1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298 3299 3300 3301 3302 3303 3304 3305 3306 3307 3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319 3320 3321 3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333 3334 3335 3336 3337 3338 3339 3340 3341 3342 3343 3344 3345 3346 3347 3348 3349 3350 3351 3352 3353 3354 3355 3356 3357 3358 3359 3360 3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397 3398 3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467 3468 3469 3470 3471 3472 3473 3474 3475 3476 3477 3478 3479 3480 3481 3482 3483 3484 3485 3486 3487 3488 3489 3490 3491 3492 3493 3494 3495 3496 3497 3498 3499 3500 3501 3502 3503 3504 3505 3506 3507 3508 3509 3510 3511 3512 3513 3514 3515 3516 3517 3518 3519 3520 3521 3522 3523 3524 3525 3526 3527 3528 3529 3530 3531 3532 3533 3534 3535 3536 3537 3538 3539 3540 3541 3542 3543 3544 3545 3546 3547 3548 3549 3550 3551 3552 3553 3554 3555 3556 3557 3558 3559 3560 3561 3562 3563 3564 3565 3566 3567 3568 3569 3570 3571 3572 3573 3574 3575 3576 3577 3578 3579 3580 3581 3582 3583 3584 3585 3586 3587 3588 3589 3590 3591 3592 3593 3594 3595 3596 3597 3598 3599 3600 3601 3602 3603 3604 3605 3606 3607 3608 3609 3610 3611 3612 3613 3614 3615 3616 3617 3618 3619 3620 3621 3622 3623 3624 3625 3626 3627 3628 3629 3630 3631 3632 3633 3634 3635 3636 3637 3638 3639 3640 3641 3642 3643 3644 3645 3646 3647 3648 3649 3650 3651 3652 3653 3654 3655 3656 3657 3658 3659 3660 3661 3662 3663 3664 3665 3666 3667 3668 3669 3670 3671 3672 3673 3674 3675 3676 3677 3678 3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690 3691 3692 3693 3694 3695 3696 3697 3698 3699 3700 3701 3702 3703 3704 3705 3706 3707 3708 3709 3710 3711 3712 3713 3714 3715 3716 3717 3718 3719 3720 3721 3722 3723 3724 3725 3726 3727 3728 3729 3730 3731 3732 3733 3734 3735 3736 3737 3738 3739 3740 3741 3742 3743 3744 3745 3746 3747 3748 3749 3750 3751 3752 3753 3754 3755 3756 3757 3758 3759 3760 3761 3762 3763 3764 3765 3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799 3800 3801 3802 3803 3804 3805 3806 3807 3808 3809 3810 3811 3812 3813 3814 3815 3816 3817 3818 3819 3820 3821 3822 3823 3824 3825 3826 3827 3828 3829 3830 3831 3832 3833 3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854 3855 3856 3857 3858 3859 3860 3861 3862 3863 3864 3865 3866 3867 3868 3869 3870 3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885 3886 3887 3888 3889 3890 3891 3892 3893 3894 3895 3896 3897 3898 3899 3900 3901 3902 3903 3904 3905 3906 3907 3908 3909 3910 3911 3912 3913 3914 3915 3916 3917 3918 3919 3920 3921 3922 3923 3924 3925 3926 3927 3928 3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947 3948 3949 3950 3951 3952 3953 3954 3955 3956 3957 3958 3959 3960 3961 3962 3963 3964 3965 3966 3967 3968 3969 3970 3971 3972 3973 3974 3975 3976 3977 3978 3979 3980 3981 3982 3983 3984 3985 3986 3987 3988 3989 3990 3991 3992 3993 3994 3995 3996 3997 3998 3999 4000 4001 4002 4003 4004 4005 4006 4007 4008 4009 4010 4011 4012 4013 4014 4015 4016 4017 4018 4019 4020 4021 4022 4023 4024 4025 4026 4027 4028 4029 4030 4031 4032 4033 4034 4035 4036 4037 4038 4039 4040 4041 4042 4043 4044 4045 4046 4047 4048 4049 4050 4051 4052 4053 4054 4055 4056 4057 4058 4059 4060 4061 4062 4063 4064 4065 4066 4067 4068 4069 4070 4071 4072 4073 4074 4075 4076 4077 4078 4079 4080 4081 4082 4083 4084 4085 4086 4087 4088 4089 4090 4091 4092 4093 4094 4095 4096 4097 4098 4099 4100 4101 4102 4103 4104 4105 4106 4107 4108 4109 4110 4111 4112 4113 4114 4115 4116 4117 4118 4119 4120 4121 4122 4123 4124 4125 4126 4127 4128 4129 4130 4131 4132 4133 4134 4135 4136 4137 4138 4139 4140 4141 4142 4143 4144 4145 4146 4147 4148 4149 4150 4151 4152 4153 4154 4155 4156 4157 4158 4159 4160 4161 4162 4163 4164 4165 4166 4167 4168 4169 4170 4171 4172 4173 4174 4175 4176 4177 4178 4179 4180 4181 4182 4183 4184 4185 4186 4187 4188 4189 4190 4191 4192 4193 4194 4195 4196 4197 4198 4199 4200 4201 4202 4203 4204 4205 4206 4207 4208 4209 4210 4211 4212 4213 4214 4215 4216 4217 4218 4219 4220 4221 4222 4223 4224 4225 4226 4227 4228 4229 4230 4231 4232 4233 4234 4235 4236 4237 4238 4239 4240 4241 4242 4243 4244 4245 4246 4247 4248 4249 4250 4251 4252 4253 4254 4255 4256 4257 4258 4259 4260 4261 4262 4263 4264 4265 4266 4267 4268 4269 4270 4271 4272 4273 4274 4275 4276 4277 4278 4279 4280 4281 4282 4283 4284 4285 4286 4287 4288 4289 4290 4291 4292 4293 4294 4295 4296 4297 4298 4299 4300 4301 4302 4303 4304 4305 4306 4307 4308 4309 4310 4311 4312 4313 4314 4315 4316 4317 4318 4319 4320 4321 4322 4323 4324 4325 4326 4327 4328 4329 4330 4331 4332 4333 4334 4335 4336 4337 4338 4339 4340 4341 4342 4343 4344 4345 4346 4347 4348 4349 4350 4351 4352 4353 4354 4355 4356 4357 4358 4359 4360 4361 4362 4363 4364 4365 4366 4367 4368 4369 4370 4371 4372 4373 4374 4375 4376 4377 4378 4379 4380 4381 4382 4383 4384 4385 4386 4387 4388 4389 4390 4391 4392 4393 4394 4395 4396 4397 4398 4399 4400 4401 4402 4403 4404 4405 4406 4407 4408 4409 4410 4411 4412 4413 4414 4415 4416 4417 4418 4419 4420 4421 4422 4423 4424 4425 4426 4427 4428 4429 4430 4431 4432 4433 4434 4435 4436 4437 4438 4439 4440 4441 4442 4443 4444 4445 4446 4447 4448 4449 4450 4451 4452 4453 4454 4455 4456 4457 4458 4459 4460 4461 4462 4463 4464 4465 4466 4467 4468 4469 4470 4471 4472 4473 4474 4475 4476 4477 4478 4479 4480 4481 4482 4483 4484 4485 4486 4487 4488 4489 4490 4491 4492 4493 4494 4495 4496 4497 4498 4499 4500 4501 4502 4503 4504 4505 4506 4507 4508 4509 4510 4511 4512 4513 4514 4515 4516 4517 4518 4519 4520 4521 4522 4523 4524 4525 4526 4527 4528 4529 4530 4531 4532 4533 4534 4535 4536 4537 4538 4539 4540 4541 4542 4543 4544 4545 4546 4547 4548 4549 4550 4551 4552 4553 4554 4555 4556 4557 4558 4559 4560 4561 4562 4563 4564 4565 4566 4567 4568 4569 4570 4571 4572 4573 4574 4575 4576 4577 4578 4579 4580 4581 4582 4583 4584 4585 4586 4587 4588 4589 4590 4591 4592 4593 4594 4595 4596 4597 4598 4599 4600 4601 4602 4603 4604 4605 4606 4607 4608 4609 4610 4611 4612 4613 4614 4615 4616 4617 4618 4619 4620 4621 4622 4623 4624 4625 4626 4627 4628 4629 4630 4631 4632 4633 4634 4635 4636 4637 4638 4639 4640 4641 4642 4643 4644 4645 4646 4647 4648 4649 4650 4651 4652 4653 4654 4655 4656 4657 4658 4659 4660 4661 4662 4663 4664 4665 4666 4667 4668 4669 4670 4671 4672 4673 4674 4675 4676 4677 4678 4679 4680 4681 4682 4683 4684 4685 4686 4687 4688 4689 4690 4691 4692 4693 4694 4695 4696 4697 4698 4699 4700 4701 4702 4703 4704 4705 4706 4707 4708 4709 4710 4711 4712 4713 4714 4715 4716 4717 4718 4719 4720 4721 4722 4723 4724 4725 4726 4727 4728 4729 4730 4731 4732 4733 4734 4735 4736 4737 4738 4739 4740 4741 4742 4743 4744 4745 4746 4747 4748 4749 4750 4751 4752 4753 4754 4755 4756 4757 4758 4759 4760 4761 4762 4763 4764 4765 4766 4767 4768 4769 4770 4771 4772 4773 4774 4775 4776 4777 4778 4779 4780 4781 4782 4783 4784 4785 4786 4787 4788 4789 4790 4791 4792 4793 4794 4795 4796 4797 4798 4799 4800 4801 4802 4803 4804 4805 4806 4807 4808 4809 4810 4811 4812 4813 4814 4815 4816 4817 4818 4819 4820 4821 4822 4823 4824 4825 4826 4827 4828 4829 4830 4831 4832 4833 4834 4835 4836 4837 4838 4839 4840 4841 4842 4843 4844 4845 4846 4847 4848 4849 4850 4851 4852 4853 4854 4855 4856 4857 4858 4859 4860 4861 4862 4863 4864 4865 4866 4867 4868 4869 4870 4871 4872 4873 4874 4875 4876 4877 4878 4879 4880 4881 4882 4883 4884 4885 4886 4887 4888 4889 4890 4891 4892 4893 4894 4895 4896 4897 4898 4899 4900 4901 4902 4903 4904 4905 4906 4907 4908 4909 4910 4911 4912 4913 4914 4915 4916 4917 4918 4919 4920 4921 4922 4923 4924 4925 4926 4927 4928 4929 4930 4931 4932 4933 4934 4935 4936 4937 4938 4939 4940 4941 4942 4943 4944 4945 4946 4947 4948 4949 4950 4951 4952 4953 4954 4955 4956 4957 4958 4959 4960 4961 4962 4963 4964 4965 4966 4967 4968 4969 4970 4971 4972 4973 4974 4975 4976 4977 4978 4979 4980 4981 4982 4983 4984 4985 4986 4987 4988 4989 4990 4991 4992 4993 4994 4995 4996 4997 4998 4999 5000 5001 5002 5003 5004 5005 5006 5007 5008 5009 5010 5011 5012 5013 5014 5015 5016 5017 5018 5019 5020 5021 5022 5023 5024 5025 5026 5027 5028 5029 5030 5031 5032 5033 5034 5035 5036 5037 5038 5039 5040 5041 5042 5043 5044 5045 5046 5047 5048 5049 5050 5051 5052 5053 5054 5055 5056 5057 5058 5059 5060 5061 5062 5063 5064 5065 5066 5067 5068 5069 5070 5071 5072 5073 5074 5075 5076 5077 5078 5079 5080 5081 5082 5083 5084 5085 5086 5087 5088 5089 5090 5091 5092 5093 5094 5095 5096 5097 5098 5099 5100 5101 5102 5103 5104 5105 5106 5107 5108 5109 5110 5111 5112 5113 5114 5115 5116 5117 5118 5119 5120 5121 5122 5123 5124 5125 5126 5127 5128 5129 5130 5131 5132 5133 5134 5135 5136 5137 5138 5139 5140 5141 5142 5143 5144 5145 5146 5147 5148 5149 5150 5151 5152 5153 5154 5155 5156 5157 5158 5159 5160 5161 5162 5163 5164 5165 5166 5167 5168 5169 5170 5171 5172 5173 5174 5175 5176 5177 5178 5179 5180 5181 5182 5183 5184 5185 5186 5187 5188 5189 5190 5191 5192 5193 5194 5195 5196 5197 5198 5199 5200 5201 5202 5203 5204 5205 5206 5207 5208 5209 5210 5211 5212 5213 5214 5215 5216 5217 5218 5219 5220 5221 5222 5223 5224 5225 5226 5227 5228 5229 5230 5231 5232 5233 5234 5235 5236 5237 5238 5239 5240 5241 5242 5243 5244 5245 5246 5247 5248 5249 5250 5251 5252 5253 5254 5255 5256 5257 5258 5259 5260 5261 5262 5263 5264 5265 5266 5267 5268 5269 5270 5271 5272 5273 5274 5275 5276 5277 5278 5279 5280 5281 5282 5283 5284 5285 5286 5287 5288 5289 5290 5291 5292 5293 5294 5295 5296 5297 5298 5299 5300 5301 5302 5303 5304 5305 5306 5307 5308 5309 5310 5311 5312 5313 5314 5315 5316 5317 5318 5319 5320 5321 5322 5323 5324 5325 5326 5327 5328 5329 5330 5331 5332 5333 5334 5335 5336 5337 5338 5339 5340 5341 5342 5343 5344 5345 5346 5347 5348 5349 5350 5351 5352 5353 5354 5355 5356 5357 5358 5359 5360 5361 5362 5363 5364 5365 5366 5367 5368 5369 5370 5371 5372 5373 5374 5375 5376 5377 5378 5379 5380 5381 5382 5383 5384 5385 5386 5387 5388 5389 5390 5391 5392 5393 5394 5395 5396 5397 5398 5399 5400 5401 5402 5403 5404 5405 5406 5407 5408 5409 5410 5411 5412 5413 5414 5415 5416 5417 5418 5419 5420 5421 5422 5423 5424 5425 5426 5427 5428 5429 5430 5431 5432 5433 5434 5435 5436 5437 5438 5439 5440 5441 5442 5443 5444 5445 5446 5447 5448 5449 5450 5451 5452 5453 5454 5455 5456 5457 5458 5459 5460 5461 5462 5463 5464 5465 5466 5467 5468 5469 5470 5471 5472 5473 5474 5475 5476 5477 5478 5479 5480 5481 5482 5483 5484 5485 5486 5487 5488 5489 5490 5491 5492 5493 5494 5495 5496 5497 5498 5499 5500 5501 5502 5503 5504 5505 5506 5507 5508 5509 5510 5511 5512 5513 5514 5515 5516 5517 5518 5519 5520 5521 5522 5523 5524 5525 5526 5527 5528 5529 5530 5531 5532 5533 5534 5535 5536 5537 5538 5539 5540 5541 5542 5543 5544 5545 5546 5547 5548 5549 5550 5551 5552 5553 5554 5555 5556 5557 5558 5559 5560 5561 5562 5563 5564 5565 5566 5567 5568 5569 5570 5571 5572 5573 5574 5575 5576 5577 5578 5579 5580 5581 5582 5583 5584 5585 5586 5587 5588 5589 5590 5591 5592 5593 5594 5595 5596 5597 5598 5599 5600 5601 5602 5603 5604 5605 5606 5607 5608 5609 5610 5611 5612 5613 5614 5615 5616 5617 5618 5619 5620 5621 5622 5623 5624 5625 5626 5627 5628 5629 5630 5631 5632 5633 5634 5635 5636 5637 5638 5639 5640 5641 5642 5643 5644 5645 5646 5647 5648 5649 5650 5651 5652 5653 5654 5655 5656 5657 5658 5659 5660 5661 5662 5663 5664 5665 5666 5667 5668 5669 5670 5671 5672 5673 5674 5675 5676 5677 5678 5679 5680 5681 5682 5683 5684 5685 5686 5687 5688 5689 5690 5691 5692 5693 5694 5695 5696 5697 5698 5699 5700 5701 5702 5703 5704 5705 5706 5707 5708 5709 5710 5711 5712 5713 5714 5715 5716 5717 5718 5719 5720 5721 5722 5723 5724 5725 5726 5727 5728 5729 5730 5731 5732 5733 5734 5735 5736 5737 5738 5739 5740 5741 5742 5743 5744 5745 5746 5747 5748 5749 5750 5751 5752 5753 5754 5755 5756 5757 5758 5759 5760 5761 5762 5763 5764 5765 5766 5767 5768 5769 5770 5771 5772 5773 5774 5775 5776 5777 5778 5779 5780 5781 5782 5783 5784 5785 5786 5787 5788 5789 5790 5791 5792 5793 5794 5795 5796 5797 5798 5799 5800 5801 5802 5803 5804 5805 5806 5807 5808 5809 5810 5811 5812 5813 5814 5815 5816 5817 5818 5819 5820 5821 5822 5823 5824 5825 5826 5827 5828 5829 5830 5831 5832 5833 5834 5835 5836 5837 5838 5839 5840 5841 5842 5843 5844 5845 5846 5847 5848 5849 5850 5851 5852 5853 5854 5855 5856 5857 5858 5859 5860 5861 5862 5863 5864 5865 5866 5867 5868 5869 5870 5871 5872 5873 5874 5875 5876 5877 5878 5879 5880 5881 5882 5883 5884 5885 5886 5887 5888 5889 5890 5891 5892 5893 5894 5895 5896 5897 5898 5899 5900 5901 5902 5903 5904 5905 5906 5907 5908 5909 5910 5911 5912 5913 5914 5915 5916 5917 5918 5919 5920 5921 5922 5923 5924 5925 5926 5927 5928 5929 5930 5931 5932 5933 5934 5935 5936 5937 5938 5939 5940 5941 5942 5943 5944 5945 5946 5947 5948 5949 5950 5951 5952 5953 5954 5955 5956 5957 5958 5959 5960 5961 5962 5963 5964 5965 5966 5967 5968 5969 5970 5971 5972 5973 5974 5975 5976 5977 5978 5979 5980 5981 5982 5983 5984 5985 5986 5987 5988 5989 5990 5991 5992 5993 5994 5995 5996 5997 5998 5999 6000 6001 6002 6003 6004 6005 6006 6007 6008 6009 6010 6011 6012 6013 6014 6015 6016 6017 6018 6019 6020 6021 6022 6023 6024 6025 6026 6027 6028 6029 6030 6031 6032 6033 6034 6035 6036 6037 6038 6039 6040 6041 6042 6043 6044 6045 6046 6047 6048 6049 6050 6051 6052 6053 6054 6055 6056 6057 6058 6059 6060 6061 6062 6063 6064 6065 6066 6067 6068 6069 6070 6071 6072 6073 6074 6075 6076 6077 6078 6079 6080 6081 6082 6083 6084 6085 6086 6087 6088 6089 6090 6091 6092 6093 6094 6095 6096 6097 6098 6099 6100 6101 6102 6103 6104 6105 6106 6107 6108 6109 6110 6111 6112 6113 6114 6115 6116 6117 6118 6119 6120 6121 6122 6123 6124 6125 6126 6127 6128 6129 6130 6131 6132 6133 6134 6135 6136 6137 6138 6139 6140 6141 6142 6143 6144 6145 6146 6147 6148 6149 6150 6151 6152 6153 6154 6155 6156 6157 6158 6159 6160 6161 6162 6163 6164 6165 6166 6167 6168 6169 6170 6171 6172 6173 6174 6175 6176 6177 6178 6179 6180 6181 6182 6183 6184 6185 6186 6187 6188 6189 6190 6191 6192 6193 6194 6195 6196 6197 6198 6199 6200 6201 6202 6203 6204 6205 6206 6207 6208 6209 6210 6211 6212 6213 6214 6215 6216 6217 6218 6219 6220 6221 6222 6223 6224 6225 6226 6227 6228 6229 6230 6231 6232 6233 6234 6235 6236 6237 6238 6239 6240 6241 6242 6243 6244 6245 6246 6247 6248 6249 6250 6251 6252 6253 6254 6255 6256 6257 6258 6259 6260 6261 6262 6263 6264 6265 6266 6267 6268 6269 6270 6271 6272 6273 6274 6275 6276 6277 6278 6279 6280 6281 6282 6283 6284 6285 6286 6287 6288 6289 6290 6291 6292 6293 6294 6295 6296 6297 6298 6299 6300 6301 6302 6303 6304 6305 6306 6307 6308 6309 6310 6311 6312 6313 6314 6315 6316 6317 6318 6319 6320 6321 6322 6323 6324 6325 6326 6327 6328 6329 6330 6331 6332 6333 6334 6335 6336 6337 6338 6339 6340 6341 6342 6343 6344 6345 6346 6347 6348 6349 6350 6351 6352 6353 6354 6355 6356 6357 6358 6359 6360 6361 6362 6363 6364 6365 6366 6367 6368 6369 6370 6371 6372 6373 6374 6375 6376 6377 6378 6379 6380 6381 6382 6383 6384 6385 6386 6387 6388 6389 6390 6391 6392 6393 6394 6395 6396 6397 6398 6399 6400 6401 6402 6403 6404 6405 6406 6407 6408 6409 6410 6411 6412 6413 6414 6415 6416 6417 6418 6419 6420 6421 6422 6423 6424 6425 6426 6427 6428 6429 6430 6431 6432 6433 6434 6435 6436 6437 6438 6439 6440 6441 6442 6443 6444 6445 6446 6447 6448 6449 6450 6451 6452 6453 6454 6455 6456 6457 6458 6459 6460 6461 6462 6463 6464 6465 6466 6467 6468 6469 6470 6471 6472 6473 6474 6475 6476 6477 6478 6479 6480 6481 6482 6483 6484 6485 6486 6487 6488 6489 6490 6491 6492 6493 6494 6495 6496 6497 6498 6499 6500 6501 6502 6503 6504 6505 6506 6507 6508 6509 6510 6511 6512 6513 6514 6515 6516 6517 6518 6519 6520 6521 6522 6523 6524 6525 6526 6527 6528 6529 6530 6531 6532 6533 6534 6535 6536 6537 6538 6539 6540 6541 6542 6543 6544 6545 6546 6547 6548 6549 6550 6551 6552 6553 6554 6555 6556 6557 6558 6559 6560 6561 6562 6563 6564 6565 6566 6567 6568 6569 6570 6571 6572 6573 6574 6575 6576 6577 6578 6579 6580 6581 6582 6583 6584 6585 6586 6587 6588 6589 6590 6591 6592 6593 6594 6595 6596 6597 6598 6599 6600 6601 6602 6603 6604 6605 6606 6607 6608 6609 6610 6611 6612 6613 6614 6615 6616 6617 6618 6619 6620 6621 6622 6623 6624 6625 6626 6627 6628 6629 6630 6631 6632 6633 6634 6635 6636 6637 6638 6639 6640 6641 6642 6643 6644 6645 6646 6647 6648 6649 6650 6651 6652 6653 6654 6655 6656 6657 6658 6659 6660 6661 6662 6663 6664 6665 6666 6667 6668 6669 6670 6671 6672 6673 6674 6675 6676 6677 6678 6679 6680 6681 6682 6683 6684 6685 6686 6687 6688 6689 6690 6691 6692 6693 6694 6695 6696 6697 6698 6699 6700 6701 6702 6703 6704 6705 6706 6707 6708 6709 6710 6711 6712 6713 6714 6715 6716 6717 6718 6719 6720 6721 6722 6723 6724 6725 6726 6727 6728 6729 6730 6731 6732 6733 6734 6735 6736 6737 6738 6739 6740 6741 6742 6743 6744 6745 6746 6747 6748 6749 6750 6751 6752 6753 6754 6755 6756 6757 6758 6759 6760 6761 6762 6763 6764 6765 6766 6767 6768 6769 6770 6771 6772 6773 6774 6775 6776 6777 6778 6779 6780 6781 6782 6783 6784 6785 6786 6787 6788 6789 6790 6791 6792 6793 6794 6795 6796 6797 6798 6799 6800 6801 6802 6803 6804 6805 6806 6807 6808 6809 6810 6811 6812 6813 6814 6815 6816 6817 6818 6819 6820 6821 6822 6823 6824 6825 6826 6827 6828 6829 6830 6831 6832 6833 6834 6835 6836 6837 6838 6839 6840 6841 6842 6843 6844 6845 6846 6847 6848 6849 6850 6851 6852 6853 6854 6855 6856 6857 6858 6859 6860 6861 6862 6863 6864 6865 6866 6867 6868 6869 6870 6871 6872 6873 6874 6875 6876 6877 6878 6879 6880 6881 6882 6883 6884 6885 6886 6887 6888 6889 6890 6891 6892 6893 6894 6895 6896 6897 6898 6899 6900 6901 6902 6903 6904 6905 6906 6907 6908 6909 6910 6911 6912 6913 6914 6915 6916 6917 6918 6919 6920 6921 6922 6923 6924 6925 6926 6927 6928 6929 6930 6931 6932 6933 6934 6935 6936 6937 6938 6939 6940 6941 6942 6943 6944 6945 6946 6947 6948 6949 6950 6951 6952 6953 6954 6955 6956 6957 6958 6959 6960 6961 6962 6963 6964 6965 6966 6967 6968 6969 6970 6971 6972 6973 6974 6975 6976 6977 6978 6979 6980 6981 6982 6983 6984 6985 6986 6987 6988 6989 6990 6991 6992 6993 6994 6995 6996 6997 6998 6999 7000 7001 7002 7003 7004 7005 7006 7007 7008 7009 7010 7011 7012 7013 7014 7015 7016 7017 7018 7019 7020 7021 7022 7023 7024 7025 7026 7027 7028 7029 7030 7031 7032 7033 7034 7035 7036 7037 7038 7039 7040 7041 7042 7043 7044 7045 7046 7047 7048 7049 7050 7051 7052 7053 7054 7055 7056 7057 7058 7059 7060 7061 7062 7063 7064 7065 7066 7067 7068 7069 7070 7071 7072 7073 7074 7075 7076 7077 7078 7079 7080 7081 7082 7083 7084 7085 7086 7087 7088 7089 7090 7091 7092 7093 7094 7095 7096 7097 7098 7099 7100 7101 7102 7103 7104 7105 7106 7107 7108 7109 7110 7111 7112 7113 7114 7115 7116 7117 7118 7119 7120 7121 7122 7123 7124 7125 7126 7127 7128 7129 7130 7131 7132 7133 7134 7135 7136 7137 7138 7139 7140 7141 7142 7143 7144 7145 7146 7147 7148 7149 7150 7151 7152 7153 7154 7155 7156 7157 7158 7159 7160 7161 7162 7163 7164 7165 7166 7167 7168 7169 7170 7171 7172 7173 7174 7175 7176 7177 7178 7179 7180 7181 7182 7183 7184 7185 7186 7187 7188 7189 7190 7191 7192 7193 7194 7195 7196 7197 7198 7199 7200 7201 7202 7203 7204 7205 7206 7207 7208 7209 7210 7211 7212 7213 7214 7215 7216 7217 7218 7219 7220 7221 7222 7223 7224 7225 7226 7227 7228 7229 7230 7231 7232 7233 7234 7235 7236 7237 7238 7239 7240 7241 7242 7243 7244 7245 7246 7247 7248 7249 7250 7251 7252 7253 7254 7255 7256 7257 7258 7259 7260 7261 7262 7263 7264 7265 7266 7267 7268 7269 7270 7271 7272 7273 7274 7275 7276 7277 7278 7279 7280 7281 7282 7283 7284 7285 7286 7287 7288 7289 7290 7291 7292 7293 7294 7295 7296 7297 7298 7299 7300 7301 7302 7303 7304 7305 7306 7307 7308 7309 7310 7311 7312 7313 7314 7315 7316 7317 7318 7319 7320 7321 7322 7323 7324 7325 7326 7327 7328 7329 7330 7331 7332 7333 7334 7335 7336 7337 7338 7339 7340 7341 7342 7343 7344 7345 7346 7347 7348 7349 7350 7351 7352 7353 7354 7355 7356 7357 7358 7359 7360 7361 7362 7363 7364 7365 7366 7367 7368 7369 7370 7371 7372 7373 7374 7375 7376 7377 7378 7379 7380 7381 7382 7383 7384 7385 7386 7387 7388 7389 7390 7391 7392 7393 7394 7395 7396 7397 7398 7399 7400 7401 7402 7403 7404 7405 7406 7407 7408 7409 7410 7411 7412 7413 7414 7415 7416 7417 7418 7419 7420 7421 7422 7423 7424 7425 7426 7427 7428 7429 7430 7431 7432 7433 7434 7435 7436 7437 7438 7439 7440 7441 7442 7443 7444 7445 7446 7447 7448 7449 7450 7451 7452 7453 7454 7455 7456 7457 7458 7459 7460 7461 7462 7463 7464 7465 7466 7467 7468 7469 7470 7471 7472 7473 7474 7475 7476 7477 7478 7479 7480 7481 7482 7483 7484 7485 7486 7487 7488 7489 7490 7491 7492 7493 7494 7495 7496 7497 7498 7499 7500 7501 7502 7503 7504 7505 7506 7507 7508 7509 7510 7511 7512 7513 7514 7515 7516 7517 7518 7519 7520 7521 7522 7523 7524 7525 7526 7527 7528 7529 7530 7531 7532 7533 7534 7535 7536 7537 7538 7539 7540 7541 7542 7543 7544 7545 7546 7547 7548 7549 7550 7551 7552 7553 7554 7555 7556 7557 7558 7559 7560 7561 7562 7563 7564 7565 7566 7567 7568 7569 7570 7571 7572 7573 7574 7575 7576 7577 7578 7579 7580 7581 7582 7583 7584 7585 7586 7587 7588 7589 7590 7591 7592 7593 7594 7595 7596 7597 7598 7599 7600 7601 7602 7603 7604 7605 7606 7607 7608 7609 7610 7611 7612 7613 7614 7615 7616 7617 7618 7619 7620 7621 7622 7623 7624 7625 7626 7627 7628 7629 7630 7631 7632 7633 7634 7635 7636 7637 7638 7639 7640 7641 7642 7643 7644 7645 7646 7647 7648 7649 7650 7651 7652 7653 7654 7655 7656 7657 7658 7659 7660 7661 7662 7663 7664 7665 7666 7667 7668 7669 7670 7671 7672 7673 7674 7675 7676 7677 7678 7679 7680 7681 7682 7683 7684 7685 7686 7687 7688 7689 7690 7691 7692 7693 7694 7695 7696 7697 7698 7699 7700 7701 7702 7703 7704 7705 7706 7707 7708 7709 7710 7711 7712 7713 7714 7715 7716 7717 7718 7719 7720 7721 7722 7723 7724 7725 7726 7727 7728 7729 7730 7731 7732 7733 7734 7735 7736 7737 7738 7739 7740 7741 7742 7743 7744 7745 7746 7747 7748 7749 7750 7751 7752 7753 7754 7755 7756 7757 7758 7759 7760 7761 7762 7763 7764 7765 7766 7767 7768 7769 7770 7771 7772 7773 7774 7775 7776 7777 7778 7779 7780 7781 7782 7783 7784 7785 7786 7787 7788 7789 7790 7791 7792 7793 7794 7795 7796 7797 7798 7799 7800 7801 7802 7803 7804 7805 7806 7807 7808 7809 7810 7811 7812 7813 7814 7815 7816 7817 7818 7819 7820 7821 7822 7823 7824 7825 7826 7827 7828 7829 7830 7831 7832 7833 7834 7835 7836 7837 7838 7839 7840 7841 7842 7843 7844 7845 7846 7847 7848 7849 7850 7851 7852 7853 7854 7855 7856 7857 7858 7859 7860 7861 7862 7863 7864 7865 7866 7867 7868 7869 7870 7871 7872 7873 7874 7875 7876 7877 7878 7879 7880 7881 7882 7883 7884 7885 7886 7887 7888 7889 7890 7891 7892 7893 7894 7895 7896 7897 7898 7899 7900 7901 7902 7903 7904 7905 7906 7907 7908 7909 7910 7911 7912 7913 7914 7915 7916 7917 7918 7919 7920 7921 7922 7923 7924 7925 7926 7927 7928 7929 7930 7931 7932 7933 7934 7935 7936 7937
|
/* CFString.c
Copyright (c) 1998-2019, Apple Inc. and the Swift project authors
Portions Copyright (c) 2014-2019, Apple Inc. and the Swift project authors
Licensed under Apache License v2.0 with Runtime Library Exception
See http://swift.org/LICENSE.txt for license information
See http://swift.org/CONTRIBUTORS.txt for the list of Swift project authors
Responsibility: Ali Ozer
!!! For performance reasons, it's important that all functions marked CF_INLINE in this file are inlined.
*/
#include "CFBase.h"
#include "CFString.h"
#include "CFDictionary.h"
#include "CFStringEncodingConverterExt.h"
#include "CFStringEncodingConverterPriv.h"
#include "CFUniChar.h"
#include "CFUnicodeDecomposition.h"
#include "CFUnicodePrecomposition.h"
#include "CFPriv.h"
#include "CFNumber.h"
#include "CFNumberFormatter.h"
#include "CFError_Private.h"
#include "CFInternal.h"
#include "CFUniCharPriv.h"
#include "CFString_Internal.h"
#include "CFRuntime_Internal.h"
#include <assert.h>
#include <_foundation_unicode/uchar.h>
#if TARGET_OS_MAC || TARGET_OS_WIN32 || TARGET_OS_LINUX || TARGET_OS_BSD || TARGET_OS_WASI
#include "CFConstantKeys.h"
#include "CFStringLocalizedFormattingInternal.h"
#endif
#include <stdarg.h>
#include <stdio.h>
#include <string.h>
#if TARGET_OS_MAC || TARGET_OS_LINUX || TARGET_OS_BSD
#include <unistd.h>
#endif
#if TARGET_OS_WASI
#include <sys/types.h> // for u_char
#endif
#if defined(__GNUC__)
#define LONG_DOUBLE_SUPPORT 1
#else
#define LONG_DOUBLE_SUPPORT 0
#endif
CONST_STRING_DECL(_kCFStringFormatMetadataReplacementIndexKey, "Index");
CONST_STRING_DECL(_kCFStringFormatMetadataSpecifierRangeLocationInFormatStringKey, "SpecLocation");
CONST_STRING_DECL(_kCFStringFormatMetadataSpecifierRangeLengthInFormatStringKey, "SpecLength");
CONST_STRING_DECL(_kCFStringFormatMetadataReplacementRangeLocationKey, "ReplacementLocation");
CONST_STRING_DECL(_kCFStringFormatMetadataReplacementRangeLengthKey, "ReplacementLength");
CONST_STRING_DECL(_kCFStringFormatMetadataArgumentObjectKey, "Object");
CONST_STRING_DECL(_kCFStringFormatMetadataArgumentNumberKey, "Number");
#define USE_STRING_ROM 0
#ifndef INSTRUMENT_SHARED_STRINGS
#define INSTRUMENT_SHARED_STRINGS 0
#endif
#if INSTRUMENT_SHARED_STRINGS
#include <sys/stat.h> /* for umask() */
#include <unistd.h> /* for confstr */
#include <crt_externs.h>
static void __CFRecordStringAllocationEvent(const char *encoding, const char *bytes, CFIndex byteCount) {
static CFLock_t lock = CFLockInit;
if (memchr(bytes, '\n', byteCount)) return; //never record string allocation events for strings with newlines, because those confuse our parser and because they'll never go into the ROM
__CFLock(&lock);
static int fd = -1;
static dispatch_once_t pred;
dispatch_once(&pred, ^{
const char *name = *_NSGetProgname();
if (! name) name = "UNKNOWN";
umask(0);
char temp_dir[PATH_MAX];
char path[PATH_MAX];
size_t temp_dir_len = confstr(_CS_DARWIN_USER_TEMP_DIR, temp_dir, PATH_MAX);
if (temp_dir_len == 0){
fprintf(stderr, "confstr failed due to %s", strerror(errno));
} else {
snprintf(path, sizeof(path), "%s/CFSharedStringInstrumentation_%s_%d.txt", temp_dir, name, getpid());
fd = open(path, O_WRONLY | O_APPEND | O_CREAT, 0666);
if (fd <= 0) {
int error = errno;
const char *errString = strerror(error);
fprintf(stderr, "open() failed with error %d (%s)\n", error, errString);
}
}
});
if (fd > 0) {
char *buffer = NULL;
char formatString[256];
snprintf(formatString, sizeof(formatString), "%%-8d\t%%-16s\t%%.%lds\n", byteCount);
int resultCount = asprintf(&buffer, formatString, getpid(), encoding, bytes);
if (buffer && resultCount > 0) write(fd, buffer, resultCount);
else puts("Couldn't record allocation event");
free(buffer);
}
__CFUnlock(&lock);
}
#endif //INSTRUMENT_SHARED_STRINGS
#if TARGET_OS_MAC
extern size_t malloc_good_size(size_t size);
#endif
extern void __CFStrConvertBytesToUnicode(const uint8_t *bytes, UniChar *buffer, CFIndex numChars);
CF_PRIVATE uint32_t CFUniCharGetConditionalCaseMappingFlags(UTF32Char theChar, UTF16Char *buffer, CFIndex currentIndex, CFIndex length, uint32_t type, const uint8_t *langCode, uint32_t lastFlags);
static Boolean __CFStringAppendFormatCore(CFMutableStringRef outputString, CFStringRef (*copyDescFunc)(void *, const void *), CFStringRef (*contextDescFunc)(void *, const void *, const void *, bool, bool *), CFDictionaryRef formatOptions, CFDictionaryRef stringsDictConfig, CFStringRef validFormatSpecifiers, CFStringRef formatString, CFIndex initialArgPosition, const void *origValues, CFIndex originalValuesSize, va_list args, CFArrayRef *outReplacementMetadata, CFErrorRef *errorPtr);
static inline const char * _CFStringGetCStringPtrInternal(CFStringRef str, CFStringEncoding encoding, Boolean requiresNullTermination, Boolean requiresBridgingCheck);
CF_INLINE void _CFStringInitInlineBufferInternal(CFStringRef str, CFStringInlineBuffer *buf, CFRange range, Boolean requiresBridgingCheck) {
buf->theString = str;
buf->rangeToBuffer = range;
buf->directCStringBuffer = (buf->directUniCharBuffer = CFStringGetCharactersPtr(str)) ? NULL : _CFStringGetCStringPtrInternal(str, kCFStringEncodingASCII, false, requiresBridgingCheck);
buf->bufferedRangeStart = buf->bufferedRangeEnd = 0;
}
#if defined(DEBUG)
// We put this into C & Pascal strings if we can't convert
#define CONVERSIONFAILURESTR "CFString conversion failed"
// We set this to true when purging the constant string table, so CFStringDeallocate doesn't assert
static Boolean __CFConstantStringTableBeingFreed = false;
#endif
// Two constant strings used by CFString; these are initialized in CFStringInitialize
CONST_STRING_DECL(kCFEmptyString, "")
// This is separate for C++
struct __notInlineMutable {
void *buffer;
CFIndex length;
CFIndex capacity; // Capacity in bytes
unsigned int hasGap:1; // Currently unused
unsigned int isFixedCapacity:1;
unsigned int isExternalMutable:1;
unsigned int capacityProvidedExternally:1;
#if __LP64__
unsigned long desiredCapacity:60;
#elif __LLP64__
unsigned long long desiredCapacity:60;
#else
unsigned long desiredCapacity:28;
#endif
CFAllocatorRef contentsAllocator; // Optional
}; // The only mutable variant for CFString
/* !!! Never do sizeof(CFString); the union is here just to make it easier to access some fields.
*/
struct __attribute__((__aligned__(8))) __CFString {
CFRuntimeBase base;
union { // In many cases the allocated structs are smaller than these
struct __inline1 {
CFIndex length;
} inline1; // Bytes follow the length
struct __notInlineImmutable1 {
void *buffer; // Note that the buffer is in the same place for all non-inline variants of CFString
CFIndex length;
CFAllocatorRef contentsDeallocator; // Optional; just the dealloc func is used
} notInlineImmutable1; // This is the usual not-inline immutable CFString
struct __notInlineImmutable2 {
void *buffer;
CFAllocatorRef contentsDeallocator; // Optional; just the dealloc func is used
} notInlineImmutable2; // This is the not-inline immutable CFString when length is stored with the contents (first byte)
struct __notInlineMutable notInlineMutable;
} variants;
};
/*
I = is immutable
E = not inline contents
U = is Unicode
N = has NULL byte
L = has length byte
D = explicit deallocator for contents (for mutable objects, allocator)
C = length field is CFIndex (rather than UInt32); only meaningful for 64-bit, really
if needed this bit (valuable real-estate) can be given up for another bit elsewhere, since this info is needed just for 64-bit
Also need (only for mutable)
F = is fixed
G = has gap
Cap, DesCap = capacity
B7 B6 B5 B4 B3 B2 B1 B0
U N L C I
B6 B5
0 0 inline contents
0 1 E (freed with default allocator)
1 0 E (not freed)
1 1 E D
!!! Note: Constant CFStrings use the bit patterns:
C8 (11001000 = default allocator, not inline, not freed contents; 8-bit; has NULL byte; doesn't have length; is immutable)
D0 (11010000 = default allocator, not inline, not freed contents; Unicode; is immutable)
The bit usages should not be modified in a way that would effect these bit patterns.
Note that some of the bit patterns in the enum below overlap and are duplicated. Keep this in mind as you do searches for use cases.
*/
enum {
// These are bit numbers - do not use them as masks
__kCFIsMutable = 0,
// !!! Bit 1 has been freed up
__kCFHasLengthByte = 2,
__kCFHasNullByte = 3,
__kCFIsUnicode = 4,
};
typedef enum {
// These are values in bit numbers 5 & 6
__kCFHasInlineContents = 0,
__kCFNotInlineContentsDefaultFree = 1, // Use allocator's free function
__kCFNotInlineContentsNoFree = 2, // Don't free
__kCFNotInlineContentsCustomFree = 3, // Use a specially provided free function
} _CFStringInlineContents;
CF_INLINE void __CFStrSetInlineContents(CFStringRef str, _CFStringInlineContents contents) {__CFRuntimeSetValue(str, 6, 5, contents);}
CF_INLINE Boolean __CFStrIsInline(CFStringRef str) {return __CFRuntimeGetValue(str, 6, 5) == __kCFHasInlineContents;}
CF_INLINE Boolean __CFStrFreeContentsWhenDone(CFStringRef str) {
// Contents of this flag are shared with the inline contents field
return __CFRuntimeGetFlag(str, 5);
}
CF_INLINE Boolean __CFStrHasContentsDeallocator(CFStringRef str) {return __CFRuntimeGetValue(str, 6, 5) == __kCFNotInlineContentsCustomFree;}
CF_INLINE Boolean __CFStrHasContentsAllocator(CFStringRef str) {return __CFRuntimeGetValue(str, 6, 5) == __kCFNotInlineContentsCustomFree;}
// !!! Assumptions:
// Mutable strings are not inline
// Compile-time constant strings are not inline
// Mutable strings always have explicit length (but they might also have length byte and null byte)
// If there is an explicit length, always use that instead of the length byte (length byte is useful for quickly returning pascal strings)
// Never look at the length byte for the length; use __CFStrLength or __CFStrLength2
/* The following set of functions and macros need to be updated on change to the bit configuration
*/
CF_INLINE Boolean __CFStrIsMutable(CFStringRef str) {return __CFRuntimeGetFlag(str, __kCFIsMutable);}
CF_INLINE Boolean __CFStrIsUnicode(CFStringRef str) {return __CFRuntimeGetFlag(str, __kCFIsUnicode);}
CF_INLINE Boolean __CFStrIsEightBit(CFStringRef str) {return !__CFRuntimeGetFlag(str, __kCFIsUnicode);}
CF_INLINE Boolean __CFStrHasNullByte(CFStringRef str) {return __CFRuntimeGetFlag(str, __kCFHasNullByte);}
CF_INLINE Boolean __CFStrHasLengthByte(CFStringRef str) {return __CFRuntimeGetFlag(str, __kCFHasLengthByte);}
CF_INLINE Boolean __CFStrHasExplicitLength(CFStringRef str) {
// Has explicit length if (1) mutable or (2) not mutable and no length byte
const uint8_t isMutableMask = 1 | 4; // is_mutable_mask | has_length_byte_mask
const uint8_t hasLengthByteMask = 4; // has_length_byte_mask
return (__CFRuntimeGetValue(str, 2, 0) & isMutableMask) != hasLengthByteMask;
}
CF_INLINE void __CFStrSetIsMutable(CFStringRef str) {__CFRuntimeSetFlag(str, __kCFIsMutable, true);}
CF_INLINE void __CFStrSetHasNullByte(CFStringRef str, Boolean flag) {__CFRuntimeSetFlag(str, __kCFHasNullByte, flag);}
CF_INLINE void __CFStrSetHasLengthByte(CFStringRef str, Boolean flag) {__CFRuntimeSetFlag(str, __kCFHasLengthByte, flag);}
CF_INLINE void __CFStrSetUnicode(CFMutableStringRef str, Boolean flag) {__CFRuntimeSetFlag(str, __kCFIsUnicode, flag);}
CF_INLINE void __CFStrSetHasLengthAndNullBytes(CFMutableStringRef str) {
__CFStrSetHasLengthByte(str, true);
__CFStrSetHasNullByte(str, true);
}
CF_INLINE void __CFStrClearHasLengthAndNullBytes(CFMutableStringRef str) {
__CFStrSetHasLengthByte(str, false);
__CFStrSetHasNullByte(str, false);
}
CF_INLINE Boolean __CFStrIsConstant(CFStringRef str) {
#if DEPLOYMENT_RUNTIME_SWIFT
return str->base._swift_rc & _CF_SWIFT_RC_PINNED_FLAG;
#else
return __CFRuntimeIsConstant(str);
#endif
}
CF_INLINE SInt32 __CFStrSkipAnyLengthByte(CFStringRef str) {return __CFRuntimeGetFlag(str, __kCFHasLengthByte) ? 1 : 0;} // Number of bytes to skip over the length byte in the contents
/* Returns ptr to the buffer (which might include the length byte).
*/
CF_INLINE const void * _Nullable __CFStrContents(CFStringRef str) {
if (__CFStrIsInline(str)) {
return (const void *)(((uintptr_t)&(str->variants)) + (__CFStrHasExplicitLength(str) ? sizeof(CFIndex) : 0));
} else { // Not inline; pointer is always word 2
return str->variants.notInlineImmutable1.buffer;
}
}
static CFAllocatorRef *__CFStrContentsDeallocatorPtr(CFStringRef str) {
return __CFStrHasExplicitLength(str) ? &(((CFMutableStringRef)str)->variants.notInlineImmutable1.contentsDeallocator) : &(((CFMutableStringRef)str)->variants.notInlineImmutable2.contentsDeallocator); }
// Assumption: Called with immutable strings only, and on strings that are known to have a contentsDeallocator
CF_INLINE CFAllocatorRef __CFStrContentsDeallocator(CFStringRef str) {
return *__CFStrContentsDeallocatorPtr(str);
}
// Assumption: Called with immutable strings only, and on strings that are known to have a contentsDeallocator
CF_INLINE void __CFStrSetContentsDeallocator(CFStringRef str, CFAllocatorRef allocator) {
CFRetain(allocator);
*__CFStrContentsDeallocatorPtr(str) = allocator;
}
static CFAllocatorRef *__CFStrContentsAllocatorPtr(CFStringRef str) {
CFAssert(!__CFStrIsInline(str), __kCFLogAssertion, "Asking for contents allocator of inline string");
CFAssert(__CFStrIsMutable(str), __kCFLogAssertion, "Asking for contents allocator of an immutable string");
return (CFAllocatorRef *)&(str->variants.notInlineMutable.contentsAllocator);
}
// Assumption: Called with strings that have a contents allocator; also, contents allocator follows custom
CF_INLINE CFAllocatorRef __CFStrContentsAllocator(CFMutableStringRef str) {
return *(__CFStrContentsAllocatorPtr(str));
}
// Assumption: Called with strings that have a contents allocator; also, contents allocator follows custom
CF_INLINE void __CFStrSetContentsAllocator(CFMutableStringRef str, CFAllocatorRef allocator) {
CFRetain(allocator);
*(__CFStrContentsAllocatorPtr(str)) = allocator;
}
/* Returns length; use __CFStrLength2 if contents buffer pointer has already been computed.
*/
CF_INLINE CFIndex __CFStrLength(CFStringRef str) {
if (__CFStrHasExplicitLength(str)) {
if (__CFStrIsInline(str)) {
return str->variants.inline1.length;
} else {
return str->variants.notInlineImmutable1.length;
}
} else {
return (CFIndex)(*((uint8_t *)__CFStrContents(str)));
}
}
CF_INLINE CFIndex __CFStrLength2(CFStringRef str, const void *buffer) {
if (__CFStrHasExplicitLength(str)) {
if (__CFStrIsInline(str)) {
return str->variants.inline1.length;
} else {
return str->variants.notInlineImmutable1.length;
}
} else {
return (CFIndex)(*((uint8_t *)buffer));
}
}
Boolean __CFStringIsEightBit(CFStringRef str) {
return __CFStrIsEightBit(str);
}
/* Sets the content pointer for immutable or mutable strings.
*/
CF_INLINE void __CFStrSetContentPtr(CFStringRef str, const void *p) {
// XXX_PCB catch all writes for mutable string case.
*((void **)&((CFMutableStringRef)str)->variants.notInlineImmutable1.buffer) = (void *)p;
}
CF_INLINE void __CFStrSetExplicitLength(CFStringRef str, CFIndex v) {
if (__CFStrIsInline(str)) {
((CFMutableStringRef)str)->variants.inline1.length = v;
} else {
((CFMutableStringRef)str)->variants.notInlineImmutable1.length = v;
}
}
// Assumption: The following set of inlines (using str->variants.notInlineMutable) are called with mutable strings only
CF_INLINE Boolean __CFStrIsFixed(CFStringRef str) {return str->variants.notInlineMutable.isFixedCapacity;}
CF_INLINE Boolean __CFStrIsExternalMutable(CFStringRef str) {return str->variants.notInlineMutable.isExternalMutable;}
CF_INLINE void __CFStrSetIsFixed(CFMutableStringRef str) {str->variants.notInlineMutable.isFixedCapacity = 1;}
CF_INLINE void __CFStrSetIsExternalMutable(CFMutableStringRef str) {str->variants.notInlineMutable.isExternalMutable = 1;}
//CF_INLINE void __CFStrSetHasGap(CFMutableStringRef str) {str->variants.notInlineMutable.hasGap = 1;} currently unused
// If capacity is provided externally, we only change it when we need to grow beyond it
CF_INLINE Boolean __CFStrCapacityProvidedExternally(CFStringRef str) {return str->variants.notInlineMutable.capacityProvidedExternally;}
CF_INLINE void __CFStrSetCapacityProvidedExternally(CFMutableStringRef str) {str->variants.notInlineMutable.capacityProvidedExternally = 1;}
CF_INLINE void __CFStrClearCapacityProvidedExternally(CFMutableStringRef str) {str->variants.notInlineMutable.capacityProvidedExternally = 0;}
// "Capacity" is stored in number of bytes, not characters. It indicates the total number of bytes in the contents buffer.
CF_INLINE CFIndex __CFStrCapacity(CFStringRef str) {return str->variants.notInlineMutable.capacity;}
CF_INLINE void __CFStrSetCapacity(CFMutableStringRef str, CFIndex cap) {str->variants.notInlineMutable.capacity = cap;}
// "Desired capacity" is in number of characters; it is the client requested capacity; if fixed, it is the upper bound on the mutable string backing store.
CF_INLINE CFIndex __CFStrDesiredCapacity(CFStringRef str) {return str->variants.notInlineMutable.desiredCapacity;}
CF_INLINE void __CFStrSetDesiredCapacity(CFMutableStringRef str, CFIndex size) {str->variants.notInlineMutable.desiredCapacity = size;}
static void *__CFStrAllocateMutableContents(CFMutableStringRef str, CFIndex size) {
void *ptr;
CFAllocatorRef alloc = (__CFStrHasContentsAllocator(str)) ? __CFStrContentsAllocator(str) : __CFGetAllocator(str);
ptr = CFAllocatorAllocate(alloc, size, 0);
if (__CFOASafe) __CFSetLastAllocationEventName(ptr, "CFString (store)");
return ptr;
}
static void __CFStrDeallocateMutableContents(CFMutableStringRef str, void *buffer) {
CFAllocatorRef alloc = (__CFStrHasContentsAllocator(str)) ? __CFStrContentsAllocator(str) : __CFGetAllocator(str);
if (__CFStrIsMutable(str) && __CFStrHasContentsAllocator(str) && (0)) {
// do nothing
} else {
CFAllocatorDeallocate(alloc, buffer);
}
}
/* CFString specific init flags
Note that you cannot count on the external buffer not being copied.
Also, if you specify an external buffer, you should not change it behind the CFString's back.
*/
enum {
__kCFThinUnicodeIfPossible = 0x1000000, /* See if the Unicode contents can be thinned down to 8-bit */
kCFStringPascal = 0x10000, /* Indicating that the string data has a Pascal string structure (length byte at start) */
kCFStringNoCopyProvidedContents = 0x20000, /* Don't copy the provided string contents if possible; free it when no longer needed */
kCFStringNoCopyNoFreeProvidedContents = 0x30000 /* Don't copy the provided string contents if possible; don't free it when no longer needed */
};
/* System Encoding.
*/
static CFStringEncoding __CFDefaultSystemEncoding = kCFStringEncodingInvalidId;
static CFStringEncoding __CFDefaultFileSystemEncoding = kCFStringEncodingInvalidId;
CFStringEncoding __CFDefaultEightBitStringEncoding = kCFStringEncodingInvalidId;
#if TARGET_OS_MAC
#define __defaultEncoding kCFStringEncodingMacRoman
#elif TARGET_OS_LINUX || TARGET_OS_WASI
#define __defaultEncoding kCFStringEncodingUTF8
#elif TARGET_OS_WIN32
#define __defaultEncoding kCFStringEncodingWindowsLatin1
#else
#warning This value must match __CFGetConverter condition in CFStringEncodingConverter.c
#define __defaultEncoding kCFStringEncodingISOLatin1
#endif
CFStringEncoding CFStringGetSystemEncoding(void) {
if (__CFDefaultSystemEncoding == kCFStringEncodingInvalidId) {
__CFDefaultSystemEncoding = __defaultEncoding;
const CFStringEncodingConverter *converter = CFStringEncodingGetConverter(__CFDefaultSystemEncoding);
__CFSetCharToUniCharFunc(converter->encodingClass == kCFStringEncodingConverterCheapEightBit ? converter->toUnicode.cheapEightBit : NULL);
}
return __CFDefaultSystemEncoding;
}
// Fast version for internal use
CF_INLINE CFStringEncoding __CFStringGetSystemEncoding(void) {
if (__CFDefaultSystemEncoding == kCFStringEncodingInvalidId) (void)CFStringGetSystemEncoding();
return __CFDefaultSystemEncoding;
}
CFStringEncoding CFStringFileSystemEncoding(void) {
if (__CFDefaultFileSystemEncoding == kCFStringEncodingInvalidId) {
#if TARGET_OS_MAC || TARGET_OS_WIN32 || TARGET_OS_BSD
__CFDefaultFileSystemEncoding = kCFStringEncodingUTF8;
#else
__CFDefaultFileSystemEncoding = CFStringGetSystemEncoding();
#endif
}
return __CFDefaultFileSystemEncoding;
}
/* ??? Is returning length when no other answer is available the right thing?
!!! All of the (length > (LONG_MAX / N)) type checks are to avoid wrap-around and eventual malloc overflow in the client
*/
CFIndex CFStringGetMaximumSizeForEncoding(CFIndex length, CFStringEncoding encoding) {
if (encoding == kCFStringEncodingUTF8) {
return (length > (LONG_MAX / 3)) ? kCFNotFound : (length * 3);
}
if ((encoding == kCFStringEncodingUTF32) || (encoding == kCFStringEncodingUTF32BE) || (encoding == kCFStringEncodingUTF32LE)) { // UTF-32
return (length > (LONG_MAX / sizeof(UTF32Char))) ? kCFNotFound : (length * sizeof(UTF32Char));
}
switch (encoding & 0xFFF) { // Mask off non-base part
case kCFStringEncodingUnicode:
return (length > (LONG_MAX / sizeof(UniChar))) ? kCFNotFound : (length * sizeof(UniChar));
case kCFStringEncodingNonLossyASCII:
return (length > (LONG_MAX / 6)) ? kCFNotFound : (length * 6); // 1 Unichar can expand to 6 bytes
case kCFStringEncodingMacRoman:
case kCFStringEncodingWindowsLatin1:
case kCFStringEncodingISOLatin1:
case kCFStringEncodingNextStepLatin:
case kCFStringEncodingASCII:
return length / sizeof(uint8_t);
default:
return length / sizeof(uint8_t);
}
}
/* Returns whether the indicated encoding can be stored in 8-bit chars
*/
CF_INLINE Boolean __CFStrEncodingCanBeStoredInEightBit(CFStringEncoding encoding) {
switch (encoding & 0xFFF) { // just use encoding base
case kCFStringEncodingInvalidId:
case kCFStringEncodingUnicode:
case kCFStringEncodingNonLossyASCII:
return false;
case kCFStringEncodingMacRoman:
case kCFStringEncodingWindowsLatin1:
case kCFStringEncodingISOLatin1:
case kCFStringEncodingNextStepLatin:
case kCFStringEncodingASCII:
return true;
default: return false;
}
}
/* Returns the encoding used in eight bit CFStrings (can't be any encoding which isn't 1-to-1 with Unicode)
For 10.9-linked apps, we've set this encoding to ASCII for all cases; see <rdar://problem/3597233>
*/
CFStringEncoding __CFStringComputeEightBitStringEncoding(void) {
// This flag prevents recursive entry into __CFStringComputeEightBitStringEncoding
static Boolean __CFStringIsBeingInitialized2 = false;
if (__CFStringIsBeingInitialized2) return kCFStringEncodingASCII;
__CFStringIsBeingInitialized2 = true;
Boolean useAscii = true;
__CFStringIsBeingInitialized2 = false;
if (useAscii) {
__CFDefaultEightBitStringEncoding = kCFStringEncodingASCII;
} else {
if (__CFDefaultEightBitStringEncoding == kCFStringEncodingInvalidId) {
CFStringEncoding systemEncoding = CFStringGetSystemEncoding();
if (systemEncoding == kCFStringEncodingInvalidId) { // We're right in the middle of querying system encoding from default database. Delaying to set until system encoding is determined.
return kCFStringEncodingASCII;
} else if (__CFStrEncodingCanBeStoredInEightBit(systemEncoding)) {
__CFDefaultEightBitStringEncoding = systemEncoding;
} else {
__CFDefaultEightBitStringEncoding = kCFStringEncodingASCII;
}
}
}
return __CFDefaultEightBitStringEncoding;
}
/* Returns whether the provided bytes can be stored in ASCII
*/
CF_INLINE Boolean __CFBytesInASCII(const uint8_t *bytes, CFIndex len) {
#if TARGET_RT_64_BIT
uint64_t align_mask = 7;
#else
uint32_t align_mask = 3;
#endif
/* Read bytes until the buffer is aligned. */
while (((uintptr_t)bytes & align_mask) && len > 0) {
if (*bytes++ & 0x80) return false;
len--;
}
#if TARGET_RT_64_BIT
/* A bit of unrolling; go by 32s, 16s, and 8s first */
while (len >= 4 * sizeof(uint64_t)) {
uint64_t val;
memcpy(&val, bytes, sizeof(uint64_t));
uint64_t hiBits = (val & 0x8080808080808080ULL); // More efficient to collect this rather than do a conditional at every step
bytes += sizeof(uint64_t);
memcpy(&val, bytes, sizeof(uint64_t));
hiBits |= (val & 0x8080808080808080ULL);
bytes += sizeof(uint64_t);
memcpy(&val, bytes, sizeof(uint64_t));
hiBits |= (val & 0x8080808080808080ULL);
bytes += sizeof(uint64_t);
memcpy(&val, bytes, sizeof(uint64_t));
if (hiBits | (val & 0x8080808080808080ULL)) return false;
bytes += sizeof(uint64_t);
len -= 4 * sizeof(uint64_t);
}
while (len >= 2 * sizeof(uint64_t)) {
uint64_t val;
memcpy(&val, bytes, sizeof(uint64_t));
uint64_t hiBits = (val & 0x8080808080808080ULL);
bytes += sizeof(uint64_t);
memcpy(&val, bytes, sizeof(uint64_t));
if (hiBits | (val & 0x8080808080808080ULL)) return false;
bytes += sizeof(uint64_t);
len -= 2 * sizeof(uint64_t);
}
while (len >= sizeof(uint64_t)) {
uint64_t val;
memcpy(&val, bytes, sizeof(uint64_t));
if (val & 0x8080808080808080ULL) return false;
bytes += sizeof(uint64_t);
len -= sizeof(uint64_t);
}
#endif
/* Go by 4s */
while (len >= sizeof(uint32_t)) {
uint32_t val;
memcpy(&val, bytes, sizeof(uint32_t));
if (val & 0x80808080U) return false;
bytes += sizeof(uint32_t);
len -= sizeof(uint32_t);
}
/* Handle the rest one byte at a time */
while (len--) {
if (*bytes++ & 0x80) return false;
}
return true;
}
/* Returns whether the provided 8-bit string in the specified encoding can be stored in an 8-bit CFString.
*/
CF_INLINE Boolean __CFCanUseEightBitCFStringForBytes(const uint8_t *bytes, CFIndex len, CFStringEncoding encoding) {
// If the encoding is the same as the 8-bit CFString encoding, we can just use the bytes as-is.
// One exception is ASCII, which unfortunately needs to mean ISOLatin1 for compatibility reasons <rdar://problem/5458321>.
if (encoding == __CFStringGetEightBitStringEncoding() && encoding != kCFStringEncodingASCII) return true;
if (__CFStringEncodingIsSupersetOfASCII(encoding) && __CFBytesInASCII(bytes, len)) return true;
return false;
}
/* Returns whether a length byte can be tacked on to a string of the indicated length.
*/
CF_INLINE Boolean __CFCanUseLengthByte(CFIndex len) {
#define __kCFMaxPascalStrLen 255
return (len <= __kCFMaxPascalStrLen) ? true : false;
}
/* Various string assertions
*/
#define __CFAssertIsString(cf) __CFGenericValidateType(cf, _kCFRuntimeIDCFString)
#define __CFAssertIndexIsInStringBounds(cf, idx) CFAssert3((idx) >= 0 && (idx) < __CFStrLength(cf), __kCFLogAssertion, "%s(): string index %ld out of bounds (length %ld)", __PRETTY_FUNCTION__, idx, __CFStrLength(cf))
#define __CFAssertRangeIsInStringBounds(cf, idx, count) CFAssert4((idx) >= 0 && (idx + count) <= __CFStrLength(cf), __kCFLogAssertion, "%s(): string range %ld,%ld out of bounds (length %ld)", __PRETTY_FUNCTION__, idx, count, __CFStrLength(cf))
#define __CFAssertIsStringAndMutable(cf) {__CFGenericValidateType(cf, _kCFRuntimeIDCFString); CFAssert1(__CFStrIsMutable(cf), __kCFLogAssertion, "%s(): string not mutable", __PRETTY_FUNCTION__);}
#define __CFAssertIsStringAndExternalMutable(cf) {__CFGenericValidateType(cf, _kCFRuntimeIDCFString); CFAssert1(__CFStrIsMutable(cf) && __CFStrIsExternalMutable(cf), __kCFLogAssertion, "%s(): string not external mutable", __PRETTY_FUNCTION__);}
#define __CFAssertIsNotNegative(idx) CFAssert2(idx >= 0, __kCFLogAssertion, "%s(): index %ld is negative", __PRETTY_FUNCTION__, idx)
#define __CFAssertIfFixedLengthIsOK(cf, reqLen) CFAssert2(!__CFStrIsFixed(cf) || (reqLen <= __CFStrDesiredCapacity(cf)), __kCFLogAssertion, "%s(): length %ld too large", __PRETTY_FUNCTION__, reqLen)
#define CF_RETURN_IF_NOT_MUTABLE(cf) do { \
if(!__CFStrIsMutable(cf)) {\
fprintf(stderr, "CFString: %s(): Expect mutable string\n", __PRETTY_FUNCTION__);\
return;\
} \
} while (0)
/* Basic algorithm is to shrink memory when capacity is SHRINKFACTOR times the required capacity or to allocate memory when the capacity is less than GROWFACTOR times the required capacity. This function will return -1 if the new capacity is just too big (> LONG_MAX).
Additional complications are applied in the following order:
- desiredCapacity, which is the minimum (except initially things can be at zero)
- rounding up to factor of 8
- compressing (to fit the number if 16 bits), which effectively rounds up to factor of 256
- we need to make sure GROWFACTOR computation doesn't suffer from overflow issues on 32-bit, hence the casting to unsigned. Normally for required capacity of C bytes, the allocated space is (3C+1)/2. If C > ULONG_MAX/3, we instead simply return LONG_MAX
*/
#define SHRINKFACTOR(c) (c / 2)
#if TARGET_RT_64_BIT
#define GROWFACTOR(c) ((c * 3 + 1) / 2)
#else
#define GROWFACTOR(c) (((c) >= (ULONG_MAX / 3UL)) ? __CFMax(LONG_MAX - 4095, (c)) : (((unsigned long)c * 3 + 1) / 2))
#endif
CF_INLINE CFIndex __CFStrNewCapacity(CFMutableStringRef str, unsigned long reqCapacity, CFIndex capacity, Boolean leaveExtraRoom, CFIndex charSize) {
if (capacity != 0 || reqCapacity != 0) { /* If initially zero, and space not needed, leave it at that... */
if ((capacity < reqCapacity) || /* We definitely need the room... */
(!__CFStrCapacityProvidedExternally(str) && /* Assuming we control the capacity... */
((reqCapacity < SHRINKFACTOR(capacity)) || /* ...we have too much room! */
(!leaveExtraRoom && (reqCapacity < capacity))))) { /* ...we need to eliminate the extra space... */
if (reqCapacity > LONG_MAX) return -1; /* Too big any way you cut it */
unsigned long newCapacity = leaveExtraRoom ? GROWFACTOR(reqCapacity) : reqCapacity; /* Grow by 3/2 if extra room is desired */
CFIndex desiredCapacity = __CFStrDesiredCapacity(str) * charSize;
if (newCapacity < desiredCapacity) { /* If less than desired, bump up to desired */
newCapacity = desiredCapacity;
} else if (__CFStrIsFixed(str)) { /* Otherwise, if fixed, no need to go above the desired (fixed) capacity */
newCapacity = __CFMax(desiredCapacity, reqCapacity); /* !!! So, fixed is not really fixed, but "tight" */
}
if (__CFStrHasContentsAllocator(str)) { /* Also apply any preferred size from the allocator */
newCapacity = CFAllocatorGetPreferredSizeForSize(__CFStrContentsAllocator(str), newCapacity, 0);
#if TARGET_OS_MAC
} else {
newCapacity = malloc_good_size(newCapacity);
#endif
}
return (newCapacity > LONG_MAX) ? -1 : (CFIndex)newCapacity; // If packing: __CFStrUnpackNumber(__CFStrPackNumber(newCapacity));
}
}
return capacity;
}
/* rearrangeBlocks() rearranges the blocks of data within the buffer so that they are "evenly spaced". buffer is assumed to have enough room for the result.
numBlocks is current total number of blocks within buffer.
blockSize is the size of each block in bytes
ranges and numRanges hold the ranges that are no longer needed; ranges are stored sorted in increasing order, and don't overlap
insertLength is the final spacing between the remaining blocks
Example: buffer = A B C D E F G H, blockSize = 1, ranges = { (2,1) , (4,2) } (so we want to "delete" C and E F), fromEnd = NO
if insertLength = 4, result = A B ? ? ? ? D ? ? ? ? G H
if insertLength = 0, result = A B D G H
Example: buffer = A B C D E F G H I J K L M N O P Q R S T U, blockSize = 1, ranges { (1,1), (3,1), (5,11), (17,1), (19,1) }, fromEnd = NO
if insertLength = 3, result = A ? ? ? C ? ? ? E ? ? ? Q ? ? ? S ? ? ? U
*/
typedef struct _CFStringDeferredRange {
CFIndex beginning;
CFIndex length;
CFIndex shift;
} CFStringDeferredRange;
typedef struct _CFStringStackInfo {
CFIndex capacity; // Capacity (if capacity == count, need to realloc to add another)
CFIndex count; // Number of elements actually stored
CFStringDeferredRange *stack;
Boolean hasMalloced; // Indicates "stack" is allocated and needs to be deallocated when done
char _padding[3];
} CFStringStackInfo;
CF_INLINE void pop (CFStringStackInfo *si, CFStringDeferredRange *topRange) {
si->count = si->count - 1;
*topRange = si->stack[si->count];
}
CF_INLINE void push (CFStringStackInfo *si, const CFStringDeferredRange *newRange) {
if (si->count == si->capacity) {
// increase size of the stack
si->capacity = (si->capacity + 4) * 2;
if (si->hasMalloced) {
si->stack = __CFSafelyReallocateWithAllocator(kCFAllocatorSystemDefault, si->stack, si->capacity * sizeof(CFStringDeferredRange), 0, NULL);
} else {
CFStringDeferredRange *newStack = (CFStringDeferredRange *)CFAllocatorAllocate(kCFAllocatorSystemDefault, si->capacity * sizeof(CFStringDeferredRange), 0);
memmove(newStack, si->stack, si->count * sizeof(CFStringDeferredRange));
si->stack = newStack;
si->hasMalloced = true;
}
}
si->stack[si->count] = *newRange;
si->count = si->count + 1;
}
static void rearrangeBlocks(
uint8_t *buffer,
CFIndex numBlocks,
CFIndex blockSize,
const CFRange *ranges,
CFIndex numRanges,
CFIndex insertLength) {
#define origStackSize 10
CFStringDeferredRange origStack[origStackSize];
CFStringStackInfo si = {origStackSize, 0, origStack, false, {0, 0, 0}};
CFStringDeferredRange currentNonRange = {0, 0, 0};
CFIndex currentRange = 0;
CFIndex amountShifted = 0;
// must have at least 1 range left.
while (currentRange < numRanges) {
currentNonRange.beginning = (ranges[currentRange].location + ranges[currentRange].length) * blockSize;
if ((numRanges - currentRange) == 1) {
// at the end.
currentNonRange.length = numBlocks * blockSize - currentNonRange.beginning;
if (currentNonRange.length == 0) break;
} else {
currentNonRange.length = (ranges[currentRange + 1].location * blockSize) - currentNonRange.beginning;
}
currentNonRange.shift = amountShifted + (insertLength * blockSize) - (ranges[currentRange].length * blockSize);
amountShifted = currentNonRange.shift;
if (amountShifted <= 0) {
// process current item and rest of stack
if (currentNonRange.shift && currentNonRange.length) memmove (&buffer[currentNonRange.beginning + currentNonRange.shift], &buffer[currentNonRange.beginning], currentNonRange.length);
while (si.count > 0) {
pop (&si, ¤tNonRange); // currentNonRange now equals the top element of the stack.
if (currentNonRange.shift && currentNonRange.length) memmove (&buffer[currentNonRange.beginning + currentNonRange.shift], &buffer[currentNonRange.beginning], currentNonRange.length);
}
} else {
// add currentNonRange to stack.
push (&si, ¤tNonRange);
}
currentRange++;
}
// no more ranges. if anything is on the stack, process.
while (si.count > 0) {
pop (&si, ¤tNonRange); // currentNonRange now equals the top element of the stack.
if (currentNonRange.shift && currentNonRange.length) memmove (&buffer[currentNonRange.beginning + currentNonRange.shift], &buffer[currentNonRange.beginning], currentNonRange.length);
}
if (si.hasMalloced) CFAllocatorDeallocate (kCFAllocatorSystemDefault, si.stack);
}
/* See comments for rearrangeBlocks(); this is the same, but the string is assembled in another buffer (dstBuffer), so the algorithm is much easier. We also take care of the case where the source is not-Unicode but destination is. (The reverse case is not supported.)
*/
static void copyBlocks(
const uint8_t *srcBuffer,
uint8_t *dstBuffer,
CFIndex srcLength,
Boolean srcIsUnicode,
Boolean dstIsUnicode,
const CFRange *ranges,
CFIndex numRanges,
CFIndex insertLength) {
CFIndex srcLocationInBytes = 0; // in order to avoid multiplying all the time, this is in terms of bytes, not blocks
CFIndex dstLocationInBytes = 0; // ditto
CFIndex srcBlockSize = srcIsUnicode ? sizeof(UniChar) : sizeof(uint8_t);
CFIndex insertLengthInBytes = insertLength * (dstIsUnicode ? sizeof(UniChar) : sizeof(uint8_t));
CFIndex rangeIndex = 0;
CFIndex srcToDstMultiplier = (srcIsUnicode == dstIsUnicode) ? 1 : (sizeof(UniChar) / sizeof(uint8_t));
// Loop over the ranges, copying the range to be preserved (right before each range)
while (rangeIndex < numRanges) {
CFIndex srcLengthInBytes = ranges[rangeIndex].location * srcBlockSize - srcLocationInBytes; // srcLengthInBytes is in terms of bytes, not blocks; represents length of region to be preserved
if (srcLengthInBytes > 0) {
if (srcIsUnicode == dstIsUnicode) {
memmove(dstBuffer + dstLocationInBytes, srcBuffer + srcLocationInBytes, srcLengthInBytes);
} else {
__CFStrConvertBytesToUnicode(srcBuffer + srcLocationInBytes, (UniChar *)(dstBuffer + dstLocationInBytes), srcLengthInBytes);
}
}
srcLocationInBytes += srcLengthInBytes + ranges[rangeIndex].length * srcBlockSize; // Skip over the just-copied and to-be-deleted stuff
dstLocationInBytes += srcLengthInBytes * srcToDstMultiplier + insertLengthInBytes;
rangeIndex++;
}
// Do last range (the one beyond last range)
if (srcLocationInBytes < srcLength * srcBlockSize) {
if (srcIsUnicode == dstIsUnicode) {
memmove(dstBuffer + dstLocationInBytes, srcBuffer + srcLocationInBytes, srcLength * srcBlockSize - srcLocationInBytes);
} else {
__CFStrConvertBytesToUnicode(srcBuffer + srcLocationInBytes, (UniChar *)(dstBuffer + dstLocationInBytes), srcLength * srcBlockSize - srcLocationInBytes);
}
}
}
/* Call the callback; if it doesn't exist or returns false, then log
*/
__attribute__((cold))
void __CFStringHandleOutOfMemory(CFTypeRef _Nullable obj) CLANG_ANALYZER_NORETURN {
CFStringRef msg = CFSTR("Out of memory. We suggest restarting the application. If you have an unsaved document, create a backup copy in Finder, then try to save.");
}
/* Reallocates the backing store of the string to accomodate the new length. Space is reserved or characters are deleted as indicated by insertLength and the ranges in deleteRanges. The length is updated to reflect the new state. Will also maintain a length byte and a null byte in 8-bit strings. If length cannot fit in length byte, the space will still be reserved, but will be 0. (Hence the reason the length byte should never be looked at as length unless there is no explicit length.)
*/
static void __CFStringChangeSizeMultiple(CFMutableStringRef str, const CFRange *deleteRanges, CFIndex numDeleteRanges, CFIndex insertLength, Boolean makeUnicode) {
const uint8_t *curContents = (uint8_t *)__CFStrContents(str);
CFIndex curLength = curContents ? __CFStrLength2(str, curContents) : 0;
unsigned long newLength; // We use unsigned to better keep track of overflow
// Compute new length of the string
if (numDeleteRanges == 1) {
newLength = curLength + insertLength - deleteRanges[0].length;
} else {
CFIndex cnt;
newLength = curLength + insertLength * numDeleteRanges;
for (cnt = 0; cnt < numDeleteRanges; cnt++) newLength -= deleteRanges[cnt].length;
}
// Disabled: <rdar://problem/23208702> Questionable assert in CFString.c
//__CFAssertIfFixedLengthIsOK(str, newLength);
if (newLength == 0) {
// An somewhat optimized code-path for this special case, with the following implicit values:
// newIsUnicode = false
// useLengthAndNullBytes = false
// newCharSize = sizeof(uint8_t)
// If the newCapacity happens to be the same as the old, we don't free the buffer; otherwise we just free it totally
// instead of doing a potentially useless reallocation (as the needed capacity later might turn out to be different anyway)
CFIndex curCapacity = __CFStrCapacity(str);
CFIndex newCapacity = __CFStrNewCapacity(str, 0, curCapacity, true, sizeof(uint8_t));
if (newCapacity != curCapacity) { // If we're reallocing anyway (larger or smaller --- larger could happen if desired capacity was changed in the meantime), let's just free it all
if (curContents) __CFStrDeallocateMutableContents(str, (uint8_t *)curContents);
__CFStrSetContentPtr(str, NULL);
__CFStrSetCapacity(str, 0);
__CFStrClearCapacityProvidedExternally(str);
__CFStrClearHasLengthAndNullBytes(str);
if (!__CFStrIsExternalMutable(str)) __CFStrSetUnicode(str, false); // External mutable implies Unicode
} else {
if (!__CFStrIsExternalMutable(str)) {
__CFStrSetUnicode(str, false);
if (curCapacity >= (int)(sizeof(uint8_t) * 2)) { // If there's room
if (!curContents) { CRSetCrashLogMessage("String had a capacity but NULL buffer pointer"); HALT; }
__CFStrSetHasLengthAndNullBytes(str);
((uint8_t *)curContents)[0] = ((uint8_t *)curContents)[1] = 0;
} else {
__CFStrClearHasLengthAndNullBytes(str);
}
}
}
__CFStrSetExplicitLength(str, 0);
} else { /* This else-clause assumes newLength > 0 */
Boolean oldIsUnicode = __CFStrIsUnicode(str);
Boolean newIsUnicode = makeUnicode || (oldIsUnicode /* && (newLength > 0) - implicit */ ) || __CFStrIsExternalMutable(str);
CFIndex newCharSize = newIsUnicode ? sizeof(UniChar) : sizeof(uint8_t);
Boolean useLengthAndNullBytes = !newIsUnicode /* && (newLength > 0) - implicit */;
CFIndex numExtraBytes = useLengthAndNullBytes ? 2 : 0; /* 2 extra bytes to keep the length byte & null... */
CFIndex curCapacity = __CFStrCapacity(str);
if (newLength > (LONG_MAX - numExtraBytes) / newCharSize) __CFStringHandleOutOfMemory(str); // Does not return
CFIndex newCapacity = __CFStrNewCapacity(str, newLength * newCharSize + numExtraBytes, curCapacity, true, newCharSize);
if (newCapacity == -1) __CFStringHandleOutOfMemory(str); // Does not return
Boolean allocNewBuffer = (curContents == NULL) || (newCapacity != curCapacity) || (curLength > 0 && !oldIsUnicode && newIsUnicode); /* We alloc new buffer if oldIsUnicode != newIsUnicode because the contents have to be copied */
uint8_t *newContents;
if (allocNewBuffer) {
newContents = (uint8_t *)__CFStrAllocateMutableContents(str, newCapacity);
if (!newContents) { // Try allocating without extra room
newCapacity = __CFStrNewCapacity(str, newLength * newCharSize + numExtraBytes, curCapacity, false, newCharSize);
// Since we checked for this above, it shouldn't be the case here, but just in case
if (newCapacity == -1) __CFStringHandleOutOfMemory(str); // Does not return
newContents = (uint8_t *)__CFStrAllocateMutableContents(str, newCapacity);
if (!newContents) __CFStringHandleOutOfMemory(str); // Does not return
}
} else {
newContents = (uint8_t *)curContents;
}
Boolean hasLengthAndNullBytes = __CFStrHasLengthByte(str);
CFAssert1(hasLengthAndNullBytes == __CFStrHasNullByte(str), __kCFLogAssertion, "%s(): Invalid state in 8-bit string", __PRETTY_FUNCTION__);
// Calculate pointers to the actual string content (skipping over the length byte, if present). Note that keeping a reference to the base is needed for newContents under GC, since the copy may take a long time.
const uint8_t *curContentsBody = hasLengthAndNullBytes ? (curContents+1) : curContents;
uint8_t *newContentsBody = useLengthAndNullBytes ? (newContents+1) : newContents;
if (curContents) {
if (oldIsUnicode == newIsUnicode) {
if (newContentsBody == curContentsBody) {
rearrangeBlocks(newContentsBody, curLength, newCharSize, deleteRanges, numDeleteRanges, insertLength);
} else {
copyBlocks(curContentsBody, newContentsBody, curLength, oldIsUnicode, newIsUnicode, deleteRanges, numDeleteRanges, insertLength);
}
} else if (newIsUnicode) { /* this implies we have a new buffer */
copyBlocks(curContentsBody, newContentsBody, curLength, oldIsUnicode, newIsUnicode, deleteRanges, numDeleteRanges, insertLength);
}
if (allocNewBuffer && __CFStrFreeContentsWhenDone(str)) __CFStrDeallocateMutableContents(str, (void *)curContents);
}
if (!newIsUnicode) {
if (useLengthAndNullBytes) {
newContentsBody[newLength] = 0; /* Always have null byte, if not unicode */
newContents[0] = __CFCanUseLengthByte(newLength) ? (uint8_t)newLength : 0;
if (!hasLengthAndNullBytes) __CFStrSetHasLengthAndNullBytes(str);
} else {
if (hasLengthAndNullBytes) __CFStrClearHasLengthAndNullBytes(str);
}
if (oldIsUnicode) __CFStrSetUnicode(str, false);
} else { // New is unicode...
if (!oldIsUnicode) __CFStrSetUnicode(str, true);
if (hasLengthAndNullBytes) __CFStrClearHasLengthAndNullBytes(str);
}
__CFStrSetExplicitLength(str, newLength);
if (allocNewBuffer) {
__CFStrSetCapacity(str, newCapacity);
__CFStrClearCapacityProvidedExternally(str);
// __CFStrEnsureContentsFreeable(str); // Commented out until we clarify: <rdar://problem/27151105>. Until then will leak: <rdar://problem/26346533>
__CFStrSetContentPtr(str, newContents);
}
}
}
/* Same as above, but takes one range (very common case)
*/
CF_INLINE void __CFStringChangeSize(CFMutableStringRef str, CFRange range, CFIndex insertLength, Boolean makeUnicode) {
__CFStringChangeSizeMultiple(str, &range, 1, insertLength, makeUnicode);
}
#if defined(DEBUG)
static Boolean __CFStrIsConstantString(CFStringRef str);
#endif
static void __CFStringDeallocate(CFTypeRef cf) {
CFStringRef str = (CFStringRef)cf;
// If in DEBUG mode, check to see if the string a CFSTR, and complain.
CFAssert1(__CFConstantStringTableBeingFreed || !__CFStrIsConstantString((CFStringRef)cf), __kCFLogAssertion, "Tried to deallocate CFSTR(\"%@\")", str);
if (!__CFStrIsInline(str)) {
uint8_t *contents;
Boolean isMutable = __CFStrIsMutable(str);
if (__CFStrFreeContentsWhenDone(str) && (contents = (uint8_t *)__CFStrContents(str))) {
if (isMutable) {
__CFStrDeallocateMutableContents((CFMutableStringRef)str, contents);
} else {
if (__CFStrHasContentsDeallocator(str)) {
CFAllocatorRef allocator = __CFStrContentsDeallocator(str);
CFAllocatorDeallocate(allocator, contents);
CFRelease(allocator);
} else {
CFAllocatorRef alloc = __CFGetAllocator(str);
CFAllocatorDeallocate(alloc, contents);
}
}
}
if (isMutable && __CFStrHasContentsAllocator(str)) {
CFAllocatorRef allocator = __CFStrContentsAllocator((CFMutableStringRef)str);
CFRelease(allocator);
}
}
}
static Boolean __CFStringEqual(CFTypeRef cf1, CFTypeRef cf2) {
CFStringRef str1 = (CFStringRef)cf1;
CFStringRef str2 = (CFStringRef)cf2;
const uint8_t *contents1;
const uint8_t *contents2;
CFIndex len1;
/* !!! We do not need IsString assertions, as the CFBase runtime assures this */
/* !!! We do not need == test, as the CFBase runtime assures this */
contents1 = (uint8_t *)__CFStrContents(str1);
contents2 = (uint8_t *)__CFStrContents(str2);
len1 = __CFStrLength2(str1, contents1);
if (len1 != __CFStrLength2(str2, contents2)) return false;
contents1 += __CFStrSkipAnyLengthByte(str1);
contents2 += __CFStrSkipAnyLengthByte(str2);
if (__CFStrIsEightBit(str1) && __CFStrIsEightBit(str2)) {
return memcmp((const char *)contents1, (const char *)contents2, len1) ? false : true;
} else if (__CFStrIsEightBit(str1)) { /* One string has Unicode contents */
CFStringInlineBuffer buf;
CFIndex buf_idx = 0;
_CFStringInitInlineBufferInternal(str1, &buf, CFRangeMake(0, len1), false);
for (buf_idx = 0; buf_idx < len1; buf_idx++) {
if (__CFStringGetCharacterFromInlineBufferQuick(&buf, buf_idx) != ((UniChar *)contents2)[buf_idx]) return false;
}
} else if (__CFStrIsEightBit(str2)) { /* One string has Unicode contents */
CFStringInlineBuffer buf;
CFIndex buf_idx = 0;
_CFStringInitInlineBufferInternal(str2, &buf, CFRangeMake(0, len1), false);
for (buf_idx = 0; buf_idx < len1; buf_idx++) {
if (__CFStringGetCharacterFromInlineBufferQuick(&buf, buf_idx) != ((UniChar *)contents1)[buf_idx]) return false;
}
} else { /* Both strings have Unicode contents */
CFIndex idx;
for (idx = 0; idx < len1; idx++) {
if (((UniChar *)contents1)[idx] != ((UniChar *)contents2)[idx]) return false;
}
}
return true;
}
CF_PRIVATE Boolean _CFStringEqual(CFStringRef cf1, CFStringRef cf2) {
return __CFStringEqual(cf1, cf2);
}
/* String hashing: Should give the same results whatever the encoding; so we hash UniChars.
If the length is less than or equal to 96, then the hash function is simply the
following (n is the nth UniChar character, starting from 0):
hash(-1) = length
hash(n) = hash(n-1) * 257 + unichar(n);
Hash = hash(length-1) * ((length & 31) + 1)
If the length is greater than 96, then the above algorithm applies to
characters 0..31, (length/2)-16..(length/2)+15, and length-32..length-1, inclusive;
thus the first, middle, and last 32 characters.
Note that the loops below are unrolled; and: 67503105 is 257^4 - 256^4
If hashcode is changed from UInt32 to something else, this last piece needs to be readjusted.
!!! We haven't updated for LP64 yet
NOTE: The hash algorithm used to be duplicated in CF and Foundation; but now it should only be in the four functions below.
Hash function was changed between Panther and Tiger, and Tiger and Leopard.
*/
#define HashEverythingLimit 96
#define HashNextFourUniChars(accessStart, accessEnd, pointer) \
{result = result * 67503105U + (((accessStart 0 accessEnd) * 257U + (accessStart 1 accessEnd)) * 257U + (accessStart 2 accessEnd)) * 257U + (accessStart 3 accessEnd); pointer += 4;}
#define HashNextUniChar(accessStart, accessEnd, pointer) \
{result = result * 257U + (accessStart 0 accessEnd); pointer++;}
/* In this function, actualLen is the length of the original string; but len is the number of characters in buffer. The buffer is expected to contain the parts of the string relevant to hashing.
*/
CF_INLINE CFHashCode __CFStrHashCharacters(const UniChar *uContents, CFIndex len, CFIndex actualLen) {
CFHashCode result = actualLen;
if (len <= HashEverythingLimit) {
const UniChar *end4 = uContents + (len & ~3);
const UniChar *end = uContents + len;
while (uContents < end4) HashNextFourUniChars(uContents[, ], uContents); // First count in fours
while (uContents < end) HashNextUniChar(uContents[, ], uContents); // Then for the last <4 chars, count in ones...
} else {
const UniChar *contents, *end;
contents = uContents;
end = contents + 32;
while (contents < end) HashNextFourUniChars(contents[, ], contents);
contents = uContents + (len >> 1) - 16;
end = contents + 32;
while (contents < end) HashNextFourUniChars(contents[, ], contents);
end = uContents + len;
contents = end - 32;
while (contents < end) HashNextFourUniChars(contents[, ], contents);
}
return result + (result << (actualLen & 31));
}
/* This hashes cString in the eight bit string encoding. It also includes the little debug-time sanity check.
*/
CF_INLINE CFHashCode __CFStrHashEightBit(const uint8_t *cContents, CFIndex len) {
#if defined(DEBUG)
if (!__CFCharToUniCharFunc) { // A little sanity verification: If this is not set, trying to hash high byte chars would be a bad idea
CFIndex cnt;
Boolean err = false;
if (len <= HashEverythingLimit) {
for (cnt = 0; cnt < len; cnt++) if (cContents[cnt] >= 128) err = true;
} else {
for (cnt = 0; cnt < 32; cnt++) if (cContents[cnt] >= 128) err = true;
for (cnt = (len >> 1) - 16; cnt < (len >> 1) + 16; cnt++) if (cContents[cnt] >= 128) err = true;
for (cnt = (len - 32); cnt < len; cnt++) if (cContents[cnt] >= 128) err = true;
}
if (err) {
// Can't do log here, as it might be too early
fprintf(stderr, "Warning: CFHash() attempting to hash CFString containing high bytes before properly initialized to do so\n");
}
}
#endif
CFHashCode result = len;
if (len <= HashEverythingLimit) {
const uint8_t *end4 = cContents + (len & ~3);
const uint8_t *end = cContents + len;
while (cContents < end4) HashNextFourUniChars(__CFCharToUniCharTable[cContents[, ]], cContents); // First count in fours
while (cContents < end) HashNextUniChar(__CFCharToUniCharTable[cContents[, ]], cContents); // Then for the last <4 chars, count in ones...
} else {
const uint8_t *contents, *end;
contents = cContents;
end = contents + 32;
while (contents < end) HashNextFourUniChars(__CFCharToUniCharTable[contents[, ]], contents);
contents = cContents + (len >> 1) - 16;
end = contents + 32;
while (contents < end) HashNextFourUniChars(__CFCharToUniCharTable[contents[, ]], contents);
end = cContents + len;
contents = end - 32;
while (contents < end) HashNextFourUniChars(__CFCharToUniCharTable[contents[, ]], contents);
}
return result + (result << (len & 31));
}
// This is for NSStringROMKeySet.
CF_PRIVATE CFHashCode __CFStrHashEightBit2(const uint8_t *cContents, CFIndex len) {
return __CFStrHashEightBit(cContents, len);
}
CFHashCode CFStringHashISOLatin1CString(const uint8_t *bytes, CFIndex len) {
CFHashCode result = len;
if (len <= HashEverythingLimit) {
const uint8_t *end4 = bytes + (len & ~3);
const uint8_t *end = bytes + len;
while (bytes < end4) HashNextFourUniChars(bytes[, ], bytes); // First count in fours
while (bytes < end) HashNextUniChar(bytes[, ], bytes); // Then for the last <4 chars, count in ones...
} else {
const uint8_t *contents, *end;
contents = bytes;
end = contents + 32;
while (contents < end) HashNextFourUniChars(contents[, ], contents);
contents = bytes + (len >> 1) - 16;
end = contents + 32;
while (contents < end) HashNextFourUniChars(contents[, ], contents);
end = bytes + len;
contents = end - 32;
while (contents < end) HashNextFourUniChars(contents[, ], contents);
}
return result + (result << (len & 31));
}
CFHashCode CFStringHashCString(const uint8_t *bytes, CFIndex len) {
return __CFStrHashEightBit(bytes, len);
}
CFHashCode CFStringHashCharacters(const UniChar *characters, CFIndex len) {
return __CFStrHashCharacters(characters, len, len);
}
/* This is meant to be called from NSString or subclassers only. It is an error for this to be called without the ObjC runtime or an argument which is not an NSString or subclass. It can be called with NSCFString, although that would be inefficient (causing indirection) and won't normally happen anyway, as NSCFString overrides hash.
*/
CFHashCode CFStringHashNSString(CFStringRef str) {
UniChar buffer[HashEverythingLimit];
CFIndex bufLen; // Number of characters in the buffer for hashing
CFIndex len = 0; // Actual length of the string
#if DEPLOYMENT_RUNTIME_SWIFT
len = CF_SWIFT_CALLV(str, NSString.length);
if (len <= HashEverythingLimit) {
(void)CF_SWIFT_CALLV(str, NSString.getCharacters, CFRangeMake(0, len), buffer);
bufLen = len;
} else {
(void)CF_SWIFT_CALLV(str, NSString.getCharacters, CFRangeMake(0, 32), buffer);
(void)CF_SWIFT_CALLV(str, NSString.getCharacters, CFRangeMake((len >> 1) - 16, 32), buffer+32);
(void)CF_SWIFT_CALLV(str, NSString.getCharacters, CFRangeMake(len - 32, 32), buffer+64);
bufLen = HashEverythingLimit;
}
#else
len = CF_OBJC_CALLV((NSString *)str, length);
if (len <= HashEverythingLimit) {
(void)CF_OBJC_CALLV((NSString *)str, getCharacters:buffer range:NSMakeRange(0, len));
bufLen = len;
} else {
(void)CF_OBJC_CALLV((NSString *)str, getCharacters:buffer range:NSMakeRange(0, 32));
(void)CF_OBJC_CALLV((NSString *)str, getCharacters:buffer+32 range:NSMakeRange((len >> 1) - 16, 32));
(void)CF_OBJC_CALLV((NSString *)str, getCharacters:buffer+64 range:NSMakeRange(len - 32, 32));
bufLen = HashEverythingLimit;
}
#endif
return __CFStrHashCharacters(buffer, bufLen, len);
}
CFHashCode __CFStringHash(CFTypeRef cf) {
/* !!! We do not need an IsString assertion here, as this is called by the CFBase runtime only */
CFStringRef str = (CFStringRef)cf;
const uint8_t *contents = (uint8_t *)__CFStrContents(str);
CFIndex len = __CFStrLength2(str, contents);
if (__CFStrIsEightBit(str)) {
contents += __CFStrSkipAnyLengthByte(str);
return __CFStrHashEightBit(contents, len);
} else {
return __CFStrHashCharacters((const UniChar *)contents, len, len);
}
}
static CFStringRef __CFStringCopyDescription(CFTypeRef cf) {
return CFStringCreateWithFormat(kCFAllocatorSystemDefault, NULL, CFSTR("<CFString %p [%p]>{contents = \"%@\"}"), cf, __CFGetAllocator(cf), cf);
}
static CFStringRef __CFStringCopyFormattingDescription(CFTypeRef cf, CFDictionaryRef formatOptions) {
return (CFStringRef)CFStringCreateCopy(__CFGetAllocator(cf), (CFStringRef)cf);
}
CF_PRIVATE CFStringRef _CFNonObjCStringCreateCopy(CFAllocatorRef alloc, CFStringRef str);
typedef CFTypeRef (*CF_STRING_CREATE_COPY)(CFAllocatorRef alloc, CFTypeRef theString);
const CFRuntimeClass __CFStringClass = {
_kCFRuntimeScannedObject,
"CFString",
NULL, // init
(CF_STRING_CREATE_COPY)_CFNonObjCStringCreateCopy,
__CFStringDeallocate,
__CFStringEqual,
__CFStringHash,
__CFStringCopyFormattingDescription,
__CFStringCopyDescription
};
CFTypeID CFStringGetTypeID(void) {
return _kCFRuntimeIDCFString;
}
static Boolean CFStrIsUnicode(CFStringRef str) {
CF_SWIFT_FUNCDISPATCHV(_kCFRuntimeIDCFString, Boolean, str, NSString._encodingCantBeStoredInEightBitCFString);
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, Boolean, (NSString *)str, _encodingCantBeStoredInEightBitCFString);
return __CFStrIsUnicode(str);
}
#define ALLOCATORSFREEFUNC ((CFAllocatorRef)-1)
/* contentsDeallocator indicates how to free the data if it's noCopy == true:
kCFAllocatorNull: don't free
ALLOCATORSFREEFUNC: free with main allocator's free func (don't pass in the real func ptr here)
NULL: default allocator
otherwise it's the allocator that should be used (it will be explicitly stored)
if noCopy == false, then freeFunc should be ALLOCATORSFREEFUNC
hasLengthByte, hasNullByte: refers to bytes; used only if encoding != Unicode
possiblyExternalFormat indicates that the bytes might have BOM and be swapped
tryToReduceUnicode means that the Unicode should be checked to see if it contains just ASCII (and reduce it if so)
numBytes contains the actual number of bytes in "bytes", including Length byte,
BUT not the NULL byte at the end
bytes should not contain BOM characters
!!! Various flags should be combined to reduce number of arguments, if possible
*/
CF_PRIVATE CFStringRef __CFStringCreateImmutableFunnel3(
CFAllocatorRef alloc, const void *bytes, CFIndex numBytes, CFStringEncoding encoding,
Boolean possiblyExternalFormat, Boolean tryToReduceUnicode, Boolean hasLengthByte, Boolean hasNullByte, Boolean noCopy,
CFAllocatorRef contentsDeallocator, UInt32 converterFlags) {
CFMutableStringRef str = NULL;
CFVarWidthCharBuffer vBuf;
CFIndex size;
Boolean useLengthByte = false;
Boolean useNullByte = false;
Boolean useInlineData = false;
#if INSTRUMENT_SHARED_STRINGS
const char *recordedEncoding;
char encodingBuffer[128];
if (encoding == kCFStringEncodingUnicode) recordedEncoding = "Unicode";
else if (encoding == kCFStringEncodingASCII) recordedEncoding = "ASCII";
else if (encoding == kCFStringEncodingUTF8) recordedEncoding = "UTF8";
else if (encoding == kCFStringEncodingMacRoman) recordedEncoding = "MacRoman";
else {
snprintf(encodingBuffer, sizeof(encodingBuffer), "0x%lX", (unsigned long)encoding);
recordedEncoding = encodingBuffer;
}
#endif
if (alloc == NULL) alloc = __CFGetDefaultAllocator();
if (contentsDeallocator == ALLOCATORSFREEFUNC) {
contentsDeallocator = alloc;
} else if (contentsDeallocator == NULL) {
contentsDeallocator = __CFGetDefaultAllocator();
}
if ((NULL != kCFEmptyString) && (numBytes == 0) && _CFAllocatorIsSystemDefault(alloc)) { // If we are using the system default allocator, and the string is empty, then use the empty string!
if (noCopy && (contentsDeallocator != kCFAllocatorNull)) { // See 2365208... This change was done after Sonata; before we didn't free the bytes at all (leak).
CFAllocatorDeallocate(contentsDeallocator, (void *)bytes);
}
return (CFStringRef)CFRetain(kCFEmptyString); // Quick exit; won't catch all empty strings, but most
}
// At this point, contentsDeallocator is either same as alloc, or kCFAllocatorNull, or something else, but not NULL
vBuf.shouldFreeChars = false; // We use this to remember to free the buffer possibly allocated by decode
// Record whether we're starting out with an ASCII-superset string, because we need to know this later for the string ROM; this may get changed later if we successfully convert down from Unicode. We only record this once because __CFCanUseEightBitCFStringForBytes() can be expensive.
Boolean stringSupportsEightBitCFRepresentation = encoding != kCFStringEncodingUnicode && __CFCanUseEightBitCFStringForBytes((const uint8_t *)bytes, numBytes, encoding);
// We may also change noCopy within this function if we have to decode the string into an external buffer. We do not want to avoid the use of the string ROM merely because we tried to be efficient and reuse the decoded buffer for the CFString's external storage. Therefore, we use this variable to track whether we actually can ignore the noCopy flag (which may or may not be set anyways).
Boolean stringROMShouldIgnoreNoCopy = false;
// First check to see if the data needs to be converted...
// ??? We could be more efficient here and in some cases (Unicode data) eliminate a copy
if ((encoding == kCFStringEncodingUnicode && possiblyExternalFormat) || (encoding != kCFStringEncodingUnicode && ! stringSupportsEightBitCFRepresentation)) {
const void *realBytes = (uint8_t *) bytes + (hasLengthByte ? 1 : 0);
CFIndex realNumBytes = numBytes - (hasLengthByte ? 1 : 0);
Boolean usingPassedInMemory = false;
vBuf.allocator = kCFAllocatorSystemDefault; // We don't want to use client's allocator for temp stuff
vBuf.chars.unicode = NULL; // This will cause the decode function to allocate memory if necessary
if (!__CFStringDecodeByteStream3((const uint8_t *)realBytes, realNumBytes, encoding, false, &vBuf, &usingPassedInMemory, converterFlags)) {
// Note that if the string can't be created, we don't free the buffer, even if there is a contents deallocator. This is on purpose.
return NULL;
}
encoding = vBuf.isASCII ? kCFStringEncodingASCII : kCFStringEncodingUnicode;
// Update our flag according to whether the decoded buffer is ASCII
stringSupportsEightBitCFRepresentation = vBuf.isASCII;
if (!usingPassedInMemory) {
// Because __CFStringDecodeByteStream3() allocated our buffer, it's OK for us to free it if we can get the string from the ROM.
stringROMShouldIgnoreNoCopy = true;
// Make the parameters fit the new situation
numBytes = vBuf.isASCII ? vBuf.numChars : (vBuf.numChars * sizeof(UniChar));
hasLengthByte = hasNullByte = false;
// Get rid of the original buffer if its not being used
if (noCopy && (contentsDeallocator != kCFAllocatorNull)) {
CFAllocatorDeallocate(contentsDeallocator, (void *)bytes);
}
contentsDeallocator = alloc; // At this point we are using the string's allocator, as the original buffer is gone...
// See if we can reuse any storage the decode func might have allocated
// We do this only for Unicode, as otherwise we would not have NULL and Length bytes
if (vBuf.shouldFreeChars && (alloc == vBuf.allocator) && encoding == kCFStringEncodingUnicode) {
vBuf.shouldFreeChars = false; // Transferring ownership to the CFString
bytes = __CFSafelyReallocateWithAllocator(vBuf.allocator, (void *)vBuf.chars.unicode, numBytes, 0, NULL); // Tighten up the storage
noCopy = true;
#if INSTRUMENT_SHARED_STRINGS
if (encoding == kCFStringEncodingASCII) recordedEncoding = "ForeignASCII-NoCopy";
else recordedEncoding = "ForeignUnicode-NoCopy";
#endif
} else {
#if INSTRUMENT_SHARED_STRINGS
if (encoding == kCFStringEncodingASCII) recordedEncoding = "ForeignASCII-Copy";
else recordedEncoding = "ForeignUnicode-Copy";
#endif
bytes = vBuf.chars.unicode;
noCopy = false; // Can't do noCopy anymore
// If vBuf.shouldFreeChars is true, the buffer will be freed as intended near the end of this func
}
}
// At this point, all necessary input arguments have been changed to reflect the new state
} else if (encoding == kCFStringEncodingUnicode && tryToReduceUnicode) { // Check to see if we can reduce Unicode to ASCII
CFIndex cnt;
CFIndex len = numBytes / sizeof(UniChar);
Boolean allASCII = true;
for (cnt = 0; cnt < len; cnt++) if (((const UniChar *)bytes)[cnt] > 127) {
allASCII = false;
break;
}
if (allASCII) { // Yes we can!
uint8_t *ptr, *mem;
Boolean newHasLengthByte = __CFCanUseLengthByte(len);
numBytes = (len + 1 + (newHasLengthByte ? 1 : 0)) * sizeof(uint8_t); // NULL and possible length byte
// See if we can use that temporary local buffer in vBuf...
if (numBytes >= __kCFVarWidthLocalBufferSize) {
mem = ptr = (uint8_t *)CFAllocatorAllocate(alloc, numBytes, 0);
if (__CFOASafe) __CFSetLastAllocationEventName(mem, "CFString (store)");
} else {
mem = ptr = (uint8_t *)(vBuf.localBuffer);
}
if (mem) { // If we can't allocate memory for some reason, use what we had (that is, as if we didn't have all ASCII)
// Copy the Unicode bytes into the new ASCII buffer
hasLengthByte = newHasLengthByte;
hasNullByte = true;
if (hasLengthByte) *ptr++ = (uint8_t)len;
for (cnt = 0; cnt < len; cnt++) ptr[cnt] = (uint8_t)(((const UniChar *)bytes)[cnt]);
ptr[len] = 0;
if (noCopy && (contentsDeallocator != kCFAllocatorNull)) {
CFAllocatorDeallocate(contentsDeallocator, (void *)bytes);
}
// Now make everything look like we had an ASCII buffer to start with
bytes = mem;
encoding = kCFStringEncodingASCII;
contentsDeallocator = alloc; // At this point we are using the string's allocator, as the original buffer is gone...
noCopy = (numBytes >= __kCFVarWidthLocalBufferSize); // If we had to allocate it, make sure it's kept around
numBytes--; // Should not contain the NULL byte at end...
stringSupportsEightBitCFRepresentation = true; // We're ASCII now!
stringROMShouldIgnoreNoCopy = true; // We allocated this buffer, so we should feel free to get rid of it if we can use the string ROM
#if INSTRUMENT_SHARED_STRINGS
recordedEncoding = "U->A";
#endif
}
}
// At this point, all necessary input arguments have been changed to reflect the new state
}
#if USE_STRING_ROM || ENABLE_TAGGED_POINTER_STRINGS || INSTRUMENT_SHARED_STRINGS
CFIndex lengthByte = (hasLengthByte ? 1 : 0);
CFIndex realNumBytes = numBytes - lengthByte;
const uint8_t *realBytes = bytes + lengthByte;
#endif
if (!str) {
// Now determine the necessary size
#if INSTRUMENT_SHARED_STRINGS || USE_STRING_ROM
Boolean stringSupportsROM = stringSupportsEightBitCFRepresentation;
#endif
#if INSTRUMENT_SHARED_STRINGS
if (stringSupportsROM) __CFRecordStringAllocationEvent(recordedEncoding, (const char *)realBytes, realNumBytes);
#endif
#if USE_STRING_ROM
CFStringRef romResult = NULL;
if (stringSupportsROM) {
// Disable the string ROM if necessary
static Boolean sDisableStringROM = false;
static dispatch_once_t onceToken;
dispatch_once(&onceToken, ^{
sDisableStringROM = getenv("CFStringDisableROM") != NULL;
});
if (!sDisableStringROM) romResult = __CFSearchStringROM((const char *)realBytes, realNumBytes);
}
/* if we get a result from our ROM, and noCopy is set, then deallocate the buffer immediately */
if (romResult) {
if (noCopy && (contentsDeallocator != kCFAllocatorNull)) {
CFAllocatorDeallocate(contentsDeallocator, (void *)bytes);
}
/* these don't get used again, but clear them for consistency */
noCopy = false;
bytes = NULL;
/* set our result to the ROM result which is not really mutable, of course, but that's OK because we don't try to modify it. */
str = (CFMutableStringRef)CFRetain(romResult);
#if INSTRUMENT_TAGGED_POINTER_STRINGS
_CFTaggedPointerStringStats.stringROMCount++;
#endif
}
if (! romResult) {
#else
if (1) {
#endif
#if INSTRUMENT_SHARED_STRINGS
if (stringSupportsROM) __CFRecordStringAllocationEvent(recordedEncoding, (const char *)realBytes, realNumBytes);
#endif
#if INSTRUMENT_TAGGED_POINTER_STRINGS
_CFTaggedPointerStringStats.otherStringCount++;
#endif
// Now determine the necessary size
if (noCopy) {
size = sizeof(void *); // Pointer to the buffer
if ((0) || (contentsDeallocator != alloc && contentsDeallocator != kCFAllocatorNull)) {
size += sizeof(void *); // The contentsDeallocator
}
if (!hasLengthByte) size += sizeof(CFIndex); // Explicit length
useLengthByte = hasLengthByte;
useNullByte = hasNullByte;
} else { // Inline data; reserve space for it
useInlineData = true;
size = numBytes;
if (hasLengthByte) {
useLengthByte = true;
} else if (encoding != kCFStringEncodingUnicode && __CFCanUseLengthByte(numBytes)) {
useLengthByte = true;
size += 1;
} else {
size += sizeof(CFIndex); // Explicit length
}
if (hasNullByte || encoding != kCFStringEncodingUnicode) {
useNullByte = true;
size += 1;
}
}
#ifdef STRING_SIZE_STATS
// Dump alloced CFString size info every so often
static int cnt = 0;
static unsigned sizes[256] = {0};
int allocedSize = size + sizeof(CFRuntimeBase);
if (allocedSize < 255) sizes[allocedSize]++; else sizes[255]++;
if ((++cnt % 1000) == 0) {
printf ("\nTotal: %d\n", cnt);
int i; for (i = 0; i < 256; i++) printf("%03d: %5d%s", i, sizes[i], ((i % 8) == 7) ? "\n" : " ");
}
#endif
// Finally, allocate!
#if DEPLOYMENT_RUNTIME_SWIFT
// Swift.String is 3 pointers, so we have to allocate to the largest of the two (some variants of __CFString are smaller than swift Strings)
CFIndex swiftStringSize = sizeof(CFRuntimeBase) + (sizeof(void *) * 3);
if (swiftStringSize > size) size = swiftStringSize;
#endif
str = (CFMutableStringRef)_CFRuntimeCreateInstance(alloc, _kCFRuntimeIDCFString, size, NULL);
if (str) {
if (__CFOASafe) __CFSetLastAllocationEventName(str, "CFString (immutable)");
_CFStringInlineContents allocBits = contentsDeallocator == alloc ? __kCFNotInlineContentsDefaultFree : (contentsDeallocator == kCFAllocatorNull ? __kCFNotInlineContentsNoFree : __kCFNotInlineContentsCustomFree);
__CFStrSetInlineContents(str, useInlineData ? __kCFHasInlineContents : allocBits);
__CFStrSetUnicode(str, encoding == kCFStringEncodingUnicode);
__CFStrSetHasNullByte(str, useNullByte);
__CFStrSetHasLengthByte(str, useLengthByte);
if (!useLengthByte) {
CFIndex length = numBytes - (hasLengthByte ? 1 : 0);
if (encoding == kCFStringEncodingUnicode) length /= sizeof(UniChar);
__CFStrSetExplicitLength(str, length);
}
if (useInlineData) {
uint8_t *contents = (uint8_t *)__CFStrContents(str);
if (useLengthByte && !hasLengthByte) *contents++ = (uint8_t)numBytes;
memmove(contents, bytes, numBytes);
if (useNullByte) contents[numBytes] = 0;
} else {
__CFStrSetContentPtr(str, bytes);
if (__CFStrHasContentsDeallocator(str)) __CFStrSetContentsDeallocator(str, contentsDeallocator);
}
} else {
if (noCopy && (contentsDeallocator != kCFAllocatorNull)) {
CFAllocatorDeallocate(contentsDeallocator, (void *)bytes);
}
}
}
}
if (vBuf.shouldFreeChars) CFAllocatorDeallocate(vBuf.allocator, (void *)bytes);
#if 0
#warning Debug code
const uint8_t *contents = (uint8_t *)__CFStrContents(str);
CFIndex len = __CFStrLength2(str, contents);
if (__CFStrIsEightBit(str)) {
contents += __CFStrSkipAnyLengthByte(str);
if (!__CFBytesInASCII(contents, len)) {
printf("CFString with 8 bit backing store not ASCII: %p, \"%.*s\"\n", str, (int)len, contents);
}
}
#endif
return str;
}
/* !!! __CFStringCreateImmutableFunnel2() is kept around for compatibility; it should be deprecated
*/
CFStringRef __CFStringCreateImmutableFunnel2(
CFAllocatorRef alloc, const void *bytes, CFIndex numBytes, CFStringEncoding encoding,
Boolean possiblyExternalFormat, Boolean tryToReduceUnicode, Boolean hasLengthByte, Boolean hasNullByte, Boolean noCopy,
CFAllocatorRef contentsDeallocator) {
return __CFStringCreateImmutableFunnel3(alloc, bytes, numBytes, encoding, possiblyExternalFormat, tryToReduceUnicode, hasLengthByte, hasNullByte, noCopy, contentsDeallocator, 0);
}
CFStringRef CFStringCreateWithPascalString(CFAllocatorRef alloc, ConstStringPtr pStr, CFStringEncoding encoding) {
CFIndex len = (CFIndex)(*(uint8_t *)pStr);
return __CFStringCreateImmutableFunnel3(alloc, pStr, len+1, encoding, false, false, true, false, false, ALLOCATORSFREEFUNC, 0);
}
CFStringRef CFStringCreateWithCString(CFAllocatorRef alloc, const char *cStr, CFStringEncoding encoding) {
CFIndex len = strlen(cStr);
return __CFStringCreateImmutableFunnel3(alloc, cStr, len, encoding, false, false, false, true, false, ALLOCATORSFREEFUNC, 0);
}
CFStringRef CFStringCreateWithPascalStringNoCopy(CFAllocatorRef alloc, ConstStringPtr pStr, CFStringEncoding encoding, CFAllocatorRef contentsDeallocator) {
CFIndex len = (CFIndex)(*(uint8_t *)pStr);
return __CFStringCreateImmutableFunnel3(alloc, pStr, len+1, encoding, false, false, true, false, true, contentsDeallocator, 0);
}
CFStringRef CFStringCreateWithCStringNoCopy(CFAllocatorRef alloc, const char *cStr, CFStringEncoding encoding, CFAllocatorRef contentsDeallocator) {
CFIndex len = strlen(cStr);
return __CFStringCreateImmutableFunnel3(alloc, cStr, len, encoding, false, false, false, true, true, contentsDeallocator, 0);
}
CFStringRef CFStringCreateWithCharacters(CFAllocatorRef alloc, const UniChar *chars, CFIndex numChars) {
return __CFStringCreateImmutableFunnel3(alloc, chars, numChars * sizeof(UniChar), kCFStringEncodingUnicode, false, true, false, false, false, ALLOCATORSFREEFUNC, 0);
}
CFStringRef CFStringCreateWithCharactersNoCopy(CFAllocatorRef alloc, const UniChar *chars, CFIndex numChars, CFAllocatorRef contentsDeallocator) {
return __CFStringCreateImmutableFunnel3(alloc, chars, numChars * sizeof(UniChar), kCFStringEncodingUnicode, false, false, false, false, true, contentsDeallocator, 0);
}
CFStringRef CFStringCreateWithBytes(CFAllocatorRef alloc, const uint8_t *bytes, CFIndex numBytes, CFStringEncoding encoding, Boolean externalFormat) {
return __CFStringCreateImmutableFunnel3(alloc, bytes, numBytes, encoding, externalFormat, true, false, false, false, ALLOCATORSFREEFUNC, 0);
}
CFStringRef _CFStringCreateWithBytesNoCopy(CFAllocatorRef alloc, const uint8_t *bytes, CFIndex numBytes, CFStringEncoding encoding, Boolean externalFormat, CFAllocatorRef contentsDeallocator) {
return __CFStringCreateImmutableFunnel3(alloc, bytes, numBytes, encoding, externalFormat, true, false, false, true, contentsDeallocator, 0);
}
CFStringRef CFStringCreateWithBytesNoCopy(CFAllocatorRef alloc, const uint8_t *bytes, CFIndex numBytes, CFStringEncoding encoding, Boolean externalFormat, CFAllocatorRef contentsDeallocator) {
return __CFStringCreateImmutableFunnel3(alloc, bytes, numBytes, encoding, externalFormat, true, false, false, true, contentsDeallocator, 0);
}
CFStringRef CFStringCreateStringWithValidatedFormat(CFAllocatorRef alloc, CFDictionaryRef formatOptions, CFStringRef validFormatSpecifiers, CFStringRef format, va_list arguments, CFErrorRef *errorPtr) {
CFMutableStringRef outputString = CFStringCreateMutable(kCFAllocatorSystemDefault, 0); //should use alloc if no copy/release
__CFStrSetDesiredCapacity(outputString, 120); // Given this will be tightened later, choosing a larger working string is fine
if (__CFStringAppendFormatCore(outputString, NULL, NULL, formatOptions, NULL, validFormatSpecifiers, format, 0, NULL, 0, arguments, NULL, errorPtr)) {
// ??? copy/release should not be necessary here -- just make immutable, compress if possible
// (However, this does make the string inline, and cause the supplied allocator to be used...)
CFStringRef str = (CFStringRef)CFStringCreateCopy(alloc, outputString);
CFRelease(outputString);
return str;
} else {
CFRelease(outputString);
// errorPtr set by __CFStringAppendFormatCore
return NULL;
}
}
CFStringRef CFStringCreateWithFormatAndArguments(CFAllocatorRef alloc, CFDictionaryRef formatOptions, CFStringRef format, va_list arguments) {
return _CFStringCreateWithFormatAndArgumentsAux2(alloc, NULL, NULL, formatOptions, format, arguments);
}
CFStringRef _CFStringCreateWithFormatAndArgumentsAux2(CFAllocatorRef alloc, CFStringRef (*copyDescFunc)(void *, const void *), CFStringRef (*contextDescFunc)(void *, const void *, const void *, bool , bool *), CFDictionaryRef formatOptions, CFStringRef format, va_list arguments) {
return _CFStringCreateWithFormatAndArgumentsReturningMetadata(alloc, copyDescFunc, contextDescFunc, formatOptions, NULL, format, NULL, arguments);
}
CFStringRef _CFStringCreateWithFormatAndArgumentsReturningMetadata(CFAllocatorRef alloc, CFStringRef (*copyDescFunc)(void *, const void *), CFStringRef (*contextDescFunc)(void *, const void *, const void *, bool , bool *), CFDictionaryRef formatOptions, CFDictionaryRef formatConfiguration, CFStringRef format, CFArrayRef *outMetadata, va_list arguments) {
CFStringRef str = NULL;
CFMutableStringRef outputString = CFStringCreateMutable(kCFAllocatorSystemDefault, 0); //should use alloc if no copy/release
__CFStrSetDesiredCapacity(outputString, 120); // Given this will be tightened later, choosing a larger working string is fine
CFErrorRef error;
if (__CFStringAppendFormatCore(outputString, copyDescFunc, contextDescFunc, formatOptions, formatConfiguration, NULL, format, 0, NULL, 0, arguments, outMetadata, &error)) {
// ??? copy/release should not be necessary here -- just make immutable, compress if possible
// (However, this does make the string inline, and cause the supplied allocator to be used...)
str = (CFStringRef)CFStringCreateCopy(alloc, outputString);
} else {
CFLog(kCFLogLevelError, CFSTR("ERROR: Failed to format string: %@"), error);
if (error) CFRelease(error);
}
CFRelease(outputString);
return str;
}
CFStringRef _CFStringCreateWithFormatAndArgumentsAux(CFAllocatorRef alloc, CFStringRef (*copyDescFunc)(void *, const void *), CFDictionaryRef formatOptions, CFStringRef format, va_list arguments) {
return _CFStringCreateWithFormatAndArgumentsAux2(alloc, copyDescFunc, NULL, formatOptions, format, arguments);
}
CFStringRef CFStringCreateWithFormat(CFAllocatorRef alloc, CFDictionaryRef formatOptions, CFStringRef format, ...) {
CFStringRef result;
va_list argList;
va_start(argList, format);
result = CFStringCreateWithFormatAndArguments(alloc, formatOptions, format, argList);
va_end(argList);
return result;
}
CFStringRef CFStringCreateWithSubstring(CFAllocatorRef alloc, CFStringRef str, CFRange range) {
CF_SWIFT_FUNCDISPATCHV(_kCFRuntimeIDCFString, CFStringRef, (CFSwiftRef)str, NSString._createSubstringWithRange, range);
// CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, CFStringRef , (NSString *)str, _createSubstringWithRange:NSMakeRange(range.location, range.length));
__CFAssertIsString(str);
__CFAssertRangeIsInStringBounds(str, range.location, range.length);
if ((range.location == 0) && (range.length == __CFStrLength(str))) { /* The substring is the whole string... */
return (CFStringRef)_CFNonObjCStringCreateCopy(alloc, str);
} else if (__CFStrIsEightBit(str)) {
const uint8_t *contents = (const uint8_t *)__CFStrContents(str);
return __CFStringCreateImmutableFunnel3(alloc, contents + range.location + __CFStrSkipAnyLengthByte(str), range.length, __CFStringGetEightBitStringEncoding(), false, false, false, false, false, ALLOCATORSFREEFUNC, 0);
} else {
const UniChar *contents = (UniChar *)__CFStrContents(str);
return __CFStringCreateImmutableFunnel3(alloc, contents + range.location, range.length * sizeof(UniChar), kCFStringEncodingUnicode, false, true, false, false, false, ALLOCATORSFREEFUNC, 0);
}
}
static CFStringRef _CFStringSlowPathCopyBundleUnloadingProtectedString(CFStringRef str) {
CFIndex const len = CFStringGetLength(str);
if (len == 0) {
// Check this first to both avoid an allocation, and avoid potentially stack-allocating a zero-length buffer below.
return CFSTR("");
}
CFStringEncoding const fastestEncoding = CFStringGetFastestEncoding(str);
const char * const cStr = _CFStringGetCStringPtrInternal(str, fastestEncoding, false, true);
if (cStr) {
return CFStringCreateWithBytes(kCFAllocatorSystemDefault, (const uint8_t *)cStr, len, fastestEncoding, false);
}
const UniChar * const charsPtr = CFStringGetCharactersPtr(str);
if (charsPtr) {
return CFStringCreateWithCharacters(kCFAllocatorSystemDefault, charsPtr, len);
}
CFIndex const maxByteCount = CFStringGetMaximumSizeForEncoding(len, fastestEncoding);
CFIndex byteCount = 0;
CFStringRef result = NULL;
SAFE_STACK_BUFFER_DECL(uint8_t, buffer, maxByteCount, 256 /* malloc for buffers longer than 256 bytes */); // `str` here is currently only ever a bundle ID. Bundle IDs are rarely this long.
if (CFStringGetBytes(str, CFRangeMake(0, len), fastestEncoding, 0, false, buffer, maxByteCount, &byteCount)) {
result = CFStringCreateWithBytes(kCFAllocatorSystemDefault, buffer, byteCount, fastestEncoding, false);
} else {
result = CFStringCreateMutableCopy(kCFAllocatorSystemDefault, 0, str);
}
SAFE_STACK_BUFFER_CLEANUP(buffer);
return result;
}
CF_PRIVATE CFStringRef _CFStringCopyBundleUnloadingProtectedString(CFStringRef str) {
return _CFStringSlowPathCopyBundleUnloadingProtectedString(str);
}
CF_PRIVATE CFStringRef _CFNonObjCStringCreateCopy(CFAllocatorRef alloc, CFStringRef str) {
__CFAssertIsString(str);
if (!__CFStrIsMutable((CFStringRef)str) && // If the string is not mutable
((alloc ? alloc : __CFGetDefaultAllocator()) == __CFGetAllocator(str)) && // and it has the same allocator as the one we're using
(__CFStrIsInline((CFStringRef)str) || __CFStrFreeContentsWhenDone((CFStringRef)str) || __CFStrIsConstant((CFStringRef)str))) { // and the characters are inline, or are owned by the string, or the string is constant
return _CFNonObjCRetain(str); // Then just retain instead of making a true copy
}
if (__CFStrIsEightBit((CFStringRef)str)) {
const uint8_t *contents = (const uint8_t *)__CFStrContents((CFStringRef)str);
return __CFStringCreateImmutableFunnel3(alloc, contents + __CFStrSkipAnyLengthByte((CFStringRef)str), __CFStrLength2((CFStringRef)str, contents), __CFStringGetEightBitStringEncoding(), false, false, false, false, false, ALLOCATORSFREEFUNC, 0);
} else {
const UniChar *contents = (const UniChar *)__CFStrContents((CFStringRef)str);
return __CFStringCreateImmutableFunnel3(alloc, contents, __CFStrLength2((CFStringRef)str, contents) * sizeof(UniChar), kCFStringEncodingUnicode, false, true, false, false, false, ALLOCATORSFREEFUNC, 0);
}
}
CFStringRef CFStringCreateCopy(CFAllocatorRef alloc, CFStringRef str) {
CF_SWIFT_FUNCDISPATCHV(_kCFRuntimeIDCFString, CFStringRef, (CFSwiftRef)str, NSString.copy);
// CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, CFStringRef, (NSString *)str, copy);
return _CFNonObjCStringCreateCopy(alloc, str);
}
/*** Constant string stuff... ***/
/* Table which holds constant strings created with CFSTR, when -fconstant-cfstrings option is not used. These dynamically created constant strings are stored in constantStringTable. The keys are the 8-bit constant C-strings from the compiler; the values are the CFStrings created for them. _CFSTRLock protects this table.
*/
static CFMutableDictionaryRef constantStringTable = NULL;
static CFLock_t _CFSTRLock = CFLockInit;
static CFStringRef __cStrCopyDescription(const void *ptr) {
return CFStringCreateWithCStringNoCopy(kCFAllocatorSystemDefault, (const char *)ptr, __CFStringGetEightBitStringEncoding(), kCFAllocatorNull);
}
static Boolean __cStrEqual(const void *ptr1, const void *ptr2) {
return (strcmp((const char *)ptr1, (const char *)ptr2) == 0);
}
static CFHashCode __cStrHash(const void *ptr) {
// It doesn't quite matter if we convert to Unicode correctly, as long as we do it consistently
const char *cStr = (const char *)ptr;
CFIndex len = strlen(cStr);
CFHashCode result = 0;
if (len <= 4) { // All chars
unsigned cnt = len;
while (cnt--) result += (result << 8) + *cStr++;
} else { // First and last 2 chars
result += (result << 8) + cStr[0];
result += (result << 8) + cStr[1];
result += (result << 8) + cStr[len-2];
result += (result << 8) + cStr[len-1];
}
result += (result << (len & 31));
return result;
}
#if DEPLOYMENT_RUNTIME_SWIFT
#else
CFStringRef __CFStringMakeConstantString(const char *cStr) {
CFStringRef result;
#if defined(DEBUG)
// StringTest checks that we share kCFEmptyString, which is defeated by constantStringAllocatorForDebugging
if ('\0' == *cStr) return kCFEmptyString;
#endif
if (constantStringTable == NULL) {
CFDictionaryKeyCallBacks constantStringCallBacks = {0, NULL, NULL, __cStrCopyDescription, __cStrEqual, __cStrHash};
CFDictionaryValueCallBacks constantStringValueCallBacks = kCFTypeDictionaryValueCallBacks;
constantStringValueCallBacks.equal = NULL; // So that we only find strings that are ==
CFMutableDictionaryRef table = CFDictionaryCreateMutable(kCFAllocatorSystemDefault, 0, &constantStringCallBacks, &constantStringValueCallBacks);
_CFDictionarySetCapacity(table, 2500); // avoid lots of rehashing
__CFLock(&_CFSTRLock);
if (constantStringTable == NULL) constantStringTable = table;
__CFUnlock(&_CFSTRLock);
if (constantStringTable != table) CFRelease(table);
}
__CFLock(&_CFSTRLock);
if ((result = (CFStringRef)CFDictionaryGetValue(constantStringTable, cStr))) {
__CFUnlock(&_CFSTRLock);
} else {
__CFUnlock(&_CFSTRLock);
{
char *key = NULL;
Boolean isASCII = true;
// Given this code path is rarer these days, OK to do this extra work to verify the strings
const char *tmp = cStr;
while (*tmp) {
if (*(tmp++) & 0x80) {
isASCII = false;
break;
}
}
if (!isASCII) {
CFMutableStringRef ms = CFStringCreateMutable(kCFAllocatorSystemDefault, 0);
tmp = cStr;
while (*tmp) {
CFStringAppendFormat(ms, NULL, (*tmp & 0x80) ? CFSTR("\\%3o") : CFSTR("%1c"), *tmp);
tmp++;
}
CFLog(kCFLogLevelWarning, CFSTR("WARNING: CFSTR(\"%@\") has non-7 bit chars, interpreting using MacOS Roman encoding for now, but this will change. Please eliminate usages of non-7 bit chars (including escaped characters above \\177 octal) in CFSTR()."), ms);
CFRelease(ms);
}
// Treat non-7 bit chars in CFSTR() as MacOSRoman, for compatibility
result = CFStringCreateWithCString(kCFAllocatorSystemDefault, cStr, kCFStringEncodingMacRoman);
if (result == NULL) {
CFLog(__kCFLogAssertion, CFSTR("Can't interpret CFSTR() as MacOS Roman, crashing"));
HALT;
}
Boolean isTaggedPointerString = CF_IS_OBJC(_kCFRuntimeIDCFString, result);
if (!isTaggedPointerString) {
if (__CFOASafe) __CFSetLastAllocationEventName((void *)result, "CFString (CFSTR)");
if (__CFStrIsEightBit(result)) key = (char *)__CFStrContents(result) + __CFStrSkipAnyLengthByte(result);
}
if (!key) { // Either the string is not 8-bit or it's a tagged pointer string
CFIndex keySize = strlen(cStr) + 1;
key = (char *)CFAllocatorAllocate(kCFAllocatorSystemDefault, keySize, 0);
if (__CFOASafe) __CFSetLastAllocationEventName((void *)key, "CFString (CFSTR key)");
strlcpy(key, cStr, keySize); // !!! We will leak this, if the string is removed from the table (or table is freed)
}
{
CFStringRef resultToBeReleased = result;
CFIndex count;
__CFLock(&_CFSTRLock);
count = CFDictionaryGetCount(constantStringTable);
CFDictionaryAddValue(constantStringTable, key, result);
if (CFDictionaryGetCount(constantStringTable) == count) { // add did nothing, someone already put it there
result = (CFStringRef)CFDictionaryGetValue(constantStringTable, key);
} else if (!isTaggedPointerString && !__CFRuntimeIsConstant(result)) {
// As of rdar://22175031 constant strings in CF are in read-only memory in the dyld shared cache
// So don't call this if __CFRuntimeIsConstant is true for cases like kCFEmptyString or strings in the string ROM
__CFRuntimeSetRC(result, 0);
}
__CFUnlock(&_CFSTRLock);
// This either eliminates the extra retain on the freshly created string, or frees it, if it was actually not inserted into the table
CFRelease(resultToBeReleased);
}
}
}
return result;
}
#endif
#if defined(DEBUG)
static Boolean __CFStrIsConstantString(CFStringRef str) {
Boolean found = false;
if (constantStringTable) {
__CFLock(&_CFSTRLock);
found = CFDictionaryContainsValue(constantStringTable, str);
__CFUnlock(&_CFSTRLock);
}
return found;
}
#endif
#if TARGET_OS_WIN32
void __CFStringCleanup (void) {
/* in case library is unloaded, release store for the constant string table */
if (constantStringTable != NULL) {
#if defined(DEBUG)
__CFConstantStringTableBeingFreed = true;
CFRelease(constantStringTable);
__CFConstantStringTableBeingFreed = false;
#else
CFRelease(constantStringTable);
#endif
constantStringTable = NULL;
}
}
#endif
// Can pass in NSString as replacement string
// Call with numRanges > 0, and incrementing ranges
static int __CFStringReplaceMultiple(CFMutableStringRef str, CFRange *ranges, CFIndex numRanges, CFStringRef replacement) {
if (!__CFStrIsMutable(str)) return _CFStringErrNotMutable;
int cnt;
CFStringRef copy = NULL;
if (replacement == str) copy = replacement = CFStringCreateCopy(kCFAllocatorSystemDefault, replacement); // Very special and hopefully rare case
CFIndex replacementLength = CFStringGetLength(replacement);
__CFStringChangeSizeMultiple(str, ranges, numRanges, replacementLength, (replacementLength > 0) && CFStrIsUnicode(replacement));
if (__CFStrIsUnicode(str)) {
UniChar *contents = (UniChar *)__CFStrContents(str);
UniChar *firstReplacement = contents + ranges[0].location;
// Extract the replacementString into the first location, then copy from there
CFStringGetCharacters(replacement, CFRangeMake(0, replacementLength), firstReplacement);
for (cnt = 1; cnt < numRanges; cnt++) {
// The ranges are in terms of the original string; so offset by the change in length due to insertion
contents += replacementLength - ranges[cnt - 1].length;
memmove(contents + ranges[cnt].location, firstReplacement, replacementLength * sizeof(UniChar));
}
} else {
uint8_t *contents = (uint8_t *)__CFStrContents(str);
uint8_t *firstReplacement = contents + ranges[0].location + __CFStrSkipAnyLengthByte(str);
// Extract the replacementString into the first location, then copy from there
CFStringGetBytes(replacement, CFRangeMake(0, replacementLength), __CFStringGetEightBitStringEncoding(), 0, false, firstReplacement, replacementLength, NULL);
contents += __CFStrSkipAnyLengthByte(str); // Now contents will simply track the location to insert next string into
for (cnt = 1; cnt < numRanges; cnt++) {
// The ranges are in terms of the original string; so offset by the change in length due to insertion
contents += replacementLength - ranges[cnt - 1].length;
memmove(contents + ranges[cnt].location, firstReplacement, replacementLength);
}
}
if (copy) CFRelease(copy);
return _CFStringErrNone;
}
// Can pass in NSString as replacement string
CF_INLINE void __CFStringReplace(CFMutableStringRef str, CFRange range, CFStringRef replacement) {
CFStringRef copy = NULL;
if (replacement == str) copy = replacement = (CFStringRef)CFStringCreateCopy(kCFAllocatorSystemDefault, replacement); // Very special and hopefully rare case
CFIndex replacementLength = CFStringGetLength(replacement);
__CFStringChangeSize(str, range, replacementLength, (replacementLength > 0) && CFStrIsUnicode(replacement));
if (__CFStrIsUnicode(str)) {
UniChar *contents = (UniChar *)__CFStrContents(str);
if (contents) {
CFStringGetCharacters(replacement, CFRangeMake(0, replacementLength), contents + range.location);
}
} else {
uint8_t *contents = (uint8_t *)__CFStrContents(str);
CFStringGetBytes(replacement, CFRangeMake(0, replacementLength), __CFStringGetEightBitStringEncoding(), 0, false, contents + range.location + __CFStrSkipAnyLengthByte(str), replacementLength, NULL);
}
if (copy) CFRelease(copy);
}
/* If client does not provide a minimum capacity
*/
#define DEFAULTMINCAPACITY 32
CF_INLINE CFMutableStringRef __CFStringCreateMutableFunnel(CFAllocatorRef alloc, CFIndex maxLength, _CFStringInlineContents inlineContents, Boolean isUnicode) {
CFMutableStringRef str;
Boolean hasExternalContentsAllocator = (inlineContents == __kCFNotInlineContentsCustomFree);
if (alloc == NULL) alloc = __CFGetDefaultAllocator();
// Note that if there is an externalContentsAllocator, then we also have the storage for the string allocator...
str = (CFMutableStringRef)_CFRuntimeCreateInstance(alloc, _kCFRuntimeIDCFString, sizeof(struct __notInlineMutable) - (hasExternalContentsAllocator ? 0 : sizeof(CFAllocatorRef)), NULL);
if (str) {
if (__CFOASafe) __CFSetLastAllocationEventName(str, "CFString (mutable)");
__CFStrSetInlineContents(str, inlineContents);
__CFStrSetUnicode(str, isUnicode);
__CFStrSetIsMutable(str);
str->variants.notInlineMutable.buffer = NULL;
__CFStrSetExplicitLength(str, 0);
str->variants.notInlineMutable.hasGap = str->variants.notInlineMutable.isFixedCapacity = str->variants.notInlineMutable.isExternalMutable = str->variants.notInlineMutable.capacityProvidedExternally = 0;
if (maxLength != 0) __CFStrSetIsFixed(str);
__CFStrSetDesiredCapacity(str, (maxLength == 0) ? DEFAULTMINCAPACITY : maxLength);
__CFStrSetCapacity(str, 0);
if (__CFStrHasContentsAllocator(str)) {
// contents allocator starts out as the string's own allocator
__CFStrSetContentsAllocator(str, alloc);
}
}
return str;
}
CFMutableStringRef CFStringCreateMutableWithExternalCharactersNoCopy(CFAllocatorRef alloc, UniChar *chars, CFIndex numChars, CFIndex capacity, CFAllocatorRef externalCharactersAllocator) {
_CFStringInlineContents contentsAllocationBits = externalCharactersAllocator ? ((externalCharactersAllocator == kCFAllocatorNull) ? __kCFNotInlineContentsNoFree : __kCFNotInlineContentsCustomFree) : __kCFNotInlineContentsDefaultFree;
CFMutableStringRef string = __CFStringCreateMutableFunnel(alloc, 0, contentsAllocationBits, true);
if (string) {
__CFStrSetIsExternalMutable(string);
if (__CFStrHasContentsAllocator(string)) {
CFAllocatorRef allocator = __CFStrContentsAllocator((CFMutableStringRef)string);
CFRelease(allocator);
// If externalCharactersAllocator == NULL, contentsAllocationBits is set to __kCFNotInlineContentsDefaultFree, which gives the string a default (non-custom) contents allocator.
// In that case, __CFStrHasContentsAllocator() returns FALSE, and we don't fall into here.
_CLANG_ANALYZER_ASSERT(externalCharactersAllocator != NULL);
__CFStrSetContentsAllocator(string, externalCharactersAllocator);
}
CFStringSetExternalCharactersNoCopy(string, chars, numChars, capacity);
}
return string;
}
CFMutableStringRef CFStringCreateMutable(CFAllocatorRef alloc, CFIndex maxLength) {
return __CFStringCreateMutableFunnel(alloc, maxLength, __kCFNotInlineContentsDefaultFree, false);
}
CFMutableStringRef CFStringCreateMutableCopy(CFAllocatorRef alloc, CFIndex maxLength, CFStringRef string) {
CFMutableStringRef newString;
// CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, CFMutableStringRef, (NSString *)string, mutableCopy);
__CFAssertIsString(string);
newString = CFStringCreateMutable(alloc, maxLength);
__CFStringReplace(newString, CFRangeMake(0, 0), string);
return newString;
}
CF_PRIVATE void _CFStrSetDesiredCapacity(CFMutableStringRef str, CFIndex len) {
__CFAssertIsStringAndMutable(str);
__CFStrSetDesiredCapacity(str, len);
}
/* This one is for CF
*/
CFIndex CFStringGetLength(CFStringRef str) {
CF_SWIFT_FUNCDISPATCHV(_kCFRuntimeIDCFString, CFIndex, (CFSwiftRef)str, NSString.length);
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, CFIndex, (NSString *)str, length);
__CFAssertIsString(str);
return __CFStrLength(str);
}
/* This one is for NSCFString; it does not ObjC dispatch or assertion check
*/
CFIndex _CFStringGetLength2(CFStringRef str) {
return __CFStrLength(str);
}
/* Guts of CFStringGetCharacterAtIndex(); called from the two functions below. Don't call it from elsewhere.
*/
CF_INLINE UniChar __CFStringGetCharacterAtIndexGuts(CFStringRef str, CFIndex idx, const uint8_t *contents) {
if (__CFStrIsEightBit(str)) {
contents += __CFStrSkipAnyLengthByte(str);
#if defined(DEBUG)
if (!__CFCharToUniCharFunc && (contents[idx] >= 128)) {
// Can't do log here, as it might be too early
fprintf(stderr, "Warning: CFStringGetCharacterAtIndex() attempted on CFString containing high bytes before properly initialized to do so\n");
}
#endif
return __CFCharToUniCharTable[contents[idx]];
}
return ((UniChar *)contents)[idx];
}
/* This one is for the CF API
*/
UniChar CFStringGetCharacterAtIndex(CFStringRef str, CFIndex idx) {
CF_SWIFT_FUNCDISPATCHV(_kCFRuntimeIDCFString, UniChar, (CFSwiftRef)str, NSString.characterAtIndex, idx);
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, UniChar, (NSString *)str, characterAtIndex:(NSUInteger)idx);
__CFAssertIsString(str);
__CFAssertIndexIsInStringBounds(str, idx);
return __CFStringGetCharacterAtIndexGuts(str, idx, (const uint8_t *)__CFStrContents(str));
}
/* This one is for NSCFString usage; it doesn't do ObjC dispatch; but it does do range check
*/
int _CFStringCheckAndGetCharacterAtIndex(CFStringRef str, CFIndex idx, UniChar *ch) {
const uint8_t *contents = (const uint8_t *)__CFStrContents(str);
if (idx < 0 || idx >= __CFStrLength2(str, contents)) { return _CFStringErrBounds; }
*ch = __CFStringGetCharacterAtIndexGuts(str, idx, contents);
return _CFStringErrNone;
}
/* Guts of CFStringGetCharacters(); called from the two functions below. Don't call it from elsewhere.
*/
CF_INLINE void __CFStringGetCharactersGuts(CFStringRef str, CFRange range, UniChar *buffer, const uint8_t *contents) {
if (__CFStrIsEightBit(str)) {
__CFStrConvertBytesToUnicode(((uint8_t *)contents) + (range.location + __CFStrSkipAnyLengthByte(str)), buffer, range.length);
} else {
const UniChar *uContents = ((UniChar *)contents) + range.location;
memmove(buffer, uContents, range.length * sizeof(UniChar));
}
}
/* This one is for the CF API
*/
void CFStringGetCharacters(CFStringRef str, CFRange range, UniChar *buffer) {
CF_SWIFT_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (CFSwiftRef)str, NSString.getCharacters, range, buffer);
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (NSString *)str, getCharacters:(unichar *)buffer range:NSMakeRange(range.location, range.length));
__CFAssertIsString(str);
__CFAssertRangeIsInStringBounds(str, range.location, range.length);
__CFStringGetCharactersGuts(str, range, buffer, (const uint8_t *)__CFStrContents(str));
}
/* This one is for NSCFString usage; it doesn't do ObjC dispatch; but it does do range check
*/
int _CFStringCheckAndGetCharacters(CFStringRef str, CFRange range, UniChar *buffer) {
const uint8_t *contents = (const uint8_t *)__CFStrContents(str);
if (range.location + range.length > __CFStrLength2(str, contents)) return _CFStringErrBounds;
__CFStringGetCharactersGuts(str, range, buffer, contents);
return _CFStringErrNone;
}
CFIndex CFStringGetBytes(CFStringRef str, CFRange range, CFStringEncoding encoding, uint8_t lossByte, Boolean isExternalRepresentation, uint8_t * _Nullable buffer, CFIndex maxBufLen, CFIndex *usedBufLen) {
#if DEPLOYMENT_RUNTIME_SWIFT
if (CF_IS_SWIFT(_kCFRuntimeIDCFString, str) && __CFSwiftBridge.NSString.__getBytes != NULL) {
return __CFSwiftBridge.NSString.__getBytes(str, encoding, range, buffer, maxBufLen, usedBufLen);
}
#endif
__CFAssertIsNotNegative(maxBufLen);
{
__CFAssertIsString(str);
__CFAssertRangeIsInStringBounds(str, range.location, range.length);
if (__CFStrIsEightBit(str) && ((__CFStringGetEightBitStringEncoding() == encoding) || (__CFStringGetEightBitStringEncoding() == kCFStringEncodingASCII && __CFStringEncodingIsSupersetOfASCII(encoding)))) { // Requested encoding is equal to the encoding in string
const unsigned char *contents = (const unsigned char *)__CFStrContents(str);
CFIndex cLength = range.length;
if (buffer) {
if (cLength > maxBufLen) cLength = maxBufLen;
memmove(buffer, contents + __CFStrSkipAnyLengthByte(str) + range.location, cLength);
}
if (usedBufLen) *usedBufLen = cLength;
return cLength;
}
}
return __CFStringEncodeByteStream(str, range.location, range.length, isExternalRepresentation, encoding, lossByte, buffer, maxBufLen, usedBufLen);
}
ConstStringPtr CFStringGetPascalStringPtr (CFStringRef str, CFStringEncoding encoding) {
if (!CF_IS_OBJC(_kCFRuntimeIDCFString, str) && !CF_IS_SWIFT(_kCFRuntimeIDCFString, str)) { /* ??? Hope the compiler optimizes this away if OBJC_MAPPINGS is not on */
__CFAssertIsString(str);
if (__CFStrHasLengthByte(str) && __CFStrIsEightBit(str) && ((__CFStringGetEightBitStringEncoding() == encoding) || (__CFStringGetEightBitStringEncoding() == kCFStringEncodingASCII && __CFStringEncodingIsSupersetOfASCII(encoding)))) { // Requested encoding is equal to the encoding in string || the contents is in ASCII
const uint8_t *contents = (const uint8_t *)__CFStrContents(str);
if (__CFStrHasExplicitLength(str) && (__CFStrLength2(str, contents) != (SInt32)(*contents))) return NULL; // Invalid length byte
return (ConstStringPtr)contents;
}
// ??? Also check for encoding = SystemEncoding and perhaps bytes are all ASCII?
}
return NULL;
}
static inline const char * _CFStringGetCStringPtrInternal(CFStringRef str, CFStringEncoding encoding, Boolean requiresNullTermination, Boolean requiresBridgingCheck) {
if (encoding != __CFStringGetEightBitStringEncoding() && (kCFStringEncodingASCII != __CFStringGetEightBitStringEncoding() || !__CFStringEncodingIsSupersetOfASCII(encoding))) return NULL;
// ??? Also check for encoding = SystemEncoding and perhaps bytes are all ASCII?
if (str == NULL) return NULL; // Should really just crash, but for compatibility... see <rdar://problem/12340248>
if (requiresBridgingCheck) {
CF_SWIFT_FUNCDISPATCHV(_kCFRuntimeIDCFString, const char *, (CFSwiftRef)str, NSString._fastCStringContents, requiresNullTermination);
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, const char *, (NSString *)str, _fastCStringContents:requiresNullTermination);
}
__CFAssertIsString(str);
if ((!requiresNullTermination && __CFStrIsEightBit(str)) || __CFStrHasNullByte(str)) {
// Note: this is called a lot, 27000 times to open a small xcode project with one file open.
// Of these uses about 1500 are for cStrings/utf8strings.
return (const char *)__CFStrContents(str) + __CFStrSkipAnyLengthByte(str);
} else {
return NULL;
}
}
const char * _CFNonObjCStringGetCStringPtr(CFStringRef str, CFStringEncoding encoding, Boolean requiresNullTermination) {
return _CFStringGetCStringPtrInternal(str, encoding, requiresNullTermination, false);
}
const char * CFStringGetCStringPtr(CFStringRef str, CFStringEncoding encoding) {
return _CFStringGetCStringPtrInternal(str, encoding, true, true);
}
const UniChar *CFStringGetCharactersPtr(CFStringRef str) {
CF_SWIFT_FUNCDISPATCHV(_kCFRuntimeIDCFString, const UniChar *, (CFSwiftRef)str, NSString._fastCharacterContents);
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, const UniChar *, (NSString *)str, _fastCharacterContents);
__CFAssertIsString(str);
if (__CFStrIsUnicode(str)) return (const UniChar *)__CFStrContents(str);
return NULL;
}
Boolean CFStringGetPascalString(CFStringRef str, Str255 buffer, CFIndex bufferSize, CFStringEncoding encoding) {
CFIndex length;
CFIndex usedLen;
__CFAssertIsNotNegative(bufferSize);
if (bufferSize < 1) return false;
if (CF_IS_OBJC(_kCFRuntimeIDCFString, str) || CF_IS_SWIFT(_kCFRuntimeIDCFString, str)) { /* ??? Hope the compiler optimizes this away if OBJC_MAPPINGS is not on */
length = CFStringGetLength(str);
if (!__CFCanUseLengthByte(length)) return false; // Can't fit into pstring
} else {
const uint8_t *contents;
__CFAssertIsString(str);
contents = (const uint8_t *)__CFStrContents(str);
length = __CFStrLength2(str, contents);
if (!__CFCanUseLengthByte(length)) return false; // Can't fit into pstring
if (__CFStrIsEightBit(str) && ((__CFStringGetEightBitStringEncoding() == encoding) || (__CFStringGetEightBitStringEncoding() == kCFStringEncodingASCII && __CFStringEncodingIsSupersetOfASCII(encoding)))) { // Requested encoding is equal to the encoding in string
if (length >= bufferSize) return false;
memmove((void*)(1 + (const char*)buffer), (__CFStrSkipAnyLengthByte(str) + contents), length);
*buffer = (unsigned char)length;
return true;
}
}
if (__CFStringEncodeByteStream(str, 0, length, false, encoding, false, (UInt8 *)(1 + (uint8_t *)buffer), bufferSize - 1, &usedLen) != length) {
#if defined(DEBUG)
if (bufferSize > 0) {
strlcpy((char *)buffer + 1, CONVERSIONFAILURESTR, bufferSize - 1);
buffer[0] = (unsigned char)((CFIndex)sizeof(CONVERSIONFAILURESTR) < (bufferSize - 1) ? (CFIndex)sizeof(CONVERSIONFAILURESTR) : (bufferSize - 1));
}
#else
if (bufferSize > 0) buffer[0] = 0;
#endif
return false;
}
*buffer = (unsigned char)usedLen;
return true;
}
Boolean CFStringGetCString(CFStringRef str, char *buffer, CFIndex bufferSize, CFStringEncoding encoding) {
const uint8_t *contents;
CFIndex len;
__CFAssertIsNotNegative(bufferSize);
if (bufferSize < 1) return false;
CF_SWIFT_FUNCDISPATCHV(_kCFRuntimeIDCFString, Boolean, (CFSwiftRef)str, NSString._getCString, buffer, bufferSize - 1, encoding);
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, Boolean, (NSString *)str, _getCString:buffer maxLength:(NSUInteger)bufferSize - 1 encoding:encoding);
__CFAssertIsString(str);
contents = (const uint8_t *)__CFStrContents(str);
len = __CFStrLength2(str, contents);
if (__CFStrIsEightBit(str) && ((__CFStringGetEightBitStringEncoding() == encoding) || (__CFStringGetEightBitStringEncoding() == kCFStringEncodingASCII && __CFStringEncodingIsSupersetOfASCII(encoding)))) { // Requested encoding is equal to the encoding in string
if (len >= bufferSize) {
buffer[0] = 0;
return false;
}
memmove(buffer, contents + __CFStrSkipAnyLengthByte(str), len);
buffer[len] = 0;
return true;
} else {
CFIndex usedLen;
if (__CFStringEncodeByteStream(str, 0, len, false, encoding, false, (unsigned char*) buffer, bufferSize - 1, &usedLen) == len) {
buffer[usedLen] = '\0';
return true;
} else {
#if defined(DEBUG)
strlcpy(buffer, CONVERSIONFAILURESTR, bufferSize);
#else
if (bufferSize > 0) buffer[0] = 0;
#endif
return false;
}
}
}
extern Boolean __CFLocaleGetDoesNotRequireSpecialCaseHandling(struct __CFLocale *locale);
extern void __CFLocaleSetDoesNotRequireSpecialCaseHandling(struct __CFLocale *locale);
// Returns the language code if the given locale is one of the "special" languages that
// requires special handing during case mapping:
// - "az": Azerbaijani
// - "lt": Lithuanian
// - "tr": Turkish
// - "nl": Dutch
// - "el": Greek
// For all other locales such as en_US, this function returs NULL.
// See `CFUniCharMapCaseTo`
// See https://www.unicode.org/Public/UNIDATA/SpecialCasing.txt
static const char *_CFStrGetSpecialCaseHandlingLanguageIdentifierForLocale(CFLocaleRef locale, bool collatorOnly) {
CFStringRef localeID;
const char *langID = NULL;
static const void *lastLocale = NULL;
static const char *lastLangID = NULL;
static CFLock_t lock = CFLockInit;
if (__CFLocaleGetDoesNotRequireSpecialCaseHandling((struct __CFLocale *)locale)) return NULL;
__CFLock(&lock);
if ((NULL != lastLocale) && (lastLocale == locale)) {
__CFUnlock(&lock);
return lastLangID;
}
__CFUnlock(&lock);
if (collatorOnly) {
localeID = (CFStringRef)CFLocaleGetValue(locale, __kCFLocaleCollatorID);
} else {
localeID = (CFStringRef)CFLocaleGetIdentifier(locale);
}
CFIndex length = CFStringGetLength(localeID);
if (length > 1) {
uint8_t buffer[2];
const uint8_t *contents = (const uint8_t *)CFStringGetCStringPtr(localeID, kCFStringEncodingUTF8);
if (!contents) {
if (2 == CFStringGetBytes(localeID, CFRangeMake(0,2), kCFStringEncodingUTF8, 0, false, buffer, sizeof(buffer), NULL)) contents = buffer;
}
if (contents) {
const char *string = (const char *)contents;
if (!strncmp(string, "az", 2)) { // Azerbaijani
langID = "az";
} else if (!strncmp(string, "lt", 2)) { // Lithuanian
langID = "lt";
} else if (!strncmp(string, "tr", 2)) { // Turkish
langID = "tr";
} else if (!strncmp(string, "nl", 2)) { // Dutch
langID = "nl";
} else if (!strncmp(string, "el", 2)) { // Greek
langID = "el";
}
}
}
if (langID == NULL) __CFLocaleSetDoesNotRequireSpecialCaseHandling((struct __CFLocale *)locale);
__CFLock(&lock);
lastLocale = locale;
lastLangID = langID;
__CFUnlock(&lock);
return langID;
}
CF_INLINE bool _CFCanUseLocale(CFLocaleRef locale) {
if (locale) {
return true;
}
return false;
}
#define MAX_CASE_MAPPING_BUF (8)
#define WHITE_SPACE_CHARACTER (0x0020)
#define ZERO_WIDTH_JOINER (0x200D)
#define COMBINING_GRAPHEME_JOINER (0x034F)
// Hangul ranges
#define HANGUL_CHOSEONG_START (0x1100)
#define HANGUL_CHOSEONG_END (0x115F)
#define HANGUL_JUNGSEONG_START (0x1160)
#define HANGUL_JUNGSEONG_END (0x11A2)
#define HANGUL_JONGSEONG_START (0x11A8)
#define HANGUL_JONGSEONG_END (0x11F9)
#define HANGUL_SYLLABLE_START (0xAC00)
#define HANGUL_SYLLABLE_END (0xD7AF)
// Returns the length of characters filled into outCharacters. If no change, returns 0. maxBufLen shoule be at least 8
static CFIndex __CFStringFoldCharacterClusterAtIndex(UTF32Char character, CFStringInlineBuffer *buffer, CFIndex index, CFOptionFlags flags, const uint8_t *langCode, UTF32Char *outCharacters, CFIndex maxBufferLength, CFIndex *consumedLength, bool *insufficientBufferSpace) {
CFIndex filledLength = 0, currentIndex = index;
if (0 != character) {
UTF16Char lowSurrogate;
CFIndex planeNo = (character >> 16);
bool isTurkikCapitalI = false;
static const uint8_t *decompBMP = NULL;
static const uint8_t *graphemeBMP = NULL;
if (NULL == decompBMP) {
decompBMP = CFUniCharGetBitmapPtrForPlane(kCFUniCharCanonicalDecomposableCharacterSet, 0);
graphemeBMP = CFUniCharGetBitmapPtrForPlane(kCFUniCharGraphemeExtendCharacterSet, 0);
}
currentIndex += ((character > 0xFFFF) ? 2 : 1);
if ((character < 0x0080) && ((NULL == langCode) || (character != 'I'))) { // ASCII
if ((flags & kCFCompareCaseInsensitive) && (character >= 'A') && (character <= 'Z')) {
character += ('a' - 'A');
*outCharacters = character;
filledLength = 1;
}
} else {
// do width-insensitive mapping
if ((flags & kCFCompareWidthInsensitive) && (character >= 0xFF00) && (character <= 0xFFEF)) {
(void)CFUniCharCompatibilityDecompose(&character, 1, 1);
*outCharacters = character;
filledLength = 1;
}
// map surrogates
if ((0 == planeNo) && CFUniCharIsSurrogateHighCharacter(character) && CFUniCharIsSurrogateLowCharacter((lowSurrogate = CFStringGetCharacterFromInlineBuffer(buffer, currentIndex)))) {
character = CFUniCharGetLongCharacterForSurrogatePair(character, lowSurrogate);
++currentIndex;
planeNo = (character >> 16);
}
// decompose
if (flags & (kCFCompareDiacriticInsensitive|kCFCompareNonliteral)) {
if (CFUniCharIsMemberOfBitmap(character, ((0 == planeNo) ? decompBMP : CFUniCharGetBitmapPtrForPlane(kCFUniCharCanonicalDecomposableCharacterSet, planeNo)))) {
UTF32Char original = character;
filledLength = CFUniCharDecomposeCharacter(character, outCharacters, maxBufferLength);
character = *outCharacters;
if ((flags & kCFCompareDiacriticInsensitive) && (character < 0x0510)) {
filledLength = 1; // reset if Roman, Greek, Cyrillic
} else if (0 == (flags & kCFCompareNonliteral)) {
character = original;
filledLength = 0;
} else if (filledLength == 0 && NULL != insufficientBufferSpace) {
*insufficientBufferSpace = true;
}
}
}
// fold case
if (flags & kCFCompareCaseInsensitive) {
const uint8_t *nonBaseBitmap;
bool filterNonBase = (((flags & kCFCompareDiacriticInsensitive) && (character < 0x0510)) ? true : false);
static const uint8_t *lowerBMP = NULL;
static const uint8_t *caseFoldBMP = NULL;
if (NULL == lowerBMP) {
lowerBMP = CFUniCharGetBitmapPtrForPlane(kCFUniCharHasNonSelfLowercaseCharacterSet, 0);
caseFoldBMP = CFUniCharGetBitmapPtrForPlane(kCFUniCharHasNonSelfCaseFoldingCharacterSet, 0);
}
if ((NULL != langCode) && ('I' == character) && ((0 == strcmp((const char *)langCode, "tr")) || (0 == strcmp((const char *)langCode, "az")))) { // do Turkik special-casing
if (filledLength > 1) {
if (0x0307 == outCharacters[1]) {
if (--filledLength > 1) memmove((outCharacters + 1), (outCharacters + 2), sizeof(UTF32Char) * (filledLength - 1));
character = *outCharacters = 'i';
isTurkikCapitalI = true;
}
} else if (0x0307 == CFStringGetCharacterFromInlineBuffer(buffer, currentIndex)) {
character = *outCharacters = 'i';
filledLength = 1;
++currentIndex;
isTurkikCapitalI = true;
}
}
if (!isTurkikCapitalI && (CFUniCharIsMemberOfBitmap(character, ((0 == planeNo) ? lowerBMP : CFUniCharGetBitmapPtrForPlane(kCFUniCharHasNonSelfLowercaseCharacterSet, planeNo))) || CFUniCharIsMemberOfBitmap(character, ((0 == planeNo) ? caseFoldBMP : CFUniCharGetBitmapPtrForPlane(kCFUniCharHasNonSelfCaseFoldingCharacterSet, planeNo))))) {
UTF16Char caseFoldBuffer[MAX_CASE_MAPPING_BUF];
const UTF16Char *bufferP = caseFoldBuffer, *bufferLimit;
UTF32Char *outCharactersP = outCharacters;
uint32_t bufferLength = CFUniCharMapCaseTo(character, caseFoldBuffer, MAX_CASE_MAPPING_BUF, kCFUniCharCaseFold, 0, langCode);
bufferLimit = bufferP + bufferLength;
if (filledLength > 0) --filledLength; // decrement filledLength (will add back later)
// make space for casefold characters
if ((filledLength > 0) && (bufferLength > 1)) {
CFIndex totalScalerLength = 0;
while (bufferP < bufferLimit) {
if (CFUniCharIsSurrogateHighCharacter(*(bufferP++)) && (bufferP < bufferLimit) && CFUniCharIsSurrogateLowCharacter(*bufferP)) ++bufferP;
++totalScalerLength;
}
memmove(outCharacters + totalScalerLength, outCharacters + 1, filledLength * sizeof(UTF32Char));
bufferP = caseFoldBuffer;
}
// fill
while (bufferP < bufferLimit) {
character = *(bufferP++);
if (CFUniCharIsSurrogateHighCharacter(character) && (bufferP < bufferLimit) && CFUniCharIsSurrogateLowCharacter(*bufferP)) {
character = CFUniCharGetLongCharacterForSurrogatePair(character, *(bufferP++));
nonBaseBitmap = CFUniCharGetBitmapPtrForPlane(kCFUniCharGraphemeExtendCharacterSet, (character >> 16));
} else {
nonBaseBitmap = graphemeBMP;
}
if (!filterNonBase || !CFUniCharIsMemberOfBitmap(character, nonBaseBitmap)) {
*(outCharactersP++) = character;
++filledLength;
}
}
}
}
}
// collect following combining marks
if (flags & (kCFCompareDiacriticInsensitive|kCFCompareNonliteral)) {
const uint8_t *nonBaseBitmap;
const uint8_t *decompBitmap;
bool doFill = (((flags & kCFCompareDiacriticInsensitive) && (character < 0x0510)) ? false : true);
if (0 == filledLength) {
*outCharacters = character; // filledLength will be updated below on demand
if (doFill) { // check if really needs to fill
UTF32Char nonBaseCharacter = CFStringGetCharacterFromInlineBuffer(buffer, currentIndex);
if (CFUniCharIsSurrogateHighCharacter(nonBaseCharacter) && CFUniCharIsSurrogateLowCharacter((lowSurrogate = CFStringGetCharacterFromInlineBuffer(buffer, currentIndex + 1)))) {
nonBaseCharacter = CFUniCharGetLongCharacterForSurrogatePair(nonBaseCharacter, lowSurrogate);
nonBaseBitmap = CFUniCharGetBitmapPtrForPlane(kCFUniCharGraphemeExtendCharacterSet, (nonBaseCharacter >> 16));
decompBitmap = CFUniCharGetBitmapPtrForPlane(kCFUniCharCanonicalDecomposableCharacterSet, (nonBaseCharacter >> 16));
} else {
nonBaseBitmap = graphemeBMP;
decompBitmap = decompBMP;
}
if (CFUniCharIsMemberOfBitmap(nonBaseCharacter, nonBaseBitmap)) {
filledLength = 1; // For the base character
if ((0 == (flags & kCFCompareDiacriticInsensitive)) || (nonBaseCharacter > 0x050F)) {
if (CFUniCharIsMemberOfBitmap(nonBaseCharacter, decompBitmap)) {
CFIndex decomposedLength = CFUniCharDecomposeCharacter(nonBaseCharacter, &(outCharacters[filledLength]), maxBufferLength - filledLength);
filledLength += decomposedLength;
if (decomposedLength == 0 && NULL != insufficientBufferSpace) {
*insufficientBufferSpace = true;
}
} else {
outCharacters[filledLength++] = nonBaseCharacter;
}
}
currentIndex += ((nonBaseBitmap == graphemeBMP) ? 1 : 2);
} else {
doFill = false;
}
}
}
bool endedCharacterCluster = false;
while (filledLength < maxBufferLength) { // do the rest
character = CFStringGetCharacterFromInlineBuffer(buffer, currentIndex);
if (CFUniCharIsSurrogateHighCharacter(character) && CFUniCharIsSurrogateLowCharacter((lowSurrogate = CFStringGetCharacterFromInlineBuffer(buffer, currentIndex + 1)))) {
character = CFUniCharGetLongCharacterForSurrogatePair(character, lowSurrogate);
nonBaseBitmap = CFUniCharGetBitmapPtrForPlane(kCFUniCharGraphemeExtendCharacterSet, (character >> 16));
decompBitmap = CFUniCharGetBitmapPtrForPlane(kCFUniCharCanonicalDecomposableCharacterSet, (character >> 16));
} else {
nonBaseBitmap = graphemeBMP;
decompBitmap = decompBMP;
}
if (isTurkikCapitalI) {
isTurkikCapitalI = false;
} else if (CFUniCharIsMemberOfBitmap(character, nonBaseBitmap)) {
if (doFill) {
if (CFUniCharIsMemberOfBitmap(character, decompBitmap)) {
CFIndex currentLength = CFUniCharDecomposeCharacter(character, &(outCharacters[filledLength]), maxBufferLength - filledLength);
if (0 == currentLength) break; // didn't fit
filledLength += currentLength;
} else {
outCharacters[filledLength++] = character;
}
} else if (0 == filledLength) {
filledLength = 1; // For the base character
}
currentIndex += ((nonBaseBitmap == graphemeBMP) ? 1 : 2);
} else {
endedCharacterCluster = true;
break;
}
}
if (!endedCharacterCluster && NULL != insufficientBufferSpace) {
*insufficientBufferSpace = true;
}
if (filledLength > 1) {
UTF32Char *sortCharactersLimit = outCharacters + filledLength;
UTF32Char *sortCharacters = sortCharactersLimit - 1;
while ((outCharacters < sortCharacters) && CFUniCharIsMemberOfBitmap(*sortCharacters, ((*sortCharacters < 0x10000) ? graphemeBMP : CFUniCharGetBitmapPtrForPlane(kCFUniCharGraphemeExtendCharacterSet, (*sortCharacters >> 16))))) --sortCharacters;
if ((sortCharactersLimit - sortCharacters) > 1) CFUniCharPrioritySort(sortCharacters, (sortCharactersLimit - sortCharacters)); // priority sort
}
}
}
if ((filledLength > 0) && (NULL != consumedLength)) *consumedLength = (currentIndex - index);
return filledLength;
}
static bool __CFStringFillCharacterSetInlineBuffer(CFCharacterSetInlineBuffer *buffer, CFStringCompareFlags compareOptions) {
if (0 != (compareOptions & kCFCompareIgnoreNonAlphanumeric)) {
static CFCharacterSetRef nonAlnumChars = NULL;
if (NULL == nonAlnumChars) {
CFMutableCharacterSetRef cset = CFCharacterSetCreateMutableCopy(kCFAllocatorSystemDefault, CFCharacterSetGetPredefined(kCFCharacterSetAlphaNumeric));
CFCharacterSetInvert(cset);
#pragma GCC diagnostic push
#pragma GCC diagnostic ignored "-Wdeprecated"
if (!OSAtomicCompareAndSwapPtrBarrier(NULL, cset, (void **)&nonAlnumChars)) CFRelease(cset);
#pragma GCC diagnostic pop
}
CFCharacterSetInitInlineBuffer(nonAlnumChars, buffer);
return true;
}
return false;
}
#define kCFStringStackBufferLength (__kCFStringInlineBufferLength)
static const u_char __ASCII_LOWERCASE_TABLE[] = {
0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x09, 0x0A, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, 0x18, 0x19, 0x1A, 0x1B, 0x1C, 0x1D, 0x1E, 0x1F,
0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27, 0x28, 0x29, 0x2A, 0x2B, 0x2C, 0x2D, 0x2E, 0x2F,
0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, 0x38, 0x39, 0x3A, 0x3B, 0x3C, 0x3D, 0x3E, 0x3F,
0x40,
/* The A-Z range should become a-z */
0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F,
0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7A,
0x5B, 0x5C, 0x5D, 0x5E, 0x5F,
0x60, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F,
0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7A, 0x7B, 0x7C, 0x7D, 0x7E, 0x7F,
};
// This function is an implementation of strncasecmp_l that does not stop comparing at embedded null bytes
// We are not calling to LibC APIs such as tolower_l here because calling to those APIs (as compared to using a lookup table) introduced significant performance regressions
CF_INLINE int __CFStringCompareASCIICaseInsensitive(const u_char *str1, const u_char *str2, size_t n) {
if (n != 0) {
do {
u_char a = __ASCII_LOWERCASE_TABLE[*str1++];
u_char b = __ASCII_LOWERCASE_TABLE[*str2++];
if (a != b) {
return a - b;
}
} while (--n != 0);
}
return 0;
}
CFComparisonResult CFStringCompareWithOptionsAndLocale(CFStringRef string, CFStringRef string2, CFRange rangeToCompare, CFStringCompareFlags compareOptions, CFLocaleRef locale) {
/* No objc dispatch needed here since CFStringInlineBuffer works with both CFString and NSString */
UTF32Char strBuf1[kCFStringStackBufferLength];
UTF32Char strBuf2[kCFStringStackBufferLength];
CFStringInlineBuffer inlineBuf1, inlineBuf2;
UTF32Char str1Char, str2Char;
CFIndex str1UsedLen, str2UsedLen;
CFIndex str1Index = 0, str2Index = 0, strBuf1Index = 0, strBuf2Index = 0, strBuf1Len = 0, strBuf2Len = 0;
CFIndex str1LocalizedIndex = 0, str2LocalizedIndex = 0;
CFIndex forcedIndex1 = 0, forcedIndex2 = 0;
CFIndex str2Len = CFStringGetLength(string2);
bool caseInsensitive = ((compareOptions & kCFCompareCaseInsensitive) ? true : false);
bool diacriticsInsensitive = ((compareOptions & kCFCompareDiacriticInsensitive) ? true : false);
bool equalityOptions = ((compareOptions & (kCFCompareCaseInsensitive|kCFCompareNonliteral|kCFCompareDiacriticInsensitive|kCFCompareWidthInsensitive)) ? true : false);
bool numerically = ((compareOptions & kCFCompareNumerically) ? true : false);
bool forceOrdering = ((compareOptions & kCFCompareForcedOrdering) ? true : false);
const uint8_t *langCode;
CFComparisonResult compareResult = kCFCompareEqualTo;
UTF16Char otherChar;
Boolean freeLocale = false;
CFCharacterSetInlineBuffer *ignoredChars = NULL;
CFCharacterSetInlineBuffer csetBuffer;
bool numericEquivalence = false;
if ((compareOptions & kCFCompareLocalized) && (NULL == locale)) {
locale = CFLocaleCopyCurrent();
freeLocale = true;
}
langCode = ((NULL == locale) ? NULL : (const uint8_t *)_CFStrGetSpecialCaseHandlingLanguageIdentifierForLocale(locale, true));
if (__CFStringFillCharacterSetInlineBuffer(&csetBuffer, compareOptions)) {
ignoredChars = &csetBuffer;
equalityOptions = true;
}
if ((NULL == locale) && (NULL == ignoredChars) && !numerically) { // could do binary comp (be careful when adding new flags)
CFStringEncoding eightBitEncoding = __CFStringGetEightBitStringEncoding();
const uint8_t *str1Bytes = (const uint8_t *)_CFStringGetCStringPtrInternal(string, eightBitEncoding, false, true);
const uint8_t *str2Bytes = (const uint8_t *)_CFStringGetCStringPtrInternal(string2, eightBitEncoding, false, true);
CFIndex factor = sizeof(uint8_t);
if ((NULL != str1Bytes) && (NULL != str2Bytes)) {
compareOptions &= ~kCFCompareNonliteral; // remove non-literal
if ((kCFStringEncodingASCII == eightBitEncoding) && (false == forceOrdering)) {
if (caseInsensitive) {
// Here we call our own __CFStringCompareASCIICaseInsensitive rather than strncasecmp_l to continue comparing after embedded null bytes
int cmpResult = __CFStringCompareASCIICaseInsensitive(str1Bytes + rangeToCompare.location, str2Bytes, __CFMin(rangeToCompare.length, str2Len));
if (0 == cmpResult) cmpResult = rangeToCompare.length - str2Len;
return ((0 == cmpResult) ? kCFCompareEqualTo : ((cmpResult < 0) ? kCFCompareLessThan : kCFCompareGreaterThan));
}
} else if (caseInsensitive || diacriticsInsensitive) {
CFIndex limitLength = __CFMin(rangeToCompare.length, str2Len);
str1Bytes += rangeToCompare.location;
while (str1Index < limitLength) {
str1Char = str1Bytes[str1Index];
str2Char = str2Bytes[str1Index];
if (str1Char != str2Char) {
if ((str1Char < 0x80) && (str2Char < 0x80)) {
if (forceOrdering && (kCFCompareEqualTo == compareResult) && (str1Char != str2Char)) compareResult = ((str1Char < str2Char) ? kCFCompareLessThan : kCFCompareGreaterThan);
if (caseInsensitive) {
if ((str1Char >= 'A') && (str1Char <= 'Z')) str1Char += ('a' - 'A');
if ((str2Char >= 'A') && (str2Char <= 'Z')) str2Char += ('a' - 'A');
}
if (str1Char != str2Char) return ((str1Char < str2Char) ? kCFCompareLessThan : kCFCompareGreaterThan);
} else {
str1Bytes = NULL;
break;
}
}
++str1Index;
}
str2Index = str1Index;
if (str1Index == limitLength) {
int cmpResult = rangeToCompare.length - str2Len;
return ((0 == cmpResult) ? compareResult : ((cmpResult < 0) ? kCFCompareLessThan : kCFCompareGreaterThan));
}
}
} else if (!equalityOptions && (NULL == str1Bytes) && (NULL == str2Bytes)) {
str1Bytes = (const uint8_t *)CFStringGetCharactersPtr(string);
str2Bytes = (const uint8_t *)CFStringGetCharactersPtr(string2);
factor = sizeof(UTF16Char);
#if __LITTLE_ENDIAN__
if ((NULL != str1Bytes) && (NULL != str2Bytes)) { // we cannot use memcmp
const UTF16Char *str1 = ((const UTF16Char *)str1Bytes) + rangeToCompare.location;
const UTF16Char *str1Limit = str1 + __CFMin(rangeToCompare.length, str2Len);
const UTF16Char *str2 = (const UTF16Char *)str2Bytes;
CFIndex cmpResult = 0;
while ((0 == cmpResult) && (str1 < str1Limit)) cmpResult = (CFIndex)*(str1++) - (CFIndex)*(str2++);
if (0 == cmpResult) cmpResult = rangeToCompare.length - str2Len;
return ((0 == cmpResult) ? kCFCompareEqualTo : ((cmpResult < 0) ? kCFCompareLessThan : kCFCompareGreaterThan));
}
#endif /* __LITTLE_ENDIAN__ */
}
if ((NULL != str1Bytes) && (NULL != str2Bytes)) {
int cmpResult = memcmp(str1Bytes + (rangeToCompare.location * factor), str2Bytes, __CFMin(rangeToCompare.length, str2Len) * factor);
if (0 == cmpResult) cmpResult = rangeToCompare.length - str2Len;
return ((0 == cmpResult) ? kCFCompareEqualTo : ((cmpResult < 0) ? kCFCompareLessThan : kCFCompareGreaterThan));
}
}
const uint8_t *graphemeBMP = CFUniCharGetBitmapPtrForPlane(kCFUniCharGraphemeExtendCharacterSet, 0);
_CFStringInitInlineBufferInternal(string, &inlineBuf1, rangeToCompare, true);
_CFStringInitInlineBufferInternal(string2, &inlineBuf2, CFRangeMake(0, str2Len), true);
if (NULL != locale) {
str1LocalizedIndex = str1Index;
str2LocalizedIndex = str2Index;
// We temporarily disable kCFCompareDiacriticInsensitive for SL <rdar://problem/6767096>. Should be revisited in NMOS <rdar://problem/7003830>
if (forceOrdering) {
diacriticsInsensitive = false;
compareOptions &= ~kCFCompareDiacriticInsensitive;
}
}
CFIndex preventStr1FoldingUntil = 0, preventStr2FoldingUntil = 0;
while ((str1Index < rangeToCompare.length) && (str2Index < str2Len)) {
if (strBuf1Len == 0) {
str1Char = CFStringGetCharacterFromInlineBuffer(&inlineBuf1, str1Index);
if (caseInsensitive && (str1Char >= 'A') && (str1Char <= 'Z') && ((NULL == langCode) || (str1Char != 'I')) && ((false == forceOrdering) || (kCFCompareEqualTo != compareResult))) str1Char += ('a' - 'A');
str1UsedLen = 1;
} else {
str1Char = strBuf1[strBuf1Index++];
}
if (strBuf2Len == 0) {
str2Char = CFStringGetCharacterFromInlineBuffer(&inlineBuf2, str2Index);
if (caseInsensitive && (str2Char >= 'A') && (str2Char <= 'Z') && ((NULL == langCode) || (str2Char != 'I')) && ((false == forceOrdering) || (kCFCompareEqualTo != compareResult))) str2Char += ('a' - 'A');
str2UsedLen = 1;
} else {
str2Char = strBuf2[strBuf2Index++];
}
if (numerically && ((0 == strBuf1Len) && (str1Char <= '9') && (str1Char >= '0')) && ((0 == strBuf2Len) && (str2Char <= '9') && (str2Char >= '0'))) { // If both are not ASCII digits, then don't do numerical comparison here
uint64_t intValue1 = 0, intValue2 = 0; // !!! Doesn't work if numbers are > max uint64_t
CFIndex str1NumRangeIndex = str1Index;
CFIndex str2NumRangeIndex = str2Index;
do {
intValue1 = (intValue1 * 10) + (str1Char - '0');
str1Char = CFStringGetCharacterFromInlineBuffer(&inlineBuf1, ++str1Index);
} while ((str1Char <= '9') && (str1Char >= '0'));
do {
intValue2 = intValue2 * 10 + (str2Char - '0');
str2Char = CFStringGetCharacterFromInlineBuffer(&inlineBuf2, ++str2Index);
} while ((str2Char <= '9') && (str2Char >= '0'));
if (intValue1 == intValue2) {
if (forceOrdering && (kCFCompareEqualTo == compareResult) && ((str1Index - str1NumRangeIndex) != (str2Index - str2NumRangeIndex))) {
compareResult = (((str1Index - str1NumRangeIndex) < (str2Index - str2NumRangeIndex)) ? kCFCompareLessThan : kCFCompareGreaterThan);
numericEquivalence = true;
forcedIndex1 = str1NumRangeIndex;
forcedIndex2 = str2NumRangeIndex;
}
continue;
} else if (intValue1 < intValue2) {
if (freeLocale && locale) {
CFRelease(locale);
}
return kCFCompareLessThan;
} else {
if (freeLocale && locale) {
CFRelease(locale);
}
return kCFCompareGreaterThan;
}
}
if (str1Char != str2Char) {
if (!equalityOptions) {
compareResult = ((NULL == locale) ? ((str1Char < str2Char) ? kCFCompareLessThan : kCFCompareGreaterThan) : _CFCompareStringsWithLocale(&inlineBuf1, CFRangeMake(str1Index, rangeToCompare.length - str1Index), &inlineBuf2, CFRangeMake(str2Index, str2Len - str2Index), compareOptions, locale));
if (freeLocale && locale) {
CFRelease(locale);
}
return compareResult;
}
if (forceOrdering && (kCFCompareEqualTo == compareResult)) {
compareResult = ((str1Char < str2Char) ? kCFCompareLessThan : kCFCompareGreaterThan);
forcedIndex1 = str1LocalizedIndex;
forcedIndex2 = str2LocalizedIndex;
}
if ((str1Char < 0x80) && (str2Char < 0x80) && (NULL == ignoredChars)) {
if (NULL != locale) {
compareResult = _CFCompareStringsWithLocale(&inlineBuf1, CFRangeMake(str1Index, rangeToCompare.length - str1Index), &inlineBuf2, CFRangeMake(str2Index, str2Len - str2Index), compareOptions, locale);
if (freeLocale && locale) {
CFRelease(locale);
}
return compareResult;
} else if (!caseInsensitive) {
if (freeLocale && locale) {
CFRelease(locale);
}
return ((str1Char < str2Char) ? kCFCompareLessThan : kCFCompareGreaterThan);
}
}
if (CFUniCharIsSurrogateHighCharacter(str1Char) && CFUniCharIsSurrogateLowCharacter((otherChar = CFStringGetCharacterFromInlineBuffer(&inlineBuf1, str1Index + 1)))) {
str1Char = CFUniCharGetLongCharacterForSurrogatePair(str1Char, otherChar);
str1UsedLen = 2;
}
if (CFUniCharIsSurrogateHighCharacter(str2Char) && CFUniCharIsSurrogateLowCharacter((otherChar = CFStringGetCharacterFromInlineBuffer(&inlineBuf2, str2Index + 1)))) {
str2Char = CFUniCharGetLongCharacterForSurrogatePair(str2Char, otherChar);
str2UsedLen = 2;
}
if (NULL != ignoredChars) {
if (CFCharacterSetInlineBufferIsLongCharacterMember(ignoredChars, str1Char)) {
if ((strBuf1Len > 0) && (strBuf1Index == strBuf1Len)) strBuf1Len = 0;
if (strBuf1Len == 0) str1Index += str1UsedLen;
if (strBuf2Len > 0) --strBuf2Index;
continue;
}
if (CFCharacterSetInlineBufferIsLongCharacterMember(ignoredChars, str2Char)) {
if ((strBuf2Len > 0) && (strBuf2Index == strBuf2Len)) strBuf2Len = 0;
if (strBuf2Len == 0) str2Index += str2UsedLen;
if (strBuf1Len > 0) -- strBuf1Index;
continue;
}
}
if (diacriticsInsensitive && (str1Index > 0)) {
bool str1Skip = false;
bool str2Skip = false;
if ((0 == strBuf1Len) && CFUniCharIsMemberOfBitmap(str1Char, ((str1Char < 0x10000) ? graphemeBMP : CFUniCharGetBitmapPtrForPlane(kCFUniCharGraphemeExtendCharacterSet, (str1Char >> 16))))) {
str1Char = str2Char;
str1Skip = true;
}
if ((0 == strBuf2Len) && CFUniCharIsMemberOfBitmap(str2Char, ((str2Char < 0x10000) ? graphemeBMP : CFUniCharGetBitmapPtrForPlane(kCFUniCharGraphemeExtendCharacterSet, (str2Char >> 16))))) {
str2Char = str1Char;
str2Skip = true;
}
if (str1Skip != str2Skip) {
if (str1Skip) str2Index -= str2UsedLen;
if (str2Skip) str1Index -= str1UsedLen;
}
}
if (str1Char != str2Char) {
if (0 == strBuf1Len && (preventStr1FoldingUntil == 0 || preventStr1FoldingUntil == str1Index)) {
preventStr1FoldingUntil = 0;
bool insufficientBuffer = false;
strBuf1Len = __CFStringFoldCharacterClusterAtIndex(str1Char, &inlineBuf1, str1Index, compareOptions, langCode, strBuf1, kCFStringStackBufferLength, &str1UsedLen, &insufficientBuffer);
if (strBuf1Len > 0) {
str1Char = *strBuf1;
strBuf1Index = 1;
}
if (insufficientBuffer) {
// We have a character cluster larger than our maximum folding size. This is likely a malformed string, so do not fold the remainder of this cluster
CFRange currentCluster = CFStringGetRangeOfCharacterClusterAtIndex(string, str1Index, kCFStringGraphemeCluster);
preventStr1FoldingUntil = currentCluster.location + currentCluster.length;
}
}
if ((0 == strBuf1Len) && (0 < strBuf2Len)) {
compareResult = ((NULL == locale) ? ((str1Char < str2Char) ? kCFCompareLessThan : kCFCompareGreaterThan) : _CFCompareStringsWithLocale(&inlineBuf1, CFRangeMake(str1LocalizedIndex, rangeToCompare.length - str1LocalizedIndex), &inlineBuf2, CFRangeMake(str2LocalizedIndex, str2Len - str2LocalizedIndex), compareOptions, locale));
if (freeLocale && locale) {
CFRelease(locale);
}
return compareResult;
}
if ((0 == strBuf2Len) && ((0 == strBuf1Len) || (str1Char != str2Char)) && (preventStr2FoldingUntil == 0 || preventStr2FoldingUntil == str2Index)) {
preventStr2FoldingUntil = 0;
bool insufficientBuffer = false;
strBuf2Len = __CFStringFoldCharacterClusterAtIndex(str2Char, &inlineBuf2, str2Index, compareOptions, langCode, strBuf2, kCFStringStackBufferLength, &str2UsedLen, &insufficientBuffer);
if (strBuf2Len > 0) {
str2Char = *strBuf2;
strBuf2Index = 1;
}
if ((0 == strBuf2Len) || (str1Char != str2Char)) {
compareResult = ((NULL == locale) ? ((str1Char < str2Char) ? kCFCompareLessThan : kCFCompareGreaterThan) : _CFCompareStringsWithLocale(&inlineBuf1, CFRangeMake(str1LocalizedIndex, rangeToCompare.length - str1LocalizedIndex), &inlineBuf2, CFRangeMake(str2LocalizedIndex, str2Len - str2LocalizedIndex), compareOptions, locale));
if (freeLocale && locale) {
CFRelease(locale);
}
return compareResult;
}
if (insufficientBuffer) {
// We have a character cluster larger than our maximum folding size. This is likely a malformed string, so do not fold the remainder of this cluster
CFRange currentCluster = CFStringGetRangeOfCharacterClusterAtIndex(string2, str2Index, kCFStringGraphemeCluster);
preventStr2FoldingUntil = currentCluster.location + currentCluster.length;
}
}
}
if ((strBuf1Len > 0) && (strBuf2Len > 0)) {
while ((strBuf1Index < strBuf1Len) && (strBuf2Index < strBuf2Len)) {
if (strBuf1[strBuf1Index] != strBuf2[strBuf2Index]) break;
++strBuf1Index; ++strBuf2Index;
}
if ((strBuf1Index < strBuf1Len) && (strBuf2Index < strBuf2Len)) {
CFComparisonResult res = ((NULL == locale) ? ((strBuf1[strBuf1Index] < strBuf2[strBuf2Index]) ? kCFCompareLessThan : kCFCompareGreaterThan) : _CFCompareStringsWithLocale(&inlineBuf1, CFRangeMake(str1LocalizedIndex, rangeToCompare.length - str1LocalizedIndex), &inlineBuf2, CFRangeMake(str2LocalizedIndex, str2Len - str2LocalizedIndex), compareOptions, locale));
if (freeLocale && locale) {
CFRelease(locale);
}
return res;
}
}
}
if ((strBuf1Len > 0) && (strBuf1Index == strBuf1Len)) strBuf1Len = 0;
if ((strBuf2Len > 0) && (strBuf2Index == strBuf2Len)) strBuf2Len = 0;
if (strBuf1Len == 0) str1Index += str1UsedLen;
if (strBuf2Len == 0) str2Index += str2UsedLen;
if ((strBuf1Len == 0) && (strBuf2Len == 0)) {
str1LocalizedIndex = str1Index;
str2LocalizedIndex = str2Index;
}
}
if (diacriticsInsensitive || (NULL != ignoredChars)) {
while (str1Index < rangeToCompare.length) {
str1Char = CFStringGetCharacterFromInlineBuffer(&inlineBuf1, str1Index);
if ((str1Char < 0x80) && (NULL == ignoredChars)) break; // found ASCII
if (CFUniCharIsSurrogateHighCharacter(str1Char) && CFUniCharIsSurrogateLowCharacter((otherChar = CFStringGetCharacterFromInlineBuffer(&inlineBuf1, str1Index + 1)))) str1Char = CFUniCharGetLongCharacterForSurrogatePair(str1Char, otherChar);
if ((!diacriticsInsensitive || !CFUniCharIsMemberOfBitmap(str1Char, ((str1Char < 0x10000) ? graphemeBMP : CFUniCharGetBitmapPtrForPlane(kCFUniCharGraphemeExtendCharacterSet, (str1Char >> 16))))) && ((NULL == ignoredChars) || !CFCharacterSetInlineBufferIsLongCharacterMember(ignoredChars, str1Char))) break;
str1Index += ((str1Char < 0x10000) ? 1 : 2);
}
while (str2Index < str2Len) {
str2Char = CFStringGetCharacterFromInlineBuffer(&inlineBuf2, str2Index);
if ((str2Char < 0x80) && (NULL == ignoredChars)) break; // found ASCII
if (CFUniCharIsSurrogateHighCharacter(str2Char) && CFUniCharIsSurrogateLowCharacter((otherChar = CFStringGetCharacterFromInlineBuffer(&inlineBuf2, str2Index + 1)))) str2Char = CFUniCharGetLongCharacterForSurrogatePair(str2Char, otherChar);
if ((!diacriticsInsensitive || !CFUniCharIsMemberOfBitmap(str2Char, ((str2Char < 0x10000) ? graphemeBMP : CFUniCharGetBitmapPtrForPlane(kCFUniCharGraphemeExtendCharacterSet, (str2Char >> 16))))) && ((NULL == ignoredChars) || !CFCharacterSetInlineBufferIsLongCharacterMember(ignoredChars, str2Char))) break;
str2Index += ((str2Char < 0x10000) ? 1 : 2);
}
}
// Need to recalc localized result here for forced ordering, ICU cannot do numericEquivalence
if (!numericEquivalence && (NULL != locale) && (kCFCompareEqualTo != compareResult) && (str1Index == rangeToCompare.length) && (str2Index == str2Len)) compareResult = _CFCompareStringsWithLocale(&inlineBuf1, CFRangeMake(forcedIndex1, rangeToCompare.length - forcedIndex1), &inlineBuf2, CFRangeMake(forcedIndex2, str2Len - forcedIndex2), compareOptions, locale);
if (freeLocale && locale) {
CFRelease(locale);
}
return ((str1Index < rangeToCompare.length) ? kCFCompareGreaterThan : ((str2Index < str2Len) ? kCFCompareLessThan : compareResult));
}
CFComparisonResult CFStringCompareWithOptions(CFStringRef string, CFStringRef string2, CFRange rangeToCompare, CFStringCompareFlags compareOptions) { return CFStringCompareWithOptionsAndLocale(string, string2, rangeToCompare, compareOptions, NULL); }
CFComparisonResult CFStringCompare(CFStringRef string, CFStringRef str2, CFStringCompareFlags options) {
return CFStringCompareWithOptions(string, str2, CFRangeMake(0, CFStringGetLength(string)), options);
}
Boolean CFStringFindWithOptionsAndLocale(CFStringRef string, CFStringRef stringToFind, CFRange rangeToSearch, CFStringCompareFlags compareOptions, CFLocaleRef locale, CFRange *result) {
/* No objc dispatch needed here since CFStringInlineBuffer works with both CFString and NSString */
CFIndex findStrLen = CFStringGetLength(stringToFind);
Boolean didFind = false;
bool lengthVariants = ((compareOptions & (kCFCompareCaseInsensitive|kCFCompareNonliteral|kCFCompareDiacriticInsensitive)) ? true : false);
CFCharacterSetInlineBuffer *ignoredChars = NULL;
CFCharacterSetInlineBuffer csetBuffer;
if (__CFStringFillCharacterSetInlineBuffer(&csetBuffer, compareOptions)) {
ignoredChars = &csetBuffer;
lengthVariants = true;
}
if ((findStrLen > 0) && (rangeToSearch.length > 0) && ((findStrLen <= rangeToSearch.length) || lengthVariants)) {
UTF32Char strBuf1[kCFStringStackBufferLength];
UTF32Char strBuf2[kCFStringStackBufferLength];
CFStringInlineBuffer inlineBuf1, inlineBuf2;
UTF32Char str1Char = 0, str2Char = 0;
CFStringEncoding eightBitEncoding = __CFStringGetEightBitStringEncoding();
const uint8_t *str1Bytes = (const uint8_t *)_CFStringGetCStringPtrInternal(string, eightBitEncoding, false, true);
const uint8_t *str2Bytes = (const uint8_t *)_CFStringGetCStringPtrInternal(stringToFind, eightBitEncoding, false, true);
const UTF32Char *characters, *charactersLimit;
const uint8_t *langCode = NULL;
CFIndex fromLoc, toLoc;
CFIndex str1Index, str2Index;
CFIndex strBuf1Len, strBuf2Len;
CFIndex maxStr1Index = (rangeToSearch.location + rangeToSearch.length);
CFIndex lastStr1FoldIndex = 0, lastStr1FoldLength = 0, lastStr1FoldUsed = 0, preventStr1FoldingUntil = 0;
CFIndex lastStr2FoldIndex = 0, lastStr2FoldLength = 0, lastStr2FoldUsed = 0, preventStr2FoldingUntil = 0;
bool equalityOptions = ((lengthVariants || (compareOptions & kCFCompareWidthInsensitive)) ? true : false);
bool caseInsensitive = ((compareOptions & kCFCompareCaseInsensitive) ? true : false);
bool forwardAnchor = ((kCFCompareAnchored == (compareOptions & (kCFCompareBackwards|kCFCompareAnchored))) ? true : false);
bool backwardAnchor = (((kCFCompareBackwards|kCFCompareAnchored) == (compareOptions & (kCFCompareBackwards|kCFCompareAnchored))) ? true : false);
int8_t delta;
if (NULL == locale) {
if (compareOptions & kCFCompareLocalized) {
CFLocaleRef currentLocale = CFLocaleCopyCurrent();
langCode = (const uint8_t *)_CFStrGetSpecialCaseHandlingLanguageIdentifierForLocale(currentLocale, true);
CFRelease(currentLocale);
}
} else {
langCode = (const uint8_t *)_CFStrGetSpecialCaseHandlingLanguageIdentifierForLocale(locale, true);
}
_CFStringInitInlineBufferInternal(string, &inlineBuf1, CFRangeMake(0, rangeToSearch.location + rangeToSearch.length), true);
_CFStringInitInlineBufferInternal(stringToFind, &inlineBuf2, CFRangeMake(0, findStrLen), true);
if (compareOptions & kCFCompareBackwards) {
fromLoc = rangeToSearch.location + rangeToSearch.length - (lengthVariants ? 1 : findStrLen);
toLoc = (((compareOptions & kCFCompareAnchored) && !lengthVariants) ? fromLoc : rangeToSearch.location);
} else {
fromLoc = rangeToSearch.location;
toLoc = ((compareOptions & kCFCompareAnchored) ? fromLoc : rangeToSearch.location + rangeToSearch.length - (lengthVariants ? 1 : findStrLen));
}
delta = ((fromLoc <= toLoc) ? 1 : -1);
if ((NULL != str1Bytes) && (NULL != str2Bytes)) {
uint8_t str1Byte, str2Byte;
while (1) {
str1Index = fromLoc;
str2Index = 0;
while ((str1Index < maxStr1Index) && (str2Index < findStrLen)) {
str1Byte = str1Bytes[str1Index];
str2Byte = str2Bytes[str2Index];
if (str1Byte != str2Byte) {
if (equalityOptions) {
if ((str1Byte < 0x80) && ((NULL == langCode) || ('I' != str1Byte))) {
if (caseInsensitive && (str1Byte >= 'A') && (str1Byte <= 'Z')) str1Byte += ('a' - 'A');
*strBuf1 = str1Byte;
strBuf1Len = 1;
} else {
str1Char = CFStringGetCharacterFromInlineBuffer(&inlineBuf1, str1Index);
strBuf1Len = __CFStringFoldCharacterClusterAtIndex(str1Char, &inlineBuf1, str1Index, compareOptions, langCode, strBuf1, kCFStringStackBufferLength, NULL, NULL);
if (1 > strBuf1Len) {
*strBuf1 = str1Char;
strBuf1Len = 1;
}
}
if ((NULL != ignoredChars) && (forwardAnchor || (str1Index != fromLoc)) && CFCharacterSetInlineBufferIsLongCharacterMember(ignoredChars, ((str1Byte < 0x80) ? str1Byte : str1Char))) {
++str1Index;
continue;
}
if ((str2Byte < 0x80) && ((NULL == langCode) || ('I' != str2Byte))) {
if (caseInsensitive && (str2Byte >= 'A') && (str2Byte <= 'Z')) str2Byte += ('a' - 'A');
*strBuf2 = str2Byte;
strBuf2Len = 1;
} else {
str2Char = CFStringGetCharacterFromInlineBuffer(&inlineBuf2, str2Index);
strBuf2Len = __CFStringFoldCharacterClusterAtIndex(str2Char, &inlineBuf2, str2Index, compareOptions, langCode, strBuf2, kCFStringStackBufferLength, NULL, NULL);
if (1 > strBuf2Len) {
*strBuf2 = str2Char;
strBuf2Len = 1;
}
}
if ((NULL != ignoredChars) && CFCharacterSetInlineBufferIsLongCharacterMember(ignoredChars, ((str2Byte < 0x80) ? str2Byte : str2Char))) {
++str2Index;
continue;
}
if ((1 == strBuf1Len) && (1 == strBuf2Len)) { // normal case
if (*strBuf1 != *strBuf2) break;
} else {
CFIndex delta;
if (!caseInsensitive && (strBuf1Len != strBuf2Len)) break;
if (memcmp(strBuf1, strBuf2, sizeof(UTF32Char) * __CFMin(strBuf1Len, strBuf2Len))) break;
if (strBuf1Len < strBuf2Len) {
delta = strBuf2Len - strBuf1Len;
if ((str1Index + strBuf1Len + delta) > maxStr1Index) break;
characters = &(strBuf2[strBuf1Len]);
charactersLimit = characters + delta;
while (characters < charactersLimit) {
strBuf1Len = __CFStringFoldCharacterClusterAtIndex(CFStringGetCharacterFromInlineBuffer(&inlineBuf1, str1Index + 1), &inlineBuf1, str1Index + 1, compareOptions, langCode, strBuf1, kCFStringStackBufferLength, NULL, NULL);
if ((strBuf1Len > 0) || (*characters != *strBuf1)) break;
++characters; ++str1Index;
}
if (characters < charactersLimit) break;
} else if (strBuf2Len < strBuf1Len) {
delta = strBuf1Len - strBuf2Len;
if ((str2Index + strBuf2Len + delta) > findStrLen) break;
characters = &(strBuf1[strBuf2Len]);
charactersLimit = characters + delta;
while (characters < charactersLimit) {
strBuf2Len = __CFStringFoldCharacterClusterAtIndex(CFStringGetCharacterFromInlineBuffer(&inlineBuf2, str1Index + 1), &inlineBuf2, str2Index + 1, compareOptions, langCode, strBuf2, kCFStringStackBufferLength, NULL, NULL);
if ((strBuf2Len > 0) || (*characters != *strBuf2)) break;
++characters; ++str2Index;
}
if (characters < charactersLimit) break;
}
}
} else {
break;
}
}
++str1Index; ++str2Index;
}
if ((NULL != ignoredChars) && (str1Index == maxStr1Index) && (str2Index < findStrLen)) { // Process the stringToFind tail
while (str2Index < findStrLen) {
str2Char = CFStringGetCharacterFromInlineBuffer(&inlineBuf2, str2Index);
if (!CFCharacterSetInlineBufferIsLongCharacterMember(ignoredChars, str2Char)) break;
++str2Index;
}
}
if (str2Index == findStrLen) {
if ((NULL != ignoredChars) && backwardAnchor && (str1Index < maxStr1Index)) { // Process the anchor tail
while (str1Index < maxStr1Index) {
str1Char = CFStringGetCharacterFromInlineBuffer(&inlineBuf1, str1Index);
if (!CFCharacterSetInlineBufferIsLongCharacterMember(ignoredChars, str1Char)) break;
++str1Index;
}
}
if (!backwardAnchor || (str1Index == maxStr1Index)) {
didFind = true;
if (NULL != result) *result = CFRangeMake(fromLoc, str1Index - fromLoc);
}
break;
}
if (fromLoc == toLoc) break;
fromLoc += delta;
}
} else if (equalityOptions) {
UTF16Char otherChar;
CFIndex str1UsedLen, str2UsedLen, strBuf1Index = 0, strBuf2Index = 0;
bool diacriticsInsensitive = ((compareOptions & kCFCompareDiacriticInsensitive) ? true : false);
const uint8_t *graphemeBMP = CFUniCharGetBitmapPtrForPlane(kCFUniCharGraphemeExtendCharacterSet, 0);
const uint8_t *combClassBMP = (const uint8_t *)CFUniCharGetUnicodePropertyDataForPlane(kCFUniCharCombiningProperty, 0);
while (1) {
str1Index = fromLoc;
str2Index = 0;
lastStr2FoldIndex = lastStr2FoldUsed = lastStr2FoldLength = preventStr2FoldingUntil = 0;
strBuf1Len = strBuf2Len = 0;
while (str2Index < findStrLen) {
if (strBuf1Len == 0) {
str1Char = CFStringGetCharacterFromInlineBuffer(&inlineBuf1, str1Index);
if (caseInsensitive && (str1Char >= 'A') && (str1Char <= 'Z') && ((NULL == langCode) || (str1Char != 'I'))) str1Char += ('a' - 'A');
str1UsedLen = 1;
} else {
str1Char = strBuf1[strBuf1Index++];
}
if (strBuf2Len == 0) {
str2Char = CFStringGetCharacterFromInlineBuffer(&inlineBuf2, str2Index);
if (caseInsensitive && (str2Char >= 'A') && (str2Char <= 'Z') && ((NULL == langCode) || (str2Char != 'I'))) str2Char += ('a' - 'A');
str2UsedLen = 1;
} else {
str2Char = strBuf2[strBuf2Index++];
}
if (str1Char != str2Char) {
if ((str1Char < 0x80) && (str2Char < 0x80) && (NULL == ignoredChars) && ((NULL == langCode) || !caseInsensitive)) break;
if (CFUniCharIsSurrogateHighCharacter(str1Char) && CFUniCharIsSurrogateLowCharacter((otherChar = CFStringGetCharacterFromInlineBuffer(&inlineBuf1, str1Index + 1)))) {
str1Char = CFUniCharGetLongCharacterForSurrogatePair(str1Char, otherChar);
str1UsedLen = 2;
}
if (CFUniCharIsSurrogateHighCharacter(str2Char) && CFUniCharIsSurrogateLowCharacter((otherChar = CFStringGetCharacterFromInlineBuffer(&inlineBuf2, str2Index + 1)))) {
str2Char = CFUniCharGetLongCharacterForSurrogatePair(str2Char, otherChar);
str2UsedLen = 2;
}
if (NULL != ignoredChars) {
if ((forwardAnchor || (str1Index != fromLoc)) && (str1Index < maxStr1Index) && CFCharacterSetInlineBufferIsLongCharacterMember(ignoredChars, str1Char)) {
if ((strBuf1Len > 0) && (strBuf1Index == strBuf1Len)) strBuf1Len = 0;
if (strBuf1Len == 0) str1Index += str1UsedLen;
if (strBuf2Len > 0) --strBuf2Index;
continue;
}
if (CFCharacterSetInlineBufferIsLongCharacterMember(ignoredChars, str2Char)) {
if ((strBuf2Len > 0) && (strBuf2Index == strBuf2Len)) strBuf2Len = 0;
if (strBuf2Len == 0) str2Index += str2UsedLen;
if (strBuf1Len > 0) -- strBuf1Index;
continue;
}
}
if (diacriticsInsensitive && (str1Index > fromLoc)) {
bool str1Skip = false;
bool str2Skip = false;
if ((0 == strBuf1Len) && CFUniCharIsMemberOfBitmap(str1Char, ((str1Char < 0x10000) ? graphemeBMP : CFUniCharGetBitmapPtrForPlane(kCFUniCharGraphemeExtendCharacterSet, (str1Char >> 16))))) {
str1Char = str2Char;
str1Skip = true;
}
if ((0 == strBuf2Len) && CFUniCharIsMemberOfBitmap(str2Char, ((str2Char < 0x10000) ? graphemeBMP : CFUniCharGetBitmapPtrForPlane(kCFUniCharGraphemeExtendCharacterSet, (str2Char >> 16))))) {
str2Char = str1Char;
str2Skip = true;
}
if (str1Skip != str2Skip) {
if (str1Skip) str2Index -= str2UsedLen;
if (str2Skip) str1Index -= str1UsedLen;
}
}
if (str1Char != str2Char) {
if (0 == strBuf1Len && (preventStr1FoldingUntil == 0 || preventStr1FoldingUntil == str1Index)) {
preventStr1FoldingUntil = 0;
// Check `strBuf1Index` isn't going to be larger than `strBuf1`
strBuf1Index = str1Index - lastStr1FoldIndex + 1;
if (lastStr1FoldLength > 0 && str1Index >= lastStr1FoldIndex && str1Index < lastStr1FoldIndex + lastStr1FoldUsed && strBuf1Index < lastStr1FoldLength) {
strBuf1Len = lastStr1FoldLength;
str1Char = strBuf1[strBuf1Index - 1];
} else {
bool insufficientBuffer = false;
strBuf1Len = __CFStringFoldCharacterClusterAtIndex(str1Char, &inlineBuf1, str1Index, compareOptions, langCode, strBuf1, kCFStringStackBufferLength, &str1UsedLen, &insufficientBuffer);
if (strBuf1Len > 0) {
str1Char = *strBuf1;
strBuf1Index = 1;
}
lastStr1FoldLength = strBuf1Len;
lastStr1FoldIndex = str1Index;
lastStr1FoldUsed = str1UsedLen;
if (insufficientBuffer) {
// We have a character cluster larger than our maximum folding size. This is likely a malformed string, so do not fold the remainder of this cluster
CFRange currentCluster = CFStringGetRangeOfCharacterClusterAtIndex(string, str1Index, kCFStringGraphemeCluster);
if (delta == 1) {
preventStr1FoldingUntil = currentCluster.location + currentCluster.length;
} else {
preventStr1FoldingUntil = MAX(currentCluster.location - 1, 1);
}
}
}
}
if ((0 == strBuf1Len) && (0 < strBuf2Len)) break;
if ((0 == strBuf2Len) && ((0 == strBuf1Len) || (str1Char != str2Char))) {
if (preventStr2FoldingUntil == 0 || preventStr2FoldingUntil == str2Index) {
preventStr2FoldingUntil = 0;
// Check `strBuf2Index` isn't going to be larger than `strBuf2`
strBuf2Index = str2Index - lastStr2FoldIndex + 1;
if (lastStr2FoldLength > 0 && str2Index >= lastStr2FoldIndex && str2Index < lastStr2FoldIndex + lastStr2FoldUsed && strBuf2Index < lastStr2FoldLength) {
strBuf2Len = lastStr2FoldLength;
str2Char = strBuf2[strBuf2Index - 1];
if (str1Char != str2Char) break;
} else {
bool insufficientBuffer = false;
strBuf2Len = __CFStringFoldCharacterClusterAtIndex(str2Char, &inlineBuf2, str2Index, compareOptions, langCode, strBuf2, kCFStringStackBufferLength, &str2UsedLen, &insufficientBuffer);
lastStr2FoldLength = strBuf2Len;
lastStr2FoldIndex = str2Index;
lastStr2FoldUsed = str2UsedLen;
if (insufficientBuffer) {
// We have a character cluster larger than our maximum folding size. This is likely a malformed string, so do not fold the remainder of this cluster
CFRange currentCluster = CFStringGetRangeOfCharacterClusterAtIndex(stringToFind, str2Index, kCFStringGraphemeCluster);
preventStr2FoldingUntil = currentCluster.location + currentCluster.length;
}
if ((0 == strBuf2Len) || (str1Char != *strBuf2)) break;
strBuf2Index = 1;
}
} else {
if (str1Char != str2Char) break;
}
}
}
if ((strBuf1Len > 0) && (strBuf2Len > 0)) {
while ((strBuf1Index < strBuf1Len) && (strBuf2Index < strBuf2Len)) {
if (strBuf1[strBuf1Index] != strBuf2[strBuf2Index]) break;
++strBuf1Index; ++strBuf2Index;
}
if ((strBuf1Index < strBuf1Len) && (strBuf2Index < strBuf2Len)) break;
}
}
if ((strBuf1Len > 0) && (strBuf1Index == strBuf1Len)) strBuf1Len = 0;
if ((strBuf2Len > 0) && (strBuf2Index == strBuf2Len)) strBuf2Len = 0;
if (strBuf1Len == 0) str1Index += str1UsedLen;
if (strBuf2Len == 0) str2Index += str2UsedLen;
}
if ((NULL != ignoredChars) && (str1Index == maxStr1Index) && (str2Index < findStrLen)) { // Process the stringToFind tail
while (str2Index < findStrLen) {
str2Char = CFStringGetCharacterFromInlineBuffer(&inlineBuf2, str2Index);
if (CFUniCharIsSurrogateHighCharacter(str2Char) && CFUniCharIsSurrogateLowCharacter((otherChar = CFStringGetCharacterFromInlineBuffer(&inlineBuf2, str2Index + 1)))) {
str2Char = CFUniCharGetLongCharacterForSurrogatePair(str2Char, otherChar);
}
if (!CFCharacterSetInlineBufferIsLongCharacterMember(ignoredChars, str2Char)) break;
str2Index += ((str2Char < 0x10000) ? 1 : 2);
}
}
if (str2Index == findStrLen) {
bool match = true;
if (strBuf1Len > 0) {
match = false;
if (diacriticsInsensitive && (strBuf1[0] < 0x0510)) {
while (strBuf1Index < strBuf1Len) {
if (!CFUniCharIsMemberOfBitmap(strBuf1[strBuf1Index], ((strBuf1[strBuf1Index] < 0x10000) ? graphemeBMP : CFUniCharGetBitmapPtrForPlane(kCFUniCharCanonicalDecomposableCharacterSet, (strBuf1[strBuf1Index] >> 16))))) break;
++strBuf1Index;
}
if (strBuf1Index == strBuf1Len) {
str1Index += str1UsedLen;
match = true;
}
}
}
if (match && (compareOptions & (kCFCompareDiacriticInsensitive|kCFCompareNonliteral)) && (str1Index < maxStr1Index)) {
const uint8_t *nonBaseBitmap;
str1Char = CFStringGetCharacterFromInlineBuffer(&inlineBuf1, str1Index);
if (CFUniCharIsSurrogateHighCharacter(str1Char) && CFUniCharIsSurrogateLowCharacter((otherChar = CFStringGetCharacterFromInlineBuffer(&inlineBuf1, str1Index + 1)))) {
str1Char = CFUniCharGetLongCharacterForSurrogatePair(str1Char, otherChar);
nonBaseBitmap = CFUniCharGetBitmapPtrForPlane(kCFUniCharGraphemeExtendCharacterSet, (str1Char >> 16));
} else {
nonBaseBitmap = graphemeBMP;
}
if (CFUniCharIsMemberOfBitmap(str1Char, nonBaseBitmap)) {
if (diacriticsInsensitive) {
if (str1Char < 0x10000) {
CFIndex index = str1Index;
do {
str1Char = CFStringGetCharacterFromInlineBuffer(&inlineBuf1, --index);
// <rdar://problem/36547482> Possible lost optimization in CFString
} while (/* CFUniCharIsMemberOfBitmap(str1Char, graphemeBMP), */(rangeToSearch.location < index));
if (str1Char < 0x0510) {
while (++str1Index < maxStr1Index) if (!CFUniCharIsMemberOfBitmap(CFStringGetCharacterFromInlineBuffer(&inlineBuf1, str1Index), graphemeBMP)) break;
}
}
} else {
match = false;
}
} else if (!diacriticsInsensitive) {
otherChar = CFStringGetCharacterFromInlineBuffer(&inlineBuf1, str1Index - 1);
// this is assuming viramas are only in BMP ???
if ((str1Char == COMBINING_GRAPHEME_JOINER) || (otherChar == COMBINING_GRAPHEME_JOINER) || (otherChar == ZERO_WIDTH_JOINER) || ((otherChar >= HANGUL_CHOSEONG_START) && (otherChar <= HANGUL_JONGSEONG_END)) || (CFUniCharGetCombiningPropertyForCharacter(otherChar, combClassBMP) == 9)) {
CFRange clusterRange = CFStringGetRangeOfCharacterClusterAtIndex(string, str1Index - 1, kCFStringGraphemeCluster);
if (str1Index < (clusterRange.location + clusterRange.length)) match = false;
}
}
}
if (match) {
if ((NULL != ignoredChars) && backwardAnchor && (str1Index < maxStr1Index)) { // Process the anchor tail
while (str1Index < maxStr1Index) {
str1Char = CFStringGetCharacterFromInlineBuffer(&inlineBuf1, str1Index);
if (CFUniCharIsSurrogateHighCharacter(str1Char) && CFUniCharIsSurrogateLowCharacter((otherChar = CFStringGetCharacterFromInlineBuffer(&inlineBuf1, str1Index + 1)))) {
str1Char = CFUniCharGetLongCharacterForSurrogatePair(str1Char, otherChar);
}
if (!CFCharacterSetInlineBufferIsLongCharacterMember(ignoredChars, str1Char)) break;
str1Index += ((str1Char < 0x10000) ? 1 : 2);
}
}
if (!backwardAnchor || (str1Index == maxStr1Index)) {
didFind = true;
if (NULL != result) *result = CFRangeMake(fromLoc, str1Index - fromLoc);
}
break;
}
}
if (fromLoc == toLoc) break;
fromLoc += delta;
}
} else {
while (1) {
str1Index = fromLoc;
str2Index = 0;
while (str2Index < findStrLen) {
if (CFStringGetCharacterFromInlineBuffer(&inlineBuf1, str1Index) != CFStringGetCharacterFromInlineBuffer(&inlineBuf2, str2Index)) break;
++str1Index; ++str2Index;
}
if (str2Index == findStrLen) {
didFind = true;
if (NULL != result) *result = CFRangeMake(fromLoc, findStrLen);
break;
}
if (fromLoc == toLoc) break;
fromLoc += delta;
}
}
}
return didFind;
}
Boolean CFStringFindWithOptions(CFStringRef string, CFStringRef stringToFind, CFRange rangeToSearch, CFStringCompareFlags compareOptions, CFRange *result) { return CFStringFindWithOptionsAndLocale(string, stringToFind, rangeToSearch, compareOptions, NULL, result); }
// Functions to deal with special arrays of CFRange, CFDataRef, created by CFStringCreateArrayWithFindResults()
static const void *__rangeRetain(CFAllocatorRef allocator, const void *ptr) {
CFRetain(*(CFDataRef *)((uint8_t *)ptr + sizeof(CFRange)));
return ptr;
}
static void __rangeRelease(CFAllocatorRef allocator, const void *ptr) {
CFRelease(*(CFDataRef *)((uint8_t *)ptr + sizeof(CFRange)));
}
static CFStringRef __rangeCopyDescription(const void *ptr) {
CFRange range = *(CFRange *)ptr;
return CFStringCreateWithFormat(kCFAllocatorSystemDefault, NULL, CFSTR("{%ld, %ld}"), (long)range.location, (long)range.length);
}
static Boolean __rangeEqual(const void *ptr1, const void *ptr2) {
CFRange range1 = *(CFRange *)ptr1;
CFRange range2 = *(CFRange *)ptr2;
return (range1.location == range2.location) && (range1.length == range2.length);
}
CFArrayRef CFStringCreateArrayWithFindResults(CFAllocatorRef alloc, CFStringRef string, CFStringRef stringToFind, CFRange rangeToSearch, CFStringCompareFlags compareOptions) {
CFRange foundRange;
Boolean backwards = ((compareOptions & kCFCompareBackwards) != 0);
UInt32 endIndex = rangeToSearch.location + rangeToSearch.length;
CFMutableDataRef rangeStorage = NULL; // Basically an array of CFRange, CFDataRef (packed)
uint8_t *rangeStorageBytes = NULL;
CFIndex foundCount = 0;
CFIndex capacity = 0; // Number of CFRange, CFDataRef element slots in rangeStorage
if (alloc == NULL) alloc = __CFGetDefaultAllocator();
while ((rangeToSearch.length > 0) && CFStringFindWithOptions(string, stringToFind, rangeToSearch, compareOptions, &foundRange)) {
// Determine the next range
if (backwards) {
rangeToSearch.length = foundRange.location - rangeToSearch.location;
} else {
rangeToSearch.location = foundRange.location + foundRange.length;
rangeToSearch.length = endIndex - rangeToSearch.location;
}
// If necessary, grow the data and squirrel away the found range
if (foundCount >= capacity) {
if (rangeStorage == NULL) rangeStorage = CFDataCreateMutable(alloc, 0);
capacity = (capacity + 4) * 2;
CFDataSetLength(rangeStorage, capacity * (sizeof(CFRange) + sizeof(CFDataRef)));
rangeStorageBytes = (uint8_t *)CFDataGetMutableBytePtr(rangeStorage) + foundCount * (sizeof(CFRange) + sizeof(CFDataRef));
}
memmove(rangeStorageBytes, &foundRange, sizeof(CFRange)); // The range
memmove(rangeStorageBytes + sizeof(CFRange), &rangeStorage, sizeof(CFDataRef)); // The data
rangeStorageBytes += (sizeof(CFRange) + sizeof(CFDataRef));
foundCount++;
}
if (foundCount > 0) {
CFIndex cnt;
CFMutableArrayRef array;
const CFArrayCallBacks callbacks = {0, __rangeRetain, __rangeRelease, __rangeCopyDescription, __rangeEqual};
CFDataSetLength(rangeStorage, foundCount * (sizeof(CFRange) + sizeof(CFDataRef))); // Tighten storage up
rangeStorageBytes = (uint8_t *)CFDataGetMutableBytePtr(rangeStorage);
array = CFArrayCreateMutable(alloc, foundCount * sizeof(CFRange *), &callbacks);
for (cnt = 0; cnt < foundCount; cnt++) {
// Each element points to the appropriate CFRange in the CFData
CFArrayAppendValue(array, rangeStorageBytes + cnt * (sizeof(CFRange) + sizeof(CFDataRef)));
}
CFRelease(rangeStorage); // We want the data to go away when all CFRanges inside it are released...
return array;
} else {
return NULL;
}
}
CFRange CFStringFind(CFStringRef string, CFStringRef stringToFind, CFStringCompareFlags compareOptions) {
CFRange foundRange;
if (CFStringFindWithOptions(string, stringToFind, CFRangeMake(0, CFStringGetLength(string)), compareOptions, &foundRange)) {
return foundRange;
} else {
return CFRangeMake(kCFNotFound, 0);
}
}
Boolean CFStringHasPrefix(CFStringRef string, CFStringRef prefix) {
return CFStringFindWithOptions(string, prefix, CFRangeMake(0, CFStringGetLength(string)), kCFCompareAnchored, NULL);
}
Boolean CFStringHasSuffix(CFStringRef string, CFStringRef suffix) {
return CFStringFindWithOptions(string, suffix, CFRangeMake(0, CFStringGetLength(string)), kCFCompareAnchored|kCFCompareBackwards, NULL);
}
#define MAX_TRANSCODING_LENGTH 4
#define HANGUL_JONGSEONG_COUNT (28)
CF_INLINE bool _CFStringIsHangulLVT(UTF32Char character) {
return (((character - HANGUL_SYLLABLE_START) % HANGUL_JONGSEONG_COUNT) ? true : false);
}
static uint8_t const __CFTranscodingHintLength[] = {
2, 3, 4, 4, 4, 4, 4, 2, 2, 2, 2, 4, 0, 0, 0, 0
};
enum {
kCFStringHangulStateL,
kCFStringHangulStateV,
kCFStringHangulStateT,
kCFStringHangulStateLV,
kCFStringHangulStateLVT,
kCFStringHangulStateBreak
};
#pragma mark Pictographic Sequences
/* The following few functions serve to identify ranges of pictographic sequences (AKA emoji sequences with forwards and backwards extension) in a string around a given index. */
// Reads a character from the given inline buffer at the given index.
// If the character is a non-BMP character, this reads the matching surrogate pair character as well, and returns the effective read range through an out parameter.
static inline UTF32Char __CFStringGetLongCharacterFromInlineBuffer(CFStringInlineBuffer *buffer, CFIndex length, CFIndex idx, CFRange *readRange) {
if (idx < 0 || idx >= length) {
// Matches CFStringGetCharacterFromInlineBuffer.
if (readRange) *readRange = CFRangeMake(kCFNotFound, 0);
return 0;
}
CFRange range = CFRangeMake(idx, 1);
UTF32Char character = CFStringGetCharacterFromInlineBuffer(buffer, idx);
if (CFUniCharIsSurrogateHighCharacter(character) && idx < length - 1) {
// We need to read ahead if possible to get the low surrogate and combine.
UTF16Char surrogateLow = CFStringGetCharacterFromInlineBuffer(buffer, idx + 1);
if (CFUniCharIsSurrogateLowCharacter(surrogateLow)) {
range.length++;
character = CFUniCharGetLongCharacterForSurrogatePair(character, surrogateLow);
}
} else if (CFUniCharIsSurrogateLowCharacter(character) && idx > 0) {
// We need to read behind if possible to get the low surrogate and combine.
UTF16Char surrogateHigh = CFStringGetCharacterFromInlineBuffer(buffer, idx - 1);
if (CFUniCharIsSurrogateHighCharacter(surrogateHigh)) {
range.location--;
range.length++;
character = CFUniCharGetLongCharacterForSurrogatePair(surrogateHigh, character);
}
}
if (readRange) *readRange = range;
return character;
}
static inline bool __CFStringIsValidExtendCharacterForPictographicSequence(UTF32Char character) {
// From https://www.unicode.org/reports/tr29/#Extend:
//
// Grapheme_Extend = Yes, or
// Emoji_Modifier=Yes in emoji-data.txt
//
return u_hasBinaryProperty(character, UCHAR_GRAPHEME_EXTEND) || u_hasBinaryProperty(character, UCHAR_EMOJI_MODIFIER);
}
static inline bool __CFStringIsValidExtendedPictographicCharacterForPictographicSequence(UTF32Char character) {
return u_hasBinaryProperty(character, UCHAR_EXTENDED_PICTOGRAPHIC);
}
static inline bool __CFStringIsValidPrecoreCharacterForPictographicSequence(UTF32Char character) {
// From https://www.unicode.org/reports/tr29/#Regex_Definitions:
//
// precore := Prepend
//
// We can look up the grapheme cluster break class and use it directly.
bool isValid = (UGraphemeClusterBreak)u_getIntPropertyValue(character, UCHAR_GRAPHEME_CLUSTER_BREAK) == U_GCB_PREPEND;
return isValid;
}
static inline bool __CFStringIsValidPostcoreCharacterForPictographicSequence(UTF32Char character) {
// From https://www.unicode.org/reports/tr29/#Regex_Definitions:
//
// postcore := [Extend ZWJ SpacingMark]
//
// We already have an expression to match Extend characters (and ZWJ is trivial); we can look up the grapheme cluster break class and use it directly to determine spacing mark characters.
bool isValid = character == ZERO_WIDTH_JOINER || __CFStringIsValidExtendCharacterForPictographicSequence(character) || (UGraphemeClusterBreak)u_getIntPropertyValue(character, UCHAR_GRAPHEME_CLUSTER_BREAK) == U_GCB_SPACING_MARK;
return isValid;
}
// Represents the match information for a single component in a pictographic sequence below.
// See __CFStringGetExtendedPictographicSequenceComponent and __CFStringGetExtendedPictographicSequence for usage information.
typedef struct {
CFRange range;
CFIndex firstExtendIndex;
CFIndex zwjIndex;
CFIndex pictographIndex;
} __CFStringPictographicSequenceComponent;
// Given an index, attempts to return the range of the containing element of Grapheme Cluster Boundary Rule GB11:
//
// \p{Extended_Pictographic} (Extend* ZWJ \p{Extended_Pictographic})*
//
// Specifically, this will attempt to match either the lone starting \p{Extended_Pictographic} if the index corresponds to it, or a singular instance of (Extend* ZWJ \p{Extended_Pictographic}) which contains the `index`.
// For instance, the string @"stuff...👩❤️💋👨stuff..." is segmented this way as:
//
// 👩 ❤ Var_Sel 💋 👨
// ┌──────────┐│┌───────┐┌─────┐┌──────┐┌──────┐┌─────┐┌───────┐┌─────┐┌───────┐│┌──────────┐
// │ stuff... │ │ 1F469 ││ ZWJ ││ 2764 ││ FE0F ││ ZWJ ││ 1F48B ││ ZWJ ││ 1F468 │ │ stuff... │
// └──────────┘│└───────┘└─────┘└──────┘└──────┘└─────┘└───────┘└─────┘└───────┘│└──────────┘
// ──r0─── ─────r1────── ─────────r2─────────── ──────r3──────
//
// Each of ranges r0-r3 is one "component" of this sequence that we're looking to match. Given any index which falls in one of these ranges, we should return the same match information.
// Here, r0 is matched as a \p{Extended_Pictographic}, while each of r1-r3 are matched as (Extend* ZWJ \p{Extended_Pictographic}).
//
// If a match is found for the given index, the match information is returned through the `outComponent` parameter.
static inline bool __CFStringGetExtendedPictographicSequenceComponent(CFStringInlineBuffer *buffer, CFIndex length, CFIndex index, __CFStringPictographicSequenceComponent *outComponent) {
if (index < 0 || index >= length) {
// This is relied upon in __CFStringGetExtendedPictographicSequence to prevent reading invalid components without additional checking.
return false;
}
__CFStringPictographicSequenceComponent match = {{kCFNotFound, 0}, -1, -1, -1};
// The given index can point to any part of any component in a sequence as above.
// Start by rewinding backwards as far as we can to see if we we're in the type of component which has a ZWJ or not.
CFRange currentRange = CFRangeMake(index, 0);
while (currentRange.location >= 0) {
// This adjusts currentRange to match the actual range of a long character, if necessary.
UTF32Char character = __CFStringGetLongCharacterFromInlineBuffer(buffer, length, currentRange.location, ¤tRange);
if (__CFStringIsValidExtendCharacterForPictographicSequence(character)) {
// This is an extend character; we're at the beginning of the cluster.
match.firstExtendIndex = currentRange.location;
} else if (character == ZERO_WIDTH_JOINER) {
if (match.firstExtendIndex != -1 || match.zwjIndex != -1) {
// A ZWJ isn't valid here — we've already previously seen a ZWJ or Extend characters (i.e. this ZWJ is not part of this component).
// For example, we've extended backwards and hit another ZWJ:
//
// ┌───────────────────────┐ ┌─────┐ ┌────────┐ ┌─────┐ ┌───────────────────────┐
// │ Extended_Pictographic │ │ ZWJ │ │ Extend │ │ ZWJ │ │ Extended_Pictographic │
// └───────────────────────┘ └─────┘ └────────┘ └─────┘ └───────────────────────┘
// ▲ │
// └───────────────────────────────────┘
//
// Note that this sequence is not valid (and we will reject it one level up), but we'll stop here.
break;
}
match.zwjIndex = currentRange.location;
} else if (__CFStringIsValidExtendedPictographicCharacterForPictographicSequence(character)) {
if (match.pictographIndex != -1 || match.zwjIndex != -1 || match.firstExtendIndex != -1) {
// We've already either seen a pictograph before, or we've seen other characters which come before a pictograph.
// For example, we've extended far enough backwards to find the previous pictograph:
//
// ┌───────────────────────┐ ┌────────┐ ┌─────┐ ┌───────────────────────┐
// │ Extended_Pictographic │ │ Extend │ │ ZWJ │ │ Extended_Pictographic │
// └───────────────────────┘ └────────┘ └─────┘ └───────────────────────┘
// ▲ │
// └────────────────────────────────────────────┘
break;
}
match.pictographIndex = currentRange.location;
} else {
// This isn't a character which is valid to include in this pictograph sequence.
break;
}
match.range.location = currentRange.location;
match.range.length += currentRange.length;
currentRange.location--;
}
// We've advanced as far back as we can go. At this point, we should either have matched
//
// ┌───────────────────────┐
// │ Extended_Pictographic │
// └───────────────────────┘
//
// or at least some subset of
//
// ┌─────────┐ ┌──────┐ ┌────────────────────────┐
// │ Extend* │ │ ZWJ* │ │ Extended_Pictographic* │
// └─────────┘ └──────┘ └────────────────────────┘
//
// If we didn't match _anything_ then we're not looking at a valid component.
if (match.pictographIndex == -1) {
// No pictograph yet...
if (match.zwjIndex == -1 && match.firstExtendIndex == -1) {
// ... nor anything else. Advancing forward won't be any use here; this isn't a pictographic sequence component.
return false;
}
// We've matched the start of a component here; continue below.
} else {
// We've got a pictograph, so there's nothing left to find by searching forward.
// Possible options for what we've matched:
//
// 1. <Extended_Pictographic>
// 2. <ZWJ> <Extended_Pictographic>
// 3. <Extend> <Extended_Pictographic>
// 4. <Extend> <ZWJ> <Extended_Pictographic>
//
// Of these, options 1, 2, and 4 are valid, since we either need a ZWJ or don't.
if (match.firstExtendIndex != -1 && match.zwjIndex == -1) {
// This is option 3 above. It's likely that we're matching the FE0F (or similar) from a preceding cluster.
// For example, we may have extended backwards from the start of 👨👦 to the end of 👱♀️ in the string @"👱♀️👨👦" with no ZWJ in between:
//
//
// 👱 ZWJ ♀ Var_Sel 👨 ZWJ 👦
// ┌───────┐ ┌──────┐ ┌──────┐ ┌──────┐ │ ┌───────┐ ┌──────┐ ┌───────┐
// │ 1F471 │ │ 200D │ │ 2640 │ │ FE0F │ │ 1F468 │ │ 200D │ │ 1F466 │
// └───────┘ └──────┘ └──────┘ └──────┘ │ └───────┘ └──────┘ └───────┘
// ▲ │
// └──────────┘
//
// FE0F is a variant selector (a valid extend character) and we matched it with no intervening ZWJ.
// In this case, simply ignore the extend characters we've found and return the base pictograph itself as the start of a new sequence looking forward.
match.range.location = match.pictographIndex;
match.range.length -= (match.pictographIndex - match.firstExtendIndex);
}
if (outComponent) *outComponent = match;
return true;
}
// We don't have a full component yet — we might have some Extend characters and/or a ZWJ, but no pictograph yet.
// Extend forward as far as we can.
currentRange.location = match.range.location + match.range.length;
currentRange.length = 0;
while (match.pictographIndex == -1 && currentRange.location < length) {
// This adjusts currentRange to match the actual range of a long character, if necessary.
UTF32Char character = __CFStringGetLongCharacterFromInlineBuffer(buffer, length, currentRange.location, ¤tRange);
if (__CFStringIsValidExtendCharacterForPictographicSequence(character)) {
if (match.zwjIndex != -1) {
// We've already seen a ZWJ, so further <Extend>* characters are not valid here.
// For example:
//
// ┌───────────────────────┐ ┌─────┐ ┌────────┐ ┌─────┐ ┌───────────────────────┐
// │ Extended_Pictographic │ │ ZWJ │ │ Extend │ │ ZWJ │ │ Extended_Pictographic │
// └───────────────────────┘ └─────┘ └────────┘ └─────┘ └───────────────────────┘
// │ ▲
// └────────┘
//
// Note that this sequence is not valid (and we will reject all trailing characters as part of the sequence), but we'll stop here.
break;
}
// When extending backwards, we updated component.firstExtendIndex; here we don't update it because we're extending forward.
} else if (character == ZERO_WIDTH_JOINER) {
if (match.zwjIndex != -1) {
// We've already seen a ZWJ, so another ZWJ is not valid here.
// For example:
//
// ┌───────────────────────┐ ┌─────┐ ┌─────┐ ┌───────────────────────┐
// │ Extended_Pictographic │ │ ZWJ │ │ ZWJ │ │ Extended_Pictographic │
// └───────────────────────┘ └─────┘ └─────┘ └───────────────────────┘
// │ ▲
// └───────┘
//
// Note that this sequence is not valid (and we will reject all trailing characters as part of the sequence), but we'll stop here.
break;
}
match.zwjIndex = currentRange.location;
} else if (__CFStringIsValidExtendedPictographicCharacterForPictographicSequence(character)) {
// We're matching the pictograph we've been looking for, e.g.
//
// ┌───────────────────────┐ ┌─────┐ ┌───────────────────────┐
// │ Extended_Pictographic │ │ ZWJ │ │ Extended_Pictographic │
// └───────────────────────┘ └─────┘ └───────────────────────┘
// │ ▲
// └────────────────┘
//
// The loop condition means we'll break out after matching this.
match.pictographIndex = currentRange.location;
} else {
// This isn't a character which is valid to include in this pictograph sequence.
break;
}
match.range.length += currentRange.length;
currentRange.location += currentRange.length;
currentRange.length = 0;
}
if (match.pictographIndex == -1) {
// Still no pictograph. We're done.
return false;
} else {
// At this point we should have everything.
if (outComponent) *outComponent = match;
return true;
}
}
// Given an index into a buffer, attempts to match an extended pictographic sequence containing the character at that index.
// Specifically, we're looking to match an instance of the extended grapheme cluster grammar in Table 1b. of UAX#29 (http://unicode.org/reports/tr29/) as it concerns pictographic sequences:
//
// precore* core postcore*
//
// where in our case, we care about
//
// precore := Prepend
// core := \p{Extended_Pictographic} (Extend* ZWJ \p{Extended_Pictographic})*
// postcore := [Extend ZWJ SpacingMark]
//
// In the future, we can extend core to match the full definition of
//
// core := hangul-syllable | ri-sequence | xpicto-sequence | [^Control CR LF]
//
// to more generalize the implementation of CFStringGetRangeOfCharacterClusterAtIndex.
//
// To do this, we look to match instances of precore, core, and postcore characters around `index`.
// __CFStringGetExtendedPictographicSequenceComponent above will be used to match instances of `core`, and the general strategy involved tries to extend backwards to the start of the sequence, then forwards to figure out where/what we're matching.
//
// For instance, a string like @"stuff...👩🏿✈️stuff..." is segmented as
//
// 👩 🏿 ✈ Var_Sel
// ┌──────────┐│┌───────┐┌───────┐┌─────┐┌──────┐┌──────┐│┌──────────┐
// │ stuff... │ │ 1F469 ││ 1F3FF ││ ZWJ ││ 2708 ││ FE0F │ │ stuff... │
// └──────────┘│└───────┘└───────┘└─────┘└──────┘└──────┘│└──────────┘
// ──r0─── ──────────r1────────── ──r2──
//
// Given any index which falls within ranges r0-r2, we're looking to return the full match from the beginning of r0 to the end of r2 in `outRange`.
// In this case,
//
// * r0 is the base \p{Extended_Pictographic} matched from `core`,
// * r1 is the extension (Extend* ZWJ \p{Extended_Pictographic}) matched from `core`,
// * and r2 is a `postcore` extend character
//
// `core` characters are matched using __CFStringGetExtendedPictographicSequenceComponent, and we extend with `precore` and `postcore` here.
static inline bool __CFStringGetExtendedPictographicSequence(CFStringInlineBuffer *buffer, CFIndex length, CFIndex index, CFRange *outRange) {
if (index < 0 || index >= length) {
return false;
}
// We want to find the base character here of the whole cluster, so let's rewind backwards.
// We may be at the end of the cluster, so let's start rewinding backwards until we hit a boundary.
CFRange currentRange;
UTF32Char currentCharacter = __CFStringGetLongCharacterFromInlineBuffer(buffer, length, index, ¤tRange);
// We'll start by matching postcore characters. One difficulty here is that allowable postcore characters are also present in the middle of core matches (like Extend and ZWJs).
// This means that if we match some characters here, they may actually fall in the middle of a core match:
//
// ┌───────────────────────┐ ┌─────┐ ┌───────────────────────┐
// │ Extended_Pictographic │ │ ZWJ │ │ Extended_Pictographic │
// └───────────────────────┘ └─────┘ └───────────────────────┘
// ▲
// └─── could be postcore or could be core (in this case, it's core, but we don't know it yet)
//
// We can reconcile this later; for now, we'll keep track of this range until we've concluded whether or not these are really postcore characters.
CFRange postcoreRange = CFRangeMake(currentRange.length, 0);
while (__CFStringIsValidPostcoreCharacterForPictographicSequence(currentCharacter)) {
postcoreRange.location = currentRange.location;
postcoreRange.length += currentRange.length;
if (postcoreRange.location == 0) {
// We've managed to only match postcore characters and can't extend further; there's no pictographic sequence here, so no point in continuing to look.
return false;
}
// This adjusts currentRange to match the actual range of a long character, if necessary.
currentCharacter = __CFStringGetLongCharacterFromInlineBuffer(buffer, length, postcoreRange.location - 1, ¤tRange);
}
// We may or may not have matched any postcore characters, but either way, now try to match a pictographic sequence if we can, extending backwards.
__CFStringPictographicSequenceComponent currentComponent = {{kCFNotFound, 0}, -1, -1, -1};
CFRange coreRange = CFRangeMake(currentRange.location, 0);
while (__CFStringGetExtendedPictographicSequenceComponent(buffer, length, currentRange.location, ¤tComponent)) {
coreRange.location = currentComponent.range.location;
coreRange.length += currentComponent.range.length;
currentRange.location = currentComponent.range.location - 1;
currentRange.length = 0;
if (currentComponent.zwjIndex == -1) {
// This component is the start of the sequence; stop trying to look for more.
break;
}
}
bool shouldLookForPrecoreCharacters = true;
if (currentComponent.firstExtendIndex != -1 || currentComponent.zwjIndex != -1) {
// The last component we found had characters preceding the pictograph, but we stopped looking.
// This can either be because we've hit the beginning of the string (i.e. there's no preceding component to find), or the character preceding those would not form a valid component itself, e.g.:
//
// │ ┌─────┐ ┌───────┐ ┌─────┐
// │ ZWJ │ │ 1F469 │ │ ... │
// │ └─────┘ └───────┘ └─────┘
// ▲ │
// └────────┘
//
// OR
//
// ┌────────────┐ ┌────────┐ ┌─────┐ ┌───────┐ ┌─────┐
// │ Garbage... │ | Extend | │ ZWJ │ │ 1F469 │ │ ... │
// └────────────┘ └────────┘ └─────┘ └───────┘ └─────┘
// ▲ │
// └──────────────────┘
//
// or similar.
// In this case, we'll ignore these preceding characters and use the pictograph as the start of the sequence looking forward.
coreRange.location = currentComponent.pictographIndex;
coreRange.length -= currentComponent.pictographIndex - currentComponent.range.location;
currentRange.location = currentComponent.pictographIndex + 1;
// There's also no point in further looking for precore characters because Extend and ZWJ are not valid precore chars.
shouldLookForPrecoreCharacters = false;
}
if (postcoreRange.length > 0 && coreRange.length == 0) {
// We matched some postcore characters but no pictographic sequence components.
// There's no point in looking for precore characters, or looking forwards afterwards.
return false;
}
// We've now extended backwards to the start of the emoji sequence, if one exists. We can now try to match precore if there are any to be found.
CFRange precoreRange = CFRangeMake(currentRange.location, 0);
if (shouldLookForPrecoreCharacters) {
// Extend backwards as far as possible.
if (currentRange.location >= 0) {
// This adjusts currentRange to match the actual range of a long character, if necessary.
currentCharacter = __CFStringGetLongCharacterFromInlineBuffer(buffer, length, currentRange.location, ¤tRange);
while (__CFStringIsValidPrecoreCharacterForPictographicSequence(currentCharacter)) {
precoreRange.location = currentRange.location;
precoreRange.length += currentRange.length;
if (precoreRange.location == 0) {
break;
}
currentCharacter = __CFStringGetLongCharacterFromInlineBuffer(buffer, length, precoreRange.location - 1, ¤tRange);
}
}
// Then forwards...
currentRange = CFRangeMake(precoreRange.location + precoreRange.length, 0);
while (currentRange.location < length) {
// This adjusts currentRange to match the actual range of a long character, if necessary.
currentCharacter = __CFStringGetLongCharacterFromInlineBuffer(buffer, length, currentRange.location, ¤tRange);
if (__CFStringIsValidPrecoreCharacterForPictographicSequence(currentCharacter)) {
precoreRange.length += currentRange.length;
currentRange.location += currentRange.length;
} else {
break;
}
}
}
// We've now extended backwards as far as we can go...
if (precoreRange.length == 0 && coreRange.length == 0) {
// ... and haven't matched anything meaningful that we could get to by searching forward from here (i.e. we can't match precore/core characters after any postcore ones).
return false;
}
// We can now extend forwards to increase our match.
// If we found no core characters, we'll look for them right past any precore characters we've found. If we have found core characters, we'll continue extending them forward.
if (coreRange.length == 0) {
coreRange = CFRangeMake(precoreRange.location + precoreRange.length, 0);
currentRange = coreRange;
} else {
currentRange = CFRangeMake(coreRange.location + coreRange.length, 0);
}
while (__CFStringGetExtendedPictographicSequenceComponent(buffer, length, currentRange.location, ¤tComponent)) {
if (coreRange.length > 0 && currentComponent.zwjIndex == -1) {
// This component had no ZWJ; it's the start of the next sequence, and we shouldn't include it.
break;
}
coreRange.length += currentComponent.range.length;
currentRange.location += currentComponent.range.length;
}
// Now before looking for more postcore characters, we should evaluate whether the ones we've already seen are actual postcore characters or not.
// It's entirely possible we matched something like a ZWJ up-front that's really part of the core match.
if (postcoreRange.length > 0) {
CFIndex onePastCore = coreRange.location + coreRange.length;
CFIndex onePastPostcore = postcoreRange.location + postcoreRange.length;
if (onePastCore >= onePastPostcore) {
// We've subsumed the entire postcore range, e.g., our initial example:
//
// ┌───────────────────────┐ ┌─────┐ ┌───────────────────────┐ ┌─────┐
// │ Extended_Pictographic │ │ ZWJ │ │ Extended_Pictographic │ | ... |
// └───────────────────────┘ └─────┘ └───────────────────────┘ └─────┘
// ▲ ▲
// │ └─ coreRange ends here
// └─── appeared to be postcore but is part of coreRange
//
// We can look for more postcore characters past the whole of the currently matched range.
postcoreRange = CFRangeMake(onePastCore, 0);
}
currentRange = CFRangeMake(postcoreRange.location + postcoreRange.length, 0);
} else {
// We didn't find any postcore characters; currentRange points just past the end of coreRange, so advance to there and if we find any characters, we'll append them to this range.
postcoreRange = currentRange;
}
if (currentRange.location < length) {
// There may be further trailing post-core characters.
// This adjusts currentRange to match the actual range of a long character, if necessary.
currentCharacter = __CFStringGetLongCharacterFromInlineBuffer(buffer, length, currentRange.location, ¤tRange);
while (__CFStringIsValidPostcoreCharacterForPictographicSequence(currentCharacter)) {
postcoreRange.length += currentRange.length;
currentRange.location += currentRange.length;
currentCharacter = __CFStringGetLongCharacterFromInlineBuffer(buffer, length, currentRange.location, ¤tRange);
}
}
// We're only willing to return full matches, which necessitate finding a core character -- otherwise this is not an eligible xpicto-sequence.
bool const haveMatch = coreRange.length > 0;
if (haveMatch && outRange) {
// At this point, the union of {precoreRange, coreRange, postcoreRange} gives us the full range of the match.
*outRange = coreRange;
if (precoreRange.length > 0) {
outRange->location = precoreRange.location;
outRange->length += precoreRange.length;
}
if (postcoreRange.length > 0) {
outRange->length += postcoreRange.length;
}
}
return haveMatch;
}
#pragma mark Composed Character Sequences
#define RI_SURROGATE_HI (0xD83C)
static inline bool __CFStringIsRegionalIndicatorSurrogateLow(UTF16Char character) { return (character >= 0xDDE6) && (character <= 0xDDFF) ? true : false; }
static inline bool __CFStringIsRegionalIndicatorAtIndex(CFStringInlineBuffer *buffer, CFIndex index) {
return ((CFStringGetCharacterFromInlineBuffer(buffer, index) == RI_SURROGATE_HI) && __CFStringIsRegionalIndicatorSurrogateLow(CFStringGetCharacterFromInlineBuffer(buffer, index + 1))) ? true : false;
}
static inline bool __CFStringIsFitzpatrickModifiers(UTF32Char character) { return ((character >= 0x1F3FB) && (character <= 0x1F3FF) ? true : false); }
static inline bool __CFStringIsTagSequence(UTF32Char character) { return ((character >= 0xE0020) && (character <= 0xE007F) ? true : false); }
static CFRange _CFStringInlineBufferGetComposedRange(CFStringInlineBuffer *buffer, CFIndex start, CFStringCharacterClusterType type, const uint8_t *bmpBitmap, CFIndex csetType) {
CFIndex end = start + 1;
const uint8_t *bitmap = bmpBitmap;
UTF32Char character;
UTF16Char otherSurrogate;
uint8_t step;
character = CFStringGetCharacterFromInlineBuffer(buffer, start);
// We don't combine characters in Armenian ~ Limbu range for backward deletion
if ((type != kCFStringBackwardDeletionCluster) || (character < 0x0530) || (character > 0x194F)) {
// Check if the current is surrogate
if (CFUniCharIsSurrogateHighCharacter(character) && CFUniCharIsSurrogateLowCharacter((otherSurrogate = CFStringGetCharacterFromInlineBuffer(buffer, start + 1)))) {
++end;
character = CFUniCharGetLongCharacterForSurrogatePair(character, otherSurrogate);
bitmap = CFUniCharGetBitmapPtrForPlane(csetType, (character >> 16));
}
// Extend backward
while (start > 0) {
if ((type == kCFStringBackwardDeletionCluster) && (character >= 0x0530) && (character < 0x1950)) break;
if (character < 0x10000) { // the first round could be already be non-BMP
if (CFUniCharIsSurrogateLowCharacter(character) && CFUniCharIsSurrogateHighCharacter((otherSurrogate = CFStringGetCharacterFromInlineBuffer(buffer, start - 1)))) {
character = CFUniCharGetLongCharacterForSurrogatePair(otherSurrogate, character);
bitmap = CFUniCharGetBitmapPtrForPlane(csetType, (character >> 16));
if (--start == 0) break; // starting with non-BMP combining mark
} else {
bitmap = bmpBitmap;
}
}
Boolean isRelevantFitzpatrickModifier = (start > 0 && __CFStringIsFitzpatrickModifiers(character));
Boolean isInBitmap = CFUniCharIsMemberOfBitmap(character, bitmap);
Boolean isTagSequence = __CFStringIsTagSequence(character);
Boolean behavesLikeCombiningMark = (character == 0xFF9E) || (character == 0xFF9F) || ((character & 0x1FFFF0) == 0xF870 /* variant tag */);
if (!isRelevantFitzpatrickModifier && !isInBitmap && ! isTagSequence && !behavesLikeCombiningMark) {
// Nothing to extend backward for.
break;
}
--start;
character = CFStringGetCharacterFromInlineBuffer(buffer, start);
}
}
// Hangul
if (((character >= HANGUL_CHOSEONG_START) && (character <= HANGUL_JONGSEONG_END)) || ((character >= HANGUL_SYLLABLE_START) && (character <= HANGUL_SYLLABLE_END))) {
uint8_t state;
uint8_t initialState;
if (character < HANGUL_JUNGSEONG_START) {
state = kCFStringHangulStateL;
} else if (character < HANGUL_JONGSEONG_START) {
state = kCFStringHangulStateV;
} else if (character < HANGUL_SYLLABLE_START) {
state = kCFStringHangulStateT;
} else {
state = (_CFStringIsHangulLVT(character) ? kCFStringHangulStateLVT : kCFStringHangulStateLV);
}
initialState = state;
// Extend backward
while (((character = CFStringGetCharacterFromInlineBuffer(buffer, start - 1)) >= HANGUL_CHOSEONG_START) && (character <= HANGUL_SYLLABLE_END) && ((character <= HANGUL_JONGSEONG_END) || (character >= HANGUL_SYLLABLE_START))) {
switch (state) {
case kCFStringHangulStateV:
if (character <= HANGUL_CHOSEONG_END) {
state = kCFStringHangulStateL;
} else if ((character >= HANGUL_SYLLABLE_START) && (character <= HANGUL_SYLLABLE_END) && !_CFStringIsHangulLVT(character)) {
state = kCFStringHangulStateLV;
} else if (character > HANGUL_JUNGSEONG_END) {
state = kCFStringHangulStateBreak;
}
break;
case kCFStringHangulStateT:
if ((character >= HANGUL_JUNGSEONG_START) && (character <= HANGUL_JUNGSEONG_END)) {
state = kCFStringHangulStateV;
} else if ((character >= HANGUL_SYLLABLE_START) && (character <= HANGUL_SYLLABLE_END)) {
state = (_CFStringIsHangulLVT(character) ? kCFStringHangulStateLVT : kCFStringHangulStateLV);
} else if (character < HANGUL_JUNGSEONG_START) {
state = kCFStringHangulStateBreak;
}
break;
default:
state = ((character < HANGUL_JUNGSEONG_START) ? kCFStringHangulStateL : kCFStringHangulStateBreak);
break;
}
if (state == kCFStringHangulStateBreak) break;
--start;
}
// Extend forward
state = initialState;
while (((character = CFStringGetCharacterFromInlineBuffer(buffer, end)) > 0) && (((character >= HANGUL_CHOSEONG_START) && (character <= HANGUL_JONGSEONG_END)) || ((character >= HANGUL_SYLLABLE_START) && (character <= HANGUL_SYLLABLE_END)))) {
switch (state) {
case kCFStringHangulStateLV:
case kCFStringHangulStateV:
if ((character >= HANGUL_JUNGSEONG_START) && (character <= HANGUL_JONGSEONG_END)) {
state = ((character < HANGUL_JONGSEONG_START) ? kCFStringHangulStateV : kCFStringHangulStateT);
} else {
state = kCFStringHangulStateBreak;
}
break;
case kCFStringHangulStateLVT:
case kCFStringHangulStateT:
state = (((character >= HANGUL_JONGSEONG_START) && (character <= HANGUL_JONGSEONG_END)) ? kCFStringHangulStateT : kCFStringHangulStateBreak);
break;
default:
if (character < HANGUL_JUNGSEONG_START) {
state = kCFStringHangulStateL;
} else if (character < HANGUL_JONGSEONG_START) {
state = kCFStringHangulStateV;
} else if (character >= HANGUL_SYLLABLE_START) {
state = (_CFStringIsHangulLVT(character) ? kCFStringHangulStateLVT : kCFStringHangulStateLV);
} else {
state = kCFStringHangulStateBreak;
}
break;
}
if (state == kCFStringHangulStateBreak) break;
++end;
}
}
// Extend forward
while ((character = CFStringGetCharacterFromInlineBuffer(buffer, end)) > 0) {
if ((type == kCFStringBackwardDeletionCluster) && (character >= 0x0530) && (character < 0x1950)) break;
if (CFUniCharIsSurrogateHighCharacter(character) && CFUniCharIsSurrogateLowCharacter((otherSurrogate = CFStringGetCharacterFromInlineBuffer(buffer, end + 1)))) {
character = CFUniCharGetLongCharacterForSurrogatePair(character, otherSurrogate);
bitmap = CFUniCharGetBitmapPtrForPlane(csetType, (character >> 16));
step = 2;
} else {
bitmap = bmpBitmap;
step = 1;
}
Boolean isRelevantFitzpatrickModifier = __CFStringIsFitzpatrickModifiers(character);
Boolean isInBitmap = CFUniCharIsMemberOfBitmap(character, bitmap);
Boolean isTagSequence = __CFStringIsTagSequence(character);
Boolean behavesLikeCombiningMark = (character == 0xFF9E) || (character == 0xFF9F) || ((character & 0x1FFFF0) == 0xF870 /* variant tag */);
if (!isRelevantFitzpatrickModifier && !isInBitmap && ! isTagSequence && !behavesLikeCombiningMark) {
// Nothing to extend backward for.
break;
}
end += step;
}
return CFRangeMake(start, end - start);
}
CF_INLINE bool _CFStringIsVirama(UTF32Char character, const uint8_t *combClassBMP) {
return ((character == COMBINING_GRAPHEME_JOINER) || (CFUniCharGetCombiningPropertyForCharacter(character, (const uint8_t *)((character < 0x10000) ? combClassBMP : CFUniCharGetUnicodePropertyDataForPlane(kCFUniCharCombiningProperty, (character >> 16)))) == 9) ? true : false);
}
CFRange CFStringGetRangeOfCharacterClusterAtIndex(CFStringRef string, CFIndex charIndex, CFStringCharacterClusterType type) {
CFRange range;
CFIndex currentIndex;
CFIndex length = CFStringGetLength(string);
CFIndex csetType = ((kCFStringGraphemeCluster == type) ? kCFUniCharGraphemeExtendCharacterSet : kCFUniCharNonBaseCharacterSet);
CFStringInlineBuffer stringBuffer;
const uint8_t *bmpBitmap;
const uint8_t *letterBMP;
static const uint8_t *combClassBMP = NULL;
UTF32Char character;
UTF16Char otherSurrogate;
if (charIndex >= length) return CFRangeMake(kCFNotFound, 0);
/* Fast case. If we're eight-bit, it's either the default encoding is cheap or the content is all ASCII. Watch out when (or if) adding more 8bit Mac-scripts in CFStringEncodingConverters
*/
if (!CF_IS_OBJC(_kCFRuntimeIDCFString, string) && !CF_IS_SWIFT(_kCFRuntimeIDCFString, string) && __CFStrIsEightBit(string)) return CFRangeMake(charIndex, 1);
bmpBitmap = CFUniCharGetBitmapPtrForPlane(csetType, 0);
letterBMP = CFUniCharGetBitmapPtrForPlane(kCFUniCharLetterCharacterSet, 0);
if (NULL == combClassBMP) combClassBMP = (const uint8_t *)CFUniCharGetUnicodePropertyDataForPlane(kCFUniCharCombiningProperty, 0);
CFStringInitInlineBuffer(string, &stringBuffer, CFRangeMake(0, length));
// Get composed character sequence first
range = _CFStringInlineBufferGetComposedRange(&stringBuffer, charIndex, type, bmpBitmap, csetType);
// Do grapheme joiners
if (type < kCFStringCursorMovementCluster) {
const uint8_t *letter = letterBMP;
// Check to see if we have a letter at the beginning of initial cluster
character = CFStringGetCharacterFromInlineBuffer(&stringBuffer, range.location);
if ((range.length > 1) && CFUniCharIsSurrogateHighCharacter(character) && CFUniCharIsSurrogateLowCharacter((otherSurrogate = CFStringGetCharacterFromInlineBuffer(&stringBuffer, range.location + 1)))) {
character = CFUniCharGetLongCharacterForSurrogatePair(character, otherSurrogate);
letter = CFUniCharGetBitmapPtrForPlane(kCFUniCharLetterCharacterSet, (character >> 16));
}
if ((character == ZERO_WIDTH_JOINER) || CFUniCharIsMemberOfBitmap(character, letter)) {
CFRange otherRange;
// Check if preceded by grapheme joiners (U034F and viramas)
otherRange.location = currentIndex = range.location;
while (currentIndex > 1) {
character = CFStringGetCharacterFromInlineBuffer(&stringBuffer, --currentIndex);
// ??? We're assuming viramas only in BMP
if ((_CFStringIsVirama(character, combClassBMP) || ((character == ZERO_WIDTH_JOINER) && _CFStringIsVirama(CFStringGetCharacterFromInlineBuffer(&stringBuffer, --currentIndex), combClassBMP))) && (currentIndex > 0)) {
--currentIndex;
} else {
break;
}
currentIndex = _CFStringInlineBufferGetComposedRange(&stringBuffer, currentIndex, type, bmpBitmap, csetType).location;
character = CFStringGetCharacterFromInlineBuffer(&stringBuffer, currentIndex);
if (CFUniCharIsSurrogateLowCharacter(character) && CFUniCharIsSurrogateHighCharacter((otherSurrogate = CFStringGetCharacterFromInlineBuffer(&stringBuffer, currentIndex - 1)))) {
character = CFUniCharGetLongCharacterForSurrogatePair(character, otherSurrogate);
letter = CFUniCharGetBitmapPtrForPlane(kCFUniCharLetterCharacterSet, (character >> 16));
--currentIndex;
} else {
letter = letterBMP;
}
if (!CFUniCharIsMemberOfBitmap(character, letter)) break;
range.location = currentIndex;
}
range.length += otherRange.location - range.location;
// Check if followed by grapheme joiners
if ((range.length > 1) && ((range.location + range.length) < length)) {
otherRange = range;
currentIndex = otherRange.location + otherRange.length;
do {
character = CFStringGetCharacterFromInlineBuffer(&stringBuffer, currentIndex - 1);
// ??? We're assuming viramas only in BMP
if ((character != ZERO_WIDTH_JOINER) && !_CFStringIsVirama(character, combClassBMP)) break;
character = CFStringGetCharacterFromInlineBuffer(&stringBuffer, currentIndex);
if (character == ZERO_WIDTH_JOINER) character = CFStringGetCharacterFromInlineBuffer(&stringBuffer, ++currentIndex);
if (CFUniCharIsSurrogateHighCharacter(character) && CFUniCharIsSurrogateLowCharacter((otherSurrogate = CFStringGetCharacterFromInlineBuffer(&stringBuffer, currentIndex + 1)))) {
character = CFUniCharGetLongCharacterForSurrogatePair(character, otherSurrogate);
letter = CFUniCharGetBitmapPtrForPlane(kCFUniCharLetterCharacterSet, (character >> 16));
} else {
letter = letterBMP;
}
// We only conjoin letters
if (!CFUniCharIsMemberOfBitmap(character, letter)) break;
otherRange = _CFStringInlineBufferGetComposedRange(&stringBuffer, currentIndex, type, bmpBitmap, csetType);
currentIndex = otherRange.location + otherRange.length;
} while ((otherRange.location + otherRange.length) < length);
range.length = currentIndex - range.location;
}
}
}
// Check if we're part of prefix transcoding hints
CFIndex otherIndex;
currentIndex = (range.location + range.length) - (MAX_TRANSCODING_LENGTH + 1);
if (currentIndex < 0) currentIndex = 0;
while (currentIndex <= range.location) {
character = CFStringGetCharacterFromInlineBuffer(&stringBuffer, currentIndex);
if ((character & 0x1FFFF0) == 0xF860) { // transcoding hint
otherIndex = currentIndex + __CFTranscodingHintLength[(character - 0xF860)] + 1;
if (otherIndex >= (range.location + range.length)) {
if (otherIndex <= length) {
for (CFIndex checkIndex = currentIndex + 1; checkIndex < otherIndex;) {
CFRange checkRange = _CFStringInlineBufferGetComposedRange(&stringBuffer, checkIndex, type, bmpBitmap, csetType);
checkIndex = checkRange.location + checkRange.length;
// Don't include any part of a composed range that extends beyond the hint range
if (checkIndex > otherIndex) {
otherIndex = checkRange.location;
break;
}
}
range.location = currentIndex;
range.length = otherIndex - currentIndex;
}
break;
}
}
++currentIndex;
}
// Regional flag
if ((range.length == 2) && __CFStringIsRegionalIndicatorAtIndex(&stringBuffer, range.location)) { // RI
// Extend backward
currentIndex = range.location;
while ((currentIndex > 1) && __CFStringIsRegionalIndicatorAtIndex(&stringBuffer, currentIndex - 2)) currentIndex -= 2;
if ((range.location > currentIndex) && (0 != ((range.location - currentIndex) % 4))) { // currentIndex is the 2nd RI
range.location -= 2;
range.length += 2;
}
if ((range.length == 2) && ((range.location + range.length + 2) <= length) && __CFStringIsRegionalIndicatorAtIndex(&stringBuffer, range.location + range.length)) {
range.length += 2;
}
}
// Attempt to expand to match pictographic sequences (Emoji and otherwise).
CFRange cluster;
if (__CFStringGetExtendedPictographicSequence(&stringBuffer, length, range.location, &cluster)) {
// We've found a pictographic cluster by the definition given in https://www.unicode.org/reports/tr29/#Regex_Definitions
// However, we have to be careful -- if we are trying to match a composed character sequence, we might end up with `cluster` not aligning with `range`.
//
// For instance,
//
// \U0001F498\u108F\u103D ≡ [HEART WITH ARROW, MYANMAR SIGN RUMAI PALAUNG TONE-5, MYANMAR CONSONANT SIGN MEDIAL WA]
//
// is in its entirety a valid composed character sequence (1F498 is a Symbol; 108F and 103D are both Marks).
//
// However, it does not form a pictographic sequence -- 1F498 is a valid Extended_Pictographic character, but 108F is specifically excluded from the SpacingMark property.
//
// So, `range` here might be {0, 4}, while `cluster` can be `{0, 2}`. We should only allow `cluster` to _extend_ `range`, not shrink it.
CFIndex const rangeEnd = range.location + range.length;
CFIndex const clusterEnd = cluster.location + cluster.length;
/* We want cluster to exclusively extend range, not just intersect it. Cases:
1. range: [ ] range.location == cluster.location && rangeEnd == clusterEnd
cluster: [ ]
2. range: [ ] range.location == cluster.location && rangeEnd < clusterEnd
cluster: [ ]
3. range: [ ] range.location > cluster.location && rangeEnd == clusterEnd
cluster: [ ]
4. range: [ ] range.location > cluster.location && rangeEnd < clusterEnd
cluster: [ ]
*/
Boolean const clusterContainsRange = (range.location >= cluster.location && rangeEnd <= clusterEnd);
// If the above is not true, the match we've found here is not useful for the semantics we're applying, and we'll ignore it.
if (clusterContainsRange) {
range = cluster;
}
}
// Gather the final grapheme extends
CFRange finalCluster;
// Backwards
if ((range.location > 0) && (range.length == 1) && (ZERO_WIDTH_JOINER == CFStringGetCharacterFromInlineBuffer(&stringBuffer, range.location))) {
finalCluster = _CFStringInlineBufferGetComposedRange(&stringBuffer, range.location - 1, type, bmpBitmap, csetType);
if (range.location == (finalCluster.location + finalCluster.length)) {
range = finalCluster;
++range.length;
}
}
// Forwards
if ((range.location + range.length) < length) {
if (ZERO_WIDTH_JOINER == CFStringGetCharacterFromInlineBuffer(&stringBuffer, range.location + range.length)) {
++range.length;
}
}
return range;
}
CFRange CFStringGetRangeOfComposedCharactersAtIndex(CFStringRef theString, CFIndex theIndex) {
return CFStringGetRangeOfCharacterClusterAtIndex(theString, theIndex, kCFStringComposedCharacterCluster);
}
/*!
@function CFStringFindCharacterFromSet
Query the range of characters contained in the specified character set.
@param theString The CFString which is to be searched. If this
parameter is not a valid CFString, the behavior is
undefined.
@param theSet The CFCharacterSet against which the membership
of characters is checked. If this parameter is not a valid
CFCharacterSet, the behavior is undefined.
@param range The range of characters within the string to search. If
the range location or end point (defined by the location
plus length minus 1) are outside the index space of the
string (0 to N-1 inclusive, where N is the length of the
string), the behavior is undefined. If the range length is
negative, the behavior is undefined. The range may be empty
(length 0), in which case no search is performed.
@param searchOptions The bitwise-or'ed option flags to control
the search behavior. The supported options are
kCFCompareBackwards andkCFCompareAnchored.
If other option flags are specified, the behavior
is undefined.
@param result The pointer to a CFRange supplied by the caller in
which the search result is stored. If a pointer to an invalid
memory is specified, the behavior is undefined.
@result true, if at least a character which is a member of the character
set is found and result is filled, otherwise, false.
*/
#define SURROGATE_START 0xD800
#define SURROGATE_END 0xDFFF
CF_EXPORT Boolean CFStringFindCharacterFromSet(CFStringRef theString, CFCharacterSetRef theSet, CFRange rangeToSearch, CFStringCompareFlags searchOptions, CFRange *result) {
CFStringInlineBuffer stringBuffer;
CFCharacterSetInlineBuffer csetBuffer;
UniChar ch;
CFIndex step;
CFIndex fromLoc, toLoc, cnt; // fromLoc and toLoc are inclusive
Boolean found = false;
Boolean done = false;
//#warning FIX ME !! Should support kCFCompareNonliteral
if ((rangeToSearch.location + rangeToSearch.length > CFStringGetLength(theString)) || (rangeToSearch.length == 0)) return false;
if (searchOptions & kCFCompareBackwards) {
fromLoc = rangeToSearch.location + rangeToSearch.length - 1;
toLoc = rangeToSearch.location;
} else {
fromLoc = rangeToSearch.location;
toLoc = rangeToSearch.location + rangeToSearch.length - 1;
}
if (searchOptions & kCFCompareAnchored) {
toLoc = fromLoc;
}
step = (fromLoc <= toLoc) ? 1 : -1;
cnt = fromLoc;
_CFStringInitInlineBufferInternal(theString, &stringBuffer, rangeToSearch, true);
CFCharacterSetInitInlineBuffer(theSet, &csetBuffer);
do {
ch = CFStringGetCharacterFromInlineBuffer(&stringBuffer, cnt - rangeToSearch.location);
if ((ch >= SURROGATE_START) && (ch <= SURROGATE_END)) {
int otherCharIndex = cnt + step;
if (((step < 0) && (otherCharIndex < toLoc)) || ((step > 0) && (otherCharIndex > toLoc))) {
done = true;
} else {
UniChar highChar;
UniChar lowChar = CFStringGetCharacterFromInlineBuffer(&stringBuffer, otherCharIndex - rangeToSearch.location);
if (cnt < otherCharIndex) {
highChar = ch;
} else {
highChar = lowChar;
lowChar = ch;
}
if (CFUniCharIsSurrogateHighCharacter(highChar) && CFUniCharIsSurrogateLowCharacter(lowChar) && CFCharacterSetInlineBufferIsLongCharacterMember(&csetBuffer, CFUniCharGetLongCharacterForSurrogatePair(highChar, lowChar))) {
if (result) *result = CFRangeMake((cnt < otherCharIndex ? cnt : otherCharIndex), 2);
return true;
} else if (otherCharIndex == toLoc) {
done = true;
} else {
cnt = otherCharIndex + step;
}
}
} else if (CFCharacterSetInlineBufferIsLongCharacterMember(&csetBuffer, ch)) {
done = found = true;
} else if (cnt == toLoc) {
done = true;
} else {
cnt += step;
}
} while (!done);
if (found && result) *result = CFRangeMake(cnt, 1);
return found;
}
/* Line range code */
#define CarriageReturn '\r' /* 0x0d */
#define NewLine '\n' /* 0x0a */
#define NextLine 0x0085
#define LineSeparator 0x2028
#define ParaSeparator 0x2029
CF_INLINE Boolean isALineSeparatorTypeCharacter(UniChar ch, Boolean includeLineEndings) {
if (ch > CarriageReturn && ch < NextLine) return false; /* Quick test to cover most chars */
return (ch == NewLine || ch == CarriageReturn || ch == ParaSeparator || (includeLineEndings && (ch == NextLine || ch == LineSeparator))) ? true : false;
}
static void __CFStringGetLineOrParagraphBounds(CFStringRef string, CFRange range, CFIndex *lineBeginIndex, CFIndex *lineEndIndex, CFIndex *contentsEndIndex, Boolean includeLineEndings) {
CFIndex len;
CFStringInlineBuffer buf;
UniChar ch;
__CFAssertIsString(string);
__CFAssertRangeIsInStringBounds(string, range.location, range.length);
len = __CFStrLength(string);
if (lineBeginIndex) {
CFIndex start;
if (range.location == 0) {
start = 0;
} else {
_CFStringInitInlineBufferInternal(string, &buf, CFRangeMake(0, len), false);
CFIndex buf_idx = range.location;
/* Take care of the special case where start happens to fall right between \r and \n */
ch = CFStringGetCharacterFromInlineBuffer(&buf, buf_idx);
buf_idx--;
if ((ch == NewLine) && (CFStringGetCharacterFromInlineBuffer(&buf, buf_idx) == CarriageReturn)) {
buf_idx--;
}
while (1) {
if (buf_idx < 0) {
start = 0;
break;
} else if (isALineSeparatorTypeCharacter(CFStringGetCharacterFromInlineBuffer(&buf, buf_idx), includeLineEndings)) {
start = buf_idx + 1;
break;
} else {
buf_idx--;
}
}
}
*lineBeginIndex = start;
}
/* Now find the ending point */
if (lineEndIndex || contentsEndIndex) {
CFIndex endOfContents, lineSeparatorLength = 1; /* 1 by default */
_CFStringInitInlineBufferInternal(string, &buf, CFRangeMake(0, len), false);
CFIndex buf_idx = range.location + range.length - (range.length ? 1 : 0);
/* First look at the last char in the range (if the range is zero length, the char after the range) to see if we're already on or within a end of line sequence... */
ch = __CFStringGetCharacterFromInlineBufferAux(&buf, buf_idx);
if (ch == NewLine) {
endOfContents = buf_idx;
buf_idx--;
if (__CFStringGetCharacterFromInlineBufferAux(&buf, buf_idx) == CarriageReturn) {
lineSeparatorLength = 2;
endOfContents--;
}
} else {
while (1) {
if (isALineSeparatorTypeCharacter(ch, includeLineEndings)) {
endOfContents = buf_idx; /* This is actually end of contentsRange */
buf_idx++; /* OK for this to go past the end */
if ((ch == CarriageReturn) && (__CFStringGetCharacterFromInlineBufferAux(&buf, buf_idx) == NewLine)) {
lineSeparatorLength = 2;
}
break;
} else if (buf_idx >= len) {
endOfContents = len;
lineSeparatorLength = 0;
break;
} else {
buf_idx++;
ch = __CFStringGetCharacterFromInlineBufferAux(&buf, buf_idx);
}
}
}
if (contentsEndIndex) *contentsEndIndex = endOfContents;
if (lineEndIndex) *lineEndIndex = endOfContents + lineSeparatorLength;
}
}
void CFStringGetLineBounds(CFStringRef string, CFRange range, CFIndex *lineBeginIndex, CFIndex *lineEndIndex, CFIndex *contentsEndIndex) {
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (NSString *)string, getLineStart:(NSUInteger *)lineBeginIndex end:(NSUInteger *)lineEndIndex contentsEnd:(NSUInteger *)contentsEndIndex forRange:NSMakeRange(range.location, range.length));
__CFStringGetLineOrParagraphBounds(string, range, lineBeginIndex, lineEndIndex, contentsEndIndex, true);
}
void CFStringGetParagraphBounds(CFStringRef string, CFRange range, CFIndex *parBeginIndex, CFIndex *parEndIndex, CFIndex *contentsEndIndex) {
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (NSString *)string, getParagraphStart:(NSUInteger *)parBeginIndex end:(NSUInteger *)parEndIndex contentsEnd:(NSUInteger *)contentsEndIndex forRange:NSMakeRange(range.location, range.length));
__CFStringGetLineOrParagraphBounds(string, range, parBeginIndex, parEndIndex, contentsEndIndex, false);
}
CFStringRef CFStringCreateByCombiningStrings(CFAllocatorRef alloc, CFArrayRef array, CFStringRef separatorString) {
CFIndex numChars;
CFIndex separatorNumByte;
CFIndex stringCount = CFArrayGetCount(array);
Boolean isSepCFString = !CF_IS_OBJC(_kCFRuntimeIDCFString, separatorString) && !CF_IS_SWIFT(_kCFRuntimeIDCFString, separatorString);
Boolean canBeEightbit = isSepCFString && __CFStrIsEightBit(separatorString);
CFIndex idx;
CFStringRef otherString;
void *buffer;
uint8_t *bufPtr;
const void *separatorContents = NULL;
if (stringCount == 0) {
return CFStringCreateWithCharacters(alloc, NULL, 0);
} else if (stringCount == 1) {
return CFStringCreateCopy(alloc, (CFStringRef)CFArrayGetValueAtIndex(array, 0));
}
if (alloc == NULL) alloc = __CFGetDefaultAllocator();
numChars = CFStringGetLength(separatorString) * (stringCount - 1);
for (idx = 0; idx < stringCount; idx++) {
otherString = (CFStringRef)CFArrayGetValueAtIndex(array, idx);
numChars += CFStringGetLength(otherString);
// canBeEightbit is already false if the separator is an NSString...
if (CF_IS_OBJC(_kCFRuntimeIDCFString, otherString) || CF_IS_SWIFT(_kCFRuntimeIDCFString, otherString) || ! __CFStrIsEightBit(otherString)) canBeEightbit = false;
}
buffer = (uint8_t *)CFAllocatorAllocate(alloc, canBeEightbit ? ((numChars + 1) * sizeof(uint8_t)) : (numChars * sizeof(UniChar)), 0);
bufPtr = (uint8_t *)buffer;
// check that bufPtr actually got allocated
if (!bufPtr) {
__CFStringHandleOutOfMemory(NULL);
}
if (__CFOASafe) __CFSetLastAllocationEventName(buffer, "CFString (store)");
separatorNumByte = CFStringGetLength(separatorString) * (canBeEightbit ? sizeof(uint8_t) : sizeof(UniChar));
for (idx = 0; idx < stringCount; idx++) {
if (idx) { // add separator here unless first string
if (separatorContents) {
memmove(bufPtr, separatorContents, separatorNumByte);
} else {
if (!isSepCFString) { // NSString
CFStringGetCharacters(separatorString, CFRangeMake(0, CFStringGetLength(separatorString)), (UniChar *)bufPtr);
} else if (canBeEightbit || __CFStrIsUnicode(separatorString)) {
memmove(bufPtr, (const uint8_t *)__CFStrContents(separatorString) + __CFStrSkipAnyLengthByte(separatorString), separatorNumByte);
} else {
__CFStrConvertBytesToUnicode((uint8_t *)__CFStrContents(separatorString) + __CFStrSkipAnyLengthByte(separatorString), (UniChar *)bufPtr, __CFStrLength(separatorString));
}
separatorContents = bufPtr;
}
bufPtr += separatorNumByte;
}
otherString = (CFStringRef )CFArrayGetValueAtIndex(array, idx);
if (CF_IS_OBJC(_kCFRuntimeIDCFString, otherString) || CF_IS_SWIFT(_kCFRuntimeIDCFString, otherString)) {
CFIndex otherLength = CFStringGetLength(otherString);
CFStringGetCharacters(otherString, CFRangeMake(0, otherLength), (UniChar *)bufPtr);
bufPtr += otherLength * sizeof(UniChar);
} else {
const uint8_t * otherContents = (const uint8_t *)__CFStrContents(otherString);
CFIndex otherNumByte = __CFStrLength2(otherString, otherContents) * (canBeEightbit ? sizeof(uint8_t) : sizeof(UniChar));
if (canBeEightbit || __CFStrIsUnicode(otherString)) {
memmove(bufPtr, otherContents + __CFStrSkipAnyLengthByte(otherString), otherNumByte);
} else {
__CFStrConvertBytesToUnicode(otherContents + __CFStrSkipAnyLengthByte(otherString), (UniChar *)bufPtr, __CFStrLength2(otherString, otherContents));
}
bufPtr += otherNumByte;
}
}
if (canBeEightbit) *bufPtr = 0; // NULL byte;
return canBeEightbit ?
CFStringCreateWithCStringNoCopy(alloc, (const char*)buffer, __CFStringGetEightBitStringEncoding(), alloc) :
CFStringCreateWithCharactersNoCopy(alloc, (UniChar *)buffer, numChars, alloc);
}
CFArrayRef CFStringCreateArrayBySeparatingStrings(CFAllocatorRef alloc, CFStringRef string, CFStringRef separatorString) {
CFArrayRef separatorRanges;
CFIndex length = CFStringGetLength(string);
/* No objc dispatch needed here since CFStringCreateArrayWithFindResults() works with both CFString and NSString */
if (!(separatorRanges = CFStringCreateArrayWithFindResults(alloc, string, separatorString, CFRangeMake(0, length), 0))) {
return CFArrayCreate(alloc, (const void **)&string, 1, & kCFTypeArrayCallBacks);
} else {
CFIndex idx;
CFIndex count = CFArrayGetCount(separatorRanges);
CFIndex startIndex = 0;
CFIndex numChars;
CFMutableArrayRef array = CFArrayCreateMutable(alloc, count + 2, & kCFTypeArrayCallBacks);
const CFRange *currentRange;
CFStringRef substring;
for (idx = 0;idx < count;idx++) {
currentRange = (const CFRange *)CFArrayGetValueAtIndex(separatorRanges, idx);
numChars = currentRange->location - startIndex;
substring = CFStringCreateWithSubstring(alloc, string, CFRangeMake(startIndex, numChars));
CFArrayAppendValue(array, substring);
CFRelease(substring);
startIndex = currentRange->location + currentRange->length;
}
substring = CFStringCreateWithSubstring(alloc, string, CFRangeMake(startIndex, length - startIndex));
CFArrayAppendValue(array, substring);
CFRelease(substring);
CFRelease(separatorRanges);
return array;
}
}
CFStringRef CFStringCreateFromExternalRepresentation(CFAllocatorRef alloc, CFDataRef data, CFStringEncoding encoding) {
return CFStringCreateWithBytes(alloc, CFDataGetBytePtr(data), CFDataGetLength(data), encoding, true);
}
CFDataRef CFStringCreateExternalRepresentation(CFAllocatorRef alloc, CFStringRef string, CFStringEncoding encoding, uint8_t lossByte) {
CFIndex length;
CFIndex guessedByteLength;
uint8_t *bytes;
CFIndex usedLength;
SInt32 result;
if (CF_IS_OBJC(_kCFRuntimeIDCFString, string) || CF_IS_SWIFT(_kCFRuntimeIDCFString, string)) { /* ??? Hope the compiler optimizes this away if OBJC_MAPPINGS is not on */
length = CFStringGetLength(string);
} else {
__CFAssertIsString(string);
length = __CFStrLength(string);
if (__CFStrIsEightBit(string) && ((__CFStringGetEightBitStringEncoding() == encoding) || (__CFStringGetEightBitStringEncoding() == kCFStringEncodingASCII && __CFStringEncodingIsSupersetOfASCII(encoding)))) { // Requested encoding is equal to the encoding in string
return CFDataCreate(alloc, ((uint8_t *)__CFStrContents(string) + __CFStrSkipAnyLengthByte(string)), __CFStrLength(string));
}
}
if (alloc == NULL) alloc = __CFGetDefaultAllocator();
if (((encoding & 0x0FFF) == kCFStringEncodingUnicode) && ((encoding == kCFStringEncodingUnicode) || ((encoding > kCFStringEncodingUTF8) && (encoding <= kCFStringEncodingUTF32LE)))) {
guessedByteLength = (length + 1) * ((((encoding >> 26) & 2) == 0) ? sizeof(UTF16Char) : sizeof(UTF32Char)); // UTF32 format has the bit set
} else if (((guessedByteLength = CFStringGetMaximumSizeForEncoding(length, encoding)) > length) && !CF_IS_OBJC(_kCFRuntimeIDCFString, string) && !CF_IS_SWIFT(_kCFRuntimeIDCFString, string)) { // Multi byte encoding
#if TARGET_OS_MAC || TARGET_OS_LINUX || TARGET_OS_BSD
if (__CFStrIsUnicode(string)) {
CFIndex aLength = CFStringEncodingByteLengthForCharacters(encoding, kCFStringEncodingPrependBOM, __CFStrContents(string), __CFStrLength(string));
if (aLength > 0) guessedByteLength = aLength;
} else {
#endif
result = __CFStringEncodeByteStream(string, 0, length, true, encoding, lossByte, NULL, LONG_MAX, &guessedByteLength);
// if result == length, we always succeed
// otherwise, if result == 0, we fail
// otherwise, if there was a lossByte but still result != length, we fail
if ((result != length) && (!result || !lossByte)) return NULL;
if (guessedByteLength == length && __CFStrIsEightBit(string) && __CFStringEncodingIsSupersetOfASCII(encoding)) { // It's all ASCII !!
return CFDataCreate(alloc, ((uint8_t *)__CFStrContents(string) + __CFStrSkipAnyLengthByte(string)), __CFStrLength(string));
}
#if TARGET_OS_MAC || TARGET_OS_LINUX || TARGET_OS_BSD
}
#endif
}
bytes = (uint8_t *)CFAllocatorAllocate(alloc, guessedByteLength, 0);
if (__CFOASafe) __CFSetLastAllocationEventName(bytes, "CFData (store)");
result = __CFStringEncodeByteStream(string, 0, length, true, encoding, lossByte, bytes, guessedByteLength, &usedLength);
if ((result != length) && (!result || !lossByte)) { // see comment above about what this means
CFAllocatorDeallocate(alloc, bytes);
return NULL;
}
return CFDataCreateWithBytesNoCopy(alloc, (uint8_t *)bytes, usedLength, alloc);
}
CFStringEncoding CFStringGetSmallestEncoding(CFStringRef str) {
CFIndex len;
if (CF_IS_SWIFT(_kCFRuntimeIDCFString, str)) {
return kCFStringEncodingUnicode;
}
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, CFStringEncoding, (NSString *)str, _smallestEncodingInCFStringEncoding);
__CFAssertIsString(str);
if (__CFStrIsEightBit(str)) return __CFStringGetEightBitStringEncoding();
len = __CFStrLength(str);
if (__CFStringEncodeByteStream(str, 0, len, false, __CFStringGetEightBitStringEncoding(), 0, NULL, LONG_MAX, NULL) == len) return __CFStringGetEightBitStringEncoding();
if ((__CFStringGetEightBitStringEncoding() != __CFStringGetSystemEncoding()) && (__CFStringEncodeByteStream(str, 0, len, false, __CFStringGetSystemEncoding(), 0, NULL, LONG_MAX, NULL) == len)) return __CFStringGetSystemEncoding();
return kCFStringEncodingUnicode; /* ??? */
}
CFStringEncoding CFStringGetFastestEncoding(CFStringRef str) {
if (CF_IS_SWIFT(_kCFRuntimeIDCFString, str)) {
return kCFStringEncodingUnicode;
}
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, CFStringEncoding, (NSString *)str, _fastestEncodingInCFStringEncoding);
__CFAssertIsString(str);
return __CFStrIsEightBit(str) ? __CFStringGetEightBitStringEncoding() : kCFStringEncodingUnicode; /* ??? */
}
SInt32 CFStringGetIntValue(CFStringRef str) {
Boolean success;
SInt32 result;
SInt32 idx = 0;
CFStringInlineBuffer buf;
_CFStringInitInlineBufferInternal(str, &buf, CFRangeMake(0, CFStringGetLength(str)), true);
success = __CFStringScanInteger(&buf, NULL, &idx, false, &result);
return success ? result : 0;
}
double CFStringGetDoubleValue(CFStringRef str) {
Boolean success;
double result;
SInt32 idx = 0;
CFStringInlineBuffer buf;
_CFStringInitInlineBufferInternal(str, &buf, CFRangeMake(0, CFStringGetLength(str)), true);
success = __CFStringScanDouble(&buf, NULL, &idx, &result);
return success ? result : 0.0;
}
/*** Mutable functions... ***/
void CFStringSetExternalCharactersNoCopy(CFMutableStringRef string, UniChar *chars, CFIndex length, CFIndex capacity) {
__CFAssertIsNotNegative(length);
__CFAssertIsStringAndExternalMutable(string);
CF_RETURN_IF_NOT_MUTABLE(string);
CFAssert4((length <= capacity) && ((capacity == 0) || ((capacity > 0) && chars)), __kCFLogAssertion, "%s(): Invalid args: characters %p length %ld capacity %ld", __PRETTY_FUNCTION__, chars, length, capacity);
__CFStrSetContentPtr(string, chars);
__CFStrSetExplicitLength(string, length);
__CFStrSetCapacity(string, capacity * sizeof(UniChar));
__CFStrSetCapacityProvidedExternally(string);
}
void CFStringInsert(CFMutableStringRef str, CFIndex idx, CFStringRef insertedStr) {
CF_SWIFT_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (CFSwiftRef)str, NSMutableString.insertString, idx, (CFSwiftRef)insertedStr);
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (NSMutableString *)str, insertString:(NSString *)insertedStr atIndex:(NSUInteger)idx);
CF_RETURN_IF_NOT_MUTABLE(str);
CFAssert3(idx >= 0 && idx <= __CFStrLength(str), __kCFLogAssertion, "%s(): string index %ld out of bounds (length %ld)", __PRETTY_FUNCTION__, idx, __CFStrLength(str));
__CFStringReplace(str, CFRangeMake(idx, 0), insertedStr);
}
void CFStringDelete(CFMutableStringRef str, CFRange range) {
CF_SWIFT_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (CFSwiftRef)str, NSMutableString.deleteCharactersInRange, range);
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (NSMutableString *)str, deleteCharactersInRange:NSMakeRange(range.location, range.length));
CF_RETURN_IF_NOT_MUTABLE(str);
__CFAssertRangeIsInStringBounds(str, range.location, range.length);
__CFStringChangeSize(str, range, 0, false);
}
void CFStringReplace(CFMutableStringRef str, CFRange range, CFStringRef replacement) {
CF_SWIFT_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (CFSwiftRef)str, NSMutableString.replaceCharactersInRange, range, (CFSwiftRef)replacement);
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (NSMutableString *)str, replaceCharactersInRange:NSMakeRange(range.location, range.length) withString:(NSString *)replacement);
CF_RETURN_IF_NOT_MUTABLE(str);
__CFAssertRangeIsInStringBounds(str, range.location, range.length);
__CFStringReplace(str, range, replacement);
}
void CFStringReplaceAll(CFMutableStringRef str, CFStringRef replacement) {
CF_SWIFT_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (CFSwiftRef)str, NSMutableString.setString, (CFSwiftRef)replacement);
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (NSMutableString *)str, setString:(NSString *)replacement);
CF_RETURN_IF_NOT_MUTABLE(str);
__CFStringReplace(str, CFRangeMake(0, __CFStrLength(str)), replacement);
}
void CFStringAppend(CFMutableStringRef str, CFStringRef appended) {
CF_SWIFT_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (CFSwiftRef)str, NSMutableString.appendString, (CFSwiftRef)appended);
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (NSMutableString *)str, appendString:(NSString *)appended);
CF_RETURN_IF_NOT_MUTABLE(str);
__CFStringReplace(str, CFRangeMake(__CFStrLength(str), 0), appended);
}
void CFStringAppendCharacters(CFMutableStringRef str, const UniChar *chars, CFIndex appendedLength) {
CFIndex strLength, idx;
__CFAssertIsNotNegative(appendedLength);
CF_SWIFT_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (CFSwiftRef)str, NSMutableString.appendCharacters, chars, appendedLength);
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (NSMutableString *)str, appendCharacters:chars length:(NSUInteger)appendedLength);
CF_RETURN_IF_NOT_MUTABLE(str);
strLength = __CFStrLength(str);
if (__CFStrIsUnicode(str)) {
__CFStringChangeSize(str, CFRangeMake(strLength, 0), appendedLength, true);
memmove((UniChar *)__CFStrContents(str) + strLength, chars, appendedLength * sizeof(UniChar));
} else {
uint8_t *contents;
bool isASCII = true;
for (idx = 0; isASCII && idx < appendedLength; idx++) isASCII = (chars[idx] < 0x80);
__CFStringChangeSize(str, CFRangeMake(strLength, 0), appendedLength, !isASCII);
if (!isASCII) {
memmove((UniChar *)__CFStrContents(str) + strLength, chars, appendedLength * sizeof(UniChar));
} else {
contents = (uint8_t *)__CFStrContents(str) + strLength + __CFStrSkipAnyLengthByte(str);
for (idx = 0; idx < appendedLength; idx++) contents[idx] = (uint8_t)chars[idx];
}
}
}
void __CFStringAppendBytes(CFMutableStringRef str, const char *cStr, CFIndex appendedLength, CFStringEncoding encoding) {
Boolean appendedIsUnicode = false;
Boolean freeCStrWhenDone = false;
Boolean demoteAppendedUnicode = false;
CFVarWidthCharBuffer vBuf;
__CFAssertIsNotNegative(appendedLength);
if (encoding == kCFStringEncodingASCII || encoding == __CFStringGetEightBitStringEncoding()) {
// appendedLength now denotes length in UniChars
} else if (encoding == kCFStringEncodingUnicode) {
UniChar *chars = (UniChar *)cStr;
CFIndex idx, length = appendedLength / sizeof(UniChar);
bool isASCII = true;
for (idx = 0; isASCII && idx < length; idx++) isASCII = (chars[idx] < 0x80);
if (!isASCII) {
appendedIsUnicode = true;
} else {
demoteAppendedUnicode = true;
}
appendedLength = length;
} else {
Boolean usingPassedInMemory = false;
vBuf.allocator = __CFGetDefaultAllocator(); // We don't want to use client's allocator for temp stuff
vBuf.chars.unicode = NULL; // This will cause the decode function to allocate memory if necessary
if (!__CFStringDecodeByteStream3((const uint8_t *)cStr, appendedLength, encoding, __CFStrIsUnicode(str), &vBuf, &usingPassedInMemory, 0)) {
CFAssert1(0, __kCFLogAssertion, "Supplied bytes could not be converted specified encoding %ud", (unsigned int)encoding);
return;
}
// If not ASCII, appendedLength now denotes length in UniChars
appendedLength = vBuf.numChars;
appendedIsUnicode = !vBuf.isASCII;
cStr = (const char *)vBuf.chars.ascii;
freeCStrWhenDone = !usingPassedInMemory && vBuf.shouldFreeChars;
}
if (CF_IS_OBJC(_kCFRuntimeIDCFString, str)) {
if (!appendedIsUnicode && !demoteAppendedUnicode) {
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (NSMutableString *)str, _cfAppendCString:(const unsigned char *)cStr length:(NSInteger)appendedLength);
} else {
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (NSMutableString *)str, appendCharacters:(const unichar *)cStr length:(NSUInteger)appendedLength);
}
}
#if DEPLOYMENT_RUNTIME_SWIFT
else if (CF_IS_SWIFT(_kCFRuntimeIDCFString, str)) {
if (!appendedIsUnicode && !demoteAppendedUnicode) {
CF_SWIFT_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (CFSwiftRef)str, NSMutableString._cfAppendCString,(const char *)cStr, appendedLength);
} else {
CF_SWIFT_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (CFSwiftRef)str, NSMutableString.appendCharacters, (const UniChar *)cStr, appendedLength);
}
}
#endif
else {
CFIndex strLength;
__CFAssertIsStringAndMutable(str);
strLength = __CFStrLength(str);
__CFStringChangeSize(str, CFRangeMake(strLength, 0), appendedLength, appendedIsUnicode || __CFStrIsUnicode(str));
if (__CFStrIsUnicode(str)) {
UniChar *contents = (UniChar *)__CFStrContents(str);
if (appendedIsUnicode) {
memmove(contents + strLength, cStr, appendedLength * sizeof(UniChar));
} else {
__CFStrConvertBytesToUnicode((const uint8_t *)cStr, contents + strLength, appendedLength);
}
} else {
if (demoteAppendedUnicode) {
UniChar *chars = (UniChar *)cStr;
CFIndex idx;
uint8_t *contents = (uint8_t *)__CFStrContents(str) + strLength + __CFStrSkipAnyLengthByte(str);
for (idx = 0; idx < appendedLength; idx++) contents[idx] = (uint8_t)chars[idx];
} else {
uint8_t *contents = (uint8_t *)__CFStrContents(str);
memmove(contents + strLength + __CFStrSkipAnyLengthByte(str), cStr, appendedLength);
}
}
}
if (freeCStrWhenDone) CFAllocatorDeallocate(__CFGetDefaultAllocator(), (void *)cStr);
}
void CFStringAppendPascalString(CFMutableStringRef str, ConstStringPtr pStr, CFStringEncoding encoding) {
CF_RETURN_IF_NOT_MUTABLE(str);
__CFStringAppendBytes(str, (const char *)(pStr + 1), (CFIndex)*pStr, encoding);
}
void CFStringAppendCString(CFMutableStringRef str, const char *cStr, CFStringEncoding encoding) {
CF_RETURN_IF_NOT_MUTABLE(str);
__CFStringAppendBytes(str, cStr, strlen(cStr), encoding);
}
void CFStringAppendFormat(CFMutableStringRef str, CFDictionaryRef formatOptions, CFStringRef format, ...) {
CF_RETURN_IF_NOT_MUTABLE(str);
va_list argList;
va_start(argList, format);
CFStringAppendFormatAndArguments(str, formatOptions, format, argList);
va_end(argList);
}
CFIndex CFStringFindAndReplace(CFMutableStringRef string, CFStringRef stringToFind, CFStringRef replacementString, CFRange rangeToSearch, CFStringCompareFlags compareOptions) {
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, CFIndex, (NSMutableString *)string, replaceOccurrencesOfString:(NSString *)stringToFind withString:(NSString *)replacementString options:(NSStringCompareOptions)compareOptions range:NSMakeRange(rangeToSearch.location, rangeToSearch.length));
CFRange foundRange;
Boolean backwards = ((compareOptions & kCFCompareBackwards) != 0);
UInt32 endIndex = rangeToSearch.location + rangeToSearch.length;
#define MAX_RANGES_ON_STACK (1000 / sizeof(CFRange))
CFRange rangeBuffer[MAX_RANGES_ON_STACK]; // Used to avoid allocating memory
CFRange *ranges = rangeBuffer;
CFIndex foundCount = 0;
CFIndex capacity = MAX_RANGES_ON_STACK;
__CFAssertRangeIsInStringBounds(string, rangeToSearch.location, rangeToSearch.length);
// Note: This code is very similar to the one in CFStringCreateArrayWithFindResults().
while ((rangeToSearch.length > 0) && CFStringFindWithOptions(string, stringToFind, rangeToSearch, compareOptions, &foundRange)) {
// Determine the next range
if (backwards) {
rangeToSearch.length = foundRange.location - rangeToSearch.location;
} else {
rangeToSearch.location = foundRange.location + foundRange.length;
rangeToSearch.length = endIndex - rangeToSearch.location;
}
// If necessary, grow the array
if (foundCount >= capacity) {
bool firstAlloc = (ranges == rangeBuffer) ? true : false;
capacity = (capacity + 4) * 2;
// Note that reallocate with NULL previous pointer is same as allocate
ranges = __CFSafelyReallocateWithAllocator(kCFAllocatorSystemDefault, firstAlloc ? NULL : ranges, capacity * sizeof(CFRange), 0, NULL);
if (firstAlloc) memmove(ranges, rangeBuffer, MAX_RANGES_ON_STACK * sizeof(CFRange));
}
ranges[foundCount] = foundRange;
foundCount++;
}
if (foundCount > 0) {
if (backwards) { // Reorder the ranges to be incrementing (better to do this here, then to check other places)
int head = 0;
int tail = foundCount - 1;
while (head < tail) {
CFRange temp = ranges[head];
ranges[head] = ranges[tail];
ranges[tail] = temp;
head++;
tail--;
}
}
int err = __CFStringReplaceMultiple(string, ranges, foundCount, replacementString);
if (err == _CFStringErrNotMutable) {
os_log_fault(_CFOSLog(), "CFString: %s(): Expect mutable string", __PRETTY_FUNCTION__);
}
if (ranges != rangeBuffer) CFAllocatorDeallocate(kCFAllocatorSystemDefault, ranges);
}
return foundCount;
}
// This function is here for NSString purposes
// It allows checking for mutability before mutating; this allows NSString to catch invalid mutations
int __CFStringCheckAndReplace(CFMutableStringRef str, CFRange range, CFStringRef replacement) {
if (!__CFStrIsMutable(str)) return _CFStringErrNotMutable; // These three ifs are always here, for NSString usage
if (!replacement) return _CFStringErrNilArg;
// This attempts to catch bad ranges including those described in 3375535 (-1,1)
unsigned long endOfRange = (unsigned long)(range.location) + (unsigned long)(range.length); // NSRange uses unsigned quantities, hence the casting
if ((endOfRange > (unsigned long)__CFStrLength(str)) || (endOfRange < (unsigned long)(range.location))) return _CFStringErrBounds;
__CFAssertIsStringAndMutable(str);
__CFAssertRangeIsInStringBounds(str, range.location, range.length);
__CFStringReplace(str, range, replacement);
return _CFStringErrNone;
}
// This function determines whether errors which would cause string exceptions should
// be ignored or not
Boolean __CFStringNoteErrors(void) {
return true;
}
void CFStringPad(CFMutableStringRef string, CFStringRef padString, CFIndex length, CFIndex indexIntoPad) {
CFIndex originalLength;
__CFAssertIsNotNegative(length);
__CFAssertIsNotNegative(indexIntoPad);
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (NSMutableString *)string, _cfPad:padString length:(uint32_t)length padIndex:(uint32_t)indexIntoPad);
CF_RETURN_IF_NOT_MUTABLE(string);
originalLength = __CFStrLength(string);
if (length < originalLength) {
__CFStringChangeSize(string, CFRangeMake(length, originalLength - length), 0, false);
} else if (originalLength < length) {
uint8_t *contents;
Boolean isUnicode;
CFIndex charSize;
CFIndex padStringLength;
CFIndex padLength;
CFIndex padRemaining = length - originalLength;
if (CF_IS_OBJC(_kCFRuntimeIDCFString, padString) || CF_IS_SWIFT(_kCFRuntimeIDCFString, padString)) { /* ??? Hope the compiler optimizes this away if OBJC_MAPPINGS is not on */
padStringLength = CFStringGetLength(padString);
isUnicode = true; /* !!! Bad for now */
} else {
__CFAssertIsString(padString);
padStringLength = __CFStrLength(padString);
isUnicode = __CFStrIsUnicode(string) || __CFStrIsUnicode(padString);
}
charSize = isUnicode ? sizeof(UniChar) : sizeof(uint8_t);
__CFStringChangeSize(string, CFRangeMake(originalLength, 0), padRemaining, isUnicode);
contents = (uint8_t *)__CFStrContents(string) + charSize * originalLength + __CFStrSkipAnyLengthByte(string);
padLength = padStringLength - indexIntoPad;
padLength = padRemaining < padLength ? padRemaining : padLength;
while (padRemaining > 0) {
if (isUnicode) {
CFStringGetCharacters(padString, CFRangeMake(indexIntoPad, padLength), (UniChar *)contents);
} else {
CFStringGetBytes(padString, CFRangeMake(indexIntoPad, padLength), __CFStringGetEightBitStringEncoding(), 0, false, contents, padRemaining * charSize, NULL);
}
contents += padLength * charSize;
padRemaining -= padLength;
indexIntoPad = 0;
padLength = padRemaining < padLength ? padRemaining : padStringLength;
}
}
}
void CFStringTrim(CFMutableStringRef string, CFStringRef trimString) {
CFRange range;
CFIndex newStartIndex;
CFIndex length;
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (NSMutableString *)string, _cfTrim:trimString);
CF_RETURN_IF_NOT_MUTABLE(string);
newStartIndex = 0;
length = __CFStrLength(string);
while (CFStringFindWithOptions(string, trimString, CFRangeMake(newStartIndex, length - newStartIndex), kCFCompareAnchored, &range)) {
newStartIndex = range.location + range.length;
}
if (newStartIndex < length) {
CFIndex charSize = __CFStrIsUnicode(string) ? sizeof(UniChar) : sizeof(uint8_t);
uint8_t *contents = (uint8_t *)__CFStrContents(string) + __CFStrSkipAnyLengthByte(string);
length -= newStartIndex;
if (CFStringGetLength(trimString) < length) {
while (CFStringFindWithOptions(string, trimString, CFRangeMake(newStartIndex, length), kCFCompareAnchored|kCFCompareBackwards, &range)) {
length = range.location - newStartIndex;
}
}
memmove(contents, contents + newStartIndex * charSize, length * charSize);
__CFStringChangeSize(string, CFRangeMake(length, __CFStrLength(string) - length), 0, false);
} else { // Only trimString in string, trim all
__CFStringChangeSize(string, CFRangeMake(0, length), 0, false);
}
}
void CFStringTrimWhitespace(CFMutableStringRef string) {
CFIndex newStartIndex;
CFIndex length;
CFStringInlineBuffer buffer;
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (NSMutableString *)string, _cfTrimWS);
CF_RETURN_IF_NOT_MUTABLE(string);
newStartIndex = 0;
length = __CFStrLength(string);
_CFStringInitInlineBufferInternal(string, &buffer, CFRangeMake(0, length), false /* already did CF_OBJC_FUNCDISPATCHV above */);
CFIndex buffer_idx = 0;
while (buffer_idx < length && CFUniCharIsMemberOf(__CFStringGetCharacterFromInlineBufferQuick(&buffer, buffer_idx), kCFUniCharWhitespaceAndNewlineCharacterSet))
buffer_idx++;
newStartIndex = buffer_idx;
if (newStartIndex < length) {
uint8_t *contents = (uint8_t *)__CFStrContents(string) + __CFStrSkipAnyLengthByte(string);
CFIndex charSize = (__CFStrIsUnicode(string) ? sizeof(UniChar) : sizeof(uint8_t));
buffer_idx = length - 1;
while (0 <= buffer_idx && CFUniCharIsMemberOf(__CFStringGetCharacterFromInlineBufferQuick(&buffer, buffer_idx), kCFUniCharWhitespaceAndNewlineCharacterSet))
buffer_idx--;
length = buffer_idx - newStartIndex + 1;
memmove(contents, contents + newStartIndex * charSize, length * charSize);
__CFStringChangeSize(string, CFRangeMake(length, __CFStrLength(string) - length), 0, false);
} else { // Whitespace only string
__CFStringChangeSize(string, CFRangeMake(0, length), 0, false);
}
}
void CFStringLowercase(CFMutableStringRef string, CFLocaleRef locale) {
CFIndex currentIndex = 0;
CFIndex length;
const uint8_t *langCode;
Boolean isEightBit = __CFStrIsEightBit(string);
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (NSMutableString *)string, _cfLowercase:(const void *)locale);
CF_RETURN_IF_NOT_MUTABLE(string);
length = __CFStrLength(string);
langCode = (const uint8_t *)(_CFCanUseLocale(locale) ? _CFStrGetSpecialCaseHandlingLanguageIdentifierForLocale(locale, false) : NULL);
if (!langCode && isEightBit) {
uint8_t *contents = (uint8_t *)__CFStrContents(string) + __CFStrSkipAnyLengthByte(string);
for (;currentIndex < length;currentIndex++) {
if (contents[currentIndex] >= 'A' && contents[currentIndex] <= 'Z') {
contents[currentIndex] += 'a' - 'A';
} else if (contents[currentIndex] > 127) {
break;
}
}
}
if (currentIndex < length) {
UTF16Char *contents;
UniChar mappedCharacters[MAX_CASE_MAPPING_BUF];
CFIndex mappedLength;
UTF32Char currentChar;
UInt32 flags = 0;
if (isEightBit) __CFStringChangeSize(string, CFRangeMake(0, 0), 0, true);
contents = (UniChar *)__CFStrContents(string);
for (;currentIndex < length;currentIndex++) {
if (CFUniCharIsSurrogateHighCharacter(contents[currentIndex]) && (currentIndex + 1 < length) && CFUniCharIsSurrogateLowCharacter(contents[currentIndex + 1])) {
currentChar = CFUniCharGetLongCharacterForSurrogatePair(contents[currentIndex], contents[currentIndex + 1]);
} else {
currentChar = contents[currentIndex];
}
flags = ((langCode || (currentChar == 0x03A3)) ? CFUniCharGetConditionalCaseMappingFlags(currentChar, contents, currentIndex, length, kCFUniCharToLowercase, langCode, flags) : 0);
mappedLength = CFUniCharMapCaseTo(currentChar, mappedCharacters, MAX_CASE_MAPPING_BUF, kCFUniCharToLowercase, flags, langCode);
if (mappedLength > 0) contents[currentIndex] = *mappedCharacters;
if (currentChar > 0xFFFF) { // Non-BMP char
switch (mappedLength) {
case 0:
__CFStringChangeSize(string, CFRangeMake(currentIndex, 2), 0, true);
contents = (UniChar *)__CFStrContents(string);
length -= 2;
break;
case 1:
__CFStringChangeSize(string, CFRangeMake(currentIndex + 1, 1), 0, true);
contents = (UniChar *)__CFStrContents(string);
--length;
break;
case 2:
contents[++currentIndex] = mappedCharacters[1];
break;
default:
--mappedLength; // Skip the current char
__CFStringChangeSize(string, CFRangeMake(currentIndex + 1, 0), mappedLength - 1, true);
contents = (UniChar *)__CFStrContents(string);
memmove(contents + currentIndex + 1, mappedCharacters + 1, mappedLength * sizeof(UniChar));
length += (mappedLength - 1);
currentIndex += mappedLength;
break;
}
} else if (mappedLength == 0) {
__CFStringChangeSize(string, CFRangeMake(currentIndex, 1), 0, true);
contents = (UniChar *)__CFStrContents(string);
--length;
} else if (mappedLength > 1) {
--mappedLength; // Skip the current char
__CFStringChangeSize(string, CFRangeMake(currentIndex + 1, 0), mappedLength, true);
contents = (UniChar *)__CFStrContents(string);
memmove(contents + currentIndex + 1, mappedCharacters + 1, mappedLength * sizeof(UniChar));
length += mappedLength;
currentIndex += mappedLength;
}
}
}
}
void CFStringUppercase(CFMutableStringRef string, CFLocaleRef locale) {
CFIndex currentIndex = 0;
CFIndex length;
const uint8_t *langCode;
Boolean isEightBit = __CFStrIsEightBit(string);
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (NSMutableString *)string, _cfUppercase:(const void *)locale);
CF_RETURN_IF_NOT_MUTABLE(string);
length = __CFStrLength(string);
langCode = (const uint8_t *)(_CFCanUseLocale(locale) ? _CFStrGetSpecialCaseHandlingLanguageIdentifierForLocale(locale, false) : NULL);
if (!langCode && isEightBit) {
uint8_t *contents = (uint8_t *)__CFStrContents(string) + __CFStrSkipAnyLengthByte(string);
for (;currentIndex < length;currentIndex++) {
if (contents[currentIndex] >= 'a' && contents[currentIndex] <= 'z') {
contents[currentIndex] -= 'a' - 'A';
} else if (contents[currentIndex] > 127) {
break;
}
}
}
if (currentIndex < length) {
UniChar *contents;
UniChar mappedCharacters[MAX_CASE_MAPPING_BUF];
CFIndex mappedLength;
UTF32Char currentChar;
UInt32 flags = 0;
if (isEightBit) __CFStringChangeSize(string, CFRangeMake(0, 0), 0, true);
contents = (UniChar *)__CFStrContents(string);
for (;currentIndex < length;currentIndex++) {
if (CFUniCharIsSurrogateHighCharacter(contents[currentIndex]) && (currentIndex + 1 < length) && CFUniCharIsSurrogateLowCharacter(contents[currentIndex + 1])) {
currentChar = CFUniCharGetLongCharacterForSurrogatePair(contents[currentIndex], contents[currentIndex + 1]);
} else {
currentChar = contents[currentIndex];
}
flags = (langCode ? CFUniCharGetConditionalCaseMappingFlags(currentChar, contents, currentIndex, length, kCFUniCharToUppercase, langCode, flags) : 0);
mappedLength = CFUniCharMapCaseTo(currentChar, mappedCharacters, MAX_CASE_MAPPING_BUF, kCFUniCharToUppercase, flags, langCode);
if (mappedLength > 0) contents[currentIndex] = *mappedCharacters;
if (currentChar > 0xFFFF) { // Non-BMP char
switch (mappedLength) {
case 0:
__CFStringChangeSize(string, CFRangeMake(currentIndex, 2), 0, true);
contents = (UniChar *)__CFStrContents(string);
length -= 2;
break;
case 1:
__CFStringChangeSize(string, CFRangeMake(currentIndex + 1, 1), 0, true);
contents = (UniChar *)__CFStrContents(string);
--length;
break;
case 2:
contents[++currentIndex] = mappedCharacters[1];
break;
default:
--mappedLength; // Skip the current char
__CFStringChangeSize(string, CFRangeMake(currentIndex + 1, 0), mappedLength - 1, true);
contents = (UniChar *)__CFStrContents(string);
memmove(contents + currentIndex + 1, mappedCharacters + 1, mappedLength * sizeof(UniChar));
length += (mappedLength - 1);
currentIndex += mappedLength;
break;
}
} else if (mappedLength == 0) {
__CFStringChangeSize(string, CFRangeMake(currentIndex, 1), 0, true);
contents = (UniChar *)__CFStrContents(string);
--length;
} else if (mappedLength > 1) {
--mappedLength; // Skip the current char
__CFStringChangeSize(string, CFRangeMake(currentIndex + 1, 0), mappedLength, true);
contents = (UniChar *)__CFStrContents(string);
memmove(contents + currentIndex + 1, mappedCharacters + 1, mappedLength * sizeof(UniChar));
length += mappedLength;
currentIndex += mappedLength;
}
}
}
}
void CFStringCapitalize(CFMutableStringRef string, CFLocaleRef locale) {
CFIndex currentIndex = 0;
CFIndex length;
const uint8_t *langCode;
Boolean isEightBit = __CFStrIsEightBit(string);
Boolean isLastCased = false;
const uint8_t *caseIgnorableForBMP;
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (NSMutableString *)string, _cfCapitalize:(const void *)locale);
CF_RETURN_IF_NOT_MUTABLE(string);
length = __CFStrLength(string);
caseIgnorableForBMP = CFUniCharGetBitmapPtrForPlane(kCFUniCharCaseIgnorableCharacterSet, 0);
langCode = (const uint8_t *)(_CFCanUseLocale(locale) ? _CFStrGetSpecialCaseHandlingLanguageIdentifierForLocale(locale, false) : NULL);
if (!langCode && isEightBit) {
uint8_t *contents = (uint8_t *)__CFStrContents(string) + __CFStrSkipAnyLengthByte(string);
for (;currentIndex < length;currentIndex++) {
if (contents[currentIndex] > 127) {
break;
} else if (contents[currentIndex] >= 'A' && contents[currentIndex] <= 'Z') {
contents[currentIndex] += (isLastCased ? 'a' - 'A' : 0);
isLastCased = true;
} else if (contents[currentIndex] >= 'a' && contents[currentIndex] <= 'z') {
contents[currentIndex] -= (!isLastCased ? 'a' - 'A' : 0);
isLastCased = true;
} else if (!CFUniCharIsMemberOfBitmap(contents[currentIndex], caseIgnorableForBMP)) {
isLastCased = false;
}
}
}
if (currentIndex < length) {
UniChar *contents;
UniChar mappedCharacters[MAX_CASE_MAPPING_BUF];
CFIndex mappedLength;
UTF32Char currentChar;
UInt32 flags = 0;
if (isEightBit) __CFStringChangeSize(string, CFRangeMake(0, 0), 0, true);
contents = (UniChar *)__CFStrContents(string);
for (;currentIndex < length;currentIndex++) {
if (CFUniCharIsSurrogateHighCharacter(contents[currentIndex]) && (currentIndex + 1 < length) && CFUniCharIsSurrogateLowCharacter(contents[currentIndex + 1])) {
currentChar = CFUniCharGetLongCharacterForSurrogatePair(contents[currentIndex], contents[currentIndex + 1]);
} else {
currentChar = contents[currentIndex];
}
flags = ((langCode || ((currentChar == 0x03A3) && isLastCased)) ? CFUniCharGetConditionalCaseMappingFlags(currentChar, contents, currentIndex, length, (isLastCased ? kCFUniCharToLowercase : kCFUniCharToTitlecase), langCode, flags) : 0);
mappedLength = CFUniCharMapCaseTo(currentChar, mappedCharacters, MAX_CASE_MAPPING_BUF, (isLastCased ? kCFUniCharToLowercase : kCFUniCharToTitlecase), flags, langCode);
if (mappedLength > 0) contents[currentIndex] = *mappedCharacters;
if (currentChar > 0xFFFF) { // Non-BMP char
switch (mappedLength) {
case 0:
__CFStringChangeSize(string, CFRangeMake(currentIndex, 2), 0, true);
contents = (UniChar *)__CFStrContents(string);
length -= 2;
break;
case 1:
__CFStringChangeSize(string, CFRangeMake(currentIndex + 1, 1), 0, true);
contents = (UniChar *)__CFStrContents(string);
--length;
break;
case 2:
contents[++currentIndex] = mappedCharacters[1];
break;
default:
--mappedLength; // Skip the current char
__CFStringChangeSize(string, CFRangeMake(currentIndex + 1, 0), mappedLength - 1, true);
contents = (UniChar *)__CFStrContents(string);
memmove(contents + currentIndex + 1, mappedCharacters + 1, mappedLength * sizeof(UniChar));
length += (mappedLength - 1);
currentIndex += mappedLength;
break;
}
} else if (mappedLength == 0) {
__CFStringChangeSize(string, CFRangeMake(currentIndex, 1), 0, true);
contents = (UniChar *)__CFStrContents(string);
--length;
} else if (mappedLength > 1) {
--mappedLength; // Skip the current char
__CFStringChangeSize(string, CFRangeMake(currentIndex + 1, 0), mappedLength, true);
contents = (UniChar *)__CFStrContents(string);
memmove(contents + currentIndex + 1, mappedCharacters + 1, mappedLength * sizeof(UniChar));
length += mappedLength;
currentIndex += mappedLength;
}
if (!((currentChar > 0xFFFF) ? CFUniCharIsMemberOf(currentChar, kCFUniCharCaseIgnorableCharacterSet) : CFUniCharIsMemberOfBitmap(currentChar, caseIgnorableForBMP))) { // We have non-caseignorable here
isLastCased = ((CFUniCharIsMemberOf(currentChar, kCFUniCharUppercaseLetterCharacterSet) || CFUniCharIsMemberOf(currentChar, kCFUniCharLowercaseLetterCharacterSet)) ? true : false);
}
}
}
}
#define MAX_DECOMP_BUF 64
#define HANGUL_SBASE 0xAC00
#define HANGUL_LBASE 0x1100
#define HANGUL_VBASE 0x1161
#define HANGUL_TBASE 0x11A7
#define HANGUL_SCOUNT 11172
#define HANGUL_LCOUNT 19
#define HANGUL_VCOUNT 21
#define HANGUL_TCOUNT 28
#define HANGUL_NCOUNT (HANGUL_VCOUNT * HANGUL_TCOUNT)
CF_INLINE uint32_t __CFGetUTF16Length(const UTF32Char *characters, uint32_t utf32Length) {
const UTF32Char *limit = characters + utf32Length;
uint32_t length = 0;
while (characters < limit) length += (*(characters++) > 0xFFFF ? 2 : 1);
return length;
}
CF_INLINE void __CFFillInUTF16(const UTF32Char *characters, UTF16Char *dst, uint32_t utf32Length) {
const UTF32Char *limit = characters + utf32Length;
UTF32Char currentChar;
while (characters < limit) {
currentChar = *(characters++);
if (currentChar > 0xFFFF) {
currentChar -= 0x10000;
*(dst++) = (UTF16Char)((currentChar >> 10) + 0xD800UL);
*(dst++) = (UTF16Char)((currentChar & 0x3FF) + 0xDC00UL);
} else {
*(dst++) = currentChar;
}
}
}
void CFStringNormalize(CFMutableStringRef string, CFStringNormalizationForm theForm) {
CFIndex currentIndex = 0;
CFIndex length;
bool needToReorder = true;
CF_OBJC_FUNCDISPATCHV(_kCFRuntimeIDCFString, void, (NSMutableString *)string, _cfNormalize:theForm);
CF_RETURN_IF_NOT_MUTABLE(string);
length = __CFStrLength(string);
if (__CFStrIsEightBit(string)) {
uint8_t *contents;
if (theForm == kCFStringNormalizationFormC) return; // 8bit form has no decomposition
contents = (uint8_t *)__CFStrContents(string) + __CFStrSkipAnyLengthByte(string);
for (;currentIndex < length;currentIndex++) {
if (contents[currentIndex] > 127) {
__CFStringChangeSize(string, CFRangeMake(0, 0), 0, true); // need to do harm way
needToReorder = false;
break;
}
}
}
if (currentIndex < length) {
UTF16Char *limit = (UTF16Char *)__CFStrContents(string) + length;
UTF16Char *contents = (UTF16Char *)__CFStrContents(string) + currentIndex;
UTF32Char buffer[MAX_DECOMP_BUF];
UTF32Char *mappedCharacters = buffer;
CFIndex allocatedLength = MAX_DECOMP_BUF;
CFIndex mappedLength;
CFIndex currentLength;
UTF32Char currentChar;
const uint8_t *decompBMP = CFUniCharGetBitmapPtrForPlane(kCFUniCharCanonicalDecomposableCharacterSet, 0);
const uint8_t *nonBaseBMP = CFUniCharGetBitmapPtrForPlane(kCFUniCharNonBaseCharacterSet, 0);
const uint8_t *combiningBMP = (const uint8_t *)CFUniCharGetUnicodePropertyDataForPlane(kCFUniCharCombiningProperty, 0);
while (contents < limit) {
if (CFUniCharIsSurrogateHighCharacter(*contents) && (contents + 1 < limit) && CFUniCharIsSurrogateLowCharacter(*(contents + 1))) {
currentChar = CFUniCharGetLongCharacterForSurrogatePair(*contents, *(contents + 1));
currentLength = 2;
contents += 2;
} else {
currentChar = *(contents++);
currentLength = 1;
}
mappedLength = 0;
if (CFUniCharIsMemberOfBitmap(currentChar, ((currentChar < 0x10000) ? decompBMP : CFUniCharGetBitmapPtrForPlane(kCFUniCharCanonicalDecomposableCharacterSet, (currentChar >> 16)))) && (0 == CFUniCharGetCombiningPropertyForCharacter(currentChar, ((currentChar < 0x10000) ? combiningBMP : (const uint8_t *)CFUniCharGetUnicodePropertyDataForPlane(kCFUniCharCombiningProperty, (currentChar >> 16)))))) {
if ((theForm & kCFStringNormalizationFormC) == 0 || currentChar < HANGUL_SBASE || currentChar > (HANGUL_SBASE + HANGUL_SCOUNT)) { // We don't have to decompose Hangul Syllables if we're precomposing again
mappedLength = CFUniCharDecomposeCharacter(currentChar, mappedCharacters, MAX_DECOMP_BUF);
}
}
if ((needToReorder || (theForm & kCFStringNormalizationFormC)) && ((contents < limit) || (mappedLength == 0))) {
if (mappedLength > 0) {
if (CFUniCharIsSurrogateHighCharacter(*contents) && (contents + 1 < limit) && CFUniCharIsSurrogateLowCharacter(*(contents + 1))) {
currentChar = CFUniCharGetLongCharacterForSurrogatePair(*contents, *(contents + 1));
} else {
currentChar = *contents;
}
}
if (0 != CFUniCharGetCombiningPropertyForCharacter(currentChar, (const uint8_t *)((currentChar < 0x10000) ? combiningBMP : CFUniCharGetUnicodePropertyDataForPlane(kCFUniCharCombiningProperty, (currentChar >> 16))))) {
uint32_t decompLength;
if (mappedLength == 0) {
contents -= (currentChar & 0xFFFF0000 ? 2 : 1);
if (currentIndex > 0) {
if (CFUniCharIsSurrogateLowCharacter(*(contents - 1)) && (currentIndex > 1) && CFUniCharIsSurrogateHighCharacter(*(contents - 2))) {
*mappedCharacters = CFUniCharGetLongCharacterForSurrogatePair(*(contents - 2), *(contents - 1));
currentIndex -= 2;
currentLength += 2;
} else {
*mappedCharacters = *(contents - 1);
--currentIndex;
++currentLength;
}
mappedLength = 1;
}
} else {
currentLength += (currentChar & 0xFFFF0000 ? 2 : 1);
}
contents += (currentChar & 0xFFFF0000 ? 2 : 1);
if (CFUniCharIsMemberOfBitmap(currentChar, ((currentChar < 0x10000) ? decompBMP : CFUniCharGetBitmapPtrForPlane(kCFUniCharCanonicalDecomposableCharacterSet, (currentChar >> 16))))) { // Vietnamese accent, etc.
decompLength = CFUniCharDecomposeCharacter(currentChar, mappedCharacters + mappedLength, MAX_DECOMP_BUF - mappedLength);
mappedLength += decompLength;
} else {
mappedCharacters[mappedLength++] = currentChar;
}
while (contents < limit) {
if (CFUniCharIsSurrogateHighCharacter(*contents) && (contents + 1 < limit) && CFUniCharIsSurrogateLowCharacter(*(contents + 1))) {
currentChar = CFUniCharGetLongCharacterForSurrogatePair(*contents, *(contents + 1));
} else {
currentChar = *contents;
}
if (0 == CFUniCharGetCombiningPropertyForCharacter(currentChar, (const uint8_t *)((currentChar < 0x10000) ? combiningBMP : CFUniCharGetUnicodePropertyDataForPlane(kCFUniCharCombiningProperty, (currentChar >> 16))))) break;
if (currentChar & 0xFFFF0000) {
contents += 2;
currentLength += 2;
} else {
++contents;
++currentLength;
}
if (mappedLength == allocatedLength) {
allocatedLength += MAX_DECOMP_BUF;
if (mappedCharacters == buffer) {
mappedCharacters = (UTF32Char *)CFAllocatorAllocate(kCFAllocatorSystemDefault, allocatedLength * sizeof(UTF32Char), 0);
memmove(mappedCharacters, buffer, MAX_DECOMP_BUF * sizeof(UTF32Char));
} else {
mappedCharacters = __CFSafelyReallocateWithAllocator(kCFAllocatorSystemDefault, mappedCharacters, allocatedLength * sizeof(UTF32Char), 0, NULL);
}
}
if (CFUniCharIsMemberOfBitmap(currentChar, ((currentChar < 0x10000) ? decompBMP : CFUniCharGetBitmapPtrForPlane(kCFUniCharCanonicalDecomposableCharacterSet, (currentChar >> 16))))) { // Vietnamese accent, etc.
decompLength = CFUniCharDecomposeCharacter(currentChar, mappedCharacters + mappedLength, MAX_DECOMP_BUF - mappedLength);
mappedLength += decompLength;
} else {
mappedCharacters[mappedLength++] = currentChar;
}
}
}
if (needToReorder && mappedLength > 1) CFUniCharPrioritySort(mappedCharacters, mappedLength);
}
if (theForm & kCFStringNormalizationFormKD) {
CFIndex newLength = 0;
if (mappedLength == 0 && CFUniCharIsMemberOf(currentChar, kCFUniCharCompatibilityDecomposableCharacterSet)) {
mappedCharacters[mappedLength++] = currentChar;
}
while (newLength < mappedLength) {
newLength = CFUniCharCompatibilityDecompose(mappedCharacters, mappedLength, allocatedLength);
if (newLength == 0) {
allocatedLength += MAX_DECOMP_BUF;
if (mappedCharacters == buffer) {
mappedCharacters = (UTF32Char *)CFAllocatorAllocate(kCFAllocatorSystemDefault, allocatedLength * sizeof(UTF32Char), 0);
memmove(mappedCharacters, buffer, MAX_DECOMP_BUF * sizeof(UTF32Char));
} else {
mappedCharacters = __CFSafelyReallocateWithAllocator(kCFAllocatorSystemDefault, mappedCharacters, allocatedLength * sizeof(UTF32Char), 0, NULL);
}
}
}
mappedLength = newLength;
}
if (theForm & kCFStringNormalizationFormC) {
UTF32Char nextChar;
if (mappedLength > 1) {
CFIndex consumedLength = 1;
UTF32Char *currentBase = mappedCharacters;
uint8_t currentClass, lastClass = 0;
bool didCombine = false;
currentChar = *mappedCharacters;
while (consumedLength < mappedLength) {
nextChar = mappedCharacters[consumedLength];
currentClass = CFUniCharGetCombiningPropertyForCharacter(nextChar, (const uint8_t *)((nextChar < 0x10000) ? combiningBMP : CFUniCharGetUnicodePropertyDataForPlane(kCFUniCharCombiningProperty, (nextChar >> 16))));
if (theForm & kCFStringNormalizationFormKD) {
if ((currentChar >= HANGUL_LBASE) && (currentChar < (HANGUL_LBASE + 0xFF))) {
SInt8 lIndex = currentChar - HANGUL_LBASE;
if ((0 <= lIndex) && (lIndex <= HANGUL_LCOUNT)) {
SInt16 vIndex = nextChar - HANGUL_VBASE;
if ((vIndex >= 0) && (vIndex <= HANGUL_VCOUNT)) {
SInt16 tIndex = 0;
CFIndex usedLength = mappedLength;
mappedCharacters[consumedLength++] = 0xFFFD;
if (consumedLength < mappedLength) {
tIndex = mappedCharacters[consumedLength] - HANGUL_TBASE;
if ((tIndex < 0) || (tIndex > HANGUL_TCOUNT)) {
tIndex = 0;
} else {
mappedCharacters[consumedLength++] = 0xFFFD;
}
}
*currentBase = (lIndex * HANGUL_VCOUNT + vIndex) * HANGUL_TCOUNT + tIndex + HANGUL_SBASE;
while (--usedLength > 0) {
if (mappedCharacters[usedLength] == 0xFFFD) {
--mappedLength;
--consumedLength;
memmove(mappedCharacters + usedLength, mappedCharacters + usedLength + 1, (mappedLength - usedLength) * sizeof(UTF32Char));
}
}
currentBase = mappedCharacters + consumedLength;
currentChar = *currentBase;
++consumedLength;
continue;
}
}
}
if (!CFUniCharIsMemberOfBitmap(nextChar, ((nextChar < 0x10000) ? nonBaseBMP : CFUniCharGetBitmapPtrForPlane(kCFUniCharNonBaseCharacterSet, (nextChar >> 16))))) {
*currentBase = currentChar;
currentBase = mappedCharacters + consumedLength;
currentChar = nextChar;
++consumedLength;
continue;
}
}
if ((lastClass == 0) || (currentClass > lastClass)) {
nextChar = CFUniCharPrecomposeCharacter(currentChar, nextChar);
if (nextChar == 0xFFFD) {
lastClass = currentClass;
} else {
mappedCharacters[consumedLength] = 0xFFFD;
didCombine = true;
currentChar = nextChar;
}
}
++consumedLength;
}
*currentBase = currentChar;
if (didCombine) {
consumedLength = mappedLength;
while (--consumedLength > 0) {
if (mappedCharacters[consumedLength] == 0xFFFD) {
--mappedLength;
memmove(mappedCharacters + consumedLength, mappedCharacters + consumedLength + 1, (mappedLength - consumedLength) * sizeof(UTF32Char));
}
}
}
} else if ((currentChar >= HANGUL_LBASE) && (currentChar < (HANGUL_LBASE + 0xFF))) { // Hangul Jamo
SInt8 lIndex = currentChar - HANGUL_LBASE;
if ((contents < limit) && (0 <= lIndex) && (lIndex <= HANGUL_LCOUNT)) {
SInt16 vIndex = *contents - HANGUL_VBASE;
if ((vIndex >= 0) && (vIndex <= HANGUL_VCOUNT)) {
SInt16 tIndex = 0;
++contents; ++currentLength;
if (contents < limit) {
tIndex = *contents - HANGUL_TBASE;
if ((tIndex < 0) || (tIndex > HANGUL_TCOUNT)) {
tIndex = 0;
} else {
++contents; ++currentLength;
}
}
*mappedCharacters = (lIndex * HANGUL_VCOUNT + vIndex) * HANGUL_TCOUNT + tIndex + HANGUL_SBASE;
mappedLength = 1;
}
}
} else { // collect class 0 non-base characters
while (contents < limit) {
nextChar = *contents;
if (CFUniCharIsSurrogateHighCharacter(nextChar) && ((contents + 1) < limit) && CFUniCharIsSurrogateLowCharacter(*(contents + 1))) {
nextChar = CFUniCharGetLongCharacterForSurrogatePair(nextChar, *(contents + 1));
if (!CFUniCharIsMemberOfBitmap(nextChar, (const uint8_t *)CFUniCharGetBitmapPtrForPlane(kCFUniCharNonBaseCharacterSet, (nextChar >> 16))) || (0 != CFUniCharGetCombiningPropertyForCharacter(nextChar, (const uint8_t *)CFUniCharGetUnicodePropertyDataForPlane(kCFUniCharCombiningProperty, (nextChar >> 16))))) break;
} else {
if (!CFUniCharIsMemberOfBitmap(nextChar, nonBaseBMP) || (0 != CFUniCharGetCombiningPropertyForCharacter(nextChar, combiningBMP))) break;
}
currentChar = CFUniCharPrecomposeCharacter(currentChar, nextChar);
if (0xFFFD == currentChar) break;
if (nextChar < 0x10000) {
++contents; ++currentLength;
} else {
contents += 2;
currentLength += 2;
}
*mappedCharacters = currentChar;
mappedLength = 1;
}
}
}
if (mappedLength > 0) {
CFIndex utf16Length = __CFGetUTF16Length(mappedCharacters, mappedLength);
if (utf16Length != currentLength) {
__CFStringChangeSize(string, CFRangeMake(currentIndex, currentLength), utf16Length, true);
currentLength = utf16Length;
}
contents = (UTF16Char *)__CFStrContents(string);
limit = contents + __CFStrLength(string);
contents += currentIndex;
__CFFillInUTF16(mappedCharacters, contents, mappedLength);
contents += utf16Length;
}
currentIndex += currentLength;
}
if (mappedCharacters != buffer) CFAllocatorDeallocate(kCFAllocatorSystemDefault, mappedCharacters);
}
}
void CFStringFold(CFMutableStringRef theString, CFStringCompareFlags theFlags, CFLocaleRef locale) {
CF_RETURN_IF_NOT_MUTABLE(theString);
CFStringInlineBuffer stringBuffer;
CFIndex length = CFStringGetLength(theString);
CFIndex currentIndex = 0;
CFIndex bufferLength = 0;
UTF32Char buffer[kCFStringStackBufferLength];
const uint8_t *cString;
const uint8_t *langCode;
CFStringEncoding eightBitEncoding;
bool caseInsensitive = ((theFlags & kCFCompareCaseInsensitive) ? true : false);
bool isObjcOrSwift = CF_IS_OBJC(_kCFRuntimeIDCFString, theString) || CF_IS_SWIFT(_kCFRuntimeIDCFString, theString);
CFLocaleRef theLocale = locale;
if ((theFlags & kCFCompareLocalized) && (NULL == locale)) {
theLocale = CFLocaleCopyCurrent();
}
theFlags &= (kCFCompareCaseInsensitive|kCFCompareDiacriticInsensitive|kCFCompareWidthInsensitive);
if ((0 == theFlags) || (0 == length)) goto bail; // nothing to do
langCode = ((NULL == theLocale) ? NULL : (const uint8_t *)_CFStrGetSpecialCaseHandlingLanguageIdentifierForLocale(theLocale, true));
eightBitEncoding = __CFStringGetEightBitStringEncoding();
cString = (const uint8_t *)_CFStringGetCStringPtrInternal(theString, eightBitEncoding, false, isObjcOrSwift);
if ((NULL != cString) && !caseInsensitive && (kCFStringEncodingASCII == eightBitEncoding)) goto bail; // All ASCII
_CFStringInitInlineBufferInternal(theString, &stringBuffer, CFRangeMake(0, length), isObjcOrSwift);
if ((NULL != cString) && (theFlags & (kCFCompareCaseInsensitive|kCFCompareDiacriticInsensitive))) {
const uint8_t *cStringPtr = cString;
const uint8_t *cStringLimit = cString + length;
uint8_t *cStringContents = (isObjcOrSwift ? NULL : (uint8_t *)__CFStrContents(theString) + __CFStrSkipAnyLengthByte(theString));
while (cStringPtr < cStringLimit) {
if ((*cStringPtr < 0x80) && (NULL == langCode)) {
if (caseInsensitive && (*cStringPtr >= 'A') && (*cStringPtr <= 'Z')) {
if (NULL == cStringContents) {
break;
} else {
cStringContents[cStringPtr - cString] += ('a' - 'A');
}
}
} else {
if ((bufferLength = __CFStringFoldCharacterClusterAtIndex((UTF32Char)__CFCharToUniCharTable[*cStringPtr], &stringBuffer, cStringPtr - cString, theFlags, langCode, buffer, kCFStringStackBufferLength, NULL, NULL)) > 0) {
if ((*buffer > 0x7F) || (bufferLength > 1) || (NULL == cStringContents)) break;
cStringContents[cStringPtr - cString] = *buffer;
}
}
++cStringPtr;
}
currentIndex = cStringPtr - cString;
}
if (currentIndex < length) {
UTF16Char *contents;
CFMutableStringRef cfString = NULL;
CFRange range;
if (isObjcOrSwift) {
range = CFRangeMake(currentIndex, length - currentIndex);
contents = (UTF16Char *)CFAllocatorAllocate(kCFAllocatorSystemDefault, sizeof(UTF16Char) * range.length, 0);
CFStringGetCharacters(theString, range, contents);
cfString = CFStringCreateMutableWithExternalCharactersNoCopy(kCFAllocatorSystemDefault, contents, range.length, range.length, NULL);
}
if (cfString) {
CFStringFold(cfString, theFlags, theLocale);
CFStringReplace(theString, range, cfString);
CFRelease(cfString);
} else {
const UTF32Char *characters;
const UTF32Char *charactersLimit;
UTF32Char character;
CFIndex consumedLength;
contents = NULL;
if (bufferLength > 0) {
__CFStringChangeSize(theString, CFRangeMake(currentIndex + 1, 0), bufferLength - 1, true);
length = __CFStrLength(theString);
_CFStringInitInlineBufferInternal(theString, &stringBuffer, CFRangeMake(0, length), isObjcOrSwift);
contents = (UTF16Char *)__CFStrContents(theString) + currentIndex;
characters = buffer;
charactersLimit = characters + bufferLength;
while (characters < charactersLimit) *(contents++) = (UTF16Char)*(characters++);
++currentIndex;
}
while (currentIndex < length) {
character = __CFStringGetCharacterFromInlineBufferQuick(&stringBuffer, currentIndex);
consumedLength = 0;
if ((NULL == langCode) && (character < 0x80) && (0 == (theFlags & kCFCompareDiacriticInsensitive))) {
if (caseInsensitive && (character >= 'A') && (character <= 'Z')) {
consumedLength = 1;
bufferLength = 1;
*buffer = character + ('a' - 'A');
}
} else {
if (CFUniCharIsSurrogateHighCharacter(character) && ((currentIndex + 1) < length)) {
UTF16Char lowSurrogate = __CFStringGetCharacterFromInlineBufferQuick(&stringBuffer, currentIndex + 1);
if (CFUniCharIsSurrogateLowCharacter(lowSurrogate)) character = CFUniCharGetLongCharacterForSurrogatePair(character, lowSurrogate);
}
bufferLength = __CFStringFoldCharacterClusterAtIndex(character, &stringBuffer, currentIndex, theFlags, langCode, buffer, kCFStringStackBufferLength, &consumedLength, NULL);
}
if (consumedLength > 0) {
CFIndex utf16Length = bufferLength;
characters = buffer;
charactersLimit = characters + bufferLength;
while (characters < charactersLimit) if (*(characters++) > 0xFFFF) ++utf16Length; // Extend bufferLength to the UTF-16 length
if ((utf16Length != consumedLength) || __CFStrIsEightBit(theString)) {
CFRange range;
CFIndex insertLength;
if (consumedLength < utf16Length) { // Need to expand
range = CFRangeMake(currentIndex + consumedLength, 0);
insertLength = utf16Length - consumedLength;
} else {
range = CFRangeMake(currentIndex + utf16Length, consumedLength - utf16Length);
insertLength = 0;
}
__CFStringChangeSize(theString, range, insertLength, true);
length = __CFStrLength(theString);
_CFStringInitInlineBufferInternal(theString, &stringBuffer, CFRangeMake(0, length), isObjcOrSwift);
}
(void)CFUniCharFromUTF32(buffer, bufferLength, (UTF16Char *)__CFStrContents(theString) + currentIndex, true, __CF_BIG_ENDIAN__);
currentIndex += utf16Length;
} else {
++currentIndex;
}
}
}
}
bail:
if (NULL == locale && theLocale) {
CFRelease(theLocale);
}
}
static bool _CFStringHasStrongRTL(CFStringRef str, CFRange range) {
CFIndex charIndex = 0;
CFStringInlineBuffer stringBuffer;
const uint8_t *strongRightBMP = CFUniCharGetBitmapPtrForPlane(kCFUniCharStrongRightToLeftCharacterSet, 0); // Most RTL strong chars are in BMP
CFStringInitInlineBuffer(str, &stringBuffer, range);
while (charIndex < range.length) { // Both line break & strong right characters are all in BMP so no need to process surrogates
UTF32Char character = CFStringGetCharacterFromInlineBuffer(&stringBuffer, charIndex);
const uint8_t *strongRight = strongRightBMP;
if (CFUniCharIsSurrogateHighCharacter(character)) {
UTF16Char otherChar = CFStringGetCharacterFromInlineBuffer(&stringBuffer, ++charIndex);
uint32_t plane;
if (!CFUniCharIsSurrogateLowCharacter(otherChar)) continue;
character = CFUniCharGetLongCharacterForSurrogatePair(character, otherChar);
plane = ((character >> 16) & 0x1F);
if (0 != plane) strongRight = CFUniCharGetBitmapPtrForPlane(kCFUniCharStrongRightToLeftCharacterSet, plane);
}
if (CFUniCharIsMemberOfBitmap(character, strongRight)) return true;
++charIndex;
}
return false;
}
/* String formatting */
enum {
kCFStringFormatZeroFlag = (1 << 0), // if not, padding is space char
kCFStringFormatMinusFlag = (1 << 1), // if not, no flag implied
kCFStringFormatPlusFlag = (1 << 2), // if not, no flag implied, overrides space
kCFStringFormatSpaceFlag = (1 << 3), // if not, no flag implied
kCFStringFormatExternalSpecFlag = (1 << 4), // using config dict
kCFStringFormatLocalizable = (1 << 5), // explicitly mark the specs we can localize
kCFStringFormatEntityMarkerFlag = (1 << 6), // using entity marker
kCFStringFormatPercentReplacementFlag = (1 << 7), // marks a '%%' replacement, used for metadata collection
};
typedef struct {
int16_t size;
int16_t type;
SInt32 loc;
SInt32 len;
SInt32 widthArg;
SInt32 precArg;
uint32_t flags;
int8_t mainArgNum;
int8_t precArgNum;
int8_t widthArgNum;
int8_t configDictIndex;
int8_t numericFormatStyle; // Only set for localizable numeric quantities
} CFFormatSpec;
typedef struct {
int16_t type;
int16_t size;
union {
int64_t int64Value;
double doubleValue;
#if LONG_DOUBLE_SUPPORT
long double longDoubleValue;
#endif
void *pointerValue;
} value;
} CFPrintValue;
enum {
CFFormatDefaultSize = 0,
CFFormatSize1 = 1,
CFFormatSize2 = 2,
CFFormatSize4 = 3,
CFFormatSize8 = 4,
CFFormatSize16 = 5,
#if TARGET_RT_64_BIT
CFFormatSizeLong = CFFormatSize8,
CFFormatSizePointer = CFFormatSize8
#else
CFFormatSizeLong = CFFormatSize4,
CFFormatSizePointer = CFFormatSize4
#endif
};
enum {
CFFormatStyleDecimal = (1 << 0),
CFFormatStyleScientific = (1 << 1),
CFFormatStyleDecimalOrScientific = CFFormatStyleDecimal|CFFormatStyleScientific,
CFFormatStyleUnsigned = (1 << 2)
};
enum {
CFFormatLiteralType = 32,
CFFormatLongType = 33,
CFFormatDoubleType = 34,
CFFormatPointerType = 35,
CFFormatCFType = 37, /* handled specially; this is the general object type */
CFFormatUnicharsType = 38, /* handled specially */
CFFormatCharsType = 39, /* handled specially */
CFFormatPascalCharsType = 40, /* handled specially */
CFFormatSingleUnicharType = 41, /* handled specially */
CFFormatDummyPointerType = 42, /* special case for %n */
CFFormatIncompleteSpecifierType = 43 /* special case for a trailing incomplete specifier */
};
#if TARGET_OS_MAC || TARGET_OS_WIN32 || TARGET_OS_LINUX || TARGET_OS_WASI
/* Only come in here if spec->type is CFFormatLongType or CFFormatDoubleType. Pass in 0 for width or precision if not specified. Returns false if couldn't do the format (with the assumption the caller falls back to unlocalized).
*/
static Boolean __CFStringFormatLocalizedNumber(CFMutableStringRef output, CFLocaleRef locale, const CFPrintValue *values, const CFFormatSpec *spec, SInt32 width, SInt32 precision, Boolean hasPrecision) {
static CFLock_t formatterLock = CFLockInit;
// These formatters are recached if the locale argument is different
static CFNumberFormatterRef decimalFormatter = NULL;
static CFNumberFormatterRef scientificFormatter = NULL;
static CFNumberFormatterRef gFormatter = NULL; // for %g
static SInt32 groupingSize = 0;
static SInt32 secondaryGroupingSize = 0;
// !!! This code should be removed before shipping
static int disableLocalizedFormatting = -1;
if (disableLocalizedFormatting == -1) disableLocalizedFormatting = (getenv("CFStringDisableLocalizedNumberFormatting") != NULL) ? 1 : 0;
if (disableLocalizedFormatting) return false;
CFNumberFormatterRef formatter;
__CFLock(&formatterLock); // We use the formatter from one thread at one time; if this proves to be a bottleneck we need to get fancier
switch (spec->numericFormatStyle) {
case CFFormatStyleUnsigned:
case CFFormatStyleDecimal:
if (!decimalFormatter || !CFEqual(CFNumberFormatterGetLocale(decimalFormatter), locale)) { // cache or recache if the locale is different
if (decimalFormatter) CFRelease(decimalFormatter);
decimalFormatter = CFNumberFormatterCreate(NULL, locale, kCFNumberFormatterDecimalStyle); // since this is shared, remember to reset all its properties!
}
formatter = decimalFormatter;
break;
case CFFormatStyleScientific:
if (!scientificFormatter || !CFEqual(CFNumberFormatterGetLocale(scientificFormatter), locale)) { // cache or recache if the locale is different
if (scientificFormatter) CFRelease(scientificFormatter);
scientificFormatter = CFNumberFormatterCreate(NULL, locale, kCFNumberFormatterScientificStyle);
CFStringRef pattern = CFSTR("#E+00"); // the default pattern does not have the sign if the exponent is positive and it is single digit
CFNumberFormatterSetFormat(scientificFormatter, pattern);
CFNumberFormatterSetProperty(scientificFormatter, kCFNumberFormatterUseSignificantDigitsKey, kCFBooleanTrue);
}
formatter = scientificFormatter;
break;
case CFFormatStyleDecimalOrScientific:
if (!gFormatter || !CFEqual(CFNumberFormatterGetLocale(gFormatter), locale)) { // cache or recache if the locale is different
if (gFormatter) CFRelease(gFormatter);
gFormatter = CFNumberFormatterCreate(NULL, locale, kCFNumberFormatterDecimalStyle);
// when we update the locale in gFormatter, we also need to update the two grouping sizes
CFNumberRef num = (CFNumberRef) CFNumberFormatterCopyProperty(gFormatter, kCFNumberFormatterGroupingSizeKey);
CFNumberGetValue(num, kCFNumberSInt32Type, &groupingSize);
CFRelease(num);
num = (CFNumberRef) CFNumberFormatterCopyProperty(gFormatter, kCFNumberFormatterSecondaryGroupingSizeKey);
CFNumberGetValue(num, kCFNumberSInt32Type, &secondaryGroupingSize);
CFRelease(num);
}
formatter = gFormatter;
break;
default:
// HALT, or else CFNumberGetFormat below will be called on uninitialized memory
CRSetCrashLogMessage("Unexpected formatter style");
HALT;
}
CFStringRef origFormat = CFStringCreateCopy(NULL, CFNumberFormatterGetFormat(formatter)); // Need to hang on to this in case the format changes while we are setting properties below
SInt32 prec = hasPrecision ? precision : ((spec->type == CFFormatLongType) ? 0 : 6); // default precision of printf is 6
// pattern must be set before setting width and padding
// otherwise, the pattern will take over those settings
if (spec->numericFormatStyle == CFFormatStyleDecimalOrScientific) {
if (prec == 0) prec = 1; // at least one sig fig
CFMutableStringRef pattern = CFStringCreateMutable(NULL, 0);
// use significant digits pattern
CFStringAppendCString(pattern, "@", kCFStringEncodingASCII);
CFStringPad(pattern, CFSTR("#"), prec, 0);
double targetValue = values[spec->mainArgNum].value.doubleValue;
#if LONG_DOUBLE_SUPPORT
if (CFFormatSize16 == values[spec->mainArgNum].size) {
targetValue = values[spec->mainArgNum].value.longDoubleValue; // losing precision
}
#endif
double max = pow(10.0, (double)prec); // if the value requires more digits than the number of sig figs, we need to use scientific format
double min = 0.0001; // if the value is less than 10E-4, scientific format is the shorter form
if (((targetValue > 0 && (targetValue > max || targetValue < min)) || (targetValue < 0 && (targetValue < -max || targetValue > -min)))){
CFStringAppendCString(pattern, "E+00", kCFStringEncodingASCII);
} else if (prec > groupingSize && groupingSize != 0) {
CFStringInsert(pattern, prec-groupingSize, CFSTR(",")); // if we are not using scientific format, we need to set the pattern to use grouping separator
if (secondaryGroupingSize != 0 && prec > (groupingSize + secondaryGroupingSize)) CFStringInsert(pattern, prec-groupingSize-secondaryGroupingSize, CFSTR(","));
}
CFNumberFormatterSetFormat(formatter, pattern);
CFRelease(pattern);
}
// clear the padding, we will add it later if we need it
const SInt32 z = 0;
CFNumberRef zero = CFNumberCreate(NULL, kCFNumberSInt32Type, &z);
CFNumberFormatterSetProperty(formatter, kCFNumberFormatterFormatWidthKey, zero);
CFNumberRef tmp = CFNumberCreate(NULL, kCFNumberSInt32Type, &prec);
CFNumberFormatterSetProperty(formatter, kCFNumberFormatterMaxFractionDigitsKey, tmp);
if (spec->type == CFFormatDoubleType) {
CFNumberFormatterSetProperty(formatter, kCFNumberFormatterMinFractionDigitsKey, tmp);
} else {
CFNumberFormatterSetProperty(formatter, kCFNumberFormatterMinFractionDigitsKey, zero);
}
CFRelease(tmp);
CFRelease(zero);
Boolean isNegative = false;
switch (values[spec->mainArgNum].type) {
case CFFormatLongType:
if (values[spec->mainArgNum].value.int64Value < 0) isNegative = true;
break;
case CFFormatDoubleType:
#if LONG_DOUBLE_SUPPORT
if ((CFFormatSize16 == values[spec->mainArgNum].size) && (values[spec->mainArgNum].value.longDoubleValue < 0)) isNegative = true;
else
#endif
if (values[spec->mainArgNum].value.doubleValue < 0) isNegative = true;
break;
}
CFStringRef currentPattern = CFNumberFormatterGetFormat(formatter);
if ((spec->flags & kCFStringFormatPlusFlag) && !isNegative) {
if (CFStringGetCharacterAtIndex(currentPattern, 0) != '+') {
CFMutableStringRef newPattern = CFStringCreateMutableCopy(NULL, 0, CFSTR("+"));
CFStringAppend(newPattern, currentPattern);
CFNumberFormatterSetFormat(formatter, newPattern);
CFRelease(newPattern);
}
} else {
if (CFStringGetCharacterAtIndex(currentPattern, 0) == '+') {
CFStringRef newPattern = CFStringCreateWithSubstring(NULL, currentPattern, CFRangeMake(1, CFStringGetLength(currentPattern)-1));
CFNumberFormatterSetFormat(formatter, newPattern);
CFRelease(newPattern);
}
}
Boolean padZero = spec->flags & kCFStringFormatZeroFlag;
// width == 0 seems to be CFNumberFormatter's default setting
if (hasPrecision && spec->type == CFFormatLongType) { // if we have precision and %d or %u, we pad 0 according to precision first
tmp = CFNumberCreate(NULL, kCFNumberSInt32Type, &prec);
} else {
tmp = CFNumberCreate(NULL, kCFNumberSInt32Type, &width);
}
CFNumberFormatterSetProperty(formatter, kCFNumberFormatterFormatWidthKey, tmp);
if (hasPrecision && spec->type == CFFormatLongType) { // if we have precision and %d or %u, we pad 0
padZero = true;
}
// Left (default) or right padding
SInt32 p = (spec->flags & kCFStringFormatMinusFlag) ? kCFNumberFormatterPadAfterSuffix : (padZero ? kCFNumberFormatterPadAfterPrefix : kCFNumberFormatterPadBeforePrefix);
SInt32 minDigits = 0;
CFNumberGetValue(tmp, kCFNumberSInt32Type, &minDigits);
// Ensure we're padding up to the minimum number of digits
if (padZero && (minDigits > 0) && p) {
// For floating-point values, the previously set `kCFNumberFormatterFormatWidthKey` value is sufficient to calculate the minimum integer digits needed, accounting for plus/minus sign, grouping separators, radix, etc.
// We need this to force padding for integer values (e.g., 42 -> "%04d" => "0,042", or 234 -> "%10.5d" => " 00,234").
if (spec->type != CFFormatDoubleType) {
CFNumberFormatterSetProperty(formatter, kCFNumberFormatterMinIntegerDigitsKey, tmp);
}
}
CFRelease(tmp);
if (hasPrecision && spec->type == CFFormatLongType) {
SInt32 tmpP = kCFNumberFormatterPadAfterPrefix;
tmp = CFNumberCreate(NULL, kCFNumberSInt32Type, &tmpP);
} else {
tmp = CFNumberCreate(NULL, kCFNumberSInt32Type, &p);
}
CFNumberFormatterSetProperty(formatter, kCFNumberFormatterPaddingPositionKey, tmp);
CFRelease(tmp);
if (!padZero) { // If we're padding with 0, the zero needs to be localized. Setting MinIntegerDigitsKey localizes, setting PaddingCharacterKey does not
CFNumberFormatterSetProperty(formatter, kCFNumberFormatterPaddingCharacterKey, CFSTR(" "));
}
// Work around until ICU allows for localizing of custom format patterns
// <rdar://problem/22745439> Custom scientific notation formatting does not work with padding
if (padZero && spec->numericFormatStyle == CFFormatStyleDecimalOrScientific) {
CFNumberFormatterSetProperty(formatter, kCFNumberFormatterPaddingCharacterKey, CFSTR("0"));
}
if (spec->numericFormatStyle == CFFormatStyleScientific) {
prec++; // for %e, precision+1 is the number of sig fig
tmp = CFNumberCreate(NULL, kCFNumberSInt32Type, &prec);
CFNumberFormatterSetProperty(formatter, kCFNumberFormatterMinSignificantDigitsKey, tmp);
CFNumberFormatterSetProperty(formatter, kCFNumberFormatterMaxSignificantDigitsKey, tmp);
CFRelease(tmp);
}
CFStringRef localizedNumberString = NULL;
switch (spec->type) {
case CFFormatLongType:
// ??? Need to do unsigned
localizedNumberString = CFNumberFormatterCreateStringWithValue(NULL, formatter, kCFNumberSInt64Type, &(values[spec->mainArgNum].value.int64Value));
break;
case CFFormatDoubleType: {
#if LONG_DOUBLE_SUPPORT
if (CFFormatSize16 == values[spec->mainArgNum].size) {
double doubleValue = values[spec->mainArgNum].value.longDoubleValue; // losing precision
localizedNumberString = CFNumberFormatterCreateStringWithValue(NULL, formatter, kCFNumberDoubleType, &doubleValue);
} else
#endif
{
localizedNumberString = CFNumberFormatterCreateStringWithValue(NULL, formatter, kCFNumberDoubleType, &(values[spec->mainArgNum].value.doubleValue));
}
break;
}
}
CFNumberFormatterSetFormat(formatter, origFormat); // Need to reset the format in case it changed
CFRelease(origFormat);
__CFUnlock(&formatterLock);
if (localizedNumberString) {
// we need to pad space if we have %d or %u
if (spec->type == CFFormatLongType && hasPrecision && CFStringGetLength(localizedNumberString) < width) {
CFMutableStringRef finalStr = NULL;
if (p == kCFNumberFormatterPadAfterSuffix) {
finalStr = CFStringCreateMutableCopy(NULL, 0, localizedNumberString);
CFStringPad(finalStr, CFSTR(" "), width, 0);
} else {
finalStr = CFStringCreateMutable(NULL, 0);
CFStringPad(finalStr, CFSTR(" "), width - CFStringGetLength(localizedNumberString), 0);
CFStringAppend(finalStr, localizedNumberString);
}
CFRelease(localizedNumberString);
localizedNumberString = finalStr;
}
CFStringAppend(output, localizedNumberString);
CFRelease(localizedNumberString);
return true;
}
return false;
}
#endif
#if !DEPLOYMENT_RUNTIME_OBJC
// Open-source Core Foundation cannot rely on Foundation headers being present, so we redefine this here.
// This must match the code that is vended by FoundationErrors.h and swift-corelibs-foundation.
static const CFIndex NSFormattingError = 2048;
#endif // !DEPLOYMENT_RUNTIME_OBJC
CF_INLINE CFErrorRef __CFCreateOverflowError(void) {
CFAllocatorRef tmpAlloc = __CFGetDefaultAllocator();
CFMutableDictionaryRef userInfo = CFDictionaryCreateMutable(tmpAlloc, 0, &kCFCopyStringDictionaryKeyCallBacks, &kCFTypeDictionaryValueCallBacks);
CFDictionarySetValue(userInfo, kCFErrorDebugDescriptionKey, CFSTR("Overflow occurred"));
CFErrorRef result = CFErrorCreate(tmpAlloc, kCFErrorDomainCocoa, NSFormattingError, userInfo);
CFRelease(userInfo);
return result;
}
CF_INLINE Boolean __CFParseFormatSpec(const UniChar *uformat, const uint8_t *cformat, SInt32 *fmtIdx, SInt32 fmtLen, CFFormatSpec *spec, CFStringRef *configKeyPointer, CFErrorRef *errorPtr) {
Boolean seenDot = false;
Boolean seenSharp = false;
Boolean seenOpenBracket = false;
Boolean validBracketSequence = false;
CFIndex keyLength = 0;
CFIndex keyIndex = kCFNotFound;
for (;;) {
UniChar ch;
if (fmtLen <= *fmtIdx) { /* no type */
spec->type = CFFormatIncompleteSpecifierType;
return true;
}
if (cformat) ch = (UniChar)cformat[(*fmtIdx)++]; else ch = uformat[(*fmtIdx)++];
if (keyIndex >= 0) {
if ((ch < '0') || ((ch > '9') && (ch < 'A')) || ((ch > 'Z') && (ch < 'a') && (ch != '_')) || (ch > 'z')) {
if (ch == ']') {
if (seenOpenBracket) {
validBracketSequence = true;
keyLength = (*fmtIdx) - 1 - keyIndex;
}
} else if (ch == '@') {
if (validBracketSequence) {
spec->flags |= kCFStringFormatEntityMarkerFlag;
} else {
keyLength = (*fmtIdx) - 1 - keyIndex;
}
spec->flags |= kCFStringFormatExternalSpecFlag;
spec->type = CFFormatCFType;
spec->size = CFFormatSizePointer; // 4 or 8 depending on LP64
if ((NULL != configKeyPointer) && (keyLength > 0)) {
if (cformat) {
*configKeyPointer = CFStringCreateWithBytes(NULL, cformat + keyIndex, keyLength, __CFStringGetEightBitStringEncoding(), FALSE);
} else {
*configKeyPointer = CFStringCreateWithCharactersNoCopy(NULL, uformat + keyIndex, keyLength, kCFAllocatorNull);
}
}
return true;
} else {
keyIndex = kCFNotFound;
}
}
continue;
}
reswtch:switch (ch) {
case '#': // ignored for now
seenSharp = true;
break;
case '[':
if (!seenOpenBracket) { // We can only have one
seenOpenBracket = true;
keyIndex = *fmtIdx;
}
break;
case 0x20:
if (!(spec->flags & kCFStringFormatPlusFlag)) spec->flags |= kCFStringFormatSpaceFlag;
break;
case '-':
spec->flags |= kCFStringFormatMinusFlag;
spec->flags &= ~kCFStringFormatZeroFlag; // remove zero flag
break;
case '+':
spec->flags |= kCFStringFormatPlusFlag;
spec->flags &= ~kCFStringFormatSpaceFlag; // remove space flag
break;
case '0':
if (seenDot) { // after we see '.' and then we see '0', it is 0 precision. We should not see '.' after '0' if '0' is the zero padding flag
spec->precArg = 0;
break;
}
if (!(spec->flags & kCFStringFormatMinusFlag)) spec->flags |= kCFStringFormatZeroFlag;
break;
case 'h':
if (*fmtIdx < fmtLen) {
// fetch next character, don't increment fmtIdx
if (cformat) ch = (UniChar)cformat[(*fmtIdx)]; else ch = uformat[(*fmtIdx)];
if ('h' == ch) { // 'hh' for char, like 'c'
(*fmtIdx)++;
spec->size = CFFormatSize1;
break;
}
}
spec->size = CFFormatSize2;
break;
case 'l':
if (*fmtIdx < fmtLen) {
// fetch next character, don't increment fmtIdx
if (cformat) ch = (UniChar)cformat[(*fmtIdx)]; else ch = uformat[(*fmtIdx)];
if ('l' == ch) { // 'll' for long long, like 'q'
(*fmtIdx)++;
spec->size = CFFormatSize8;
break;
}
}
spec->size = CFFormatSizeLong; // 4 or 8 depending on LP64
break;
#if LONG_DOUBLE_SUPPORT
case 'L':
spec->size = CFFormatSize16;
break;
#endif
case 'q':
spec->size = CFFormatSize8;
break;
case 't': case 'z':
spec->size = CFFormatSizeLong; // 4 or 8 depending on LP64
break;
case 'j':
spec->size = CFFormatSize8;
break;
case 'c':
spec->type = CFFormatLongType;
spec->size = CFFormatSize1;
return true;
case 'D': case 'd': case 'i': case 'U': case 'u':
// we can localize all but octal or hex
if (_CFExecutableLinkedOnOrAfter(CFSystemVersionMountainLion)) spec->flags |= kCFStringFormatLocalizable;
spec->numericFormatStyle = CFFormatStyleDecimal;
if (ch == 'u' || ch == 'U') spec->numericFormatStyle = CFFormatStyleUnsigned;
// fall thru
case 'O': case 'o': case 'x': case 'X':
spec->type = CFFormatLongType;
// Seems like if spec->size == 0, we should spec->size = CFFormatSize4. However, 0 is handled correctly.
return true;
case 'f': case 'F': case 'g': case 'G': case 'e': case 'E': {
// we can localize all but hex float output
if (_CFExecutableLinkedOnOrAfter(CFSystemVersionMountainLion)) spec->flags |= kCFStringFormatLocalizable;
char lch = (ch >= 'A' && ch <= 'Z') ? (ch - 'A' + 'a') : ch;
spec->numericFormatStyle = ((lch == 'e' || lch == 'g') ? CFFormatStyleScientific : 0) | ((lch == 'f' || lch == 'g') ? CFFormatStyleDecimal : 0);
if (seenDot && spec->precArg == -1 && spec->precArgNum == -1) { // for the cases that we have '.' but no precision followed, not even '*'
spec->precArg = 0;
}
}
// fall thru
case 'a': case 'A':
spec->type = CFFormatDoubleType;
if (spec->size != CFFormatSize16) spec->size = CFFormatSize8;
return true;
case 'n': /* %n is not handled correctly; for Leopard or newer apps, we disable it further */
spec->type = CFFormatDummyPointerType;
spec->size = CFFormatSizePointer; // 4 or 8 depending on LP64
return true;
case 'p':
spec->type = CFFormatPointerType;
spec->size = CFFormatSizePointer; // 4 or 8 depending on LP64
return true;
case 's':
spec->type = CFFormatCharsType;
spec->size = CFFormatSizePointer; // 4 or 8 depending on LP64
return true;
case 'S':
spec->type = CFFormatUnicharsType;
spec->size = CFFormatSizePointer; // 4 or 8 depending on LP64
return true;
case 'C':
spec->type = CFFormatSingleUnicharType;
spec->size = CFFormatSize2;
return true;
case 'P':
spec->type = CFFormatPascalCharsType;
spec->size = CFFormatSizePointer; // 4 or 8 depending on LP64
return true;
case '@':
if (seenSharp) {
seenSharp = false;
keyIndex = *fmtIdx;
break;
} else {
spec->type = CFFormatCFType;
spec->size = CFFormatSizePointer; // 4 or 8 depending on LP64
return true;
}
case '1': case '2': case '3': case '4': case '5': case '6': case '7': case '8': case '9': {
long long number = 0;
do {
if (__builtin_smulll_overflow(number, 10, &number) || __builtin_saddll_overflow(number, ch - '0', &number) || number > INT64_MAX) {
if (errorPtr) *errorPtr = __CFCreateOverflowError();
return false;
}
if (cformat) ch = (UniChar)cformat[(*fmtIdx)++]; else ch = uformat[(*fmtIdx)++];
} while ((UInt32)(ch - '0') <= 9);
if ('$' == ch) {
if (number > INT8_MAX) {
if (errorPtr) *errorPtr = __CFCreateOverflowError();
return false;
}
if (-2 == spec->precArgNum) {
spec->precArgNum = (int8_t)number - 1; // Arg numbers start from 1
} else if (-2 == spec->widthArgNum) {
spec->widthArgNum = (int8_t)number - 1; // Arg numbers start from 1
} else {
spec->mainArgNum = (int8_t)number - 1; // Arg numbers start from 1
}
break;
} else if (seenDot) { /* else it's either precision or width */
if (number > INT32_MAX) {
if (errorPtr) *errorPtr = __CFCreateOverflowError();
return false;
}
spec->precArg = (SInt32)number;
} else {
if (number > INT32_MAX) {
if (errorPtr) *errorPtr = __CFCreateOverflowError();
return false;
}
spec->widthArg = (SInt32)number;
}
goto reswtch;
}
case '*':
spec->widthArgNum = -2;
break;
case '.':
seenDot = true;
if (cformat) ch = (UniChar)cformat[(*fmtIdx)++]; else ch = uformat[(*fmtIdx)++];
if ('*' == ch) {
spec->precArgNum = -2;
break;
}
goto reswtch;
default:
spec->type = CFFormatLiteralType;
return true;
}
}
return true;
}
// Length of the buffer to call sprintf() with
#define BUFFER_LEN 512
#if TARGET_OS_MAC
#define SNPRINTF(TYPE, WHAT) { \
TYPE value = (TYPE) WHAT; \
if (-1 != specs[curSpec].widthArgNum) { \
if (-1 != specs[curSpec].precArgNum) { \
snprintf_l(buffer, BUFFER_LEN-1, NULL, formatBuffer, width, precision, value); \
} else { \
snprintf_l(buffer, BUFFER_LEN-1, NULL, formatBuffer, width, value); \
} \
} else { \
if (-1 != specs[curSpec].precArgNum) { \
snprintf_l(buffer, BUFFER_LEN-1, NULL, formatBuffer, precision, value); \
} else { \
snprintf_l(buffer, BUFFER_LEN-1, NULL, formatBuffer, value); \
} \
}}
#else
#define SNPRINTF(TYPE, WHAT) { \
TYPE value = (TYPE) WHAT; \
if (-1 != specs[curSpec].widthArgNum) { \
if (-1 != specs[curSpec].precArgNum) { \
sprintf(buffer, formatBuffer, width, precision, value); \
} else { \
sprintf(buffer, formatBuffer, width, value); \
} \
} else { \
if (-1 != specs[curSpec].precArgNum) { \
sprintf(buffer, formatBuffer, precision, value); \
} else { \
sprintf(buffer, formatBuffer, value); \
} \
}}
#endif
CF_INLINE void _CFStringFormatReplacementDictionaryAppendRange(CFMutableDictionaryRef replacement, SInt32 specLoc, SInt32 specLen, CFIndex lengthBefore, CFIndex lengthAfter) {
CFNumberRef specLocation = CFNumberCreate(kCFAllocatorSystemDefault, kCFNumberSInt32Type, &specLoc);
CFDictionarySetValue(replacement, _kCFStringFormatMetadataSpecifierRangeLocationInFormatStringKey, specLocation);
CFRelease(specLocation);
CFNumberRef specLength = CFNumberCreate(kCFAllocatorSystemDefault, kCFNumberSInt32Type, &specLen);
CFDictionarySetValue(replacement, _kCFStringFormatMetadataSpecifierRangeLengthInFormatStringKey, specLength);
CFRelease(specLength);
CFNumberRef rangeLocationObject = CFNumberCreate(kCFAllocatorSystemDefault, kCFNumberCFIndexType, &lengthBefore);
CFDictionarySetValue(replacement, _kCFStringFormatMetadataReplacementRangeLocationKey, rangeLocationObject);
CFRelease(rangeLocationObject);
CFIndex length = MAX(lengthAfter - lengthBefore, 0);
CFNumberRef rangeLengthObject = CFNumberCreate(kCFAllocatorSystemDefault, kCFNumberCFIndexType, &length);
CFDictionarySetValue(replacement, _kCFStringFormatMetadataReplacementRangeLengthKey, rangeLengthObject);
CFRelease(rangeLengthObject);
}
CF_INLINE void _CFStringFormatReplacementDictionaryAppendArgumentIndex(CFMutableDictionaryRef replacement, int16_t type, int8_t mainArgNum, CFIndex valuesCount) {
if (mainArgNum < 0 || mainArgNum >= valuesCount || type == CFFormatLiteralType) {
return;
}
CFIndex userVisibleIndex = mainArgNum + 1;
CFNumberRef indexObject = CFNumberCreate(kCFAllocatorSystemDefault, kCFNumberCFIndexType, &userVisibleIndex);
CFDictionarySetValue(replacement, _kCFStringFormatMetadataReplacementIndexKey, indexObject);
CFRelease(indexObject);
}
CF_INLINE void _CFStringFormatReplacementDictionaryAppendArgumentValue(CFMutableDictionaryRef replacement, CFPrintValue *values, int16_t type, int8_t mainArgNum, CFIndex valuesCount) {
if (mainArgNum < 0 || mainArgNum >= valuesCount) {
return;
}
CFNumberRef numberValue = NULL;
CFPrintValue const value = values[mainArgNum];
switch (value.type) {
case CFFormatCFType:
if (value.value.pointerValue) {
CFDictionarySetValue(replacement, _kCFStringFormatMetadataArgumentObjectKey, value.value.pointerValue);
}
break;
case CFFormatLongType:
numberValue = CFNumberCreate(kCFAllocatorSystemDefault, kCFNumberSInt64Type, &value.value.int64Value);
CFDictionarySetValue(replacement, _kCFStringFormatMetadataArgumentNumberKey, numberValue);
CFRelease(numberValue);
break;
case CFFormatDoubleType:
numberValue = CFNumberCreate(kCFAllocatorSystemDefault, kCFNumberDoubleType, &value.value.doubleValue);
CFDictionarySetValue(replacement, _kCFStringFormatMetadataArgumentNumberKey, numberValue);
CFRelease(numberValue);
break;
default:
break;
}
}
static void _CFStringFormatAppendMetadata(CFMutableArrayRef *outReplacementMetadata, CFIndex specsCount, CFPrintValue *values, CFIndex valuesCount, CFFormatSpec spec, CFIndex lengthBefore, CFIndex lengthAfter) {
if (!outReplacementMetadata) {
return;
}
if (spec.type == CFFormatLiteralType) {
if ((spec.flags & kCFStringFormatPercentReplacementFlag) == 0) {
// This is a literal, unreplaced chunk. Do not add it as a replacement.
return;
}
} else if (spec.type != CFFormatIncompleteSpecifierType) {
if (spec.mainArgNum < 0) {
return;
}
assert(spec.mainArgNum < valuesCount);
if (spec.mainArgNum >= valuesCount) {
return;
}
}
if (!*outReplacementMetadata) {
*outReplacementMetadata = CFArrayCreateMutable(kCFAllocatorSystemDefault, specsCount, &kCFTypeArrayCallBacks);
}
CFMutableDictionaryRef replacement = CFDictionaryCreateMutable(kCFAllocatorSystemDefault, 2, &kCFTypeDictionaryKeyCallBacks, &kCFTypeDictionaryValueCallBacks);
_CFStringFormatReplacementDictionaryAppendRange(replacement, spec.loc, spec.len, lengthBefore, lengthAfter);
_CFStringFormatReplacementDictionaryAppendArgumentIndex(replacement, spec.type, spec.mainArgNum, valuesCount);
_CFStringFormatReplacementDictionaryAppendArgumentValue(replacement, values, spec.type, spec.mainArgNum, valuesCount);
CFArrayAppendValue(*outReplacementMetadata, replacement);
CFRelease(replacement);
}
/* These three functions are the external entry points for string formatting.
*/
void CFStringAppendFormatAndArguments(CFMutableStringRef outputString, CFDictionaryRef formatOptions, CFStringRef formatString, va_list args) {
CF_RETURN_IF_NOT_MUTABLE(outputString);
CFErrorRef error;
if (!__CFStringAppendFormatCore(outputString, NULL, NULL, formatOptions, NULL, NULL, formatString, 0, NULL, 0, args, NULL, &error)) {
CFLog(kCFLogLevelError, CFSTR("ERROR: Failed to format string: %@"), error);
if (error) CFRelease(error);
}
}
void _CFStringAppendFormatAndArgumentsAux2(CFMutableStringRef outputString, CFStringRef (*copyDescFunc)(void *, const void *), CFStringRef (*contextDescFunc)(void *, const void *, const void *, bool, bool *), CFDictionaryRef formatOptions, CFStringRef formatString, va_list args) {
CFErrorRef error;
if (!__CFStringAppendFormatCore(outputString, copyDescFunc, contextDescFunc, formatOptions, NULL, NULL, formatString, 0, NULL, 0, args, NULL, &error)) {
CFLog(kCFLogLevelError, CFSTR("ERROR: Failed to format string: %@"), error);
if (error) CFRelease(error);
}
}
void _CFStringAppendFormatAndArgumentsAux(CFMutableStringRef outputString, CFStringRef (*copyDescFunc)(void *, const void *), CFDictionaryRef formatOptions, CFStringRef formatString, va_list args) {
_CFStringAppendFormatAndArgumentsAux2(outputString, copyDescFunc, NULL, formatOptions, formatString, args);
}
SInt32 __CFStringFindFormatSpecifiersInString(const uint8_t *cformat, const UniChar *uformat, CFIndex formatLen, CFFormatSpec *specs, CFStringRef *formatSpecs, CFIndex *numFormatSpecs) {
SInt32 curSpec, formatIdx;
/* Collect format specification information from the format string */
for (curSpec = 0, formatIdx = 0; formatIdx < formatLen; curSpec++) {
SInt32 newFmtIdx;
specs[curSpec].loc = formatIdx;
specs[curSpec].len = 0;
specs[curSpec].size = 0;
specs[curSpec].type = 0;
specs[curSpec].flags = 0;
specs[curSpec].widthArg = -1;
specs[curSpec].precArg = -1;
specs[curSpec].mainArgNum = -1;
specs[curSpec].precArgNum = -1;
specs[curSpec].widthArgNum = -1;
specs[curSpec].configDictIndex = -1;
if (cformat) {
for (newFmtIdx = formatIdx; newFmtIdx < formatLen && '%' != cformat[newFmtIdx]; newFmtIdx++);
} else {
for (newFmtIdx = formatIdx; newFmtIdx < formatLen && '%' != uformat[newFmtIdx]; newFmtIdx++);
}
if (newFmtIdx != formatIdx) { /* Literal chunk */
if (curSpec > -1) {
curSpec--; /* Skip by writing the next spec over this one */
}
} else {
newFmtIdx++; /* Skip % */
__CFParseFormatSpec(uformat, cformat, &newFmtIdx, formatLen, &(specs[curSpec]), NULL, NULL);
if (CFFormatLiteralType == specs[curSpec].type) {
if (curSpec > -1) {
curSpec--; /* Skip literal chunks by writing the next spec over this one*/
}
} else {
specs[curSpec].len = newFmtIdx - formatIdx;
// Copy the format string out
switch (specs[curSpec].type) {
case CFFormatLongType:
case CFFormatDoubleType:
case CFFormatPointerType:
{
if (formatSpecs && numFormatSpecs) {
char formatBuffer[128];
SInt32 cidx, idx, loc;
loc = specs[curSpec].loc;
if (cformat) {
for (idx = 0, cidx = 0; cidx < specs[curSpec].len && idx < 128; idx++, cidx++) {
if ('$' == cformat[loc + cidx]) {
if (idx > -1) {
for (idx--; idx >= 0 && '0' <= formatBuffer[idx] && formatBuffer[idx] <= '9'; idx--);
}
} else {
formatBuffer[idx] = cformat[loc + cidx];
}
}
} else {
for (idx = 0, cidx = 0; cidx < specs[curSpec].len && idx < 128; idx++, cidx++) {
if ('$' == uformat[loc + cidx]) {
if (idx > -1) {
for (idx--; idx >= 0 && '0' <= formatBuffer[idx] && formatBuffer[idx] <= '9'; idx--);
}
} else {
formatBuffer[idx] = (int8_t)uformat[loc + cidx];
}
}
}
formatBuffer[idx] = '\0';
formatSpecs[(*numFormatSpecs)++] = CFStringCreateWithCString(kCFAllocatorDefault, formatBuffer, kCFStringEncodingUTF8);
}
}
break;
default:
break;
}
}
}
formatIdx = newFmtIdx;
}
return curSpec;
}
#define FORMAT_BUFFER_LEN 400
#define VPRINTF_BUFFER_LEN 61
static void __CFStringSetUpFormatAndSpecBuffers(CFStringRef formatString, CFIndex formatLen, const uint8_t **cformat, const UniChar **uformat, UniChar **formatChars, UniChar *localFormatBuffer, CFFormatSpec **specs, CFFormatSpec *localSpecsBuffer, CFStringRef **formatSpecs, CFStringRef *localFormatSpecsBuffer) {
SInt32 formatIdx, sizeSpecs = 0;
CFAllocatorRef tmpAlloc = __CFGetDefaultAllocator();
if (!CF_IS_OBJC(_kCFRuntimeIDCFString, formatString) && !CF_IS_SWIFT(CFStringGetTypeID(), formatString)) {
__CFAssertIsString(formatString);
if (!__CFStrIsUnicode(formatString)) {
*cformat = (const uint8_t *)__CFStrContents(formatString);
if (*cformat) *cformat += __CFStrSkipAnyLengthByte(formatString);
} else {
*uformat = (const UniChar *)__CFStrContents(formatString);
}
}
if (!(*cformat) && !(*uformat)) {
*formatChars = (formatLen > FORMAT_BUFFER_LEN) ? (UniChar *)CFAllocatorAllocate(tmpAlloc, formatLen * sizeof(UniChar), 0) : localFormatBuffer;
if (*formatChars != localFormatBuffer && __CFOASafe) __CFSetLastAllocationEventName(*formatChars, "CFString (temp)");
CFStringGetCharacters(formatString, CFRangeMake(0, formatLen), *formatChars);
*uformat = *formatChars;
}
if (*cformat) {
for (formatIdx = 0; formatIdx < formatLen; formatIdx++) if ('%' == (*cformat)[formatIdx]) sizeSpecs++;
} else {
for (formatIdx = 0; formatIdx < formatLen; formatIdx++) if ('%' == (*uformat)[formatIdx]) sizeSpecs++;
}
*specs = ((2 * sizeSpecs + 1) > VPRINTF_BUFFER_LEN) ? (CFFormatSpec *)CFAllocatorAllocate(tmpAlloc, (2 * sizeSpecs + 1) * sizeof(CFFormatSpec), 0) : localSpecsBuffer;
if (*specs != localSpecsBuffer && __CFOASafe) __CFSetLastAllocationEventName(*specs, "CFString (temp)");
*formatSpecs = ((2 * sizeSpecs + 1) > VPRINTF_BUFFER_LEN) ? (CFStringRef *)CFAllocatorAllocate(tmpAlloc, (2 * sizeSpecs + 1) * sizeof(CFStringRef), 0) : localFormatSpecsBuffer;
if (*formatSpecs != localFormatSpecsBuffer && __CFOASafe) __CFSetLastAllocationEventName(*formatSpecs, "CFString (temp)");
}
static CFIndex __CFStringValidateFormat(CFStringRef expected, CFStringRef untrustedFormat, CFIndex alreadyValidated, CFErrorRef *errorPtr) {
bool verified = true;
SInt32 numSpecsUntrusted = 0, numSpecsExpected = 0;
CFIndex formatLenUntrusted = 0, formatLenExpected = 0, numFormatSpecsUntrusted = 0, numFormatSpecsExpected = 0;
CFAllocatorRef tmpAlloc = __CFGetDefaultAllocator();
const uint8_t *cformatUntrusted = NULL;
const UniChar *uformatUntrusted = NULL;
const uint8_t *cformatExpected = NULL;
const UniChar *uformatExpected = NULL;
UniChar *formatCharsUntrusted = NULL;
UniChar *formatCharsExpected = NULL;
UniChar localFormatBufferUntrusted[FORMAT_BUFFER_LEN];
UniChar localFormatBufferExpected[FORMAT_BUFFER_LEN];
CFFormatSpec *specsUntrusted = NULL;
CFFormatSpec *specsExpected = NULL;
CFFormatSpec localSpecsBufferUntrusted[VPRINTF_BUFFER_LEN];
CFFormatSpec localSpecsBufferExpected[VPRINTF_BUFFER_LEN];
CFStringRef *formatSpecsUntrusted = NULL;
CFStringRef *formatSpecsExpected = NULL;
CFStringRef localFormatSpecsBufferUntrusted[VPRINTF_BUFFER_LEN];
CFStringRef localFormatSpecsBufferExpected[VPRINTF_BUFFER_LEN];
/* Set up */
// Untrusted
formatLenUntrusted = CFStringGetLength(untrustedFormat);
__CFStringSetUpFormatAndSpecBuffers(untrustedFormat, formatLenUntrusted, &cformatUntrusted, &uformatUntrusted, &formatCharsUntrusted, localFormatBufferUntrusted, &specsUntrusted, localSpecsBufferUntrusted, &formatSpecsUntrusted, localFormatSpecsBufferUntrusted);
// Expected
formatLenExpected = CFStringGetLength(expected);
__CFStringSetUpFormatAndSpecBuffers(expected, formatLenExpected, &cformatExpected, &uformatExpected, &formatCharsExpected, localFormatBufferExpected, &specsExpected, localSpecsBufferExpected, &formatSpecsExpected, localFormatSpecsBufferExpected);
/* Get info about format specifiers in both strings */
numSpecsUntrusted = __CFStringFindFormatSpecifiersInString(cformatUntrusted, uformatUntrusted, formatLenUntrusted, specsUntrusted, formatSpecsUntrusted, &numFormatSpecsUntrusted);
numSpecsExpected = __CFStringFindFormatSpecifiersInString(cformatExpected, uformatExpected, formatLenExpected, specsExpected, formatSpecsExpected, &numFormatSpecsExpected);
// if we accept zero untrusted specs we should accept fewer than expected as well
if ((numSpecsUntrusted <= (numSpecsExpected - alreadyValidated)) && (numFormatSpecsUntrusted <= numFormatSpecsExpected)) {
CFIndex idx;
for (idx = 0; idx < numSpecsUntrusted; idx++) {
int8_t argNum = specsUntrusted[idx].mainArgNum != -1 ? specsUntrusted[idx].mainArgNum : idx;
if (((argNum + alreadyValidated) >= numSpecsExpected) || (specsUntrusted[idx].type != specsExpected[argNum + alreadyValidated].type) || (specsUntrusted[idx].size != specsExpected[argNum].size)) {
verified = false;
break;
}
}
if (verified) {
for (idx = 0; idx < numFormatSpecsUntrusted; idx++) {
CFComparisonResult comp = CFStringCompare(formatSpecsUntrusted[idx], formatSpecsExpected[idx], 0);
if (comp != kCFCompareEqualTo) {
if (specsUntrusted[idx].numericFormatStyle != specsExpected[idx + alreadyValidated].numericFormatStyle) {
verified = false;
break;
}
}
}
}
} else {
// If the untrusted string doesn't have any format specifiers in it, we're still ok.
if (numSpecsUntrusted != 0) {
verified = false;
}
}
if (!verified) {
if (errorPtr) {
#if TARGET_OS_MAC || TARGET_OS_WIN32
CFStringRef debugMsg = CFStringCreateWithFormat(tmpAlloc, NULL, CFSTR("Format '%@' does not match expected '%@'"), untrustedFormat, expected);
CFMutableDictionaryRef userInfo = CFDictionaryCreateMutable(tmpAlloc, 0, &kCFCopyStringDictionaryKeyCallBacks, &kCFTypeDictionaryValueCallBacks);
CFDictionarySetValue(userInfo, kCFErrorDebugDescriptionKey, debugMsg);
*errorPtr = CFErrorCreate(tmpAlloc, kCFErrorDomainCocoa, NSFormattingError, userInfo);
CFRelease(userInfo);
CFRelease(debugMsg);
#endif
}
}
if (formatCharsUntrusted && (formatCharsUntrusted != localFormatBufferUntrusted)) CFAllocatorDeallocate(tmpAlloc, formatCharsUntrusted);
if (formatCharsExpected && (formatCharsExpected != localFormatBufferExpected)) CFAllocatorDeallocate(tmpAlloc, formatCharsExpected);
if (specsUntrusted != localSpecsBufferUntrusted) CFAllocatorDeallocate(tmpAlloc, specsUntrusted);
if (specsExpected != localSpecsBufferExpected) CFAllocatorDeallocate(tmpAlloc, specsExpected);
// Free allocated strings
CFIndex idx;
for (idx = 0; idx < numFormatSpecsUntrusted; idx++) {
if (formatSpecsUntrusted[idx]) { CFRelease(formatSpecsUntrusted[idx]); }
}
for (idx = 0; idx < numFormatSpecsExpected; idx++) {
if (formatSpecsExpected[idx]) { CFRelease(formatSpecsExpected[idx]); }
}
if (formatSpecsUntrusted != localFormatSpecsBufferUntrusted) CFAllocatorDeallocate(tmpAlloc, formatSpecsUntrusted);
if (formatSpecsExpected != localFormatSpecsBufferExpected) CFAllocatorDeallocate(tmpAlloc, formatSpecsExpected);
return (verified ? numSpecsUntrusted + alreadyValidated : -1);
}
/*
__CFStringAppendFormatCore(): The core for all string formatting.
outputString: Mutable string being formatted into. Results will be appended.
copyDescFunc: Callback for formatting strings. Can be NULL. Foundation calls with a function that will invoke descriptionWithLocale: or description. Second argument is the locale (a dictionary or locale).
contextDescFunc: Callback for doing context-based formatting. Can be NULL. <<!!! Describe the arguments>>
formatOptions: Locale specific info. Used to be a CFDictionary, now CFLocale, but either is still possible. If !NULL, localized formatting is assumed.
stringsDictConfig: Only used for recursive calls when doing stringsDict formatting, or as a fallback for when the config is not stored in the passed formatString (formatting an NSAttributedString). Otherwise NULL.
validFormatSpecifiers: Only used to validate the format specifiers in the formatString. A string that contains an in order sequence of the valid format specifiers.
formatString: The actual format string.
initialArgPosition: Only used for recursive calls when doing stringsDict formatting. Otherwise 0. <<!!! Confirm>>
origValues: Only used for recursive calls when doing stringsDict formatting. Otherwise NULL. <<!!! Confirm>>
originalValuesSize: Only used for recursive calls when doing stringsDict formatting. Otherwise 0. <<!!! Confirm>>
args: The arguments to be formatted.
outReplacementMetadata: On return, contains information on the replacements applied for the specifiers. If NULL, no metadata is generated.
errorPtr: Only used when validating the formatString against valid format specfiers. Nil unless the validation fails. If error is set, return value is false.
??? %s depends on handling of encodings by __CFStringAppendBytes
*/
static Boolean __CFStringAppendFormatCore(CFMutableStringRef outputString, CFStringRef (*copyDescFunc)(void *, const void *), CFStringRef (*contextDescFunc)(void *, const void *, const void *, bool, bool *), CFDictionaryRef formatOptions, CFDictionaryRef stringsDictConfig, CFStringRef validFormatSpecifiers, CFStringRef formatString, CFIndex initialArgPosition, const void *origValues, CFIndex originalValuesSize, va_list args, CFArrayRef *outReplacementMetadata, CFErrorRef *errorPtr) {
int32_t numSpecs, sizeSpecs, sizeArgNum, formatIdx, curSpec, argNum;
CFIndex formatLen;
const uint8_t *cformat = NULL;
const UniChar *uformat = NULL;
UniChar *formatChars = NULL;
UniChar localFormatBuffer[FORMAT_BUFFER_LEN];
CFFormatSpec localSpecsBuffer[VPRINTF_BUFFER_LEN];
CFFormatSpec *specs;
CFPrintValue localValuesBuffer[VPRINTF_BUFFER_LEN];
CFPrintValue *values;
const CFPrintValue *originalValues = (const CFPrintValue *)origValues;
CFDictionaryRef localConfigs[VPRINTF_BUFFER_LEN];
CFDictionaryRef *configs;
CFMutableDictionaryRef formattingConfig = NULL;
CFIndex numConfigs;
CFAllocatorRef tmpAlloc = NULL;
bool localizedFormatting = formatOptions && (CFGetTypeID(formatOptions) == CFLocaleGetTypeID());
CFMutableArrayRef metadataStorage = NULL;
CFMutableArrayRef *metadata = outReplacementMetadata ? &metadataStorage : NULL;
intmax_t dummyLocation; // A place for %n to do its thing in; should be the widest possible int value
numSpecs = 0;
sizeSpecs = 0;
sizeArgNum = 0;
numConfigs = 0;
specs = NULL;
values = NULL;
configs = NULL;
if (validFormatSpecifiers) {
const uint8_t *cExpectedFormat = NULL;
const UniChar *uExpectedFormat = NULL;
UniChar *expectedFormatChars = NULL;
UniChar expectedLocalFormatBuffer[FORMAT_BUFFER_LEN];
CFFormatSpec *expectedSpecs = NULL;
CFFormatSpec localExpectedSpecsBuffer[VPRINTF_BUFFER_LEN];
CFStringRef *expectedFormatSpecs = NULL;
CFStringRef expectedFormatSpecsBuffer[VPRINTF_BUFFER_LEN];
CFIndex expectedFormatLen = CFStringGetLength(validFormatSpecifiers);
__CFStringSetUpFormatAndSpecBuffers(validFormatSpecifiers, expectedFormatLen, &cExpectedFormat, &uExpectedFormat, &expectedFormatChars, expectedLocalFormatBuffer, &expectedSpecs, localExpectedSpecsBuffer, &expectedFormatSpecs, expectedFormatSpecsBuffer);
SInt32 numExpectedSpecs = __CFStringFindFormatSpecifiersInString(cExpectedFormat, uExpectedFormat, expectedFormatLen, expectedSpecs, NULL, NULL);
if (expectedFormatChars != expectedLocalFormatBuffer) CFAllocatorDeallocate(tmpAlloc, expectedFormatChars);
if (expectedSpecs != localExpectedSpecsBuffer) CFAllocatorDeallocate(tmpAlloc, expectedSpecs);
if (expectedFormatSpecs != expectedFormatSpecsBuffer) CFAllocatorDeallocate(tmpAlloc, expectedFormatSpecs);
if (expectedFormatLen == 0 || numExpectedSpecs == 0) {
if (errorPtr) {
CFStringRef debugMsg = CFStringCreateWithFormat(tmpAlloc, NULL, CFSTR("Expected format '%@' is invalid"), validFormatSpecifiers);
CFMutableDictionaryRef userInfo = CFDictionaryCreateMutable(tmpAlloc, 0, &kCFCopyStringDictionaryKeyCallBacks, &kCFTypeDictionaryValueCallBacks);
CFDictionarySetValue(userInfo, kCFErrorDebugDescriptionKey, debugMsg);
*errorPtr = CFErrorCreate(tmpAlloc, kCFErrorDomainCocoa, NSFormattingError, userInfo);
CFRelease(debugMsg);
CFRelease(userInfo);
}
return false;
}
/* Validate expected format specifiers against untrusted format string */
if (__CFStringValidateFormat(validFormatSpecifiers, formatString, 0, errorPtr) < 0) {
return false;
}
}
Boolean success = true;
formatLen = CFStringGetLength(formatString);
/* Avoid overflow of formatIdx in loops below that compare to formatIdx */
if (formatLen > INT32_MAX) {
if (errorPtr) *errorPtr = __CFCreateOverflowError();
success = false;
goto cleanup;
}
if (!CF_IS_OBJC(_kCFRuntimeIDCFString, formatString) && !CF_IS_SWIFT(CFStringGetTypeID(), formatString)) {
__CFAssertIsString(formatString);
if (!__CFStrIsUnicode(formatString)) {
cformat = (const uint8_t *)__CFStrContents(formatString);
if (cformat) cformat += __CFStrSkipAnyLengthByte(formatString);
} else {
uformat = (const UniChar *)__CFStrContents(formatString);
}
}
if (!cformat && !uformat) {
formatChars = (formatLen > FORMAT_BUFFER_LEN) ? (UniChar *)CFAllocatorAllocate(tmpAlloc = __CFGetDefaultAllocator(), formatLen * sizeof(UniChar), 0) : localFormatBuffer;
if (formatChars != localFormatBuffer && __CFOASafe) __CFSetLastAllocationEventName(formatChars, "CFString (temp)");
CFStringGetCharacters(formatString, CFRangeMake(0, formatLen), formatChars);
uformat = formatChars;
}
/* Compute an upper bound for the number of format specifications */
if (cformat) {
for (formatIdx = 0; formatIdx < formatLen; formatIdx++) {
if ('%' == cformat[formatIdx]) {
if (__builtin_sadd_overflow(sizeSpecs, 1, &sizeSpecs)) {
if (errorPtr) *errorPtr = __CFCreateOverflowError();
success = false;
goto cleanup;
}
}
}
} else {
for (formatIdx = 0; formatIdx < formatLen; formatIdx++) {
if ('%' == uformat[formatIdx]) {
if (__builtin_sadd_overflow(sizeSpecs, 1, &sizeSpecs)) {
if (errorPtr) *errorPtr = __CFCreateOverflowError();
success = false;
goto cleanup;
}
}
}
}
/* The code following this point makes multiple integer calcuations based off sizeSpecs, which is an upper-bound of the number of tokens in the string based on the number of '%' characters found. In order to avoid integer overflow at multiple points throughout this function, we will cap sizeSpec to an amount that won't overflow, but is still plenty large enough to support any reasonable format strings */
#define MAX_SIZE_SPECS (0xfffff) /* Won't overflow a 32-bit integer up to and including a multiplicative factor of 256 */
if (sizeSpecs > MAX_SIZE_SPECS) {
if (errorPtr) *errorPtr = __CFCreateOverflowError();
success = false;
goto cleanup;
}
tmpAlloc = __CFGetDefaultAllocator();
specs = ((2 * sizeSpecs + 1) > VPRINTF_BUFFER_LEN) ? (CFFormatSpec *)CFAllocatorAllocate(tmpAlloc, (2 * sizeSpecs + 1) * sizeof(CFFormatSpec), 0) : localSpecsBuffer;
if (specs != localSpecsBuffer && __CFOASafe) __CFSetLastAllocationEventName(specs, "CFString (temp)");
configs = ((sizeSpecs < VPRINTF_BUFFER_LEN) ? localConfigs : (CFDictionaryRef *)CFAllocatorAllocate(tmpAlloc, sizeof(CFStringRef) * sizeSpecs, 0));
/* Collect format specification information from the format string */
for (curSpec = 0, formatIdx = 0; formatIdx < formatLen; curSpec++) {
SInt32 newFmtIdx;
specs[curSpec].loc = formatIdx;
specs[curSpec].len = 0;
specs[curSpec].size = 0;
specs[curSpec].type = 0;
specs[curSpec].flags = 0;
specs[curSpec].widthArg = -1;
specs[curSpec].precArg = -1;
specs[curSpec].mainArgNum = -1;
specs[curSpec].precArgNum = -1;
specs[curSpec].widthArgNum = -1;
specs[curSpec].configDictIndex = -1;
if (cformat) {
for (newFmtIdx = formatIdx; newFmtIdx < formatLen && '%' != cformat[newFmtIdx]; newFmtIdx++);
} else {
for (newFmtIdx = formatIdx; newFmtIdx < formatLen && '%' != uformat[newFmtIdx]; newFmtIdx++);
}
if (newFmtIdx != formatIdx) { /* Literal chunk */
specs[curSpec].type = CFFormatLiteralType;
specs[curSpec].len = newFmtIdx - formatIdx;
} else {
CFStringRef configKey = NULL;
newFmtIdx++; /* Skip % */
if (!__CFParseFormatSpec(uformat, cformat, &newFmtIdx, formatLen, &(specs[curSpec]), &configKey, errorPtr)) {
success = false;
goto cleanup;
}
if (CFFormatLiteralType == specs[curSpec].type) {
specs[curSpec].loc = formatIdx + 1;
specs[curSpec].len = 1;
specs[curSpec].flags |= kCFStringFormatPercentReplacementFlag;
} else {
specs[curSpec].len = newFmtIdx - formatIdx;
}
}
formatIdx = newFmtIdx;
// fprintf(stderr, "specs[%d] = {\n size = %d,\n type = %d,\n loc = %d,\n len = %d,\n mainArgNum = %d,\n precArgNum = %d,\n widthArgNum = %d\n}\n", curSpec, specs[curSpec].size, specs[curSpec].type, specs[curSpec].loc, specs[curSpec].len, specs[curSpec].mainArgNum, specs[curSpec].precArgNum, specs[curSpec].widthArgNum);
}
numSpecs = curSpec;
if (originalValues == NULL) {
// Max of three args per spec, reasoning thus: 1 width, 1 prec, 1 value
sizeArgNum = 3 * sizeSpecs + 1;
} else {
#define MAX_SIZE_ORIGINAL_VALUES (0x2ffffe) /* Max size of original values is (3 * MAX_SIZE_SPECS + 1) */
if (originalValuesSize > MAX_SIZE_ORIGINAL_VALUES) {
if (errorPtr) *errorPtr = __CFCreateOverflowError();
success = false;
goto cleanup;
}
sizeArgNum = originalValuesSize;
}
values = (sizeArgNum > VPRINTF_BUFFER_LEN) ? (CFPrintValue *)CFAllocatorAllocate(tmpAlloc, sizeArgNum * sizeof(CFPrintValue), 0) : localValuesBuffer;
if (values != localValuesBuffer && __CFOASafe) __CFSetLastAllocationEventName(values, "CFString (temp)");
memset(values, 0, sizeArgNum * sizeof(CFPrintValue));
va_list copiedArgs;
if (numConfigs > 0) va_copy(copiedArgs, args); // we need to preserve the original state for passing down
/* Compute values array */
argNum = initialArgPosition;
CFIndex validatedDictSpecs = 0;
for (curSpec = 0; curSpec < numSpecs; curSpec++) {
SInt32 newMaxArgNum;
if (0 == specs[curSpec].type) continue;
if (CFFormatLiteralType == specs[curSpec].type) continue;
if (CFFormatIncompleteSpecifierType == specs[curSpec].type) continue;
newMaxArgNum = sizeArgNum;
if (newMaxArgNum < specs[curSpec].mainArgNum) {
newMaxArgNum = specs[curSpec].mainArgNum;
}
if (newMaxArgNum < specs[curSpec].precArgNum) {
newMaxArgNum = specs[curSpec].precArgNum;
}
if (newMaxArgNum < specs[curSpec].widthArgNum) {
newMaxArgNum = specs[curSpec].widthArgNum;
}
if (sizeArgNum < newMaxArgNum) {
if (specs != localSpecsBuffer) CFAllocatorDeallocate(tmpAlloc, specs);
if (values != localValuesBuffer) CFAllocatorDeallocate(tmpAlloc, values);
if (formatChars && (formatChars != localFormatBuffer)) CFAllocatorDeallocate(tmpAlloc, formatChars);
// More arguments than expected - not an error case though.
return true;
}
/* It is actually incorrect to reorder some specs and not all; we just do some random garbage here */
if (-2 == specs[curSpec].widthArgNum) {
specs[curSpec].widthArgNum = argNum++;
}
if (-2 == specs[curSpec].precArgNum) {
specs[curSpec].precArgNum = argNum++;
}
if (-1 == specs[curSpec].mainArgNum) {
specs[curSpec].mainArgNum = argNum++;
}
values[specs[curSpec].mainArgNum].size = specs[curSpec].size;
values[specs[curSpec].mainArgNum].type = specs[curSpec].type;
if (-1 != specs[curSpec].widthArgNum) {
values[specs[curSpec].widthArgNum].size = 0;
values[specs[curSpec].widthArgNum].type = CFFormatLongType;
}
if (-1 != specs[curSpec].precArgNum) {
values[specs[curSpec].precArgNum].size = 0;
values[specs[curSpec].precArgNum].type = CFFormatLongType;
}
}
CFIndex validatedInnerSpecs = 0;
/* Collect the arguments in correct type from vararg list */
for (argNum = 0; argNum < sizeArgNum; argNum++) {
if ((NULL != originalValues) && (0 == values[argNum].type)) values[argNum] = originalValues[argNum];
switch (values[argNum].type) {
case 0:
case CFFormatIncompleteSpecifierType:
case CFFormatLiteralType:
break;
case CFFormatLongType:
case CFFormatSingleUnicharType:
if (CFFormatSize1 == values[argNum].size) {
values[argNum].value.int64Value = (int64_t)(int8_t)va_arg(args, int);
} else if (CFFormatSize2 == values[argNum].size) {
values[argNum].value.int64Value = (int64_t)(int16_t)va_arg(args, int);
} else if (CFFormatSize4 == values[argNum].size) {
values[argNum].value.int64Value = (int64_t)va_arg(args, int32_t);
} else if (CFFormatSize8 == values[argNum].size) {
values[argNum].value.int64Value = (int64_t)va_arg(args, int64_t);
} else {
values[argNum].value.int64Value = (int64_t)va_arg(args, int);
}
break;
case CFFormatDoubleType:
#if LONG_DOUBLE_SUPPORT
if (CFFormatSize16 == values[argNum].size) {
values[argNum].value.longDoubleValue = va_arg(args, long double);
} else
#endif
{
values[argNum].value.doubleValue = va_arg(args, double);
}
break;
case CFFormatPointerType:
case CFFormatCFType:
case CFFormatUnicharsType:
case CFFormatCharsType:
case CFFormatPascalCharsType:
values[argNum].value.pointerValue = va_arg(args, void *);
break;
case CFFormatDummyPointerType:
(void)va_arg(args, void *); // Skip the provided argument
values[argNum].value.pointerValue = &dummyLocation;
break;
}
}
va_end(args);
/* Format the pieces together */
if (NULL == originalValues) {
originalValues = values;
originalValuesSize = sizeArgNum;
}
SInt32 numSpecsContext = 0;
CFFormatSpec *specsContext = NULL;
if (numSpecs > 0) {
specsContext = (CFFormatSpec *)calloc(numSpecs, sizeof(CFFormatSpec));
}
const CFStringRef replacement = CFSTR("%@NSCONTEXT");
for (curSpec = 0; curSpec < numSpecs; curSpec++) {
SInt32 width = 0, precision = 0;
UniChar *up, ch;
Boolean hasWidth = false, hasPrecision = false;
// widthArgNum and widthArg are never set at the same time; same for precArg*
if (-1 != specs[curSpec].widthArgNum) {
width = (SInt32)values[specs[curSpec].widthArgNum].value.int64Value;
hasWidth = true;
}
if (-1 != specs[curSpec].precArgNum) {
precision = (SInt32)values[specs[curSpec].precArgNum].value.int64Value;
hasPrecision = true;
}
if (-1 != specs[curSpec].widthArg) {
width = specs[curSpec].widthArg;
hasWidth = true;
}
if (-1 != specs[curSpec].precArg) {
precision = specs[curSpec].precArg;
hasPrecision = true;
}
#define __CFStringFormatOutputLengthIfNeeded() (metadata ? CFStringGetLength(outputString) : 0)
CFIndex oldLength = 0;
switch (specs[curSpec].type) {
case CFFormatLongType:
case CFFormatDoubleType:
#if TARGET_OS_MAC || TARGET_OS_WIN32 || TARGET_OS_LINUX || TARGET_OS_WASI
if (localizedFormatting && (specs[curSpec].flags & kCFStringFormatLocalizable)) { // We have a locale, so we do localized formatting
oldLength = __CFStringFormatOutputLengthIfNeeded();
if (__CFStringFormatLocalizedNumber(outputString, (CFLocaleRef)formatOptions, values, &specs[curSpec], width, precision, hasPrecision)) {
_CFStringFormatAppendMetadata(metadata, numSpecs, values, sizeArgNum, specs[curSpec], oldLength, __CFStringFormatOutputLengthIfNeeded());
break;
}
}
/* Otherwise fall-thru to the next case! */
#endif
case CFFormatPointerType: {
char stackFormatBuffer[128];
char *dynamicFormatBuffer = NULL;
char *formatBuffer = stackFormatBuffer;
if (specs[curSpec].len + 1 > sizeof(stackFormatBuffer)) {
// Leave space for the null terminator (+1)
dynamicFormatBuffer = (char *)CFAllocatorAllocate(kCFAllocatorSystemDefault, specs[curSpec].len + 1, 0);
formatBuffer = dynamicFormatBuffer;
}
#define EXTRA_BUFFER_LEN_FOR_WIDTH_PRECISION (16)
char stackBuffer[BUFFER_LEN + EXTRA_BUFFER_LEN_FOR_WIDTH_PRECISION];
char *dynamicBuffer = NULL;
char *buffer = stackBuffer;
size_t bufferSize = sizeof(stackBuffer);
if (width + precision > EXTRA_BUFFER_LEN_FOR_WIDTH_PRECISION) {
bufferSize = BUFFER_LEN + width + precision;
dynamicBuffer = (char *)CFAllocatorAllocate(kCFAllocatorSystemDefault, bufferSize, 0);
buffer = dynamicBuffer;
}
SInt32 cidx, idx, loc;
Boolean appended = false;
loc = specs[curSpec].loc;
// In preparation to call snprintf(), copy the format string out
if (cformat) {
for (idx = 0, cidx = 0; cidx < specs[curSpec].len; idx++, cidx++) {
if ('$' == cformat[loc + cidx]) {
if (idx > -1) {
for (idx--; '0' <= formatBuffer[idx] && formatBuffer[idx] <= '9'; idx--);
}
} else {
formatBuffer[idx] = cformat[loc + cidx];
}
}
} else {
for (idx = 0, cidx = 0; cidx < specs[curSpec].len; idx++, cidx++) {
if ('$' == uformat[loc + cidx]) {
if (idx > -1) {
for (idx--; '0' <= formatBuffer[idx] && formatBuffer[idx] <= '9'; idx--);
}
} else {
formatBuffer[idx] = (int8_t)uformat[loc + cidx];
}
}
}
formatBuffer[idx] = '\0';
// Should modify format buffer here if necessary; for example, to translate %qd to
// the equivalent, on architectures which do not have %q.
buffer[bufferSize - 1] = '\0';
oldLength = __CFStringFormatOutputLengthIfNeeded();
switch (specs[curSpec].type) {
case CFFormatLongType:
if (CFFormatSize8 == specs[curSpec].size) {
SNPRINTF(int64_t, values[specs[curSpec].mainArgNum].value.int64Value)
} else {
SNPRINTF(SInt32, values[specs[curSpec].mainArgNum].value.int64Value)
}
break;
case CFFormatPointerType:
case CFFormatDummyPointerType:
SNPRINTF(void *, values[specs[curSpec].mainArgNum].value.pointerValue)
break;
case CFFormatDoubleType:
#if LONG_DOUBLE_SUPPORT
if (CFFormatSize16 == specs[curSpec].size) {
SNPRINTF(long double, values[specs[curSpec].mainArgNum].value.longDoubleValue)
} else
#endif
{
SNPRINTF(double, values[specs[curSpec].mainArgNum].value.doubleValue)
}
// See if we need to localize the decimal point
if (formatOptions) { // We have localization info
#if TARGET_OS_MAC || TARGET_OS_WIN32 || TARGET_OS_LINUX || TARGET_OS_BSD
CFStringRef decimalSeparator = (CFGetTypeID(formatOptions) == CFLocaleGetTypeID()) ? (CFStringRef)CFLocaleGetValue((CFLocaleRef)formatOptions, kCFLocaleDecimalSeparatorKey) : (CFStringRef)CFDictionaryGetValue(formatOptions, CFSTR("NSDecimalSeparator"));
#else
CFStringRef decimalSeparator = CFSTR(".");
#endif
if (decimalSeparator != NULL) { // We have a decimal separator in there
CFIndex decimalPointLoc = 0;
while (buffer[decimalPointLoc] != 0 && buffer[decimalPointLoc] != '.') decimalPointLoc++;
if (buffer[decimalPointLoc] == '.') { // And we have a decimal point in the formatted string
buffer[decimalPointLoc] = 0;
CFStringAppendCString(outputString, (const char *)buffer, __CFStringGetEightBitStringEncoding());
CFStringAppend(outputString, decimalSeparator);
CFStringAppendCString(outputString, (const char *)(buffer + decimalPointLoc + 1), __CFStringGetEightBitStringEncoding());
appended = true;
}
}
}
break;
}
if (!appended) CFStringAppendCString(outputString, (const char *)buffer, __CFStringGetEightBitStringEncoding());
_CFStringFormatAppendMetadata(metadata, numSpecs, values, sizeArgNum, specs[curSpec], oldLength, __CFStringFormatOutputLengthIfNeeded());
if (dynamicBuffer) {
CFAllocatorDeallocate(kCFAllocatorSystemDefault, dynamicBuffer);
}
if (dynamicFormatBuffer) {
CFAllocatorDeallocate(kCFAllocatorSystemDefault, dynamicFormatBuffer);
}
}
break;
case CFFormatLiteralType:
oldLength = __CFStringFormatOutputLengthIfNeeded();
if (cformat) {
__CFStringAppendBytes(outputString, (const char *)(cformat+specs[curSpec].loc), specs[curSpec].len, __CFStringGetEightBitStringEncoding());
} else {
CFStringAppendCharacters(outputString, uformat+specs[curSpec].loc, specs[curSpec].len);
}
_CFStringFormatAppendMetadata(metadata, numSpecs, values, sizeArgNum, specs[curSpec], oldLength, __CFStringFormatOutputLengthIfNeeded());
break;
case CFFormatIncompleteSpecifierType:
oldLength = __CFStringFormatOutputLengthIfNeeded();
_CFStringFormatAppendMetadata(metadata, numSpecs, values, sizeArgNum, specs[curSpec], oldLength, oldLength);
break;
case CFFormatPascalCharsType:
case CFFormatCharsType:
oldLength = __CFStringFormatOutputLengthIfNeeded();
if (values[specs[curSpec].mainArgNum].value.pointerValue == NULL) {
CFStringAppendCString(outputString, "(null)", kCFStringEncodingASCII);
} else {
int len;
const char *str = (const char *)values[specs[curSpec].mainArgNum].value.pointerValue;
if (specs[curSpec].type == CFFormatPascalCharsType) { // Pascal string case
len = ((unsigned char *)str)[0];
str++;
if (hasPrecision && precision < len) len = precision;
} else { // C-string case
if (!hasPrecision) { // No precision, so rely on the terminating null character
len = strlen(str);
} else { // Don't blindly call strlen() if there is a precision; the string might not have a terminating null (3131988)
const char *terminatingNull = (const char *)memchr(str, 0, precision); // Basically strlen() on only the first precision characters of str
if (terminatingNull) { // There was a null in the first precision characters
len = terminatingNull - str;
} else {
len = precision;
}
}
}
// Since the spec says the behavior of the ' ', '0', '#', and '+' flags is undefined for
// '%s', and since we have ignored them in the past, the behavior is hereby cast in stone
// to ignore those flags (and, say, never pad with '0' instead of space).
if (specs[curSpec].flags & kCFStringFormatMinusFlag) {
__CFStringAppendBytes(outputString, str, len, __CFStringGetSystemEncoding());
if (hasWidth && width > len) {
int w = width - len; // We need this many spaces; do it ten at a time
do {__CFStringAppendBytes(outputString, " ", (w > 10 ? 10 : w), kCFStringEncodingASCII);} while ((w -= 10) > 0);
}
} else {
if (hasWidth && width > len) {
int w = width - len; // We need this many spaces; do it ten at a time
do {__CFStringAppendBytes(outputString, " ", (w > 10 ? 10 : w), kCFStringEncodingASCII);} while ((w -= 10) > 0);
}
__CFStringAppendBytes(outputString, str, len, __CFStringGetSystemEncoding());
}
}
_CFStringFormatAppendMetadata(metadata, numSpecs, values, sizeArgNum, specs[curSpec], oldLength, __CFStringFormatOutputLengthIfNeeded());
break;
case CFFormatSingleUnicharType:
oldLength = __CFStringFormatOutputLengthIfNeeded();
ch = (UniChar)values[specs[curSpec].mainArgNum].value.int64Value;
CFStringAppendCharacters(outputString, &ch, 1);
_CFStringFormatAppendMetadata(metadata, numSpecs, values, sizeArgNum, specs[curSpec], oldLength, __CFStringFormatOutputLengthIfNeeded());
break;
case CFFormatUnicharsType:
oldLength = __CFStringFormatOutputLengthIfNeeded();
up = (UniChar *)values[specs[curSpec].mainArgNum].value.pointerValue;
if (NULL == up) {
CFStringAppendCString(outputString, "(null)", kCFStringEncodingASCII);
} else {
int len;
if (hasPrecision) {
// if we have precision, still pay attention to earlier null termination, as we do with %s (19784466)
for (len = 0; (len < precision) && (0 != up[len]); len++);
} else {
// if no precision, then simply find the length
for (len = 0; 0 != up[len]; len++);
}
// Since the spec says the behavior of the ' ', '0', '#', and '+' flags is undefined for
// '%s', and since we have ignored them in the past, the behavior is hereby cast in stone
// to ignore those flags (and, say, never pad with '0' instead of space).
if (specs[curSpec].flags & kCFStringFormatMinusFlag) {
CFStringAppendCharacters(outputString, up, len);
if (hasWidth && width > len) {
int w = width - len; // We need this many spaces; do it ten at a time
do {__CFStringAppendBytes(outputString, " ", (w > 10 ? 10 : w), kCFStringEncodingASCII);} while ((w -= 10) > 0);
}
} else {
if (hasWidth && width > len) {
int w = width - len; // We need this many spaces; do it ten at a time
do {__CFStringAppendBytes(outputString, " ", (w > 10 ? 10 : w), kCFStringEncodingASCII);} while ((w -= 10) > 0);
}
CFStringAppendCharacters(outputString, up, len);
}
}
_CFStringFormatAppendMetadata(metadata, numSpecs, values, sizeArgNum, specs[curSpec], oldLength, __CFStringFormatOutputLengthIfNeeded());
break;
case CFFormatCFType:
oldLength = __CFStringFormatOutputLengthIfNeeded();
if (specs[curSpec].configDictIndex != -1) { // config dict
CFTypeRef object = NULL;
switch (values[specs[curSpec].mainArgNum].type) {
case CFFormatLongType:
object = CFNumberCreate(tmpAlloc, kCFNumberSInt64Type, &(values[specs[curSpec].mainArgNum].value.int64Value));
break;
case CFFormatDoubleType:
#if LONG_DOUBLE_SUPPORT
if (CFFormatSize16 == values[specs[curSpec].mainArgNum].size) {
double aValue = values[specs[curSpec].mainArgNum].value.longDoubleValue; // losing precision
object = CFNumberCreate(tmpAlloc, kCFNumberDoubleType, &aValue);
} else
#endif
{
object = CFNumberCreate(tmpAlloc, kCFNumberDoubleType, &(values[specs[curSpec].mainArgNum].value.doubleValue));
}
break;
case CFFormatPointerType:
object = CFNumberCreate(tmpAlloc, kCFNumberCFIndexType, &(values[specs[curSpec].mainArgNum].value.pointerValue));
break;
case CFFormatPascalCharsType:
case CFFormatCharsType:
if (NULL != values[specs[curSpec].mainArgNum].value.pointerValue) {
CFMutableStringRef aString = CFStringCreateMutable(tmpAlloc, 0);
int len;
const char *str = (const char *)values[specs[curSpec].mainArgNum].value.pointerValue;
if (specs[curSpec].type == CFFormatPascalCharsType) { // Pascal string case
len = ((unsigned char *)str)[0];
str++;
if (hasPrecision && precision < len) len = precision;
} else { // C-string case
if (!hasPrecision) { // No precision, so rely on the terminating null character
len = strlen(str);
} else { // Don't blindly call strlen() if there is a precision; the string might not have a terminating null (3131988)
const char *terminatingNull = (const char *)memchr(str, 0, precision); // Basically strlen() on only the first precision characters of str
if (terminatingNull) { // There was a null in the first precision characters
len = terminatingNull - str;
} else {
len = precision;
}
}
}
// Since the spec says the behavior of the ' ', '0', '#', and '+' flags is undefined for
// '%s', and since we have ignored them in the past, the behavior is hereby cast in stone
// to ignore those flags (and, say, never pad with '0' instead of space).
if (specs[curSpec].flags & kCFStringFormatMinusFlag) {
__CFStringAppendBytes(aString, str, len, __CFStringGetSystemEncoding());
if (hasWidth && width > len) {
int w = width - len; // We need this many spaces; do it ten at a time
do {__CFStringAppendBytes(aString, " ", (w > 10 ? 10 : w), kCFStringEncodingASCII);} while ((w -= 10) > 0);
}
} else {
if (hasWidth && width > len) {
int w = width - len; // We need this many spaces; do it ten at a time
do {__CFStringAppendBytes(aString, " ", (w > 10 ? 10 : w), kCFStringEncodingASCII);} while ((w -= 10) > 0);
}
__CFStringAppendBytes(aString, str, len, __CFStringGetSystemEncoding());
}
object = aString;
}
break;
case CFFormatSingleUnicharType:
ch = (UniChar)values[specs[curSpec].mainArgNum].value.int64Value;
object = CFStringCreateWithCharactersNoCopy(tmpAlloc, &ch, 1, kCFAllocatorNull);
break;
case CFFormatUnicharsType:
//??? need to handle width, precision, and padding arguments
up = (UniChar *)values[specs[curSpec].mainArgNum].value.pointerValue;
if (NULL != up) {
CFMutableStringRef aString = CFStringCreateMutable(tmpAlloc, 0);
int len;
for (len = 0; 0 != up[len]; len++);
// Since the spec says the behavior of the ' ', '0', '#', and '+' flags is undefined for
// '%s', and since we have ignored them in the past, the behavior is hereby cast in stone
// to ignore those flags (and, say, never pad with '0' instead of space).
if (hasPrecision && precision < len) len = precision;
if (specs[curSpec].flags & kCFStringFormatMinusFlag) {
CFStringAppendCharacters(aString, up, len);
if (hasWidth && width > len) {
int w = width - len; // We need this many spaces; do it ten at a time
do {__CFStringAppendBytes(aString, " ", (w > 10 ? 10 : w), kCFStringEncodingASCII);} while ((w -= 10) > 0);
}
} else {
if (hasWidth && width > len) {
int w = width - len; // We need this many spaces; do it ten at a time
do {__CFStringAppendBytes(aString, " ", (w > 10 ? 10 : w), kCFStringEncodingASCII);} while ((w -= 10) > 0);
}
CFStringAppendCharacters(aString, up, len);
}
object = aString;
}
break;
case CFFormatCFType:
if (NULL != values[specs[curSpec].mainArgNum].value.pointerValue) object = CFRetain(values[specs[curSpec].mainArgNum].value.pointerValue);
break;
}
if (NULL != object) CFRelease(object);
} else if (NULL != values[specs[curSpec].mainArgNum].value.pointerValue) {
CFStringRef str = NULL;
if (contextDescFunc) {
bool found = NO;
str = contextDescFunc(values[specs[curSpec].mainArgNum].value.pointerValue, formatString, replacement, NO, &found);
if (found) {
str = CFRetain(replacement);
specsContext[numSpecsContext] = specs[curSpec];
numSpecsContext++;
}
}
if (!str) {
if (copyDescFunc) {
str = copyDescFunc(values[specs[curSpec].mainArgNum].value.pointerValue, formatOptions);
} else {
str = __CFCopyFormattingDescription(values[specs[curSpec].mainArgNum].value.pointerValue, formatOptions);
if (NULL == str) {
str = CFCopyDescription(values[specs[curSpec].mainArgNum].value.pointerValue);
}
}
}
if (str) {
CFStringAppend(outputString, str);
CFRelease(str);
} else {
CFStringAppendCString(outputString, "(null description)", kCFStringEncodingASCII);
}
} else {
CFStringAppendCString(outputString, "(null)", kCFStringEncodingASCII);
}
_CFStringFormatAppendMetadata(metadata, numSpecs, values, sizeArgNum, specs[curSpec], oldLength, __CFStringFormatOutputLengthIfNeeded());
break;
}
if (validFormatSpecifiers && specs[curSpec].type != CFFormatLiteralType && specs[curSpec].configDictIndex == -1) {
// We already took care of confirming special strings with config dictionaries;
// this makes other types of specs as validated as well
validatedInnerSpecs += 1;
}
}
for (SInt32 i = 0; i < numSpecsContext; i++) {
CFRange range = CFStringFind(outputString, replacement, 0);
CFStringRef str = contextDescFunc(values[specsContext[i].mainArgNum].value.pointerValue, outputString, replacement, true, NULL);
if (str) {
CFStringReplace(outputString, range, str);
CFRelease(str);
}
}
free(specsContext);
cleanup:
if (numConfigs > 0) va_end(copiedArgs);
if (specs != localSpecsBuffer) CFAllocatorDeallocate(tmpAlloc, specs);
if (values != localValuesBuffer) CFAllocatorDeallocate(tmpAlloc, values);
if (formatChars && (formatChars != localFormatBuffer)) CFAllocatorDeallocate(tmpAlloc, formatChars);
if (configs != localConfigs) CFAllocatorDeallocate(tmpAlloc, configs);
if (formattingConfig != NULL) CFRelease(formattingConfig);
if (metadataStorage && outReplacementMetadata) {
*outReplacementMetadata = CFArrayCreateCopy(kCFAllocatorSystemDefault, metadataStorage);
CFRelease(metadataStorage);
}
return success;
}
#undef SNPRINTF
void CFShowStr(CFStringRef str) {
CFAllocatorRef alloc;
if (!str) {
fprintf(stdout, "(null)\n");
return;
}
if (CF_IS_OBJC(_kCFRuntimeIDCFString, str) || CF_IS_SWIFT(_kCFRuntimeIDCFString, str)) {
fprintf(stdout, "This is an NSString, not CFString\n");
return;
}
alloc = CFGetAllocator(str);
fprintf(stdout, "\nLength %d\nIsEightBit %d\n", (int)__CFStrLength(str), __CFStrIsEightBit(str));
fprintf(stdout, "HasLengthByte %d\nHasNullByte %d\nInlineContents %d\n",
__CFStrHasLengthByte(str), __CFStrHasNullByte(str), __CFStrIsInline(str));
fprintf(stdout, "Allocator ");
if (alloc != kCFAllocatorSystemDefault) {
fprintf(stdout, "%p\n", (void *)alloc);
} else {
fprintf(stdout, "SystemDefault\n");
}
fprintf(stdout, "Mutable %d\n", __CFStrIsMutable(str));
if (!__CFStrIsMutable(str) && __CFStrHasContentsDeallocator(str)) {
if (__CFStrContentsDeallocator(str)) fprintf(stdout, "ContentsDeallocatorFunc %p\n", (void *)__CFStrContentsDeallocator(str));
else fprintf(stdout, "ContentsDeallocatorFunc None\n");
} else if (__CFStrIsMutable(str) && __CFStrHasContentsAllocator(str)) {
fprintf(stdout, "ExternalContentsAllocator %p\n", (void *)__CFStrContentsAllocator((CFMutableStringRef)str));
}
if (__CFStrIsMutable(str)) {
fprintf(stdout, "CurrentCapacity %d\n%sCapacity %d\n", (int)__CFStrCapacity(str), __CFStrIsFixed(str) ? "Fixed" : "Desired", (int)__CFStrDesiredCapacity(str));
}
fprintf(stdout, "Contents %p\n", (void *)__CFStrContents(str));
}
#undef HANGUL_SBASE
#undef HANGUL_LBASE
#undef HANGUL_VBASE
#undef HANGUL_TBASE
#undef HANGUL_SCOUNT
#undef HANGUL_LCOUNT
#undef HANGUL_VCOUNT
#undef HANGUL_TCOUNT
#undef HANGUL_NCOUNT
|