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
|
/*
** 2004 April 6
**
** The author disclaims copyright to this source code. In place of
** a legal notice, here is a blessing:
**
** May you do good and not evil.
** May you find forgiveness for yourself and forgive others.
** May you share freely, never taking more than you give.
**
*************************************************************************
** $Id: btree.c,v 1.396 2007/08/13 14:56:44 drh Exp $
**
** This file implements a external (disk-based) database using BTrees.
** See the header comment on "btreeInt.h" for additional information.
** Including a description of file format and an overview of operation.
*/
#include "btreeInt.h"
/*
** The header string that appears at the beginning of every
** SQLite database.
*/
static const char zMagicHeader[] = SQLITE_FILE_HEADER;
/*
** Set this global variable to 1 to enable tracing using the TRACE
** macro.
*/
#if SQLITE_TEST
int sqlite3_btree_trace=0; /* True to enable tracing */
#endif
/*
** Forward declaration
*/
static int checkReadLocks(Btree*,Pgno,BtCursor*);
#ifdef SQLITE_OMIT_SHARED_CACHE
/*
** The functions queryTableLock(), lockTable() and unlockAllTables()
** manipulate entries in the BtShared.pLock linked list used to store
** shared-cache table level locks. If the library is compiled with the
** shared-cache feature disabled, then there is only ever one user
** of each BtShared structure and so this locking is not necessary.
** So define the lock related functions as no-ops.
*/
#define queryTableLock(a,b,c) SQLITE_OK
#define lockTable(a,b,c) SQLITE_OK
#define unlockAllTables(a)
#else
/*
** Query to see if btree handle p may obtain a lock of type eLock
** (READ_LOCK or WRITE_LOCK) on the table with root-page iTab. Return
** SQLITE_OK if the lock may be obtained (by calling lockTable()), or
** SQLITE_LOCKED if not.
*/
static int queryTableLock(Btree *p, Pgno iTab, u8 eLock){
BtShared *pBt = p->pBt;
BtLock *pIter;
/* This is a no-op if the shared-cache is not enabled */
if( 0==sqlite3ThreadDataReadOnly()->useSharedData ){
return SQLITE_OK;
}
/* This (along with lockTable()) is where the ReadUncommitted flag is
** dealt with. If the caller is querying for a read-lock and the flag is
** set, it is unconditionally granted - even if there are write-locks
** on the table. If a write-lock is requested, the ReadUncommitted flag
** is not considered.
**
** In function lockTable(), if a read-lock is demanded and the
** ReadUncommitted flag is set, no entry is added to the locks list
** (BtShared.pLock).
**
** To summarize: If the ReadUncommitted flag is set, then read cursors do
** not create or respect table locks. The locking procedure for a
** write-cursor does not change.
*/
if(
!p->pSqlite ||
0==(p->pSqlite->flags&SQLITE_ReadUncommitted) ||
eLock==WRITE_LOCK ||
iTab==MASTER_ROOT
){
for(pIter=pBt->pLock; pIter; pIter=pIter->pNext){
if( pIter->pBtree!=p && pIter->iTable==iTab &&
(pIter->eLock!=eLock || eLock!=READ_LOCK) ){
return SQLITE_LOCKED;
}
}
}
return SQLITE_OK;
}
/*
** Add a lock on the table with root-page iTable to the shared-btree used
** by Btree handle p. Parameter eLock must be either READ_LOCK or
** WRITE_LOCK.
**
** SQLITE_OK is returned if the lock is added successfully. SQLITE_BUSY and
** SQLITE_NOMEM may also be returned.
*/
static int lockTable(Btree *p, Pgno iTable, u8 eLock){
BtShared *pBt = p->pBt;
BtLock *pLock = 0;
BtLock *pIter;
/* This is a no-op if the shared-cache is not enabled */
if( 0==sqlite3ThreadDataReadOnly()->useSharedData ){
return SQLITE_OK;
}
assert( SQLITE_OK==queryTableLock(p, iTable, eLock) );
/* If the read-uncommitted flag is set and a read-lock is requested,
** return early without adding an entry to the BtShared.pLock list. See
** comment in function queryTableLock() for more info on handling
** the ReadUncommitted flag.
*/
if(
(p->pSqlite) &&
(p->pSqlite->flags&SQLITE_ReadUncommitted) &&
(eLock==READ_LOCK) &&
iTable!=MASTER_ROOT
){
return SQLITE_OK;
}
/* First search the list for an existing lock on this table. */
for(pIter=pBt->pLock; pIter; pIter=pIter->pNext){
if( pIter->iTable==iTable && pIter->pBtree==p ){
pLock = pIter;
break;
}
}
/* If the above search did not find a BtLock struct associating Btree p
** with table iTable, allocate one and link it into the list.
*/
if( !pLock ){
pLock = (BtLock *)sqliteMalloc(sizeof(BtLock));
if( !pLock ){
return SQLITE_NOMEM;
}
pLock->iTable = iTable;
pLock->pBtree = p;
pLock->pNext = pBt->pLock;
pBt->pLock = pLock;
}
/* Set the BtLock.eLock variable to the maximum of the current lock
** and the requested lock. This means if a write-lock was already held
** and a read-lock requested, we don't incorrectly downgrade the lock.
*/
assert( WRITE_LOCK>READ_LOCK );
if( eLock>pLock->eLock ){
pLock->eLock = eLock;
}
return SQLITE_OK;
}
/*
** Release all the table locks (locks obtained via calls to the lockTable()
** procedure) held by Btree handle p.
*/
static void unlockAllTables(Btree *p){
BtLock **ppIter = &p->pBt->pLock;
/* If the shared-cache extension is not enabled, there should be no
** locks in the BtShared.pLock list, making this procedure a no-op. Assert
** that this is the case.
*/
assert( sqlite3ThreadDataReadOnly()->useSharedData || 0==*ppIter );
while( *ppIter ){
BtLock *pLock = *ppIter;
if( pLock->pBtree==p ){
*ppIter = pLock->pNext;
sqliteFree(pLock);
}else{
ppIter = &pLock->pNext;
}
}
}
#endif /* SQLITE_OMIT_SHARED_CACHE */
static void releasePage(MemPage *pPage); /* Forward reference */
#ifndef SQLITE_OMIT_INCRBLOB
/*
** Invalidate the overflow page-list cache for cursor pCur, if any.
*/
static void invalidateOverflowCache(BtCursor *pCur){
sqliteFree(pCur->aOverflow);
pCur->aOverflow = 0;
}
/*
** Invalidate the overflow page-list cache for all cursors opened
** on the shared btree structure pBt.
*/
static void invalidateAllOverflowCache(BtShared *pBt){
BtCursor *p;
for(p=pBt->pCursor; p; p=p->pNext){
invalidateOverflowCache(p);
}
}
#else
#define invalidateOverflowCache(x)
#define invalidateAllOverflowCache(x)
#endif
/*
** Save the current cursor position in the variables BtCursor.nKey
** and BtCursor.pKey. The cursor's state is set to CURSOR_REQUIRESEEK.
*/
static int saveCursorPosition(BtCursor *pCur){
int rc;
assert( CURSOR_VALID==pCur->eState );
assert( 0==pCur->pKey );
rc = sqlite3BtreeKeySize(pCur, &pCur->nKey);
/* If this is an intKey table, then the above call to BtreeKeySize()
** stores the integer key in pCur->nKey. In this case this value is
** all that is required. Otherwise, if pCur is not open on an intKey
** table, then malloc space for and store the pCur->nKey bytes of key
** data.
*/
if( rc==SQLITE_OK && 0==pCur->pPage->intKey){
void *pKey = sqliteMalloc(pCur->nKey);
if( pKey ){
rc = sqlite3BtreeKey(pCur, 0, pCur->nKey, pKey);
if( rc==SQLITE_OK ){
pCur->pKey = pKey;
}else{
sqliteFree(pKey);
}
}else{
rc = SQLITE_NOMEM;
}
}
assert( !pCur->pPage->intKey || !pCur->pKey );
if( rc==SQLITE_OK ){
releasePage(pCur->pPage);
pCur->pPage = 0;
pCur->eState = CURSOR_REQUIRESEEK;
}
invalidateOverflowCache(pCur);
return rc;
}
/*
** Save the positions of all cursors except pExcept open on the table
** with root-page iRoot. Usually, this is called just before cursor
** pExcept is used to modify the table (BtreeDelete() or BtreeInsert()).
*/
static int saveAllCursors(BtShared *pBt, Pgno iRoot, BtCursor *pExcept){
BtCursor *p;
for(p=pBt->pCursor; p; p=p->pNext){
if( p!=pExcept && (0==iRoot || p->pgnoRoot==iRoot) &&
p->eState==CURSOR_VALID ){
int rc = saveCursorPosition(p);
if( SQLITE_OK!=rc ){
return rc;
}
}
}
return SQLITE_OK;
}
/*
** Clear the current cursor position.
*/
static void clearCursorPosition(BtCursor *pCur){
sqliteFree(pCur->pKey);
pCur->pKey = 0;
pCur->eState = CURSOR_INVALID;
}
/*
** Restore the cursor to the position it was in (or as close to as possible)
** when saveCursorPosition() was called. Note that this call deletes the
** saved position info stored by saveCursorPosition(), so there can be
** at most one effective restoreOrClearCursorPosition() call after each
** saveCursorPosition().
**
** If the second argument argument - doSeek - is false, then instead of
** returning the cursor to it's saved position, any saved position is deleted
** and the cursor state set to CURSOR_INVALID.
*/
int sqlite3BtreeRestoreOrClearCursorPosition(BtCursor *pCur){
int rc;
assert( pCur->eState==CURSOR_REQUIRESEEK );
#ifndef SQLITE_OMIT_INCRBLOB
if( pCur->isIncrblobHandle ){
return SQLITE_ABORT;
}
#endif
pCur->eState = CURSOR_INVALID;
rc = sqlite3BtreeMoveto(pCur, pCur->pKey, pCur->nKey, 0, &pCur->skip);
if( rc==SQLITE_OK ){
sqliteFree(pCur->pKey);
pCur->pKey = 0;
assert( pCur->eState==CURSOR_VALID || pCur->eState==CURSOR_INVALID );
}
return rc;
}
#define restoreOrClearCursorPosition(p) \
(p->eState==CURSOR_REQUIRESEEK ? \
sqlite3BtreeRestoreOrClearCursorPosition(p) : \
SQLITE_OK)
#ifndef SQLITE_OMIT_AUTOVACUUM
/*
** Given a page number of a regular database page, return the page
** number for the pointer-map page that contains the entry for the
** input page number.
*/
static Pgno ptrmapPageno(BtShared *pBt, Pgno pgno){
int nPagesPerMapPage = (pBt->usableSize/5)+1;
int iPtrMap = (pgno-2)/nPagesPerMapPage;
int ret = (iPtrMap*nPagesPerMapPage) + 2;
if( ret==PENDING_BYTE_PAGE(pBt) ){
ret++;
}
return ret;
}
/*
** Write an entry into the pointer map.
**
** This routine updates the pointer map entry for page number 'key'
** so that it maps to type 'eType' and parent page number 'pgno'.
** An error code is returned if something goes wrong, otherwise SQLITE_OK.
*/
static int ptrmapPut(BtShared *pBt, Pgno key, u8 eType, Pgno parent){
DbPage *pDbPage; /* The pointer map page */
u8 *pPtrmap; /* The pointer map data */
Pgno iPtrmap; /* The pointer map page number */
int offset; /* Offset in pointer map page */
int rc;
/* The master-journal page number must never be used as a pointer map page */
assert( 0==PTRMAP_ISPAGE(pBt, PENDING_BYTE_PAGE(pBt)) );
assert( pBt->autoVacuum );
if( key==0 ){
return SQLITE_CORRUPT_BKPT;
}
iPtrmap = PTRMAP_PAGENO(pBt, key);
rc = sqlite3PagerGet(pBt->pPager, iPtrmap, &pDbPage);
if( rc!=SQLITE_OK ){
return rc;
}
offset = PTRMAP_PTROFFSET(pBt, key);
pPtrmap = (u8 *)sqlite3PagerGetData(pDbPage);
if( eType!=pPtrmap[offset] || get4byte(&pPtrmap[offset+1])!=parent ){
TRACE(("PTRMAP_UPDATE: %d->(%d,%d)\n", key, eType, parent));
rc = sqlite3PagerWrite(pDbPage);
if( rc==SQLITE_OK ){
pPtrmap[offset] = eType;
put4byte(&pPtrmap[offset+1], parent);
}
}
sqlite3PagerUnref(pDbPage);
return rc;
}
/*
** Read an entry from the pointer map.
**
** This routine retrieves the pointer map entry for page 'key', writing
** the type and parent page number to *pEType and *pPgno respectively.
** An error code is returned if something goes wrong, otherwise SQLITE_OK.
*/
static int ptrmapGet(BtShared *pBt, Pgno key, u8 *pEType, Pgno *pPgno){
DbPage *pDbPage; /* The pointer map page */
int iPtrmap; /* Pointer map page index */
u8 *pPtrmap; /* Pointer map page data */
int offset; /* Offset of entry in pointer map */
int rc;
iPtrmap = PTRMAP_PAGENO(pBt, key);
rc = sqlite3PagerGet(pBt->pPager, iPtrmap, &pDbPage);
if( rc!=0 ){
return rc;
}
pPtrmap = (u8 *)sqlite3PagerGetData(pDbPage);
offset = PTRMAP_PTROFFSET(pBt, key);
assert( pEType!=0 );
*pEType = pPtrmap[offset];
if( pPgno ) *pPgno = get4byte(&pPtrmap[offset+1]);
sqlite3PagerUnref(pDbPage);
if( *pEType<1 || *pEType>5 ) return SQLITE_CORRUPT_BKPT;
return SQLITE_OK;
}
#endif /* SQLITE_OMIT_AUTOVACUUM */
/*
** Given a btree page and a cell index (0 means the first cell on
** the page, 1 means the second cell, and so forth) return a pointer
** to the cell content.
**
** This routine works only for pages that do not contain overflow cells.
*/
#define findCell(pPage, iCell) \
((pPage)->aData + get2byte(&(pPage)->aData[(pPage)->cellOffset+2*(iCell)]))
#ifdef SQLITE_TEST
u8 *sqlite3BtreeFindCell(MemPage *pPage, int iCell){
assert( iCell>=0 );
assert( iCell<get2byte(&pPage->aData[pPage->hdrOffset+3]) );
return findCell(pPage, iCell);
}
#endif
/*
** This a more complex version of sqlite3BtreeFindCell() that works for
** pages that do contain overflow cells. See insert
*/
static u8 *findOverflowCell(MemPage *pPage, int iCell){
int i;
for(i=pPage->nOverflow-1; i>=0; i--){
int k;
struct _OvflCell *pOvfl;
pOvfl = &pPage->aOvfl[i];
k = pOvfl->idx;
if( k<=iCell ){
if( k==iCell ){
return pOvfl->pCell;
}
iCell--;
}
}
return findCell(pPage, iCell);
}
/*
** Parse a cell content block and fill in the CellInfo structure. There
** are two versions of this function. sqlite3BtreeParseCell() takes a
** cell index as the second argument and sqlite3BtreeParseCellPtr()
** takes a pointer to the body of the cell as its second argument.
**
** Within this file, the parseCell() macro can be called instead of
** sqlite3BtreeParseCellPtr(). Using some compilers, this will be faster.
*/
void sqlite3BtreeParseCellPtr(
MemPage *pPage, /* Page containing the cell */
u8 *pCell, /* Pointer to the cell text. */
CellInfo *pInfo /* Fill in this structure */
){
int n; /* Number bytes in cell content header */
u32 nPayload; /* Number of bytes of cell payload */
pInfo->pCell = pCell;
assert( pPage->leaf==0 || pPage->leaf==1 );
n = pPage->childPtrSize;
assert( n==4-4*pPage->leaf );
if( pPage->hasData ){
n += getVarint32(&pCell[n], &nPayload);
}else{
nPayload = 0;
}
pInfo->nData = nPayload;
if( pPage->intKey ){
n += getVarint(&pCell[n], (u64 *)&pInfo->nKey);
}else{
u32 x;
n += getVarint32(&pCell[n], &x);
pInfo->nKey = x;
nPayload += x;
}
pInfo->nPayload = nPayload;
pInfo->nHeader = n;
if( nPayload<=pPage->maxLocal ){
/* This is the (easy) common case where the entire payload fits
** on the local page. No overflow is required.
*/
int nSize; /* Total size of cell content in bytes */
pInfo->nLocal = nPayload;
pInfo->iOverflow = 0;
nSize = nPayload + n;
if( nSize<4 ){
nSize = 4; /* Minimum cell size is 4 */
}
pInfo->nSize = nSize;
}else{
/* If the payload will not fit completely on the local page, we have
** to decide how much to store locally and how much to spill onto
** overflow pages. The strategy is to minimize the amount of unused
** space on overflow pages while keeping the amount of local storage
** in between minLocal and maxLocal.
**
** Warning: changing the way overflow payload is distributed in any
** way will result in an incompatible file format.
*/
int minLocal; /* Minimum amount of payload held locally */
int maxLocal; /* Maximum amount of payload held locally */
int surplus; /* Overflow payload available for local storage */
minLocal = pPage->minLocal;
maxLocal = pPage->maxLocal;
surplus = minLocal + (nPayload - minLocal)%(pPage->pBt->usableSize - 4);
if( surplus <= maxLocal ){
pInfo->nLocal = surplus;
}else{
pInfo->nLocal = minLocal;
}
pInfo->iOverflow = pInfo->nLocal + n;
pInfo->nSize = pInfo->iOverflow + 4;
}
}
#define parseCell(pPage, iCell, pInfo) \
sqlite3BtreeParseCellPtr((pPage), findCell((pPage), (iCell)), (pInfo))
void sqlite3BtreeParseCell(
MemPage *pPage, /* Page containing the cell */
int iCell, /* The cell index. First cell is 0 */
CellInfo *pInfo /* Fill in this structure */
){
parseCell(pPage, iCell, pInfo);
}
/*
** Compute the total number of bytes that a Cell needs in the cell
** data area of the btree-page. The return number includes the cell
** data header and the local payload, but not any overflow page or
** the space used by the cell pointer.
*/
#ifndef NDEBUG
static int cellSize(MemPage *pPage, int iCell){
CellInfo info;
sqlite3BtreeParseCell(pPage, iCell, &info);
return info.nSize;
}
#endif
static int cellSizePtr(MemPage *pPage, u8 *pCell){
CellInfo info;
sqlite3BtreeParseCellPtr(pPage, pCell, &info);
return info.nSize;
}
#ifndef SQLITE_OMIT_AUTOVACUUM
/*
** If the cell pCell, part of page pPage contains a pointer
** to an overflow page, insert an entry into the pointer-map
** for the overflow page.
*/
static int ptrmapPutOvflPtr(MemPage *pPage, u8 *pCell){
if( pCell ){
CellInfo info;
sqlite3BtreeParseCellPtr(pPage, pCell, &info);
assert( (info.nData+(pPage->intKey?0:info.nKey))==info.nPayload );
if( (info.nData+(pPage->intKey?0:info.nKey))>info.nLocal ){
Pgno ovfl = get4byte(&pCell[info.iOverflow]);
return ptrmapPut(pPage->pBt, ovfl, PTRMAP_OVERFLOW1, pPage->pgno);
}
}
return SQLITE_OK;
}
/*
** If the cell with index iCell on page pPage contains a pointer
** to an overflow page, insert an entry into the pointer-map
** for the overflow page.
*/
static int ptrmapPutOvfl(MemPage *pPage, int iCell){
u8 *pCell;
pCell = findOverflowCell(pPage, iCell);
return ptrmapPutOvflPtr(pPage, pCell);
}
#endif
/*
** Defragment the page given. All Cells are moved to the
** end of the page and all free space is collected into one
** big FreeBlk that occurs in between the header and cell
** pointer array and the cell content area.
*/
static int defragmentPage(MemPage *pPage){
int i; /* Loop counter */
int pc; /* Address of a i-th cell */
int addr; /* Offset of first byte after cell pointer array */
int hdr; /* Offset to the page header */
int size; /* Size of a cell */
int usableSize; /* Number of usable bytes on a page */
int cellOffset; /* Offset to the cell pointer array */
int brk; /* Offset to the cell content area */
int nCell; /* Number of cells on the page */
unsigned char *data; /* The page data */
unsigned char *temp; /* Temp area for cell content */
assert( sqlite3PagerIswriteable(pPage->pDbPage) );
assert( pPage->pBt!=0 );
assert( pPage->pBt->usableSize <= SQLITE_MAX_PAGE_SIZE );
assert( pPage->nOverflow==0 );
temp = sqliteMalloc( pPage->pBt->pageSize );
if( temp==0 ) return SQLITE_NOMEM;
data = pPage->aData;
hdr = pPage->hdrOffset;
cellOffset = pPage->cellOffset;
nCell = pPage->nCell;
assert( nCell==get2byte(&data[hdr+3]) );
usableSize = pPage->pBt->usableSize;
brk = get2byte(&data[hdr+5]);
memcpy(&temp[brk], &data[brk], usableSize - brk);
brk = usableSize;
for(i=0; i<nCell; i++){
u8 *pAddr; /* The i-th cell pointer */
pAddr = &data[cellOffset + i*2];
pc = get2byte(pAddr);
assert( pc<pPage->pBt->usableSize );
size = cellSizePtr(pPage, &temp[pc]);
brk -= size;
memcpy(&data[brk], &temp[pc], size);
put2byte(pAddr, brk);
}
assert( brk>=cellOffset+2*nCell );
put2byte(&data[hdr+5], brk);
data[hdr+1] = 0;
data[hdr+2] = 0;
data[hdr+7] = 0;
addr = cellOffset+2*nCell;
memset(&data[addr], 0, brk-addr);
sqliteFree(temp);
return SQLITE_OK;
}
/*
** Allocate nByte bytes of space on a page.
**
** Return the index into pPage->aData[] of the first byte of
** the new allocation. Or return 0 if there is not enough free
** space on the page to satisfy the allocation request.
**
** If the page contains nBytes of free space but does not contain
** nBytes of contiguous free space, then this routine automatically
** calls defragementPage() to consolidate all free space before
** allocating the new chunk.
*/
static int allocateSpace(MemPage *pPage, int nByte){
int addr, pc, hdr;
int size;
int nFrag;
int top;
int nCell;
int cellOffset;
unsigned char *data;
data = pPage->aData;
assert( sqlite3PagerIswriteable(pPage->pDbPage) );
assert( pPage->pBt );
if( nByte<4 ) nByte = 4;
if( pPage->nFree<nByte || pPage->nOverflow>0 ) return 0;
pPage->nFree -= nByte;
hdr = pPage->hdrOffset;
nFrag = data[hdr+7];
if( nFrag<60 ){
/* Search the freelist looking for a slot big enough to satisfy the
** space request. */
addr = hdr+1;
while( (pc = get2byte(&data[addr]))>0 ){
size = get2byte(&data[pc+2]);
if( size>=nByte ){
if( size<nByte+4 ){
memcpy(&data[addr], &data[pc], 2);
data[hdr+7] = nFrag + size - nByte;
return pc;
}else{
put2byte(&data[pc+2], size-nByte);
return pc + size - nByte;
}
}
addr = pc;
}
}
/* Allocate memory from the gap in between the cell pointer array
** and the cell content area.
*/
top = get2byte(&data[hdr+5]);
nCell = get2byte(&data[hdr+3]);
cellOffset = pPage->cellOffset;
if( nFrag>=60 || cellOffset + 2*nCell > top - nByte ){
if( defragmentPage(pPage) ) return 0;
top = get2byte(&data[hdr+5]);
}
top -= nByte;
assert( cellOffset + 2*nCell <= top );
put2byte(&data[hdr+5], top);
return top;
}
/*
** Return a section of the pPage->aData to the freelist.
** The first byte of the new free block is pPage->aDisk[start]
** and the size of the block is "size" bytes.
**
** Most of the effort here is involved in coalesing adjacent
** free blocks into a single big free block.
*/
static void freeSpace(MemPage *pPage, int start, int size){
int addr, pbegin, hdr;
unsigned char *data = pPage->aData;
assert( pPage->pBt!=0 );
assert( sqlite3PagerIswriteable(pPage->pDbPage) );
assert( start>=pPage->hdrOffset+6+(pPage->leaf?0:4) );
assert( (start + size)<=pPage->pBt->usableSize );
if( size<4 ) size = 4;
#ifdef SQLITE_SECURE_DELETE
/* Overwrite deleted information with zeros when the SECURE_DELETE
** option is enabled at compile-time */
memset(&data[start], 0, size);
#endif
/* Add the space back into the linked list of freeblocks */
hdr = pPage->hdrOffset;
addr = hdr + 1;
while( (pbegin = get2byte(&data[addr]))<start && pbegin>0 ){
assert( pbegin<=pPage->pBt->usableSize-4 );
assert( pbegin>addr );
addr = pbegin;
}
assert( pbegin<=pPage->pBt->usableSize-4 );
assert( pbegin>addr || pbegin==0 );
put2byte(&data[addr], start);
put2byte(&data[start], pbegin);
put2byte(&data[start+2], size);
pPage->nFree += size;
/* Coalesce adjacent free blocks */
addr = pPage->hdrOffset + 1;
while( (pbegin = get2byte(&data[addr]))>0 ){
int pnext, psize;
assert( pbegin>addr );
assert( pbegin<=pPage->pBt->usableSize-4 );
pnext = get2byte(&data[pbegin]);
psize = get2byte(&data[pbegin+2]);
if( pbegin + psize + 3 >= pnext && pnext>0 ){
int frag = pnext - (pbegin+psize);
assert( frag<=data[pPage->hdrOffset+7] );
data[pPage->hdrOffset+7] -= frag;
put2byte(&data[pbegin], get2byte(&data[pnext]));
put2byte(&data[pbegin+2], pnext+get2byte(&data[pnext+2])-pbegin);
}else{
addr = pbegin;
}
}
/* If the cell content area begins with a freeblock, remove it. */
if( data[hdr+1]==data[hdr+5] && data[hdr+2]==data[hdr+6] ){
int top;
pbegin = get2byte(&data[hdr+1]);
memcpy(&data[hdr+1], &data[pbegin], 2);
top = get2byte(&data[hdr+5]);
put2byte(&data[hdr+5], top + get2byte(&data[pbegin+2]));
}
}
/*
** Decode the flags byte (the first byte of the header) for a page
** and initialize fields of the MemPage structure accordingly.
*/
static void decodeFlags(MemPage *pPage, int flagByte){
BtShared *pBt; /* A copy of pPage->pBt */
assert( pPage->hdrOffset==(pPage->pgno==1 ? 100 : 0) );
pPage->intKey = (flagByte & (PTF_INTKEY|PTF_LEAFDATA))!=0;
pPage->zeroData = (flagByte & PTF_ZERODATA)!=0;
pPage->leaf = (flagByte & PTF_LEAF)!=0;
pPage->childPtrSize = 4*(pPage->leaf==0);
pBt = pPage->pBt;
if( flagByte & PTF_LEAFDATA ){
pPage->leafData = 1;
pPage->maxLocal = pBt->maxLeaf;
pPage->minLocal = pBt->minLeaf;
}else{
pPage->leafData = 0;
pPage->maxLocal = pBt->maxLocal;
pPage->minLocal = pBt->minLocal;
}
pPage->hasData = !(pPage->zeroData || (!pPage->leaf && pPage->leafData));
}
/*
** Initialize the auxiliary information for a disk block.
**
** The pParent parameter must be a pointer to the MemPage which
** is the parent of the page being initialized. The root of a
** BTree has no parent and so for that page, pParent==NULL.
**
** Return SQLITE_OK on success. If we see that the page does
** not contain a well-formed database page, then return
** SQLITE_CORRUPT. Note that a return of SQLITE_OK does not
** guarantee that the page is well-formed. It only shows that
** we failed to detect any corruption.
*/
int sqlite3BtreeInitPage(
MemPage *pPage, /* The page to be initialized */
MemPage *pParent /* The parent. Might be NULL */
){
int pc; /* Address of a freeblock within pPage->aData[] */
int hdr; /* Offset to beginning of page header */
u8 *data; /* Equal to pPage->aData */
BtShared *pBt; /* The main btree structure */
int usableSize; /* Amount of usable space on each page */
int cellOffset; /* Offset from start of page to first cell pointer */
int nFree; /* Number of unused bytes on the page */
int top; /* First byte of the cell content area */
pBt = pPage->pBt;
assert( pBt!=0 );
assert( pParent==0 || pParent->pBt==pBt );
assert( pPage->pgno==sqlite3PagerPagenumber(pPage->pDbPage) );
assert( pPage->aData == &((unsigned char*)pPage)[-pBt->pageSize] );
if( pPage->pParent!=pParent && (pPage->pParent!=0 || pPage->isInit) ){
/* The parent page should never change unless the file is corrupt */
return SQLITE_CORRUPT_BKPT;
}
if( pPage->isInit ) return SQLITE_OK;
if( pPage->pParent==0 && pParent!=0 ){
pPage->pParent = pParent;
sqlite3PagerRef(pParent->pDbPage);
}
hdr = pPage->hdrOffset;
data = pPage->aData;
decodeFlags(pPage, data[hdr]);
pPage->nOverflow = 0;
pPage->idxShift = 0;
usableSize = pBt->usableSize;
pPage->cellOffset = cellOffset = hdr + 12 - 4*pPage->leaf;
top = get2byte(&data[hdr+5]);
pPage->nCell = get2byte(&data[hdr+3]);
if( pPage->nCell>MX_CELL(pBt) ){
/* To many cells for a single page. The page must be corrupt */
return SQLITE_CORRUPT_BKPT;
}
if( pPage->nCell==0 && pParent!=0 && pParent->pgno!=1 ){
/* All pages must have at least one cell, except for root pages */
return SQLITE_CORRUPT_BKPT;
}
/* Compute the total free space on the page */
pc = get2byte(&data[hdr+1]);
nFree = data[hdr+7] + top - (cellOffset + 2*pPage->nCell);
while( pc>0 ){
int next, size;
if( pc>usableSize-4 ){
/* Free block is off the page */
return SQLITE_CORRUPT_BKPT;
}
next = get2byte(&data[pc]);
size = get2byte(&data[pc+2]);
if( next>0 && next<=pc+size+3 ){
/* Free blocks must be in accending order */
return SQLITE_CORRUPT_BKPT;
}
nFree += size;
pc = next;
}
pPage->nFree = nFree;
if( nFree>=usableSize ){
/* Free space cannot exceed total page size */
return SQLITE_CORRUPT_BKPT;
}
pPage->isInit = 1;
return SQLITE_OK;
}
/*
** Set up a raw page so that it looks like a database page holding
** no entries.
*/
static void zeroPage(MemPage *pPage, int flags){
unsigned char *data = pPage->aData;
BtShared *pBt = pPage->pBt;
int hdr = pPage->hdrOffset;
int first;
assert( sqlite3PagerPagenumber(pPage->pDbPage)==pPage->pgno );
assert( &data[pBt->pageSize] == (unsigned char*)pPage );
assert( sqlite3PagerIswriteable(pPage->pDbPage) );
memset(&data[hdr], 0, pBt->usableSize - hdr);
data[hdr] = flags;
first = hdr + 8 + 4*((flags&PTF_LEAF)==0);
memset(&data[hdr+1], 0, 4);
data[hdr+7] = 0;
put2byte(&data[hdr+5], pBt->usableSize);
pPage->nFree = pBt->usableSize - first;
decodeFlags(pPage, flags);
pPage->hdrOffset = hdr;
pPage->cellOffset = first;
pPage->nOverflow = 0;
pPage->idxShift = 0;
pPage->nCell = 0;
pPage->isInit = 1;
}
/*
** Get a page from the pager. Initialize the MemPage.pBt and
** MemPage.aData elements if needed.
**
** If the noContent flag is set, it means that we do not care about
** the content of the page at this time. So do not go to the disk
** to fetch the content. Just fill in the content with zeros for now.
** If in the future we call sqlite3PagerWrite() on this page, that
** means we have started to be concerned about content and the disk
** read should occur at that point.
*/
int sqlite3BtreeGetPage(
BtShared *pBt, /* The btree */
Pgno pgno, /* Number of the page to fetch */
MemPage **ppPage, /* Return the page in this parameter */
int noContent /* Do not load page content if true */
){
int rc;
MemPage *pPage;
DbPage *pDbPage;
rc = sqlite3PagerAcquire(pBt->pPager, pgno, (DbPage**)&pDbPage, noContent);
if( rc ) return rc;
pPage = (MemPage *)sqlite3PagerGetExtra(pDbPage);
pPage->aData = sqlite3PagerGetData(pDbPage);
pPage->pDbPage = pDbPage;
pPage->pBt = pBt;
pPage->pgno = pgno;
pPage->hdrOffset = pPage->pgno==1 ? 100 : 0;
*ppPage = pPage;
return SQLITE_OK;
}
/*
** Get a page from the pager and initialize it. This routine
** is just a convenience wrapper around separate calls to
** sqlite3BtreeGetPage() and sqlite3BtreeInitPage().
*/
static int getAndInitPage(
BtShared *pBt, /* The database file */
Pgno pgno, /* Number of the page to get */
MemPage **ppPage, /* Write the page pointer here */
MemPage *pParent /* Parent of the page */
){
int rc;
if( pgno==0 ){
return SQLITE_CORRUPT_BKPT;
}
rc = sqlite3BtreeGetPage(pBt, pgno, ppPage, 0);
if( rc==SQLITE_OK && (*ppPage)->isInit==0 ){
rc = sqlite3BtreeInitPage(*ppPage, pParent);
}
return rc;
}
/*
** Release a MemPage. This should be called once for each prior
** call to sqlite3BtreeGetPage.
*/
static void releasePage(MemPage *pPage){
if( pPage ){
assert( pPage->aData );
assert( pPage->pBt );
assert( &pPage->aData[pPage->pBt->pageSize]==(unsigned char*)pPage );
sqlite3PagerUnref(pPage->pDbPage);
}
}
/*
** This routine is called when the reference count for a page
** reaches zero. We need to unref the pParent pointer when that
** happens.
*/
static void pageDestructor(DbPage *pData, int pageSize){
MemPage *pPage;
assert( (pageSize & 7)==0 );
pPage = (MemPage *)sqlite3PagerGetExtra(pData);
if( pPage->pParent ){
MemPage *pParent = pPage->pParent;
pPage->pParent = 0;
releasePage(pParent);
}
pPage->isInit = 0;
}
/*
** During a rollback, when the pager reloads information into the cache
** so that the cache is restored to its original state at the start of
** the transaction, for each page restored this routine is called.
**
** This routine needs to reset the extra data section at the end of the
** page to agree with the restored data.
*/
static void pageReinit(DbPage *pData, int pageSize){
MemPage *pPage;
assert( (pageSize & 7)==0 );
pPage = (MemPage *)sqlite3PagerGetExtra(pData);
if( pPage->isInit ){
pPage->isInit = 0;
sqlite3BtreeInitPage(pPage, pPage->pParent);
}
}
/*
** Open a database file.
**
** zFilename is the name of the database file. If zFilename is NULL
** a new database with a random name is created. This randomly named
** database file will be deleted when sqlite3BtreeClose() is called.
*/
int sqlite3BtreeOpen(
const char *zFilename, /* Name of the file containing the BTree database */
sqlite3 *pSqlite, /* Associated database handle */
Btree **ppBtree, /* Pointer to new Btree object written here */
int flags /* Options */
){
BtShared *pBt; /* Shared part of btree structure */
Btree *p; /* Handle to return */
int rc = SQLITE_OK;
int nReserve;
unsigned char zDbHeader[100];
#if !defined(SQLITE_OMIT_SHARED_CACHE) && !defined(SQLITE_OMIT_DISKIO)
const ThreadData *pTsdro;
#endif
/* Set the variable isMemdb to true for an in-memory database, or
** false for a file-based database. This symbol is only required if
** either of the shared-data or autovacuum features are compiled
** into the library.
*/
#if !defined(SQLITE_OMIT_SHARED_CACHE) || !defined(SQLITE_OMIT_AUTOVACUUM)
#ifdef SQLITE_OMIT_MEMORYDB
const int isMemdb = 0;
#else
const int isMemdb = zFilename && !strcmp(zFilename, ":memory:");
#endif
#endif
p = sqliteMalloc(sizeof(Btree));
if( !p ){
return SQLITE_NOMEM;
}
p->inTrans = TRANS_NONE;
p->pSqlite = pSqlite;
/* Try to find an existing Btree structure opened on zFilename. */
#if !defined(SQLITE_OMIT_SHARED_CACHE) && !defined(SQLITE_OMIT_DISKIO)
pTsdro = sqlite3ThreadDataReadOnly();
if( pTsdro->useSharedData && zFilename && !isMemdb ){
char *zFullPathname = sqlite3OsFullPathname(zFilename);
if( !zFullPathname ){
sqliteFree(p);
return SQLITE_NOMEM;
}
for(pBt=pTsdro->pBtree; pBt; pBt=pBt->pNext){
assert( pBt->nRef>0 );
if( 0==strcmp(zFullPathname, sqlite3PagerFilename(pBt->pPager)) ){
p->pBt = pBt;
*ppBtree = p;
pBt->nRef++;
sqliteFree(zFullPathname);
return SQLITE_OK;
}
}
sqliteFree(zFullPathname);
}
#endif
/*
** The following asserts make sure that structures used by the btree are
** the right size. This is to guard against size changes that result
** when compiling on a different architecture.
*/
assert( sizeof(i64)==8 || sizeof(i64)==4 );
assert( sizeof(u64)==8 || sizeof(u64)==4 );
assert( sizeof(u32)==4 );
assert( sizeof(u16)==2 );
assert( sizeof(Pgno)==4 );
pBt = sqliteMalloc( sizeof(*pBt) );
if( pBt==0 ){
rc = SQLITE_NOMEM;
goto btree_open_out;
}
rc = sqlite3PagerOpen(&pBt->pPager, zFilename, EXTRA_SIZE, flags);
if( rc==SQLITE_OK ){
rc = sqlite3PagerReadFileheader(pBt->pPager,sizeof(zDbHeader),zDbHeader);
}
if( rc!=SQLITE_OK ){
goto btree_open_out;
}
p->pBt = pBt;
sqlite3PagerSetDestructor(pBt->pPager, pageDestructor);
sqlite3PagerSetReiniter(pBt->pPager, pageReinit);
pBt->pCursor = 0;
pBt->pPage1 = 0;
pBt->readOnly = sqlite3PagerIsreadonly(pBt->pPager);
pBt->pageSize = get2byte(&zDbHeader[16]);
if( pBt->pageSize<512 || pBt->pageSize>SQLITE_MAX_PAGE_SIZE
|| ((pBt->pageSize-1)&pBt->pageSize)!=0 ){
pBt->pageSize = SQLITE_DEFAULT_PAGE_SIZE;
pBt->maxEmbedFrac = 64; /* 25% */
pBt->minEmbedFrac = 32; /* 12.5% */
pBt->minLeafFrac = 32; /* 12.5% */
#ifndef SQLITE_OMIT_AUTOVACUUM
/* If the magic name ":memory:" will create an in-memory database, then
** leave the autoVacuum mode at 0 (do not auto-vacuum), even if
** SQLITE_DEFAULT_AUTOVACUUM is true. On the other hand, if
** SQLITE_OMIT_MEMORYDB has been defined, then ":memory:" is just a
** regular file-name. In this case the auto-vacuum applies as per normal.
*/
if( zFilename && !isMemdb ){
pBt->autoVacuum = (SQLITE_DEFAULT_AUTOVACUUM ? 1 : 0);
pBt->incrVacuum = (SQLITE_DEFAULT_AUTOVACUUM==2 ? 1 : 0);
}
#endif
nReserve = 0;
}else{
nReserve = zDbHeader[20];
pBt->maxEmbedFrac = zDbHeader[21];
pBt->minEmbedFrac = zDbHeader[22];
pBt->minLeafFrac = zDbHeader[23];
pBt->pageSizeFixed = 1;
#ifndef SQLITE_OMIT_AUTOVACUUM
pBt->autoVacuum = (get4byte(&zDbHeader[36 + 4*4])?1:0);
pBt->incrVacuum = (get4byte(&zDbHeader[36 + 7*4])?1:0);
#endif
}
pBt->usableSize = pBt->pageSize - nReserve;
assert( (pBt->pageSize & 7)==0 ); /* 8-byte alignment of pageSize */
sqlite3PagerSetPagesize(pBt->pPager, pBt->pageSize);
#if !defined(SQLITE_OMIT_SHARED_CACHE) && !defined(SQLITE_OMIT_DISKIO)
/* Add the new btree to the linked list starting at ThreadData.pBtree.
** There is no chance that a malloc() may fail inside of the
** sqlite3ThreadData() call, as the ThreadData structure must have already
** been allocated for pTsdro->useSharedData to be non-zero.
*/
if( pTsdro->useSharedData && zFilename && !isMemdb ){
pBt->pNext = pTsdro->pBtree;
sqlite3ThreadData()->pBtree = pBt;
}
#endif
pBt->nRef = 1;
*ppBtree = p;
btree_open_out:
if( rc!=SQLITE_OK ){
if( pBt && pBt->pPager ){
sqlite3PagerClose(pBt->pPager);
}
sqliteFree(pBt);
sqliteFree(p);
*ppBtree = 0;
}
return rc;
}
/*
** Close an open database and invalidate all cursors.
*/
int sqlite3BtreeClose(Btree *p){
BtShared *pBt = p->pBt;
BtCursor *pCur;
#ifndef SQLITE_OMIT_SHARED_CACHE
ThreadData *pTsd;
#endif
/* Close all cursors opened via this handle. */
pCur = pBt->pCursor;
while( pCur ){
BtCursor *pTmp = pCur;
pCur = pCur->pNext;
if( pTmp->pBtree==p ){
sqlite3BtreeCloseCursor(pTmp);
}
}
/* Rollback any active transaction and free the handle structure.
** The call to sqlite3BtreeRollback() drops any table-locks held by
** this handle.
*/
sqlite3BtreeRollback(p);
sqliteFree(p);
#ifndef SQLITE_OMIT_SHARED_CACHE
/* If there are still other outstanding references to the shared-btree
** structure, return now. The remainder of this procedure cleans
** up the shared-btree.
*/
assert( pBt->nRef>0 );
pBt->nRef--;
if( pBt->nRef ){
return SQLITE_OK;
}
/* Remove the shared-btree from the thread wide list. Call
** ThreadDataReadOnly() and then cast away the const property of the
** pointer to avoid allocating thread data if it is not really required.
*/
pTsd = (ThreadData *)sqlite3ThreadDataReadOnly();
if( pTsd->pBtree==pBt ){
assert( pTsd==sqlite3ThreadData() );
pTsd->pBtree = pBt->pNext;
}else{
BtShared *pPrev;
for(pPrev=pTsd->pBtree; pPrev && pPrev->pNext!=pBt; pPrev=pPrev->pNext){}
if( pPrev ){
assert( pTsd==sqlite3ThreadData() );
pPrev->pNext = pBt->pNext;
}
}
#endif
/* Close the pager and free the shared-btree structure */
assert( !pBt->pCursor );
sqlite3PagerClose(pBt->pPager);
if( pBt->xFreeSchema && pBt->pSchema ){
pBt->xFreeSchema(pBt->pSchema);
}
sqliteFree(pBt->pSchema);
sqliteFree(pBt);
return SQLITE_OK;
}
/*
** Change the busy handler callback function.
*/
int sqlite3BtreeSetBusyHandler(Btree *p, BusyHandler *pHandler){
BtShared *pBt = p->pBt;
pBt->pBusyHandler = pHandler;
sqlite3PagerSetBusyhandler(pBt->pPager, pHandler);
return SQLITE_OK;
}
/*
** Change the limit on the number of pages allowed in the cache.
**
** The maximum number of cache pages is set to the absolute
** value of mxPage. If mxPage is negative, the pager will
** operate asynchronously - it will not stop to do fsync()s
** to insure data is written to the disk surface before
** continuing. Transactions still work if synchronous is off,
** and the database cannot be corrupted if this program
** crashes. But if the operating system crashes or there is
** an abrupt power failure when synchronous is off, the database
** could be left in an inconsistent and unrecoverable state.
** Synchronous is on by default so database corruption is not
** normally a worry.
*/
int sqlite3BtreeSetCacheSize(Btree *p, int mxPage){
BtShared *pBt = p->pBt;
sqlite3PagerSetCachesize(pBt->pPager, mxPage);
return SQLITE_OK;
}
/*
** Change the way data is synced to disk in order to increase or decrease
** how well the database resists damage due to OS crashes and power
** failures. Level 1 is the same as asynchronous (no syncs() occur and
** there is a high probability of damage) Level 2 is the default. There
** is a very low but non-zero probability of damage. Level 3 reduces the
** probability of damage to near zero but with a write performance reduction.
*/
#ifndef SQLITE_OMIT_PAGER_PRAGMAS
int sqlite3BtreeSetSafetyLevel(Btree *p, int level, int fullSync){
BtShared *pBt = p->pBt;
sqlite3PagerSetSafetyLevel(pBt->pPager, level, fullSync);
return SQLITE_OK;
}
#endif
/*
** Return TRUE if the given btree is set to safety level 1. In other
** words, return TRUE if no sync() occurs on the disk files.
*/
int sqlite3BtreeSyncDisabled(Btree *p){
BtShared *pBt = p->pBt;
assert( pBt && pBt->pPager );
return sqlite3PagerNosync(pBt->pPager);
}
#if !defined(SQLITE_OMIT_PAGER_PRAGMAS) || !defined(SQLITE_OMIT_VACUUM)
/*
** Change the default pages size and the number of reserved bytes per page.
**
** The page size must be a power of 2 between 512 and 65536. If the page
** size supplied does not meet this constraint then the page size is not
** changed.
**
** Page sizes are constrained to be a power of two so that the region
** of the database file used for locking (beginning at PENDING_BYTE,
** the first byte past the 1GB boundary, 0x40000000) needs to occur
** at the beginning of a page.
**
** If parameter nReserve is less than zero, then the number of reserved
** bytes per page is left unchanged.
*/
int sqlite3BtreeSetPageSize(Btree *p, int pageSize, int nReserve){
BtShared *pBt = p->pBt;
if( pBt->pageSizeFixed ){
return SQLITE_READONLY;
}
if( nReserve<0 ){
nReserve = pBt->pageSize - pBt->usableSize;
}
if( pageSize>=512 && pageSize<=SQLITE_MAX_PAGE_SIZE &&
((pageSize-1)&pageSize)==0 ){
assert( (pageSize & 7)==0 );
assert( !pBt->pPage1 && !pBt->pCursor );
pBt->pageSize = sqlite3PagerSetPagesize(pBt->pPager, pageSize);
}
pBt->usableSize = pBt->pageSize - nReserve;
return SQLITE_OK;
}
/*
** Return the currently defined page size
*/
int sqlite3BtreeGetPageSize(Btree *p){
return p->pBt->pageSize;
}
int sqlite3BtreeGetReserve(Btree *p){
return p->pBt->pageSize - p->pBt->usableSize;
}
/*
** Set the maximum page count for a database if mxPage is positive.
** No changes are made if mxPage is 0 or negative.
** Regardless of the value of mxPage, return the maximum page count.
*/
int sqlite3BtreeMaxPageCount(Btree *p, int mxPage){
return sqlite3PagerMaxPageCount(p->pBt->pPager, mxPage);
}
#endif /* !defined(SQLITE_OMIT_PAGER_PRAGMAS) || !defined(SQLITE_OMIT_VACUUM) */
/*
** Change the 'auto-vacuum' property of the database. If the 'autoVacuum'
** parameter is non-zero, then auto-vacuum mode is enabled. If zero, it
** is disabled. The default value for the auto-vacuum property is
** determined by the SQLITE_DEFAULT_AUTOVACUUM macro.
*/
int sqlite3BtreeSetAutoVacuum(Btree *p, int autoVacuum){
#ifdef SQLITE_OMIT_AUTOVACUUM
return SQLITE_READONLY;
#else
BtShared *pBt = p->pBt;
int av = (autoVacuum?1:0);
if( pBt->pageSizeFixed && av!=pBt->autoVacuum ){
return SQLITE_READONLY;
}
pBt->autoVacuum = av;
return SQLITE_OK;
#endif
}
/*
** Return the value of the 'auto-vacuum' property. If auto-vacuum is
** enabled 1 is returned. Otherwise 0.
*/
int sqlite3BtreeGetAutoVacuum(Btree *p){
#ifdef SQLITE_OMIT_AUTOVACUUM
return BTREE_AUTOVACUUM_NONE;
#else
return (
(!p->pBt->autoVacuum)?BTREE_AUTOVACUUM_NONE:
(!p->pBt->incrVacuum)?BTREE_AUTOVACUUM_FULL:
BTREE_AUTOVACUUM_INCR
);
#endif
}
/*
** Get a reference to pPage1 of the database file. This will
** also acquire a readlock on that file.
**
** SQLITE_OK is returned on success. If the file is not a
** well-formed database file, then SQLITE_CORRUPT is returned.
** SQLITE_BUSY is returned if the database is locked. SQLITE_NOMEM
** is returned if we run out of memory.
*/
static int lockBtree(BtShared *pBt){
int rc, pageSize;
MemPage *pPage1;
if( pBt->pPage1 ) return SQLITE_OK;
rc = sqlite3BtreeGetPage(pBt, 1, &pPage1, 0);
if( rc!=SQLITE_OK ) return rc;
/* Do some checking to help insure the file we opened really is
** a valid database file.
*/
rc = SQLITE_NOTADB;
if( sqlite3PagerPagecount(pBt->pPager)>0 ){
u8 *page1 = pPage1->aData;
if( memcmp(page1, zMagicHeader, 16)!=0 ){
goto page1_init_failed;
}
if( page1[18]>1 ){
pBt->readOnly = 1;
}
if( page1[19]>1 ){
goto page1_init_failed;
}
pageSize = get2byte(&page1[16]);
if( ((pageSize-1)&pageSize)!=0 || pageSize<512 ){
goto page1_init_failed;
}
assert( (pageSize & 7)==0 );
pBt->pageSize = pageSize;
pBt->usableSize = pageSize - page1[20];
if( pBt->usableSize<500 ){
goto page1_init_failed;
}
pBt->maxEmbedFrac = page1[21];
pBt->minEmbedFrac = page1[22];
pBt->minLeafFrac = page1[23];
#ifndef SQLITE_OMIT_AUTOVACUUM
pBt->autoVacuum = (get4byte(&page1[36 + 4*4])?1:0);
pBt->incrVacuum = (get4byte(&page1[36 + 7*4])?1:0);
#endif
}
/* maxLocal is the maximum amount of payload to store locally for
** a cell. Make sure it is small enough so that at least minFanout
** cells can will fit on one page. We assume a 10-byte page header.
** Besides the payload, the cell must store:
** 2-byte pointer to the cell
** 4-byte child pointer
** 9-byte nKey value
** 4-byte nData value
** 4-byte overflow page pointer
** So a cell consists of a 2-byte poiner, a header which is as much as
** 17 bytes long, 0 to N bytes of payload, and an optional 4 byte overflow
** page pointer.
*/
pBt->maxLocal = (pBt->usableSize-12)*pBt->maxEmbedFrac/255 - 23;
pBt->minLocal = (pBt->usableSize-12)*pBt->minEmbedFrac/255 - 23;
pBt->maxLeaf = pBt->usableSize - 35;
pBt->minLeaf = (pBt->usableSize-12)*pBt->minLeafFrac/255 - 23;
if( pBt->minLocal>pBt->maxLocal || pBt->maxLocal<0 ){
goto page1_init_failed;
}
assert( pBt->maxLeaf + 23 <= MX_CELL_SIZE(pBt) );
pBt->pPage1 = pPage1;
return SQLITE_OK;
page1_init_failed:
releasePage(pPage1);
pBt->pPage1 = 0;
return rc;
}
/*
** This routine works like lockBtree() except that it also invokes the
** busy callback if there is lock contention.
*/
static int lockBtreeWithRetry(Btree *pRef){
int rc = SQLITE_OK;
if( pRef->inTrans==TRANS_NONE ){
u8 inTransaction = pRef->pBt->inTransaction;
btreeIntegrity(pRef);
rc = sqlite3BtreeBeginTrans(pRef, 0);
pRef->pBt->inTransaction = inTransaction;
pRef->inTrans = TRANS_NONE;
if( rc==SQLITE_OK ){
pRef->pBt->nTransaction--;
}
btreeIntegrity(pRef);
}
return rc;
}
/*
** If there are no outstanding cursors and we are not in the middle
** of a transaction but there is a read lock on the database, then
** this routine unrefs the first page of the database file which
** has the effect of releasing the read lock.
**
** If there are any outstanding cursors, this routine is a no-op.
**
** If there is a transaction in progress, this routine is a no-op.
*/
static void unlockBtreeIfUnused(BtShared *pBt){
if( pBt->inTransaction==TRANS_NONE && pBt->pCursor==0 && pBt->pPage1!=0 ){
if( sqlite3PagerRefcount(pBt->pPager)>=1 ){
if( pBt->pPage1->aData==0 ){
MemPage *pPage = pBt->pPage1;
pPage->aData = &((u8*)pPage)[-pBt->pageSize];
pPage->pBt = pBt;
pPage->pgno = 1;
}
releasePage(pBt->pPage1);
}
pBt->pPage1 = 0;
pBt->inStmt = 0;
}
}
/*
** Create a new database by initializing the first page of the
** file.
*/
static int newDatabase(BtShared *pBt){
MemPage *pP1;
unsigned char *data;
int rc;
if( sqlite3PagerPagecount(pBt->pPager)>0 ) return SQLITE_OK;
pP1 = pBt->pPage1;
assert( pP1!=0 );
data = pP1->aData;
rc = sqlite3PagerWrite(pP1->pDbPage);
if( rc ) return rc;
memcpy(data, zMagicHeader, sizeof(zMagicHeader));
assert( sizeof(zMagicHeader)==16 );
put2byte(&data[16], pBt->pageSize);
data[18] = 1;
data[19] = 1;
data[20] = pBt->pageSize - pBt->usableSize;
data[21] = pBt->maxEmbedFrac;
data[22] = pBt->minEmbedFrac;
data[23] = pBt->minLeafFrac;
memset(&data[24], 0, 100-24);
zeroPage(pP1, PTF_INTKEY|PTF_LEAF|PTF_LEAFDATA );
pBt->pageSizeFixed = 1;
#ifndef SQLITE_OMIT_AUTOVACUUM
assert( pBt->autoVacuum==1 || pBt->autoVacuum==0 );
assert( pBt->incrVacuum==1 || pBt->incrVacuum==0 );
put4byte(&data[36 + 4*4], pBt->autoVacuum);
put4byte(&data[36 + 7*4], pBt->incrVacuum);
#endif
return SQLITE_OK;
}
/*
** Attempt to start a new transaction. A write-transaction
** is started if the second argument is nonzero, otherwise a read-
** transaction. If the second argument is 2 or more and exclusive
** transaction is started, meaning that no other process is allowed
** to access the database. A preexisting transaction may not be
** upgraded to exclusive by calling this routine a second time - the
** exclusivity flag only works for a new transaction.
**
** A write-transaction must be started before attempting any
** changes to the database. None of the following routines
** will work unless a transaction is started first:
**
** sqlite3BtreeCreateTable()
** sqlite3BtreeCreateIndex()
** sqlite3BtreeClearTable()
** sqlite3BtreeDropTable()
** sqlite3BtreeInsert()
** sqlite3BtreeDelete()
** sqlite3BtreeUpdateMeta()
**
** If an initial attempt to acquire the lock fails because of lock contention
** and the database was previously unlocked, then invoke the busy handler
** if there is one. But if there was previously a read-lock, do not
** invoke the busy handler - just return SQLITE_BUSY. SQLITE_BUSY is
** returned when there is already a read-lock in order to avoid a deadlock.
**
** Suppose there are two processes A and B. A has a read lock and B has
** a reserved lock. B tries to promote to exclusive but is blocked because
** of A's read lock. A tries to promote to reserved but is blocked by B.
** One or the other of the two processes must give way or there can be
** no progress. By returning SQLITE_BUSY and not invoking the busy callback
** when A already has a read lock, we encourage A to give up and let B
** proceed.
*/
int sqlite3BtreeBeginTrans(Btree *p, int wrflag){
BtShared *pBt = p->pBt;
int rc = SQLITE_OK;
btreeIntegrity(p);
/* If the btree is already in a write-transaction, or it
** is already in a read-transaction and a read-transaction
** is requested, this is a no-op.
*/
if( p->inTrans==TRANS_WRITE || (p->inTrans==TRANS_READ && !wrflag) ){
return SQLITE_OK;
}
/* Write transactions are not possible on a read-only database */
if( pBt->readOnly && wrflag ){
return SQLITE_READONLY;
}
/* If another database handle has already opened a write transaction
** on this shared-btree structure and a second write transaction is
** requested, return SQLITE_BUSY.
*/
if( pBt->inTransaction==TRANS_WRITE && wrflag ){
return SQLITE_BUSY;
}
do {
if( pBt->pPage1==0 ){
rc = lockBtree(pBt);
}
if( rc==SQLITE_OK && wrflag ){
if( pBt->readOnly ){
rc = SQLITE_READONLY;
}else{
rc = sqlite3PagerBegin(pBt->pPage1->pDbPage, wrflag>1);
if( rc==SQLITE_OK ){
rc = newDatabase(pBt);
}
}
}
if( rc==SQLITE_OK ){
if( wrflag ) pBt->inStmt = 0;
}else{
unlockBtreeIfUnused(pBt);
}
}while( rc==SQLITE_BUSY && pBt->inTransaction==TRANS_NONE &&
sqlite3InvokeBusyHandler(pBt->pBusyHandler) );
if( rc==SQLITE_OK ){
if( p->inTrans==TRANS_NONE ){
pBt->nTransaction++;
}
p->inTrans = (wrflag?TRANS_WRITE:TRANS_READ);
if( p->inTrans>pBt->inTransaction ){
pBt->inTransaction = p->inTrans;
}
}
btreeIntegrity(p);
return rc;
}
#ifndef SQLITE_OMIT_AUTOVACUUM
/*
** Set the pointer-map entries for all children of page pPage. Also, if
** pPage contains cells that point to overflow pages, set the pointer
** map entries for the overflow pages as well.
*/
static int setChildPtrmaps(MemPage *pPage){
int i; /* Counter variable */
int nCell; /* Number of cells in page pPage */
int rc; /* Return code */
BtShared *pBt = pPage->pBt;
int isInitOrig = pPage->isInit;
Pgno pgno = pPage->pgno;
rc = sqlite3BtreeInitPage(pPage, pPage->pParent);
if( rc!=SQLITE_OK ){
goto set_child_ptrmaps_out;
}
nCell = pPage->nCell;
for(i=0; i<nCell; i++){
u8 *pCell = findCell(pPage, i);
rc = ptrmapPutOvflPtr(pPage, pCell);
if( rc!=SQLITE_OK ){
goto set_child_ptrmaps_out;
}
if( !pPage->leaf ){
Pgno childPgno = get4byte(pCell);
rc = ptrmapPut(pBt, childPgno, PTRMAP_BTREE, pgno);
if( rc!=SQLITE_OK ) goto set_child_ptrmaps_out;
}
}
if( !pPage->leaf ){
Pgno childPgno = get4byte(&pPage->aData[pPage->hdrOffset+8]);
rc = ptrmapPut(pBt, childPgno, PTRMAP_BTREE, pgno);
}
set_child_ptrmaps_out:
pPage->isInit = isInitOrig;
return rc;
}
/*
** Somewhere on pPage, which is guarenteed to be a btree page, not an overflow
** page, is a pointer to page iFrom. Modify this pointer so that it points to
** iTo. Parameter eType describes the type of pointer to be modified, as
** follows:
**
** PTRMAP_BTREE: pPage is a btree-page. The pointer points at a child
** page of pPage.
**
** PTRMAP_OVERFLOW1: pPage is a btree-page. The pointer points at an overflow
** page pointed to by one of the cells on pPage.
**
** PTRMAP_OVERFLOW2: pPage is an overflow-page. The pointer points at the next
** overflow page in the list.
*/
static int modifyPagePointer(MemPage *pPage, Pgno iFrom, Pgno iTo, u8 eType){
if( eType==PTRMAP_OVERFLOW2 ){
/* The pointer is always the first 4 bytes of the page in this case. */
if( get4byte(pPage->aData)!=iFrom ){
return SQLITE_CORRUPT_BKPT;
}
put4byte(pPage->aData, iTo);
}else{
int isInitOrig = pPage->isInit;
int i;
int nCell;
sqlite3BtreeInitPage(pPage, 0);
nCell = pPage->nCell;
for(i=0; i<nCell; i++){
u8 *pCell = findCell(pPage, i);
if( eType==PTRMAP_OVERFLOW1 ){
CellInfo info;
sqlite3BtreeParseCellPtr(pPage, pCell, &info);
if( info.iOverflow ){
if( iFrom==get4byte(&pCell[info.iOverflow]) ){
put4byte(&pCell[info.iOverflow], iTo);
break;
}
}
}else{
if( get4byte(pCell)==iFrom ){
put4byte(pCell, iTo);
break;
}
}
}
if( i==nCell ){
if( eType!=PTRMAP_BTREE ||
get4byte(&pPage->aData[pPage->hdrOffset+8])!=iFrom ){
return SQLITE_CORRUPT_BKPT;
}
put4byte(&pPage->aData[pPage->hdrOffset+8], iTo);
}
pPage->isInit = isInitOrig;
}
return SQLITE_OK;
}
/*
** Move the open database page pDbPage to location iFreePage in the
** database. The pDbPage reference remains valid.
*/
static int relocatePage(
BtShared *pBt, /* Btree */
MemPage *pDbPage, /* Open page to move */
u8 eType, /* Pointer map 'type' entry for pDbPage */
Pgno iPtrPage, /* Pointer map 'page-no' entry for pDbPage */
Pgno iFreePage /* The location to move pDbPage to */
){
MemPage *pPtrPage; /* The page that contains a pointer to pDbPage */
Pgno iDbPage = pDbPage->pgno;
Pager *pPager = pBt->pPager;
int rc;
assert( eType==PTRMAP_OVERFLOW2 || eType==PTRMAP_OVERFLOW1 ||
eType==PTRMAP_BTREE || eType==PTRMAP_ROOTPAGE );
/* Move page iDbPage from it's current location to page number iFreePage */
TRACE(("AUTOVACUUM: Moving %d to free page %d (ptr page %d type %d)\n",
iDbPage, iFreePage, iPtrPage, eType));
rc = sqlite3PagerMovepage(pPager, pDbPage->pDbPage, iFreePage);
if( rc!=SQLITE_OK ){
return rc;
}
pDbPage->pgno = iFreePage;
/* If pDbPage was a btree-page, then it may have child pages and/or cells
** that point to overflow pages. The pointer map entries for all these
** pages need to be changed.
**
** If pDbPage is an overflow page, then the first 4 bytes may store a
** pointer to a subsequent overflow page. If this is the case, then
** the pointer map needs to be updated for the subsequent overflow page.
*/
if( eType==PTRMAP_BTREE || eType==PTRMAP_ROOTPAGE ){
rc = setChildPtrmaps(pDbPage);
if( rc!=SQLITE_OK ){
return rc;
}
}else{
Pgno nextOvfl = get4byte(pDbPage->aData);
if( nextOvfl!=0 ){
rc = ptrmapPut(pBt, nextOvfl, PTRMAP_OVERFLOW2, iFreePage);
if( rc!=SQLITE_OK ){
return rc;
}
}
}
/* Fix the database pointer on page iPtrPage that pointed at iDbPage so
** that it points at iFreePage. Also fix the pointer map entry for
** iPtrPage.
*/
if( eType!=PTRMAP_ROOTPAGE ){
rc = sqlite3BtreeGetPage(pBt, iPtrPage, &pPtrPage, 0);
if( rc!=SQLITE_OK ){
return rc;
}
rc = sqlite3PagerWrite(pPtrPage->pDbPage);
if( rc!=SQLITE_OK ){
releasePage(pPtrPage);
return rc;
}
rc = modifyPagePointer(pPtrPage, iDbPage, iFreePage, eType);
releasePage(pPtrPage);
if( rc==SQLITE_OK ){
rc = ptrmapPut(pBt, iFreePage, eType, iPtrPage);
}
}
return rc;
}
/* Forward declaration required by incrVacuumStep(). */
static int allocateBtreePage(BtShared *, MemPage **, Pgno *, Pgno, u8);
/*
** Perform a single step of an incremental-vacuum. If successful,
** return SQLITE_OK. If there is no work to do (and therefore no
** point in calling this function again), return SQLITE_DONE.
**
** More specificly, this function attempts to re-organize the
** database so that the last page of the file currently in use
** is no longer in use.
**
** If the nFin parameter is non-zero, the implementation assumes
** that the caller will keep calling incrVacuumStep() until
** it returns SQLITE_DONE or an error, and that nFin is the
** number of pages the database file will contain after this
** process is complete.
*/
static int incrVacuumStep(BtShared *pBt, Pgno nFin){
Pgno iLastPg; /* Last page in the database */
Pgno nFreeList; /* Number of pages still on the free-list */
iLastPg = pBt->nTrunc;
if( iLastPg==0 ){
iLastPg = sqlite3PagerPagecount(pBt->pPager);
}
if( !PTRMAP_ISPAGE(pBt, iLastPg) && iLastPg!=PENDING_BYTE_PAGE(pBt) ){
int rc;
u8 eType;
Pgno iPtrPage;
nFreeList = get4byte(&pBt->pPage1->aData[36]);
if( nFreeList==0 || nFin==iLastPg ){
return SQLITE_DONE;
}
rc = ptrmapGet(pBt, iLastPg, &eType, &iPtrPage);
if( rc!=SQLITE_OK ){
return rc;
}
if( eType==PTRMAP_ROOTPAGE ){
return SQLITE_CORRUPT_BKPT;
}
if( eType==PTRMAP_FREEPAGE ){
if( nFin==0 ){
/* Remove the page from the files free-list. This is not required
** if nFin is non-zero. In that case, the free-list will be
** truncated to zero after this function returns, so it doesn't
** matter if it still contains some garbage entries.
*/
Pgno iFreePg;
MemPage *pFreePg;
rc = allocateBtreePage(pBt, &pFreePg, &iFreePg, iLastPg, 1);
if( rc!=SQLITE_OK ){
return rc;
}
assert( iFreePg==iLastPg );
releasePage(pFreePg);
}
} else {
Pgno iFreePg; /* Index of free page to move pLastPg to */
MemPage *pLastPg;
rc = sqlite3BtreeGetPage(pBt, iLastPg, &pLastPg, 0);
if( rc!=SQLITE_OK ){
return rc;
}
/* If nFin is zero, this loop runs exactly once and page pLastPg
** is swapped with the first free page pulled off the free list.
**
** On the other hand, if nFin is greater than zero, then keep
** looping until a free-page located within the first nFin pages
** of the file is found.
*/
do {
MemPage *pFreePg;
rc = allocateBtreePage(pBt, &pFreePg, &iFreePg, 0, 0);
if( rc!=SQLITE_OK ){
releasePage(pLastPg);
return rc;
}
releasePage(pFreePg);
}while( nFin!=0 && iFreePg>nFin );
assert( iFreePg<iLastPg );
rc = sqlite3PagerWrite(pLastPg->pDbPage);
if( rc!=SQLITE_OK ){
return rc;
}
rc = relocatePage(pBt, pLastPg, eType, iPtrPage, iFreePg);
releasePage(pLastPg);
if( rc!=SQLITE_OK ){
return rc;
}
}
}
pBt->nTrunc = iLastPg - 1;
while( pBt->nTrunc==PENDING_BYTE_PAGE(pBt)||PTRMAP_ISPAGE(pBt, pBt->nTrunc) ){
pBt->nTrunc--;
}
return SQLITE_OK;
}
/*
** A write-transaction must be opened before calling this function.
** It performs a single unit of work towards an incremental vacuum.
**
** If the incremental vacuum is finished after this function has run,
** SQLITE_DONE is returned. If it is not finished, but no error occured,
** SQLITE_OK is returned. Otherwise an SQLite error code.
*/
int sqlite3BtreeIncrVacuum(Btree *p){
BtShared *pBt = p->pBt;
assert( pBt->inTransaction==TRANS_WRITE && p->inTrans==TRANS_WRITE );
if( !pBt->autoVacuum ){
return SQLITE_DONE;
}
invalidateAllOverflowCache(pBt);
return incrVacuumStep(pBt, 0);
}
/*
** This routine is called prior to sqlite3PagerCommit when a transaction
** is commited for an auto-vacuum database.
**
** If SQLITE_OK is returned, then *pnTrunc is set to the number of pages
** the database file should be truncated to during the commit process.
** i.e. the database has been reorganized so that only the first *pnTrunc
** pages are in use.
*/
static int autoVacuumCommit(BtShared *pBt, Pgno *pnTrunc){
int rc = SQLITE_OK;
Pager *pPager = pBt->pPager;
#ifndef NDEBUG
int nRef = sqlite3PagerRefcount(pPager);
#endif
invalidateAllOverflowCache(pBt);
assert(pBt->autoVacuum);
if( !pBt->incrVacuum ){
Pgno nFin = 0;
if( pBt->nTrunc==0 ){
Pgno nFree;
Pgno nPtrmap;
const int pgsz = pBt->pageSize;
Pgno nOrig = sqlite3PagerPagecount(pBt->pPager);
if( PTRMAP_ISPAGE(pBt, nOrig) ){
return SQLITE_CORRUPT_BKPT;
}
if( nOrig==PENDING_BYTE_PAGE(pBt) ){
nOrig--;
}
nFree = get4byte(&pBt->pPage1->aData[36]);
nPtrmap = (nFree-nOrig+PTRMAP_PAGENO(pBt, nOrig)+pgsz/5)/(pgsz/5);
nFin = nOrig - nFree - nPtrmap;
if( nOrig>PENDING_BYTE_PAGE(pBt) && nFin<=PENDING_BYTE_PAGE(pBt) ){
nFin--;
}
while( PTRMAP_ISPAGE(pBt, nFin) || nFin==PENDING_BYTE_PAGE(pBt) ){
nFin--;
}
}
while( rc==SQLITE_OK ){
rc = incrVacuumStep(pBt, nFin);
}
if( rc==SQLITE_DONE ){
assert(nFin==0 || pBt->nTrunc==0 || nFin<=pBt->nTrunc);
rc = SQLITE_OK;
if( pBt->nTrunc ){
rc = sqlite3PagerWrite(pBt->pPage1->pDbPage);
put4byte(&pBt->pPage1->aData[32], 0);
put4byte(&pBt->pPage1->aData[36], 0);
pBt->nTrunc = nFin;
}
}
if( rc!=SQLITE_OK ){
sqlite3PagerRollback(pPager);
}
}
if( rc==SQLITE_OK ){
*pnTrunc = pBt->nTrunc;
pBt->nTrunc = 0;
}
assert( nRef==sqlite3PagerRefcount(pPager) );
return rc;
}
#endif
/*
** This routine does the first phase of a two-phase commit. This routine
** causes a rollback journal to be created (if it does not already exist)
** and populated with enough information so that if a power loss occurs
** the database can be restored to its original state by playing back
** the journal. Then the contents of the journal are flushed out to
** the disk. After the journal is safely on oxide, the changes to the
** database are written into the database file and flushed to oxide.
** At the end of this call, the rollback journal still exists on the
** disk and we are still holding all locks, so the transaction has not
** committed. See sqlite3BtreeCommit() for the second phase of the
** commit process.
**
** This call is a no-op if no write-transaction is currently active on pBt.
**
** Otherwise, sync the database file for the btree pBt. zMaster points to
** the name of a master journal file that should be written into the
** individual journal file, or is NULL, indicating no master journal file
** (single database transaction).
**
** When this is called, the master journal should already have been
** created, populated with this journal pointer and synced to disk.
**
** Once this is routine has returned, the only thing required to commit
** the write-transaction for this database file is to delete the journal.
*/
int sqlite3BtreeCommitPhaseOne(Btree *p, const char *zMaster){
int rc = SQLITE_OK;
if( p->inTrans==TRANS_WRITE ){
BtShared *pBt = p->pBt;
Pgno nTrunc = 0;
#ifndef SQLITE_OMIT_AUTOVACUUM
if( pBt->autoVacuum ){
rc = autoVacuumCommit(pBt, &nTrunc);
if( rc!=SQLITE_OK ){
return rc;
}
}
#endif
rc = sqlite3PagerCommitPhaseOne(pBt->pPager, zMaster, nTrunc);
}
return rc;
}
/*
** Commit the transaction currently in progress.
**
** This routine implements the second phase of a 2-phase commit. The
** sqlite3BtreeSync() routine does the first phase and should be invoked
** prior to calling this routine. The sqlite3BtreeSync() routine did
** all the work of writing information out to disk and flushing the
** contents so that they are written onto the disk platter. All this
** routine has to do is delete or truncate the rollback journal
** (which causes the transaction to commit) and drop locks.
**
** This will release the write lock on the database file. If there
** are no active cursors, it also releases the read lock.
*/
int sqlite3BtreeCommitPhaseTwo(Btree *p){
BtShared *pBt = p->pBt;
btreeIntegrity(p);
/* If the handle has a write-transaction open, commit the shared-btrees
** transaction and set the shared state to TRANS_READ.
*/
if( p->inTrans==TRANS_WRITE ){
int rc;
assert( pBt->inTransaction==TRANS_WRITE );
assert( pBt->nTransaction>0 );
rc = sqlite3PagerCommitPhaseTwo(pBt->pPager);
if( rc!=SQLITE_OK ){
return rc;
}
pBt->inTransaction = TRANS_READ;
pBt->inStmt = 0;
}
unlockAllTables(p);
/* If the handle has any kind of transaction open, decrement the transaction
** count of the shared btree. If the transaction count reaches 0, set
** the shared state to TRANS_NONE. The unlockBtreeIfUnused() call below
** will unlock the pager.
*/
if( p->inTrans!=TRANS_NONE ){
pBt->nTransaction--;
if( 0==pBt->nTransaction ){
pBt->inTransaction = TRANS_NONE;
}
}
/* Set the handles current transaction state to TRANS_NONE and unlock
** the pager if this call closed the only read or write transaction.
*/
p->inTrans = TRANS_NONE;
unlockBtreeIfUnused(pBt);
btreeIntegrity(p);
return SQLITE_OK;
}
/*
** Do both phases of a commit.
*/
int sqlite3BtreeCommit(Btree *p){
int rc;
rc = sqlite3BtreeCommitPhaseOne(p, 0);
if( rc==SQLITE_OK ){
rc = sqlite3BtreeCommitPhaseTwo(p);
}
return rc;
}
#ifndef NDEBUG
/*
** Return the number of write-cursors open on this handle. This is for use
** in assert() expressions, so it is only compiled if NDEBUG is not
** defined.
*/
static int countWriteCursors(BtShared *pBt){
BtCursor *pCur;
int r = 0;
for(pCur=pBt->pCursor; pCur; pCur=pCur->pNext){
if( pCur->wrFlag ) r++;
}
return r;
}
#endif
/*
** Rollback the transaction in progress. All cursors will be
** invalided by this operation. Any attempt to use a cursor
** that was open at the beginning of this operation will result
** in an error.
**
** This will release the write lock on the database file. If there
** are no active cursors, it also releases the read lock.
*/
int sqlite3BtreeRollback(Btree *p){
int rc;
BtShared *pBt = p->pBt;
MemPage *pPage1;
rc = saveAllCursors(pBt, 0, 0);
#ifndef SQLITE_OMIT_SHARED_CACHE
if( rc!=SQLITE_OK ){
/* This is a horrible situation. An IO or malloc() error occured whilst
** trying to save cursor positions. If this is an automatic rollback (as
** the result of a constraint, malloc() failure or IO error) then
** the cache may be internally inconsistent (not contain valid trees) so
** we cannot simply return the error to the caller. Instead, abort
** all queries that may be using any of the cursors that failed to save.
*/
while( pBt->pCursor ){
sqlite3 *db = pBt->pCursor->pBtree->pSqlite;
if( db ){
sqlite3AbortOtherActiveVdbes(db, 0);
}
}
}
#endif
btreeIntegrity(p);
unlockAllTables(p);
if( p->inTrans==TRANS_WRITE ){
int rc2;
#ifndef SQLITE_OMIT_AUTOVACUUM
pBt->nTrunc = 0;
#endif
assert( TRANS_WRITE==pBt->inTransaction );
rc2 = sqlite3PagerRollback(pBt->pPager);
if( rc2!=SQLITE_OK ){
rc = rc2;
}
/* The rollback may have destroyed the pPage1->aData value. So
** call sqlite3BtreeGetPage() on page 1 again to make
** sure pPage1->aData is set correctly. */
if( sqlite3BtreeGetPage(pBt, 1, &pPage1, 0)==SQLITE_OK ){
releasePage(pPage1);
}
assert( countWriteCursors(pBt)==0 );
pBt->inTransaction = TRANS_READ;
}
if( p->inTrans!=TRANS_NONE ){
assert( pBt->nTransaction>0 );
pBt->nTransaction--;
if( 0==pBt->nTransaction ){
pBt->inTransaction = TRANS_NONE;
}
}
p->inTrans = TRANS_NONE;
pBt->inStmt = 0;
unlockBtreeIfUnused(pBt);
btreeIntegrity(p);
return rc;
}
/*
** Start a statement subtransaction. The subtransaction can
** can be rolled back independently of the main transaction.
** You must start a transaction before starting a subtransaction.
** The subtransaction is ended automatically if the main transaction
** commits or rolls back.
**
** Only one subtransaction may be active at a time. It is an error to try
** to start a new subtransaction if another subtransaction is already active.
**
** Statement subtransactions are used around individual SQL statements
** that are contained within a BEGIN...COMMIT block. If a constraint
** error occurs within the statement, the effect of that one statement
** can be rolled back without having to rollback the entire transaction.
*/
int sqlite3BtreeBeginStmt(Btree *p){
int rc;
BtShared *pBt = p->pBt;
if( (p->inTrans!=TRANS_WRITE) || pBt->inStmt ){
return pBt->readOnly ? SQLITE_READONLY : SQLITE_ERROR;
}
assert( pBt->inTransaction==TRANS_WRITE );
rc = pBt->readOnly ? SQLITE_OK : sqlite3PagerStmtBegin(pBt->pPager);
pBt->inStmt = 1;
return rc;
}
/*
** Commit the statment subtransaction currently in progress. If no
** subtransaction is active, this is a no-op.
*/
int sqlite3BtreeCommitStmt(Btree *p){
int rc;
BtShared *pBt = p->pBt;
if( pBt->inStmt && !pBt->readOnly ){
rc = sqlite3PagerStmtCommit(pBt->pPager);
}else{
rc = SQLITE_OK;
}
pBt->inStmt = 0;
return rc;
}
/*
** Rollback the active statement subtransaction. If no subtransaction
** is active this routine is a no-op.
**
** All cursors will be invalidated by this operation. Any attempt
** to use a cursor that was open at the beginning of this operation
** will result in an error.
*/
int sqlite3BtreeRollbackStmt(Btree *p){
int rc = SQLITE_OK;
BtShared *pBt = p->pBt;
sqlite3MallocDisallow();
if( pBt->inStmt && !pBt->readOnly ){
rc = sqlite3PagerStmtRollback(pBt->pPager);
assert( countWriteCursors(pBt)==0 );
pBt->inStmt = 0;
}
sqlite3MallocAllow();
return rc;
}
/*
** Default key comparison function to be used if no comparison function
** is specified on the sqlite3BtreeCursor() call.
*/
static int dfltCompare(
void *NotUsed, /* User data is not used */
int n1, const void *p1, /* First key to compare */
int n2, const void *p2 /* Second key to compare */
){
int c;
c = memcmp(p1, p2, n1<n2 ? n1 : n2);
if( c==0 ){
c = n1 - n2;
}
return c;
}
/*
** Create a new cursor for the BTree whose root is on the page
** iTable. The act of acquiring a cursor gets a read lock on
** the database file.
**
** If wrFlag==0, then the cursor can only be used for reading.
** If wrFlag==1, then the cursor can be used for reading or for
** writing if other conditions for writing are also met. These
** are the conditions that must be met in order for writing to
** be allowed:
**
** 1: The cursor must have been opened with wrFlag==1
**
** 2: Other database connections that share the same pager cache
** but which are not in the READ_UNCOMMITTED state may not have
** cursors open with wrFlag==0 on the same table. Otherwise
** the changes made by this write cursor would be visible to
** the read cursors in the other database connection.
**
** 3: The database must be writable (not on read-only media)
**
** 4: There must be an active transaction.
**
** No checking is done to make sure that page iTable really is the
** root page of a b-tree. If it is not, then the cursor acquired
** will not work correctly.
**
** The comparison function must be logically the same for every cursor
** on a particular table. Changing the comparison function will result
** in incorrect operations. If the comparison function is NULL, a
** default comparison function is used. The comparison function is
** always ignored for INTKEY tables.
*/
int sqlite3BtreeCursor(
Btree *p, /* The btree */
int iTable, /* Root page of table to open */
int wrFlag, /* 1 to write. 0 read-only */
int (*xCmp)(void*,int,const void*,int,const void*), /* Key Comparison func */
void *pArg, /* First arg to xCompare() */
BtCursor **ppCur /* Write new cursor here */
){
int rc;
BtCursor *pCur;
BtShared *pBt = p->pBt;
*ppCur = 0;
if( wrFlag ){
if( pBt->readOnly ){
return SQLITE_READONLY;
}
if( checkReadLocks(p, iTable, 0) ){
return SQLITE_LOCKED;
}
}
if( pBt->pPage1==0 ){
rc = lockBtreeWithRetry(p);
if( rc!=SQLITE_OK ){
return rc;
}
if( pBt->readOnly && wrFlag ){
return SQLITE_READONLY;
}
}
pCur = sqliteMalloc( sizeof(*pCur) );
if( pCur==0 ){
rc = SQLITE_NOMEM;
goto create_cursor_exception;
}
pCur->pgnoRoot = (Pgno)iTable;
if( iTable==1 && sqlite3PagerPagecount(pBt->pPager)==0 ){
rc = SQLITE_EMPTY;
goto create_cursor_exception;
}
rc = getAndInitPage(pBt, pCur->pgnoRoot, &pCur->pPage, 0);
if( rc!=SQLITE_OK ){
goto create_cursor_exception;
}
/* Now that no other errors can occur, finish filling in the BtCursor
** variables, link the cursor into the BtShared list and set *ppCur (the
** output argument to this function).
*/
pCur->xCompare = xCmp ? xCmp : dfltCompare;
pCur->pArg = pArg;
pCur->pBtree = p;
pCur->wrFlag = wrFlag;
pCur->pNext = pBt->pCursor;
if( pCur->pNext ){
pCur->pNext->pPrev = pCur;
}
pBt->pCursor = pCur;
pCur->eState = CURSOR_INVALID;
*ppCur = pCur;
return SQLITE_OK;
create_cursor_exception:
if( pCur ){
releasePage(pCur->pPage);
sqliteFree(pCur);
}
unlockBtreeIfUnused(pBt);
return rc;
}
/*
** Close a cursor. The read lock on the database file is released
** when the last cursor is closed.
*/
int sqlite3BtreeCloseCursor(BtCursor *pCur){
BtShared *pBt = pCur->pBtree->pBt;
clearCursorPosition(pCur);
if( pCur->pPrev ){
pCur->pPrev->pNext = pCur->pNext;
}else{
pBt->pCursor = pCur->pNext;
}
if( pCur->pNext ){
pCur->pNext->pPrev = pCur->pPrev;
}
releasePage(pCur->pPage);
unlockBtreeIfUnused(pBt);
invalidateOverflowCache(pCur);
sqliteFree(pCur);
return SQLITE_OK;
}
/*
** Make a temporary cursor by filling in the fields of pTempCur.
** The temporary cursor is not on the cursor list for the Btree.
*/
void sqlite3BtreeGetTempCursor(BtCursor *pCur, BtCursor *pTempCur){
memcpy(pTempCur, pCur, sizeof(*pCur));
pTempCur->pNext = 0;
pTempCur->pPrev = 0;
if( pTempCur->pPage ){
sqlite3PagerRef(pTempCur->pPage->pDbPage);
}
}
/*
** Delete a temporary cursor such as was made by the CreateTemporaryCursor()
** function above.
*/
void sqlite3BtreeReleaseTempCursor(BtCursor *pCur){
if( pCur->pPage ){
sqlite3PagerUnref(pCur->pPage->pDbPage);
}
}
/*
** Make sure the BtCursor* given in the argument has a valid
** BtCursor.info structure. If it is not already valid, call
** sqlite3BtreeParseCell() to fill it in.
**
** BtCursor.info is a cache of the information in the current cell.
** Using this cache reduces the number of calls to sqlite3BtreeParseCell().
**
** 2007-06-25: There is a bug in some versions of MSVC that cause the
** compiler to crash when getCellInfo() is implemented as a macro.
** But there is a measureable speed advantage to using the macro on gcc
** (when less compiler optimizations like -Os or -O0 are used and the
** compiler is not doing agressive inlining.) So we use a real function
** for MSVC and a macro for everything else. Ticket #2457.
*/
#ifndef NDEBUG
static void assertCellInfo(BtCursor *pCur){
CellInfo info;
memset(&info, 0, sizeof(info));
sqlite3BtreeParseCell(pCur->pPage, pCur->idx, &info);
assert( memcmp(&info, &pCur->info, sizeof(info))==0 );
}
#else
#define assertCellInfo(x)
#endif
#ifdef _MSC_VER
/* Use a real function in MSVC to work around bugs in that compiler. */
static void getCellInfo(BtCursor *pCur){
if( pCur->info.nSize==0 ){
sqlite3BtreeParseCell(pCur->pPage, pCur->idx, &pCur->info);
}else{
assertCellInfo(pCur);
}
}
#else /* if not _MSC_VER */
/* Use a macro in all other compilers so that the function is inlined */
#define getCellInfo(pCur) \
if( pCur->info.nSize==0 ){ \
sqlite3BtreeParseCell(pCur->pPage, pCur->idx, &pCur->info); \
}else{ \
assertCellInfo(pCur); \
}
#endif /* _MSC_VER */
/*
** Set *pSize to the size of the buffer needed to hold the value of
** the key for the current entry. If the cursor is not pointing
** to a valid entry, *pSize is set to 0.
**
** For a table with the INTKEY flag set, this routine returns the key
** itself, not the number of bytes in the key.
*/
int sqlite3BtreeKeySize(BtCursor *pCur, i64 *pSize){
int rc = restoreOrClearCursorPosition(pCur);
if( rc==SQLITE_OK ){
assert( pCur->eState==CURSOR_INVALID || pCur->eState==CURSOR_VALID );
if( pCur->eState==CURSOR_INVALID ){
*pSize = 0;
}else{
getCellInfo(pCur);
*pSize = pCur->info.nKey;
}
}
return rc;
}
/*
** Set *pSize to the number of bytes of data in the entry the
** cursor currently points to. Always return SQLITE_OK.
** Failure is not possible. If the cursor is not currently
** pointing to an entry (which can happen, for example, if
** the database is empty) then *pSize is set to 0.
*/
int sqlite3BtreeDataSize(BtCursor *pCur, u32 *pSize){
int rc = restoreOrClearCursorPosition(pCur);
if( rc==SQLITE_OK ){
assert( pCur->eState==CURSOR_INVALID || pCur->eState==CURSOR_VALID );
if( pCur->eState==CURSOR_INVALID ){
/* Not pointing at a valid entry - set *pSize to 0. */
*pSize = 0;
}else{
getCellInfo(pCur);
*pSize = pCur->info.nData;
}
}
return rc;
}
/*
** Given the page number of an overflow page in the database (parameter
** ovfl), this function finds the page number of the next page in the
** linked list of overflow pages. If possible, it uses the auto-vacuum
** pointer-map data instead of reading the content of page ovfl to do so.
**
** If an error occurs an SQLite error code is returned. Otherwise:
**
** Unless pPgnoNext is NULL, the page number of the next overflow
** page in the linked list is written to *pPgnoNext. If page ovfl
** is the last page in it's linked list, *pPgnoNext is set to zero.
**
** If ppPage is not NULL, *ppPage is set to the MemPage* handle
** for page ovfl. The underlying pager page may have been requested
** with the noContent flag set, so the page data accessable via
** this handle may not be trusted.
*/
static int getOverflowPage(
BtShared *pBt,
Pgno ovfl, /* Overflow page */
MemPage **ppPage, /* OUT: MemPage handle */
Pgno *pPgnoNext /* OUT: Next overflow page number */
){
Pgno next = 0;
int rc;
/* One of these must not be NULL. Otherwise, why call this function? */
assert(ppPage || pPgnoNext);
/* If pPgnoNext is NULL, then this function is being called to obtain
** a MemPage* reference only. No page-data is required in this case.
*/
if( !pPgnoNext ){
return sqlite3BtreeGetPage(pBt, ovfl, ppPage, 1);
}
#ifndef SQLITE_OMIT_AUTOVACUUM
/* Try to find the next page in the overflow list using the
** autovacuum pointer-map pages. Guess that the next page in
** the overflow list is page number (ovfl+1). If that guess turns
** out to be wrong, fall back to loading the data of page
** number ovfl to determine the next page number.
*/
if( pBt->autoVacuum ){
Pgno pgno;
Pgno iGuess = ovfl+1;
u8 eType;
while( PTRMAP_ISPAGE(pBt, iGuess) || iGuess==PENDING_BYTE_PAGE(pBt) ){
iGuess++;
}
if( iGuess<=sqlite3PagerPagecount(pBt->pPager) ){
rc = ptrmapGet(pBt, iGuess, &eType, &pgno);
if( rc!=SQLITE_OK ){
return rc;
}
if( eType==PTRMAP_OVERFLOW2 && pgno==ovfl ){
next = iGuess;
}
}
}
#endif
if( next==0 || ppPage ){
MemPage *pPage = 0;
rc = sqlite3BtreeGetPage(pBt, ovfl, &pPage, next!=0);
assert(rc==SQLITE_OK || pPage==0);
if( next==0 && rc==SQLITE_OK ){
next = get4byte(pPage->aData);
}
if( ppPage ){
*ppPage = pPage;
}else{
releasePage(pPage);
}
}
*pPgnoNext = next;
return rc;
}
/*
** Copy data from a buffer to a page, or from a page to a buffer.
**
** pPayload is a pointer to data stored on database page pDbPage.
** If argument eOp is false, then nByte bytes of data are copied
** from pPayload to the buffer pointed at by pBuf. If eOp is true,
** then sqlite3PagerWrite() is called on pDbPage and nByte bytes
** of data are copied from the buffer pBuf to pPayload.
**
** SQLITE_OK is returned on success, otherwise an error code.
*/
static int copyPayload(
void *pPayload, /* Pointer to page data */
void *pBuf, /* Pointer to buffer */
int nByte, /* Number of bytes to copy */
int eOp, /* 0 -> copy from page, 1 -> copy to page */
DbPage *pDbPage /* Page containing pPayload */
){
if( eOp ){
/* Copy data from buffer to page (a write operation) */
int rc = sqlite3PagerWrite(pDbPage);
if( rc!=SQLITE_OK ){
return rc;
}
memcpy(pPayload, pBuf, nByte);
}else{
/* Copy data from page to buffer (a read operation) */
memcpy(pBuf, pPayload, nByte);
}
return SQLITE_OK;
}
/*
** This function is used to read or overwrite payload information
** for the entry that the pCur cursor is pointing to. If the eOp
** parameter is 0, this is a read operation (data copied into
** buffer pBuf). If it is non-zero, a write (data copied from
** buffer pBuf).
**
** A total of "amt" bytes are read or written beginning at "offset".
** Data is read to or from the buffer pBuf.
**
** This routine does not make a distinction between key and data.
** It just reads or writes bytes from the payload area. Data might
** appear on the main page or be scattered out on multiple overflow
** pages.
**
** If the BtCursor.isIncrblobHandle flag is set, and the current
** cursor entry uses one or more overflow pages, this function
** allocates space for and lazily popluates the overflow page-list
** cache array (BtCursor.aOverflow). Subsequent calls use this
** cache to make seeking to the supplied offset more efficient.
**
** Once an overflow page-list cache has been allocated, it may be
** invalidated if some other cursor writes to the same table, or if
** the cursor is moved to a different row. Additionally, in auto-vacuum
** mode, the following events may invalidate an overflow page-list cache.
**
** * An incremental vacuum,
** * A commit in auto_vacuum="full" mode,
** * Creating a table (may require moving an overflow page).
*/
static int accessPayload(
BtCursor *pCur, /* Cursor pointing to entry to read from */
int offset, /* Begin reading this far into payload */
int amt, /* Read this many bytes */
unsigned char *pBuf, /* Write the bytes into this buffer */
int skipKey, /* offset begins at data if this is true */
int eOp /* zero to read. non-zero to write. */
){
unsigned char *aPayload;
int rc = SQLITE_OK;
u32 nKey;
int iIdx = 0;
MemPage *pPage = pCur->pPage; /* Btree page of current cursor entry */
BtShared *pBt = pCur->pBtree->pBt; /* Btree this cursor belongs to */
assert( pPage );
assert( pCur->eState==CURSOR_VALID );
assert( pCur->idx>=0 && pCur->idx<pPage->nCell );
assert( offset>=0 );
getCellInfo(pCur);
aPayload = pCur->info.pCell + pCur->info.nHeader;
nKey = (pPage->intKey ? 0 : pCur->info.nKey);
if( skipKey ){
offset += nKey;
}
if( offset+amt > nKey+pCur->info.nData ){
/* Trying to read or write past the end of the data is an error */
return SQLITE_ERROR;
}
/* Check if data must be read/written to/from the btree page itself. */
if( offset<pCur->info.nLocal ){
int a = amt;
if( a+offset>pCur->info.nLocal ){
a = pCur->info.nLocal - offset;
}
rc = copyPayload(&aPayload[offset], pBuf, a, eOp, pPage->pDbPage);
offset = 0;
pBuf += a;
amt -= a;
}else{
offset -= pCur->info.nLocal;
}
if( rc==SQLITE_OK && amt>0 ){
const int ovflSize = pBt->usableSize - 4; /* Bytes content per ovfl page */
Pgno nextPage;
nextPage = get4byte(&aPayload[pCur->info.nLocal]);
#ifndef SQLITE_OMIT_INCRBLOB
/* If the isIncrblobHandle flag is set and the BtCursor.aOverflow[]
** has not been allocated, allocate it now. The array is sized at
** one entry for each overflow page in the overflow chain. The
** page number of the first overflow page is stored in aOverflow[0],
** etc. A value of 0 in the aOverflow[] array means "not yet known"
** (the cache is lazily populated).
*/
if( pCur->isIncrblobHandle && !pCur->aOverflow ){
int nOvfl = (pCur->info.nPayload-pCur->info.nLocal+ovflSize-1)/ovflSize;
pCur->aOverflow = (Pgno *)sqliteMalloc(sizeof(Pgno)*nOvfl);
if( nOvfl && !pCur->aOverflow ){
rc = SQLITE_NOMEM;
}
}
/* If the overflow page-list cache has been allocated and the
** entry for the first required overflow page is valid, skip
** directly to it.
*/
if( pCur->aOverflow && pCur->aOverflow[offset/ovflSize] ){
iIdx = (offset/ovflSize);
nextPage = pCur->aOverflow[iIdx];
offset = (offset%ovflSize);
}
#endif
for( ; rc==SQLITE_OK && amt>0 && nextPage; iIdx++){
#ifndef SQLITE_OMIT_INCRBLOB
/* If required, populate the overflow page-list cache. */
if( pCur->aOverflow ){
assert(!pCur->aOverflow[iIdx] || pCur->aOverflow[iIdx]==nextPage);
pCur->aOverflow[iIdx] = nextPage;
}
#endif
if( offset>=ovflSize ){
/* The only reason to read this page is to obtain the page
** number for the next page in the overflow chain. The page
** data is not required. So first try to lookup the overflow
** page-list cache, if any, then fall back to the getOverflowPage()
** function.
*/
#ifndef SQLITE_OMIT_INCRBLOB
if( pCur->aOverflow && pCur->aOverflow[iIdx+1] ){
nextPage = pCur->aOverflow[iIdx+1];
} else
#endif
rc = getOverflowPage(pBt, nextPage, 0, &nextPage);
offset -= ovflSize;
}else{
/* Need to read this page properly. It contains some of the
** range of data that is being read (eOp==0) or written (eOp!=0).
*/
DbPage *pDbPage;
int a = amt;
rc = sqlite3PagerGet(pBt->pPager, nextPage, &pDbPage);
if( rc==SQLITE_OK ){
aPayload = sqlite3PagerGetData(pDbPage);
nextPage = get4byte(aPayload);
if( a + offset > ovflSize ){
a = ovflSize - offset;
}
rc = copyPayload(&aPayload[offset+4], pBuf, a, eOp, pDbPage);
sqlite3PagerUnref(pDbPage);
offset = 0;
amt -= a;
pBuf += a;
}
}
}
}
if( rc==SQLITE_OK && amt>0 ){
return SQLITE_CORRUPT_BKPT;
}
return rc;
}
/*
** Read part of the key associated with cursor pCur. Exactly
** "amt" bytes will be transfered into pBuf[]. The transfer
** begins at "offset".
**
** Return SQLITE_OK on success or an error code if anything goes
** wrong. An error is returned if "offset+amt" is larger than
** the available payload.
*/
int sqlite3BtreeKey(BtCursor *pCur, u32 offset, u32 amt, void *pBuf){
int rc = restoreOrClearCursorPosition(pCur);
if( rc==SQLITE_OK ){
assert( pCur->eState==CURSOR_VALID );
assert( pCur->pPage!=0 );
if( pCur->pPage->intKey ){
return SQLITE_CORRUPT_BKPT;
}
assert( pCur->pPage->intKey==0 );
assert( pCur->idx>=0 && pCur->idx<pCur->pPage->nCell );
rc = accessPayload(pCur, offset, amt, (unsigned char*)pBuf, 0, 0);
}
return rc;
}
/*
** Read part of the data associated with cursor pCur. Exactly
** "amt" bytes will be transfered into pBuf[]. The transfer
** begins at "offset".
**
** Return SQLITE_OK on success or an error code if anything goes
** wrong. An error is returned if "offset+amt" is larger than
** the available payload.
*/
int sqlite3BtreeData(BtCursor *pCur, u32 offset, u32 amt, void *pBuf){
int rc = restoreOrClearCursorPosition(pCur);
if( rc==SQLITE_OK ){
assert( pCur->eState==CURSOR_VALID );
assert( pCur->pPage!=0 );
assert( pCur->idx>=0 && pCur->idx<pCur->pPage->nCell );
rc = accessPayload(pCur, offset, amt, pBuf, 1, 0);
}
return rc;
}
/*
** Return a pointer to payload information from the entry that the
** pCur cursor is pointing to. The pointer is to the beginning of
** the key if skipKey==0 and it points to the beginning of data if
** skipKey==1. The number of bytes of available key/data is written
** into *pAmt. If *pAmt==0, then the value returned will not be
** a valid pointer.
**
** This routine is an optimization. It is common for the entire key
** and data to fit on the local page and for there to be no overflow
** pages. When that is so, this routine can be used to access the
** key and data without making a copy. If the key and/or data spills
** onto overflow pages, then accessPayload() must be used to reassembly
** the key/data and copy it into a preallocated buffer.
**
** The pointer returned by this routine looks directly into the cached
** page of the database. The data might change or move the next time
** any btree routine is called.
*/
static const unsigned char *fetchPayload(
BtCursor *pCur, /* Cursor pointing to entry to read from */
int *pAmt, /* Write the number of available bytes here */
int skipKey /* read beginning at data if this is true */
){
unsigned char *aPayload;
MemPage *pPage;
u32 nKey;
int nLocal;
assert( pCur!=0 && pCur->pPage!=0 );
assert( pCur->eState==CURSOR_VALID );
pPage = pCur->pPage;
assert( pCur->idx>=0 && pCur->idx<pPage->nCell );
getCellInfo(pCur);
aPayload = pCur->info.pCell;
aPayload += pCur->info.nHeader;
if( pPage->intKey ){
nKey = 0;
}else{
nKey = pCur->info.nKey;
}
if( skipKey ){
aPayload += nKey;
nLocal = pCur->info.nLocal - nKey;
}else{
nLocal = pCur->info.nLocal;
if( nLocal>nKey ){
nLocal = nKey;
}
}
*pAmt = nLocal;
return aPayload;
}
/*
** For the entry that cursor pCur is point to, return as
** many bytes of the key or data as are available on the local
** b-tree page. Write the number of available bytes into *pAmt.
**
** The pointer returned is ephemeral. The key/data may move
** or be destroyed on the next call to any Btree routine.
**
** These routines is used to get quick access to key and data
** in the common case where no overflow pages are used.
*/
const void *sqlite3BtreeKeyFetch(BtCursor *pCur, int *pAmt){
if( pCur->eState==CURSOR_VALID ){
return (const void*)fetchPayload(pCur, pAmt, 0);
}
return 0;
}
const void *sqlite3BtreeDataFetch(BtCursor *pCur, int *pAmt){
if( pCur->eState==CURSOR_VALID ){
return (const void*)fetchPayload(pCur, pAmt, 1);
}
return 0;
}
/*
** Move the cursor down to a new child page. The newPgno argument is the
** page number of the child page to move to.
*/
static int moveToChild(BtCursor *pCur, u32 newPgno){
int rc;
MemPage *pNewPage;
MemPage *pOldPage;
BtShared *pBt = pCur->pBtree->pBt;
assert( pCur->eState==CURSOR_VALID );
rc = getAndInitPage(pBt, newPgno, &pNewPage, pCur->pPage);
if( rc ) return rc;
pNewPage->idxParent = pCur->idx;
pOldPage = pCur->pPage;
pOldPage->idxShift = 0;
releasePage(pOldPage);
pCur->pPage = pNewPage;
pCur->idx = 0;
pCur->info.nSize = 0;
if( pNewPage->nCell<1 ){
return SQLITE_CORRUPT_BKPT;
}
return SQLITE_OK;
}
/*
** Return true if the page is the virtual root of its table.
**
** The virtual root page is the root page for most tables. But
** for the table rooted on page 1, sometime the real root page
** is empty except for the right-pointer. In such cases the
** virtual root page is the page that the right-pointer of page
** 1 is pointing to.
*/
int sqlite3BtreeIsRootPage(MemPage *pPage){
MemPage *pParent = pPage->pParent;
if( pParent==0 ) return 1;
if( pParent->pgno>1 ) return 0;
if( get2byte(&pParent->aData[pParent->hdrOffset+3])==0 ) return 1;
return 0;
}
/*
** Move the cursor up to the parent page.
**
** pCur->idx is set to the cell index that contains the pointer
** to the page we are coming from. If we are coming from the
** right-most child page then pCur->idx is set to one more than
** the largest cell index.
*/
void sqlite3BtreeMoveToParent(BtCursor *pCur){
MemPage *pParent;
MemPage *pPage;
int idxParent;
assert( pCur->eState==CURSOR_VALID );
pPage = pCur->pPage;
assert( pPage!=0 );
assert( !sqlite3BtreeIsRootPage(pPage) );
pParent = pPage->pParent;
assert( pParent!=0 );
idxParent = pPage->idxParent;
sqlite3PagerRef(pParent->pDbPage);
releasePage(pPage);
pCur->pPage = pParent;
pCur->info.nSize = 0;
assert( pParent->idxShift==0 );
pCur->idx = idxParent;
}
/*
** Move the cursor to the root page
*/
static int moveToRoot(BtCursor *pCur){
MemPage *pRoot;
int rc = SQLITE_OK;
BtShared *pBt = pCur->pBtree->pBt;
if( pCur->eState==CURSOR_REQUIRESEEK ){
clearCursorPosition(pCur);
}
pRoot = pCur->pPage;
if( pRoot && pRoot->pgno==pCur->pgnoRoot ){
assert( pRoot->isInit );
}else{
if(
SQLITE_OK!=(rc = getAndInitPage(pBt, pCur->pgnoRoot, &pRoot, 0))
){
pCur->eState = CURSOR_INVALID;
return rc;
}
releasePage(pCur->pPage);
pCur->pPage = pRoot;
}
pCur->idx = 0;
pCur->info.nSize = 0;
if( pRoot->nCell==0 && !pRoot->leaf ){
Pgno subpage;
assert( pRoot->pgno==1 );
subpage = get4byte(&pRoot->aData[pRoot->hdrOffset+8]);
assert( subpage>0 );
pCur->eState = CURSOR_VALID;
rc = moveToChild(pCur, subpage);
}
pCur->eState = ((pCur->pPage->nCell>0)?CURSOR_VALID:CURSOR_INVALID);
return rc;
}
/*
** Move the cursor down to the left-most leaf entry beneath the
** entry to which it is currently pointing.
**
** The left-most leaf is the one with the smallest key - the first
** in ascending order.
*/
static int moveToLeftmost(BtCursor *pCur){
Pgno pgno;
int rc;
MemPage *pPage;
assert( pCur->eState==CURSOR_VALID );
while( !(pPage = pCur->pPage)->leaf ){
assert( pCur->idx>=0 && pCur->idx<pPage->nCell );
pgno = get4byte(findCell(pPage, pCur->idx));
rc = moveToChild(pCur, pgno);
if( rc ) return rc;
}
return SQLITE_OK;
}
/*
** Move the cursor down to the right-most leaf entry beneath the
** page to which it is currently pointing. Notice the difference
** between moveToLeftmost() and moveToRightmost(). moveToLeftmost()
** finds the left-most entry beneath the *entry* whereas moveToRightmost()
** finds the right-most entry beneath the *page*.
**
** The right-most entry is the one with the largest key - the last
** key in ascending order.
*/
static int moveToRightmost(BtCursor *pCur){
Pgno pgno;
int rc;
MemPage *pPage;
assert( pCur->eState==CURSOR_VALID );
while( !(pPage = pCur->pPage)->leaf ){
pgno = get4byte(&pPage->aData[pPage->hdrOffset+8]);
pCur->idx = pPage->nCell;
rc = moveToChild(pCur, pgno);
if( rc ) return rc;
}
pCur->idx = pPage->nCell - 1;
pCur->info.nSize = 0;
return SQLITE_OK;
}
/* Move the cursor to the first entry in the table. Return SQLITE_OK
** on success. Set *pRes to 0 if the cursor actually points to something
** or set *pRes to 1 if the table is empty.
*/
int sqlite3BtreeFirst(BtCursor *pCur, int *pRes){
int rc;
rc = moveToRoot(pCur);
if( rc ) return rc;
if( pCur->eState==CURSOR_INVALID ){
assert( pCur->pPage->nCell==0 );
*pRes = 1;
return SQLITE_OK;
}
assert( pCur->pPage->nCell>0 );
*pRes = 0;
rc = moveToLeftmost(pCur);
return rc;
}
/* Move the cursor to the last entry in the table. Return SQLITE_OK
** on success. Set *pRes to 0 if the cursor actually points to something
** or set *pRes to 1 if the table is empty.
*/
int sqlite3BtreeLast(BtCursor *pCur, int *pRes){
int rc;
rc = moveToRoot(pCur);
if( rc ) return rc;
if( CURSOR_INVALID==pCur->eState ){
assert( pCur->pPage->nCell==0 );
*pRes = 1;
return SQLITE_OK;
}
assert( pCur->eState==CURSOR_VALID );
*pRes = 0;
rc = moveToRightmost(pCur);
return rc;
}
/* Move the cursor so that it points to an entry near pKey/nKey.
** Return a success code.
**
** For INTKEY tables, only the nKey parameter is used. pKey is
** ignored. For other tables, nKey is the number of bytes of data
** in pKey. The comparison function specified when the cursor was
** created is used to compare keys.
**
** If an exact match is not found, then the cursor is always
** left pointing at a leaf page which would hold the entry if it
** were present. The cursor might point to an entry that comes
** before or after the key.
**
** The result of comparing the key with the entry to which the
** cursor is written to *pRes if pRes!=NULL. The meaning of
** this value is as follows:
**
** *pRes<0 The cursor is left pointing at an entry that
** is smaller than pKey or if the table is empty
** and the cursor is therefore left point to nothing.
**
** *pRes==0 The cursor is left pointing at an entry that
** exactly matches pKey.
**
** *pRes>0 The cursor is left pointing at an entry that
** is larger than pKey.
*/
int sqlite3BtreeMoveto(
BtCursor *pCur, /* The cursor to be moved */
const void *pKey, /* The key content for indices. Not used by tables */
i64 nKey, /* Size of pKey. Or the key for tables */
int biasRight, /* If true, bias the search to the high end */
int *pRes /* Search result flag */
){
int rc;
rc = moveToRoot(pCur);
if( rc ) return rc;
assert( pCur->pPage );
assert( pCur->pPage->isInit );
if( pCur->eState==CURSOR_INVALID ){
*pRes = -1;
assert( pCur->pPage->nCell==0 );
return SQLITE_OK;
}
for(;;){
int lwr, upr;
Pgno chldPg;
MemPage *pPage = pCur->pPage;
int c = -1; /* pRes return if table is empty must be -1 */
lwr = 0;
upr = pPage->nCell-1;
if( !pPage->intKey && pKey==0 ){
return SQLITE_CORRUPT_BKPT;
}
if( biasRight ){
pCur->idx = upr;
}else{
pCur->idx = (upr+lwr)/2;
}
if( lwr<=upr ) for(;;){
void *pCellKey;
i64 nCellKey;
pCur->info.nSize = 0;
if( pPage->intKey ){
u8 *pCell;
pCell = findCell(pPage, pCur->idx) + pPage->childPtrSize;
if( pPage->hasData ){
u32 dummy;
pCell += getVarint32(pCell, &dummy);
}
getVarint(pCell, (u64 *)&nCellKey);
if( nCellKey<nKey ){
c = -1;
}else if( nCellKey>nKey ){
c = +1;
}else{
c = 0;
}
}else{
int available;
pCellKey = (void *)fetchPayload(pCur, &available, 0);
nCellKey = pCur->info.nKey;
if( available>=nCellKey ){
c = pCur->xCompare(pCur->pArg, nCellKey, pCellKey, nKey, pKey);
}else{
pCellKey = sqliteMallocRaw( nCellKey );
if( pCellKey==0 ) return SQLITE_NOMEM;
rc = sqlite3BtreeKey(pCur, 0, nCellKey, (void *)pCellKey);
c = pCur->xCompare(pCur->pArg, nCellKey, pCellKey, nKey, pKey);
sqliteFree(pCellKey);
if( rc ) return rc;
}
}
if( c==0 ){
if( pPage->leafData && !pPage->leaf ){
lwr = pCur->idx;
upr = lwr - 1;
break;
}else{
if( pRes ) *pRes = 0;
return SQLITE_OK;
}
}
if( c<0 ){
lwr = pCur->idx+1;
}else{
upr = pCur->idx-1;
}
if( lwr>upr ){
break;
}
pCur->idx = (lwr+upr)/2;
}
assert( lwr==upr+1 );
assert( pPage->isInit );
if( pPage->leaf ){
chldPg = 0;
}else if( lwr>=pPage->nCell ){
chldPg = get4byte(&pPage->aData[pPage->hdrOffset+8]);
}else{
chldPg = get4byte(findCell(pPage, lwr));
}
if( chldPg==0 ){
assert( pCur->idx>=0 && pCur->idx<pCur->pPage->nCell );
if( pRes ) *pRes = c;
return SQLITE_OK;
}
pCur->idx = lwr;
pCur->info.nSize = 0;
rc = moveToChild(pCur, chldPg);
if( rc ){
return rc;
}
}
/* NOT REACHED */
}
/*
** Return TRUE if the cursor is not pointing at an entry of the table.
**
** TRUE will be returned after a call to sqlite3BtreeNext() moves
** past the last entry in the table or sqlite3BtreePrev() moves past
** the first entry. TRUE is also returned if the table is empty.
*/
int sqlite3BtreeEof(BtCursor *pCur){
/* TODO: What if the cursor is in CURSOR_REQUIRESEEK but all table entries
** have been deleted? This API will need to change to return an error code
** as well as the boolean result value.
*/
return (CURSOR_VALID!=pCur->eState);
}
/*
** Advance the cursor to the next entry in the database. If
** successful then set *pRes=0. If the cursor
** was already pointing to the last entry in the database before
** this routine was called, then set *pRes=1.
*/
int sqlite3BtreeNext(BtCursor *pCur, int *pRes){
int rc;
MemPage *pPage;
rc = restoreOrClearCursorPosition(pCur);
if( rc!=SQLITE_OK ){
return rc;
}
assert( pRes!=0 );
pPage = pCur->pPage;
if( CURSOR_INVALID==pCur->eState ){
*pRes = 1;
return SQLITE_OK;
}
if( pCur->skip>0 ){
pCur->skip = 0;
*pRes = 0;
return SQLITE_OK;
}
pCur->skip = 0;
assert( pPage->isInit );
assert( pCur->idx<pPage->nCell );
pCur->idx++;
pCur->info.nSize = 0;
if( pCur->idx>=pPage->nCell ){
if( !pPage->leaf ){
rc = moveToChild(pCur, get4byte(&pPage->aData[pPage->hdrOffset+8]));
if( rc ) return rc;
rc = moveToLeftmost(pCur);
*pRes = 0;
return rc;
}
do{
if( sqlite3BtreeIsRootPage(pPage) ){
*pRes = 1;
pCur->eState = CURSOR_INVALID;
return SQLITE_OK;
}
sqlite3BtreeMoveToParent(pCur);
pPage = pCur->pPage;
}while( pCur->idx>=pPage->nCell );
*pRes = 0;
if( pPage->leafData ){
rc = sqlite3BtreeNext(pCur, pRes);
}else{
rc = SQLITE_OK;
}
return rc;
}
*pRes = 0;
if( pPage->leaf ){
return SQLITE_OK;
}
rc = moveToLeftmost(pCur);
return rc;
}
/*
** Step the cursor to the back to the previous entry in the database. If
** successful then set *pRes=0. If the cursor
** was already pointing to the first entry in the database before
** this routine was called, then set *pRes=1.
*/
int sqlite3BtreePrevious(BtCursor *pCur, int *pRes){
int rc;
Pgno pgno;
MemPage *pPage;
rc = restoreOrClearCursorPosition(pCur);
if( rc!=SQLITE_OK ){
return rc;
}
if( CURSOR_INVALID==pCur->eState ){
*pRes = 1;
return SQLITE_OK;
}
if( pCur->skip<0 ){
pCur->skip = 0;
*pRes = 0;
return SQLITE_OK;
}
pCur->skip = 0;
pPage = pCur->pPage;
assert( pPage->isInit );
assert( pCur->idx>=0 );
if( !pPage->leaf ){
pgno = get4byte( findCell(pPage, pCur->idx) );
rc = moveToChild(pCur, pgno);
if( rc ) return rc;
rc = moveToRightmost(pCur);
}else{
while( pCur->idx==0 ){
if( sqlite3BtreeIsRootPage(pPage) ){
pCur->eState = CURSOR_INVALID;
*pRes = 1;
return SQLITE_OK;
}
sqlite3BtreeMoveToParent(pCur);
pPage = pCur->pPage;
}
pCur->idx--;
pCur->info.nSize = 0;
if( pPage->leafData && !pPage->leaf ){
rc = sqlite3BtreePrevious(pCur, pRes);
}else{
rc = SQLITE_OK;
}
}
*pRes = 0;
return rc;
}
/*
** Allocate a new page from the database file.
**
** The new page is marked as dirty. (In other words, sqlite3PagerWrite()
** has already been called on the new page.) The new page has also
** been referenced and the calling routine is responsible for calling
** sqlite3PagerUnref() on the new page when it is done.
**
** SQLITE_OK is returned on success. Any other return value indicates
** an error. *ppPage and *pPgno are undefined in the event of an error.
** Do not invoke sqlite3PagerUnref() on *ppPage if an error is returned.
**
** If the "nearby" parameter is not 0, then a (feeble) effort is made to
** locate a page close to the page number "nearby". This can be used in an
** attempt to keep related pages close to each other in the database file,
** which in turn can make database access faster.
**
** If the "exact" parameter is not 0, and the page-number nearby exists
** anywhere on the free-list, then it is guarenteed to be returned. This
** is only used by auto-vacuum databases when allocating a new table.
*/
static int allocateBtreePage(
BtShared *pBt,
MemPage **ppPage,
Pgno *pPgno,
Pgno nearby,
u8 exact
){
MemPage *pPage1;
int rc;
int n; /* Number of pages on the freelist */
int k; /* Number of leaves on the trunk of the freelist */
MemPage *pTrunk = 0;
MemPage *pPrevTrunk = 0;
pPage1 = pBt->pPage1;
n = get4byte(&pPage1->aData[36]);
if( n>0 ){
/* There are pages on the freelist. Reuse one of those pages. */
Pgno iTrunk;
u8 searchList = 0; /* If the free-list must be searched for 'nearby' */
/* If the 'exact' parameter was true and a query of the pointer-map
** shows that the page 'nearby' is somewhere on the free-list, then
** the entire-list will be searched for that page.
*/
#ifndef SQLITE_OMIT_AUTOVACUUM
if( exact && nearby<=sqlite3PagerPagecount(pBt->pPager) ){
u8 eType;
assert( nearby>0 );
assert( pBt->autoVacuum );
rc = ptrmapGet(pBt, nearby, &eType, 0);
if( rc ) return rc;
if( eType==PTRMAP_FREEPAGE ){
searchList = 1;
}
*pPgno = nearby;
}
#endif
/* Decrement the free-list count by 1. Set iTrunk to the index of the
** first free-list trunk page. iPrevTrunk is initially 1.
*/
rc = sqlite3PagerWrite(pPage1->pDbPage);
if( rc ) return rc;
put4byte(&pPage1->aData[36], n-1);
/* The code within this loop is run only once if the 'searchList' variable
** is not true. Otherwise, it runs once for each trunk-page on the
** free-list until the page 'nearby' is located.
*/
do {
pPrevTrunk = pTrunk;
if( pPrevTrunk ){
iTrunk = get4byte(&pPrevTrunk->aData[0]);
}else{
iTrunk = get4byte(&pPage1->aData[32]);
}
rc = sqlite3BtreeGetPage(pBt, iTrunk, &pTrunk, 0);
if( rc ){
pTrunk = 0;
goto end_allocate_page;
}
k = get4byte(&pTrunk->aData[4]);
if( k==0 && !searchList ){
/* The trunk has no leaves and the list is not being searched.
** So extract the trunk page itself and use it as the newly
** allocated page */
assert( pPrevTrunk==0 );
rc = sqlite3PagerWrite(pTrunk->pDbPage);
if( rc ){
goto end_allocate_page;
}
*pPgno = iTrunk;
memcpy(&pPage1->aData[32], &pTrunk->aData[0], 4);
*ppPage = pTrunk;
pTrunk = 0;
TRACE(("ALLOCATE: %d trunk - %d free pages left\n", *pPgno, n-1));
}else if( k>pBt->usableSize/4 - 8 ){
/* Value of k is out of range. Database corruption */
rc = SQLITE_CORRUPT_BKPT;
goto end_allocate_page;
#ifndef SQLITE_OMIT_AUTOVACUUM
}else if( searchList && nearby==iTrunk ){
/* The list is being searched and this trunk page is the page
** to allocate, regardless of whether it has leaves.
*/
assert( *pPgno==iTrunk );
*ppPage = pTrunk;
searchList = 0;
rc = sqlite3PagerWrite(pTrunk->pDbPage);
if( rc ){
goto end_allocate_page;
}
if( k==0 ){
if( !pPrevTrunk ){
memcpy(&pPage1->aData[32], &pTrunk->aData[0], 4);
}else{
memcpy(&pPrevTrunk->aData[0], &pTrunk->aData[0], 4);
}
}else{
/* The trunk page is required by the caller but it contains
** pointers to free-list leaves. The first leaf becomes a trunk
** page in this case.
*/
MemPage *pNewTrunk;
Pgno iNewTrunk = get4byte(&pTrunk->aData[8]);
rc = sqlite3BtreeGetPage(pBt, iNewTrunk, &pNewTrunk, 0);
if( rc!=SQLITE_OK ){
goto end_allocate_page;
}
rc = sqlite3PagerWrite(pNewTrunk->pDbPage);
if( rc!=SQLITE_OK ){
releasePage(pNewTrunk);
goto end_allocate_page;
}
memcpy(&pNewTrunk->aData[0], &pTrunk->aData[0], 4);
put4byte(&pNewTrunk->aData[4], k-1);
memcpy(&pNewTrunk->aData[8], &pTrunk->aData[12], (k-1)*4);
releasePage(pNewTrunk);
if( !pPrevTrunk ){
put4byte(&pPage1->aData[32], iNewTrunk);
}else{
rc = sqlite3PagerWrite(pPrevTrunk->pDbPage);
if( rc ){
goto end_allocate_page;
}
put4byte(&pPrevTrunk->aData[0], iNewTrunk);
}
}
pTrunk = 0;
TRACE(("ALLOCATE: %d trunk - %d free pages left\n", *pPgno, n-1));
#endif
}else{
/* Extract a leaf from the trunk */
int closest;
Pgno iPage;
unsigned char *aData = pTrunk->aData;
rc = sqlite3PagerWrite(pTrunk->pDbPage);
if( rc ){
goto end_allocate_page;
}
if( nearby>0 ){
int i, dist;
closest = 0;
dist = get4byte(&aData[8]) - nearby;
if( dist<0 ) dist = -dist;
for(i=1; i<k; i++){
int d2 = get4byte(&aData[8+i*4]) - nearby;
if( d2<0 ) d2 = -d2;
if( d2<dist ){
closest = i;
dist = d2;
}
}
}else{
closest = 0;
}
iPage = get4byte(&aData[8+closest*4]);
if( !searchList || iPage==nearby ){
*pPgno = iPage;
if( *pPgno>sqlite3PagerPagecount(pBt->pPager) ){
/* Free page off the end of the file */
return SQLITE_CORRUPT_BKPT;
}
TRACE(("ALLOCATE: %d was leaf %d of %d on trunk %d"
": %d more free pages\n",
*pPgno, closest+1, k, pTrunk->pgno, n-1));
if( closest<k-1 ){
memcpy(&aData[8+closest*4], &aData[4+k*4], 4);
}
put4byte(&aData[4], k-1);
rc = sqlite3BtreeGetPage(pBt, *pPgno, ppPage, 1);
if( rc==SQLITE_OK ){
sqlite3PagerDontRollback((*ppPage)->pDbPage);
rc = sqlite3PagerWrite((*ppPage)->pDbPage);
if( rc!=SQLITE_OK ){
releasePage(*ppPage);
}
}
searchList = 0;
}
}
releasePage(pPrevTrunk);
pPrevTrunk = 0;
}while( searchList );
}else{
/* There are no pages on the freelist, so create a new page at the
** end of the file */
*pPgno = sqlite3PagerPagecount(pBt->pPager) + 1;
#ifndef SQLITE_OMIT_AUTOVACUUM
if( pBt->nTrunc ){
/* An incr-vacuum has already run within this transaction. So the
** page to allocate is not from the physical end of the file, but
** at pBt->nTrunc.
*/
*pPgno = pBt->nTrunc+1;
if( *pPgno==PENDING_BYTE_PAGE(pBt) ){
(*pPgno)++;
}
}
if( pBt->autoVacuum && PTRMAP_ISPAGE(pBt, *pPgno) ){
/* If *pPgno refers to a pointer-map page, allocate two new pages
** at the end of the file instead of one. The first allocated page
** becomes a new pointer-map page, the second is used by the caller.
*/
TRACE(("ALLOCATE: %d from end of file (pointer-map page)\n", *pPgno));
assert( *pPgno!=PENDING_BYTE_PAGE(pBt) );
(*pPgno)++;
}
if( pBt->nTrunc ){
pBt->nTrunc = *pPgno;
}
#endif
assert( *pPgno!=PENDING_BYTE_PAGE(pBt) );
rc = sqlite3BtreeGetPage(pBt, *pPgno, ppPage, 0);
if( rc ) return rc;
rc = sqlite3PagerWrite((*ppPage)->pDbPage);
if( rc!=SQLITE_OK ){
releasePage(*ppPage);
}
TRACE(("ALLOCATE: %d from end of file\n", *pPgno));
}
assert( *pPgno!=PENDING_BYTE_PAGE(pBt) );
end_allocate_page:
releasePage(pTrunk);
releasePage(pPrevTrunk);
return rc;
}
/*
** Add a page of the database file to the freelist.
**
** sqlite3PagerUnref() is NOT called for pPage.
*/
static int freePage(MemPage *pPage){
BtShared *pBt = pPage->pBt;
MemPage *pPage1 = pBt->pPage1;
int rc, n, k;
/* Prepare the page for freeing */
assert( pPage->pgno>1 );
pPage->isInit = 0;
releasePage(pPage->pParent);
pPage->pParent = 0;
/* Increment the free page count on pPage1 */
rc = sqlite3PagerWrite(pPage1->pDbPage);
if( rc ) return rc;
n = get4byte(&pPage1->aData[36]);
put4byte(&pPage1->aData[36], n+1);
#ifdef SQLITE_SECURE_DELETE
/* If the SQLITE_SECURE_DELETE compile-time option is enabled, then
** always fully overwrite deleted information with zeros.
*/
rc = sqlite3PagerWrite(pPage->pDbPage);
if( rc ) return rc;
memset(pPage->aData, 0, pPage->pBt->pageSize);
#endif
#ifndef SQLITE_OMIT_AUTOVACUUM
/* If the database supports auto-vacuum, write an entry in the pointer-map
** to indicate that the page is free.
*/
if( pBt->autoVacuum ){
rc = ptrmapPut(pBt, pPage->pgno, PTRMAP_FREEPAGE, 0);
if( rc ) return rc;
}
#endif
if( n==0 ){
/* This is the first free page */
rc = sqlite3PagerWrite(pPage->pDbPage);
if( rc ) return rc;
memset(pPage->aData, 0, 8);
put4byte(&pPage1->aData[32], pPage->pgno);
TRACE(("FREE-PAGE: %d first\n", pPage->pgno));
}else{
/* Other free pages already exist. Retrive the first trunk page
** of the freelist and find out how many leaves it has. */
MemPage *pTrunk;
rc = sqlite3BtreeGetPage(pBt, get4byte(&pPage1->aData[32]), &pTrunk, 0);
if( rc ) return rc;
k = get4byte(&pTrunk->aData[4]);
if( k>=pBt->usableSize/4 - 8 ){
/* The trunk is full. Turn the page being freed into a new
** trunk page with no leaves. */
rc = sqlite3PagerWrite(pPage->pDbPage);
if( rc ) return rc;
put4byte(pPage->aData, pTrunk->pgno);
put4byte(&pPage->aData[4], 0);
put4byte(&pPage1->aData[32], pPage->pgno);
TRACE(("FREE-PAGE: %d new trunk page replacing %d\n",
pPage->pgno, pTrunk->pgno));
}else{
/* Add the newly freed page as a leaf on the current trunk */
rc = sqlite3PagerWrite(pTrunk->pDbPage);
if( rc==SQLITE_OK ){
put4byte(&pTrunk->aData[4], k+1);
put4byte(&pTrunk->aData[8+k*4], pPage->pgno);
#ifndef SQLITE_SECURE_DELETE
sqlite3PagerDontWrite(pPage->pDbPage);
#endif
}
TRACE(("FREE-PAGE: %d leaf on trunk page %d\n",pPage->pgno,pTrunk->pgno));
}
releasePage(pTrunk);
}
return rc;
}
/*
** Free any overflow pages associated with the given Cell.
*/
static int clearCell(MemPage *pPage, unsigned char *pCell){
BtShared *pBt = pPage->pBt;
CellInfo info;
Pgno ovflPgno;
int rc;
int nOvfl;
int ovflPageSize;
sqlite3BtreeParseCellPtr(pPage, pCell, &info);
if( info.iOverflow==0 ){
return SQLITE_OK; /* No overflow pages. Return without doing anything */
}
ovflPgno = get4byte(&pCell[info.iOverflow]);
ovflPageSize = pBt->usableSize - 4;
nOvfl = (info.nPayload - info.nLocal + ovflPageSize - 1)/ovflPageSize;
assert( ovflPgno==0 || nOvfl>0 );
while( nOvfl-- ){
MemPage *pOvfl;
if( ovflPgno==0 || ovflPgno>sqlite3PagerPagecount(pBt->pPager) ){
return SQLITE_CORRUPT_BKPT;
}
rc = getOverflowPage(pBt, ovflPgno, &pOvfl, (nOvfl==0)?0:&ovflPgno);
if( rc ) return rc;
rc = freePage(pOvfl);
sqlite3PagerUnref(pOvfl->pDbPage);
if( rc ) return rc;
}
return SQLITE_OK;
}
/*
** Create the byte sequence used to represent a cell on page pPage
** and write that byte sequence into pCell[]. Overflow pages are
** allocated and filled in as necessary. The calling procedure
** is responsible for making sure sufficient space has been allocated
** for pCell[].
**
** Note that pCell does not necessary need to point to the pPage->aData
** area. pCell might point to some temporary storage. The cell will
** be constructed in this temporary area then copied into pPage->aData
** later.
*/
static int fillInCell(
MemPage *pPage, /* The page that contains the cell */
unsigned char *pCell, /* Complete text of the cell */
const void *pKey, i64 nKey, /* The key */
const void *pData,int nData, /* The data */
int nZero, /* Extra zero bytes to append to pData */
int *pnSize /* Write cell size here */
){
int nPayload;
const u8 *pSrc;
int nSrc, n, rc;
int spaceLeft;
MemPage *pOvfl = 0;
MemPage *pToRelease = 0;
unsigned char *pPrior;
unsigned char *pPayload;
BtShared *pBt = pPage->pBt;
Pgno pgnoOvfl = 0;
int nHeader;
CellInfo info;
/* Fill in the header. */
nHeader = 0;
if( !pPage->leaf ){
nHeader += 4;
}
if( pPage->hasData ){
nHeader += putVarint(&pCell[nHeader], nData+nZero);
}else{
nData = nZero = 0;
}
nHeader += putVarint(&pCell[nHeader], *(u64*)&nKey);
sqlite3BtreeParseCellPtr(pPage, pCell, &info);
assert( info.nHeader==nHeader );
assert( info.nKey==nKey );
assert( info.nData==nData+nZero );
/* Fill in the payload */
nPayload = nData + nZero;
if( pPage->intKey ){
pSrc = pData;
nSrc = nData;
nData = 0;
}else{
nPayload += nKey;
pSrc = pKey;
nSrc = nKey;
}
*pnSize = info.nSize;
spaceLeft = info.nLocal;
pPayload = &pCell[nHeader];
pPrior = &pCell[info.iOverflow];
while( nPayload>0 ){
if( spaceLeft==0 ){
int isExact = 0;
#ifndef SQLITE_OMIT_AUTOVACUUM
Pgno pgnoPtrmap = pgnoOvfl; /* Overflow page pointer-map entry page */
if( pBt->autoVacuum ){
do{
pgnoOvfl++;
} while(
PTRMAP_ISPAGE(pBt, pgnoOvfl) || pgnoOvfl==PENDING_BYTE_PAGE(pBt)
);
if( pgnoOvfl>1 ){
/* isExact = 1; */
}
}
#endif
rc = allocateBtreePage(pBt, &pOvfl, &pgnoOvfl, pgnoOvfl, isExact);
#ifndef SQLITE_OMIT_AUTOVACUUM
/* If the database supports auto-vacuum, and the second or subsequent
** overflow page is being allocated, add an entry to the pointer-map
** for that page now.
**
** If this is the first overflow page, then write a partial entry
** to the pointer-map. If we write nothing to this pointer-map slot,
** then the optimistic overflow chain processing in clearCell()
** may misinterpret the uninitialised values and delete the
** wrong pages from the database.
*/
if( pBt->autoVacuum && rc==SQLITE_OK ){
u8 eType = (pgnoPtrmap?PTRMAP_OVERFLOW2:PTRMAP_OVERFLOW1);
rc = ptrmapPut(pBt, pgnoOvfl, eType, pgnoPtrmap);
if( rc ){
releasePage(pOvfl);
}
}
#endif
if( rc ){
releasePage(pToRelease);
return rc;
}
put4byte(pPrior, pgnoOvfl);
releasePage(pToRelease);
pToRelease = pOvfl;
pPrior = pOvfl->aData;
put4byte(pPrior, 0);
pPayload = &pOvfl->aData[4];
spaceLeft = pBt->usableSize - 4;
}
n = nPayload;
if( n>spaceLeft ) n = spaceLeft;
if( nSrc>0 ){
if( n>nSrc ) n = nSrc;
assert( pSrc );
memcpy(pPayload, pSrc, n);
}else{
memset(pPayload, 0, n);
}
nPayload -= n;
pPayload += n;
pSrc += n;
nSrc -= n;
spaceLeft -= n;
if( nSrc==0 ){
nSrc = nData;
pSrc = pData;
}
}
releasePage(pToRelease);
return SQLITE_OK;
}
/*
** Change the MemPage.pParent pointer on the page whose number is
** given in the second argument so that MemPage.pParent holds the
** pointer in the third argument.
*/
static int reparentPage(BtShared *pBt, Pgno pgno, MemPage *pNewParent, int idx){
MemPage *pThis;
DbPage *pDbPage;
assert( pNewParent!=0 );
if( pgno==0 ) return SQLITE_OK;
assert( pBt->pPager!=0 );
pDbPage = sqlite3PagerLookup(pBt->pPager, pgno);
if( pDbPage ){
pThis = (MemPage *)sqlite3PagerGetExtra(pDbPage);
if( pThis->isInit ){
assert( pThis->aData==(sqlite3PagerGetData(pDbPage)) );
if( pThis->pParent!=pNewParent ){
if( pThis->pParent ) sqlite3PagerUnref(pThis->pParent->pDbPage);
pThis->pParent = pNewParent;
sqlite3PagerRef(pNewParent->pDbPage);
}
pThis->idxParent = idx;
}
sqlite3PagerUnref(pDbPage);
}
#ifndef SQLITE_OMIT_AUTOVACUUM
if( pBt->autoVacuum ){
return ptrmapPut(pBt, pgno, PTRMAP_BTREE, pNewParent->pgno);
}
#endif
return SQLITE_OK;
}
/*
** Change the pParent pointer of all children of pPage to point back
** to pPage.
**
** In other words, for every child of pPage, invoke reparentPage()
** to make sure that each child knows that pPage is its parent.
**
** This routine gets called after you memcpy() one page into
** another.
*/
static int reparentChildPages(MemPage *pPage){
int i;
BtShared *pBt = pPage->pBt;
int rc = SQLITE_OK;
if( pPage->leaf ) return SQLITE_OK;
for(i=0; i<pPage->nCell; i++){
u8 *pCell = findCell(pPage, i);
if( !pPage->leaf ){
rc = reparentPage(pBt, get4byte(pCell), pPage, i);
if( rc!=SQLITE_OK ) return rc;
}
}
if( !pPage->leaf ){
rc = reparentPage(pBt, get4byte(&pPage->aData[pPage->hdrOffset+8]),
pPage, i);
pPage->idxShift = 0;
}
return rc;
}
/*
** Remove the i-th cell from pPage. This routine effects pPage only.
** The cell content is not freed or deallocated. It is assumed that
** the cell content has been copied someplace else. This routine just
** removes the reference to the cell from pPage.
**
** "sz" must be the number of bytes in the cell.
*/
static void dropCell(MemPage *pPage, int idx, int sz){
int i; /* Loop counter */
int pc; /* Offset to cell content of cell being deleted */
u8 *data; /* pPage->aData */
u8 *ptr; /* Used to move bytes around within data[] */
assert( idx>=0 && idx<pPage->nCell );
assert( sz==cellSize(pPage, idx) );
assert( sqlite3PagerIswriteable(pPage->pDbPage) );
data = pPage->aData;
ptr = &data[pPage->cellOffset + 2*idx];
pc = get2byte(ptr);
assert( pc>10 && pc+sz<=pPage->pBt->usableSize );
freeSpace(pPage, pc, sz);
for(i=idx+1; i<pPage->nCell; i++, ptr+=2){
ptr[0] = ptr[2];
ptr[1] = ptr[3];
}
pPage->nCell--;
put2byte(&data[pPage->hdrOffset+3], pPage->nCell);
pPage->nFree += 2;
pPage->idxShift = 1;
}
/*
** Insert a new cell on pPage at cell index "i". pCell points to the
** content of the cell.
**
** If the cell content will fit on the page, then put it there. If it
** will not fit, then make a copy of the cell content into pTemp if
** pTemp is not null. Regardless of pTemp, allocate a new entry
** in pPage->aOvfl[] and make it point to the cell content (either
** in pTemp or the original pCell) and also record its index.
** Allocating a new entry in pPage->aCell[] implies that
** pPage->nOverflow is incremented.
**
** If nSkip is non-zero, then do not copy the first nSkip bytes of the
** cell. The caller will overwrite them after this function returns. If
** nSkip is non-zero, then pCell may not point to an invalid memory location
** (but pCell+nSkip is always valid).
*/
static int insertCell(
MemPage *pPage, /* Page into which we are copying */
int i, /* New cell becomes the i-th cell of the page */
u8 *pCell, /* Content of the new cell */
int sz, /* Bytes of content in pCell */
u8 *pTemp, /* Temp storage space for pCell, if needed */
u8 nSkip /* Do not write the first nSkip bytes of the cell */
){
int idx; /* Where to write new cell content in data[] */
int j; /* Loop counter */
int top; /* First byte of content for any cell in data[] */
int end; /* First byte past the last cell pointer in data[] */
int ins; /* Index in data[] where new cell pointer is inserted */
int hdr; /* Offset into data[] of the page header */
int cellOffset; /* Address of first cell pointer in data[] */
u8 *data; /* The content of the whole page */
u8 *ptr; /* Used for moving information around in data[] */
assert( i>=0 && i<=pPage->nCell+pPage->nOverflow );
assert( sz==cellSizePtr(pPage, pCell) );
assert( sqlite3PagerIswriteable(pPage->pDbPage) );
if( pPage->nOverflow || sz+2>pPage->nFree ){
if( pTemp ){
memcpy(pTemp+nSkip, pCell+nSkip, sz-nSkip);
pCell = pTemp;
}
j = pPage->nOverflow++;
assert( j<sizeof(pPage->aOvfl)/sizeof(pPage->aOvfl[0]) );
pPage->aOvfl[j].pCell = pCell;
pPage->aOvfl[j].idx = i;
pPage->nFree = 0;
}else{
data = pPage->aData;
hdr = pPage->hdrOffset;
top = get2byte(&data[hdr+5]);
cellOffset = pPage->cellOffset;
end = cellOffset + 2*pPage->nCell + 2;
ins = cellOffset + 2*i;
if( end > top - sz ){
int rc = defragmentPage(pPage);
if( rc!=SQLITE_OK ) return rc;
top = get2byte(&data[hdr+5]);
assert( end + sz <= top );
}
idx = allocateSpace(pPage, sz);
assert( idx>0 );
assert( end <= get2byte(&data[hdr+5]) );
pPage->nCell++;
pPage->nFree -= 2;
memcpy(&data[idx+nSkip], pCell+nSkip, sz-nSkip);
for(j=end-2, ptr=&data[j]; j>ins; j-=2, ptr-=2){
ptr[0] = ptr[-2];
ptr[1] = ptr[-1];
}
put2byte(&data[ins], idx);
put2byte(&data[hdr+3], pPage->nCell);
pPage->idxShift = 1;
#ifndef SQLITE_OMIT_AUTOVACUUM
if( pPage->pBt->autoVacuum ){
/* The cell may contain a pointer to an overflow page. If so, write
** the entry for the overflow page into the pointer map.
*/
CellInfo info;
sqlite3BtreeParseCellPtr(pPage, pCell, &info);
assert( (info.nData+(pPage->intKey?0:info.nKey))==info.nPayload );
if( (info.nData+(pPage->intKey?0:info.nKey))>info.nLocal ){
Pgno pgnoOvfl = get4byte(&pCell[info.iOverflow]);
int rc = ptrmapPut(pPage->pBt, pgnoOvfl, PTRMAP_OVERFLOW1, pPage->pgno);
if( rc!=SQLITE_OK ) return rc;
}
}
#endif
}
return SQLITE_OK;
}
/*
** Add a list of cells to a page. The page should be initially empty.
** The cells are guaranteed to fit on the page.
*/
static void assemblePage(
MemPage *pPage, /* The page to be assemblied */
int nCell, /* The number of cells to add to this page */
u8 **apCell, /* Pointers to cell bodies */
int *aSize /* Sizes of the cells */
){
int i; /* Loop counter */
int totalSize; /* Total size of all cells */
int hdr; /* Index of page header */
int cellptr; /* Address of next cell pointer */
int cellbody; /* Address of next cell body */
u8 *data; /* Data for the page */
assert( pPage->nOverflow==0 );
totalSize = 0;
for(i=0; i<nCell; i++){
totalSize += aSize[i];
}
assert( totalSize+2*nCell<=pPage->nFree );
assert( pPage->nCell==0 );
cellptr = pPage->cellOffset;
data = pPage->aData;
hdr = pPage->hdrOffset;
put2byte(&data[hdr+3], nCell);
if( nCell ){
cellbody = allocateSpace(pPage, totalSize);
assert( cellbody>0 );
assert( pPage->nFree >= 2*nCell );
pPage->nFree -= 2*nCell;
for(i=0; i<nCell; i++){
put2byte(&data[cellptr], cellbody);
memcpy(&data[cellbody], apCell[i], aSize[i]);
cellptr += 2;
cellbody += aSize[i];
}
assert( cellbody==pPage->pBt->usableSize );
}
pPage->nCell = nCell;
}
/*
** The following parameters determine how many adjacent pages get involved
** in a balancing operation. NN is the number of neighbors on either side
** of the page that participate in the balancing operation. NB is the
** total number of pages that participate, including the target page and
** NN neighbors on either side.
**
** The minimum value of NN is 1 (of course). Increasing NN above 1
** (to 2 or 3) gives a modest improvement in SELECT and DELETE performance
** in exchange for a larger degradation in INSERT and UPDATE performance.
** The value of NN appears to give the best results overall.
*/
#define NN 1 /* Number of neighbors on either side of pPage */
#define NB (NN*2+1) /* Total pages involved in the balance */
/* Forward reference */
static int balance(MemPage*, int);
#ifndef SQLITE_OMIT_QUICKBALANCE
/*
** This version of balance() handles the common special case where
** a new entry is being inserted on the extreme right-end of the
** tree, in other words, when the new entry will become the largest
** entry in the tree.
**
** Instead of trying balance the 3 right-most leaf pages, just add
** a new page to the right-hand side and put the one new entry in
** that page. This leaves the right side of the tree somewhat
** unbalanced. But odds are that we will be inserting new entries
** at the end soon afterwards so the nearly empty page will quickly
** fill up. On average.
**
** pPage is the leaf page which is the right-most page in the tree.
** pParent is its parent. pPage must have a single overflow entry
** which is also the right-most entry on the page.
*/
static int balance_quick(MemPage *pPage, MemPage *pParent){
int rc;
MemPage *pNew;
Pgno pgnoNew;
u8 *pCell;
int szCell;
CellInfo info;
BtShared *pBt = pPage->pBt;
int parentIdx = pParent->nCell; /* pParent new divider cell index */
int parentSize; /* Size of new divider cell */
u8 parentCell[64]; /* Space for the new divider cell */
/* Allocate a new page. Insert the overflow cell from pPage
** into it. Then remove the overflow cell from pPage.
*/
rc = allocateBtreePage(pBt, &pNew, &pgnoNew, 0, 0);
if( rc!=SQLITE_OK ){
return rc;
}
pCell = pPage->aOvfl[0].pCell;
szCell = cellSizePtr(pPage, pCell);
zeroPage(pNew, pPage->aData[0]);
assemblePage(pNew, 1, &pCell, &szCell);
pPage->nOverflow = 0;
/* Set the parent of the newly allocated page to pParent. */
pNew->pParent = pParent;
sqlite3PagerRef(pParent->pDbPage);
/* pPage is currently the right-child of pParent. Change this
** so that the right-child is the new page allocated above and
** pPage is the next-to-right child.
*/
assert( pPage->nCell>0 );
pCell = findCell(pPage, pPage->nCell-1);
sqlite3BtreeParseCellPtr(pPage, pCell, &info);
rc = fillInCell(pParent, parentCell, 0, info.nKey, 0, 0, 0, &parentSize);
if( rc!=SQLITE_OK ){
return rc;
}
assert( parentSize<64 );
rc = insertCell(pParent, parentIdx, parentCell, parentSize, 0, 4);
if( rc!=SQLITE_OK ){
return rc;
}
put4byte(findOverflowCell(pParent,parentIdx), pPage->pgno);
put4byte(&pParent->aData[pParent->hdrOffset+8], pgnoNew);
#ifndef SQLITE_OMIT_AUTOVACUUM
/* If this is an auto-vacuum database, update the pointer map
** with entries for the new page, and any pointer from the
** cell on the page to an overflow page.
*/
if( pBt->autoVacuum ){
rc = ptrmapPut(pBt, pgnoNew, PTRMAP_BTREE, pParent->pgno);
if( rc==SQLITE_OK ){
rc = ptrmapPutOvfl(pNew, 0);
}
if( rc!=SQLITE_OK ){
releasePage(pNew);
return rc;
}
}
#endif
/* Release the reference to the new page and balance the parent page,
** in case the divider cell inserted caused it to become overfull.
*/
releasePage(pNew);
return balance(pParent, 0);
}
#endif /* SQLITE_OMIT_QUICKBALANCE */
/*
** This routine redistributes Cells on pPage and up to NN*2 siblings
** of pPage so that all pages have about the same amount of free space.
** Usually NN siblings on either side of pPage is used in the balancing,
** though more siblings might come from one side if pPage is the first
** or last child of its parent. If pPage has fewer than 2*NN siblings
** (something which can only happen if pPage is the root page or a
** child of root) then all available siblings participate in the balancing.
**
** The number of siblings of pPage might be increased or decreased by one or
** two in an effort to keep pages nearly full but not over full. The root page
** is special and is allowed to be nearly empty. If pPage is
** the root page, then the depth of the tree might be increased
** or decreased by one, as necessary, to keep the root page from being
** overfull or completely empty.
**
** Note that when this routine is called, some of the Cells on pPage
** might not actually be stored in pPage->aData[]. This can happen
** if the page is overfull. Part of the job of this routine is to
** make sure all Cells for pPage once again fit in pPage->aData[].
**
** In the course of balancing the siblings of pPage, the parent of pPage
** might become overfull or underfull. If that happens, then this routine
** is called recursively on the parent.
**
** If this routine fails for any reason, it might leave the database
** in a corrupted state. So if this routine fails, the database should
** be rolled back.
*/
static int balance_nonroot(MemPage *pPage){
MemPage *pParent; /* The parent of pPage */
BtShared *pBt; /* The whole database */
int nCell = 0; /* Number of cells in apCell[] */
int nMaxCells = 0; /* Allocated size of apCell, szCell, aFrom. */
int nOld; /* Number of pages in apOld[] */
int nNew; /* Number of pages in apNew[] */
int nDiv; /* Number of cells in apDiv[] */
int i, j, k; /* Loop counters */
int idx; /* Index of pPage in pParent->aCell[] */
int nxDiv; /* Next divider slot in pParent->aCell[] */
int rc; /* The return code */
int leafCorrection; /* 4 if pPage is a leaf. 0 if not */
int leafData; /* True if pPage is a leaf of a LEAFDATA tree */
int usableSpace; /* Bytes in pPage beyond the header */
int pageFlags; /* Value of pPage->aData[0] */
int subtotal; /* Subtotal of bytes in cells on one page */
int iSpace = 0; /* First unused byte of aSpace[] */
MemPage *apOld[NB]; /* pPage and up to two siblings */
Pgno pgnoOld[NB]; /* Page numbers for each page in apOld[] */
MemPage *apCopy[NB]; /* Private copies of apOld[] pages */
MemPage *apNew[NB+2]; /* pPage and up to NB siblings after balancing */
Pgno pgnoNew[NB+2]; /* Page numbers for each page in apNew[] */
u8 *apDiv[NB]; /* Divider cells in pParent */
int cntNew[NB+2]; /* Index in aCell[] of cell after i-th page */
int szNew[NB+2]; /* Combined size of cells place on i-th page */
u8 **apCell = 0; /* All cells begin balanced */
int *szCell; /* Local size of all cells in apCell[] */
u8 *aCopy[NB]; /* Space for holding data of apCopy[] */
u8 *aSpace; /* Space to hold copies of dividers cells */
#ifndef SQLITE_OMIT_AUTOVACUUM
u8 *aFrom = 0;
#endif
/*
** Find the parent page.
*/
assert( pPage->isInit );
assert( sqlite3PagerIswriteable(pPage->pDbPage) );
pBt = pPage->pBt;
pParent = pPage->pParent;
assert( pParent );
if( SQLITE_OK!=(rc = sqlite3PagerWrite(pParent->pDbPage)) ){
return rc;
}
TRACE(("BALANCE: begin page %d child of %d\n", pPage->pgno, pParent->pgno));
#ifndef SQLITE_OMIT_QUICKBALANCE
/*
** A special case: If a new entry has just been inserted into a
** table (that is, a btree with integer keys and all data at the leaves)
** and the new entry is the right-most entry in the tree (it has the
** largest key) then use the special balance_quick() routine for
** balancing. balance_quick() is much faster and results in a tighter
** packing of data in the common case.
*/
if( pPage->leaf &&
pPage->intKey &&
pPage->leafData &&
pPage->nOverflow==1 &&
pPage->aOvfl[0].idx==pPage->nCell &&
pPage->pParent->pgno!=1 &&
get4byte(&pParent->aData[pParent->hdrOffset+8])==pPage->pgno
){
/*
** TODO: Check the siblings to the left of pPage. It may be that
** they are not full and no new page is required.
*/
return balance_quick(pPage, pParent);
}
#endif
/*
** Find the cell in the parent page whose left child points back
** to pPage. The "idx" variable is the index of that cell. If pPage
** is the rightmost child of pParent then set idx to pParent->nCell
*/
if( pParent->idxShift ){
Pgno pgno;
pgno = pPage->pgno;
assert( pgno==sqlite3PagerPagenumber(pPage->pDbPage) );
for(idx=0; idx<pParent->nCell; idx++){
if( get4byte(findCell(pParent, idx))==pgno ){
break;
}
}
assert( idx<pParent->nCell
|| get4byte(&pParent->aData[pParent->hdrOffset+8])==pgno );
}else{
idx = pPage->idxParent;
}
/*
** Initialize variables so that it will be safe to jump
** directly to balance_cleanup at any moment.
*/
nOld = nNew = 0;
sqlite3PagerRef(pParent->pDbPage);
/*
** Find sibling pages to pPage and the cells in pParent that divide
** the siblings. An attempt is made to find NN siblings on either
** side of pPage. More siblings are taken from one side, however, if
** pPage there are fewer than NN siblings on the other side. If pParent
** has NB or fewer children then all children of pParent are taken.
*/
nxDiv = idx - NN;
if( nxDiv + NB > pParent->nCell ){
nxDiv = pParent->nCell - NB + 1;
}
if( nxDiv<0 ){
nxDiv = 0;
}
nDiv = 0;
for(i=0, k=nxDiv; i<NB; i++, k++){
if( k<pParent->nCell ){
apDiv[i] = findCell(pParent, k);
nDiv++;
assert( !pParent->leaf );
pgnoOld[i] = get4byte(apDiv[i]);
}else if( k==pParent->nCell ){
pgnoOld[i] = get4byte(&pParent->aData[pParent->hdrOffset+8]);
}else{
break;
}
rc = getAndInitPage(pBt, pgnoOld[i], &apOld[i], pParent);
if( rc ) goto balance_cleanup;
apOld[i]->idxParent = k;
apCopy[i] = 0;
assert( i==nOld );
nOld++;
nMaxCells += 1+apOld[i]->nCell+apOld[i]->nOverflow;
}
/* Make nMaxCells a multiple of 2 in order to preserve 8-byte
** alignment */
nMaxCells = (nMaxCells + 1)&~1;
/*
** Allocate space for memory structures
*/
apCell = sqliteMallocRaw(
nMaxCells*sizeof(u8*) /* apCell */
+ nMaxCells*sizeof(int) /* szCell */
+ ROUND8(sizeof(MemPage))*NB /* aCopy */
+ pBt->pageSize*(5+NB) /* aSpace */
+ (ISAUTOVACUUM ? nMaxCells : 0) /* aFrom */
);
if( apCell==0 ){
rc = SQLITE_NOMEM;
goto balance_cleanup;
}
szCell = (int*)&apCell[nMaxCells];
aCopy[0] = (u8*)&szCell[nMaxCells];
assert( ((aCopy[0] - (u8*)apCell) & 7)==0 ); /* 8-byte alignment required */
for(i=1; i<NB; i++){
aCopy[i] = &aCopy[i-1][pBt->pageSize+ROUND8(sizeof(MemPage))];
assert( ((aCopy[i] - (u8*)apCell) & 7)==0 ); /* 8-byte alignment required */
}
aSpace = &aCopy[NB-1][pBt->pageSize+ROUND8(sizeof(MemPage))];
assert( ((aSpace - (u8*)apCell) & 7)==0 ); /* 8-byte alignment required */
#ifndef SQLITE_OMIT_AUTOVACUUM
if( pBt->autoVacuum ){
aFrom = &aSpace[5*pBt->pageSize];
}
#endif
/*
** Make copies of the content of pPage and its siblings into aOld[].
** The rest of this function will use data from the copies rather
** that the original pages since the original pages will be in the
** process of being overwritten.
*/
for(i=0; i<nOld; i++){
MemPage *p = apCopy[i] = (MemPage*)&aCopy[i][pBt->pageSize];
p->aData = &((u8*)p)[-pBt->pageSize];
memcpy(p->aData, apOld[i]->aData, pBt->pageSize + sizeof(MemPage));
/* The memcpy() above changes the value of p->aData so we have to
** set it again. */
p->aData = &((u8*)p)[-pBt->pageSize];
}
/*
** Load pointers to all cells on sibling pages and the divider cells
** into the local apCell[] array. Make copies of the divider cells
** into space obtained form aSpace[] and remove the the divider Cells
** from pParent.
**
** If the siblings are on leaf pages, then the child pointers of the
** divider cells are stripped from the cells before they are copied
** into aSpace[]. In this way, all cells in apCell[] are without
** child pointers. If siblings are not leaves, then all cell in
** apCell[] include child pointers. Either way, all cells in apCell[]
** are alike.
**
** leafCorrection: 4 if pPage is a leaf. 0 if pPage is not a leaf.
** leafData: 1 if pPage holds key+data and pParent holds only keys.
*/
nCell = 0;
leafCorrection = pPage->leaf*4;
leafData = pPage->leafData && pPage->leaf;
for(i=0; i<nOld; i++){
MemPage *pOld = apCopy[i];
int limit = pOld->nCell+pOld->nOverflow;
for(j=0; j<limit; j++){
assert( nCell<nMaxCells );
apCell[nCell] = findOverflowCell(pOld, j);
szCell[nCell] = cellSizePtr(pOld, apCell[nCell]);
#ifndef SQLITE_OMIT_AUTOVACUUM
if( pBt->autoVacuum ){
int a;
aFrom[nCell] = i;
for(a=0; a<pOld->nOverflow; a++){
if( pOld->aOvfl[a].pCell==apCell[nCell] ){
aFrom[nCell] = 0xFF;
break;
}
}
}
#endif
nCell++;
}
if( i<nOld-1 ){
int sz = cellSizePtr(pParent, apDiv[i]);
if( leafData ){
/* With the LEAFDATA flag, pParent cells hold only INTKEYs that
** are duplicates of keys on the child pages. We need to remove
** the divider cells from pParent, but the dividers cells are not
** added to apCell[] because they are duplicates of child cells.
*/
dropCell(pParent, nxDiv, sz);
}else{
u8 *pTemp;
assert( nCell<nMaxCells );
szCell[nCell] = sz;
pTemp = &aSpace[iSpace];
iSpace += sz;
assert( iSpace<=pBt->pageSize*5 );
memcpy(pTemp, apDiv[i], sz);
apCell[nCell] = pTemp+leafCorrection;
#ifndef SQLITE_OMIT_AUTOVACUUM
if( pBt->autoVacuum ){
aFrom[nCell] = 0xFF;
}
#endif
dropCell(pParent, nxDiv, sz);
szCell[nCell] -= leafCorrection;
assert( get4byte(pTemp)==pgnoOld[i] );
if( !pOld->leaf ){
assert( leafCorrection==0 );
/* The right pointer of the child page pOld becomes the left
** pointer of the divider cell */
memcpy(apCell[nCell], &pOld->aData[pOld->hdrOffset+8], 4);
}else{
assert( leafCorrection==4 );
if( szCell[nCell]<4 ){
/* Do not allow any cells smaller than 4 bytes. */
szCell[nCell] = 4;
}
}
nCell++;
}
}
}
/*
** Figure out the number of pages needed to hold all nCell cells.
** Store this number in "k". Also compute szNew[] which is the total
** size of all cells on the i-th page and cntNew[] which is the index
** in apCell[] of the cell that divides page i from page i+1.
** cntNew[k] should equal nCell.
**
** Values computed by this block:
**
** k: The total number of sibling pages
** szNew[i]: Spaced used on the i-th sibling page.
** cntNew[i]: Index in apCell[] and szCell[] for the first cell to
** the right of the i-th sibling page.
** usableSpace: Number of bytes of space available on each sibling.
**
*/
usableSpace = pBt->usableSize - 12 + leafCorrection;
for(subtotal=k=i=0; i<nCell; i++){
assert( i<nMaxCells );
subtotal += szCell[i] + 2;
if( subtotal > usableSpace ){
szNew[k] = subtotal - szCell[i];
cntNew[k] = i;
if( leafData ){ i--; }
subtotal = 0;
k++;
}
}
szNew[k] = subtotal;
cntNew[k] = nCell;
k++;
/*
** The packing computed by the previous block is biased toward the siblings
** on the left side. The left siblings are always nearly full, while the
** right-most sibling might be nearly empty. This block of code attempts
** to adjust the packing of siblings to get a better balance.
**
** This adjustment is more than an optimization. The packing above might
** be so out of balance as to be illegal. For example, the right-most
** sibling might be completely empty. This adjustment is not optional.
*/
for(i=k-1; i>0; i--){
int szRight = szNew[i]; /* Size of sibling on the right */
int szLeft = szNew[i-1]; /* Size of sibling on the left */
int r; /* Index of right-most cell in left sibling */
int d; /* Index of first cell to the left of right sibling */
r = cntNew[i-1] - 1;
d = r + 1 - leafData;
assert( d<nMaxCells );
assert( r<nMaxCells );
while( szRight==0 || szRight+szCell[d]+2<=szLeft-(szCell[r]+2) ){
szRight += szCell[d] + 2;
szLeft -= szCell[r] + 2;
cntNew[i-1]--;
r = cntNew[i-1] - 1;
d = r + 1 - leafData;
}
szNew[i] = szRight;
szNew[i-1] = szLeft;
}
/* Either we found one or more cells (cntnew[0])>0) or we are the
** a virtual root page. A virtual root page is when the real root
** page is page 1 and we are the only child of that page.
*/
assert( cntNew[0]>0 || (pParent->pgno==1 && pParent->nCell==0) );
/*
** Allocate k new pages. Reuse old pages where possible.
*/
assert( pPage->pgno>1 );
pageFlags = pPage->aData[0];
for(i=0; i<k; i++){
MemPage *pNew;
if( i<nOld ){
pNew = apNew[i] = apOld[i];
pgnoNew[i] = pgnoOld[i];
apOld[i] = 0;
rc = sqlite3PagerWrite(pNew->pDbPage);
nNew++;
if( rc ) goto balance_cleanup;
}else{
assert( i>0 );
rc = allocateBtreePage(pBt, &pNew, &pgnoNew[i], pgnoNew[i-1], 0);
if( rc ) goto balance_cleanup;
apNew[i] = pNew;
nNew++;
}
zeroPage(pNew, pageFlags);
}
/* Free any old pages that were not reused as new pages.
*/
while( i<nOld ){
rc = freePage(apOld[i]);
if( rc ) goto balance_cleanup;
releasePage(apOld[i]);
apOld[i] = 0;
i++;
}
/*
** Put the new pages in accending order. This helps to
** keep entries in the disk file in order so that a scan
** of the table is a linear scan through the file. That
** in turn helps the operating system to deliver pages
** from the disk more rapidly.
**
** An O(n^2) insertion sort algorithm is used, but since
** n is never more than NB (a small constant), that should
** not be a problem.
**
** When NB==3, this one optimization makes the database
** about 25% faster for large insertions and deletions.
*/
for(i=0; i<k-1; i++){
int minV = pgnoNew[i];
int minI = i;
for(j=i+1; j<k; j++){
if( pgnoNew[j]<(unsigned)minV ){
minI = j;
minV = pgnoNew[j];
}
}
if( minI>i ){
int t;
MemPage *pT;
t = pgnoNew[i];
pT = apNew[i];
pgnoNew[i] = pgnoNew[minI];
apNew[i] = apNew[minI];
pgnoNew[minI] = t;
apNew[minI] = pT;
}
}
TRACE(("BALANCE: old: %d %d %d new: %d(%d) %d(%d) %d(%d) %d(%d) %d(%d)\n",
pgnoOld[0],
nOld>=2 ? pgnoOld[1] : 0,
nOld>=3 ? pgnoOld[2] : 0,
pgnoNew[0], szNew[0],
nNew>=2 ? pgnoNew[1] : 0, nNew>=2 ? szNew[1] : 0,
nNew>=3 ? pgnoNew[2] : 0, nNew>=3 ? szNew[2] : 0,
nNew>=4 ? pgnoNew[3] : 0, nNew>=4 ? szNew[3] : 0,
nNew>=5 ? pgnoNew[4] : 0, nNew>=5 ? szNew[4] : 0));
/*
** Evenly distribute the data in apCell[] across the new pages.
** Insert divider cells into pParent as necessary.
*/
j = 0;
for(i=0; i<nNew; i++){
/* Assemble the new sibling page. */
MemPage *pNew = apNew[i];
assert( j<nMaxCells );
assert( pNew->pgno==pgnoNew[i] );
assemblePage(pNew, cntNew[i]-j, &apCell[j], &szCell[j]);
assert( pNew->nCell>0 || (nNew==1 && cntNew[0]==0) );
assert( pNew->nOverflow==0 );
#ifndef SQLITE_OMIT_AUTOVACUUM
/* If this is an auto-vacuum database, update the pointer map entries
** that point to the siblings that were rearranged. These can be: left
** children of cells, the right-child of the page, or overflow pages
** pointed to by cells.
*/
if( pBt->autoVacuum ){
for(k=j; k<cntNew[i]; k++){
assert( k<nMaxCells );
if( aFrom[k]==0xFF || apCopy[aFrom[k]]->pgno!=pNew->pgno ){
rc = ptrmapPutOvfl(pNew, k-j);
if( rc!=SQLITE_OK ){
goto balance_cleanup;
}
}
}
}
#endif
j = cntNew[i];
/* If the sibling page assembled above was not the right-most sibling,
** insert a divider cell into the parent page.
*/
if( i<nNew-1 && j<nCell ){
u8 *pCell;
u8 *pTemp;
int sz;
assert( j<nMaxCells );
pCell = apCell[j];
sz = szCell[j] + leafCorrection;
if( !pNew->leaf ){
memcpy(&pNew->aData[8], pCell, 4);
pTemp = 0;
}else if( leafData ){
/* If the tree is a leaf-data tree, and the siblings are leaves,
** then there is no divider cell in apCell[]. Instead, the divider
** cell consists of the integer key for the right-most cell of
** the sibling-page assembled above only.
*/
CellInfo info;
j--;
sqlite3BtreeParseCellPtr(pNew, apCell[j], &info);
pCell = &aSpace[iSpace];
fillInCell(pParent, pCell, 0, info.nKey, 0, 0, 0, &sz);
iSpace += sz;
assert( iSpace<=pBt->pageSize*5 );
pTemp = 0;
}else{
pCell -= 4;
pTemp = &aSpace[iSpace];
iSpace += sz;
assert( iSpace<=pBt->pageSize*5 );
/* Obscure case for non-leaf-data trees: If the cell at pCell was
** previously stored on a leaf node, and it's reported size was 4
** bytes, then it may actually be smaller than this
** (see sqlite3BtreeParseCellPtr(), 4 bytes is the minimum size of
** any cell). But it's important to pass the correct size to
** insertCell(), so reparse the cell now.
**
** Note that this can never happen in an SQLite data file, as all
** cells are at least 4 bytes. It only happens in b-trees used
** to evaluate "IN (SELECT ...)" and similar clauses.
*/
if( szCell[j]==4 ){
assert(leafCorrection==4);
sz = cellSizePtr(pParent, pCell);
}
}
rc = insertCell(pParent, nxDiv, pCell, sz, pTemp, 4);
if( rc!=SQLITE_OK ) goto balance_cleanup;
put4byte(findOverflowCell(pParent,nxDiv), pNew->pgno);
#ifndef SQLITE_OMIT_AUTOVACUUM
/* If this is an auto-vacuum database, and not a leaf-data tree,
** then update the pointer map with an entry for the overflow page
** that the cell just inserted points to (if any).
*/
if( pBt->autoVacuum && !leafData ){
rc = ptrmapPutOvfl(pParent, nxDiv);
if( rc!=SQLITE_OK ){
goto balance_cleanup;
}
}
#endif
j++;
nxDiv++;
}
}
assert( j==nCell );
assert( nOld>0 );
assert( nNew>0 );
if( (pageFlags & PTF_LEAF)==0 ){
memcpy(&apNew[nNew-1]->aData[8], &apCopy[nOld-1]->aData[8], 4);
}
if( nxDiv==pParent->nCell+pParent->nOverflow ){
/* Right-most sibling is the right-most child of pParent */
put4byte(&pParent->aData[pParent->hdrOffset+8], pgnoNew[nNew-1]);
}else{
/* Right-most sibling is the left child of the first entry in pParent
** past the right-most divider entry */
put4byte(findOverflowCell(pParent, nxDiv), pgnoNew[nNew-1]);
}
/*
** Reparent children of all cells.
*/
for(i=0; i<nNew; i++){
rc = reparentChildPages(apNew[i]);
if( rc!=SQLITE_OK ) goto balance_cleanup;
}
rc = reparentChildPages(pParent);
if( rc!=SQLITE_OK ) goto balance_cleanup;
/*
** Balance the parent page. Note that the current page (pPage) might
** have been added to the freelist so it might no longer be initialized.
** But the parent page will always be initialized.
*/
assert( pParent->isInit );
rc = balance(pParent, 0);
/*
** Cleanup before returning.
*/
balance_cleanup:
sqliteFree(apCell);
for(i=0; i<nOld; i++){
releasePage(apOld[i]);
}
for(i=0; i<nNew; i++){
releasePage(apNew[i]);
}
releasePage(pParent);
TRACE(("BALANCE: finished with %d: old=%d new=%d cells=%d\n",
pPage->pgno, nOld, nNew, nCell));
return rc;
}
/*
** This routine is called for the root page of a btree when the root
** page contains no cells. This is an opportunity to make the tree
** shallower by one level.
*/
static int balance_shallower(MemPage *pPage){
MemPage *pChild; /* The only child page of pPage */
Pgno pgnoChild; /* Page number for pChild */
int rc = SQLITE_OK; /* Return code from subprocedures */
BtShared *pBt; /* The main BTree structure */
int mxCellPerPage; /* Maximum number of cells per page */
u8 **apCell; /* All cells from pages being balanced */
int *szCell; /* Local size of all cells */
assert( pPage->pParent==0 );
assert( pPage->nCell==0 );
pBt = pPage->pBt;
mxCellPerPage = MX_CELL(pBt);
apCell = sqliteMallocRaw( mxCellPerPage*(sizeof(u8*)+sizeof(int)) );
if( apCell==0 ) return SQLITE_NOMEM;
szCell = (int*)&apCell[mxCellPerPage];
if( pPage->leaf ){
/* The table is completely empty */
TRACE(("BALANCE: empty table %d\n", pPage->pgno));
}else{
/* The root page is empty but has one child. Transfer the
** information from that one child into the root page if it
** will fit. This reduces the depth of the tree by one.
**
** If the root page is page 1, it has less space available than
** its child (due to the 100 byte header that occurs at the beginning
** of the database fle), so it might not be able to hold all of the
** information currently contained in the child. If this is the
** case, then do not do the transfer. Leave page 1 empty except
** for the right-pointer to the child page. The child page becomes
** the virtual root of the tree.
*/
pgnoChild = get4byte(&pPage->aData[pPage->hdrOffset+8]);
assert( pgnoChild>0 );
assert( pgnoChild<=sqlite3PagerPagecount(pPage->pBt->pPager) );
rc = sqlite3BtreeGetPage(pPage->pBt, pgnoChild, &pChild, 0);
if( rc ) goto end_shallow_balance;
if( pPage->pgno==1 ){
rc = sqlite3BtreeInitPage(pChild, pPage);
if( rc ) goto end_shallow_balance;
assert( pChild->nOverflow==0 );
if( pChild->nFree>=100 ){
/* The child information will fit on the root page, so do the
** copy */
int i;
zeroPage(pPage, pChild->aData[0]);
for(i=0; i<pChild->nCell; i++){
apCell[i] = findCell(pChild,i);
szCell[i] = cellSizePtr(pChild, apCell[i]);
}
assemblePage(pPage, pChild->nCell, apCell, szCell);
/* Copy the right-pointer of the child to the parent. */
put4byte(&pPage->aData[pPage->hdrOffset+8],
get4byte(&pChild->aData[pChild->hdrOffset+8]));
freePage(pChild);
TRACE(("BALANCE: child %d transfer to page 1\n", pChild->pgno));
}else{
/* The child has more information that will fit on the root.
** The tree is already balanced. Do nothing. */
TRACE(("BALANCE: child %d will not fit on page 1\n", pChild->pgno));
}
}else{
memcpy(pPage->aData, pChild->aData, pPage->pBt->usableSize);
pPage->isInit = 0;
pPage->pParent = 0;
rc = sqlite3BtreeInitPage(pPage, 0);
assert( rc==SQLITE_OK );
freePage(pChild);
TRACE(("BALANCE: transfer child %d into root %d\n",
pChild->pgno, pPage->pgno));
}
rc = reparentChildPages(pPage);
assert( pPage->nOverflow==0 );
#ifndef SQLITE_OMIT_AUTOVACUUM
if( pBt->autoVacuum ){
int i;
for(i=0; i<pPage->nCell; i++){
rc = ptrmapPutOvfl(pPage, i);
if( rc!=SQLITE_OK ){
goto end_shallow_balance;
}
}
}
#endif
releasePage(pChild);
}
end_shallow_balance:
sqliteFree(apCell);
return rc;
}
/*
** The root page is overfull
**
** When this happens, Create a new child page and copy the
** contents of the root into the child. Then make the root
** page an empty page with rightChild pointing to the new
** child. Finally, call balance_internal() on the new child
** to cause it to split.
*/
static int balance_deeper(MemPage *pPage){
int rc; /* Return value from subprocedures */
MemPage *pChild; /* Pointer to a new child page */
Pgno pgnoChild; /* Page number of the new child page */
BtShared *pBt; /* The BTree */
int usableSize; /* Total usable size of a page */
u8 *data; /* Content of the parent page */
u8 *cdata; /* Content of the child page */
int hdr; /* Offset to page header in parent */
int brk; /* Offset to content of first cell in parent */
assert( pPage->pParent==0 );
assert( pPage->nOverflow>0 );
pBt = pPage->pBt;
rc = allocateBtreePage(pBt, &pChild, &pgnoChild, pPage->pgno, 0);
if( rc ) return rc;
assert( sqlite3PagerIswriteable(pChild->pDbPage) );
usableSize = pBt->usableSize;
data = pPage->aData;
hdr = pPage->hdrOffset;
brk = get2byte(&data[hdr+5]);
cdata = pChild->aData;
memcpy(cdata, &data[hdr], pPage->cellOffset+2*pPage->nCell-hdr);
memcpy(&cdata[brk], &data[brk], usableSize-brk);
assert( pChild->isInit==0 );
rc = sqlite3BtreeInitPage(pChild, pPage);
if( rc ) goto balancedeeper_out;
memcpy(pChild->aOvfl, pPage->aOvfl, pPage->nOverflow*sizeof(pPage->aOvfl[0]));
pChild->nOverflow = pPage->nOverflow;
if( pChild->nOverflow ){
pChild->nFree = 0;
}
assert( pChild->nCell==pPage->nCell );
zeroPage(pPage, pChild->aData[0] & ~PTF_LEAF);
put4byte(&pPage->aData[pPage->hdrOffset+8], pgnoChild);
TRACE(("BALANCE: copy root %d into %d\n", pPage->pgno, pChild->pgno));
#ifndef SQLITE_OMIT_AUTOVACUUM
if( pBt->autoVacuum ){
int i;
rc = ptrmapPut(pBt, pChild->pgno, PTRMAP_BTREE, pPage->pgno);
if( rc ) goto balancedeeper_out;
for(i=0; i<pChild->nCell; i++){
rc = ptrmapPutOvfl(pChild, i);
if( rc!=SQLITE_OK ){
return rc;
}
}
}
#endif
rc = balance_nonroot(pChild);
balancedeeper_out:
releasePage(pChild);
return rc;
}
/*
** Decide if the page pPage needs to be balanced. If balancing is
** required, call the appropriate balancing routine.
*/
static int balance(MemPage *pPage, int insert){
int rc = SQLITE_OK;
if( pPage->pParent==0 ){
if( pPage->nOverflow>0 ){
rc = balance_deeper(pPage);
}
if( rc==SQLITE_OK && pPage->nCell==0 ){
rc = balance_shallower(pPage);
}
}else{
if( pPage->nOverflow>0 ||
(!insert && pPage->nFree>pPage->pBt->usableSize*2/3) ){
rc = balance_nonroot(pPage);
}
}
return rc;
}
/*
** This routine checks all cursors that point to table pgnoRoot.
** If any of those cursors were opened with wrFlag==0 in a different
** database connection (a database connection that shares the pager
** cache with the current connection) and that other connection
** is not in the ReadUncommmitted state, then this routine returns
** SQLITE_LOCKED.
**
** In addition to checking for read-locks (where a read-lock
** means a cursor opened with wrFlag==0) this routine also moves
** all write cursors so that they are pointing to the
** first Cell on the root page. This is necessary because an insert
** or delete might change the number of cells on a page or delete
** a page entirely and we do not want to leave any cursors
** pointing to non-existant pages or cells.
*/
static int checkReadLocks(Btree *pBtree, Pgno pgnoRoot, BtCursor *pExclude){
BtCursor *p;
BtShared *pBt = pBtree->pBt;
sqlite3 *db = pBtree->pSqlite;
for(p=pBt->pCursor; p; p=p->pNext){
if( p==pExclude ) continue;
if( p->eState!=CURSOR_VALID ) continue;
if( p->pgnoRoot!=pgnoRoot ) continue;
if( p->wrFlag==0 ){
sqlite3 *dbOther = p->pBtree->pSqlite;
if( dbOther==0 ||
(dbOther!=db && (dbOther->flags & SQLITE_ReadUncommitted)==0) ){
return SQLITE_LOCKED;
}
}else if( p->pPage->pgno!=p->pgnoRoot ){
moveToRoot(p);
}
}
return SQLITE_OK;
}
/*
** Insert a new record into the BTree. The key is given by (pKey,nKey)
** and the data is given by (pData,nData). The cursor is used only to
** define what table the record should be inserted into. The cursor
** is left pointing at a random location.
**
** For an INTKEY table, only the nKey value of the key is used. pKey is
** ignored. For a ZERODATA table, the pData and nData are both ignored.
*/
int sqlite3BtreeInsert(
BtCursor *pCur, /* Insert data into the table of this cursor */
const void *pKey, i64 nKey, /* The key of the new record */
const void *pData, int nData, /* The data of the new record */
int nZero, /* Number of extra 0 bytes to append to data */
int appendBias /* True if this is likely an append */
){
int rc;
int loc;
int szNew;
MemPage *pPage;
BtShared *pBt = pCur->pBtree->pBt;
unsigned char *oldCell;
unsigned char *newCell = 0;
if( pBt->inTransaction!=TRANS_WRITE ){
/* Must start a transaction before doing an insert */
return pBt->readOnly ? SQLITE_READONLY : SQLITE_ERROR;
}
assert( !pBt->readOnly );
if( !pCur->wrFlag ){
return SQLITE_PERM; /* Cursor not open for writing */
}
if( checkReadLocks(pCur->pBtree, pCur->pgnoRoot, pCur) ){
return SQLITE_LOCKED; /* The table pCur points to has a read lock */
}
/* Save the positions of any other cursors open on this table */
clearCursorPosition(pCur);
if(
SQLITE_OK!=(rc = saveAllCursors(pBt, pCur->pgnoRoot, pCur)) ||
SQLITE_OK!=(rc = sqlite3BtreeMoveto(pCur, pKey, nKey, appendBias, &loc))
){
return rc;
}
pPage = pCur->pPage;
assert( pPage->intKey || nKey>=0 );
assert( pPage->leaf || !pPage->leafData );
TRACE(("INSERT: table=%d nkey=%lld ndata=%d page=%d %s\n",
pCur->pgnoRoot, nKey, nData, pPage->pgno,
loc==0 ? "overwrite" : "new entry"));
assert( pPage->isInit );
rc = sqlite3PagerWrite(pPage->pDbPage);
if( rc ) return rc;
newCell = sqliteMallocRaw( MX_CELL_SIZE(pBt) );
if( newCell==0 ) return SQLITE_NOMEM;
rc = fillInCell(pPage, newCell, pKey, nKey, pData, nData, nZero, &szNew);
if( rc ) goto end_insert;
assert( szNew==cellSizePtr(pPage, newCell) );
assert( szNew<=MX_CELL_SIZE(pBt) );
if( loc==0 && CURSOR_VALID==pCur->eState ){
int szOld;
assert( pCur->idx>=0 && pCur->idx<pPage->nCell );
oldCell = findCell(pPage, pCur->idx);
if( !pPage->leaf ){
memcpy(newCell, oldCell, 4);
}
szOld = cellSizePtr(pPage, oldCell);
rc = clearCell(pPage, oldCell);
if( rc ) goto end_insert;
dropCell(pPage, pCur->idx, szOld);
}else if( loc<0 && pPage->nCell>0 ){
assert( pPage->leaf );
pCur->idx++;
pCur->info.nSize = 0;
}else{
assert( pPage->leaf );
}
rc = insertCell(pPage, pCur->idx, newCell, szNew, 0, 0);
if( rc!=SQLITE_OK ) goto end_insert;
rc = balance(pPage, 1);
/* sqlite3BtreePageDump(pCur->pBt, pCur->pgnoRoot, 1); */
/* fflush(stdout); */
if( rc==SQLITE_OK ){
moveToRoot(pCur);
}
end_insert:
sqliteFree(newCell);
return rc;
}
/*
** Delete the entry that the cursor is pointing to. The cursor
** is left pointing at a random location.
*/
int sqlite3BtreeDelete(BtCursor *pCur){
MemPage *pPage = pCur->pPage;
unsigned char *pCell;
int rc;
Pgno pgnoChild = 0;
BtShared *pBt = pCur->pBtree->pBt;
assert( pPage->isInit );
if( pBt->inTransaction!=TRANS_WRITE ){
/* Must start a transaction before doing a delete */
return pBt->readOnly ? SQLITE_READONLY : SQLITE_ERROR;
}
assert( !pBt->readOnly );
if( pCur->idx >= pPage->nCell ){
return SQLITE_ERROR; /* The cursor is not pointing to anything */
}
if( !pCur->wrFlag ){
return SQLITE_PERM; /* Did not open this cursor for writing */
}
if( checkReadLocks(pCur->pBtree, pCur->pgnoRoot, pCur) ){
return SQLITE_LOCKED; /* The table pCur points to has a read lock */
}
/* Restore the current cursor position (a no-op if the cursor is not in
** CURSOR_REQUIRESEEK state) and save the positions of any other cursors
** open on the same table. Then call sqlite3PagerWrite() on the page
** that the entry will be deleted from.
*/
if(
(rc = restoreOrClearCursorPosition(pCur))!=0 ||
(rc = saveAllCursors(pBt, pCur->pgnoRoot, pCur))!=0 ||
(rc = sqlite3PagerWrite(pPage->pDbPage))!=0
){
return rc;
}
/* Locate the cell within it's page and leave pCell pointing to the
** data. The clearCell() call frees any overflow pages associated with the
** cell. The cell itself is still intact.
*/
pCell = findCell(pPage, pCur->idx);
if( !pPage->leaf ){
pgnoChild = get4byte(pCell);
}
rc = clearCell(pPage, pCell);
if( rc ) return rc;
if( !pPage->leaf ){
/*
** The entry we are about to delete is not a leaf so if we do not
** do something we will leave a hole on an internal page.
** We have to fill the hole by moving in a cell from a leaf. The
** next Cell after the one to be deleted is guaranteed to exist and
** to be a leaf so we can use it.
*/
BtCursor leafCur;
unsigned char *pNext;
int szNext; /* The compiler warning is wrong: szNext is always
** initialized before use. Adding an extra initialization
** to silence the compiler slows down the code. */
int notUsed;
unsigned char *tempCell = 0;
assert( !pPage->leafData );
sqlite3BtreeGetTempCursor(pCur, &leafCur);
rc = sqlite3BtreeNext(&leafCur, ¬Used);
if( rc==SQLITE_OK ){
rc = sqlite3PagerWrite(leafCur.pPage->pDbPage);
}
if( rc==SQLITE_OK ){
TRACE(("DELETE: table=%d delete internal from %d replace from leaf %d\n",
pCur->pgnoRoot, pPage->pgno, leafCur.pPage->pgno));
dropCell(pPage, pCur->idx, cellSizePtr(pPage, pCell));
pNext = findCell(leafCur.pPage, leafCur.idx);
szNext = cellSizePtr(leafCur.pPage, pNext);
assert( MX_CELL_SIZE(pBt)>=szNext+4 );
tempCell = sqliteMallocRaw( MX_CELL_SIZE(pBt) );
if( tempCell==0 ){
rc = SQLITE_NOMEM;
}
}
if( rc==SQLITE_OK ){
rc = insertCell(pPage, pCur->idx, pNext-4, szNext+4, tempCell, 0);
}
if( rc==SQLITE_OK ){
put4byte(findOverflowCell(pPage, pCur->idx), pgnoChild);
rc = balance(pPage, 0);
}
if( rc==SQLITE_OK ){
dropCell(leafCur.pPage, leafCur.idx, szNext);
rc = balance(leafCur.pPage, 0);
}
sqliteFree(tempCell);
sqlite3BtreeReleaseTempCursor(&leafCur);
}else{
TRACE(("DELETE: table=%d delete from leaf %d\n",
pCur->pgnoRoot, pPage->pgno));
dropCell(pPage, pCur->idx, cellSizePtr(pPage, pCell));
rc = balance(pPage, 0);
}
if( rc==SQLITE_OK ){
moveToRoot(pCur);
}
return rc;
}
/*
** Create a new BTree table. Write into *piTable the page
** number for the root page of the new table.
**
** The type of type is determined by the flags parameter. Only the
** following values of flags are currently in use. Other values for
** flags might not work:
**
** BTREE_INTKEY|BTREE_LEAFDATA Used for SQL tables with rowid keys
** BTREE_ZERODATA Used for SQL indices
*/
int sqlite3BtreeCreateTable(Btree *p, int *piTable, int flags){
BtShared *pBt = p->pBt;
MemPage *pRoot;
Pgno pgnoRoot;
int rc;
if( pBt->inTransaction!=TRANS_WRITE ){
/* Must start a transaction first */
return pBt->readOnly ? SQLITE_READONLY : SQLITE_ERROR;
}
assert( !pBt->readOnly );
#ifdef SQLITE_OMIT_AUTOVACUUM
rc = allocateBtreePage(pBt, &pRoot, &pgnoRoot, 1, 0);
if( rc ) return rc;
#else
if( pBt->autoVacuum ){
Pgno pgnoMove; /* Move a page here to make room for the root-page */
MemPage *pPageMove; /* The page to move to. */
/* Creating a new table may probably require moving an existing database
** to make room for the new tables root page. In case this page turns
** out to be an overflow page, delete all overflow page-map caches
** held by open cursors.
*/
invalidateAllOverflowCache(pBt);
/* Read the value of meta[3] from the database to determine where the
** root page of the new table should go. meta[3] is the largest root-page
** created so far, so the new root-page is (meta[3]+1).
*/
rc = sqlite3BtreeGetMeta(p, 4, &pgnoRoot);
if( rc!=SQLITE_OK ) return rc;
pgnoRoot++;
/* The new root-page may not be allocated on a pointer-map page, or the
** PENDING_BYTE page.
*/
if( pgnoRoot==PTRMAP_PAGENO(pBt, pgnoRoot) ||
pgnoRoot==PENDING_BYTE_PAGE(pBt) ){
pgnoRoot++;
}
assert( pgnoRoot>=3 );
/* Allocate a page. The page that currently resides at pgnoRoot will
** be moved to the allocated page (unless the allocated page happens
** to reside at pgnoRoot).
*/
rc = allocateBtreePage(pBt, &pPageMove, &pgnoMove, pgnoRoot, 1);
if( rc!=SQLITE_OK ){
return rc;
}
if( pgnoMove!=pgnoRoot ){
/* pgnoRoot is the page that will be used for the root-page of
** the new table (assuming an error did not occur). But we were
** allocated pgnoMove. If required (i.e. if it was not allocated
** by extending the file), the current page at position pgnoMove
** is already journaled.
*/
u8 eType;
Pgno iPtrPage;
releasePage(pPageMove);
/* Move the page currently at pgnoRoot to pgnoMove. */
rc = sqlite3BtreeGetPage(pBt, pgnoRoot, &pRoot, 0);
if( rc!=SQLITE_OK ){
return rc;
}
rc = ptrmapGet(pBt, pgnoRoot, &eType, &iPtrPage);
if( rc!=SQLITE_OK || eType==PTRMAP_ROOTPAGE || eType==PTRMAP_FREEPAGE ){
releasePage(pRoot);
return rc;
}
assert( eType!=PTRMAP_ROOTPAGE );
assert( eType!=PTRMAP_FREEPAGE );
rc = sqlite3PagerWrite(pRoot->pDbPage);
if( rc!=SQLITE_OK ){
releasePage(pRoot);
return rc;
}
rc = relocatePage(pBt, pRoot, eType, iPtrPage, pgnoMove);
releasePage(pRoot);
/* Obtain the page at pgnoRoot */
if( rc!=SQLITE_OK ){
return rc;
}
rc = sqlite3BtreeGetPage(pBt, pgnoRoot, &pRoot, 0);
if( rc!=SQLITE_OK ){
return rc;
}
rc = sqlite3PagerWrite(pRoot->pDbPage);
if( rc!=SQLITE_OK ){
releasePage(pRoot);
return rc;
}
}else{
pRoot = pPageMove;
}
/* Update the pointer-map and meta-data with the new root-page number. */
rc = ptrmapPut(pBt, pgnoRoot, PTRMAP_ROOTPAGE, 0);
if( rc ){
releasePage(pRoot);
return rc;
}
rc = sqlite3BtreeUpdateMeta(p, 4, pgnoRoot);
if( rc ){
releasePage(pRoot);
return rc;
}
}else{
rc = allocateBtreePage(pBt, &pRoot, &pgnoRoot, 1, 0);
if( rc ) return rc;
}
#endif
assert( sqlite3PagerIswriteable(pRoot->pDbPage) );
zeroPage(pRoot, flags | PTF_LEAF);
sqlite3PagerUnref(pRoot->pDbPage);
*piTable = (int)pgnoRoot;
return SQLITE_OK;
}
/*
** Erase the given database page and all its children. Return
** the page to the freelist.
*/
static int clearDatabasePage(
BtShared *pBt, /* The BTree that contains the table */
Pgno pgno, /* Page number to clear */
MemPage *pParent, /* Parent page. NULL for the root */
int freePageFlag /* Deallocate page if true */
){
MemPage *pPage = 0;
int rc;
unsigned char *pCell;
int i;
if( pgno>sqlite3PagerPagecount(pBt->pPager) ){
return SQLITE_CORRUPT_BKPT;
}
rc = getAndInitPage(pBt, pgno, &pPage, pParent);
if( rc ) goto cleardatabasepage_out;
for(i=0; i<pPage->nCell; i++){
pCell = findCell(pPage, i);
if( !pPage->leaf ){
rc = clearDatabasePage(pBt, get4byte(pCell), pPage->pParent, 1);
if( rc ) goto cleardatabasepage_out;
}
rc = clearCell(pPage, pCell);
if( rc ) goto cleardatabasepage_out;
}
if( !pPage->leaf ){
rc = clearDatabasePage(pBt, get4byte(&pPage->aData[8]), pPage->pParent, 1);
if( rc ) goto cleardatabasepage_out;
}
if( freePageFlag ){
rc = freePage(pPage);
}else if( (rc = sqlite3PagerWrite(pPage->pDbPage))==0 ){
zeroPage(pPage, pPage->aData[0] | PTF_LEAF);
}
cleardatabasepage_out:
releasePage(pPage);
return rc;
}
/*
** Delete all information from a single table in the database. iTable is
** the page number of the root of the table. After this routine returns,
** the root page is empty, but still exists.
**
** This routine will fail with SQLITE_LOCKED if there are any open
** read cursors on the table. Open write cursors are moved to the
** root of the table.
*/
int sqlite3BtreeClearTable(Btree *p, int iTable){
int rc;
BtShared *pBt = p->pBt;
if( p->inTrans!=TRANS_WRITE ){
return pBt->readOnly ? SQLITE_READONLY : SQLITE_ERROR;
}
rc = checkReadLocks(p, iTable, 0);
if( rc ){
return rc;
}
/* Save the position of all cursors open on this table */
if( SQLITE_OK!=(rc = saveAllCursors(pBt, iTable, 0)) ){
return rc;
}
return clearDatabasePage(pBt, (Pgno)iTable, 0, 0);
}
/*
** Erase all information in a table and add the root of the table to
** the freelist. Except, the root of the principle table (the one on
** page 1) is never added to the freelist.
**
** This routine will fail with SQLITE_LOCKED if there are any open
** cursors on the table.
**
** If AUTOVACUUM is enabled and the page at iTable is not the last
** root page in the database file, then the last root page
** in the database file is moved into the slot formerly occupied by
** iTable and that last slot formerly occupied by the last root page
** is added to the freelist instead of iTable. In this say, all
** root pages are kept at the beginning of the database file, which
** is necessary for AUTOVACUUM to work right. *piMoved is set to the
** page number that used to be the last root page in the file before
** the move. If no page gets moved, *piMoved is set to 0.
** The last root page is recorded in meta[3] and the value of
** meta[3] is updated by this procedure.
*/
int sqlite3BtreeDropTable(Btree *p, int iTable, int *piMoved){
int rc;
MemPage *pPage = 0;
BtShared *pBt = p->pBt;
if( p->inTrans!=TRANS_WRITE ){
return pBt->readOnly ? SQLITE_READONLY : SQLITE_ERROR;
}
/* It is illegal to drop a table if any cursors are open on the
** database. This is because in auto-vacuum mode the backend may
** need to move another root-page to fill a gap left by the deleted
** root page. If an open cursor was using this page a problem would
** occur.
*/
if( pBt->pCursor ){
return SQLITE_LOCKED;
}
rc = sqlite3BtreeGetPage(pBt, (Pgno)iTable, &pPage, 0);
if( rc ) return rc;
rc = sqlite3BtreeClearTable(p, iTable);
if( rc ){
releasePage(pPage);
return rc;
}
*piMoved = 0;
if( iTable>1 ){
#ifdef SQLITE_OMIT_AUTOVACUUM
rc = freePage(pPage);
releasePage(pPage);
#else
if( pBt->autoVacuum ){
Pgno maxRootPgno;
rc = sqlite3BtreeGetMeta(p, 4, &maxRootPgno);
if( rc!=SQLITE_OK ){
releasePage(pPage);
return rc;
}
if( iTable==maxRootPgno ){
/* If the table being dropped is the table with the largest root-page
** number in the database, put the root page on the free list.
*/
rc = freePage(pPage);
releasePage(pPage);
if( rc!=SQLITE_OK ){
return rc;
}
}else{
/* The table being dropped does not have the largest root-page
** number in the database. So move the page that does into the
** gap left by the deleted root-page.
*/
MemPage *pMove;
releasePage(pPage);
rc = sqlite3BtreeGetPage(pBt, maxRootPgno, &pMove, 0);
if( rc!=SQLITE_OK ){
return rc;
}
rc = relocatePage(pBt, pMove, PTRMAP_ROOTPAGE, 0, iTable);
releasePage(pMove);
if( rc!=SQLITE_OK ){
return rc;
}
rc = sqlite3BtreeGetPage(pBt, maxRootPgno, &pMove, 0);
if( rc!=SQLITE_OK ){
return rc;
}
rc = freePage(pMove);
releasePage(pMove);
if( rc!=SQLITE_OK ){
return rc;
}
*piMoved = maxRootPgno;
}
/* Set the new 'max-root-page' value in the database header. This
** is the old value less one, less one more if that happens to
** be a root-page number, less one again if that is the
** PENDING_BYTE_PAGE.
*/
maxRootPgno--;
if( maxRootPgno==PENDING_BYTE_PAGE(pBt) ){
maxRootPgno--;
}
if( maxRootPgno==PTRMAP_PAGENO(pBt, maxRootPgno) ){
maxRootPgno--;
}
assert( maxRootPgno!=PENDING_BYTE_PAGE(pBt) );
rc = sqlite3BtreeUpdateMeta(p, 4, maxRootPgno);
}else{
rc = freePage(pPage);
releasePage(pPage);
}
#endif
}else{
/* If sqlite3BtreeDropTable was called on page 1. */
zeroPage(pPage, PTF_INTKEY|PTF_LEAF );
releasePage(pPage);
}
return rc;
}
/*
** Read the meta-information out of a database file. Meta[0]
** is the number of free pages currently in the database. Meta[1]
** through meta[15] are available for use by higher layers. Meta[0]
** is read-only, the others are read/write.
**
** The schema layer numbers meta values differently. At the schema
** layer (and the SetCookie and ReadCookie opcodes) the number of
** free pages is not visible. So Cookie[0] is the same as Meta[1].
*/
int sqlite3BtreeGetMeta(Btree *p, int idx, u32 *pMeta){
DbPage *pDbPage;
int rc;
unsigned char *pP1;
BtShared *pBt = p->pBt;
/* Reading a meta-data value requires a read-lock on page 1 (and hence
** the sqlite_master table. We grab this lock regardless of whether or
** not the SQLITE_ReadUncommitted flag is set (the table rooted at page
** 1 is treated as a special case by queryTableLock() and lockTable()).
*/
rc = queryTableLock(p, 1, READ_LOCK);
if( rc!=SQLITE_OK ){
return rc;
}
assert( idx>=0 && idx<=15 );
rc = sqlite3PagerGet(pBt->pPager, 1, &pDbPage);
if( rc ) return rc;
pP1 = (unsigned char *)sqlite3PagerGetData(pDbPage);
*pMeta = get4byte(&pP1[36 + idx*4]);
sqlite3PagerUnref(pDbPage);
/* If autovacuumed is disabled in this build but we are trying to
** access an autovacuumed database, then make the database readonly.
*/
#ifdef SQLITE_OMIT_AUTOVACUUM
if( idx==4 && *pMeta>0 ) pBt->readOnly = 1;
#endif
/* Grab the read-lock on page 1. */
rc = lockTable(p, 1, READ_LOCK);
return rc;
}
/*
** Write meta-information back into the database. Meta[0] is
** read-only and may not be written.
*/
int sqlite3BtreeUpdateMeta(Btree *p, int idx, u32 iMeta){
BtShared *pBt = p->pBt;
unsigned char *pP1;
int rc;
assert( idx>=1 && idx<=15 );
if( p->inTrans!=TRANS_WRITE ){
return pBt->readOnly ? SQLITE_READONLY : SQLITE_ERROR;
}
assert( pBt->pPage1!=0 );
pP1 = pBt->pPage1->aData;
rc = sqlite3PagerWrite(pBt->pPage1->pDbPage);
if( rc ) return rc;
put4byte(&pP1[36 + idx*4], iMeta);
if( idx==7 ){
assert( pBt->autoVacuum || iMeta==0 );
assert( iMeta==0 || iMeta==1 );
pBt->incrVacuum = iMeta;
}
return SQLITE_OK;
}
/*
** Return the flag byte at the beginning of the page that the cursor
** is currently pointing to.
*/
int sqlite3BtreeFlags(BtCursor *pCur){
/* TODO: What about CURSOR_REQUIRESEEK state? Probably need to call
** restoreOrClearCursorPosition() here.
*/
MemPage *pPage = pCur->pPage;
return pPage ? pPage->aData[pPage->hdrOffset] : 0;
}
/*
** Return the pager associated with a BTree. This routine is used for
** testing and debugging only.
*/
Pager *sqlite3BtreePager(Btree *p){
return p->pBt->pPager;
}
#ifndef SQLITE_OMIT_INTEGRITY_CHECK
/*
** Append a message to the error message string.
*/
static void checkAppendMsg(
IntegrityCk *pCheck,
char *zMsg1,
const char *zFormat,
...
){
va_list ap;
char *zMsg2;
if( !pCheck->mxErr ) return;
pCheck->mxErr--;
pCheck->nErr++;
va_start(ap, zFormat);
zMsg2 = sqlite3VMPrintf(zFormat, ap);
va_end(ap);
if( zMsg1==0 ) zMsg1 = "";
if( pCheck->zErrMsg ){
char *zOld = pCheck->zErrMsg;
pCheck->zErrMsg = 0;
sqlite3SetString(&pCheck->zErrMsg, zOld, "\n", zMsg1, zMsg2, (char*)0);
sqliteFree(zOld);
}else{
sqlite3SetString(&pCheck->zErrMsg, zMsg1, zMsg2, (char*)0);
}
sqliteFree(zMsg2);
}
#endif /* SQLITE_OMIT_INTEGRITY_CHECK */
#ifndef SQLITE_OMIT_INTEGRITY_CHECK
/*
** Add 1 to the reference count for page iPage. If this is the second
** reference to the page, add an error message to pCheck->zErrMsg.
** Return 1 if there are 2 ore more references to the page and 0 if
** if this is the first reference to the page.
**
** Also check that the page number is in bounds.
*/
static int checkRef(IntegrityCk *pCheck, int iPage, char *zContext){
if( iPage==0 ) return 1;
if( iPage>pCheck->nPage || iPage<0 ){
checkAppendMsg(pCheck, zContext, "invalid page number %d", iPage);
return 1;
}
if( pCheck->anRef[iPage]==1 ){
checkAppendMsg(pCheck, zContext, "2nd reference to page %d", iPage);
return 1;
}
return (pCheck->anRef[iPage]++)>1;
}
#ifndef SQLITE_OMIT_AUTOVACUUM
/*
** Check that the entry in the pointer-map for page iChild maps to
** page iParent, pointer type ptrType. If not, append an error message
** to pCheck.
*/
static void checkPtrmap(
IntegrityCk *pCheck, /* Integrity check context */
Pgno iChild, /* Child page number */
u8 eType, /* Expected pointer map type */
Pgno iParent, /* Expected pointer map parent page number */
char *zContext /* Context description (used for error msg) */
){
int rc;
u8 ePtrmapType;
Pgno iPtrmapParent;
rc = ptrmapGet(pCheck->pBt, iChild, &ePtrmapType, &iPtrmapParent);
if( rc!=SQLITE_OK ){
checkAppendMsg(pCheck, zContext, "Failed to read ptrmap key=%d", iChild);
return;
}
if( ePtrmapType!=eType || iPtrmapParent!=iParent ){
checkAppendMsg(pCheck, zContext,
"Bad ptr map entry key=%d expected=(%d,%d) got=(%d,%d)",
iChild, eType, iParent, ePtrmapType, iPtrmapParent);
}
}
#endif
/*
** Check the integrity of the freelist or of an overflow page list.
** Verify that the number of pages on the list is N.
*/
static void checkList(
IntegrityCk *pCheck, /* Integrity checking context */
int isFreeList, /* True for a freelist. False for overflow page list */
int iPage, /* Page number for first page in the list */
int N, /* Expected number of pages in the list */
char *zContext /* Context for error messages */
){
int i;
int expected = N;
int iFirst = iPage;
while( N-- > 0 && pCheck->mxErr ){
DbPage *pOvflPage;
unsigned char *pOvflData;
if( iPage<1 ){
checkAppendMsg(pCheck, zContext,
"%d of %d pages missing from overflow list starting at %d",
N+1, expected, iFirst);
break;
}
if( checkRef(pCheck, iPage, zContext) ) break;
if( sqlite3PagerGet(pCheck->pPager, (Pgno)iPage, &pOvflPage) ){
checkAppendMsg(pCheck, zContext, "failed to get page %d", iPage);
break;
}
pOvflData = (unsigned char *)sqlite3PagerGetData(pOvflPage);
if( isFreeList ){
int n = get4byte(&pOvflData[4]);
#ifndef SQLITE_OMIT_AUTOVACUUM
if( pCheck->pBt->autoVacuum ){
checkPtrmap(pCheck, iPage, PTRMAP_FREEPAGE, 0, zContext);
}
#endif
if( n>pCheck->pBt->usableSize/4-8 ){
checkAppendMsg(pCheck, zContext,
"freelist leaf count too big on page %d", iPage);
N--;
}else{
for(i=0; i<n; i++){
Pgno iFreePage = get4byte(&pOvflData[8+i*4]);
#ifndef SQLITE_OMIT_AUTOVACUUM
if( pCheck->pBt->autoVacuum ){
checkPtrmap(pCheck, iFreePage, PTRMAP_FREEPAGE, 0, zContext);
}
#endif
checkRef(pCheck, iFreePage, zContext);
}
N -= n;
}
}
#ifndef SQLITE_OMIT_AUTOVACUUM
else{
/* If this database supports auto-vacuum and iPage is not the last
** page in this overflow list, check that the pointer-map entry for
** the following page matches iPage.
*/
if( pCheck->pBt->autoVacuum && N>0 ){
i = get4byte(pOvflData);
checkPtrmap(pCheck, i, PTRMAP_OVERFLOW2, iPage, zContext);
}
}
#endif
iPage = get4byte(pOvflData);
sqlite3PagerUnref(pOvflPage);
}
}
#endif /* SQLITE_OMIT_INTEGRITY_CHECK */
#ifndef SQLITE_OMIT_INTEGRITY_CHECK
/*
** Do various sanity checks on a single page of a tree. Return
** the tree depth. Root pages return 0. Parents of root pages
** return 1, and so forth.
**
** These checks are done:
**
** 1. Make sure that cells and freeblocks do not overlap
** but combine to completely cover the page.
** NO 2. Make sure cell keys are in order.
** NO 3. Make sure no key is less than or equal to zLowerBound.
** NO 4. Make sure no key is greater than or equal to zUpperBound.
** 5. Check the integrity of overflow pages.
** 6. Recursively call checkTreePage on all children.
** 7. Verify that the depth of all children is the same.
** 8. Make sure this page is at least 33% full or else it is
** the root of the tree.
*/
static int checkTreePage(
IntegrityCk *pCheck, /* Context for the sanity check */
int iPage, /* Page number of the page to check */
MemPage *pParent, /* Parent page */
char *zParentContext /* Parent context */
){
MemPage *pPage;
int i, rc, depth, d2, pgno, cnt;
int hdr, cellStart;
int nCell;
u8 *data;
BtShared *pBt;
int usableSize;
char zContext[100];
char *hit;
sqlite3_snprintf(sizeof(zContext), zContext, "Page %d: ", iPage);
/* Check that the page exists
*/
pBt = pCheck->pBt;
usableSize = pBt->usableSize;
if( iPage==0 ) return 0;
if( checkRef(pCheck, iPage, zParentContext) ) return 0;
if( (rc = sqlite3BtreeGetPage(pBt, (Pgno)iPage, &pPage, 0))!=0 ){
checkAppendMsg(pCheck, zContext,
"unable to get the page. error code=%d", rc);
return 0;
}
if( (rc = sqlite3BtreeInitPage(pPage, pParent))!=0 ){
checkAppendMsg(pCheck, zContext,
"sqlite3BtreeInitPage() returns error code %d", rc);
releasePage(pPage);
return 0;
}
/* Check out all the cells.
*/
depth = 0;
for(i=0; i<pPage->nCell && pCheck->mxErr; i++){
u8 *pCell;
int sz;
CellInfo info;
/* Check payload overflow pages
*/
sqlite3_snprintf(sizeof(zContext), zContext,
"On tree page %d cell %d: ", iPage, i);
pCell = findCell(pPage,i);
sqlite3BtreeParseCellPtr(pPage, pCell, &info);
sz = info.nData;
if( !pPage->intKey ) sz += info.nKey;
assert( sz==info.nPayload );
if( sz>info.nLocal ){
int nPage = (sz - info.nLocal + usableSize - 5)/(usableSize - 4);
Pgno pgnoOvfl = get4byte(&pCell[info.iOverflow]);
#ifndef SQLITE_OMIT_AUTOVACUUM
if( pBt->autoVacuum ){
checkPtrmap(pCheck, pgnoOvfl, PTRMAP_OVERFLOW1, iPage, zContext);
}
#endif
checkList(pCheck, 0, pgnoOvfl, nPage, zContext);
}
/* Check sanity of left child page.
*/
if( !pPage->leaf ){
pgno = get4byte(pCell);
#ifndef SQLITE_OMIT_AUTOVACUUM
if( pBt->autoVacuum ){
checkPtrmap(pCheck, pgno, PTRMAP_BTREE, iPage, zContext);
}
#endif
d2 = checkTreePage(pCheck,pgno,pPage,zContext);
if( i>0 && d2!=depth ){
checkAppendMsg(pCheck, zContext, "Child page depth differs");
}
depth = d2;
}
}
if( !pPage->leaf ){
pgno = get4byte(&pPage->aData[pPage->hdrOffset+8]);
sqlite3_snprintf(sizeof(zContext), zContext,
"On page %d at right child: ", iPage);
#ifndef SQLITE_OMIT_AUTOVACUUM
if( pBt->autoVacuum ){
checkPtrmap(pCheck, pgno, PTRMAP_BTREE, iPage, 0);
}
#endif
checkTreePage(pCheck, pgno, pPage, zContext);
}
/* Check for complete coverage of the page
*/
data = pPage->aData;
hdr = pPage->hdrOffset;
hit = sqliteMalloc( usableSize );
if( hit ){
memset(hit, 1, get2byte(&data[hdr+5]));
nCell = get2byte(&data[hdr+3]);
cellStart = hdr + 12 - 4*pPage->leaf;
for(i=0; i<nCell; i++){
int pc = get2byte(&data[cellStart+i*2]);
int size = cellSizePtr(pPage, &data[pc]);
int j;
if( (pc+size-1)>=usableSize || pc<0 ){
checkAppendMsg(pCheck, 0,
"Corruption detected in cell %d on page %d",i,iPage,0);
}else{
for(j=pc+size-1; j>=pc; j--) hit[j]++;
}
}
for(cnt=0, i=get2byte(&data[hdr+1]); i>0 && i<usableSize && cnt<10000;
cnt++){
int size = get2byte(&data[i+2]);
int j;
if( (i+size-1)>=usableSize || i<0 ){
checkAppendMsg(pCheck, 0,
"Corruption detected in cell %d on page %d",i,iPage,0);
}else{
for(j=i+size-1; j>=i; j--) hit[j]++;
}
i = get2byte(&data[i]);
}
for(i=cnt=0; i<usableSize; i++){
if( hit[i]==0 ){
cnt++;
}else if( hit[i]>1 ){
checkAppendMsg(pCheck, 0,
"Multiple uses for byte %d of page %d", i, iPage);
break;
}
}
if( cnt!=data[hdr+7] ){
checkAppendMsg(pCheck, 0,
"Fragmented space is %d byte reported as %d on page %d",
cnt, data[hdr+7], iPage);
}
}
sqliteFree(hit);
releasePage(pPage);
return depth+1;
}
#endif /* SQLITE_OMIT_INTEGRITY_CHECK */
#ifndef SQLITE_OMIT_INTEGRITY_CHECK
/*
** This routine does a complete check of the given BTree file. aRoot[] is
** an array of pages numbers were each page number is the root page of
** a table. nRoot is the number of entries in aRoot.
**
** If everything checks out, this routine returns NULL. If something is
** amiss, an error message is written into memory obtained from malloc()
** and a pointer to that error message is returned. The calling function
** is responsible for freeing the error message when it is done.
*/
char *sqlite3BtreeIntegrityCheck(
Btree *p, /* The btree to be checked */
int *aRoot, /* An array of root pages numbers for individual trees */
int nRoot, /* Number of entries in aRoot[] */
int mxErr, /* Stop reporting errors after this many */
int *pnErr /* Write number of errors seen to this variable */
){
int i;
int nRef;
IntegrityCk sCheck;
BtShared *pBt = p->pBt;
nRef = sqlite3PagerRefcount(pBt->pPager);
if( lockBtreeWithRetry(p)!=SQLITE_OK ){
return sqliteStrDup("Unable to acquire a read lock on the database");
}
sCheck.pBt = pBt;
sCheck.pPager = pBt->pPager;
sCheck.nPage = sqlite3PagerPagecount(sCheck.pPager);
sCheck.mxErr = mxErr;
sCheck.nErr = 0;
*pnErr = 0;
#ifndef SQLITE_OMIT_AUTOVACUUM
if( pBt->nTrunc!=0 ){
sCheck.nPage = pBt->nTrunc;
}
#endif
if( sCheck.nPage==0 ){
unlockBtreeIfUnused(pBt);
return 0;
}
sCheck.anRef = sqliteMallocRaw( (sCheck.nPage+1)*sizeof(sCheck.anRef[0]) );
if( !sCheck.anRef ){
unlockBtreeIfUnused(pBt);
*pnErr = 1;
return sqlite3MPrintf("Unable to malloc %d bytes",
(sCheck.nPage+1)*sizeof(sCheck.anRef[0]));
}
for(i=0; i<=sCheck.nPage; i++){ sCheck.anRef[i] = 0; }
i = PENDING_BYTE_PAGE(pBt);
if( i<=sCheck.nPage ){
sCheck.anRef[i] = 1;
}
sCheck.zErrMsg = 0;
/* Check the integrity of the freelist
*/
checkList(&sCheck, 1, get4byte(&pBt->pPage1->aData[32]),
get4byte(&pBt->pPage1->aData[36]), "Main freelist: ");
/* Check all the tables.
*/
for(i=0; i<nRoot && sCheck.mxErr; i++){
if( aRoot[i]==0 ) continue;
#ifndef SQLITE_OMIT_AUTOVACUUM
if( pBt->autoVacuum && aRoot[i]>1 ){
checkPtrmap(&sCheck, aRoot[i], PTRMAP_ROOTPAGE, 0, 0);
}
#endif
checkTreePage(&sCheck, aRoot[i], 0, "List of tree roots: ");
}
/* Make sure every page in the file is referenced
*/
for(i=1; i<=sCheck.nPage && sCheck.mxErr; i++){
#ifdef SQLITE_OMIT_AUTOVACUUM
if( sCheck.anRef[i]==0 ){
checkAppendMsg(&sCheck, 0, "Page %d is never used", i);
}
#else
/* If the database supports auto-vacuum, make sure no tables contain
** references to pointer-map pages.
*/
if( sCheck.anRef[i]==0 &&
(PTRMAP_PAGENO(pBt, i)!=i || !pBt->autoVacuum) ){
checkAppendMsg(&sCheck, 0, "Page %d is never used", i);
}
if( sCheck.anRef[i]!=0 &&
(PTRMAP_PAGENO(pBt, i)==i && pBt->autoVacuum) ){
checkAppendMsg(&sCheck, 0, "Pointer map page %d is referenced", i);
}
#endif
}
/* Make sure this analysis did not leave any unref() pages
*/
unlockBtreeIfUnused(pBt);
if( nRef != sqlite3PagerRefcount(pBt->pPager) ){
checkAppendMsg(&sCheck, 0,
"Outstanding page count goes from %d to %d during this analysis",
nRef, sqlite3PagerRefcount(pBt->pPager)
);
}
/* Clean up and report errors.
*/
sqliteFree(sCheck.anRef);
*pnErr = sCheck.nErr;
return sCheck.zErrMsg;
}
#endif /* SQLITE_OMIT_INTEGRITY_CHECK */
/*
** Return the full pathname of the underlying database file.
*/
const char *sqlite3BtreeGetFilename(Btree *p){
assert( p->pBt->pPager!=0 );
return sqlite3PagerFilename(p->pBt->pPager);
}
/*
** Return the pathname of the directory that contains the database file.
*/
const char *sqlite3BtreeGetDirname(Btree *p){
assert( p->pBt->pPager!=0 );
return sqlite3PagerDirname(p->pBt->pPager);
}
/*
** Return the pathname of the journal file for this database. The return
** value of this routine is the same regardless of whether the journal file
** has been created or not.
*/
const char *sqlite3BtreeGetJournalname(Btree *p){
assert( p->pBt->pPager!=0 );
return sqlite3PagerJournalname(p->pBt->pPager);
}
#ifndef SQLITE_OMIT_VACUUM
/*
** Copy the complete content of pBtFrom into pBtTo. A transaction
** must be active for both files.
**
** The size of file pBtFrom may be reduced by this operation.
** If anything goes wrong, the transaction on pBtFrom is rolled back.
*/
int sqlite3BtreeCopyFile(Btree *pTo, Btree *pFrom){
int rc = SQLITE_OK;
Pgno i, nPage, nToPage, iSkip;
BtShared *pBtTo = pTo->pBt;
BtShared *pBtFrom = pFrom->pBt;
if( pTo->inTrans!=TRANS_WRITE || pFrom->inTrans!=TRANS_WRITE ){
return SQLITE_ERROR;
}
if( pBtTo->pCursor ) return SQLITE_BUSY;
nToPage = sqlite3PagerPagecount(pBtTo->pPager);
nPage = sqlite3PagerPagecount(pBtFrom->pPager);
iSkip = PENDING_BYTE_PAGE(pBtTo);
for(i=1; rc==SQLITE_OK && i<=nPage; i++){
DbPage *pDbPage;
if( i==iSkip ) continue;
rc = sqlite3PagerGet(pBtFrom->pPager, i, &pDbPage);
if( rc ) break;
rc = sqlite3PagerOverwrite(pBtTo->pPager, i, sqlite3PagerGetData(pDbPage));
sqlite3PagerUnref(pDbPage);
}
/* If the file is shrinking, journal the pages that are being truncated
** so that they can be rolled back if the commit fails.
*/
for(i=nPage+1; rc==SQLITE_OK && i<=nToPage; i++){
DbPage *pDbPage;
if( i==iSkip ) continue;
rc = sqlite3PagerGet(pBtTo->pPager, i, &pDbPage);
if( rc ) break;
rc = sqlite3PagerWrite(pDbPage);
sqlite3PagerDontWrite(pDbPage);
/* Yeah. It seems wierd to call DontWrite() right after Write(). But
** that is because the names of those procedures do not exactly
** represent what they do. Write() really means "put this page in the
** rollback journal and mark it as dirty so that it will be written
** to the database file later." DontWrite() undoes the second part of
** that and prevents the page from being written to the database. The
** page is still on the rollback journal, though. And that is the whole
** point of this loop: to put pages on the rollback journal. */
sqlite3PagerUnref(pDbPage);
}
if( !rc && nPage<nToPage ){
rc = sqlite3PagerTruncate(pBtTo->pPager, nPage);
}
if( rc ){
sqlite3BtreeRollback(pTo);
}
return rc;
}
#endif /* SQLITE_OMIT_VACUUM */
/*
** Return non-zero if a transaction is active.
*/
int sqlite3BtreeIsInTrans(Btree *p){
return (p && (p->inTrans==TRANS_WRITE));
}
/*
** Return non-zero if a statement transaction is active.
*/
int sqlite3BtreeIsInStmt(Btree *p){
return (p->pBt && p->pBt->inStmt);
}
/*
** Return non-zero if a read (or write) transaction is active.
*/
int sqlite3BtreeIsInReadTrans(Btree *p){
return (p && (p->inTrans!=TRANS_NONE));
}
/*
** This function returns a pointer to a blob of memory associated with
** a single shared-btree. The memory is used by client code for it's own
** purposes (for example, to store a high-level schema associated with
** the shared-btree). The btree layer manages reference counting issues.
**
** The first time this is called on a shared-btree, nBytes bytes of memory
** are allocated, zeroed, and returned to the caller. For each subsequent
** call the nBytes parameter is ignored and a pointer to the same blob
** of memory returned.
**
** Just before the shared-btree is closed, the function passed as the
** xFree argument when the memory allocation was made is invoked on the
** blob of allocated memory. This function should not call sqliteFree()
** on the memory, the btree layer does that.
*/
void *sqlite3BtreeSchema(Btree *p, int nBytes, void(*xFree)(void *)){
BtShared *pBt = p->pBt;
if( !pBt->pSchema ){
pBt->pSchema = sqliteMalloc(nBytes);
pBt->xFreeSchema = xFree;
}
return pBt->pSchema;
}
/*
** Return true if another user of the same shared btree as the argument
** handle holds an exclusive lock on the sqlite_master table.
*/
int sqlite3BtreeSchemaLocked(Btree *p){
return (queryTableLock(p, MASTER_ROOT, READ_LOCK)!=SQLITE_OK);
}
#ifndef SQLITE_OMIT_SHARED_CACHE
/*
** Obtain a lock on the table whose root page is iTab. The
** lock is a write lock if isWritelock is true or a read lock
** if it is false.
*/
int sqlite3BtreeLockTable(Btree *p, int iTab, u8 isWriteLock){
int rc = SQLITE_OK;
u8 lockType = (isWriteLock?WRITE_LOCK:READ_LOCK);
rc = queryTableLock(p, iTab, lockType);
if( rc==SQLITE_OK ){
rc = lockTable(p, iTab, lockType);
}
return rc;
}
#endif
#ifndef SQLITE_OMIT_INCRBLOB
/*
** Argument pCsr must be a cursor opened for writing on an
** INTKEY table currently pointing at a valid table entry.
** This function modifies the data stored as part of that entry.
** Only the data content may only be modified, it is not possible
** to change the length of the data stored.
*/
int sqlite3BtreePutData(BtCursor *pCsr, u32 offset, u32 amt, void *z){
assert(pCsr->isIncrblobHandle);
if( pCsr->eState==CURSOR_REQUIRESEEK ){
return SQLITE_ABORT;
}
/* Check some preconditions:
** (a) the cursor is open for writing,
** (b) there is no read-lock on the table being modified and
** (c) the cursor points at a valid row of an intKey table.
*/
if( !pCsr->wrFlag ){
return SQLITE_READONLY;
}
assert( !pCsr->pBtree->pBt->readOnly
&& pCsr->pBtree->pBt->inTransaction==TRANS_WRITE );
if( checkReadLocks(pCsr->pBtree, pCsr->pgnoRoot, pCsr) ){
return SQLITE_LOCKED; /* The table pCur points to has a read lock */
}
if( pCsr->eState==CURSOR_INVALID || !pCsr->pPage->intKey ){
return SQLITE_ERROR;
}
return accessPayload(pCsr, offset, amt, (unsigned char *)z, 0, 1);
}
/*
** Set a flag on this cursor to cache the locations of pages from the
** overflow list for the current row. This is used by cursors opened
** for incremental blob IO only.
**
** This function sets a flag only. The actual page location cache
** (stored in BtCursor.aOverflow[]) is allocated and used by function
** accessPayload() (the worker function for sqlite3BtreeData() and
** sqlite3BtreePutData()).
*/
void sqlite3BtreeCacheOverflow(BtCursor *pCur){
assert(!pCur->isIncrblobHandle);
assert(!pCur->aOverflow);
pCur->isIncrblobHandle = 1;
}
#endif
|