at the end, whose size depends on the number of capturing parentheses in
the pattern. It is not used at all if there are no capturing parentheses.
frame_size is the total size of each frame
match_data->heapframes is the pointer to the frames vector
match_data->heapframes_size is the allocated size of the vector
We must pad the frame_size for alignment to ensure subsequent frames are as
aligned as heapframe. Whilst ovector is word-aligned due to being a PCRE2_SIZE
array, that does not guarantee it is suitably aligned for pointers, as some
architectures have pointers that are larger than a size_t. */
frame_size = (offsetof(heapframe, ovector) +
re->top_bracket * 2 * sizeof(PCRE2_SIZE) + HEAPFRAME_ALIGNMENT - 1) &
~(HEAPFRAME_ALIGNMENT - 1);
/* Limits set in the pattern override the match context only if they are
smaller. */
mb->heap_limit = ((mcontext->heap_limit < re->limit_heap)?
mcontext->heap_limit : re->limit_heap);
mb->match_limit = (mcontext->match_limit < re->limit_match)?
mcontext->match_limit : re->limit_match;
mb->match_limit_depth = (mcontext->depth_limit < re->limit_depth)?
mcontext->depth_limit : re->limit_depth;
/* If a pattern has very many capturing parentheses, the frame size may be very
large. Set the initial frame vector size to ensure that there are at least 10
available frames, but enforce a minimum of START_FRAMES_SIZE. If this is
greater than the heap limit, get as large a vector as possible. */
heapframes_size = frame_size * 10;
if (heapframes_size < START_FRAMES_SIZE) heapframes_size = START_FRAMES_SIZE;
if (heapframes_size / 1024 > mb->heap_limit)
{
PCRE2_SIZE max_size = 1024 * mb->heap_limit;
if (max_size < frame_size) return PCRE2_ERROR_HEAPLIMIT;
heapframes_size = max_size;
}
/* If an existing frame vector in the match_data block is large enough, we can
use it. Otherwise, free any pre-existing vector and get a new one. */
if (match_data->heapframes_size < heapframes_size)
{
match_data->memctl.free(match_data->heapframes,
match_data->memctl.memory_data);
match_data->heapframes = match_data->memctl.malloc(heapframes_size,
match_data->memctl.memory_data);
if (match_data->heapframes == NULL)
{
match_data->heapframes_size = 0;
return PCRE2_ERROR_NOMEMORY;
}
match_data->heapframes_size = heapframes_size;
}
/* Write to the ovector within the first frame to mark every capture unset and
to avoid uninitialized memory read errors when it is copied to a new frame. */
memset((char *)(match_data->heapframes) + offsetof(heapframe, ovector), 0xff,
frame_size - offsetof(heapframe, ovector));
/* Pointers to the individual character tables */
mb->lcc = re->tables + lcc_offset;
mb->fcc = re->tables + fcc_offset;
mb->ctypes = re->tables + ctypes_offset;
/* Set up the first code unit to match, if available. If there's no first code
unit there may be a bitmap of possible first characters. */
if ((re->flags & PCRE2_FIRSTSET) != 0)
{
has_first_cu = TRUE;
first_cu = first_cu2 = (PCRE2_UCHAR)(re->first_codeunit);
if ((re->flags & PCRE2_FIRSTCASELESS) != 0)
{
first_cu2 = TABLE_GET(first_cu, mb->fcc, first_cu);
#ifdef SUPPORT_UNICODE
#if PCRE2_CODE_UNIT_WIDTH == 8
if (first_cu > 127 && ucp && !utf) first_cu2 = UCD_OTHERCASE(first_cu);
#else
if (first_cu > 127 && (utf || ucp)) first_cu2 = UCD_OTHERCASE(first_cu);
#endif
#endif /* SUPPORT_UNICODE */
}
}
else
if (!startline && (re->flags & PCRE2_FIRSTMAPSET) != 0)
start_bits = re->start_bitmap;
/* There may also be a "last known required character" set. */
if ((re->flags & PCRE2_LASTSET) != 0)
{
has_req_cu = TRUE;
req_cu = req_cu2 = (PCRE2_UCHAR)(re->last_codeunit);
if ((re->flags & PCRE2_LASTCASELESS) != 0)
{
req_cu2 = TABLE_GET(req_cu, mb->fcc, req_cu);
#ifdef SUPPORT_UNICODE
#if PCRE2_CODE_UNIT_WIDTH == 8
if (req_cu > 127 && ucp && !utf) req_cu2 = UCD_OTHERCASE(req_cu);
#else
if (req_cu > 127 && (utf || ucp)) req_cu2 = UCD_OTHERCASE(req_cu);
#endif
#endif /* SUPPORT_UNICODE */
}
}
/* ==========================================================================*/
/* Loop for handling unanchored repeated matching attempts; for anchored regexs
the loop runs just once. */
#ifdef SUPPORT_UNICODE
FRAGMENT_RESTART:
#endif
start_partial = match_partial = NULL;
mb->hitend = FALSE;
#if PCRE2_CODE_UNIT_WIDTH == 8
memchr_found_first_cu = NULL;
memchr_found_first_cu2 = NULL;
#endif
for(;;)
{
PCRE2_SPTR new_start_match;
/* ----------------- Start of match optimizations ---------------- */
/* There are some optimizations that avoid running the match if a known
starting point is not found, or if a known later code unit is not present.
However, there is an option (settable at compile time) that disables these,
for testing and for ensuring that all callouts do actually occur. */
if ((re->overall_options & PCRE2_NO_START_OPTIMIZE) == 0)
{
/* If firstline is TRUE, the start of the match is constrained to the first
line of a multiline string. That is, the match must be before or at the
first newline following the start of matching. Temporarily adjust
end_subject so that we stop the scans for a first code unit at a newline.
If the match fails at the newline, later code breaks the loop. */
if (firstline)
{
PCRE2_SPTR t = start_match;
#ifdef SUPPORT_UNICODE
if (utf)
{
while (t < end_subject && !IS_NEWLINE(t))
{
t++;
ACROSSCHAR(t < end_subject, t, t++);
}
}
else
#endif
while (t < end_subject && !IS_NEWLINE(t)) t++;
end_subject = t;
}
/* Anchored: check the first code unit if one is recorded. This may seem
pointless but it can help in detecting a no match case without scanning for
the required code unit. */
if (anchored)
{
if (has_first_cu || start_bits != NULL)
{
BOOL ok = start_match < end_subject;
if (ok)
{
PCRE2_UCHAR c = UCHAR21TEST(start_match);
ok = has_first_cu && (c == first_cu || c == first_cu2);
if (!ok && start_bits != NULL)
{
#if PCRE2_CODE_UNIT_WIDTH != 8
if (c > 255) c = 255;
#endif
ok = (start_bits[c/8] & (1u << (c&7))) != 0;
}
}
if (!ok)
{
rc = MATCH_NOMATCH;
break;
}
}
}
/* Not anchored. Advance to a unique first code unit if there is one. */
else
{
if (has_first_cu)
{
if (first_cu != first_cu2) /* Caseless */
{
/* In 16-bit and 32_bit modes we have to do our own search, so can
look for both cases at once. */
#if PCRE2_CODE_UNIT_WIDTH != 8
PCRE2_UCHAR smc;
while (start_match < end_subject &&
(smc = UCHAR21TEST(start_match)) != first_cu &&
smc != first_cu2)
start_match++;
#else
/* In 8-bit mode, the use of memchr() gives a big speed up, even
though we have to call it twice in order to find the earliest
occurrence of the code unit in either of its cases. Caching is used
to remember the positions of previously found code units. This can
make a huge difference when the strings are very long and only one
case is actually present. */
PCRE2_SPTR pp1 = NULL;
PCRE2_SPTR pp2 = NULL;
PCRE2_SIZE searchlength = end_subject - start_match;
/* If we haven't got a previously found position for first_cu, or if
the current starting position is later, we need to do a search. If
the code unit is not found, set it to the end. */
if (memchr_found_first_cu == NULL ||
start_match > memchr_found_first_cu)
{
pp1 = memchr(start_match, first_cu, searchlength);
memchr_found_first_cu = (pp1 == NULL)? end_subject : pp1;
}
/* If the start is before a previously found position, use the
previous position, or NULL if a previous search failed. */
else pp1 = (memchr_found_first_cu == end_subject)? NULL :
memchr_found_first_cu;
/* Do the same thing for the other case. */
if (memchr_found_first_cu2 == NULL ||
start_match > memchr_found_first_cu2)
{
pp2 = memchr(start_match, first_cu2, searchlength);
memchr_found_first_cu2 = (pp2 == NULL)? end_subject : pp2;
}
else pp2 = (memchr_found_first_cu2 == end_subject)? NULL :
memchr_found_first_cu2;
/* Set the start to the end of the subject if neither case was found.
Otherwise, use the earlier found point. */
if (pp1 == NULL)
start_match = (pp2 == NULL)? end_subject : pp2;
else
start_match = (pp2 == NULL || pp1 < pp2)? pp1 : pp2;
#endif /* 8-bit handling */
}
/* The caseful case is much simpler. */
else
{
#if PCRE2_CODE_UNIT_WIDTH != 8
while (start_match < end_subject && UCHAR21TEST(start_match) !=
first_cu)
start_match++;
#else
start_match = memchr(start_match, first_cu, end_subject - start_match);
if (start_match == NULL) start_match = end_subject;
#endif
}
/* If we can't find the required first code unit, having reached the
true end of the subject, break the bumpalong loop, to force a match
failure, except when doing partial matching, when we let the next cycle
run at the end of the subject. To see why, consider the pattern
/(?<=abc)def/, which partially matches "abc", even though the string
does not contain the starting character "d". If we have not reached the
true end of the subject (PCRE2_FIRSTLINE caused end_subject to be
temporarily modified) we also let the cycle run, because the matching
string is legitimately allowed to start with the first code unit of a
newline. */
if (mb->partial == 0 && start_match >= mb->end_subject)
{
rc = MATCH_NOMATCH;
break;
}
}
/* If there's no first code unit, advance to just after a linebreak for a
multiline match if required. */
else if (startline)
{
if (start_match > mb->start_subject + start_offset)
{
#ifdef SUPPORT_UNICODE
if (utf)
{
while (start_match < end_subject && !WAS_NEWLINE(start_match))
{
start_match++;
ACROSSCHAR(start_match < end_subject, start_match, start_match++);
}
}
else
#endif
while (start_match < end_subject && !WAS_NEWLINE(start_match))
start_match++;
/* If we have just passed a CR and the newline option is ANY or
ANYCRLF, and we are now at a LF, advance the match position by one
more code unit. */
if (start_match[-1] == CHAR_CR &&
(mb->nltype == NLTYPE_ANY || mb->nltype == NLTYPE_ANYCRLF) &&
start_match < end_subject &&
UCHAR21TEST(start_match) == CHAR_NL)
start_match++;
}
}
/* If there's no first code unit or a requirement for a multiline line
start, advance to a non-unique first code unit if any have been
identified. The bitmap contains only 256 bits. When code units are 16 or
32 bits wide, all code units greater than 254 set the 255 bit. */
else if (start_bits != NULL)
{
while (start_match < end_subject)
{
uint32_t c = UCHAR21TEST(start_match);
#if PCRE2_CODE_UNIT_WIDTH != 8
if (c > 255) c = 255;
#endif
if ((start_bits[c/8] & (1u << (c&7))) != 0) break;
start_match++;
}
/* See comment above in first_cu checking about the next few lines. */
if (mb->partial == 0 && start_match >= mb->end_subject)
{
rc = MATCH_NOMATCH;
break;
}
}
} /* End first code unit handling */
/* Restore fudged end_subject */
end_subject = mb->end_subject;
/* The following two optimizations must be disabled for partial matching. */
if (mb->partial == 0)
{
PCRE2_SPTR p;
/* The minimum matching length is a lower bound; no string of that length
may actually match the pattern. Although the value is, strictly, in
characters, we treat it as code units to avoid spending too much time in
this optimization. */
if (end_subject - start_match < re->minlength)
{
rc = MATCH_NOMATCH;
break;
}
/* If req_cu is set, we know that that code unit must appear in the
subject for the (non-partial) match to succeed. If the first code unit is
set, req_cu must be later in the subject; otherwise the test starts at
the match point. This optimization can save a huge amount of backtracking
in patterns with nested unlimited repeats that aren't going to match.
Writing separate code for caseful/caseless versions makes it go faster,
as does using an autoincrement and backing off on a match. As in the case
of the first code unit, using memchr() in the 8-bit library gives a big
speed up. Unlike the first_cu check above, we do not need to call
memchr() twice in the caseless case because we only need to check for the
presence of the character in either case, not find the first occurrence.
The search can be skipped if the code unit was found later than the
current starting point in a previous iteration of the bumpalong loop.
HOWEVER: when the subject string is very, very long, searching to its end
can take a long time, and give bad performance on quite ordinary
anchored patterns. This showed up when somebody was matching something
like /^\d+C/ on a 32-megabyte string... so we don't do this when the
string is sufficiently long, but it's worth searching a lot more for
unanchored patterns. */
p = start_match + (has_first_cu? 1:0);
if (has_req_cu && p > req_cu_ptr)
{
PCRE2_SIZE check_length = end_subject - start_match;
if (check_length < REQ_CU_MAX ||
(!anchored && check_length < REQ_CU_MAX * 1000))
{
if (req_cu != req_cu2) /* Caseless */
{
#if PCRE2_CODE_UNIT_WIDTH != 8
while (p < end_subject)
{
uint32_t pp = UCHAR21INCTEST(p);
if (pp == req_cu || pp == req_cu2) { p--; break; }
}
#else /* 8-bit code units */
PCRE2_SPTR pp = p;
p = memchr(pp, req_cu, end_subject - pp);
if (p == NULL)
{
p = memchr(pp, req_cu2, end_subject - pp);
if (p == NULL) p = end_subject;
}
#endif /* PCRE2_CODE_UNIT_WIDTH != 8 */
}
/* The caseful case */
else
{
#if PCRE2_CODE_UNIT_WIDTH != 8
while (p < end_subject)
{
if (UCHAR21INCTEST(p) == req_cu) { p--; break; }
}
#else /* 8-bit code units */
p = memchr(p, req_cu, end_subject - p);
if (p == NULL) p = end_subject;
#endif
}
/* If we can't find the required code unit, break the bumpalong loop,
forcing a match failure. */
if (p >= end_subject)
{
rc = MATCH_NOMATCH;
break;
}
/* If we have found the required code unit, save the point where we
found it, so that we don't search again next time round the bumpalong
loop if the start hasn't yet passed this code unit. */
req_cu_ptr = p;
}
}
}
}
/* ------------ End of start of match optimizations ------------ */
/* Give no match if we have passed the bumpalong limit. */
if (start_match > bumpalong_limit)
{
rc = MATCH_NOMATCH;
break;
}
/* OK, we can now run the match. If "hitend" is set afterwards, remember the
first starting point for which a partial match was found. */
cb.start_match = (PCRE2_SIZE)(start_match - subject);
cb.callout_flags |= PCRE2_CALLOUT_STARTMATCH;
mb->start_used_ptr = start_match;
mb->last_used_ptr = start_match;
#ifdef SUPPORT_UNICODE
mb->moptions = options | fragment_options;
#else
mb->moptions = options;
#endif
mb->match_call_count = 0;
mb->end_offset_top = 0;
mb->skip_arg_count = 0;
#ifdef DEBUG_SHOW_OPS
fprintf(stderr, "++ Calling match()\n");
#endif
rc = match(start_match, mb->start_code, re->top_bracket, frame_size,
match_data, mb);
#ifdef DEBUG_SHOW_OPS
fprintf(stderr, "++ match() returned %d\n\n", rc);
#endif
if (mb->hitend && start_partial == NULL)
{
start_partial = mb->start_used_ptr;
match_partial = start_match;
}
switch(rc)
{
/* If MATCH_SKIP_ARG reaches this level it means that a MARK that matched
the SKIP's arg was not found. In this circumstance, Perl ignores the SKIP
entirely. The only way we can do that is to re-do the match at the same
point, with a flag to force SKIP with an argument to be ignored. Just
treating this case as NOMATCH does not work because it does not check other
alternatives in patterns such as A(*SKIP:A)B|AC when the subject is AC. */
case MATCH_SKIP_ARG:
new_start_match = start_match;
mb->ignore_skip_arg = mb->skip_arg_count;
break;
/* SKIP passes back the next starting point explicitly, but if it is no
greater than the match we have just done, treat it as NOMATCH. */
case MATCH_SKIP:
if (mb->verb_skip_ptr > start_match)
{
new_start_match = mb->verb_skip_ptr;
break;
}
/* Fall through */
/* NOMATCH and PRUNE advance by one character. THEN at this level acts
exactly like PRUNE. Unset ignore SKIP-with-argument. */
case MATCH_NOMATCH:
case MATCH_PRUNE:
case MATCH_THEN:
mb->ignore_skip_arg = 0;
new_start_match = start_match + 1;
#ifdef SUPPORT_UNICODE
if (utf)
ACROSSCHAR(new_start_match < end_subject, new_start_match,
new_start_match++);
#endif
break;
/* COMMIT disables the bumpalong, but otherwise behaves as NOMATCH. */
case MATCH_COMMIT:
rc = MATCH_NOMATCH;
goto ENDLOOP;
/* Any other return is either a match, or some kind of error. */
default:
goto ENDLOOP;
}
/* Control reaches here for the various types of "no match at this point"
result. Reset the code to MATCH_NOMATCH for subsequent checking. */
rc = MATCH_NOMATCH;
/* If PCRE2_FIRSTLINE is set, the match must happen before or at the first
newline in the subject (though it may continue over the newline). Therefore,
if we have just failed to match, starting at a newline, do not continue. */
if (firstline && IS_NEWLINE(start_match)) break;
/* Advance to new matching position */
start_match = new_start_match;
/* Break the loop if the pattern is anchored or if we have passed the end of
the subject. */
if (anchored || start_match > end_subject) break;
/* If we have just passed a CR and we are now at a LF, and the pattern does
not contain any explicit matches for \r or \n, and the newline option is CRLF
or ANY or ANYCRLF, advance the match position by one more code unit. In
normal matching start_match will aways be greater than the first position at
this stage, but a failed *SKIP can cause a return at the same point, which is
why the first test exists. */
if (start_match > subject + start_offset &&
start_match[-1] == CHAR_CR &&
start_match < end_subject &&
*start_match == CHAR_NL &&
(re->flags & PCRE2_HASCRORLF) == 0 &&
(mb->nltype == NLTYPE_ANY ||
mb->nltype == NLTYPE_ANYCRLF ||
mb->nllen == 2))
start_match++;
mb->mark = NULL; /* Reset for start of next match attempt */
} /* End of for(;;) "bumpalong" loop */
/* ==========================================================================*/
/* When we reach here, one of the following stopping conditions is true:
(1) The match succeeded, either completely, or partially;
(2) The pattern is anchored or the match was failed after (*COMMIT);
(3) We are past the end of the subject or the bumpalong limit;
(4) PCRE2_FIRSTLINE is set and we have failed to match at a newline, because
this option requests that a match occur at or before the first newline in
the subject.
(5) Some kind of error occurred.
*/
ENDLOOP:
/* If end_subject != true_end_subject, it means we are handling invalid UTF,
and have just processed a non-terminal fragment. If this resulted in no match
or a partial match we must carry on to the next fragment (a partial match is
returned to the caller only at the very end of the subject). A loop is used to
avoid trying to match against empty fragments; if the pattern can match an
empty string it would have done so already. */
#ifdef SUPPORT_UNICODE
if (utf && end_subject != true_end_subject &&
(rc == MATCH_NOMATCH || rc == PCRE2_ERROR_PARTIAL))
{
for (;;)
{
/* Advance past the first bad code unit, and then skip invalid character
starting code units in 8-bit and 16-bit modes. */
start_match = end_subject + 1;
#if PCRE2_CODE_UNIT_WIDTH != 32
while (start_match < true_end_subject && NOT_FIRSTCU(*start_match))
start_match++;
#endif
/* If we have hit the end of the subject, there isn't another non-empty
fragment, so give up. */
if (start_match >= true_end_subject)
{
rc = MATCH_NOMATCH; /* In case it was partial */
match_partial = NULL;
break;
}
/* Check the rest of the subject */
mb->check_subject = start_match;
rc = PRIV(valid_utf)(start_match, length - (start_match - subject),
&(match_data->startchar));
/* The rest of the subject is valid UTF. */
if (rc == 0)
{
mb->end_subject = end_subject = true_end_subject;
fragment_options = PCRE2_NOTBOL;
goto FRAGMENT_RESTART;
}
/* A subsequent UTF error has been found; if the next fragment is
non-empty, set up to process it. Otherwise, let the loop advance. */
else if (rc < 0)
{
mb->end_subject = end_subject = start_match + match_data->startchar;
if (end_subject > start_match)
{
fragment_options = PCRE2_NOTBOL|PCRE2_NOTEOL;
goto FRAGMENT_RESTART;
}
}
}
}
#endif /* SUPPORT_UNICODE */
/* Fill in fields that are always returned in the match data. */
match_data->code = re;
match_data->mark = mb->mark;
match_data->matchedby = PCRE2_MATCHEDBY_INTERPRETER;
/* Handle a fully successful match. Set the return code to the number of
captured strings, or 0 if there were too many to fit into the ovector, and then
set the remaining returned values before returning. Make a copy of the subject
string if requested. */
if (rc == MATCH_MATCH)
{
match_data->rc = ((int)mb->end_offset_top >= 2 * match_data->oveccount)?
0 : (int)mb->end_offset_top/2 + 1;
match_data->subject_length = length;
match_data->startchar = start_match - subject;
match_data->leftchar = mb->start_used_ptr - subject;
match_data->rightchar = ((mb->last_used_ptr > mb->end_match_ptr)?
mb->last_used_ptr : mb->end_match_ptr) - subject;
if ((options & PCRE2_COPY_MATCHED_SUBJECT) != 0)
{
length = CU2BYTES(length + was_zero_terminated);
match_data->subject = match_data->memctl.malloc(length,
match_data->memctl.memory_data);
if (match_data->subject == NULL) return PCRE2_ERROR_NOMEMORY;
memcpy((void *)match_data->subject, subject, length);
match_data->flags |= PCRE2_MD_COPIED_SUBJECT;
}
else match_data->subject = subject;
return match_data->rc;
}
/* Control gets here if there has been a partial match, an error, or if the
overall match attempt has failed at all permitted starting positions. Any mark
data is in the nomatch_mark field. */
match_data->mark = mb->nomatch_mark;
/* For anything other than nomatch or partial match, just return the code. */
if (rc != MATCH_NOMATCH && rc != PCRE2_ERROR_PARTIAL) match_data->rc = rc;
/* Handle a partial match. If a "soft" partial match was requested, searching
for a complete match will have continued, and the value of rc at this point
will be MATCH_NOMATCH. For a "hard" partial match, it will already be
PCRE2_ERROR_PARTIAL. */
else if (match_partial != NULL)
{
match_data->subject = subject;
match_data->subject_length = length;
match_data->ovector[0] = match_partial - subject;
match_data->ovector[1] = end_subject - subject;
match_data->startchar = match_partial - subject;
match_data->leftchar = start_partial - subject;
match_data->rightchar = end_subject - subject;
match_data->rc = PCRE2_ERROR_PARTIAL;
}
/* Else this is the classic nomatch case. */
else match_data->rc = PCRE2_ERROR_NOMATCH;
return match_data->rc;
}
/* These #undefs are here to enable unity builds with CMake. */
#undef NLBLOCK /* Block containing newline information */
#undef PSSTART /* Field containing processed string start */
#undef PSEND /* Field containing processed string end */
/* End of pcre2_match.c */
|