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 7938 7939 7940 7941 7942 7943 7944 7945 7946 7947 7948 7949 7950 7951 7952 7953 7954 7955 7956 7957 7958 7959 7960 7961 7962 7963 7964 7965 7966 7967 7968 7969 7970 7971 7972 7973 7974 7975 7976 7977 7978 7979 7980 7981 7982 7983 7984 7985 7986 7987 7988 7989 7990 7991 7992 7993 7994 7995 7996 7997 7998 7999 8000 8001 8002 8003 8004 8005 8006 8007 8008 8009 8010 8011 8012 8013 8014 8015 8016 8017 8018 8019 8020 8021 8022 8023 8024 8025 8026 8027 8028 8029 8030 8031 8032 8033 8034 8035 8036 8037 8038 8039 8040 8041 8042 8043 8044 8045 8046 8047 8048 8049 8050 8051 8052 8053 8054 8055 8056 8057 8058 8059 8060 8061 8062 8063 8064 8065 8066 8067 8068 8069 8070 8071 8072 8073 8074 8075 8076 8077 8078 8079 8080 8081 8082 8083 8084 8085 8086 8087 8088 8089 8090 8091 8092 8093 8094 8095 8096 8097 8098 8099 8100 8101 8102 8103 8104 8105 8106 8107 8108 8109 8110 8111 8112 8113 8114 8115 8116 8117 8118 8119 8120 8121 8122 8123 8124 8125 8126 8127 8128 8129 8130 8131 8132 8133 8134 8135 8136 8137 8138 8139 8140 8141 8142 8143 8144 8145 8146 8147 8148 8149 8150 8151 8152 8153 8154 8155 8156 8157 8158 8159 8160 8161 8162 8163 8164 8165 8166 8167 8168 8169 8170 8171 8172 8173 8174 8175 8176 8177 8178 8179 8180 8181 8182 8183 8184 8185 8186 8187 8188 8189 8190 8191 8192 8193 8194 8195 8196 8197 8198 8199 8200 8201 8202 8203 8204 8205 8206 8207 8208 8209 8210 8211 8212 8213 8214 8215 8216 8217 8218 8219 8220 8221 8222 8223 8224 8225 8226 8227 8228 8229 8230 8231 8232 8233 8234 8235 8236 8237 8238 8239 8240 8241 8242 8243 8244 8245 8246 8247 8248 8249 8250 8251 8252 8253 8254 8255 8256 8257 8258 8259 8260 8261 8262 8263 8264 8265 8266 8267 8268 8269 8270 8271 8272 8273 8274 8275 8276 8277 8278 8279 8280 8281 8282 8283 8284 8285 8286 8287 8288 8289 8290 8291 8292 8293 8294 8295 8296 8297 8298 8299 8300 8301 8302 8303 8304 8305 8306 8307 8308 8309 8310 8311 8312 8313 8314 8315 8316 8317 8318 8319 8320 8321 8322 8323 8324 8325 8326 8327 8328 8329 8330 8331 8332 8333 8334 8335 8336 8337 8338 8339 8340 8341 8342 8343 8344 8345 8346 8347 8348 8349 8350 8351 8352 8353 8354 8355 8356 8357 8358 8359 8360 8361 8362 8363 8364 8365 8366 8367 8368 8369 8370 8371 8372 8373 8374 8375 8376 8377 8378 8379 8380 8381 8382 8383 8384 8385 8386 8387 8388 8389 8390 8391 8392 8393 8394 8395 8396 8397 8398 8399 8400 8401 8402 8403 8404 8405 8406 8407 8408 8409 8410 8411 8412 8413 8414 8415 8416 8417 8418 8419 8420 8421 8422 8423 8424 8425 8426 8427 8428 8429 8430 8431 8432 8433 8434 8435 8436 8437 8438 8439 8440 8441 8442 8443 8444 8445 8446 8447 8448 8449 8450 8451 8452 8453 8454 8455 8456 8457 8458 8459 8460 8461 8462 8463 8464 8465 8466 8467 8468 8469 8470 8471 8472 8473 8474 8475 8476 8477 8478 8479 8480 8481 8482 8483 8484 8485 8486 8487 8488 8489 8490 8491 8492 8493 8494 8495 8496 8497 8498 8499 8500 8501 8502 8503 8504 8505 8506 8507 8508 8509 8510 8511 8512 8513 8514 8515 8516 8517 8518 8519 8520 8521 8522 8523 8524 8525 8526 8527 8528 8529 8530 8531 8532 8533 8534 8535 8536 8537 8538 8539 8540 8541 8542 8543 8544 8545 8546 8547 8548 8549 8550 8551 8552 8553 8554 8555 8556 8557 8558 8559 8560 8561 8562 8563 8564 8565 8566 8567 8568 8569 8570 8571 8572 8573 8574 8575 8576 8577 8578 8579 8580 8581 8582 8583 8584 8585 8586 8587 8588 8589 8590 8591 8592 8593 8594 8595 8596 8597 8598 8599 8600 8601 8602 8603 8604 8605 8606 8607 8608 8609 8610 8611 8612 8613 8614 8615 8616 8617 8618 8619 8620 8621 8622 8623 8624 8625 8626 8627 8628 8629 8630 8631 8632 8633 8634 8635 8636 8637 8638 8639 8640 8641 8642 8643 8644 8645 8646 8647 8648 8649 8650 8651 8652 8653 8654 8655 8656 8657 8658 8659 8660 8661 8662 8663 8664 8665 8666 8667 8668 8669 8670 8671 8672 8673 8674 8675 8676 8677 8678 8679 8680 8681 8682 8683 8684 8685 8686 8687 8688 8689 8690 8691 8692 8693 8694 8695 8696 8697 8698 8699 8700 8701 8702 8703 8704 8705 8706 8707 8708 8709 8710 8711 8712 8713 8714 8715 8716 8717 8718 8719 8720 8721 8722 8723 8724 8725 8726 8727 8728 8729 8730 8731 8732 8733 8734 8735 8736 8737 8738 8739 8740 8741 8742 8743 8744 8745 8746 8747 8748 8749 8750 8751 8752 8753 8754 8755 8756 8757 8758 8759 8760 8761 8762 8763 8764 8765 8766 8767 8768 8769 8770 8771 8772 8773 8774 8775 8776 8777 8778 8779 8780 8781 8782 8783 8784 8785 8786 8787 8788 8789 8790 8791 8792 8793 8794 8795 8796 8797 8798 8799 8800 8801 8802 8803 8804 8805 8806 8807 8808 8809 8810 8811 8812 8813 8814 8815 8816 8817 8818 8819 8820 8821 8822 8823 8824 8825 8826 8827 8828 8829 8830 8831 8832 8833 8834 8835 8836 8837 8838 8839 8840 8841 8842 8843 8844 8845 8846 8847 8848 8849 8850 8851 8852 8853 8854 8855 8856 8857 8858 8859 8860 8861 8862 8863 8864 8865 8866 8867 8868 8869 8870 8871 8872 8873 8874 8875 8876 8877 8878 8879 8880 8881 8882 8883 8884 8885 8886 8887 8888 8889 8890 8891 8892 8893 8894 8895 8896 8897 8898 8899 8900 8901 8902 8903 8904 8905 8906 8907 8908 8909 8910 8911 8912 8913 8914 8915 8916 8917 8918 8919 8920 8921 8922 8923 8924 8925 8926 8927 8928 8929 8930 8931 8932 8933 8934 8935 8936 8937 8938 8939 8940 8941 8942 8943 8944 8945 8946 8947 8948 8949 8950 8951 8952 8953 8954 8955 8956 8957 8958 8959 8960 8961 8962 8963 8964 8965 8966 8967 8968 8969 8970 8971 8972 8973 8974 8975 8976 8977 8978 8979 8980 8981 8982 8983 8984 8985 8986 8987 8988 8989 8990 8991 8992 8993 8994 8995 8996 8997 8998 8999 9000 9001 9002 9003 9004 9005 9006 9007 9008 9009 9010 9011 9012 9013 9014 9015 9016 9017 9018 9019 9020 9021 9022 9023 9024 9025 9026 9027 9028 9029 9030 9031 9032 9033 9034 9035 9036 9037 9038 9039 9040 9041 9042 9043 9044 9045 9046 9047 9048 9049 9050 9051 9052 9053 9054 9055 9056 9057 9058 9059 9060 9061 9062 9063 9064 9065 9066 9067 9068 9069 9070 9071 9072 9073 9074 9075 9076 9077 9078 9079 9080 9081 9082 9083 9084 9085 9086 9087 9088 9089 9090 9091 9092 9093 9094 9095 9096 9097 9098 9099 9100 9101 9102 9103 9104 9105 9106 9107 9108 9109 9110 9111 9112 9113 9114 9115 9116 9117 9118 9119 9120 9121 9122 9123 9124 9125 9126 9127 9128 9129 9130 9131 9132 9133 9134 9135 9136 9137 9138 9139 9140 9141 9142 9143 9144 9145 9146 9147 9148 9149 9150 9151 9152 9153 9154 9155 9156 9157 9158 9159 9160 9161 9162 9163 9164 9165 9166 9167 9168 9169 9170 9171 9172 9173 9174 9175 9176 9177 9178 9179 9180 9181 9182 9183 9184 9185 9186 9187 9188 9189 9190 9191 9192 9193 9194 9195 9196 9197 9198 9199 9200 9201 9202 9203 9204 9205 9206 9207 9208 9209 9210 9211 9212 9213 9214 9215 9216 9217 9218 9219 9220 9221 9222 9223 9224 9225 9226 9227 9228 9229 9230 9231 9232 9233 9234 9235 9236 9237 9238 9239 9240 9241 9242 9243 9244 9245 9246 9247 9248 9249 9250 9251 9252 9253 9254 9255 9256 9257 9258 9259 9260 9261 9262 9263 9264 9265 9266 9267 9268 9269 9270 9271 9272 9273 9274 9275 9276 9277 9278 9279 9280 9281 9282 9283 9284 9285 9286 9287 9288 9289 9290 9291 9292 9293 9294 9295 9296 9297 9298 9299 9300 9301 9302 9303 9304 9305 9306 9307 9308 9309 9310 9311 9312 9313 9314 9315 9316 9317 9318 9319 9320 9321 9322 9323 9324 9325 9326 9327 9328 9329 9330 9331 9332 9333 9334 9335 9336 9337 9338 9339 9340 9341 9342 9343 9344 9345 9346 9347 9348 9349 9350 9351 9352 9353 9354 9355 9356 9357 9358 9359 9360 9361 9362 9363 9364 9365 9366 9367 9368 9369 9370 9371 9372 9373 9374 9375 9376 9377 9378 9379 9380 9381 9382 9383 9384 9385 9386 9387 9388 9389 9390 9391 9392 9393 9394 9395 9396 9397 9398 9399 9400 9401 9402 9403 9404 9405 9406 9407 9408 9409 9410 9411 9412 9413 9414 9415 9416 9417 9418 9419 9420 9421 9422 9423 9424 9425 9426 9427 9428 9429 9430 9431 9432 9433 9434 9435 9436 9437 9438 9439 9440 9441 9442 9443 9444 9445 9446 9447 9448 9449 9450 9451 9452 9453 9454 9455 9456 9457 9458 9459 9460 9461 9462 9463 9464 9465 9466 9467 9468 9469 9470 9471 9472 9473 9474 9475 9476 9477 9478 9479 9480 9481 9482 9483 9484 9485 9486 9487 9488 9489 9490 9491 9492 9493 9494 9495 9496 9497 9498 9499 9500 9501 9502 9503 9504 9505 9506 9507 9508 9509 9510 9511 9512 9513 9514 9515 9516 9517 9518 9519 9520 9521 9522 9523 9524 9525 9526 9527 9528 9529 9530 9531 9532 9533 9534 9535 9536 9537 9538 9539 9540 9541 9542 9543 9544 9545 9546 9547 9548 9549 9550 9551 9552 9553 9554 9555 9556 9557 9558 9559 9560 9561 9562 9563 9564 9565 9566 9567 9568 9569 9570 9571 9572 9573 9574 9575 9576 9577 9578 9579 9580 9581 9582 9583 9584 9585 9586 9587 9588 9589 9590 9591 9592 9593 9594 9595 9596 9597 9598 9599 9600 9601 9602 9603 9604 9605 9606 9607 9608 9609 9610 9611 9612 9613 9614 9615 9616 9617 9618 9619 9620 9621 9622 9623 9624 9625 9626 9627 9628 9629 9630 9631 9632 9633 9634 9635 9636 9637 9638 9639 9640 9641 9642 9643 9644 9645 9646 9647 9648 9649 9650 9651 9652 9653 9654 9655 9656 9657 9658 9659 9660 9661 9662 9663 9664 9665 9666 9667 9668 9669 9670 9671 9672 9673 9674 9675 9676 9677 9678 9679 9680 9681 9682 9683 9684 9685 9686 9687 9688 9689 9690 9691 9692 9693 9694 9695 9696 9697 9698 9699 9700 9701 9702 9703 9704 9705 9706 9707 9708 9709 9710 9711 9712 9713 9714 9715 9716 9717 9718 9719 9720 9721 9722 9723 9724 9725 9726 9727 9728 9729 9730 9731 9732 9733 9734 9735 9736 9737 9738 9739 9740 9741 9742 9743 9744 9745 9746 9747 9748 9749 9750 9751 9752 9753 9754 9755 9756 9757 9758 9759 9760 9761 9762 9763 9764 9765 9766 9767 9768 9769 9770 9771 9772 9773 9774 9775 9776 9777 9778 9779 9780 9781 9782 9783 9784 9785 9786 9787 9788 9789 9790 9791 9792 9793 9794 9795 9796 9797 9798 9799 9800 9801 9802 9803 9804 9805 9806 9807 9808 9809 9810 9811 9812 9813 9814 9815 9816 9817 9818 9819 9820 9821 9822 9823 9824 9825 9826 9827 9828 9829 9830 9831 9832 9833 9834 9835 9836 9837 9838 9839 9840 9841 9842 9843 9844 9845 9846 9847 9848 9849 9850 9851 9852 9853 9854 9855 9856 9857 9858 9859 9860 9861 9862 9863 9864 9865 9866 9867 9868 9869 9870 9871 9872 9873 9874 9875 9876 9877 9878 9879 9880 9881 9882 9883 9884 9885 9886 9887 9888 9889 9890 9891 9892 9893 9894 9895 9896 9897 9898 9899 9900 9901 9902 9903 9904 9905 9906 9907 9908 9909 9910 9911 9912 9913 9914 9915 9916 9917 9918 9919 9920 9921 9922 9923 9924 9925 9926 9927 9928 9929 9930 9931 9932 9933 9934 9935 9936 9937 9938 9939 9940 9941 9942 9943 9944 9945 9946 9947 9948 9949 9950 9951 9952 9953 9954 9955 9956 9957 9958 9959 9960 9961 9962 9963 9964 9965 9966 9967 9968 9969 9970 9971 9972 9973 9974 9975 9976 9977 9978 9979 9980 9981 9982 9983 9984 9985 9986 9987 9988 9989 9990 9991 9992 9993 9994 9995 9996 9997 9998 9999 10000 10001 10002 10003 10004 10005 10006 10007 10008 10009 10010 10011 10012 10013 10014 10015 10016 10017 10018 10019 10020 10021 10022 10023 10024 10025 10026 10027 10028 10029 10030 10031 10032 10033 10034 10035 10036 10037 10038 10039 10040 10041 10042 10043 10044 10045 10046 10047 10048 10049 10050 10051 10052 10053 10054 10055 10056 10057 10058 10059 10060 10061 10062 10063 10064 10065 10066 10067 10068 10069 10070 10071 10072 10073 10074 10075 10076 10077 10078 10079 10080 10081 10082 10083 10084 10085 10086 10087 10088 10089 10090 10091 10092 10093 10094 10095 10096 10097 10098 10099 10100 10101 10102 10103 10104 10105 10106 10107 10108 10109 10110 10111 10112 10113 10114 10115 10116 10117 10118 10119 10120 10121 10122 10123 10124 10125 10126 10127 10128 10129 10130 10131 10132 10133 10134 10135 10136 10137 10138 10139 10140 10141 10142 10143 10144 10145 10146 10147 10148 10149 10150 10151 10152 10153 10154 10155 10156 10157 10158 10159 10160 10161 10162 10163 10164 10165 10166 10167 10168 10169 10170 10171 10172 10173 10174 10175 10176 10177 10178 10179 10180 10181 10182 10183 10184 10185 10186 10187 10188 10189 10190 10191 10192 10193 10194 10195 10196 10197 10198 10199 10200 10201 10202 10203 10204 10205 10206 10207 10208 10209 10210 10211 10212 10213 10214 10215 10216 10217 10218 10219 10220 10221 10222 10223 10224 10225 10226 10227 10228 10229 10230 10231 10232 10233 10234 10235 10236 10237 10238 10239 10240 10241 10242 10243 10244 10245 10246 10247 10248 10249 10250 10251 10252 10253 10254 10255 10256 10257 10258 10259 10260 10261 10262 10263 10264 10265 10266 10267 10268 10269 10270 10271 10272 10273 10274 10275 10276 10277 10278 10279 10280 10281 10282 10283 10284 10285 10286 10287 10288 10289 10290 10291 10292 10293 10294 10295 10296 10297 10298 10299 10300 10301 10302 10303 10304 10305 10306 10307 10308 10309 10310 10311 10312 10313 10314 10315 10316 10317 10318 10319 10320 10321 10322 10323 10324 10325 10326 10327 10328 10329 10330 10331 10332 10333 10334 10335 10336 10337 10338 10339 10340 10341 10342 10343 10344 10345 10346 10347 10348 10349 10350 10351 10352 10353 10354 10355 10356 10357 10358 10359 10360 10361 10362 10363 10364 10365 10366 10367 10368 10369 10370 10371 10372 10373 10374 10375 10376 10377 10378 10379 10380 10381 10382 10383 10384 10385 10386 10387 10388 10389 10390 10391 10392 10393 10394 10395 10396 10397 10398 10399 10400 10401 10402 10403 10404 10405 10406 10407 10408 10409 10410 10411 10412 10413 10414 10415 10416 10417 10418 10419 10420 10421 10422 10423 10424 10425 10426 10427 10428 10429 10430 10431 10432 10433 10434 10435 10436 10437 10438 10439 10440 10441 10442 10443 10444 10445 10446 10447 10448 10449 10450 10451 10452 10453 10454 10455 10456 10457 10458 10459 10460 10461 10462 10463 10464 10465 10466 10467 10468 10469 10470 10471 10472 10473 10474 10475 10476 10477 10478 10479 10480 10481 10482 10483 10484 10485 10486 10487 10488 10489 10490 10491 10492 10493 10494 10495 10496 10497 10498 10499 10500 10501 10502 10503 10504 10505 10506 10507 10508 10509 10510 10511 10512 10513 10514 10515 10516 10517 10518 10519 10520 10521 10522 10523 10524 10525 10526 10527 10528 10529 10530 10531 10532 10533 10534 10535 10536 10537 10538 10539 10540 10541 10542 10543 10544 10545 10546 10547 10548 10549 10550 10551 10552 10553 10554 10555 10556 10557 10558 10559 10560 10561 10562 10563 10564 10565 10566 10567 10568 10569 10570 10571 10572 10573 10574 10575 10576 10577 10578 10579 10580 10581 10582 10583 10584 10585 10586 10587 10588 10589 10590 10591 10592 10593 10594 10595 10596 10597 10598 10599 10600 10601 10602 10603 10604 10605 10606 10607 10608 10609 10610 10611 10612 10613 10614 10615 10616 10617 10618 10619 10620 10621 10622 10623 10624 10625 10626 10627 10628 10629 10630 10631 10632 10633 10634 10635 10636 10637 10638 10639 10640 10641 10642 10643 10644 10645 10646 10647 10648 10649 10650 10651 10652 10653 10654 10655 10656 10657 10658 10659 10660 10661 10662 10663 10664 10665 10666 10667 10668 10669 10670 10671 10672 10673 10674 10675 10676 10677 10678 10679 10680 10681 10682 10683 10684 10685 10686 10687 10688 10689 10690 10691 10692 10693 10694 10695 10696 10697 10698 10699 10700 10701 10702 10703 10704 10705 10706 10707 10708 10709 10710 10711 10712 10713 10714 10715 10716 10717 10718 10719 10720 10721 10722 10723 10724 10725 10726 10727 10728 10729 10730 10731 10732 10733 10734 10735 10736 10737 10738 10739 10740 10741 10742 10743 10744 10745 10746 10747 10748 10749 10750 10751 10752 10753 10754 10755 10756 10757 10758 10759 10760 10761 10762 10763 10764 10765 10766 10767 10768 10769 10770 10771 10772 10773 10774 10775 10776 10777 10778 10779 10780 10781 10782 10783 10784 10785 10786 10787 10788 10789 10790 10791 10792 10793 10794 10795 10796 10797 10798 10799 10800 10801 10802 10803 10804 10805 10806 10807 10808 10809 10810 10811 10812 10813 10814 10815 10816 10817 10818 10819 10820 10821 10822 10823 10824 10825 10826 10827 10828 10829 10830 10831 10832 10833 10834 10835 10836 10837 10838 10839 10840 10841 10842 10843 10844 10845 10846 10847 10848 10849 10850 10851 10852 10853 10854 10855 10856 10857 10858 10859 10860 10861 10862 10863 10864 10865 10866 10867 10868 10869 10870 10871 10872 10873 10874 10875 10876 10877 10878 10879 10880 10881 10882 10883 10884 10885 10886 10887 10888 10889 10890 10891 10892 10893 10894 10895 10896 10897 10898 10899 10900 10901 10902 10903 10904 10905 10906 10907 10908 10909 10910 10911 10912 10913 10914 10915 10916 10917 10918 10919 10920 10921 10922 10923 10924 10925 10926 10927 10928 10929 10930 10931 10932 10933 10934 10935 10936 10937 10938 10939 10940 10941 10942 10943 10944 10945 10946 10947 10948 10949 10950 10951 10952 10953 10954 10955 10956 10957 10958 10959 10960 10961 10962 10963 10964 10965 10966 10967 10968 10969 10970 10971 10972 10973 10974 10975 10976 10977 10978 10979 10980 10981 10982 10983 10984 10985 10986 10987 10988 10989 10990 10991 10992 10993 10994 10995 10996 10997 10998 10999 11000 11001 11002 11003 11004 11005 11006 11007 11008 11009 11010 11011 11012 11013 11014 11015 11016 11017 11018 11019 11020 11021 11022 11023 11024 11025 11026 11027 11028 11029 11030 11031 11032 11033 11034 11035 11036 11037 11038 11039 11040 11041 11042 11043 11044 11045 11046 11047 11048 11049 11050 11051 11052 11053 11054 11055 11056 11057 11058 11059 11060 11061 11062 11063 11064 11065 11066 11067 11068 11069 11070 11071 11072 11073 11074 11075 11076 11077 11078 11079 11080 11081 11082 11083 11084 11085 11086 11087 11088 11089 11090 11091 11092 11093 11094 11095 11096 11097 11098 11099 11100 11101 11102 11103 11104 11105 11106 11107 11108 11109 11110 11111 11112 11113 11114 11115 11116 11117 11118 11119 11120 11121 11122 11123 11124 11125 11126 11127 11128 11129 11130 11131 11132 11133 11134 11135 11136 11137 11138 11139 11140 11141 11142 11143 11144 11145 11146 11147 11148 11149 11150 11151 11152 11153 11154 11155 11156 11157 11158 11159 11160 11161 11162 11163 11164 11165 11166 11167 11168 11169 11170 11171 11172 11173 11174 11175 11176 11177 11178 11179 11180 11181 11182 11183 11184 11185 11186 11187 11188 11189 11190 11191 11192 11193 11194 11195 11196 11197 11198 11199 11200 11201 11202 11203 11204 11205 11206 11207 11208 11209 11210 11211 11212 11213 11214 11215 11216 11217 11218 11219 11220 11221 11222 11223 11224 11225 11226 11227 11228 11229 11230 11231 11232 11233 11234 11235 11236 11237 11238 11239 11240 11241 11242 11243 11244 11245 11246 11247 11248 11249 11250 11251 11252 11253 11254 11255 11256 11257 11258 11259 11260 11261 11262 11263 11264 11265 11266 11267 11268 11269 11270 11271 11272 11273 11274 11275 11276 11277 11278 11279 11280 11281 11282 11283 11284 11285 11286 11287 11288 11289 11290 11291 11292 11293 11294 11295 11296 11297 11298 11299 11300 11301 11302 11303 11304 11305 11306 11307 11308 11309 11310 11311 11312 11313 11314 11315 11316 11317 11318 11319 11320 11321 11322 11323 11324 11325 11326 11327 11328 11329 11330 11331 11332 11333 11334 11335 11336 11337 11338 11339 11340 11341 11342 11343 11344 11345 11346 11347 11348 11349 11350 11351 11352 11353 11354 11355 11356 11357 11358 11359 11360 11361 11362 11363 11364 11365 11366 11367 11368 11369 11370 11371 11372 11373 11374 11375 11376 11377 11378 11379 11380 11381 11382 11383 11384 11385 11386 11387 11388 11389 11390 11391 11392 11393 11394 11395 11396 11397 11398 11399 11400 11401 11402 11403 11404 11405 11406 11407 11408 11409 11410 11411 11412 11413 11414 11415 11416 11417 11418 11419 11420 11421 11422 11423 11424 11425 11426 11427 11428 11429 11430 11431 11432 11433 11434 11435 11436 11437 11438 11439 11440 11441 11442 11443 11444 11445 11446 11447 11448 11449 11450 11451 11452 11453 11454 11455 11456 11457 11458 11459 11460 11461 11462 11463 11464 11465 11466 11467 11468 11469 11470 11471 11472 11473 11474 11475 11476 11477 11478 11479 11480 11481 11482 11483 11484 11485 11486 11487 11488 11489 11490 11491 11492 11493 11494 11495 11496 11497 11498 11499 11500 11501 11502 11503 11504 11505 11506 11507 11508 11509 11510 11511 11512 11513 11514 11515 11516 11517 11518 11519 11520 11521 11522 11523 11524 11525 11526 11527 11528 11529 11530 11531 11532 11533 11534 11535 11536 11537 11538 11539 11540 11541 11542 11543 11544 11545 11546 11547 11548 11549 11550 11551 11552 11553 11554 11555 11556 11557 11558 11559 11560 11561 11562 11563 11564 11565 11566 11567 11568 11569 11570 11571 11572 11573 11574 11575 11576 11577 11578 11579 11580 11581 11582 11583 11584 11585 11586 11587 11588 11589 11590 11591 11592 11593 11594 11595 11596 11597 11598 11599 11600 11601 11602 11603 11604 11605 11606 11607 11608 11609 11610 11611 11612 11613 11614 11615 11616 11617 11618 11619 11620 11621 11622 11623 11624 11625 11626 11627 11628 11629 11630 11631 11632 11633 11634 11635 11636 11637 11638 11639 11640 11641 11642 11643 11644 11645 11646 11647 11648 11649 11650 11651 11652 11653 11654 11655 11656 11657 11658 11659 11660 11661 11662 11663 11664 11665 11666 11667 11668 11669 11670 11671 11672 11673 11674 11675 11676 11677 11678 11679 11680 11681 11682 11683 11684 11685 11686 11687 11688 11689 11690 11691 11692 11693 11694 11695 11696 11697 11698 11699 11700 11701 11702 11703 11704 11705 11706 11707 11708 11709 11710 11711 11712 11713 11714 11715 11716 11717 11718 11719 11720 11721 11722 11723 11724 11725 11726 11727 11728 11729 11730 11731 11732 11733 11734 11735 11736 11737 11738 11739 11740 11741 11742 11743 11744 11745 11746 11747 11748 11749 11750 11751 11752 11753 11754 11755 11756 11757 11758 11759 11760 11761 11762 11763 11764 11765 11766 11767 11768 11769 11770 11771 11772 11773 11774 11775 11776 11777 11778 11779 11780 11781 11782 11783 11784 11785 11786 11787 11788 11789 11790 11791 11792 11793 11794 11795 11796 11797 11798 11799 11800 11801 11802 11803 11804 11805 11806 11807 11808 11809 11810 11811 11812 11813 11814 11815 11816 11817 11818 11819 11820 11821 11822 11823 11824 11825 11826 11827 11828 11829 11830 11831 11832 11833 11834 11835 11836 11837 11838 11839 11840 11841 11842 11843 11844 11845 11846 11847 11848 11849 11850 11851 11852 11853 11854 11855 11856 11857 11858 11859 11860 11861 11862 11863 11864 11865 11866 11867 11868 11869 11870 11871 11872 11873 11874 11875 11876 11877 11878 11879 11880 11881 11882 11883 11884 11885 11886 11887 11888 11889 11890 11891 11892 11893 11894 11895 11896 11897 11898 11899 11900 11901 11902 11903 11904 11905 11906 11907 11908 11909 11910 11911 11912 11913 11914 11915 11916 11917 11918 11919 11920 11921 11922 11923 11924 11925 11926 11927 11928 11929 11930 11931 11932 11933 11934 11935 11936 11937 11938 11939 11940 11941 11942 11943 11944 11945 11946 11947 11948 11949 11950 11951 11952 11953 11954 11955 11956 11957 11958 11959 11960 11961 11962 11963 11964 11965 11966 11967 11968 11969 11970 11971 11972 11973 11974 11975 11976 11977 11978 11979 11980 11981 11982 11983 11984 11985 11986 11987 11988 11989 11990 11991 11992 11993 11994 11995 11996 11997 11998 11999 12000 12001 12002 12003 12004 12005 12006 12007 12008 12009 12010 12011 12012 12013 12014 12015 12016 12017 12018 12019 12020 12021 12022 12023 12024 12025 12026 12027 12028 12029 12030 12031 12032 12033 12034 12035 12036 12037 12038 12039 12040 12041 12042 12043 12044 12045 12046 12047 12048 12049 12050 12051 12052 12053 12054 12055 12056 12057 12058 12059 12060 12061 12062 12063 12064 12065 12066 12067 12068 12069 12070 12071 12072 12073 12074 12075 12076 12077 12078 12079 12080 12081 12082 12083 12084 12085 12086 12087 12088 12089 12090 12091 12092 12093 12094 12095 12096 12097 12098 12099 12100 12101 12102 12103 12104 12105 12106 12107 12108 12109 12110 12111 12112 12113 12114 12115 12116 12117 12118 12119 12120 12121 12122 12123 12124 12125 12126 12127 12128 12129 12130 12131 12132 12133 12134 12135 12136 12137 12138 12139 12140 12141 12142 12143 12144 12145 12146 12147 12148 12149 12150 12151 12152 12153 12154 12155 12156 12157 12158 12159 12160 12161 12162 12163 12164 12165 12166 12167 12168 12169 12170 12171 12172 12173 12174 12175 12176 12177 12178 12179 12180 12181 12182 12183 12184 12185 12186 12187 12188 12189 12190 12191 12192 12193 12194 12195 12196 12197 12198 12199 12200 12201 12202 12203 12204 12205 12206 12207 12208 12209 12210 12211 12212 12213 12214 12215 12216 12217 12218 12219 12220 12221 12222 12223 12224 12225 12226 12227 12228 12229 12230 12231 12232 12233 12234 12235 12236 12237 12238 12239 12240 12241 12242 12243 12244 12245 12246 12247 12248 12249 12250 12251 12252 12253 12254 12255 12256 12257 12258 12259 12260 12261 12262 12263 12264 12265 12266 12267 12268 12269 12270 12271 12272 12273 12274 12275 12276 12277 12278 12279 12280 12281 12282 12283 12284 12285 12286 12287 12288 12289 12290 12291 12292 12293 12294 12295 12296 12297 12298 12299 12300 12301 12302 12303 12304 12305 12306 12307 12308 12309 12310 12311 12312 12313 12314 12315 12316 12317 12318 12319 12320 12321 12322 12323 12324 12325 12326 12327 12328 12329 12330 12331 12332 12333 12334 12335 12336 12337 12338 12339 12340 12341 12342 12343 12344 12345 12346 12347 12348 12349 12350 12351 12352 12353 12354 12355 12356 12357 12358 12359 12360 12361 12362 12363 12364 12365 12366 12367 12368 12369 12370 12371 12372 12373 12374 12375 12376 12377 12378 12379 12380 12381 12382 12383 12384 12385 12386 12387 12388 12389 12390 12391 12392 12393 12394 12395 12396 12397 12398 12399 12400 12401 12402 12403 12404 12405 12406 12407 12408 12409 12410 12411 12412 12413 12414 12415 12416 12417 12418 12419 12420 12421 12422 12423 12424 12425 12426 12427 12428 12429 12430 12431 12432 12433 12434 12435 12436 12437 12438 12439 12440 12441 12442 12443 12444 12445 12446 12447 12448 12449 12450 12451 12452 12453 12454 12455 12456 12457 12458 12459 12460 12461 12462 12463 12464 12465 12466 12467 12468 12469 12470 12471 12472 12473 12474 12475 12476 12477 12478 12479 12480 12481 12482 12483 12484 12485 12486 12487 12488 12489 12490 12491 12492 12493 12494 12495 12496 12497 12498 12499 12500 12501 12502 12503 12504 12505 12506 12507 12508 12509 12510 12511 12512 12513 12514 12515 12516 12517 12518 12519 12520 12521 12522 12523 12524 12525 12526 12527 12528 12529 12530 12531 12532 12533 12534 12535 12536 12537 12538 12539 12540 12541 12542 12543 12544 12545 12546 12547 12548 12549 12550 12551 12552 12553 12554 12555 12556 12557 12558 12559 12560 12561 12562 12563 12564 12565 12566 12567 12568 12569 12570 12571 12572 12573 12574 12575 12576 12577 12578 12579 12580 12581 12582 12583 12584 12585 12586 12587 12588 12589 12590 12591 12592 12593 12594 12595 12596 12597 12598 12599 12600 12601 12602 12603 12604 12605 12606 12607 12608 12609 12610 12611 12612 12613 12614 12615 12616 12617 12618 12619 12620 12621 12622 12623 12624 12625 12626 12627 12628 12629 12630 12631 12632 12633 12634 12635 12636 12637 12638 12639 12640 12641 12642 12643 12644 12645 12646 12647 12648 12649 12650 12651 12652 12653 12654 12655 12656 12657 12658 12659 12660 12661 12662 12663 12664 12665 12666 12667 12668 12669 12670 12671 12672 12673 12674 12675 12676 12677 12678 12679 12680 12681 12682 12683 12684 12685 12686 12687 12688 12689 12690 12691 12692 12693 12694 12695 12696 12697 12698 12699 12700 12701 12702 12703 12704 12705 12706 12707 12708 12709 12710 12711 12712 12713 12714 12715 12716 12717 12718 12719 12720 12721 12722 12723 12724 12725 12726 12727 12728 12729 12730 12731 12732 12733 12734 12735 12736 12737 12738 12739 12740 12741 12742 12743 12744 12745 12746 12747 12748 12749 12750 12751 12752 12753 12754 12755 12756 12757 12758 12759 12760 12761 12762 12763 12764 12765 12766 12767 12768 12769 12770 12771 12772 12773 12774 12775 12776 12777 12778 12779 12780 12781 12782 12783 12784 12785 12786 12787 12788 12789 12790 12791 12792 12793 12794 12795 12796 12797 12798 12799 12800 12801 12802 12803 12804 12805 12806 12807 12808 12809 12810 12811 12812 12813 12814 12815 12816 12817 12818 12819 12820 12821 12822 12823 12824 12825 12826 12827 12828 12829 12830 12831 12832 12833 12834 12835 12836 12837 12838 12839 12840 12841 12842 12843 12844 12845 12846 12847 12848 12849 12850 12851 12852 12853 12854 12855 12856 12857 12858 12859 12860 12861 12862 12863 12864 12865 12866 12867 12868 12869 12870 12871 12872 12873 12874 12875 12876 12877 12878 12879 12880 12881 12882 12883 12884 12885 12886 12887 12888 12889 12890 12891 12892 12893 12894 12895 12896 12897 12898 12899 12900 12901 12902 12903 12904 12905 12906 12907 12908 12909 12910 12911 12912 12913 12914 12915 12916 12917 12918 12919 12920 12921 12922 12923 12924 12925 12926 12927 12928 12929 12930 12931 12932 12933 12934 12935 12936 12937 12938 12939 12940 12941 12942 12943 12944 12945 12946 12947 12948 12949 12950 12951 12952 12953 12954 12955 12956 12957 12958 12959 12960 12961 12962 12963 12964 12965 12966 12967 12968 12969 12970 12971 12972 12973 12974 12975 12976 12977 12978 12979 12980 12981 12982 12983 12984 12985 12986 12987 12988 12989 12990 12991 12992 12993 12994 12995 12996 12997 12998 12999 13000 13001 13002 13003 13004 13005 13006 13007 13008 13009 13010 13011 13012 13013 13014 13015 13016 13017 13018 13019 13020 13021 13022 13023 13024 13025 13026 13027 13028 13029 13030 13031 13032 13033 13034 13035 13036 13037 13038 13039 13040 13041 13042 13043 13044 13045 13046 13047 13048 13049 13050 13051 13052 13053 13054 13055 13056 13057 13058 13059 13060 13061 13062 13063 13064 13065 13066 13067 13068 13069 13070 13071 13072 13073 13074 13075 13076 13077 13078 13079 13080 13081 13082 13083 13084 13085 13086 13087 13088 13089 13090 13091 13092 13093 13094 13095 13096 13097 13098 13099 13100 13101 13102 13103 13104 13105 13106 13107 13108 13109 13110 13111 13112 13113 13114 13115 13116 13117 13118 13119 13120 13121 13122 13123 13124 13125 13126 13127 13128 13129 13130 13131 13132 13133 13134 13135 13136 13137 13138 13139 13140 13141 13142 13143 13144 13145 13146 13147 13148 13149 13150 13151 13152 13153 13154 13155 13156 13157 13158 13159 13160 13161 13162 13163 13164 13165 13166 13167 13168 13169 13170 13171 13172 13173 13174 13175 13176 13177 13178 13179 13180 13181 13182 13183 13184 13185 13186 13187 13188 13189 13190 13191 13192 13193 13194 13195 13196 13197 13198 13199 13200 13201 13202 13203 13204 13205 13206 13207 13208 13209 13210 13211 13212 13213 13214 13215 13216 13217 13218 13219 13220 13221 13222 13223 13224 13225 13226 13227 13228 13229 13230 13231 13232 13233 13234 13235 13236 13237 13238 13239 13240 13241 13242 13243 13244 13245 13246 13247 13248 13249 13250 13251 13252 13253 13254 13255 13256 13257 13258 13259 13260 13261 13262 13263 13264 13265 13266 13267 13268 13269 13270 13271 13272 13273 13274 13275 13276 13277 13278 13279 13280 13281 13282 13283 13284 13285 13286 13287 13288 13289 13290 13291 13292 13293 13294 13295 13296 13297 13298 13299 13300 13301 13302 13303 13304 13305 13306 13307 13308 13309 13310 13311 13312 13313 13314 13315 13316 13317 13318 13319 13320 13321 13322 13323 13324 13325 13326 13327 13328 13329 13330 13331 13332 13333 13334 13335 13336 13337 13338 13339 13340 13341 13342 13343 13344 13345 13346 13347 13348 13349 13350 13351 13352 13353 13354 13355 13356 13357 13358 13359 13360 13361 13362 13363 13364 13365 13366 13367 13368 13369 13370 13371 13372 13373 13374 13375 13376 13377 13378 13379 13380 13381 13382 13383 13384 13385 13386 13387 13388 13389 13390 13391 13392 13393 13394 13395 13396 13397 13398 13399 13400 13401 13402 13403 13404 13405 13406 13407 13408 13409 13410 13411 13412 13413 13414 13415 13416 13417 13418 13419 13420 13421 13422 13423 13424 13425 13426 13427 13428 13429 13430 13431 13432 13433 13434 13435 13436 13437 13438 13439 13440 13441 13442 13443 13444 13445 13446 13447 13448 13449 13450 13451 13452 13453 13454 13455 13456 13457 13458 13459 13460 13461 13462 13463 13464 13465 13466 13467 13468 13469 13470 13471 13472 13473 13474 13475 13476 13477 13478 13479 13480 13481 13482 13483 13484 13485 13486 13487 13488 13489 13490 13491 13492 13493 13494 13495 13496 13497 13498 13499 13500 13501 13502 13503 13504 13505 13506 13507 13508 13509 13510 13511 13512 13513 13514 13515 13516 13517 13518 13519 13520 13521 13522 13523 13524 13525 13526 13527 13528 13529 13530 13531 13532 13533 13534 13535 13536 13537 13538 13539 13540 13541 13542 13543 13544 13545 13546 13547 13548 13549 13550 13551 13552 13553 13554 13555 13556 13557 13558 13559 13560 13561 13562 13563 13564 13565 13566 13567 13568 13569 13570 13571 13572 13573 13574 13575 13576 13577 13578 13579 13580 13581 13582 13583 13584 13585 13586 13587 13588 13589 13590 13591 13592 13593 13594 13595 13596 13597 13598 13599 13600 13601 13602 13603 13604 13605 13606 13607 13608 13609 13610 13611 13612 13613 13614 13615 13616 13617 13618 13619 13620 13621 13622 13623 13624 13625 13626 13627 13628 13629 13630 13631 13632 13633 13634 13635 13636 13637 13638 13639 13640 13641 13642 13643 13644 13645 13646 13647 13648 13649 13650 13651 13652 13653 13654 13655 13656 13657 13658 13659 13660 13661 13662 13663 13664 13665 13666 13667 13668 13669 13670 13671 13672 13673 13674 13675 13676 13677 13678 13679 13680 13681 13682 13683 13684 13685 13686 13687 13688 13689 13690 13691 13692 13693 13694 13695 13696 13697 13698 13699 13700 13701 13702 13703 13704 13705 13706 13707 13708 13709 13710 13711 13712 13713 13714 13715 13716 13717 13718 13719 13720 13721 13722 13723 13724 13725 13726 13727 13728 13729 13730 13731 13732 13733 13734 13735 13736 13737 13738 13739 13740 13741 13742 13743 13744 13745 13746 13747 13748 13749 13750 13751 13752 13753 13754 13755 13756 13757 13758 13759 13760 13761 13762 13763 13764 13765 13766 13767 13768 13769 13770 13771 13772 13773 13774 13775 13776 13777 13778 13779 13780 13781 13782 13783 13784 13785 13786 13787 13788 13789 13790 13791 13792 13793 13794 13795 13796 13797 13798 13799 13800 13801 13802 13803 13804 13805 13806 13807 13808 13809 13810 13811 13812 13813 13814 13815 13816 13817 13818 13819 13820 13821 13822 13823 13824 13825 13826 13827 13828 13829 13830 13831 13832 13833 13834 13835 13836 13837 13838 13839 13840 13841 13842 13843 13844 13845 13846 13847 13848 13849 13850 13851 13852 13853 13854 13855 13856 13857 13858 13859 13860 13861 13862 13863 13864 13865 13866 13867 13868 13869 13870 13871 13872 13873 13874 13875 13876 13877 13878 13879 13880 13881 13882 13883 13884 13885 13886 13887 13888 13889 13890 13891 13892 13893 13894 13895 13896 13897 13898 13899 13900 13901 13902 13903 13904 13905 13906 13907 13908 13909 13910 13911 13912 13913 13914 13915 13916 13917 13918 13919 13920 13921 13922 13923 13924 13925 13926 13927 13928 13929 13930 13931 13932 13933 13934 13935 13936 13937 13938 13939 13940 13941 13942 13943 13944 13945 13946 13947 13948 13949 13950 13951 13952 13953 13954 13955 13956 13957 13958 13959 13960 13961 13962 13963 13964 13965 13966 13967 13968 13969 13970 13971 13972 13973 13974 13975 13976 13977 13978 13979 13980 13981 13982 13983 13984 13985 13986 13987 13988 13989 13990 13991 13992 13993 13994 13995 13996 13997 13998 13999 14000 14001 14002 14003 14004 14005 14006 14007 14008 14009 14010 14011 14012 14013 14014 14015 14016 14017 14018 14019 14020 14021 14022 14023 14024 14025 14026 14027 14028 14029 14030 14031 14032 14033 14034 14035 14036 14037 14038 14039 14040 14041 14042 14043 14044 14045 14046 14047 14048 14049 14050 14051 14052 14053 14054 14055 14056 14057 14058 14059 14060 14061 14062 14063 14064 14065 14066 14067 14068 14069 14070 14071 14072 14073 14074 14075 14076 14077 14078 14079 14080 14081 14082 14083 14084 14085 14086 14087 14088 14089 14090 14091 14092 14093 14094 14095 14096 14097 14098 14099 14100 14101 14102 14103 14104 14105 14106 14107 14108 14109 14110 14111 14112 14113 14114 14115 14116 14117 14118 14119 14120 14121 14122 14123 14124 14125 14126 14127 14128 14129 14130 14131 14132 14133 14134 14135 14136 14137 14138 14139 14140 14141 14142 14143 14144 14145 14146 14147 14148 14149 14150 14151 14152 14153 14154 14155 14156 14157 14158 14159 14160 14161 14162 14163 14164 14165 14166 14167 14168 14169 14170 14171 14172 14173 14174 14175 14176 14177 14178 14179 14180 14181 14182 14183 14184 14185 14186 14187 14188 14189 14190 14191 14192 14193 14194 14195 14196 14197 14198 14199 14200 14201 14202 14203 14204 14205 14206 14207 14208 14209 14210 14211 14212 14213 14214 14215 14216 14217 14218 14219 14220 14221 14222 14223 14224 14225 14226 14227 14228 14229 14230 14231 14232 14233 14234 14235 14236 14237 14238 14239 14240 14241 14242 14243 14244 14245 14246 14247 14248 14249 14250 14251 14252 14253 14254 14255 14256 14257 14258 14259 14260 14261 14262 14263 14264 14265 14266 14267 14268 14269 14270 14271 14272 14273 14274 14275 14276 14277 14278 14279 14280 14281 14282 14283 14284 14285 14286 14287 14288 14289 14290 14291 14292 14293 14294 14295 14296 14297 14298 14299 14300 14301 14302 14303 14304 14305 14306 14307 14308 14309 14310 14311 14312 14313 14314 14315 14316 14317 14318 14319 14320 14321 14322 14323 14324 14325 14326 14327 14328 14329 14330 14331 14332 14333 14334 14335 14336 14337 14338 14339 14340 14341 14342 14343 14344 14345 14346 14347 14348 14349 14350 14351 14352 14353 14354 14355 14356 14357 14358 14359 14360 14361 14362 14363 14364 14365 14366 14367 14368 14369 14370 14371 14372 14373 14374 14375 14376 14377 14378 14379 14380 14381 14382 14383 14384 14385 14386 14387 14388 14389 14390 14391 14392 14393 14394 14395 14396 14397 14398 14399 14400 14401 14402 14403 14404 14405 14406 14407 14408 14409 14410 14411 14412 14413 14414 14415 14416 14417 14418 14419 14420 14421 14422 14423 14424 14425 14426 14427 14428 14429 14430 14431 14432 14433 14434 14435 14436 14437 14438 14439 14440 14441 14442 14443 14444 14445 14446 14447 14448 14449 14450 14451 14452 14453 14454 14455 14456 14457 14458 14459 14460 14461 14462 14463 14464 14465 14466 14467 14468 14469 14470 14471 14472 14473 14474 14475 14476 14477 14478 14479 14480 14481 14482 14483 14484 14485 14486 14487 14488 14489 14490 14491 14492 14493 14494 14495 14496 14497 14498 14499 14500 14501 14502 14503 14504 14505 14506 14507 14508 14509 14510 14511 14512 14513 14514 14515 14516 14517 14518 14519 14520 14521 14522 14523 14524 14525 14526 14527 14528 14529 14530 14531 14532 14533 14534 14535 14536 14537 14538 14539 14540 14541 14542 14543 14544 14545 14546 14547 14548 14549 14550 14551 14552 14553 14554 14555 14556 14557 14558 14559 14560 14561 14562 14563 14564 14565 14566 14567 14568 14569 14570 14571 14572 14573 14574 14575 14576 14577 14578 14579 14580 14581 14582 14583 14584 14585 14586 14587 14588 14589 14590 14591 14592 14593 14594 14595 14596 14597 14598 14599 14600 14601 14602 14603 14604 14605 14606 14607 14608 14609 14610 14611 14612 14613 14614 14615 14616 14617 14618 14619 14620 14621 14622 14623 14624 14625 14626 14627 14628 14629 14630 14631 14632 14633 14634 14635 14636 14637 14638 14639 14640 14641 14642 14643 14644 14645 14646 14647 14648 14649 14650 14651 14652 14653 14654 14655 14656 14657 14658 14659 14660 14661 14662 14663 14664 14665 14666 14667 14668 14669 14670 14671 14672 14673 14674 14675 14676 14677 14678 14679 14680 14681 14682 14683 14684 14685 14686 14687 14688 14689 14690 14691 14692 14693 14694 14695 14696 14697 14698 14699 14700 14701 14702 14703 14704 14705 14706 14707 14708 14709 14710 14711 14712 14713 14714 14715 14716 14717 14718 14719 14720 14721 14722 14723 14724 14725 14726 14727 14728 14729 14730 14731 14732 14733 14734 14735 14736 14737 14738 14739 14740 14741 14742 14743 14744 14745 14746 14747 14748 14749 14750 14751 14752 14753 14754 14755 14756 14757 14758 14759 14760 14761 14762 14763 14764 14765 14766 14767 14768 14769 14770 14771 14772 14773 14774 14775 14776 14777 14778 14779 14780 14781 14782 14783 14784 14785 14786 14787 14788 14789 14790 14791 14792 14793 14794 14795 14796 14797 14798 14799 14800 14801 14802 14803 14804 14805 14806 14807 14808 14809 14810 14811 14812 14813 14814 14815 14816 14817 14818 14819 14820 14821 14822 14823 14824 14825 14826 14827 14828 14829 14830 14831 14832 14833 14834 14835 14836 14837 14838 14839 14840 14841 14842 14843 14844 14845 14846 14847 14848 14849 14850 14851 14852 14853 14854 14855 14856 14857 14858 14859 14860 14861 14862 14863 14864 14865 14866 14867 14868 14869 14870 14871 14872 14873 14874 14875 14876 14877 14878 14879 14880 14881 14882 14883 14884 14885 14886 14887 14888 14889 14890 14891 14892 14893 14894 14895 14896 14897 14898 14899 14900 14901 14902 14903 14904 14905 14906 14907 14908 14909 14910 14911 14912 14913 14914 14915 14916 14917 14918 14919 14920 14921 14922 14923 14924 14925 14926 14927 14928 14929 14930 14931 14932 14933 14934 14935 14936 14937 14938 14939 14940 14941 14942 14943 14944 14945 14946 14947 14948 14949 14950 14951 14952 14953 14954 14955 14956 14957 14958 14959 14960 14961 14962 14963 14964 14965 14966 14967 14968 14969 14970 14971 14972 14973 14974 14975 14976 14977 14978 14979 14980 14981 14982 14983 14984 14985 14986 14987 14988 14989 14990 14991 14992 14993 14994 14995 14996 14997 14998 14999 15000 15001 15002 15003 15004 15005 15006 15007 15008 15009 15010 15011 15012 15013 15014 15015 15016 15017 15018 15019 15020 15021 15022 15023 15024 15025 15026 15027 15028 15029 15030 15031 15032 15033 15034 15035 15036 15037 15038 15039 15040 15041 15042 15043 15044 15045 15046 15047 15048 15049 15050 15051 15052 15053 15054 15055 15056 15057 15058 15059 15060 15061 15062 15063 15064 15065 15066 15067 15068 15069 15070 15071 15072 15073 15074 15075 15076 15077 15078 15079 15080 15081 15082 15083 15084 15085 15086 15087 15088 15089 15090 15091 15092 15093 15094 15095 15096 15097 15098 15099 15100 15101 15102 15103 15104 15105 15106 15107 15108 15109 15110 15111 15112 15113 15114 15115 15116 15117 15118 15119 15120 15121 15122 15123 15124 15125 15126 15127 15128 15129 15130 15131 15132 15133 15134 15135 15136 15137 15138 15139 15140 15141 15142 15143 15144 15145 15146 15147 15148 15149 15150 15151 15152 15153 15154 15155 15156 15157 15158 15159 15160 15161 15162 15163 15164 15165 15166 15167 15168 15169 15170 15171 15172 15173 15174 15175 15176 15177 15178 15179 15180 15181 15182 15183 15184 15185 15186 15187 15188 15189 15190 15191 15192 15193 15194 15195 15196 15197 15198 15199 15200 15201 15202 15203 15204 15205 15206 15207 15208 15209 15210 15211 15212 15213 15214 15215 15216 15217 15218 15219 15220 15221 15222 15223 15224 15225 15226 15227 15228 15229 15230 15231 15232 15233 15234 15235 15236 15237 15238 15239 15240 15241 15242 15243 15244 15245 15246 15247 15248 15249 15250 15251 15252 15253 15254 15255 15256 15257 15258 15259 15260 15261 15262 15263 15264 15265 15266 15267 15268 15269 15270 15271 15272 15273 15274 15275 15276 15277 15278 15279 15280 15281 15282 15283 15284 15285 15286 15287 15288 15289 15290 15291 15292 15293 15294 15295 15296 15297 15298 15299 15300 15301 15302 15303 15304 15305 15306 15307 15308 15309 15310 15311 15312 15313 15314 15315 15316 15317 15318 15319 15320 15321 15322 15323 15324 15325 15326 15327 15328 15329 15330 15331 15332 15333 15334 15335 15336 15337 15338 15339 15340 15341 15342 15343 15344 15345 15346 15347 15348 15349 15350 15351 15352 15353 15354 15355 15356 15357 15358 15359 15360 15361 15362 15363 15364 15365 15366 15367 15368 15369 15370 15371 15372 15373 15374 15375 15376 15377 15378 15379 15380 15381 15382 15383 15384 15385 15386 15387 15388 15389 15390 15391 15392 15393 15394 15395 15396 15397 15398 15399 15400 15401 15402 15403 15404 15405 15406 15407 15408 15409 15410 15411 15412 15413 15414 15415 15416 15417 15418 15419 15420 15421 15422 15423 15424 15425 15426 15427 15428 15429 15430 15431 15432 15433 15434 15435 15436 15437 15438 15439 15440 15441 15442 15443 15444 15445 15446 15447 15448 15449 15450 15451 15452 15453 15454 15455 15456 15457 15458 15459 15460 15461 15462 15463 15464 15465 15466 15467 15468 15469 15470 15471 15472 15473 15474 15475 15476 15477 15478 15479 15480 15481 15482 15483 15484 15485 15486 15487 15488 15489 15490 15491 15492 15493 15494 15495 15496 15497 15498 15499 15500 15501 15502 15503 15504 15505 15506 15507 15508 15509 15510 15511 15512 15513 15514 15515 15516 15517 15518 15519 15520 15521 15522 15523 15524 15525 15526 15527 15528 15529 15530 15531 15532 15533 15534 15535 15536 15537 15538 15539 15540 15541 15542 15543 15544 15545 15546 15547 15548 15549 15550 15551 15552 15553 15554 15555 15556 15557 15558 15559 15560 15561 15562 15563 15564 15565 15566 15567 15568 15569 15570 15571 15572 15573 15574 15575 15576 15577 15578 15579 15580 15581 15582 15583 15584 15585 15586 15587 15588 15589 15590 15591 15592 15593 15594 15595 15596 15597 15598 15599 15600 15601 15602 15603 15604 15605 15606 15607 15608 15609 15610 15611 15612 15613 15614 15615 15616 15617 15618 15619 15620 15621 15622 15623 15624 15625 15626 15627 15628 15629 15630 15631 15632 15633 15634 15635 15636 15637 15638 15639 15640 15641 15642 15643 15644 15645 15646 15647 15648 15649 15650 15651 15652 15653 15654 15655 15656 15657 15658 15659 15660 15661 15662 15663 15664 15665 15666 15667 15668 15669 15670 15671 15672 15673 15674 15675 15676 15677 15678 15679 15680 15681 15682 15683 15684 15685 15686 15687 15688 15689 15690 15691 15692 15693 15694 15695 15696 15697 15698 15699 15700 15701 15702 15703 15704 15705 15706 15707 15708 15709 15710 15711 15712 15713 15714 15715 15716 15717 15718 15719 15720 15721 15722 15723 15724 15725 15726 15727 15728 15729 15730 15731 15732 15733 15734 15735 15736 15737 15738 15739 15740 15741 15742 15743 15744 15745 15746 15747 15748 15749 15750 15751 15752 15753 15754 15755 15756 15757 15758 15759 15760 15761 15762 15763 15764 15765 15766 15767 15768 15769 15770 15771 15772 15773 15774 15775 15776 15777 15778 15779 15780 15781 15782 15783 15784 15785 15786 15787 15788 15789 15790 15791 15792 15793 15794 15795 15796 15797 15798 15799 15800 15801 15802 15803 15804 15805 15806 15807 15808 15809 15810 15811 15812 15813 15814 15815 15816 15817 15818 15819 15820 15821 15822 15823 15824 15825 15826 15827 15828 15829 15830 15831 15832 15833 15834 15835 15836 15837 15838 15839 15840 15841 15842 15843 15844 15845 15846 15847 15848 15849 15850 15851 15852 15853 15854 15855 15856 15857 15858 15859 15860 15861 15862 15863 15864 15865 15866 15867 15868 15869 15870 15871 15872 15873 15874 15875 15876 15877 15878 15879 15880 15881 15882 15883 15884 15885 15886 15887 15888 15889 15890 15891 15892 15893 15894 15895 15896 15897 15898 15899 15900 15901 15902 15903 15904 15905 15906 15907 15908 15909 15910 15911 15912 15913 15914 15915 15916 15917 15918 15919 15920 15921 15922 15923 15924 15925 15926 15927 15928 15929 15930 15931 15932 15933 15934 15935 15936 15937 15938 15939 15940 15941 15942 15943 15944 15945 15946 15947 15948 15949 15950 15951 15952 15953 15954 15955 15956 15957 15958 15959 15960 15961 15962 15963 15964 15965 15966 15967 15968 15969 15970 15971 15972 15973 15974 15975 15976 15977 15978 15979 15980 15981 15982 15983 15984 15985 15986 15987 15988 15989 15990 15991 15992 15993 15994 15995 15996 15997 15998 15999 16000 16001 16002 16003 16004 16005 16006 16007 16008 16009 16010 16011 16012 16013 16014 16015 16016 16017 16018 16019 16020 16021 16022 16023 16024 16025 16026 16027 16028 16029 16030 16031 16032 16033 16034 16035 16036 16037 16038 16039 16040 16041 16042 16043 16044 16045 16046 16047 16048 16049 16050 16051 16052 16053 16054 16055 16056 16057 16058 16059 16060 16061 16062 16063 16064 16065 16066 16067 16068 16069 16070 16071 16072 16073 16074 16075 16076 16077 16078 16079 16080 16081 16082 16083 16084 16085 16086 16087 16088 16089 16090 16091 16092 16093 16094 16095 16096 16097 16098 16099 16100 16101 16102 16103 16104 16105 16106 16107 16108 16109 16110 16111 16112 16113 16114 16115 16116 16117 16118 16119 16120 16121 16122 16123 16124 16125 16126 16127 16128 16129 16130 16131 16132 16133 16134 16135 16136 16137 16138 16139 16140 16141 16142 16143 16144 16145 16146 16147 16148 16149 16150 16151 16152 16153 16154 16155 16156 16157 16158 16159 16160 16161 16162 16163 16164 16165 16166 16167 16168 16169 16170 16171 16172 16173 16174 16175 16176 16177 16178 16179 16180 16181 16182 16183 16184 16185 16186 16187 16188 16189 16190 16191 16192 16193 16194 16195 16196 16197 16198 16199 16200 16201 16202 16203 16204 16205 16206 16207 16208 16209 16210 16211 16212 16213 16214 16215 16216 16217 16218 16219 16220 16221 16222 16223 16224 16225 16226 16227 16228 16229 16230 16231 16232 16233 16234 16235 16236 16237 16238 16239 16240 16241 16242 16243 16244 16245 16246 16247 16248 16249 16250 16251 16252 16253 16254 16255 16256 16257 16258 16259 16260 16261 16262 16263 16264 16265 16266 16267 16268 16269 16270 16271 16272 16273 16274 16275 16276 16277 16278 16279 16280 16281 16282 16283 16284 16285 16286 16287 16288 16289 16290 16291 16292 16293 16294 16295 16296 16297 16298 16299 16300 16301 16302 16303 16304 16305 16306 16307 16308 16309 16310 16311 16312 16313 16314 16315 16316 16317 16318 16319 16320 16321 16322 16323 16324 16325 16326 16327 16328 16329 16330 16331 16332 16333 16334 16335 16336 16337 16338 16339 16340 16341 16342 16343 16344 16345 16346 16347 16348 16349 16350 16351 16352 16353 16354 16355 16356 16357 16358 16359 16360 16361 16362 16363 16364 16365 16366 16367 16368 16369 16370 16371 16372 16373 16374 16375 16376 16377 16378 16379 16380 16381 16382 16383 16384 16385 16386 16387 16388 16389 16390 16391 16392 16393 16394 16395 16396 16397 16398 16399 16400 16401 16402 16403 16404 16405 16406 16407 16408 16409 16410 16411 16412 16413 16414 16415 16416 16417 16418 16419 16420 16421 16422 16423 16424 16425 16426 16427 16428 16429 16430 16431 16432 16433 16434 16435 16436 16437 16438 16439 16440 16441 16442 16443 16444 16445 16446 16447 16448 16449 16450 16451 16452 16453 16454 16455 16456 16457 16458 16459 16460 16461 16462 16463 16464 16465 16466 16467 16468 16469 16470 16471 16472 16473 16474 16475 16476 16477 16478 16479 16480 16481 16482 16483 16484 16485 16486 16487 16488 16489 16490 16491 16492 16493 16494 16495 16496 16497 16498 16499 16500 16501 16502 16503 16504 16505 16506 16507 16508 16509 16510 16511 16512 16513 16514 16515 16516 16517 16518 16519 16520 16521 16522 16523 16524 16525 16526 16527 16528 16529 16530 16531 16532 16533 16534 16535 16536 16537 16538 16539 16540 16541 16542 16543 16544 16545 16546 16547 16548 16549 16550 16551 16552 16553 16554 16555 16556 16557 16558 16559 16560 16561 16562 16563 16564 16565 16566 16567 16568 16569 16570 16571 16572 16573 16574 16575 16576 16577 16578 16579 16580 16581 16582 16583 16584 16585 16586 16587 16588 16589 16590 16591 16592 16593 16594 16595 16596 16597 16598 16599 16600 16601 16602 16603 16604 16605 16606 16607 16608 16609 16610 16611 16612 16613 16614 16615 16616 16617 16618 16619 16620 16621 16622 16623 16624 16625 16626 16627 16628 16629 16630 16631 16632 16633 16634 16635 16636 16637 16638 16639 16640 16641 16642 16643 16644 16645 16646 16647 16648 16649 16650 16651 16652 16653 16654 16655 16656 16657 16658 16659 16660 16661 16662 16663 16664 16665 16666 16667 16668 16669 16670 16671 16672 16673 16674 16675 16676 16677 16678 16679 16680 16681 16682 16683 16684 16685 16686 16687 16688 16689 16690 16691 16692 16693 16694 16695 16696 16697 16698 16699 16700 16701 16702 16703 16704 16705 16706 16707 16708 16709 16710 16711 16712 16713 16714 16715 16716 16717 16718 16719 16720 16721 16722 16723 16724 16725 16726 16727 16728 16729 16730 16731 16732 16733 16734 16735 16736 16737 16738 16739 16740 16741 16742 16743 16744 16745 16746 16747 16748 16749 16750 16751 16752 16753 16754 16755 16756 16757 16758 16759 16760 16761 16762 16763 16764 16765 16766 16767 16768 16769 16770 16771 16772 16773 16774 16775 16776 16777 16778 16779 16780 16781 16782 16783 16784 16785 16786 16787 16788 16789 16790 16791 16792 16793 16794 16795 16796 16797 16798 16799 16800 16801 16802 16803 16804 16805 16806 16807 16808 16809 16810 16811 16812 16813 16814 16815 16816 16817 16818 16819 16820 16821 16822 16823 16824 16825 16826 16827 16828 16829 16830 16831 16832 16833 16834 16835 16836 16837 16838 16839 16840 16841 16842 16843 16844 16845 16846 16847 16848 16849 16850 16851 16852 16853 16854 16855 16856 16857 16858 16859 16860 16861 16862 16863 16864 16865 16866 16867 16868 16869 16870 16871 16872 16873 16874 16875 16876 16877 16878 16879 16880 16881 16882 16883 16884 16885 16886 16887 16888 16889 16890 16891 16892 16893 16894 16895 16896 16897 16898 16899 16900 16901 16902 16903 16904 16905 16906 16907 16908 16909 16910 16911 16912 16913 16914 16915 16916 16917 16918 16919 16920 16921 16922 16923 16924 16925 16926 16927 16928 16929 16930 16931 16932 16933 16934 16935 16936 16937 16938 16939 16940 16941 16942 16943 16944 16945 16946 16947 16948 16949 16950 16951 16952 16953 16954 16955 16956 16957 16958 16959 16960 16961 16962 16963 16964 16965 16966 16967 16968 16969 16970 16971 16972 16973 16974 16975 16976 16977 16978 16979 16980 16981 16982 16983 16984 16985 16986 16987 16988 16989 16990 16991 16992 16993 16994 16995 16996 16997 16998 16999 17000 17001 17002 17003 17004 17005 17006 17007 17008 17009 17010 17011 17012 17013 17014 17015 17016 17017 17018 17019 17020 17021 17022 17023 17024 17025 17026 17027 17028 17029 17030 17031 17032 17033 17034 17035 17036 17037 17038 17039 17040 17041 17042 17043 17044 17045 17046 17047 17048 17049 17050 17051 17052 17053 17054 17055 17056 17057 17058 17059 17060 17061 17062 17063 17064 17065 17066 17067 17068 17069 17070 17071 17072 17073 17074 17075 17076 17077 17078 17079 17080 17081 17082 17083 17084 17085 17086 17087 17088 17089 17090 17091 17092 17093 17094 17095 17096 17097 17098 17099 17100 17101 17102 17103 17104 17105 17106 17107 17108 17109 17110 17111 17112 17113 17114 17115 17116 17117 17118 17119 17120 17121 17122 17123 17124 17125 17126 17127 17128 17129 17130 17131 17132 17133 17134 17135 17136 17137 17138 17139 17140 17141 17142 17143 17144 17145 17146 17147 17148 17149 17150 17151 17152 17153 17154 17155 17156 17157 17158 17159 17160 17161 17162 17163 17164 17165 17166 17167 17168 17169 17170 17171 17172 17173 17174 17175 17176 17177 17178 17179 17180 17181 17182 17183 17184 17185 17186 17187 17188 17189 17190 17191 17192 17193 17194 17195 17196 17197 17198 17199 17200 17201 17202 17203 17204 17205 17206 17207 17208 17209 17210 17211 17212 17213 17214 17215 17216 17217 17218 17219 17220 17221 17222 17223 17224 17225 17226 17227 17228 17229 17230 17231 17232 17233 17234 17235 17236 17237 17238 17239 17240 17241 17242 17243 17244 17245 17246 17247 17248 17249 17250 17251 17252 17253 17254 17255 17256 17257 17258 17259 17260 17261 17262 17263 17264 17265 17266 17267 17268 17269 17270 17271 17272 17273 17274 17275 17276 17277 17278 17279 17280 17281 17282 17283 17284 17285 17286 17287 17288 17289 17290 17291 17292 17293 17294 17295 17296 17297 17298 17299 17300 17301 17302 17303 17304 17305 17306 17307 17308 17309 17310 17311 17312 17313 17314 17315 17316 17317 17318 17319 17320 17321 17322 17323 17324 17325 17326 17327 17328 17329 17330 17331 17332 17333 17334 17335 17336 17337 17338 17339 17340 17341 17342 17343 17344 17345 17346 17347 17348 17349 17350 17351 17352 17353 17354 17355 17356 17357 17358 17359 17360 17361 17362 17363 17364 17365 17366 17367 17368 17369 17370 17371 17372 17373 17374 17375 17376 17377 17378 17379 17380 17381 17382 17383 17384 17385 17386 17387 17388 17389 17390 17391 17392 17393 17394 17395 17396 17397 17398 17399 17400 17401 17402 17403 17404 17405 17406 17407 17408 17409 17410 17411 17412 17413 17414 17415 17416 17417 17418 17419 17420 17421 17422 17423 17424 17425 17426 17427 17428 17429 17430 17431 17432 17433 17434 17435 17436 17437 17438 17439 17440 17441 17442 17443 17444 17445 17446 17447 17448 17449 17450 17451 17452 17453 17454 17455 17456 17457 17458 17459 17460 17461 17462 17463 17464 17465 17466 17467 17468 17469 17470 17471 17472 17473 17474 17475 17476 17477 17478 17479 17480 17481 17482 17483 17484 17485 17486 17487
|
#include "prism.h"
/**
* The prism version and the serialization format.
*/
const char *
pm_version(void) {
return PRISM_VERSION;
}
/**
* In heredocs, tabs automatically complete up to the next 8 spaces. This is
* defined in CRuby as TAB_WIDTH.
*/
#define PM_TAB_WHITESPACE_SIZE 8
#ifndef PM_DEBUG_LOGGING
/**
* Debugging logging will provide you with additional debugging functions as
* well as automatically replace some functions with their debugging
* counterparts.
*/
#define PM_DEBUG_LOGGING 0
#endif
#if PM_DEBUG_LOGGING
/******************************************************************************/
/* Debugging */
/******************************************************************************/
PRISM_ATTRIBUTE_UNUSED static const char *
debug_context(pm_context_t context) {
switch (context) {
case PM_CONTEXT_BEGIN: return "BEGIN";
case PM_CONTEXT_CLASS: return "CLASS";
case PM_CONTEXT_CASE_IN: return "CASE_IN";
case PM_CONTEXT_CASE_WHEN: return "CASE_WHEN";
case PM_CONTEXT_DEF: return "DEF";
case PM_CONTEXT_DEF_PARAMS: return "DEF_PARAMS";
case PM_CONTEXT_DEFAULT_PARAMS: return "DEFAULT_PARAMS";
case PM_CONTEXT_ENSURE: return "ENSURE";
case PM_CONTEXT_ENSURE_DEF: return "ENSURE_DEF";
case PM_CONTEXT_ELSE: return "ELSE";
case PM_CONTEXT_ELSIF: return "ELSIF";
case PM_CONTEXT_EMBEXPR: return "EMBEXPR";
case PM_CONTEXT_BLOCK_BRACES: return "BLOCK_BRACES";
case PM_CONTEXT_BLOCK_KEYWORDS: return "BLOCK_KEYWORDS";
case PM_CONTEXT_FOR: return "FOR";
case PM_CONTEXT_FOR_INDEX: return "FOR_INDEX";
case PM_CONTEXT_IF: return "IF";
case PM_CONTEXT_MAIN: return "MAIN";
case PM_CONTEXT_MODULE: return "MODULE";
case PM_CONTEXT_PARENS: return "PARENS";
case PM_CONTEXT_POSTEXE: return "POSTEXE";
case PM_CONTEXT_PREDICATE: return "PREDICATE";
case PM_CONTEXT_PREEXE: return "PREEXE";
case PM_CONTEXT_RESCUE: return "RESCUE";
case PM_CONTEXT_RESCUE_ELSE: return "RESCUE_ELSE";
case PM_CONTEXT_RESCUE_ELSE_DEF: return "RESCUE_ELSE_DEF";
case PM_CONTEXT_RESCUE_DEF: return "RESCUE_DEF";
case PM_CONTEXT_SCLASS: return "SCLASS";
case PM_CONTEXT_UNLESS: return "UNLESS";
case PM_CONTEXT_UNTIL: return "UNTIL";
case PM_CONTEXT_WHILE: return "WHILE";
case PM_CONTEXT_LAMBDA_BRACES: return "LAMBDA_BRACES";
case PM_CONTEXT_LAMBDA_DO_END: return "LAMBDA_DO_END";
}
return NULL;
}
PRISM_ATTRIBUTE_UNUSED static void
debug_contexts(pm_parser_t *parser) {
pm_context_node_t *context_node = parser->current_context;
fprintf(stderr, "CONTEXTS: ");
if (context_node != NULL) {
while (context_node != NULL) {
fprintf(stderr, "%s", debug_context(context_node->context));
context_node = context_node->prev;
if (context_node != NULL) {
fprintf(stderr, " <- ");
}
}
} else {
fprintf(stderr, "NONE");
}
fprintf(stderr, "\n");
}
PRISM_ATTRIBUTE_UNUSED static void
debug_node(const pm_parser_t *parser, const pm_node_t *node) {
pm_buffer_t output_buffer = { 0 };
pm_prettyprint(&output_buffer, parser, node);
fprintf(stderr, "%.*s", (int) output_buffer.length, output_buffer.value);
pm_buffer_free(&output_buffer);
}
PRISM_ATTRIBUTE_UNUSED static void
debug_lex_mode(pm_parser_t *parser) {
pm_lex_mode_t *lex_mode = parser->lex_modes.current;
bool first = true;
while (lex_mode != NULL) {
if (first) {
first = false;
} else {
fprintf(stderr, " <- ");
}
switch (lex_mode->mode) {
case PM_LEX_DEFAULT: fprintf(stderr, "DEFAULT"); break;
case PM_LEX_EMBEXPR: fprintf(stderr, "EMBEXPR"); break;
case PM_LEX_EMBVAR: fprintf(stderr, "EMBVAR"); break;
case PM_LEX_HEREDOC: fprintf(stderr, "HEREDOC"); break;
case PM_LEX_LIST: fprintf(stderr, "LIST (terminator=%c, interpolation=%d)", lex_mode->as.list.terminator, lex_mode->as.list.interpolation); break;
case PM_LEX_REGEXP: fprintf(stderr, "REGEXP (terminator=%c)", lex_mode->as.regexp.terminator); break;
case PM_LEX_STRING: fprintf(stderr, "STRING (terminator=%c, interpolation=%d)", lex_mode->as.string.terminator, lex_mode->as.string.interpolation); break;
}
lex_mode = lex_mode->prev;
}
fprintf(stderr, "\n");
}
PRISM_ATTRIBUTE_UNUSED static void
debug_state(pm_parser_t *parser) {
fprintf(stderr, "STATE: ");
bool first = true;
if (parser->lex_state == PM_LEX_STATE_NONE) {
fprintf(stderr, "NONE\n");
return;
}
#define CHECK_STATE(state) \
if (parser->lex_state & state) { \
if (!first) fprintf(stderr, "|"); \
fprintf(stderr, "%s", #state); \
first = false; \
}
CHECK_STATE(PM_LEX_STATE_BEG)
CHECK_STATE(PM_LEX_STATE_END)
CHECK_STATE(PM_LEX_STATE_ENDARG)
CHECK_STATE(PM_LEX_STATE_ENDFN)
CHECK_STATE(PM_LEX_STATE_ARG)
CHECK_STATE(PM_LEX_STATE_CMDARG)
CHECK_STATE(PM_LEX_STATE_MID)
CHECK_STATE(PM_LEX_STATE_FNAME)
CHECK_STATE(PM_LEX_STATE_DOT)
CHECK_STATE(PM_LEX_STATE_CLASS)
CHECK_STATE(PM_LEX_STATE_LABEL)
CHECK_STATE(PM_LEX_STATE_LABELED)
CHECK_STATE(PM_LEX_STATE_FITEM)
#undef CHECK_STATE
fprintf(stderr, "\n");
}
PRISM_ATTRIBUTE_UNUSED static void
debug_token(pm_token_t * token) {
fprintf(stderr, "%s: \"%.*s\"\n", pm_token_type_to_str(token->type), (int) (token->end - token->start), token->start);
}
#endif
// Macros for min/max.
#define MIN(a,b) (((a)<(b))?(a):(b))
#define MAX(a,b) (((a)>(b))?(a):(b))
/******************************************************************************/
/* Lex mode manipulations */
/******************************************************************************/
/**
* Returns the incrementor character that should be used to increment the
* nesting count if one is possible.
*/
static inline uint8_t
lex_mode_incrementor(const uint8_t start) {
switch (start) {
case '(':
case '[':
case '{':
case '<':
return start;
default:
return '\0';
}
}
/**
* Returns the matching character that should be used to terminate a list
* beginning with the given character.
*/
static inline uint8_t
lex_mode_terminator(const uint8_t start) {
switch (start) {
case '(':
return ')';
case '[':
return ']';
case '{':
return '}';
case '<':
return '>';
default:
return start;
}
}
/**
* Push a new lex state onto the stack. If we're still within the pre-allocated
* space of the lex state stack, then we'll just use a new slot. Otherwise we'll
* allocate a new pointer and use that.
*/
static bool
lex_mode_push(pm_parser_t *parser, pm_lex_mode_t lex_mode) {
lex_mode.prev = parser->lex_modes.current;
parser->lex_modes.index++;
if (parser->lex_modes.index > PM_LEX_STACK_SIZE - 1) {
parser->lex_modes.current = (pm_lex_mode_t *) malloc(sizeof(pm_lex_mode_t));
if (parser->lex_modes.current == NULL) return false;
*parser->lex_modes.current = lex_mode;
} else {
parser->lex_modes.stack[parser->lex_modes.index] = lex_mode;
parser->lex_modes.current = &parser->lex_modes.stack[parser->lex_modes.index];
}
return true;
}
/**
* Push on a new list lex mode.
*/
static inline bool
lex_mode_push_list(pm_parser_t *parser, bool interpolation, uint8_t delimiter) {
uint8_t incrementor = lex_mode_incrementor(delimiter);
uint8_t terminator = lex_mode_terminator(delimiter);
pm_lex_mode_t lex_mode = {
.mode = PM_LEX_LIST,
.as.list = {
.nesting = 0,
.interpolation = interpolation,
.incrementor = incrementor,
.terminator = terminator
}
};
// These are the places where we need to split up the content of the list.
// We'll use strpbrk to find the first of these characters.
uint8_t *breakpoints = lex_mode.as.list.breakpoints;
memcpy(breakpoints, "\\ \t\f\r\v\n\0\0\0", sizeof(lex_mode.as.list.breakpoints));
// Now we'll add the terminator to the list of breakpoints.
size_t index = 7;
breakpoints[index++] = terminator;
// If interpolation is allowed, then we're going to check for the #
// character. Otherwise we'll only look for escapes and the terminator.
if (interpolation) {
breakpoints[index++] = '#';
}
// If there is an incrementor, then we'll check for that as well.
if (incrementor != '\0') {
breakpoints[index++] = incrementor;
}
parser->explicit_encoding = NULL;
return lex_mode_push(parser, lex_mode);
}
/**
* Push on a new list lex mode that is only used for compatibility. This is
* called when we're at the end of the file. We want the parser to be able to
* perform its normal error tolerance.
*/
static inline bool
lex_mode_push_list_eof(pm_parser_t *parser) {
return lex_mode_push_list(parser, false, '\0');
}
/**
* Push on a new regexp lex mode.
*/
static inline bool
lex_mode_push_regexp(pm_parser_t *parser, uint8_t incrementor, uint8_t terminator) {
pm_lex_mode_t lex_mode = {
.mode = PM_LEX_REGEXP,
.as.regexp = {
.nesting = 0,
.incrementor = incrementor,
.terminator = terminator
}
};
// These are the places where we need to split up the content of the
// regular expression. We'll use strpbrk to find the first of these
// characters.
uint8_t *breakpoints = lex_mode.as.regexp.breakpoints;
memcpy(breakpoints, "\n\\#\0\0", sizeof(lex_mode.as.regexp.breakpoints));
// First we'll add the terminator.
breakpoints[3] = terminator;
// Next, if there is an incrementor, then we'll check for that as well.
if (incrementor != '\0') {
breakpoints[4] = incrementor;
}
return lex_mode_push(parser, lex_mode);
}
/**
* Push on a new string lex mode.
*/
static inline bool
lex_mode_push_string(pm_parser_t *parser, bool interpolation, bool label_allowed, uint8_t incrementor, uint8_t terminator) {
pm_lex_mode_t lex_mode = {
.mode = PM_LEX_STRING,
.as.string = {
.nesting = 0,
.interpolation = interpolation,
.label_allowed = label_allowed,
.incrementor = incrementor,
.terminator = terminator
}
};
// These are the places where we need to split up the content of the
// string. We'll use strpbrk to find the first of these characters.
uint8_t *breakpoints = lex_mode.as.string.breakpoints;
memcpy(breakpoints, "\n\\\0\0\0", sizeof(lex_mode.as.string.breakpoints));
// Now add in the terminator.
size_t index = 2;
breakpoints[index++] = terminator;
// If interpolation is allowed, then we're going to check for the #
// character. Otherwise we'll only look for escapes and the terminator.
if (interpolation) {
breakpoints[index++] = '#';
}
// If we have an incrementor, then we'll add that in as a breakpoint as
// well.
if (incrementor != '\0') {
breakpoints[index++] = incrementor;
}
parser->explicit_encoding = NULL;
return lex_mode_push(parser, lex_mode);
}
/**
* Push on a new string lex mode that is only used for compatibility. This is
* called when we're at the end of the file. We want the parser to be able to
* perform its normal error tolerance.
*/
static inline bool
lex_mode_push_string_eof(pm_parser_t *parser) {
return lex_mode_push_string(parser, false, false, '\0', '\0');
}
/**
* Pop the current lex state off the stack. If we're within the pre-allocated
* space of the lex state stack, then we'll just decrement the index. Otherwise
* we'll free the current pointer and use the previous pointer.
*/
static void
lex_mode_pop(pm_parser_t *parser) {
if (parser->lex_modes.index == 0) {
parser->lex_modes.current->mode = PM_LEX_DEFAULT;
} else if (parser->lex_modes.index < PM_LEX_STACK_SIZE) {
parser->lex_modes.index--;
parser->lex_modes.current = &parser->lex_modes.stack[parser->lex_modes.index];
} else {
parser->lex_modes.index--;
pm_lex_mode_t *prev = parser->lex_modes.current->prev;
free(parser->lex_modes.current);
parser->lex_modes.current = prev;
}
}
/**
* This is the equivalent of IS_lex_state is CRuby.
*/
static inline bool
lex_state_p(pm_parser_t *parser, pm_lex_state_t state) {
return parser->lex_state & state;
}
typedef enum {
PM_IGNORED_NEWLINE_NONE = 0,
PM_IGNORED_NEWLINE_ALL,
PM_IGNORED_NEWLINE_PATTERN
} pm_ignored_newline_type_t;
static inline pm_ignored_newline_type_t
lex_state_ignored_p(pm_parser_t *parser) {
bool ignored = lex_state_p(parser, PM_LEX_STATE_BEG | PM_LEX_STATE_CLASS | PM_LEX_STATE_FNAME | PM_LEX_STATE_DOT) && !lex_state_p(parser, PM_LEX_STATE_LABELED);
if (ignored) {
return PM_IGNORED_NEWLINE_ALL;
} else if ((parser->lex_state & ~((unsigned int) PM_LEX_STATE_LABEL)) == (PM_LEX_STATE_ARG | PM_LEX_STATE_LABELED)) {
return PM_IGNORED_NEWLINE_PATTERN;
} else {
return PM_IGNORED_NEWLINE_NONE;
}
}
static inline bool
lex_state_beg_p(pm_parser_t *parser) {
return lex_state_p(parser, PM_LEX_STATE_BEG_ANY) || ((parser->lex_state & (PM_LEX_STATE_ARG | PM_LEX_STATE_LABELED)) == (PM_LEX_STATE_ARG | PM_LEX_STATE_LABELED));
}
static inline bool
lex_state_arg_p(pm_parser_t *parser) {
return lex_state_p(parser, PM_LEX_STATE_ARG_ANY);
}
static inline bool
lex_state_spcarg_p(pm_parser_t *parser, bool space_seen) {
if (parser->current.end >= parser->end) {
return false;
}
return lex_state_arg_p(parser) && space_seen && !pm_char_is_whitespace(*parser->current.end);
}
static inline bool
lex_state_end_p(pm_parser_t *parser) {
return lex_state_p(parser, PM_LEX_STATE_END_ANY);
}
/**
* This is the equivalent of IS_AFTER_OPERATOR in CRuby.
*/
static inline bool
lex_state_operator_p(pm_parser_t *parser) {
return lex_state_p(parser, PM_LEX_STATE_FNAME | PM_LEX_STATE_DOT);
}
/**
* Set the state of the lexer. This is defined as a function to be able to put a
* breakpoint in it.
*/
static inline void
lex_state_set(pm_parser_t *parser, pm_lex_state_t state) {
parser->lex_state = state;
}
#if PM_DEBUG_LOGGING
static inline void
debug_lex_state_set(pm_parser_t *parser, pm_lex_state_t state, char const * caller_name, int line_number) {
fprintf(stderr, "Caller: %s:%d\nPrevious: ", caller_name, line_number);
debug_state(parser);
lex_state_set(parser, state);
fprintf(stderr, "Now: ");
debug_state(parser);
fprintf(stderr, "\n");
}
#define lex_state_set(parser, state) debug_lex_state_set(parser, state, __func__, __LINE__)
#endif
/******************************************************************************/
/* Diagnostic-related functions */
/******************************************************************************/
/**
* Append an error to the list of errors on the parser.
*/
static inline void
pm_parser_err(pm_parser_t *parser, const uint8_t *start, const uint8_t *end, pm_diagnostic_id_t diag_id) {
pm_diagnostic_list_append(&parser->error_list, start, end, diag_id);
}
/**
* Append an error to the list of errors on the parser using a format string.
*/
#define PM_PARSER_ERR_FORMAT(parser, start, end, diag_id, ...) pm_diagnostic_list_append_format(&parser->error_list, start, end, diag_id, __VA_ARGS__)
/**
* Append an error to the list of errors on the parser using the location of the
* current token.
*/
static inline void
pm_parser_err_current(pm_parser_t *parser, pm_diagnostic_id_t diag_id) {
pm_parser_err(parser, parser->current.start, parser->current.end, diag_id);
}
/**
* Append an error to the list of errors on the parser using the given location
* using a format string.
*/
#define PM_PARSER_ERR_LOCATION_FORMAT(parser, location, diag_id, ...) pm_diagnostic_list_append_format(&parser->error_list, (location)->start, (location)->end, diag_id, __VA_ARGS__)
/**
* Append an error to the list of errors on the parser using the location of the
* given node.
*/
static inline void
pm_parser_err_node(pm_parser_t *parser, const pm_node_t *node, pm_diagnostic_id_t diag_id) {
pm_parser_err(parser, node->location.start, node->location.end, diag_id);
}
/**
* Append an error to the list of errors on the parser using the location of the
* given node and a format string.
*/
#define PM_PARSER_ERR_NODE_FORMAT(parser, node, diag_id, ...) pm_diagnostic_list_append_format(&parser->error_list, node->location.start, node->location.end, diag_id, __VA_ARGS__)
/**
* Append an error to the list of errors on the parser using the location of the
* previous token.
*/
static inline void
pm_parser_err_previous(pm_parser_t *parser, pm_diagnostic_id_t diag_id) {
pm_parser_err(parser, parser->previous.start, parser->previous.end, diag_id);
}
/**
* Append an error to the list of errors on the parser using the location of the
* given token.
*/
static inline void
pm_parser_err_token(pm_parser_t *parser, const pm_token_t *token, pm_diagnostic_id_t diag_id) {
pm_parser_err(parser, token->start, token->end, diag_id);
}
/**
* Append an error to the list of errors on the parser using the location of the
* given token and a format string.
*/
#define PM_PARSER_ERR_TOKEN_FORMAT(parser, token, diag_id, ...) pm_diagnostic_list_append_format(&parser->error_list, (token).start, (token).end, diag_id, __VA_ARGS__)
/**
* Append a warning to the list of warnings on the parser.
*/
static inline void
pm_parser_warn(pm_parser_t *parser, const uint8_t *start, const uint8_t *end, pm_diagnostic_id_t diag_id) {
if (!parser->suppress_warnings) {
pm_diagnostic_list_append(&parser->warning_list, start, end, diag_id);
}
}
/**
* Append a warning to the list of warnings on the parser using the location of
* the given token.
*/
static inline void
pm_parser_warn_token(pm_parser_t *parser, const pm_token_t *token, pm_diagnostic_id_t diag_id) {
pm_parser_warn(parser, token->start, token->end, diag_id);
}
/******************************************************************************/
/* Node-related functions */
/******************************************************************************/
/**
* Retrieve the constant pool id for the given location.
*/
static inline pm_constant_id_t
pm_parser_constant_id_location(pm_parser_t *parser, const uint8_t *start, const uint8_t *end) {
return pm_constant_pool_insert_shared(&parser->constant_pool, start, (size_t) (end - start));
}
/**
* Retrieve the constant pool id for the given string.
*/
static inline pm_constant_id_t
pm_parser_constant_id_owned(pm_parser_t *parser, const uint8_t *start, size_t length) {
return pm_constant_pool_insert_owned(&parser->constant_pool, start, length);
}
/**
* Retrieve the constant pool id for the given static literal C string.
*/
static inline pm_constant_id_t
pm_parser_constant_id_constant(pm_parser_t *parser, const char *start, size_t length) {
return pm_constant_pool_insert_constant(&parser->constant_pool, (const uint8_t *) start, length);
}
/**
* Retrieve the constant pool id for the given token.
*/
static inline pm_constant_id_t
pm_parser_constant_id_token(pm_parser_t *parser, const pm_token_t *token) {
return pm_parser_constant_id_location(parser, token->start, token->end);
}
/**
* Retrieve the constant pool id for the given token. If the token is not
* provided, then return 0.
*/
static inline pm_constant_id_t
pm_parser_optional_constant_id_token(pm_parser_t *parser, const pm_token_t *token) {
return token->type == PM_TOKEN_NOT_PROVIDED ? 0 : pm_parser_constant_id_token(parser, token);
}
/**
* Check whether or not the given node is value expression.
* If the node is value node, it returns NULL.
* If not, it returns the pointer to the node to be inspected as "void expression".
*/
static pm_node_t*
pm_check_value_expression(pm_node_t *node) {
pm_node_t* void_node = NULL;
while (node != NULL) {
switch (PM_NODE_TYPE(node)) {
case PM_RETURN_NODE:
case PM_BREAK_NODE:
case PM_NEXT_NODE:
case PM_REDO_NODE:
case PM_RETRY_NODE:
case PM_MATCH_REQUIRED_NODE:
return void_node != NULL ? void_node : node;
case PM_MATCH_PREDICATE_NODE:
return NULL;
case PM_BEGIN_NODE: {
pm_begin_node_t *cast = (pm_begin_node_t *) node;
node = (pm_node_t *) cast->statements;
break;
}
case PM_PARENTHESES_NODE: {
pm_parentheses_node_t *cast = (pm_parentheses_node_t *) node;
node = (pm_node_t *) cast->body;
break;
}
case PM_STATEMENTS_NODE: {
pm_statements_node_t *cast = (pm_statements_node_t *) node;
node = cast->body.nodes[cast->body.size - 1];
break;
}
case PM_IF_NODE: {
pm_if_node_t *cast = (pm_if_node_t *) node;
if (cast->statements == NULL || cast->consequent == NULL) {
return NULL;
}
pm_node_t *vn = pm_check_value_expression((pm_node_t *) cast->statements);
if (vn == NULL) {
return NULL;
}
if (void_node == NULL) {
void_node = vn;
}
node = cast->consequent;
break;
}
case PM_UNLESS_NODE: {
pm_unless_node_t *cast = (pm_unless_node_t *) node;
if (cast->statements == NULL || cast->consequent == NULL) {
return NULL;
}
pm_node_t *vn = pm_check_value_expression((pm_node_t *) cast->statements);
if (vn == NULL) {
return NULL;
}
if (void_node == NULL) {
void_node = vn;
}
node = (pm_node_t *) cast->consequent;
break;
}
case PM_ELSE_NODE: {
pm_else_node_t *cast = (pm_else_node_t *) node;
node = (pm_node_t *) cast->statements;
break;
}
case PM_AND_NODE: {
pm_and_node_t *cast = (pm_and_node_t *) node;
node = cast->left;
break;
}
case PM_OR_NODE: {
pm_or_node_t *cast = (pm_or_node_t *) node;
node = cast->left;
break;
}
default:
return NULL;
}
}
return NULL;
}
static inline void
pm_assert_value_expression(pm_parser_t *parser, pm_node_t *node) {
pm_node_t *void_node = pm_check_value_expression(node);
if (void_node != NULL) {
pm_parser_err_node(parser, void_node, PM_ERR_VOID_EXPRESSION);
}
}
/**
* The predicate of conditional nodes can change what would otherwise be regular
* nodes into specialized nodes. For example:
*
* if foo .. bar => RangeNode becomes FlipFlopNode
* if foo and bar .. baz => RangeNode becomes FlipFlopNode
* if /foo/ => RegularExpressionNode becomes MatchLastLineNode
* if /foo #{bar}/ => InterpolatedRegularExpressionNode becomes InterpolatedMatchLastLineNode
*/
static void
pm_conditional_predicate(pm_node_t *node) {
switch (PM_NODE_TYPE(node)) {
case PM_AND_NODE: {
pm_and_node_t *cast = (pm_and_node_t *) node;
pm_conditional_predicate(cast->left);
pm_conditional_predicate(cast->right);
break;
}
case PM_OR_NODE: {
pm_or_node_t *cast = (pm_or_node_t *) node;
pm_conditional_predicate(cast->left);
pm_conditional_predicate(cast->right);
break;
}
case PM_PARENTHESES_NODE: {
pm_parentheses_node_t *cast = (pm_parentheses_node_t *) node;
if ((cast->body != NULL) && PM_NODE_TYPE_P(cast->body, PM_STATEMENTS_NODE)) {
pm_statements_node_t *statements = (pm_statements_node_t *) cast->body;
if (statements->body.size == 1) pm_conditional_predicate(statements->body.nodes[0]);
}
break;
}
case PM_RANGE_NODE: {
pm_range_node_t *cast = (pm_range_node_t *) node;
if (cast->left) {
pm_conditional_predicate(cast->left);
}
if (cast->right) {
pm_conditional_predicate(cast->right);
}
// Here we change the range node into a flip flop node. We can do
// this since the nodes are exactly the same except for the type.
// We're only asserting against the size when we should probably
// assert against the entire layout, but we'll assume tests will
// catch this.
assert(sizeof(pm_range_node_t) == sizeof(pm_flip_flop_node_t));
node->type = PM_FLIP_FLOP_NODE;
break;
}
case PM_REGULAR_EXPRESSION_NODE:
// Here we change the regular expression node into a match last line
// node. We can do this since the nodes are exactly the same except
// for the type.
assert(sizeof(pm_regular_expression_node_t) == sizeof(pm_match_last_line_node_t));
node->type = PM_MATCH_LAST_LINE_NODE;
break;
case PM_INTERPOLATED_REGULAR_EXPRESSION_NODE:
// Here we change the interpolated regular expression node into an
// interpolated match last line node. We can do this since the nodes
// are exactly the same except for the type.
assert(sizeof(pm_interpolated_regular_expression_node_t) == sizeof(pm_interpolated_match_last_line_node_t));
node->type = PM_INTERPOLATED_MATCH_LAST_LINE_NODE;
break;
default:
break;
}
}
/**
* In a lot of places in the tree you can have tokens that are not provided but
* that do not cause an error. For example, in a method call without
* parentheses. In these cases we set the token to the "not provided" type. For
* example:
*
* pm_token_t token = not_provided(parser);
*/
static inline pm_token_t
not_provided(pm_parser_t *parser) {
return (pm_token_t) { .type = PM_TOKEN_NOT_PROVIDED, .start = parser->start, .end = parser->start };
}
#define PM_LOCATION_NULL_VALUE(parser) ((pm_location_t) { .start = parser->start, .end = parser->start })
#define PM_LOCATION_TOKEN_VALUE(token) ((pm_location_t) { .start = (token)->start, .end = (token)->end })
#define PM_LOCATION_NODE_VALUE(node) ((pm_location_t) { .start = (node)->location.start, .end = (node)->location.end })
#define PM_LOCATION_NODE_BASE_VALUE(node) ((pm_location_t) { .start = (node)->base.location.start, .end = (node)->base.location.end })
#define PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE ((pm_location_t) { .start = NULL, .end = NULL })
#define PM_OPTIONAL_LOCATION_TOKEN_VALUE(token) ((token)->type == PM_TOKEN_NOT_PROVIDED ? PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE : PM_LOCATION_TOKEN_VALUE(token))
/**
* This is a special out parameter to the parse_arguments_list function that
* includes opening and closing parentheses in addition to the arguments since
* it's so common. It is handy to use when passing argument information to one
* of the call node creation functions.
*/
typedef struct {
/** The optional location of the opening parenthesis or bracket. */
pm_location_t opening_loc;
/** The lazily-allocated optional arguments node. */
pm_arguments_node_t *arguments;
/** The optional location of the closing parenthesis or bracket. */
pm_location_t closing_loc;
/** The optional block attached to the call. */
pm_node_t *block;
} pm_arguments_t;
/**
* Retrieve the end location of a `pm_arguments_t` object.
*/
static inline const uint8_t *
pm_arguments_end(pm_arguments_t *arguments) {
if (arguments->block != NULL) {
const uint8_t *end = arguments->block->location.end;
if (arguments->closing_loc.start != NULL && arguments->closing_loc.end > end) {
end = arguments->closing_loc.end;
}
return end;
}
if (arguments->closing_loc.start != NULL) {
return arguments->closing_loc.end;
}
if (arguments->arguments != NULL) {
return arguments->arguments->base.location.end;
}
return arguments->closing_loc.end;
}
/**
* Check that we're not about to attempt to attach a brace block to a call that
* has arguments without parentheses.
*/
static void
pm_arguments_validate_block(pm_parser_t *parser, pm_arguments_t *arguments, pm_block_node_t *block) {
// First, check that we have arguments and that we don't have a closing
// location for them.
if (arguments->arguments == NULL || arguments->closing_loc.start != NULL) {
return;
}
// Next, check that we don't have a single parentheses argument. This would
// look like:
//
// foo (1) {}
//
// In this case, it's actually okay for the block to be attached to the
// call, even though it looks like it's attached to the argument.
if (arguments->arguments->arguments.size == 1 && PM_NODE_TYPE_P(arguments->arguments->arguments.nodes[0], PM_PARENTHESES_NODE)) {
return;
}
// If we didn't hit a case before this check, then at this point we need to
// add a syntax error.
pm_parser_err_node(parser, (pm_node_t *) block, PM_ERR_ARGUMENT_UNEXPECTED_BLOCK);
}
/******************************************************************************/
/* Node flag handling functions */
/******************************************************************************/
/**
* Set the given flag on the given node.
*/
static inline void
pm_node_flag_set(pm_node_t *node, pm_node_flags_t flag) {
node->flags |= flag;
}
/**
* Remove the given flag from the given node.
*/
static inline void
pm_node_flag_unset(pm_node_t *node, pm_node_flags_t flag) {
node->flags &= (pm_node_flags_t) ~flag;
}
/******************************************************************************/
/* Node creation functions */
/******************************************************************************/
/**
* Parse the decimal number represented by the range of bytes. returns
* UINT32_MAX if the number fails to parse. This function assumes that the range
* of bytes has already been validated to contain only decimal digits.
*/
static uint32_t
parse_decimal_number(pm_parser_t *parser, const uint8_t *start, const uint8_t *end) {
ptrdiff_t diff = end - start;
assert(diff > 0 && ((unsigned long) diff < SIZE_MAX));
size_t length = (size_t) diff;
char *digits = calloc(length + 1, sizeof(char));
memcpy(digits, start, length);
digits[length] = '\0';
char *endptr;
errno = 0;
unsigned long value = strtoul(digits, &endptr, 10);
if ((digits == endptr) || (*endptr != '\0') || (errno == ERANGE)) {
pm_parser_err(parser, start, end, PM_ERR_INVALID_NUMBER_DECIMAL);
value = UINT32_MAX;
}
free(digits);
if (value > UINT32_MAX) {
pm_parser_err(parser, start, end, PM_ERR_INVALID_NUMBER_DECIMAL);
value = UINT32_MAX;
}
return (uint32_t) value;
}
/**
* When you have an encoding flag on a regular expression, it takes precedence
* over all of the previously set encoding flags. So we need to mask off any
* previously set encoding flags before setting the new one.
*/
#define PM_REGULAR_EXPRESSION_ENCODING_MASK ~(PM_REGULAR_EXPRESSION_FLAGS_EUC_JP | PM_REGULAR_EXPRESSION_FLAGS_ASCII_8BIT | PM_REGULAR_EXPRESSION_FLAGS_WINDOWS_31J | PM_REGULAR_EXPRESSION_FLAGS_UTF_8)
/**
* Parse out the options for a regular expression.
*/
static inline pm_node_flags_t
pm_regular_expression_flags_create(const pm_token_t *closing) {
pm_node_flags_t flags = 0;
if (closing->type == PM_TOKEN_REGEXP_END) {
for (const uint8_t *flag = closing->start + 1; flag < closing->end; flag++) {
switch (*flag) {
case 'i': flags |= PM_REGULAR_EXPRESSION_FLAGS_IGNORE_CASE; break;
case 'm': flags |= PM_REGULAR_EXPRESSION_FLAGS_MULTI_LINE; break;
case 'x': flags |= PM_REGULAR_EXPRESSION_FLAGS_EXTENDED; break;
case 'o': flags |= PM_REGULAR_EXPRESSION_FLAGS_ONCE; break;
case 'e': flags = (pm_node_flags_t) (((pm_node_flags_t) (flags & PM_REGULAR_EXPRESSION_ENCODING_MASK)) | PM_REGULAR_EXPRESSION_FLAGS_EUC_JP); break;
case 'n': flags = (pm_node_flags_t) (((pm_node_flags_t) (flags & PM_REGULAR_EXPRESSION_ENCODING_MASK)) | PM_REGULAR_EXPRESSION_FLAGS_ASCII_8BIT); break;
case 's': flags = (pm_node_flags_t) (((pm_node_flags_t) (flags & PM_REGULAR_EXPRESSION_ENCODING_MASK)) | PM_REGULAR_EXPRESSION_FLAGS_WINDOWS_31J); break;
case 'u': flags = (pm_node_flags_t) (((pm_node_flags_t) (flags & PM_REGULAR_EXPRESSION_ENCODING_MASK)) | PM_REGULAR_EXPRESSION_FLAGS_UTF_8); break;
default: assert(false && "unreachable");
}
}
}
return flags;
}
#undef PM_REGULAR_EXPRESSION_ENCODING_MASK
static pm_statements_node_t *
pm_statements_node_create(pm_parser_t *parser);
static void
pm_statements_node_body_append(pm_statements_node_t *node, pm_node_t *statement);
static size_t
pm_statements_node_body_length(pm_statements_node_t *node);
/**
* This function is here to allow us a place to extend in the future when we
* implement our own arena allocation.
*/
static inline void *
pm_alloc_node(PRISM_ATTRIBUTE_UNUSED pm_parser_t *parser, size_t size) {
void *memory = calloc(1, size);
if (memory == NULL) {
fprintf(stderr, "Failed to allocate %zu bytes\n", size);
abort();
}
return memory;
}
#define PM_ALLOC_NODE(parser, type) (type *) pm_alloc_node(parser, sizeof(type))
/**
* Allocate a new MissingNode node.
*/
static pm_missing_node_t *
pm_missing_node_create(pm_parser_t *parser, const uint8_t *start, const uint8_t *end) {
pm_missing_node_t *node = PM_ALLOC_NODE(parser, pm_missing_node_t);
*node = (pm_missing_node_t) {{ .type = PM_MISSING_NODE, .location = { .start = start, .end = end } }};
return node;
}
/**
* Allocate and initialize a new AliasGlobalVariableNode node.
*/
static pm_alias_global_variable_node_t *
pm_alias_global_variable_node_create(pm_parser_t *parser, const pm_token_t *keyword, pm_node_t *new_name, pm_node_t *old_name) {
assert(keyword->type == PM_TOKEN_KEYWORD_ALIAS);
pm_alias_global_variable_node_t *node = PM_ALLOC_NODE(parser, pm_alias_global_variable_node_t);
*node = (pm_alias_global_variable_node_t) {
{
.type = PM_ALIAS_GLOBAL_VARIABLE_NODE,
.location = {
.start = keyword->start,
.end = old_name->location.end
},
},
.new_name = new_name,
.old_name = old_name,
.keyword_loc = PM_LOCATION_TOKEN_VALUE(keyword)
};
return node;
}
/**
* Allocate and initialize a new AliasMethodNode node.
*/
static pm_alias_method_node_t *
pm_alias_method_node_create(pm_parser_t *parser, const pm_token_t *keyword, pm_node_t *new_name, pm_node_t *old_name) {
assert(keyword->type == PM_TOKEN_KEYWORD_ALIAS);
pm_alias_method_node_t *node = PM_ALLOC_NODE(parser, pm_alias_method_node_t);
*node = (pm_alias_method_node_t) {
{
.type = PM_ALIAS_METHOD_NODE,
.location = {
.start = keyword->start,
.end = old_name->location.end
},
},
.new_name = new_name,
.old_name = old_name,
.keyword_loc = PM_LOCATION_TOKEN_VALUE(keyword)
};
return node;
}
/**
* Allocate a new AlternationPatternNode node.
*/
static pm_alternation_pattern_node_t *
pm_alternation_pattern_node_create(pm_parser_t *parser, pm_node_t *left, pm_node_t *right, const pm_token_t *operator) {
pm_alternation_pattern_node_t *node = PM_ALLOC_NODE(parser, pm_alternation_pattern_node_t);
*node = (pm_alternation_pattern_node_t) {
{
.type = PM_ALTERNATION_PATTERN_NODE,
.location = {
.start = left->location.start,
.end = right->location.end
},
},
.left = left,
.right = right,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator)
};
return node;
}
/**
* Allocate and initialize a new and node.
*/
static pm_and_node_t *
pm_and_node_create(pm_parser_t *parser, pm_node_t *left, const pm_token_t *operator, pm_node_t *right) {
pm_assert_value_expression(parser, left);
pm_and_node_t *node = PM_ALLOC_NODE(parser, pm_and_node_t);
*node = (pm_and_node_t) {
{
.type = PM_AND_NODE,
.location = {
.start = left->location.start,
.end = right->location.end
},
},
.left = left,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.right = right
};
return node;
}
/**
* Allocate an initialize a new arguments node.
*/
static pm_arguments_node_t *
pm_arguments_node_create(pm_parser_t *parser) {
pm_arguments_node_t *node = PM_ALLOC_NODE(parser, pm_arguments_node_t);
*node = (pm_arguments_node_t) {
{
.type = PM_ARGUMENTS_NODE,
.location = PM_LOCATION_NULL_VALUE(parser)
},
.arguments = { 0 }
};
return node;
}
/**
* Return the size of the given arguments node.
*/
static size_t
pm_arguments_node_size(pm_arguments_node_t *node) {
return node->arguments.size;
}
/**
* Append an argument to an arguments node.
*/
static void
pm_arguments_node_arguments_append(pm_arguments_node_t *node, pm_node_t *argument) {
if (pm_arguments_node_size(node) == 0) {
node->base.location.start = argument->location.start;
}
node->base.location.end = argument->location.end;
pm_node_list_append(&node->arguments, argument);
}
/**
* Allocate and initialize a new ArrayNode node.
*/
static pm_array_node_t *
pm_array_node_create(pm_parser_t *parser, const pm_token_t *opening) {
pm_array_node_t *node = PM_ALLOC_NODE(parser, pm_array_node_t);
*node = (pm_array_node_t) {
{
.type = PM_ARRAY_NODE,
.flags = PM_NODE_FLAG_STATIC_LITERAL,
.location = PM_LOCATION_TOKEN_VALUE(opening)
},
.opening_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(opening),
.closing_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(opening),
.elements = { 0 }
};
return node;
}
/**
* Return the size of the given array node.
*/
static inline size_t
pm_array_node_size(pm_array_node_t *node) {
return node->elements.size;
}
/**
* Append an argument to an array node.
*/
static inline void
pm_array_node_elements_append(pm_array_node_t *node, pm_node_t *element) {
if (!node->elements.size && !node->opening_loc.start) {
node->base.location.start = element->location.start;
}
pm_node_list_append(&node->elements, element);
node->base.location.end = element->location.end;
// If the element is not a static literal, then the array is not a static
// literal. Turn that flag off.
if (PM_NODE_TYPE_P(element, PM_ARRAY_NODE) || PM_NODE_TYPE_P(element, PM_HASH_NODE) || PM_NODE_TYPE_P(element, PM_RANGE_NODE) || !PM_NODE_FLAG_P(element, PM_NODE_FLAG_STATIC_LITERAL)) {
pm_node_flag_unset((pm_node_t *)node, PM_NODE_FLAG_STATIC_LITERAL);
}
if (PM_NODE_TYPE_P(element, PM_SPLAT_NODE)) {
pm_node_flag_set((pm_node_t *)node, PM_ARRAY_NODE_FLAGS_CONTAINS_SPLAT);
}
}
/**
* Set the closing token and end location of an array node.
*/
static void
pm_array_node_close_set(pm_array_node_t *node, const pm_token_t *closing) {
assert(closing->type == PM_TOKEN_BRACKET_RIGHT || closing->type == PM_TOKEN_STRING_END || closing->type == PM_TOKEN_MISSING || closing->type == PM_TOKEN_NOT_PROVIDED);
node->base.location.end = closing->end;
node->closing_loc = PM_LOCATION_TOKEN_VALUE(closing);
}
/**
* Allocate and initialize a new array pattern node. The node list given in the
* nodes parameter is guaranteed to have at least two nodes.
*/
static pm_array_pattern_node_t *
pm_array_pattern_node_node_list_create(pm_parser_t *parser, pm_node_list_t *nodes) {
pm_array_pattern_node_t *node = PM_ALLOC_NODE(parser, pm_array_pattern_node_t);
*node = (pm_array_pattern_node_t) {
{
.type = PM_ARRAY_PATTERN_NODE,
.location = {
.start = nodes->nodes[0]->location.start,
.end = nodes->nodes[nodes->size - 1]->location.end
},
},
.constant = NULL,
.rest = NULL,
.requireds = { 0 },
.posts = { 0 },
.opening_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE,
.closing_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE
};
// For now we're going to just copy over each pointer manually. This could be
// much more efficient, as we could instead resize the node list.
bool found_rest = false;
for (size_t index = 0; index < nodes->size; index++) {
pm_node_t *child = nodes->nodes[index];
if (!found_rest && (PM_NODE_TYPE_P(child, PM_SPLAT_NODE) || PM_NODE_TYPE_P(child, PM_IMPLICIT_REST_NODE))) {
node->rest = child;
found_rest = true;
} else if (found_rest) {
pm_node_list_append(&node->posts, child);
} else {
pm_node_list_append(&node->requireds, child);
}
}
return node;
}
/**
* Allocate and initialize a new array pattern node from a single rest node.
*/
static pm_array_pattern_node_t *
pm_array_pattern_node_rest_create(pm_parser_t *parser, pm_node_t *rest) {
pm_array_pattern_node_t *node = PM_ALLOC_NODE(parser, pm_array_pattern_node_t);
*node = (pm_array_pattern_node_t) {
{
.type = PM_ARRAY_PATTERN_NODE,
.location = rest->location,
},
.constant = NULL,
.rest = rest,
.requireds = { 0 },
.posts = { 0 },
.opening_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE,
.closing_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE
};
return node;
}
/**
* Allocate and initialize a new array pattern node from a constant and opening
* and closing tokens.
*/
static pm_array_pattern_node_t *
pm_array_pattern_node_constant_create(pm_parser_t *parser, pm_node_t *constant, const pm_token_t *opening, const pm_token_t *closing) {
pm_array_pattern_node_t *node = PM_ALLOC_NODE(parser, pm_array_pattern_node_t);
*node = (pm_array_pattern_node_t) {
{
.type = PM_ARRAY_PATTERN_NODE,
.location = {
.start = constant->location.start,
.end = closing->end
},
},
.constant = constant,
.rest = NULL,
.opening_loc = PM_LOCATION_TOKEN_VALUE(opening),
.closing_loc = PM_LOCATION_TOKEN_VALUE(closing),
.requireds = { 0 },
.posts = { 0 }
};
return node;
}
/**
* Allocate and initialize a new array pattern node from an opening and closing
* token.
*/
static pm_array_pattern_node_t *
pm_array_pattern_node_empty_create(pm_parser_t *parser, const pm_token_t *opening, const pm_token_t *closing) {
pm_array_pattern_node_t *node = PM_ALLOC_NODE(parser, pm_array_pattern_node_t);
*node = (pm_array_pattern_node_t) {
{
.type = PM_ARRAY_PATTERN_NODE,
.location = {
.start = opening->start,
.end = closing->end
},
},
.constant = NULL,
.rest = NULL,
.opening_loc = PM_LOCATION_TOKEN_VALUE(opening),
.closing_loc = PM_LOCATION_TOKEN_VALUE(closing),
.requireds = { 0 },
.posts = { 0 }
};
return node;
}
static inline void
pm_array_pattern_node_requireds_append(pm_array_pattern_node_t *node, pm_node_t *inner) {
pm_node_list_append(&node->requireds, inner);
}
/**
* Allocate and initialize a new assoc node.
*/
static pm_assoc_node_t *
pm_assoc_node_create(pm_parser_t *parser, pm_node_t *key, const pm_token_t *operator, pm_node_t *value) {
pm_assoc_node_t *node = PM_ALLOC_NODE(parser, pm_assoc_node_t);
const uint8_t *end;
if (value != NULL) {
end = value->location.end;
} else if (operator->type != PM_TOKEN_NOT_PROVIDED) {
end = operator->end;
} else {
end = key->location.end;
}
// If the key and value of this assoc node are both static literals, then
// we can mark this node as a static literal.
pm_node_flags_t flags = 0;
if (value && !PM_NODE_TYPE_P(value, PM_ARRAY_NODE) && !PM_NODE_TYPE_P(value, PM_HASH_NODE) && !PM_NODE_TYPE_P(value, PM_RANGE_NODE)) {
flags = key->flags & value->flags & PM_NODE_FLAG_STATIC_LITERAL;
}
*node = (pm_assoc_node_t) {
{
.type = PM_ASSOC_NODE,
.flags = flags,
.location = {
.start = key->location.start,
.end = end
},
},
.key = key,
.operator_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(operator),
.value = value
};
return node;
}
/**
* Allocate and initialize a new assoc splat node.
*/
static pm_assoc_splat_node_t *
pm_assoc_splat_node_create(pm_parser_t *parser, pm_node_t *value, const pm_token_t *operator) {
assert(operator->type == PM_TOKEN_USTAR_STAR);
pm_assoc_splat_node_t *node = PM_ALLOC_NODE(parser, pm_assoc_splat_node_t);
*node = (pm_assoc_splat_node_t) {
{
.type = PM_ASSOC_SPLAT_NODE,
.location = {
.start = operator->start,
.end = value == NULL ? operator->end : value->location.end
},
},
.value = value,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator)
};
return node;
}
/**
* Allocate a new BackReferenceReadNode node.
*/
static pm_back_reference_read_node_t *
pm_back_reference_read_node_create(pm_parser_t *parser, const pm_token_t *name) {
assert(name->type == PM_TOKEN_BACK_REFERENCE);
pm_back_reference_read_node_t *node = PM_ALLOC_NODE(parser, pm_back_reference_read_node_t);
*node = (pm_back_reference_read_node_t) {
{
.type = PM_BACK_REFERENCE_READ_NODE,
.location = PM_LOCATION_TOKEN_VALUE(name),
},
.name = pm_parser_constant_id_token(parser, name)
};
return node;
}
/**
* Allocate and initialize new a begin node.
*/
static pm_begin_node_t *
pm_begin_node_create(pm_parser_t *parser, const pm_token_t *begin_keyword, pm_statements_node_t *statements) {
pm_begin_node_t *node = PM_ALLOC_NODE(parser, pm_begin_node_t);
*node = (pm_begin_node_t) {
{
.type = PM_BEGIN_NODE,
.location = {
.start = begin_keyword->start,
.end = statements == NULL ? begin_keyword->end : statements->base.location.end
},
},
.begin_keyword_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(begin_keyword),
.statements = statements,
.end_keyword_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE
};
return node;
}
/**
* Set the rescue clause, optionally start, and end location of a begin node.
*/
static void
pm_begin_node_rescue_clause_set(pm_begin_node_t *node, pm_rescue_node_t *rescue_clause) {
// If the begin keyword doesn't exist, we set the start on the begin_node
if (!node->begin_keyword_loc.start) {
node->base.location.start = rescue_clause->base.location.start;
}
node->base.location.end = rescue_clause->base.location.end;
node->rescue_clause = rescue_clause;
}
/**
* Set the else clause and end location of a begin node.
*/
static void
pm_begin_node_else_clause_set(pm_begin_node_t *node, pm_else_node_t *else_clause) {
node->base.location.end = else_clause->base.location.end;
node->else_clause = else_clause;
}
/**
* Set the ensure clause and end location of a begin node.
*/
static void
pm_begin_node_ensure_clause_set(pm_begin_node_t *node, pm_ensure_node_t *ensure_clause) {
node->base.location.end = ensure_clause->base.location.end;
node->ensure_clause = ensure_clause;
}
/**
* Set the end keyword and end location of a begin node.
*/
static void
pm_begin_node_end_keyword_set(pm_begin_node_t *node, const pm_token_t *end_keyword) {
assert(end_keyword->type == PM_TOKEN_KEYWORD_END || end_keyword->type == PM_TOKEN_MISSING);
node->base.location.end = end_keyword->end;
node->end_keyword_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(end_keyword);
}
/**
* Allocate and initialize a new BlockArgumentNode node.
*/
static pm_block_argument_node_t *
pm_block_argument_node_create(pm_parser_t *parser, const pm_token_t *operator, pm_node_t *expression) {
pm_block_argument_node_t *node = PM_ALLOC_NODE(parser, pm_block_argument_node_t);
*node = (pm_block_argument_node_t) {
{
.type = PM_BLOCK_ARGUMENT_NODE,
.location = {
.start = operator->start,
.end = expression == NULL ? operator->end : expression->location.end
},
},
.expression = expression,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator)
};
return node;
}
/**
* Allocate and initialize a new BlockNode node.
*/
static pm_block_node_t *
pm_block_node_create(pm_parser_t *parser, pm_constant_id_list_t *locals, uint32_t locals_body_index, const pm_token_t *opening, pm_node_t *parameters, pm_node_t *body, const pm_token_t *closing) {
pm_block_node_t *node = PM_ALLOC_NODE(parser, pm_block_node_t);
*node = (pm_block_node_t) {
{
.type = PM_BLOCK_NODE,
.location = { .start = opening->start, .end = closing->end },
},
.locals = *locals,
.locals_body_index = locals_body_index,
.parameters = parameters,
.body = body,
.opening_loc = PM_LOCATION_TOKEN_VALUE(opening),
.closing_loc = PM_LOCATION_TOKEN_VALUE(closing)
};
return node;
}
/**
* Allocate and initialize a new BlockParameterNode node.
*/
static pm_block_parameter_node_t *
pm_block_parameter_node_create(pm_parser_t *parser, const pm_token_t *name, const pm_token_t *operator) {
assert(operator->type == PM_TOKEN_NOT_PROVIDED || operator->type == PM_TOKEN_UAMPERSAND || operator->type == PM_TOKEN_AMPERSAND);
pm_block_parameter_node_t *node = PM_ALLOC_NODE(parser, pm_block_parameter_node_t);
*node = (pm_block_parameter_node_t) {
{
.type = PM_BLOCK_PARAMETER_NODE,
.location = {
.start = operator->start,
.end = (name->type == PM_TOKEN_NOT_PROVIDED ? operator->end : name->end)
},
},
.name = pm_parser_optional_constant_id_token(parser, name),
.name_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(name),
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator)
};
return node;
}
/**
* Allocate and initialize a new BlockParametersNode node.
*/
static pm_block_parameters_node_t *
pm_block_parameters_node_create(pm_parser_t *parser, pm_parameters_node_t *parameters, const pm_token_t *opening) {
pm_block_parameters_node_t *node = PM_ALLOC_NODE(parser, pm_block_parameters_node_t);
const uint8_t *start;
if (opening->type != PM_TOKEN_NOT_PROVIDED) {
start = opening->start;
} else if (parameters != NULL) {
start = parameters->base.location.start;
} else {
start = NULL;
}
const uint8_t *end;
if (parameters != NULL) {
end = parameters->base.location.end;
} else if (opening->type != PM_TOKEN_NOT_PROVIDED) {
end = opening->end;
} else {
end = NULL;
}
*node = (pm_block_parameters_node_t) {
{
.type = PM_BLOCK_PARAMETERS_NODE,
.location = {
.start = start,
.end = end
}
},
.parameters = parameters,
.opening_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(opening),
.closing_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE,
.locals = { 0 }
};
return node;
}
/**
* Set the closing location of a BlockParametersNode node.
*/
static void
pm_block_parameters_node_closing_set(pm_block_parameters_node_t *node, const pm_token_t *closing) {
assert(closing->type == PM_TOKEN_PIPE || closing->type == PM_TOKEN_PARENTHESIS_RIGHT || closing->type == PM_TOKEN_MISSING);
node->base.location.end = closing->end;
node->closing_loc = PM_LOCATION_TOKEN_VALUE(closing);
}
/**
* Allocate and initialize a new BlockLocalVariableNode node.
*/
static pm_block_local_variable_node_t *
pm_block_local_variable_node_create(pm_parser_t *parser, const pm_token_t *name) {
assert(name->type == PM_TOKEN_IDENTIFIER || name->type == PM_TOKEN_MISSING);
pm_block_local_variable_node_t *node = PM_ALLOC_NODE(parser, pm_block_local_variable_node_t);
*node = (pm_block_local_variable_node_t) {
{
.type = PM_BLOCK_LOCAL_VARIABLE_NODE,
.location = PM_LOCATION_TOKEN_VALUE(name),
},
.name = pm_parser_constant_id_token(parser, name)
};
return node;
}
/**
* Append a new block-local variable to a BlockParametersNode node.
*/
static void
pm_block_parameters_node_append_local(pm_block_parameters_node_t *node, const pm_block_local_variable_node_t *local) {
pm_node_list_append(&node->locals, (pm_node_t *) local);
if (node->base.location.start == NULL) node->base.location.start = local->base.location.start;
node->base.location.end = local->base.location.end;
}
/**
* Allocate and initialize a new BreakNode node.
*/
static pm_break_node_t *
pm_break_node_create(pm_parser_t *parser, const pm_token_t *keyword, pm_arguments_node_t *arguments) {
assert(keyword->type == PM_TOKEN_KEYWORD_BREAK);
pm_break_node_t *node = PM_ALLOC_NODE(parser, pm_break_node_t);
*node = (pm_break_node_t) {
{
.type = PM_BREAK_NODE,
.location = {
.start = keyword->start,
.end = (arguments == NULL ? keyword->end : arguments->base.location.end)
},
},
.arguments = arguments,
.keyword_loc = PM_LOCATION_TOKEN_VALUE(keyword)
};
return node;
}
/**
* Allocate and initialize a new CallNode node. This sets everything to NULL or
* PM_TOKEN_NOT_PROVIDED as appropriate such that its values can be overridden
* in the various specializations of this function.
*/
static pm_call_node_t *
pm_call_node_create(pm_parser_t *parser) {
pm_call_node_t *node = PM_ALLOC_NODE(parser, pm_call_node_t);
*node = (pm_call_node_t) {
{
.type = PM_CALL_NODE,
.location = PM_LOCATION_NULL_VALUE(parser),
},
.receiver = NULL,
.call_operator_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE,
.message_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE,
.opening_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE,
.arguments = NULL,
.closing_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE,
.block = NULL,
.name = 0
};
return node;
}
/**
* Allocate and initialize a new CallNode node from an aref or an aset
* expression.
*/
static pm_call_node_t *
pm_call_node_aref_create(pm_parser_t *parser, pm_node_t *receiver, pm_arguments_t *arguments) {
pm_assert_value_expression(parser, receiver);
pm_call_node_t *node = pm_call_node_create(parser);
node->base.location.start = receiver->location.start;
node->base.location.end = pm_arguments_end(arguments);
node->receiver = receiver;
node->message_loc.start = arguments->opening_loc.start;
node->message_loc.end = arguments->closing_loc.end;
node->opening_loc = arguments->opening_loc;
node->arguments = arguments->arguments;
node->closing_loc = arguments->closing_loc;
node->block = arguments->block;
node->name = pm_parser_constant_id_constant(parser, "[]", 2);
return node;
}
/**
* Allocate and initialize a new CallNode node from a binary expression.
*/
static pm_call_node_t *
pm_call_node_binary_create(pm_parser_t *parser, pm_node_t *receiver, pm_token_t *operator, pm_node_t *argument) {
pm_assert_value_expression(parser, receiver);
pm_assert_value_expression(parser, argument);
pm_call_node_t *node = pm_call_node_create(parser);
node->base.location.start = MIN(receiver->location.start, argument->location.start);
node->base.location.end = MAX(receiver->location.end, argument->location.end);
node->receiver = receiver;
node->message_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(operator);
pm_arguments_node_t *arguments = pm_arguments_node_create(parser);
pm_arguments_node_arguments_append(arguments, argument);
node->arguments = arguments;
node->name = pm_parser_constant_id_token(parser, operator);
return node;
}
/**
* Allocate and initialize a new CallNode node from a call expression.
*/
static pm_call_node_t *
pm_call_node_call_create(pm_parser_t *parser, pm_node_t *receiver, pm_token_t *operator, pm_token_t *message, pm_arguments_t *arguments) {
pm_assert_value_expression(parser, receiver);
pm_call_node_t *node = pm_call_node_create(parser);
node->base.location.start = receiver->location.start;
const uint8_t *end = pm_arguments_end(arguments);
if (end == NULL) {
end = message->end;
}
node->base.location.end = end;
node->receiver = receiver;
node->call_operator_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(operator);
node->message_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(message);
node->opening_loc = arguments->opening_loc;
node->arguments = arguments->arguments;
node->closing_loc = arguments->closing_loc;
node->block = arguments->block;
if (operator->type == PM_TOKEN_AMPERSAND_DOT) {
pm_node_flag_set((pm_node_t *)node, PM_CALL_NODE_FLAGS_SAFE_NAVIGATION);
}
node->name = pm_parser_constant_id_token(parser, message);
return node;
}
/**
* Allocate and initialize a new CallNode node from a call to a method name
* without a receiver that could not have been a local variable read.
*/
static pm_call_node_t *
pm_call_node_fcall_create(pm_parser_t *parser, pm_token_t *message, pm_arguments_t *arguments) {
pm_call_node_t *node = pm_call_node_create(parser);
node->base.location.start = message->start;
node->base.location.end = pm_arguments_end(arguments);
node->message_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(message);
node->opening_loc = arguments->opening_loc;
node->arguments = arguments->arguments;
node->closing_loc = arguments->closing_loc;
node->block = arguments->block;
node->name = pm_parser_constant_id_token(parser, message);
return node;
}
/**
* Allocate and initialize a new CallNode node from a not expression.
*/
static pm_call_node_t *
pm_call_node_not_create(pm_parser_t *parser, pm_node_t *receiver, pm_token_t *message, pm_arguments_t *arguments) {
pm_assert_value_expression(parser, receiver);
pm_call_node_t *node = pm_call_node_create(parser);
node->base.location.start = message->start;
if (arguments->closing_loc.start != NULL) {
node->base.location.end = arguments->closing_loc.end;
} else {
node->base.location.end = receiver->location.end;
}
node->receiver = receiver;
node->message_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(message);
node->opening_loc = arguments->opening_loc;
node->arguments = arguments->arguments;
node->closing_loc = arguments->closing_loc;
node->name = pm_parser_constant_id_constant(parser, "!", 1);
return node;
}
/**
* Allocate and initialize a new CallNode node from a call shorthand expression.
*/
static pm_call_node_t *
pm_call_node_shorthand_create(pm_parser_t *parser, pm_node_t *receiver, pm_token_t *operator, pm_arguments_t *arguments) {
pm_assert_value_expression(parser, receiver);
pm_call_node_t *node = pm_call_node_create(parser);
node->base.location.start = receiver->location.start;
node->base.location.end = pm_arguments_end(arguments);
node->receiver = receiver;
node->call_operator_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(operator);
node->opening_loc = arguments->opening_loc;
node->arguments = arguments->arguments;
node->closing_loc = arguments->closing_loc;
node->block = arguments->block;
if (operator->type == PM_TOKEN_AMPERSAND_DOT) {
pm_node_flag_set((pm_node_t *)node, PM_CALL_NODE_FLAGS_SAFE_NAVIGATION);
}
node->name = pm_parser_constant_id_constant(parser, "call", 4);
return node;
}
/**
* Allocate and initialize a new CallNode node from a unary operator expression.
*/
static pm_call_node_t *
pm_call_node_unary_create(pm_parser_t *parser, pm_token_t *operator, pm_node_t *receiver, const char *name) {
pm_assert_value_expression(parser, receiver);
pm_call_node_t *node = pm_call_node_create(parser);
node->base.location.start = operator->start;
node->base.location.end = receiver->location.end;
node->receiver = receiver;
node->message_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(operator);
node->name = pm_parser_constant_id_constant(parser, name, strlen(name));
return node;
}
/**
* Allocate and initialize a new CallNode node from a call to a method name
* without a receiver that could also have been a local variable read.
*/
static pm_call_node_t *
pm_call_node_variable_call_create(pm_parser_t *parser, pm_token_t *message) {
pm_call_node_t *node = pm_call_node_create(parser);
node->base.location = PM_LOCATION_TOKEN_VALUE(message);
node->message_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(message);
node->name = pm_parser_constant_id_token(parser, message);
return node;
}
/**
* Returns whether or not this call node is a "vcall" (a call to a method name
* without a receiver that could also have been a local variable read).
*/
static inline bool
pm_call_node_variable_call_p(pm_call_node_t *node) {
return PM_NODE_FLAG_P(node, PM_CALL_NODE_FLAGS_VARIABLE_CALL);
}
/**
* Returns whether or not this call is to the [] method in the index form without a block (as
* opposed to `foo.[]` and `foo[] { }`).
*/
static inline bool
pm_call_node_index_p(pm_call_node_t *node) {
return (
(node->call_operator_loc.start == NULL) &&
(node->message_loc.start != NULL) &&
(node->message_loc.start[0] == '[') &&
(node->message_loc.end[-1] == ']') &&
(node->block == NULL || PM_NODE_TYPE_P(node->block, PM_BLOCK_ARGUMENT_NODE))
);
}
/**
* Returns whether or not this call can be used on the left-hand side of an
* operator assignment.
*/
static inline bool
pm_call_node_writable_p(pm_call_node_t *node) {
return (
(node->message_loc.start != NULL) &&
(node->message_loc.end[-1] != '!') &&
(node->message_loc.end[-1] != '?') &&
(node->opening_loc.start == NULL) &&
(node->arguments == NULL) &&
(node->block == NULL)
);
}
/**
* Initialize the read name by reading the write name and chopping off the '='.
*/
static void
pm_call_write_read_name_init(pm_parser_t *parser, pm_constant_id_t *read_name, pm_constant_id_t *write_name) {
pm_constant_t *write_constant = pm_constant_pool_id_to_constant(&parser->constant_pool, *write_name);
if (write_constant->length > 0) {
size_t length = write_constant->length - 1;
void *memory = malloc(length);
memcpy(memory, write_constant->start, length);
*read_name = pm_constant_pool_insert_owned(&parser->constant_pool, (uint8_t *) memory, length);
} else {
// We can get here if the message was missing because of a syntax error.
*read_name = pm_parser_constant_id_constant(parser, "", 0);
}
}
/**
* Allocate and initialize a new CallAndWriteNode node.
*/
static pm_call_and_write_node_t *
pm_call_and_write_node_create(pm_parser_t *parser, pm_call_node_t *target, const pm_token_t *operator, pm_node_t *value) {
assert(target->block == NULL);
assert(operator->type == PM_TOKEN_AMPERSAND_AMPERSAND_EQUAL);
pm_call_and_write_node_t *node = PM_ALLOC_NODE(parser, pm_call_and_write_node_t);
*node = (pm_call_and_write_node_t) {
{
.type = PM_CALL_AND_WRITE_NODE,
.flags = target->base.flags,
.location = {
.start = target->base.location.start,
.end = value->location.end
}
},
.receiver = target->receiver,
.call_operator_loc = target->call_operator_loc,
.message_loc = target->message_loc,
.read_name = 0,
.write_name = target->name,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value
};
pm_call_write_read_name_init(parser, &node->read_name, &node->write_name);
// Here we're going to free the target, since it is no longer necessary.
// However, we don't want to call `pm_node_destroy` because we want to keep
// around all of its children since we just reused them.
free(target);
return node;
}
/**
* Allocate and initialize a new IndexAndWriteNode node.
*/
static pm_index_and_write_node_t *
pm_index_and_write_node_create(pm_parser_t *parser, pm_call_node_t *target, const pm_token_t *operator, pm_node_t *value) {
assert(operator->type == PM_TOKEN_AMPERSAND_AMPERSAND_EQUAL);
pm_index_and_write_node_t *node = PM_ALLOC_NODE(parser, pm_index_and_write_node_t);
*node = (pm_index_and_write_node_t) {
{
.type = PM_INDEX_AND_WRITE_NODE,
.flags = target->base.flags,
.location = {
.start = target->base.location.start,
.end = value->location.end
}
},
.receiver = target->receiver,
.call_operator_loc = target->call_operator_loc,
.opening_loc = target->opening_loc,
.arguments = target->arguments,
.closing_loc = target->closing_loc,
.block = target->block,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value
};
// Here we're going to free the target, since it is no longer necessary.
// However, we don't want to call `pm_node_destroy` because we want to keep
// around all of its children since we just reused them.
free(target);
return node;
}
/**
* Allocate a new CallOperatorWriteNode node.
*/
static pm_call_operator_write_node_t *
pm_call_operator_write_node_create(pm_parser_t *parser, pm_call_node_t *target, const pm_token_t *operator, pm_node_t *value) {
assert(target->block == NULL);
pm_call_operator_write_node_t *node = PM_ALLOC_NODE(parser, pm_call_operator_write_node_t);
*node = (pm_call_operator_write_node_t) {
{
.type = PM_CALL_OPERATOR_WRITE_NODE,
.flags = target->base.flags,
.location = {
.start = target->base.location.start,
.end = value->location.end
}
},
.receiver = target->receiver,
.call_operator_loc = target->call_operator_loc,
.message_loc = target->message_loc,
.read_name = 0,
.write_name = target->name,
.operator = pm_parser_constant_id_location(parser, operator->start, operator->end - 1),
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value
};
pm_call_write_read_name_init(parser, &node->read_name, &node->write_name);
// Here we're going to free the target, since it is no longer necessary.
// However, we don't want to call `pm_node_destroy` because we want to keep
// around all of its children since we just reused them.
free(target);
return node;
}
/**
* Allocate a new IndexOperatorWriteNode node.
*/
static pm_index_operator_write_node_t *
pm_index_operator_write_node_create(pm_parser_t *parser, pm_call_node_t *target, const pm_token_t *operator, pm_node_t *value) {
pm_index_operator_write_node_t *node = PM_ALLOC_NODE(parser, pm_index_operator_write_node_t);
*node = (pm_index_operator_write_node_t) {
{
.type = PM_INDEX_OPERATOR_WRITE_NODE,
.flags = target->base.flags,
.location = {
.start = target->base.location.start,
.end = value->location.end
}
},
.receiver = target->receiver,
.call_operator_loc = target->call_operator_loc,
.opening_loc = target->opening_loc,
.arguments = target->arguments,
.closing_loc = target->closing_loc,
.block = target->block,
.operator = pm_parser_constant_id_location(parser, operator->start, operator->end - 1),
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value
};
// Here we're going to free the target, since it is no longer necessary.
// However, we don't want to call `pm_node_destroy` because we want to keep
// around all of its children since we just reused them.
free(target);
return node;
}
/**
* Allocate and initialize a new CallOrWriteNode node.
*/
static pm_call_or_write_node_t *
pm_call_or_write_node_create(pm_parser_t *parser, pm_call_node_t *target, const pm_token_t *operator, pm_node_t *value) {
assert(target->block == NULL);
assert(operator->type == PM_TOKEN_PIPE_PIPE_EQUAL);
pm_call_or_write_node_t *node = PM_ALLOC_NODE(parser, pm_call_or_write_node_t);
*node = (pm_call_or_write_node_t) {
{
.type = PM_CALL_OR_WRITE_NODE,
.flags = target->base.flags,
.location = {
.start = target->base.location.start,
.end = value->location.end
}
},
.receiver = target->receiver,
.call_operator_loc = target->call_operator_loc,
.message_loc = target->message_loc,
.read_name = 0,
.write_name = target->name,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value
};
pm_call_write_read_name_init(parser, &node->read_name, &node->write_name);
// Here we're going to free the target, since it is no longer necessary.
// However, we don't want to call `pm_node_destroy` because we want to keep
// around all of its children since we just reused them.
free(target);
return node;
}
/**
* Allocate and initialize a new IndexOrWriteNode node.
*/
static pm_index_or_write_node_t *
pm_index_or_write_node_create(pm_parser_t *parser, pm_call_node_t *target, const pm_token_t *operator, pm_node_t *value) {
assert(operator->type == PM_TOKEN_PIPE_PIPE_EQUAL);
pm_index_or_write_node_t *node = PM_ALLOC_NODE(parser, pm_index_or_write_node_t);
*node = (pm_index_or_write_node_t) {
{
.type = PM_INDEX_OR_WRITE_NODE,
.flags = target->base.flags,
.location = {
.start = target->base.location.start,
.end = value->location.end
}
},
.receiver = target->receiver,
.call_operator_loc = target->call_operator_loc,
.opening_loc = target->opening_loc,
.arguments = target->arguments,
.closing_loc = target->closing_loc,
.block = target->block,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value
};
// Here we're going to free the target, since it is no longer necessary.
// However, we don't want to call `pm_node_destroy` because we want to keep
// around all of its children since we just reused them.
free(target);
return node;
}
/**
* Allocate and initialize a new CallTargetNode node from an existing call
* node.
*/
static pm_call_target_node_t *
pm_call_target_node_create(pm_parser_t *parser, pm_call_node_t *target) {
pm_call_target_node_t *node = PM_ALLOC_NODE(parser, pm_call_target_node_t);
*node = (pm_call_target_node_t) {
{
.type = PM_CALL_TARGET_NODE,
.flags = target->base.flags,
.location = target->base.location
},
.receiver = target->receiver,
.call_operator_loc = target->call_operator_loc,
.name = target->name,
.message_loc = target->message_loc
};
// Here we're going to free the target, since it is no longer necessary.
// However, we don't want to call `pm_node_destroy` because we want to keep
// around all of its children since we just reused them.
free(target);
return node;
}
/**
* Allocate and initialize a new IndexTargetNode node from an existing call
* node.
*/
static pm_index_target_node_t *
pm_index_target_node_create(pm_parser_t *parser, pm_call_node_t *target) {
pm_index_target_node_t *node = PM_ALLOC_NODE(parser, pm_index_target_node_t);
*node = (pm_index_target_node_t) {
{
.type = PM_INDEX_TARGET_NODE,
.flags = target->base.flags,
.location = target->base.location
},
.receiver = target->receiver,
.opening_loc = target->opening_loc,
.arguments = target->arguments,
.closing_loc = target->closing_loc,
.block = target->block
};
// Here we're going to free the target, since it is no longer necessary.
// However, we don't want to call `pm_node_destroy` because we want to keep
// around all of its children since we just reused them.
free(target);
return node;
}
/**
* Allocate and initialize a new CapturePatternNode node.
*/
static pm_capture_pattern_node_t *
pm_capture_pattern_node_create(pm_parser_t *parser, pm_node_t *value, pm_node_t *target, const pm_token_t *operator) {
pm_capture_pattern_node_t *node = PM_ALLOC_NODE(parser, pm_capture_pattern_node_t);
*node = (pm_capture_pattern_node_t) {
{
.type = PM_CAPTURE_PATTERN_NODE,
.location = {
.start = value->location.start,
.end = target->location.end
},
},
.value = value,
.target = target,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator)
};
return node;
}
/**
* Allocate and initialize a new CaseNode node.
*/
static pm_case_node_t *
pm_case_node_create(pm_parser_t *parser, const pm_token_t *case_keyword, pm_node_t *predicate, const pm_token_t *end_keyword) {
pm_case_node_t *node = PM_ALLOC_NODE(parser, pm_case_node_t);
*node = (pm_case_node_t) {
{
.type = PM_CASE_NODE,
.location = {
.start = case_keyword->start,
.end = end_keyword->end
},
},
.predicate = predicate,
.consequent = NULL,
.case_keyword_loc = PM_LOCATION_TOKEN_VALUE(case_keyword),
.end_keyword_loc = PM_LOCATION_TOKEN_VALUE(end_keyword),
.conditions = { 0 }
};
return node;
}
/**
* Append a new condition to a CaseNode node.
*/
static void
pm_case_node_condition_append(pm_case_node_t *node, pm_node_t *condition) {
assert(PM_NODE_TYPE_P(condition, PM_WHEN_NODE));
pm_node_list_append(&node->conditions, condition);
node->base.location.end = condition->location.end;
}
/**
* Set the consequent of a CaseNode node.
*/
static void
pm_case_node_consequent_set(pm_case_node_t *node, pm_else_node_t *consequent) {
node->consequent = consequent;
node->base.location.end = consequent->base.location.end;
}
/**
* Set the end location for a CaseNode node.
*/
static void
pm_case_node_end_keyword_loc_set(pm_case_node_t *node, const pm_token_t *end_keyword) {
node->base.location.end = end_keyword->end;
node->end_keyword_loc = PM_LOCATION_TOKEN_VALUE(end_keyword);
}
/**
* Allocate and initialize a new CaseMatchNode node.
*/
static pm_case_match_node_t *
pm_case_match_node_create(pm_parser_t *parser, const pm_token_t *case_keyword, pm_node_t *predicate, const pm_token_t *end_keyword) {
pm_case_match_node_t *node = PM_ALLOC_NODE(parser, pm_case_match_node_t);
*node = (pm_case_match_node_t) {
{
.type = PM_CASE_MATCH_NODE,
.location = {
.start = case_keyword->start,
.end = end_keyword->end
},
},
.predicate = predicate,
.consequent = NULL,
.case_keyword_loc = PM_LOCATION_TOKEN_VALUE(case_keyword),
.end_keyword_loc = PM_LOCATION_TOKEN_VALUE(end_keyword),
.conditions = { 0 }
};
return node;
}
/**
* Append a new condition to a CaseMatchNode node.
*/
static void
pm_case_match_node_condition_append(pm_case_match_node_t *node, pm_node_t *condition) {
assert(PM_NODE_TYPE_P(condition, PM_IN_NODE));
pm_node_list_append(&node->conditions, condition);
node->base.location.end = condition->location.end;
}
/**
* Set the consequent of a CaseMatchNode node.
*/
static void
pm_case_match_node_consequent_set(pm_case_match_node_t *node, pm_else_node_t *consequent) {
node->consequent = consequent;
node->base.location.end = consequent->base.location.end;
}
/**
* Set the end location for a CaseMatchNode node.
*/
static void
pm_case_match_node_end_keyword_loc_set(pm_case_match_node_t *node, const pm_token_t *end_keyword) {
node->base.location.end = end_keyword->end;
node->end_keyword_loc = PM_LOCATION_TOKEN_VALUE(end_keyword);
}
/**
* Allocate a new ClassNode node.
*/
static pm_class_node_t *
pm_class_node_create(pm_parser_t *parser, pm_constant_id_list_t *locals, const pm_token_t *class_keyword, pm_node_t *constant_path, const pm_token_t *name, const pm_token_t *inheritance_operator, pm_node_t *superclass, pm_node_t *body, const pm_token_t *end_keyword) {
pm_class_node_t *node = PM_ALLOC_NODE(parser, pm_class_node_t);
*node = (pm_class_node_t) {
{
.type = PM_CLASS_NODE,
.location = { .start = class_keyword->start, .end = end_keyword->end },
},
.locals = *locals,
.class_keyword_loc = PM_LOCATION_TOKEN_VALUE(class_keyword),
.constant_path = constant_path,
.inheritance_operator_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(inheritance_operator),
.superclass = superclass,
.body = body,
.end_keyword_loc = PM_LOCATION_TOKEN_VALUE(end_keyword),
.name = pm_parser_constant_id_token(parser, name)
};
return node;
}
/**
* Allocate and initialize a new ClassVariableAndWriteNode node.
*/
static pm_class_variable_and_write_node_t *
pm_class_variable_and_write_node_create(pm_parser_t *parser, pm_class_variable_read_node_t *target, const pm_token_t *operator, pm_node_t *value) {
assert(operator->type == PM_TOKEN_AMPERSAND_AMPERSAND_EQUAL);
pm_class_variable_and_write_node_t *node = PM_ALLOC_NODE(parser, pm_class_variable_and_write_node_t);
*node = (pm_class_variable_and_write_node_t) {
{
.type = PM_CLASS_VARIABLE_AND_WRITE_NODE,
.location = {
.start = target->base.location.start,
.end = value->location.end
}
},
.name = target->name,
.name_loc = target->base.location,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value
};
return node;
}
/**
* Allocate and initialize a new ClassVariableOperatorWriteNode node.
*/
static pm_class_variable_operator_write_node_t *
pm_class_variable_operator_write_node_create(pm_parser_t *parser, pm_class_variable_read_node_t *target, const pm_token_t *operator, pm_node_t *value) {
pm_class_variable_operator_write_node_t *node = PM_ALLOC_NODE(parser, pm_class_variable_operator_write_node_t);
*node = (pm_class_variable_operator_write_node_t) {
{
.type = PM_CLASS_VARIABLE_OPERATOR_WRITE_NODE,
.location = {
.start = target->base.location.start,
.end = value->location.end
}
},
.name = target->name,
.name_loc = target->base.location,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value,
.operator = pm_parser_constant_id_location(parser, operator->start, operator->end - 1)
};
return node;
}
/**
* Allocate and initialize a new ClassVariableOrWriteNode node.
*/
static pm_class_variable_or_write_node_t *
pm_class_variable_or_write_node_create(pm_parser_t *parser, pm_class_variable_read_node_t *target, const pm_token_t *operator, pm_node_t *value) {
assert(operator->type == PM_TOKEN_PIPE_PIPE_EQUAL);
pm_class_variable_or_write_node_t *node = PM_ALLOC_NODE(parser, pm_class_variable_or_write_node_t);
*node = (pm_class_variable_or_write_node_t) {
{
.type = PM_CLASS_VARIABLE_OR_WRITE_NODE,
.location = {
.start = target->base.location.start,
.end = value->location.end
}
},
.name = target->name,
.name_loc = target->base.location,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value
};
return node;
}
/**
* Allocate and initialize a new ClassVariableReadNode node.
*/
static pm_class_variable_read_node_t *
pm_class_variable_read_node_create(pm_parser_t *parser, const pm_token_t *token) {
assert(token->type == PM_TOKEN_CLASS_VARIABLE);
pm_class_variable_read_node_t *node = PM_ALLOC_NODE(parser, pm_class_variable_read_node_t);
*node = (pm_class_variable_read_node_t) {
{
.type = PM_CLASS_VARIABLE_READ_NODE,
.location = PM_LOCATION_TOKEN_VALUE(token)
},
.name = pm_parser_constant_id_token(parser, token)
};
return node;
}
/**
* Initialize a new ClassVariableWriteNode node from a ClassVariableRead node.
*/
static pm_class_variable_write_node_t *
pm_class_variable_write_node_create(pm_parser_t *parser, pm_class_variable_read_node_t *read_node, pm_token_t *operator, pm_node_t *value) {
pm_class_variable_write_node_t *node = PM_ALLOC_NODE(parser, pm_class_variable_write_node_t);
*node = (pm_class_variable_write_node_t) {
{
.type = PM_CLASS_VARIABLE_WRITE_NODE,
.location = {
.start = read_node->base.location.start,
.end = value->location.end
},
},
.name = read_node->name,
.name_loc = PM_LOCATION_NODE_VALUE((pm_node_t *) read_node),
.operator_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(operator),
.value = value
};
return node;
}
/**
* Allocate and initialize a new ConstantPathAndWriteNode node.
*/
static pm_constant_path_and_write_node_t *
pm_constant_path_and_write_node_create(pm_parser_t *parser, pm_constant_path_node_t *target, const pm_token_t *operator, pm_node_t *value) {
assert(operator->type == PM_TOKEN_AMPERSAND_AMPERSAND_EQUAL);
pm_constant_path_and_write_node_t *node = PM_ALLOC_NODE(parser, pm_constant_path_and_write_node_t);
*node = (pm_constant_path_and_write_node_t) {
{
.type = PM_CONSTANT_PATH_AND_WRITE_NODE,
.location = {
.start = target->base.location.start,
.end = value->location.end
}
},
.target = target,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value
};
return node;
}
/**
* Allocate and initialize a new ConstantPathOperatorWriteNode node.
*/
static pm_constant_path_operator_write_node_t *
pm_constant_path_operator_write_node_create(pm_parser_t *parser, pm_constant_path_node_t *target, const pm_token_t *operator, pm_node_t *value) {
pm_constant_path_operator_write_node_t *node = PM_ALLOC_NODE(parser, pm_constant_path_operator_write_node_t);
*node = (pm_constant_path_operator_write_node_t) {
{
.type = PM_CONSTANT_PATH_OPERATOR_WRITE_NODE,
.location = {
.start = target->base.location.start,
.end = value->location.end
}
},
.target = target,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value,
.operator = pm_parser_constant_id_location(parser, operator->start, operator->end - 1)
};
return node;
}
/**
* Allocate and initialize a new ConstantPathOrWriteNode node.
*/
static pm_constant_path_or_write_node_t *
pm_constant_path_or_write_node_create(pm_parser_t *parser, pm_constant_path_node_t *target, const pm_token_t *operator, pm_node_t *value) {
assert(operator->type == PM_TOKEN_PIPE_PIPE_EQUAL);
pm_constant_path_or_write_node_t *node = PM_ALLOC_NODE(parser, pm_constant_path_or_write_node_t);
*node = (pm_constant_path_or_write_node_t) {
{
.type = PM_CONSTANT_PATH_OR_WRITE_NODE,
.location = {
.start = target->base.location.start,
.end = value->location.end
}
},
.target = target,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value
};
return node;
}
/**
* Allocate and initialize a new ConstantPathNode node.
*/
static pm_constant_path_node_t *
pm_constant_path_node_create(pm_parser_t *parser, pm_node_t *parent, const pm_token_t *delimiter, pm_node_t *child) {
pm_assert_value_expression(parser, parent);
pm_constant_path_node_t *node = PM_ALLOC_NODE(parser, pm_constant_path_node_t);
*node = (pm_constant_path_node_t) {
{
.type = PM_CONSTANT_PATH_NODE,
.location = {
.start = parent == NULL ? delimiter->start : parent->location.start,
.end = child->location.end
},
},
.parent = parent,
.child = child,
.delimiter_loc = PM_LOCATION_TOKEN_VALUE(delimiter)
};
return node;
}
/**
* Allocate a new ConstantPathWriteNode node.
*/
static pm_constant_path_write_node_t *
pm_constant_path_write_node_create(pm_parser_t *parser, pm_constant_path_node_t *target, const pm_token_t *operator, pm_node_t *value) {
pm_constant_path_write_node_t *node = PM_ALLOC_NODE(parser, pm_constant_path_write_node_t);
*node = (pm_constant_path_write_node_t) {
{
.type = PM_CONSTANT_PATH_WRITE_NODE,
.location = {
.start = target->base.location.start,
.end = value->location.end
},
},
.target = target,
.operator_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(operator),
.value = value
};
return node;
}
/**
* Allocate and initialize a new ConstantAndWriteNode node.
*/
static pm_constant_and_write_node_t *
pm_constant_and_write_node_create(pm_parser_t *parser, pm_constant_read_node_t *target, const pm_token_t *operator, pm_node_t *value) {
assert(operator->type == PM_TOKEN_AMPERSAND_AMPERSAND_EQUAL);
pm_constant_and_write_node_t *node = PM_ALLOC_NODE(parser, pm_constant_and_write_node_t);
*node = (pm_constant_and_write_node_t) {
{
.type = PM_CONSTANT_AND_WRITE_NODE,
.location = {
.start = target->base.location.start,
.end = value->location.end
}
},
.name = target->name,
.name_loc = target->base.location,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value
};
return node;
}
/**
* Allocate and initialize a new ConstantOperatorWriteNode node.
*/
static pm_constant_operator_write_node_t *
pm_constant_operator_write_node_create(pm_parser_t *parser, pm_constant_read_node_t *target, const pm_token_t *operator, pm_node_t *value) {
pm_constant_operator_write_node_t *node = PM_ALLOC_NODE(parser, pm_constant_operator_write_node_t);
*node = (pm_constant_operator_write_node_t) {
{
.type = PM_CONSTANT_OPERATOR_WRITE_NODE,
.location = {
.start = target->base.location.start,
.end = value->location.end
}
},
.name = target->name,
.name_loc = target->base.location,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value,
.operator = pm_parser_constant_id_location(parser, operator->start, operator->end - 1)
};
return node;
}
/**
* Allocate and initialize a new ConstantOrWriteNode node.
*/
static pm_constant_or_write_node_t *
pm_constant_or_write_node_create(pm_parser_t *parser, pm_constant_read_node_t *target, const pm_token_t *operator, pm_node_t *value) {
assert(operator->type == PM_TOKEN_PIPE_PIPE_EQUAL);
pm_constant_or_write_node_t *node = PM_ALLOC_NODE(parser, pm_constant_or_write_node_t);
*node = (pm_constant_or_write_node_t) {
{
.type = PM_CONSTANT_OR_WRITE_NODE,
.location = {
.start = target->base.location.start,
.end = value->location.end
}
},
.name = target->name,
.name_loc = target->base.location,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value
};
return node;
}
/**
* Allocate and initialize a new ConstantReadNode node.
*/
static pm_constant_read_node_t *
pm_constant_read_node_create(pm_parser_t *parser, const pm_token_t *name) {
assert(name->type == PM_TOKEN_CONSTANT || name->type == PM_TOKEN_MISSING);
pm_constant_read_node_t *node = PM_ALLOC_NODE(parser, pm_constant_read_node_t);
*node = (pm_constant_read_node_t) {
{
.type = PM_CONSTANT_READ_NODE,
.location = PM_LOCATION_TOKEN_VALUE(name)
},
.name = pm_parser_constant_id_token(parser, name)
};
return node;
}
/**
* Allocate a new ConstantWriteNode node.
*/
static pm_constant_write_node_t *
pm_constant_write_node_create(pm_parser_t *parser, pm_constant_read_node_t *target, const pm_token_t *operator, pm_node_t *value) {
pm_constant_write_node_t *node = PM_ALLOC_NODE(parser, pm_constant_write_node_t);
*node = (pm_constant_write_node_t) {
{
.type = PM_CONSTANT_WRITE_NODE,
.location = {
.start = target->base.location.start,
.end = value->location.end
}
},
.name = target->name,
.name_loc = target->base.location,
.operator_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(operator),
.value = value
};
return node;
}
/**
* Allocate and initialize a new DefNode node.
*/
static pm_def_node_t *
pm_def_node_create(
pm_parser_t *parser,
const pm_token_t *name,
pm_node_t *receiver,
pm_parameters_node_t *parameters,
pm_node_t *body,
pm_constant_id_list_t *locals,
uint32_t locals_body_index,
const pm_token_t *def_keyword,
const pm_token_t *operator,
const pm_token_t *lparen,
const pm_token_t *rparen,
const pm_token_t *equal,
const pm_token_t *end_keyword
) {
pm_def_node_t *node = PM_ALLOC_NODE(parser, pm_def_node_t);
const uint8_t *end;
if (end_keyword->type == PM_TOKEN_NOT_PROVIDED) {
end = body->location.end;
} else {
end = end_keyword->end;
}
*node = (pm_def_node_t) {
{
.type = PM_DEF_NODE,
.location = { .start = def_keyword->start, .end = end },
},
.name = pm_parser_constant_id_token(parser, name),
.name_loc = PM_LOCATION_TOKEN_VALUE(name),
.receiver = receiver,
.parameters = parameters,
.body = body,
.locals = *locals,
.locals_body_index = locals_body_index,
.def_keyword_loc = PM_LOCATION_TOKEN_VALUE(def_keyword),
.operator_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(operator),
.lparen_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(lparen),
.rparen_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(rparen),
.equal_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(equal),
.end_keyword_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(end_keyword)
};
return node;
}
/**
* Allocate a new DefinedNode node.
*/
static pm_defined_node_t *
pm_defined_node_create(pm_parser_t *parser, const pm_token_t *lparen, pm_node_t *value, const pm_token_t *rparen, const pm_location_t *keyword_loc) {
pm_defined_node_t *node = PM_ALLOC_NODE(parser, pm_defined_node_t);
*node = (pm_defined_node_t) {
{
.type = PM_DEFINED_NODE,
.location = {
.start = keyword_loc->start,
.end = (rparen->type == PM_TOKEN_NOT_PROVIDED ? value->location.end : rparen->end)
},
},
.lparen_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(lparen),
.value = value,
.rparen_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(rparen),
.keyword_loc = *keyword_loc
};
return node;
}
/**
* Allocate and initialize a new ElseNode node.
*/
static pm_else_node_t *
pm_else_node_create(pm_parser_t *parser, const pm_token_t *else_keyword, pm_statements_node_t *statements, const pm_token_t *end_keyword) {
pm_else_node_t *node = PM_ALLOC_NODE(parser, pm_else_node_t);
const uint8_t *end = NULL;
if ((end_keyword->type == PM_TOKEN_NOT_PROVIDED) && (statements != NULL)) {
end = statements->base.location.end;
} else {
end = end_keyword->end;
}
*node = (pm_else_node_t) {
{
.type = PM_ELSE_NODE,
.location = {
.start = else_keyword->start,
.end = end,
},
},
.else_keyword_loc = PM_LOCATION_TOKEN_VALUE(else_keyword),
.statements = statements,
.end_keyword_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(end_keyword)
};
return node;
}
/**
* Allocate and initialize a new EmbeddedStatementsNode node.
*/
static pm_embedded_statements_node_t *
pm_embedded_statements_node_create(pm_parser_t *parser, const pm_token_t *opening, pm_statements_node_t *statements, const pm_token_t *closing) {
pm_embedded_statements_node_t *node = PM_ALLOC_NODE(parser, pm_embedded_statements_node_t);
*node = (pm_embedded_statements_node_t) {
{
.type = PM_EMBEDDED_STATEMENTS_NODE,
.location = {
.start = opening->start,
.end = closing->end
}
},
.opening_loc = PM_LOCATION_TOKEN_VALUE(opening),
.statements = statements,
.closing_loc = PM_LOCATION_TOKEN_VALUE(closing)
};
return node;
}
/**
* Allocate and initialize a new EmbeddedVariableNode node.
*/
static pm_embedded_variable_node_t *
pm_embedded_variable_node_create(pm_parser_t *parser, const pm_token_t *operator, pm_node_t *variable) {
pm_embedded_variable_node_t *node = PM_ALLOC_NODE(parser, pm_embedded_variable_node_t);
*node = (pm_embedded_variable_node_t) {
{
.type = PM_EMBEDDED_VARIABLE_NODE,
.location = {
.start = operator->start,
.end = variable->location.end
}
},
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.variable = variable
};
return node;
}
/**
* Allocate a new EnsureNode node.
*/
static pm_ensure_node_t *
pm_ensure_node_create(pm_parser_t *parser, const pm_token_t *ensure_keyword, pm_statements_node_t *statements, const pm_token_t *end_keyword) {
pm_ensure_node_t *node = PM_ALLOC_NODE(parser, pm_ensure_node_t);
*node = (pm_ensure_node_t) {
{
.type = PM_ENSURE_NODE,
.location = {
.start = ensure_keyword->start,
.end = end_keyword->end
},
},
.ensure_keyword_loc = PM_LOCATION_TOKEN_VALUE(ensure_keyword),
.statements = statements,
.end_keyword_loc = PM_LOCATION_TOKEN_VALUE(end_keyword)
};
return node;
}
/**
* Allocate and initialize a new FalseNode node.
*/
static pm_false_node_t *
pm_false_node_create(pm_parser_t *parser, const pm_token_t *token) {
assert(token->type == PM_TOKEN_KEYWORD_FALSE);
pm_false_node_t *node = PM_ALLOC_NODE(parser, pm_false_node_t);
*node = (pm_false_node_t) {{
.type = PM_FALSE_NODE,
.flags = PM_NODE_FLAG_STATIC_LITERAL,
.location = PM_LOCATION_TOKEN_VALUE(token)
}};
return node;
}
/**
* Allocate and initialize a new find pattern node. The node list given in the
* nodes parameter is guaranteed to have at least two nodes.
*/
static pm_find_pattern_node_t *
pm_find_pattern_node_create(pm_parser_t *parser, pm_node_list_t *nodes) {
pm_find_pattern_node_t *node = PM_ALLOC_NODE(parser, pm_find_pattern_node_t);
pm_node_t *left = nodes->nodes[0];
pm_node_t *right;
if (nodes->size == 1) {
right = (pm_node_t *) pm_missing_node_create(parser, left->location.end, left->location.end);
} else {
right = nodes->nodes[nodes->size - 1];
}
*node = (pm_find_pattern_node_t) {
{
.type = PM_FIND_PATTERN_NODE,
.location = {
.start = left->location.start,
.end = right->location.end,
},
},
.constant = NULL,
.left = left,
.right = right,
.requireds = { 0 },
.opening_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE,
.closing_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE
};
// For now we're going to just copy over each pointer manually. This could be
// much more efficient, as we could instead resize the node list to only point
// to 1...-1.
for (size_t index = 1; index < nodes->size - 1; index++) {
pm_node_list_append(&node->requireds, nodes->nodes[index]);
}
return node;
}
/**
* Allocate and initialize a new FloatNode node.
*/
static pm_float_node_t *
pm_float_node_create(pm_parser_t *parser, const pm_token_t *token) {
assert(token->type == PM_TOKEN_FLOAT);
pm_float_node_t *node = PM_ALLOC_NODE(parser, pm_float_node_t);
*node = (pm_float_node_t) {{
.type = PM_FLOAT_NODE,
.flags = PM_NODE_FLAG_STATIC_LITERAL,
.location = PM_LOCATION_TOKEN_VALUE(token)
}};
return node;
}
/**
* Allocate and initialize a new FloatNode node from a FLOAT_IMAGINARY token.
*/
static pm_imaginary_node_t *
pm_float_node_imaginary_create(pm_parser_t *parser, const pm_token_t *token) {
assert(token->type == PM_TOKEN_FLOAT_IMAGINARY);
pm_imaginary_node_t *node = PM_ALLOC_NODE(parser, pm_imaginary_node_t);
*node = (pm_imaginary_node_t) {
{
.type = PM_IMAGINARY_NODE,
.flags = PM_NODE_FLAG_STATIC_LITERAL,
.location = PM_LOCATION_TOKEN_VALUE(token)
},
.numeric = (pm_node_t *) pm_float_node_create(parser, &((pm_token_t) {
.type = PM_TOKEN_FLOAT,
.start = token->start,
.end = token->end - 1
}))
};
return node;
}
/**
* Allocate and initialize a new FloatNode node from a FLOAT_RATIONAL token.
*/
static pm_rational_node_t *
pm_float_node_rational_create(pm_parser_t *parser, const pm_token_t *token) {
assert(token->type == PM_TOKEN_FLOAT_RATIONAL);
pm_rational_node_t *node = PM_ALLOC_NODE(parser, pm_rational_node_t);
*node = (pm_rational_node_t) {
{
.type = PM_RATIONAL_NODE,
.flags = PM_NODE_FLAG_STATIC_LITERAL,
.location = PM_LOCATION_TOKEN_VALUE(token)
},
.numeric = (pm_node_t *) pm_float_node_create(parser, &((pm_token_t) {
.type = PM_TOKEN_FLOAT,
.start = token->start,
.end = token->end - 1
}))
};
return node;
}
/**
* Allocate and initialize a new FloatNode node from a FLOAT_RATIONAL_IMAGINARY
* token.
*/
static pm_imaginary_node_t *
pm_float_node_rational_imaginary_create(pm_parser_t *parser, const pm_token_t *token) {
assert(token->type == PM_TOKEN_FLOAT_RATIONAL_IMAGINARY);
pm_imaginary_node_t *node = PM_ALLOC_NODE(parser, pm_imaginary_node_t);
*node = (pm_imaginary_node_t) {
{
.type = PM_IMAGINARY_NODE,
.flags = PM_NODE_FLAG_STATIC_LITERAL,
.location = PM_LOCATION_TOKEN_VALUE(token)
},
.numeric = (pm_node_t *) pm_float_node_rational_create(parser, &((pm_token_t) {
.type = PM_TOKEN_FLOAT_RATIONAL,
.start = token->start,
.end = token->end - 1
}))
};
return node;
}
/**
* Allocate and initialize a new ForNode node.
*/
static pm_for_node_t *
pm_for_node_create(
pm_parser_t *parser,
pm_node_t *index,
pm_node_t *collection,
pm_statements_node_t *statements,
const pm_token_t *for_keyword,
const pm_token_t *in_keyword,
const pm_token_t *do_keyword,
const pm_token_t *end_keyword
) {
pm_for_node_t *node = PM_ALLOC_NODE(parser, pm_for_node_t);
*node = (pm_for_node_t) {
{
.type = PM_FOR_NODE,
.location = {
.start = for_keyword->start,
.end = end_keyword->end
},
},
.index = index,
.collection = collection,
.statements = statements,
.for_keyword_loc = PM_LOCATION_TOKEN_VALUE(for_keyword),
.in_keyword_loc = PM_LOCATION_TOKEN_VALUE(in_keyword),
.do_keyword_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(do_keyword),
.end_keyword_loc = PM_LOCATION_TOKEN_VALUE(end_keyword)
};
return node;
}
/**
* Allocate and initialize a new ForwardingArgumentsNode node.
*/
static pm_forwarding_arguments_node_t *
pm_forwarding_arguments_node_create(pm_parser_t *parser, const pm_token_t *token) {
assert(token->type == PM_TOKEN_UDOT_DOT_DOT);
pm_forwarding_arguments_node_t *node = PM_ALLOC_NODE(parser, pm_forwarding_arguments_node_t);
*node = (pm_forwarding_arguments_node_t) {{ .type = PM_FORWARDING_ARGUMENTS_NODE, .location = PM_LOCATION_TOKEN_VALUE(token) }};
return node;
}
/**
* Allocate and initialize a new ForwardingParameterNode node.
*/
static pm_forwarding_parameter_node_t *
pm_forwarding_parameter_node_create(pm_parser_t *parser, const pm_token_t *token) {
assert(token->type == PM_TOKEN_UDOT_DOT_DOT);
pm_forwarding_parameter_node_t *node = PM_ALLOC_NODE(parser, pm_forwarding_parameter_node_t);
*node = (pm_forwarding_parameter_node_t) {{ .type = PM_FORWARDING_PARAMETER_NODE, .location = PM_LOCATION_TOKEN_VALUE(token) }};
return node;
}
/**
* Allocate and initialize a new ForwardingSuper node.
*/
static pm_forwarding_super_node_t *
pm_forwarding_super_node_create(pm_parser_t *parser, const pm_token_t *token, pm_arguments_t *arguments) {
assert(arguments->block == NULL || PM_NODE_TYPE_P(arguments->block, PM_BLOCK_NODE));
assert(token->type == PM_TOKEN_KEYWORD_SUPER);
pm_forwarding_super_node_t *node = PM_ALLOC_NODE(parser, pm_forwarding_super_node_t);
pm_block_node_t *block = NULL;
if (arguments->block != NULL) {
block = (pm_block_node_t *) arguments->block;
}
*node = (pm_forwarding_super_node_t) {
{
.type = PM_FORWARDING_SUPER_NODE,
.location = {
.start = token->start,
.end = block != NULL ? block->base.location.end : token->end
},
},
.block = block
};
return node;
}
/**
* Allocate and initialize a new hash pattern node from an opening and closing
* token.
*/
static pm_hash_pattern_node_t *
pm_hash_pattern_node_empty_create(pm_parser_t *parser, const pm_token_t *opening, const pm_token_t *closing) {
pm_hash_pattern_node_t *node = PM_ALLOC_NODE(parser, pm_hash_pattern_node_t);
*node = (pm_hash_pattern_node_t) {
{
.type = PM_HASH_PATTERN_NODE,
.location = {
.start = opening->start,
.end = closing->end
},
},
.constant = NULL,
.opening_loc = PM_LOCATION_TOKEN_VALUE(opening),
.closing_loc = PM_LOCATION_TOKEN_VALUE(closing),
.elements = { 0 },
.rest = NULL
};
return node;
}
/**
* Allocate and initialize a new hash pattern node.
*/
static pm_hash_pattern_node_t *
pm_hash_pattern_node_node_list_create(pm_parser_t *parser, pm_node_list_t *elements, pm_node_t *rest) {
pm_hash_pattern_node_t *node = PM_ALLOC_NODE(parser, pm_hash_pattern_node_t);
const uint8_t *start;
const uint8_t *end;
if (elements->size > 0) {
if (rest) {
start = elements->nodes[0]->location.start;
end = rest->location.end;
} else {
start = elements->nodes[0]->location.start;
end = elements->nodes[elements->size - 1]->location.end;
}
} else {
assert(rest != NULL);
start = rest->location.start;
end = rest->location.end;
}
*node = (pm_hash_pattern_node_t) {
{
.type = PM_HASH_PATTERN_NODE,
.location = {
.start = start,
.end = end
},
},
.constant = NULL,
.elements = { 0 },
.rest = rest,
.opening_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE,
.closing_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE
};
for (size_t index = 0; index < elements->size; index++) {
pm_node_t *element = elements->nodes[index];
pm_node_list_append(&node->elements, element);
}
return node;
}
/**
* Retrieve the name from a node that will become a global variable write node.
*/
static pm_constant_id_t
pm_global_variable_write_name(pm_parser_t *parser, const pm_node_t *target) {
switch (PM_NODE_TYPE(target)) {
case PM_GLOBAL_VARIABLE_READ_NODE:
return ((pm_global_variable_read_node_t *) target)->name;
case PM_BACK_REFERENCE_READ_NODE:
return ((pm_back_reference_read_node_t *) target)->name;
case PM_NUMBERED_REFERENCE_READ_NODE:
// This will only ever happen in the event of a syntax error, but we
// still need to provide something for the node.
return pm_parser_constant_id_location(parser, target->location.start, target->location.end);
default:
assert(false && "unreachable");
return (pm_constant_id_t) -1;
}
}
/**
* Allocate and initialize a new GlobalVariableAndWriteNode node.
*/
static pm_global_variable_and_write_node_t *
pm_global_variable_and_write_node_create(pm_parser_t *parser, pm_node_t *target, const pm_token_t *operator, pm_node_t *value) {
assert(operator->type == PM_TOKEN_AMPERSAND_AMPERSAND_EQUAL);
pm_global_variable_and_write_node_t *node = PM_ALLOC_NODE(parser, pm_global_variable_and_write_node_t);
*node = (pm_global_variable_and_write_node_t) {
{
.type = PM_GLOBAL_VARIABLE_AND_WRITE_NODE,
.location = {
.start = target->location.start,
.end = value->location.end
}
},
.name = pm_global_variable_write_name(parser, target),
.name_loc = target->location,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value
};
return node;
}
/**
* Allocate and initialize a new GlobalVariableOperatorWriteNode node.
*/
static pm_global_variable_operator_write_node_t *
pm_global_variable_operator_write_node_create(pm_parser_t *parser, pm_node_t *target, const pm_token_t *operator, pm_node_t *value) {
pm_global_variable_operator_write_node_t *node = PM_ALLOC_NODE(parser, pm_global_variable_operator_write_node_t);
*node = (pm_global_variable_operator_write_node_t) {
{
.type = PM_GLOBAL_VARIABLE_OPERATOR_WRITE_NODE,
.location = {
.start = target->location.start,
.end = value->location.end
}
},
.name = pm_global_variable_write_name(parser, target),
.name_loc = target->location,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value,
.operator = pm_parser_constant_id_location(parser, operator->start, operator->end - 1)
};
return node;
}
/**
* Allocate and initialize a new GlobalVariableOrWriteNode node.
*/
static pm_global_variable_or_write_node_t *
pm_global_variable_or_write_node_create(pm_parser_t *parser, pm_node_t *target, const pm_token_t *operator, pm_node_t *value) {
assert(operator->type == PM_TOKEN_PIPE_PIPE_EQUAL);
pm_global_variable_or_write_node_t *node = PM_ALLOC_NODE(parser, pm_global_variable_or_write_node_t);
*node = (pm_global_variable_or_write_node_t) {
{
.type = PM_GLOBAL_VARIABLE_OR_WRITE_NODE,
.location = {
.start = target->location.start,
.end = value->location.end
}
},
.name = pm_global_variable_write_name(parser, target),
.name_loc = target->location,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value
};
return node;
}
/**
* Allocate a new GlobalVariableReadNode node.
*/
static pm_global_variable_read_node_t *
pm_global_variable_read_node_create(pm_parser_t *parser, const pm_token_t *name) {
pm_global_variable_read_node_t *node = PM_ALLOC_NODE(parser, pm_global_variable_read_node_t);
*node = (pm_global_variable_read_node_t) {
{
.type = PM_GLOBAL_VARIABLE_READ_NODE,
.location = PM_LOCATION_TOKEN_VALUE(name),
},
.name = pm_parser_constant_id_token(parser, name)
};
return node;
}
/**
* Allocate a new GlobalVariableWriteNode node.
*/
static pm_global_variable_write_node_t *
pm_global_variable_write_node_create(pm_parser_t *parser, pm_node_t *target, const pm_token_t *operator, pm_node_t *value) {
pm_global_variable_write_node_t *node = PM_ALLOC_NODE(parser, pm_global_variable_write_node_t);
*node = (pm_global_variable_write_node_t) {
{
.type = PM_GLOBAL_VARIABLE_WRITE_NODE,
.location = {
.start = target->location.start,
.end = value->location.end
},
},
.name = pm_global_variable_write_name(parser, target),
.name_loc = PM_LOCATION_NODE_VALUE(target),
.operator_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(operator),
.value = value
};
return node;
}
/**
* Allocate a new HashNode node.
*/
static pm_hash_node_t *
pm_hash_node_create(pm_parser_t *parser, const pm_token_t *opening) {
assert(opening != NULL);
pm_hash_node_t *node = PM_ALLOC_NODE(parser, pm_hash_node_t);
*node = (pm_hash_node_t) {
{
.type = PM_HASH_NODE,
.flags = PM_NODE_FLAG_STATIC_LITERAL,
.location = PM_LOCATION_TOKEN_VALUE(opening)
},
.opening_loc = PM_LOCATION_TOKEN_VALUE(opening),
.closing_loc = PM_LOCATION_NULL_VALUE(parser),
.elements = { 0 }
};
return node;
}
/**
* Append a new element to a hash node.
*/
static inline void
pm_hash_node_elements_append(pm_hash_node_t *hash, pm_node_t *element) {
pm_node_list_append(&hash->elements, element);
bool static_literal = PM_NODE_TYPE_P(element, PM_ASSOC_NODE);
if (static_literal) {
pm_assoc_node_t *assoc = (pm_assoc_node_t *) element;
static_literal = !PM_NODE_TYPE_P(assoc->key, PM_ARRAY_NODE) && !PM_NODE_TYPE_P(assoc->key, PM_HASH_NODE) && !PM_NODE_TYPE_P(assoc->key, PM_RANGE_NODE);
static_literal = static_literal && PM_NODE_FLAG_P(assoc->key, PM_NODE_FLAG_STATIC_LITERAL);
static_literal = static_literal && PM_NODE_FLAG_P(assoc, PM_NODE_FLAG_STATIC_LITERAL);
}
if (!static_literal) {
pm_node_flag_unset((pm_node_t *)hash, PM_NODE_FLAG_STATIC_LITERAL);
}
}
static inline void
pm_hash_node_closing_loc_set(pm_hash_node_t *hash, pm_token_t *token) {
hash->base.location.end = token->end;
hash->closing_loc = PM_LOCATION_TOKEN_VALUE(token);
}
/**
* Allocate a new IfNode node.
*/
static pm_if_node_t *
pm_if_node_create(pm_parser_t *parser,
const pm_token_t *if_keyword,
pm_node_t *predicate,
const pm_token_t *then_keyword,
pm_statements_node_t *statements,
pm_node_t *consequent,
const pm_token_t *end_keyword
) {
pm_conditional_predicate(predicate);
pm_if_node_t *node = PM_ALLOC_NODE(parser, pm_if_node_t);
const uint8_t *end;
if (end_keyword->type != PM_TOKEN_NOT_PROVIDED) {
end = end_keyword->end;
} else if (consequent != NULL) {
end = consequent->location.end;
} else if (pm_statements_node_body_length(statements) != 0) {
end = statements->base.location.end;
} else {
end = predicate->location.end;
}
*node = (pm_if_node_t) {
{
.type = PM_IF_NODE,
.flags = PM_NODE_FLAG_NEWLINE,
.location = {
.start = if_keyword->start,
.end = end
},
},
.if_keyword_loc = PM_LOCATION_TOKEN_VALUE(if_keyword),
.predicate = predicate,
.then_keyword_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(then_keyword),
.statements = statements,
.consequent = consequent,
.end_keyword_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(end_keyword)
};
return node;
}
/**
* Allocate and initialize new IfNode node in the modifier form.
*/
static pm_if_node_t *
pm_if_node_modifier_create(pm_parser_t *parser, pm_node_t *statement, const pm_token_t *if_keyword, pm_node_t *predicate) {
pm_conditional_predicate(predicate);
pm_if_node_t *node = PM_ALLOC_NODE(parser, pm_if_node_t);
pm_statements_node_t *statements = pm_statements_node_create(parser);
pm_statements_node_body_append(statements, statement);
*node = (pm_if_node_t) {
{
.type = PM_IF_NODE,
.flags = PM_NODE_FLAG_NEWLINE,
.location = {
.start = statement->location.start,
.end = predicate->location.end
},
},
.if_keyword_loc = PM_LOCATION_TOKEN_VALUE(if_keyword),
.predicate = predicate,
.then_keyword_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE,
.statements = statements,
.consequent = NULL,
.end_keyword_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE
};
return node;
}
/**
* Allocate and initialize an if node from a ternary expression.
*/
static pm_if_node_t *
pm_if_node_ternary_create(pm_parser_t *parser, pm_node_t *predicate, const pm_token_t *qmark, pm_node_t *true_expression, const pm_token_t *colon, pm_node_t *false_expression) {
pm_assert_value_expression(parser, predicate);
pm_conditional_predicate(predicate);
pm_statements_node_t *if_statements = pm_statements_node_create(parser);
pm_statements_node_body_append(if_statements, true_expression);
pm_statements_node_t *else_statements = pm_statements_node_create(parser);
pm_statements_node_body_append(else_statements, false_expression);
pm_token_t end_keyword = not_provided(parser);
pm_else_node_t *else_node = pm_else_node_create(parser, colon, else_statements, &end_keyword);
pm_if_node_t *node = PM_ALLOC_NODE(parser, pm_if_node_t);
*node = (pm_if_node_t) {
{
.type = PM_IF_NODE,
.flags = PM_NODE_FLAG_NEWLINE,
.location = {
.start = predicate->location.start,
.end = false_expression->location.end,
},
},
.if_keyword_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE,
.predicate = predicate,
.then_keyword_loc = PM_LOCATION_TOKEN_VALUE(qmark),
.statements = if_statements,
.consequent = (pm_node_t *)else_node,
.end_keyword_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE
};
return node;
}
static inline void
pm_if_node_end_keyword_loc_set(pm_if_node_t *node, const pm_token_t *keyword) {
node->base.location.end = keyword->end;
node->end_keyword_loc = PM_LOCATION_TOKEN_VALUE(keyword);
}
static inline void
pm_else_node_end_keyword_loc_set(pm_else_node_t *node, const pm_token_t *keyword) {
node->base.location.end = keyword->end;
node->end_keyword_loc = PM_LOCATION_TOKEN_VALUE(keyword);
}
/**
* Allocate and initialize a new ImplicitNode node.
*/
static pm_implicit_node_t *
pm_implicit_node_create(pm_parser_t *parser, pm_node_t *value) {
pm_implicit_node_t *node = PM_ALLOC_NODE(parser, pm_implicit_node_t);
*node = (pm_implicit_node_t) {
{
.type = PM_IMPLICIT_NODE,
.location = value->location
},
.value = value
};
return node;
}
/**
* Allocate and initialize a new ImplicitRestNode node.
*/
static pm_implicit_rest_node_t *
pm_implicit_rest_node_create(pm_parser_t *parser, const pm_token_t *token) {
assert(token->type == PM_TOKEN_COMMA);
pm_implicit_rest_node_t *node = PM_ALLOC_NODE(parser, pm_implicit_rest_node_t);
*node = (pm_implicit_rest_node_t) {
{
.type = PM_IMPLICIT_REST_NODE,
.location = PM_LOCATION_TOKEN_VALUE(token)
}
};
return node;
}
/**
* Allocate and initialize a new IntegerNode node.
*/
static pm_integer_node_t *
pm_integer_node_create(pm_parser_t *parser, pm_node_flags_t base, const pm_token_t *token) {
assert(token->type == PM_TOKEN_INTEGER);
pm_integer_node_t *node = PM_ALLOC_NODE(parser, pm_integer_node_t);
*node = (pm_integer_node_t) {{
.type = PM_INTEGER_NODE,
.flags = base | PM_NODE_FLAG_STATIC_LITERAL,
.location = PM_LOCATION_TOKEN_VALUE(token)
}};
return node;
}
/**
* Allocate and initialize a new IntegerNode node from an INTEGER_IMAGINARY
* token.
*/
static pm_imaginary_node_t *
pm_integer_node_imaginary_create(pm_parser_t *parser, pm_node_flags_t base, const pm_token_t *token) {
assert(token->type == PM_TOKEN_INTEGER_IMAGINARY);
pm_imaginary_node_t *node = PM_ALLOC_NODE(parser, pm_imaginary_node_t);
*node = (pm_imaginary_node_t) {
{
.type = PM_IMAGINARY_NODE,
.flags = PM_NODE_FLAG_STATIC_LITERAL,
.location = PM_LOCATION_TOKEN_VALUE(token)
},
.numeric = (pm_node_t *) pm_integer_node_create(parser, base, &((pm_token_t) {
.type = PM_TOKEN_INTEGER,
.start = token->start,
.end = token->end - 1
}))
};
return node;
}
/**
* Allocate and initialize a new IntegerNode node from an INTEGER_RATIONAL
* token.
*/
static pm_rational_node_t *
pm_integer_node_rational_create(pm_parser_t *parser, pm_node_flags_t base, const pm_token_t *token) {
assert(token->type == PM_TOKEN_INTEGER_RATIONAL);
pm_rational_node_t *node = PM_ALLOC_NODE(parser, pm_rational_node_t);
*node = (pm_rational_node_t) {
{
.type = PM_RATIONAL_NODE,
.flags = PM_NODE_FLAG_STATIC_LITERAL,
.location = PM_LOCATION_TOKEN_VALUE(token)
},
.numeric = (pm_node_t *) pm_integer_node_create(parser, base, &((pm_token_t) {
.type = PM_TOKEN_INTEGER,
.start = token->start,
.end = token->end - 1
}))
};
return node;
}
/**
* Allocate and initialize a new IntegerNode node from an
* INTEGER_RATIONAL_IMAGINARY token.
*/
static pm_imaginary_node_t *
pm_integer_node_rational_imaginary_create(pm_parser_t *parser, pm_node_flags_t base, const pm_token_t *token) {
assert(token->type == PM_TOKEN_INTEGER_RATIONAL_IMAGINARY);
pm_imaginary_node_t *node = PM_ALLOC_NODE(parser, pm_imaginary_node_t);
*node = (pm_imaginary_node_t) {
{
.type = PM_IMAGINARY_NODE,
.flags = PM_NODE_FLAG_STATIC_LITERAL,
.location = PM_LOCATION_TOKEN_VALUE(token)
},
.numeric = (pm_node_t *) pm_integer_node_rational_create(parser, base, &((pm_token_t) {
.type = PM_TOKEN_INTEGER_RATIONAL,
.start = token->start,
.end = token->end - 1
}))
};
return node;
}
/**
* Allocate and initialize a new InNode node.
*/
static pm_in_node_t *
pm_in_node_create(pm_parser_t *parser, pm_node_t *pattern, pm_statements_node_t *statements, const pm_token_t *in_keyword, const pm_token_t *then_keyword) {
pm_in_node_t *node = PM_ALLOC_NODE(parser, pm_in_node_t);
const uint8_t *end;
if (statements != NULL) {
end = statements->base.location.end;
} else if (then_keyword->type != PM_TOKEN_NOT_PROVIDED) {
end = then_keyword->end;
} else {
end = pattern->location.end;
}
*node = (pm_in_node_t) {
{
.type = PM_IN_NODE,
.location = {
.start = in_keyword->start,
.end = end
},
},
.pattern = pattern,
.statements = statements,
.in_loc = PM_LOCATION_TOKEN_VALUE(in_keyword),
.then_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(then_keyword)
};
return node;
}
/**
* Allocate and initialize a new InstanceVariableAndWriteNode node.
*/
static pm_instance_variable_and_write_node_t *
pm_instance_variable_and_write_node_create(pm_parser_t *parser, pm_instance_variable_read_node_t *target, const pm_token_t *operator, pm_node_t *value) {
assert(operator->type == PM_TOKEN_AMPERSAND_AMPERSAND_EQUAL);
pm_instance_variable_and_write_node_t *node = PM_ALLOC_NODE(parser, pm_instance_variable_and_write_node_t);
*node = (pm_instance_variable_and_write_node_t) {
{
.type = PM_INSTANCE_VARIABLE_AND_WRITE_NODE,
.location = {
.start = target->base.location.start,
.end = value->location.end
}
},
.name = target->name,
.name_loc = target->base.location,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value
};
return node;
}
/**
* Allocate and initialize a new InstanceVariableOperatorWriteNode node.
*/
static pm_instance_variable_operator_write_node_t *
pm_instance_variable_operator_write_node_create(pm_parser_t *parser, pm_instance_variable_read_node_t *target, const pm_token_t *operator, pm_node_t *value) {
pm_instance_variable_operator_write_node_t *node = PM_ALLOC_NODE(parser, pm_instance_variable_operator_write_node_t);
*node = (pm_instance_variable_operator_write_node_t) {
{
.type = PM_INSTANCE_VARIABLE_OPERATOR_WRITE_NODE,
.location = {
.start = target->base.location.start,
.end = value->location.end
}
},
.name = target->name,
.name_loc = target->base.location,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value,
.operator = pm_parser_constant_id_location(parser, operator->start, operator->end - 1)
};
return node;
}
/**
* Allocate and initialize a new InstanceVariableOrWriteNode node.
*/
static pm_instance_variable_or_write_node_t *
pm_instance_variable_or_write_node_create(pm_parser_t *parser, pm_instance_variable_read_node_t *target, const pm_token_t *operator, pm_node_t *value) {
assert(operator->type == PM_TOKEN_PIPE_PIPE_EQUAL);
pm_instance_variable_or_write_node_t *node = PM_ALLOC_NODE(parser, pm_instance_variable_or_write_node_t);
*node = (pm_instance_variable_or_write_node_t) {
{
.type = PM_INSTANCE_VARIABLE_OR_WRITE_NODE,
.location = {
.start = target->base.location.start,
.end = value->location.end
}
},
.name = target->name,
.name_loc = target->base.location,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value
};
return node;
}
/**
* Allocate and initialize a new InstanceVariableReadNode node.
*/
static pm_instance_variable_read_node_t *
pm_instance_variable_read_node_create(pm_parser_t *parser, const pm_token_t *token) {
assert(token->type == PM_TOKEN_INSTANCE_VARIABLE);
pm_instance_variable_read_node_t *node = PM_ALLOC_NODE(parser, pm_instance_variable_read_node_t);
*node = (pm_instance_variable_read_node_t) {
{
.type = PM_INSTANCE_VARIABLE_READ_NODE,
.location = PM_LOCATION_TOKEN_VALUE(token)
},
.name = pm_parser_constant_id_token(parser, token)
};
return node;
}
/**
* Initialize a new InstanceVariableWriteNode node from an InstanceVariableRead
* node.
*/
static pm_instance_variable_write_node_t *
pm_instance_variable_write_node_create(pm_parser_t *parser, pm_instance_variable_read_node_t *read_node, pm_token_t *operator, pm_node_t *value) {
pm_instance_variable_write_node_t *node = PM_ALLOC_NODE(parser, pm_instance_variable_write_node_t);
*node = (pm_instance_variable_write_node_t) {
{
.type = PM_INSTANCE_VARIABLE_WRITE_NODE,
.location = {
.start = read_node->base.location.start,
.end = value->location.end
}
},
.name = read_node->name,
.name_loc = PM_LOCATION_NODE_BASE_VALUE(read_node),
.operator_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(operator),
.value = value
};
return node;
}
/**
* Allocate a new InterpolatedRegularExpressionNode node.
*/
static pm_interpolated_regular_expression_node_t *
pm_interpolated_regular_expression_node_create(pm_parser_t *parser, const pm_token_t *opening) {
pm_interpolated_regular_expression_node_t *node = PM_ALLOC_NODE(parser, pm_interpolated_regular_expression_node_t);
*node = (pm_interpolated_regular_expression_node_t) {
{
.type = PM_INTERPOLATED_REGULAR_EXPRESSION_NODE,
.location = {
.start = opening->start,
.end = NULL,
},
},
.opening_loc = PM_LOCATION_TOKEN_VALUE(opening),
.closing_loc = PM_LOCATION_TOKEN_VALUE(opening),
.parts = { 0 }
};
return node;
}
static inline void
pm_interpolated_regular_expression_node_append(pm_interpolated_regular_expression_node_t *node, pm_node_t *part) {
if (node->base.location.start > part->location.start) {
node->base.location.start = part->location.start;
}
if (node->base.location.end < part->location.end) {
node->base.location.end = part->location.end;
}
pm_node_list_append(&node->parts, part);
}
static inline void
pm_interpolated_regular_expression_node_closing_set(pm_interpolated_regular_expression_node_t *node, const pm_token_t *closing) {
node->closing_loc = PM_LOCATION_TOKEN_VALUE(closing);
node->base.location.end = closing->end;
pm_node_flag_set((pm_node_t *)node, pm_regular_expression_flags_create(closing));
}
/**
* Allocate and initialize a new InterpolatedStringNode node.
*/
static pm_interpolated_string_node_t *
pm_interpolated_string_node_create(pm_parser_t *parser, const pm_token_t *opening, const pm_node_list_t *parts, const pm_token_t *closing) {
pm_interpolated_string_node_t *node = PM_ALLOC_NODE(parser, pm_interpolated_string_node_t);
*node = (pm_interpolated_string_node_t) {
{
.type = PM_INTERPOLATED_STRING_NODE,
.location = {
.start = opening->start,
.end = closing->end,
},
},
.opening_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(opening),
.closing_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(closing),
.parts = { 0 }
};
if (parts != NULL) {
node->parts = *parts;
}
return node;
}
/**
* Append a part to an InterpolatedStringNode node.
*/
static inline void
pm_interpolated_string_node_append(pm_interpolated_string_node_t *node, pm_node_t *part) {
if (node->parts.size == 0 && node->opening_loc.start == NULL) {
node->base.location.start = part->location.start;
}
pm_node_list_append(&node->parts, part);
node->base.location.end = part->location.end;
}
/**
* Set the closing token of the given InterpolatedStringNode node.
*/
static void
pm_interpolated_string_node_closing_set(pm_interpolated_string_node_t *node, const pm_token_t *closing) {
node->closing_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(closing);
node->base.location.end = closing->end;
}
/**
* Allocate and initialize a new InterpolatedSymbolNode node.
*/
static pm_interpolated_symbol_node_t *
pm_interpolated_symbol_node_create(pm_parser_t *parser, const pm_token_t *opening, const pm_node_list_t *parts, const pm_token_t *closing) {
pm_interpolated_symbol_node_t *node = PM_ALLOC_NODE(parser, pm_interpolated_symbol_node_t);
*node = (pm_interpolated_symbol_node_t) {
{
.type = PM_INTERPOLATED_SYMBOL_NODE,
.location = {
.start = opening->start,
.end = closing->end,
},
},
.opening_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(opening),
.closing_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(closing),
.parts = { 0 }
};
if (parts != NULL) {
node->parts = *parts;
}
return node;
}
static inline void
pm_interpolated_symbol_node_append(pm_interpolated_symbol_node_t *node, pm_node_t *part) {
if (node->parts.size == 0 && node->opening_loc.start == NULL) {
node->base.location.start = part->location.start;
}
pm_node_list_append(&node->parts, part);
node->base.location.end = part->location.end;
}
/**
* Allocate a new InterpolatedXStringNode node.
*/
static pm_interpolated_x_string_node_t *
pm_interpolated_xstring_node_create(pm_parser_t *parser, const pm_token_t *opening, const pm_token_t *closing) {
pm_interpolated_x_string_node_t *node = PM_ALLOC_NODE(parser, pm_interpolated_x_string_node_t);
*node = (pm_interpolated_x_string_node_t) {
{
.type = PM_INTERPOLATED_X_STRING_NODE,
.location = {
.start = opening->start,
.end = closing->end
},
},
.opening_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(opening),
.closing_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(closing),
.parts = { 0 }
};
return node;
}
static inline void
pm_interpolated_xstring_node_append(pm_interpolated_x_string_node_t *node, pm_node_t *part) {
pm_node_list_append(&node->parts, part);
node->base.location.end = part->location.end;
}
static inline void
pm_interpolated_xstring_node_closing_set(pm_interpolated_x_string_node_t *node, const pm_token_t *closing) {
node->closing_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(closing);
node->base.location.end = closing->end;
}
/**
* Allocate a new KeywordHashNode node.
*/
static pm_keyword_hash_node_t *
pm_keyword_hash_node_create(pm_parser_t *parser) {
pm_keyword_hash_node_t *node = PM_ALLOC_NODE(parser, pm_keyword_hash_node_t);
*node = (pm_keyword_hash_node_t) {
.base = {
.type = PM_KEYWORD_HASH_NODE,
.location = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE,
.flags = PM_KEYWORD_HASH_NODE_FLAGS_SYMBOL_KEYS
},
.elements = { 0 }
};
return node;
}
/**
* Append an element to a KeywordHashNode node.
*/
static void
pm_keyword_hash_node_elements_append(pm_keyword_hash_node_t *hash, pm_node_t *element) {
// If the element being added is not an AssocNode or does not have a symbol key, then
// we want to turn the STATIC_KEYS flag off.
// TODO: Rename the flag to SYMBOL_KEYS instead.
if (!PM_NODE_TYPE_P(element, PM_ASSOC_NODE) || !PM_NODE_TYPE_P(((pm_assoc_node_t *) element)->key, PM_SYMBOL_NODE)) {
pm_node_flag_unset((pm_node_t *)hash, PM_KEYWORD_HASH_NODE_FLAGS_SYMBOL_KEYS);
}
pm_node_list_append(&hash->elements, element);
if (hash->base.location.start == NULL) {
hash->base.location.start = element->location.start;
}
hash->base.location.end = element->location.end;
}
/**
* Allocate and initialize a new RequiredKeywordParameterNode node.
*/
static pm_required_keyword_parameter_node_t *
pm_required_keyword_parameter_node_create(pm_parser_t *parser, const pm_token_t *name) {
pm_required_keyword_parameter_node_t *node = PM_ALLOC_NODE(parser, pm_required_keyword_parameter_node_t);
*node = (pm_required_keyword_parameter_node_t) {
{
.type = PM_REQUIRED_KEYWORD_PARAMETER_NODE,
.location = {
.start = name->start,
.end = name->end
},
},
.name = pm_parser_constant_id_location(parser, name->start, name->end - 1),
.name_loc = PM_LOCATION_TOKEN_VALUE(name),
};
return node;
}
/**
* Allocate a new OptionalKeywordParameterNode node.
*/
static pm_optional_keyword_parameter_node_t *
pm_optional_keyword_parameter_node_create(pm_parser_t *parser, const pm_token_t *name, pm_node_t *value) {
pm_optional_keyword_parameter_node_t *node = PM_ALLOC_NODE(parser, pm_optional_keyword_parameter_node_t);
*node = (pm_optional_keyword_parameter_node_t) {
{
.type = PM_OPTIONAL_KEYWORD_PARAMETER_NODE,
.location = {
.start = name->start,
.end = value->location.end
},
},
.name = pm_parser_constant_id_location(parser, name->start, name->end - 1),
.name_loc = PM_LOCATION_TOKEN_VALUE(name),
.value = value
};
return node;
}
/**
* Allocate a new KeywordRestParameterNode node.
*/
static pm_keyword_rest_parameter_node_t *
pm_keyword_rest_parameter_node_create(pm_parser_t *parser, const pm_token_t *operator, const pm_token_t *name) {
pm_keyword_rest_parameter_node_t *node = PM_ALLOC_NODE(parser, pm_keyword_rest_parameter_node_t);
*node = (pm_keyword_rest_parameter_node_t) {
{
.type = PM_KEYWORD_REST_PARAMETER_NODE,
.location = {
.start = operator->start,
.end = (name->type == PM_TOKEN_NOT_PROVIDED ? operator->end : name->end)
},
},
.name = pm_parser_optional_constant_id_token(parser, name),
.name_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(name),
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator)
};
return node;
}
/**
* Allocate a new LambdaNode node.
*/
static pm_lambda_node_t *
pm_lambda_node_create(
pm_parser_t *parser,
pm_constant_id_list_t *locals,
uint32_t locals_body_index,
const pm_token_t *operator,
const pm_token_t *opening,
const pm_token_t *closing,
pm_node_t *parameters,
pm_node_t *body
) {
pm_lambda_node_t *node = PM_ALLOC_NODE(parser, pm_lambda_node_t);
*node = (pm_lambda_node_t) {
{
.type = PM_LAMBDA_NODE,
.location = {
.start = operator->start,
.end = closing->end
},
},
.locals = *locals,
.locals_body_index = locals_body_index,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.opening_loc = PM_LOCATION_TOKEN_VALUE(opening),
.closing_loc = PM_LOCATION_TOKEN_VALUE(closing),
.parameters = parameters,
.body = body
};
return node;
}
/**
* Allocate and initialize a new LocalVariableAndWriteNode node.
*/
static pm_local_variable_and_write_node_t *
pm_local_variable_and_write_node_create(pm_parser_t *parser, pm_node_t *target, const pm_token_t *operator, pm_node_t *value, pm_constant_id_t name, uint32_t depth) {
assert(PM_NODE_TYPE_P(target, PM_LOCAL_VARIABLE_READ_NODE) || PM_NODE_TYPE_P(target, PM_CALL_NODE));
assert(operator->type == PM_TOKEN_AMPERSAND_AMPERSAND_EQUAL);
pm_local_variable_and_write_node_t *node = PM_ALLOC_NODE(parser, pm_local_variable_and_write_node_t);
*node = (pm_local_variable_and_write_node_t) {
{
.type = PM_LOCAL_VARIABLE_AND_WRITE_NODE,
.location = {
.start = target->location.start,
.end = value->location.end
}
},
.name_loc = target->location,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value,
.name = name,
.depth = depth
};
return node;
}
/**
* Allocate and initialize a new LocalVariableOperatorWriteNode node.
*/
static pm_local_variable_operator_write_node_t *
pm_local_variable_operator_write_node_create(pm_parser_t *parser, pm_node_t *target, const pm_token_t *operator, pm_node_t *value, pm_constant_id_t name, uint32_t depth) {
pm_local_variable_operator_write_node_t *node = PM_ALLOC_NODE(parser, pm_local_variable_operator_write_node_t);
*node = (pm_local_variable_operator_write_node_t) {
{
.type = PM_LOCAL_VARIABLE_OPERATOR_WRITE_NODE,
.location = {
.start = target->location.start,
.end = value->location.end
}
},
.name_loc = target->location,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value,
.name = name,
.operator = pm_parser_constant_id_location(parser, operator->start, operator->end - 1),
.depth = depth
};
return node;
}
/**
* Allocate and initialize a new LocalVariableOrWriteNode node.
*/
static pm_local_variable_or_write_node_t *
pm_local_variable_or_write_node_create(pm_parser_t *parser, pm_node_t *target, const pm_token_t *operator, pm_node_t *value, pm_constant_id_t name, uint32_t depth) {
assert(PM_NODE_TYPE_P(target, PM_LOCAL_VARIABLE_READ_NODE) || PM_NODE_TYPE_P(target, PM_CALL_NODE));
assert(operator->type == PM_TOKEN_PIPE_PIPE_EQUAL);
pm_local_variable_or_write_node_t *node = PM_ALLOC_NODE(parser, pm_local_variable_or_write_node_t);
*node = (pm_local_variable_or_write_node_t) {
{
.type = PM_LOCAL_VARIABLE_OR_WRITE_NODE,
.location = {
.start = target->location.start,
.end = value->location.end
}
},
.name_loc = target->location,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value,
.name = name,
.depth = depth
};
return node;
}
/**
* Allocate a new LocalVariableReadNode node.
*/
static pm_local_variable_read_node_t *
pm_local_variable_read_node_create(pm_parser_t *parser, const pm_token_t *name, uint32_t depth) {
pm_constant_id_t name_id = pm_parser_constant_id_token(parser, name);
if (parser->current_param_name == name_id) {
pm_parser_err_token(parser, name, PM_ERR_PARAMETER_CIRCULAR);
}
pm_local_variable_read_node_t *node = PM_ALLOC_NODE(parser, pm_local_variable_read_node_t);
*node = (pm_local_variable_read_node_t) {
{
.type = PM_LOCAL_VARIABLE_READ_NODE,
.location = PM_LOCATION_TOKEN_VALUE(name)
},
.name = name_id,
.depth = depth
};
return node;
}
/**
* Allocate and initialize a new LocalVariableWriteNode node.
*/
static pm_local_variable_write_node_t *
pm_local_variable_write_node_create(pm_parser_t *parser, pm_constant_id_t name, uint32_t depth, pm_node_t *value, const pm_location_t *name_loc, const pm_token_t *operator) {
pm_local_variable_write_node_t *node = PM_ALLOC_NODE(parser, pm_local_variable_write_node_t);
*node = (pm_local_variable_write_node_t) {
{
.type = PM_LOCAL_VARIABLE_WRITE_NODE,
.location = {
.start = name_loc->start,
.end = value->location.end
}
},
.name = name,
.depth = depth,
.value = value,
.name_loc = *name_loc,
.operator_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(operator)
};
return node;
}
/**
* Returns true if the given bounds comprise a numbered parameter (i.e., they
* are of the form /^_\d$/).
*/
static inline bool
pm_token_is_numbered_parameter(const uint8_t *start, const uint8_t *end) {
return (end - start == 2) && (start[0] == '_') && (start[1] != '0') && (pm_char_is_decimal_digit(start[1]));
}
/**
* Ensure the given bounds do not comprise a numbered parameter. If they do, add
* an appropriate error message to the parser.
*/
static inline void
pm_refute_numbered_parameter(pm_parser_t *parser, const uint8_t *start, const uint8_t *end) {
if (pm_token_is_numbered_parameter(start, end)) {
PM_PARSER_ERR_FORMAT(parser, start, end, PM_ERR_PARAMETER_NUMBERED_RESERVED, start);
}
}
/**
* Allocate and initialize a new LocalVariableTargetNode node with the given
* name and depth.
*/
static pm_local_variable_target_node_t *
pm_local_variable_target_node_create_values(pm_parser_t *parser, const pm_location_t *location, pm_constant_id_t name, uint32_t depth) {
pm_local_variable_target_node_t *node = PM_ALLOC_NODE(parser, pm_local_variable_target_node_t);
*node = (pm_local_variable_target_node_t) {
{
.type = PM_LOCAL_VARIABLE_TARGET_NODE,
.location = *location
},
.name = name,
.depth = depth
};
return node;
}
/**
* Allocate and initialize a new LocalVariableTargetNode node.
*/
static pm_local_variable_target_node_t *
pm_local_variable_target_node_create(pm_parser_t *parser, const pm_token_t *name) {
pm_refute_numbered_parameter(parser, name->start, name->end);
return pm_local_variable_target_node_create_values(
parser,
&(pm_location_t) { .start = name->start, .end = name->end },
pm_parser_constant_id_token(parser, name),
0
);
}
/**
* Allocate and initialize a new LocalVariableTargetNode node with the given depth.
*/
static pm_local_variable_target_node_t *
pm_local_variable_target_node_create_depth(pm_parser_t *parser, const pm_token_t *name, uint32_t depth) {
pm_refute_numbered_parameter(parser, name->start, name->end);
return pm_local_variable_target_node_create_values(
parser,
&(pm_location_t) { .start = name->start, .end = name->end },
pm_parser_constant_id_token(parser, name),
depth
);
}
/**
* Allocate and initialize a new MatchPredicateNode node.
*/
static pm_match_predicate_node_t *
pm_match_predicate_node_create(pm_parser_t *parser, pm_node_t *value, pm_node_t *pattern, const pm_token_t *operator) {
pm_assert_value_expression(parser, value);
pm_match_predicate_node_t *node = PM_ALLOC_NODE(parser, pm_match_predicate_node_t);
*node = (pm_match_predicate_node_t) {
{
.type = PM_MATCH_PREDICATE_NODE,
.location = {
.start = value->location.start,
.end = pattern->location.end
}
},
.value = value,
.pattern = pattern,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator)
};
return node;
}
/**
* Allocate and initialize a new MatchRequiredNode node.
*/
static pm_match_required_node_t *
pm_match_required_node_create(pm_parser_t *parser, pm_node_t *value, pm_node_t *pattern, const pm_token_t *operator) {
pm_assert_value_expression(parser, value);
pm_match_required_node_t *node = PM_ALLOC_NODE(parser, pm_match_required_node_t);
*node = (pm_match_required_node_t) {
{
.type = PM_MATCH_REQUIRED_NODE,
.location = {
.start = value->location.start,
.end = pattern->location.end
}
},
.value = value,
.pattern = pattern,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator)
};
return node;
}
/**
* Allocate and initialize a new MatchWriteNode node.
*/
static pm_match_write_node_t *
pm_match_write_node_create(pm_parser_t *parser, pm_call_node_t *call) {
pm_match_write_node_t *node = PM_ALLOC_NODE(parser, pm_match_write_node_t);
*node = (pm_match_write_node_t) {
{
.type = PM_MATCH_WRITE_NODE,
.location = call->base.location
},
.call = call,
.targets = { 0 }
};
return node;
}
/**
* Allocate a new ModuleNode node.
*/
static pm_module_node_t *
pm_module_node_create(pm_parser_t *parser, pm_constant_id_list_t *locals, const pm_token_t *module_keyword, pm_node_t *constant_path, const pm_token_t *name, pm_node_t *body, const pm_token_t *end_keyword) {
pm_module_node_t *node = PM_ALLOC_NODE(parser, pm_module_node_t);
*node = (pm_module_node_t) {
{
.type = PM_MODULE_NODE,
.location = {
.start = module_keyword->start,
.end = end_keyword->end
}
},
.locals = (locals == NULL ? ((pm_constant_id_list_t) { .ids = NULL, .size = 0, .capacity = 0 }) : *locals),
.module_keyword_loc = PM_LOCATION_TOKEN_VALUE(module_keyword),
.constant_path = constant_path,
.body = body,
.end_keyword_loc = PM_LOCATION_TOKEN_VALUE(end_keyword),
.name = pm_parser_constant_id_token(parser, name)
};
return node;
}
/**
* Allocate and initialize new MultiTargetNode node.
*/
static pm_multi_target_node_t *
pm_multi_target_node_create(pm_parser_t *parser) {
pm_multi_target_node_t *node = PM_ALLOC_NODE(parser, pm_multi_target_node_t);
*node = (pm_multi_target_node_t) {
{
.type = PM_MULTI_TARGET_NODE,
.location = { .start = NULL, .end = NULL }
},
.lefts = { 0 },
.rest = NULL,
.rights = { 0 },
.lparen_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE,
.rparen_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE
};
return node;
}
/**
* Append a target to a MultiTargetNode node.
*/
static void
pm_multi_target_node_targets_append(pm_parser_t *parser, pm_multi_target_node_t *node, pm_node_t *target) {
if (PM_NODE_TYPE_P(target, PM_SPLAT_NODE) || PM_NODE_TYPE_P(target, PM_IMPLICIT_REST_NODE)) {
if (node->rest == NULL) {
node->rest = target;
} else {
pm_parser_err_node(parser, target, PM_ERR_MULTI_ASSIGN_MULTI_SPLATS);
pm_node_list_append(&node->rights, target);
}
} else if (node->rest == NULL) {
pm_node_list_append(&node->lefts, target);
} else {
pm_node_list_append(&node->rights, target);
}
if (node->base.location.start == NULL || (node->base.location.start > target->location.start)) {
node->base.location.start = target->location.start;
}
if (node->base.location.end == NULL || (node->base.location.end < target->location.end)) {
node->base.location.end = target->location.end;
}
}
/**
* Set the opening of a MultiTargetNode node.
*/
static void
pm_multi_target_node_opening_set(pm_multi_target_node_t *node, const pm_token_t *lparen) {
node->base.location.start = lparen->start;
node->lparen_loc = PM_LOCATION_TOKEN_VALUE(lparen);
}
/**
* Set the closing of a MultiTargetNode node.
*/
static void
pm_multi_target_node_closing_set(pm_multi_target_node_t *node, const pm_token_t *rparen) {
node->base.location.end = rparen->end;
node->rparen_loc = PM_LOCATION_TOKEN_VALUE(rparen);
}
/**
* Allocate a new MultiWriteNode node.
*/
static pm_multi_write_node_t *
pm_multi_write_node_create(pm_parser_t *parser, pm_multi_target_node_t *target, const pm_token_t *operator, pm_node_t *value) {
pm_multi_write_node_t *node = PM_ALLOC_NODE(parser, pm_multi_write_node_t);
*node = (pm_multi_write_node_t) {
{
.type = PM_MULTI_WRITE_NODE,
.location = {
.start = target->base.location.start,
.end = value->location.end
}
},
.lefts = target->lefts,
.rest = target->rest,
.rights = target->rights,
.lparen_loc = target->lparen_loc,
.rparen_loc = target->rparen_loc,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value
};
// Explicitly do not call pm_node_destroy here because we want to keep
// around all of the information within the MultiWriteNode node.
free(target);
return node;
}
/**
* Allocate and initialize a new NextNode node.
*/
static pm_next_node_t *
pm_next_node_create(pm_parser_t *parser, const pm_token_t *keyword, pm_arguments_node_t *arguments) {
assert(keyword->type == PM_TOKEN_KEYWORD_NEXT);
pm_next_node_t *node = PM_ALLOC_NODE(parser, pm_next_node_t);
*node = (pm_next_node_t) {
{
.type = PM_NEXT_NODE,
.location = {
.start = keyword->start,
.end = (arguments == NULL ? keyword->end : arguments->base.location.end)
}
},
.keyword_loc = PM_LOCATION_TOKEN_VALUE(keyword),
.arguments = arguments
};
return node;
}
/**
* Allocate and initialize a new NilNode node.
*/
static pm_nil_node_t *
pm_nil_node_create(pm_parser_t *parser, const pm_token_t *token) {
assert(token->type == PM_TOKEN_KEYWORD_NIL);
pm_nil_node_t *node = PM_ALLOC_NODE(parser, pm_nil_node_t);
*node = (pm_nil_node_t) {{
.type = PM_NIL_NODE,
.flags = PM_NODE_FLAG_STATIC_LITERAL,
.location = PM_LOCATION_TOKEN_VALUE(token)
}};
return node;
}
/**
* Allocate and initialize a new NoKeywordsParameterNode node.
*/
static pm_no_keywords_parameter_node_t *
pm_no_keywords_parameter_node_create(pm_parser_t *parser, const pm_token_t *operator, const pm_token_t *keyword) {
assert(operator->type == PM_TOKEN_USTAR_STAR || operator->type == PM_TOKEN_STAR_STAR);
assert(keyword->type == PM_TOKEN_KEYWORD_NIL);
pm_no_keywords_parameter_node_t *node = PM_ALLOC_NODE(parser, pm_no_keywords_parameter_node_t);
*node = (pm_no_keywords_parameter_node_t) {
{
.type = PM_NO_KEYWORDS_PARAMETER_NODE,
.location = {
.start = operator->start,
.end = keyword->end
}
},
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.keyword_loc = PM_LOCATION_TOKEN_VALUE(keyword)
};
return node;
}
/**
* Allocate and initialize a new NumberedParametersNode node.
*/
static pm_numbered_parameters_node_t *
pm_numbered_parameters_node_create(pm_parser_t *parser, const pm_location_t *location, uint8_t maximum) {
pm_numbered_parameters_node_t *node = PM_ALLOC_NODE(parser, pm_numbered_parameters_node_t);
*node = (pm_numbered_parameters_node_t) {
{
.type = PM_NUMBERED_PARAMETERS_NODE,
.location = *location
},
.maximum = maximum
};
return node;
}
/**
* Allocate and initialize a new NthReferenceReadNode node.
*/
static pm_numbered_reference_read_node_t *
pm_numbered_reference_read_node_create(pm_parser_t *parser, const pm_token_t *name) {
assert(name->type == PM_TOKEN_NUMBERED_REFERENCE);
pm_numbered_reference_read_node_t *node = PM_ALLOC_NODE(parser, pm_numbered_reference_read_node_t);
*node = (pm_numbered_reference_read_node_t) {
{
.type = PM_NUMBERED_REFERENCE_READ_NODE,
.location = PM_LOCATION_TOKEN_VALUE(name),
},
.number = parse_decimal_number(parser, name->start + 1, name->end)
};
return node;
}
/**
* Allocate a new OptionalParameterNode node.
*/
static pm_optional_parameter_node_t *
pm_optional_parameter_node_create(pm_parser_t *parser, const pm_token_t *name, const pm_token_t *operator, pm_node_t *value) {
pm_optional_parameter_node_t *node = PM_ALLOC_NODE(parser, pm_optional_parameter_node_t);
*node = (pm_optional_parameter_node_t) {
{
.type = PM_OPTIONAL_PARAMETER_NODE,
.location = {
.start = name->start,
.end = value->location.end
}
},
.name = pm_parser_constant_id_token(parser, name),
.name_loc = PM_LOCATION_TOKEN_VALUE(name),
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.value = value
};
return node;
}
/**
* Allocate and initialize a new OrNode node.
*/
static pm_or_node_t *
pm_or_node_create(pm_parser_t *parser, pm_node_t *left, const pm_token_t *operator, pm_node_t *right) {
pm_assert_value_expression(parser, left);
pm_or_node_t *node = PM_ALLOC_NODE(parser, pm_or_node_t);
*node = (pm_or_node_t) {
{
.type = PM_OR_NODE,
.location = {
.start = left->location.start,
.end = right->location.end
}
},
.left = left,
.right = right,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator)
};
return node;
}
/**
* Allocate and initialize a new ParametersNode node.
*/
static pm_parameters_node_t *
pm_parameters_node_create(pm_parser_t *parser) {
pm_parameters_node_t *node = PM_ALLOC_NODE(parser, pm_parameters_node_t);
*node = (pm_parameters_node_t) {
{
.type = PM_PARAMETERS_NODE,
.location = PM_LOCATION_TOKEN_VALUE(&parser->current)
},
.rest = NULL,
.keyword_rest = NULL,
.block = NULL,
.requireds = { 0 },
.optionals = { 0 },
.posts = { 0 },
.keywords = { 0 }
};
return node;
}
/**
* Set the location properly for the parameters node.
*/
static void
pm_parameters_node_location_set(pm_parameters_node_t *params, pm_node_t *param) {
if (params->base.location.start == NULL) {
params->base.location.start = param->location.start;
} else {
params->base.location.start = params->base.location.start < param->location.start ? params->base.location.start : param->location.start;
}
if (params->base.location.end == NULL) {
params->base.location.end = param->location.end;
} else {
params->base.location.end = params->base.location.end > param->location.end ? params->base.location.end : param->location.end;
}
}
/**
* Append a required parameter to a ParametersNode node.
*/
static void
pm_parameters_node_requireds_append(pm_parameters_node_t *params, pm_node_t *param) {
pm_parameters_node_location_set(params, param);
pm_node_list_append(¶ms->requireds, param);
}
/**
* Append an optional parameter to a ParametersNode node.
*/
static void
pm_parameters_node_optionals_append(pm_parameters_node_t *params, pm_optional_parameter_node_t *param) {
pm_parameters_node_location_set(params, (pm_node_t *) param);
pm_node_list_append(¶ms->optionals, (pm_node_t *) param);
}
/**
* Append a post optional arguments parameter to a ParametersNode node.
*/
static void
pm_parameters_node_posts_append(pm_parameters_node_t *params, pm_node_t *param) {
pm_parameters_node_location_set(params, param);
pm_node_list_append(¶ms->posts, param);
}
/**
* Set the rest parameter on a ParametersNode node.
*/
static void
pm_parameters_node_rest_set(pm_parameters_node_t *params, pm_node_t *param) {
pm_parameters_node_location_set(params, param);
params->rest = param;
}
/**
* Append a keyword parameter to a ParametersNode node.
*/
static void
pm_parameters_node_keywords_append(pm_parameters_node_t *params, pm_node_t *param) {
pm_parameters_node_location_set(params, param);
pm_node_list_append(¶ms->keywords, param);
}
/**
* Set the keyword rest parameter on a ParametersNode node.
*/
static void
pm_parameters_node_keyword_rest_set(pm_parameters_node_t *params, pm_node_t *param) {
assert(params->keyword_rest == NULL);
pm_parameters_node_location_set(params, param);
params->keyword_rest = param;
}
/**
* Set the block parameter on a ParametersNode node.
*/
static void
pm_parameters_node_block_set(pm_parameters_node_t *params, pm_block_parameter_node_t *param) {
assert(params->block == NULL);
pm_parameters_node_location_set(params, (pm_node_t *) param);
params->block = param;
}
/**
* Allocate a new ProgramNode node.
*/
static pm_program_node_t *
pm_program_node_create(pm_parser_t *parser, pm_constant_id_list_t *locals, pm_statements_node_t *statements) {
pm_program_node_t *node = PM_ALLOC_NODE(parser, pm_program_node_t);
*node = (pm_program_node_t) {
{
.type = PM_PROGRAM_NODE,
.location = {
.start = statements == NULL ? parser->start : statements->base.location.start,
.end = statements == NULL ? parser->end : statements->base.location.end
}
},
.locals = *locals,
.statements = statements
};
return node;
}
/**
* Allocate and initialize new ParenthesesNode node.
*/
static pm_parentheses_node_t *
pm_parentheses_node_create(pm_parser_t *parser, const pm_token_t *opening, pm_node_t *body, const pm_token_t *closing) {
pm_parentheses_node_t *node = PM_ALLOC_NODE(parser, pm_parentheses_node_t);
*node = (pm_parentheses_node_t) {
{
.type = PM_PARENTHESES_NODE,
.location = {
.start = opening->start,
.end = closing->end
}
},
.body = body,
.opening_loc = PM_LOCATION_TOKEN_VALUE(opening),
.closing_loc = PM_LOCATION_TOKEN_VALUE(closing)
};
return node;
}
/**
* Allocate and initialize a new PinnedExpressionNode node.
*/
static pm_pinned_expression_node_t *
pm_pinned_expression_node_create(pm_parser_t *parser, pm_node_t *expression, const pm_token_t *operator, const pm_token_t *lparen, const pm_token_t *rparen) {
pm_pinned_expression_node_t *node = PM_ALLOC_NODE(parser, pm_pinned_expression_node_t);
*node = (pm_pinned_expression_node_t) {
{
.type = PM_PINNED_EXPRESSION_NODE,
.location = {
.start = operator->start,
.end = rparen->end
}
},
.expression = expression,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.lparen_loc = PM_LOCATION_TOKEN_VALUE(lparen),
.rparen_loc = PM_LOCATION_TOKEN_VALUE(rparen)
};
return node;
}
/**
* Allocate and initialize a new PinnedVariableNode node.
*/
static pm_pinned_variable_node_t *
pm_pinned_variable_node_create(pm_parser_t *parser, const pm_token_t *operator, pm_node_t *variable) {
pm_pinned_variable_node_t *node = PM_ALLOC_NODE(parser, pm_pinned_variable_node_t);
*node = (pm_pinned_variable_node_t) {
{
.type = PM_PINNED_VARIABLE_NODE,
.location = {
.start = operator->start,
.end = variable->location.end
}
},
.variable = variable,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator)
};
return node;
}
/**
* Allocate and initialize a new PostExecutionNode node.
*/
static pm_post_execution_node_t *
pm_post_execution_node_create(pm_parser_t *parser, const pm_token_t *keyword, const pm_token_t *opening, pm_statements_node_t *statements, const pm_token_t *closing) {
pm_post_execution_node_t *node = PM_ALLOC_NODE(parser, pm_post_execution_node_t);
*node = (pm_post_execution_node_t) {
{
.type = PM_POST_EXECUTION_NODE,
.location = {
.start = keyword->start,
.end = closing->end
}
},
.statements = statements,
.keyword_loc = PM_LOCATION_TOKEN_VALUE(keyword),
.opening_loc = PM_LOCATION_TOKEN_VALUE(opening),
.closing_loc = PM_LOCATION_TOKEN_VALUE(closing)
};
return node;
}
/**
* Allocate and initialize a new PreExecutionNode node.
*/
static pm_pre_execution_node_t *
pm_pre_execution_node_create(pm_parser_t *parser, const pm_token_t *keyword, const pm_token_t *opening, pm_statements_node_t *statements, const pm_token_t *closing) {
pm_pre_execution_node_t *node = PM_ALLOC_NODE(parser, pm_pre_execution_node_t);
*node = (pm_pre_execution_node_t) {
{
.type = PM_PRE_EXECUTION_NODE,
.location = {
.start = keyword->start,
.end = closing->end
}
},
.statements = statements,
.keyword_loc = PM_LOCATION_TOKEN_VALUE(keyword),
.opening_loc = PM_LOCATION_TOKEN_VALUE(opening),
.closing_loc = PM_LOCATION_TOKEN_VALUE(closing)
};
return node;
}
/**
* Allocate and initialize new RangeNode node.
*/
static pm_range_node_t *
pm_range_node_create(pm_parser_t *parser, pm_node_t *left, const pm_token_t *operator, pm_node_t *right) {
pm_assert_value_expression(parser, left);
pm_assert_value_expression(parser, right);
pm_range_node_t *node = PM_ALLOC_NODE(parser, pm_range_node_t);
pm_node_flags_t flags = 0;
// Indicate that this node an exclusive range if the operator is `...`.
if (operator->type == PM_TOKEN_DOT_DOT_DOT || operator->type == PM_TOKEN_UDOT_DOT_DOT) {
flags |= PM_RANGE_FLAGS_EXCLUDE_END;
}
// Indicate that this node is a static literal (i.e., can be compiled with
// a putobject in CRuby) if the left and right are implicit nil, explicit
// nil, or integers.
if (
(left == NULL || PM_NODE_TYPE_P(left, PM_NIL_NODE) || PM_NODE_TYPE_P(left, PM_INTEGER_NODE)) &&
(right == NULL || PM_NODE_TYPE_P(right, PM_NIL_NODE) || PM_NODE_TYPE_P(right, PM_INTEGER_NODE))
) {
flags |= PM_NODE_FLAG_STATIC_LITERAL;
}
*node = (pm_range_node_t) {
{
.type = PM_RANGE_NODE,
.flags = flags,
.location = {
.start = (left == NULL ? operator->start : left->location.start),
.end = (right == NULL ? operator->end : right->location.end)
}
},
.left = left,
.right = right,
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator)
};
return node;
}
/**
* Allocate and initialize a new RedoNode node.
*/
static pm_redo_node_t *
pm_redo_node_create(pm_parser_t *parser, const pm_token_t *token) {
assert(token->type == PM_TOKEN_KEYWORD_REDO);
pm_redo_node_t *node = PM_ALLOC_NODE(parser, pm_redo_node_t);
*node = (pm_redo_node_t) {{ .type = PM_REDO_NODE, .location = PM_LOCATION_TOKEN_VALUE(token) }};
return node;
}
/**
* Allocate a new initialize a new RegularExpressionNode node with the given
* unescaped string.
*/
static pm_regular_expression_node_t *
pm_regular_expression_node_create_unescaped(pm_parser_t *parser, const pm_token_t *opening, const pm_token_t *content, const pm_token_t *closing, const pm_string_t *unescaped) {
pm_regular_expression_node_t *node = PM_ALLOC_NODE(parser, pm_regular_expression_node_t);
*node = (pm_regular_expression_node_t) {
{
.type = PM_REGULAR_EXPRESSION_NODE,
.flags = pm_regular_expression_flags_create(closing) | PM_NODE_FLAG_STATIC_LITERAL,
.location = {
.start = MIN(opening->start, closing->start),
.end = MAX(opening->end, closing->end)
}
},
.opening_loc = PM_LOCATION_TOKEN_VALUE(opening),
.content_loc = PM_LOCATION_TOKEN_VALUE(content),
.closing_loc = PM_LOCATION_TOKEN_VALUE(closing),
.unescaped = *unescaped
};
return node;
}
/**
* Allocate a new initialize a new RegularExpressionNode node.
*/
static inline pm_regular_expression_node_t *
pm_regular_expression_node_create(pm_parser_t *parser, const pm_token_t *opening, const pm_token_t *content, const pm_token_t *closing) {
return pm_regular_expression_node_create_unescaped(parser, opening, content, closing, &PM_STRING_EMPTY);
}
/**
* Allocate a new RequiredParameterNode node.
*/
static pm_required_parameter_node_t *
pm_required_parameter_node_create(pm_parser_t *parser, const pm_token_t *token) {
pm_required_parameter_node_t *node = PM_ALLOC_NODE(parser, pm_required_parameter_node_t);
*node = (pm_required_parameter_node_t) {
{
.type = PM_REQUIRED_PARAMETER_NODE,
.location = PM_LOCATION_TOKEN_VALUE(token)
},
.name = pm_parser_constant_id_token(parser, token)
};
return node;
}
/**
* Allocate a new RescueModifierNode node.
*/
static pm_rescue_modifier_node_t *
pm_rescue_modifier_node_create(pm_parser_t *parser, pm_node_t *expression, const pm_token_t *keyword, pm_node_t *rescue_expression) {
pm_rescue_modifier_node_t *node = PM_ALLOC_NODE(parser, pm_rescue_modifier_node_t);
*node = (pm_rescue_modifier_node_t) {
{
.type = PM_RESCUE_MODIFIER_NODE,
.location = {
.start = expression->location.start,
.end = rescue_expression->location.end
}
},
.expression = expression,
.keyword_loc = PM_LOCATION_TOKEN_VALUE(keyword),
.rescue_expression = rescue_expression
};
return node;
}
/**
* Allocate and initiliaze a new RescueNode node.
*/
static pm_rescue_node_t *
pm_rescue_node_create(pm_parser_t *parser, const pm_token_t *keyword) {
pm_rescue_node_t *node = PM_ALLOC_NODE(parser, pm_rescue_node_t);
*node = (pm_rescue_node_t) {
{
.type = PM_RESCUE_NODE,
.location = PM_LOCATION_TOKEN_VALUE(keyword)
},
.keyword_loc = PM_LOCATION_TOKEN_VALUE(keyword),
.operator_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE,
.reference = NULL,
.statements = NULL,
.consequent = NULL,
.exceptions = { 0 }
};
return node;
}
static inline void
pm_rescue_node_operator_set(pm_rescue_node_t *node, const pm_token_t *operator) {
node->operator_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(operator);
}
/**
* Set the reference of a rescue node, and update the location of the node.
*/
static void
pm_rescue_node_reference_set(pm_rescue_node_t *node, pm_node_t *reference) {
node->reference = reference;
node->base.location.end = reference->location.end;
}
/**
* Set the statements of a rescue node, and update the location of the node.
*/
static void
pm_rescue_node_statements_set(pm_rescue_node_t *node, pm_statements_node_t *statements) {
node->statements = statements;
if (pm_statements_node_body_length(statements) > 0) {
node->base.location.end = statements->base.location.end;
}
}
/**
* Set the consequent of a rescue node, and update the location.
*/
static void
pm_rescue_node_consequent_set(pm_rescue_node_t *node, pm_rescue_node_t *consequent) {
node->consequent = consequent;
node->base.location.end = consequent->base.location.end;
}
/**
* Append an exception node to a rescue node, and update the location.
*/
static void
pm_rescue_node_exceptions_append(pm_rescue_node_t *node, pm_node_t *exception) {
pm_node_list_append(&node->exceptions, exception);
node->base.location.end = exception->location.end;
}
/**
* Allocate a new RestParameterNode node.
*/
static pm_rest_parameter_node_t *
pm_rest_parameter_node_create(pm_parser_t *parser, const pm_token_t *operator, const pm_token_t *name) {
pm_rest_parameter_node_t *node = PM_ALLOC_NODE(parser, pm_rest_parameter_node_t);
*node = (pm_rest_parameter_node_t) {
{
.type = PM_REST_PARAMETER_NODE,
.location = {
.start = operator->start,
.end = (name->type == PM_TOKEN_NOT_PROVIDED ? operator->end : name->end)
}
},
.name = pm_parser_optional_constant_id_token(parser, name),
.name_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(name),
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator)
};
return node;
}
/**
* Allocate and initialize a new RetryNode node.
*/
static pm_retry_node_t *
pm_retry_node_create(pm_parser_t *parser, const pm_token_t *token) {
assert(token->type == PM_TOKEN_KEYWORD_RETRY);
pm_retry_node_t *node = PM_ALLOC_NODE(parser, pm_retry_node_t);
*node = (pm_retry_node_t) {{ .type = PM_RETRY_NODE, .location = PM_LOCATION_TOKEN_VALUE(token) }};
return node;
}
/**
* Allocate a new ReturnNode node.
*/
static pm_return_node_t *
pm_return_node_create(pm_parser_t *parser, const pm_token_t *keyword, pm_arguments_node_t *arguments) {
pm_return_node_t *node = PM_ALLOC_NODE(parser, pm_return_node_t);
*node = (pm_return_node_t) {
{
.type = PM_RETURN_NODE,
.location = {
.start = keyword->start,
.end = (arguments == NULL ? keyword->end : arguments->base.location.end)
}
},
.keyword_loc = PM_LOCATION_TOKEN_VALUE(keyword),
.arguments = arguments
};
return node;
}
/**
* Allocate and initialize a new SelfNode node.
*/
static pm_self_node_t *
pm_self_node_create(pm_parser_t *parser, const pm_token_t *token) {
assert(token->type == PM_TOKEN_KEYWORD_SELF);
pm_self_node_t *node = PM_ALLOC_NODE(parser, pm_self_node_t);
*node = (pm_self_node_t) {{
.type = PM_SELF_NODE,
.location = PM_LOCATION_TOKEN_VALUE(token)
}};
return node;
}
/**
* Allocate a new SingletonClassNode node.
*/
static pm_singleton_class_node_t *
pm_singleton_class_node_create(pm_parser_t *parser, pm_constant_id_list_t *locals, const pm_token_t *class_keyword, const pm_token_t *operator, pm_node_t *expression, pm_node_t *body, const pm_token_t *end_keyword) {
pm_singleton_class_node_t *node = PM_ALLOC_NODE(parser, pm_singleton_class_node_t);
*node = (pm_singleton_class_node_t) {
{
.type = PM_SINGLETON_CLASS_NODE,
.location = {
.start = class_keyword->start,
.end = end_keyword->end
}
},
.locals = *locals,
.class_keyword_loc = PM_LOCATION_TOKEN_VALUE(class_keyword),
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.expression = expression,
.body = body,
.end_keyword_loc = PM_LOCATION_TOKEN_VALUE(end_keyword)
};
return node;
}
/**
* Allocate and initialize a new SourceEncodingNode node.
*/
static pm_source_encoding_node_t *
pm_source_encoding_node_create(pm_parser_t *parser, const pm_token_t *token) {
assert(token->type == PM_TOKEN_KEYWORD___ENCODING__);
pm_source_encoding_node_t *node = PM_ALLOC_NODE(parser, pm_source_encoding_node_t);
*node = (pm_source_encoding_node_t) {{
.type = PM_SOURCE_ENCODING_NODE,
.flags = PM_NODE_FLAG_STATIC_LITERAL,
.location = PM_LOCATION_TOKEN_VALUE(token)
}};
return node;
}
/**
* Allocate and initialize a new SourceFileNode node.
*/
static pm_source_file_node_t*
pm_source_file_node_create(pm_parser_t *parser, const pm_token_t *file_keyword) {
pm_source_file_node_t *node = PM_ALLOC_NODE(parser, pm_source_file_node_t);
assert(file_keyword->type == PM_TOKEN_KEYWORD___FILE__);
*node = (pm_source_file_node_t) {
{
.type = PM_SOURCE_FILE_NODE,
.flags = PM_NODE_FLAG_STATIC_LITERAL,
.location = PM_LOCATION_TOKEN_VALUE(file_keyword),
},
.filepath = parser->filepath_string,
};
return node;
}
/**
* Allocate and initialize a new SourceLineNode node.
*/
static pm_source_line_node_t *
pm_source_line_node_create(pm_parser_t *parser, const pm_token_t *token) {
assert(token->type == PM_TOKEN_KEYWORD___LINE__);
pm_source_line_node_t *node = PM_ALLOC_NODE(parser, pm_source_line_node_t);
*node = (pm_source_line_node_t) {{
.type = PM_SOURCE_LINE_NODE,
.flags = PM_NODE_FLAG_STATIC_LITERAL,
.location = PM_LOCATION_TOKEN_VALUE(token)
}};
return node;
}
/**
* Allocate a new SplatNode node.
*/
static pm_splat_node_t *
pm_splat_node_create(pm_parser_t *parser, const pm_token_t *operator, pm_node_t *expression) {
pm_splat_node_t *node = PM_ALLOC_NODE(parser, pm_splat_node_t);
*node = (pm_splat_node_t) {
{
.type = PM_SPLAT_NODE,
.location = {
.start = operator->start,
.end = (expression == NULL ? operator->end : expression->location.end)
}
},
.operator_loc = PM_LOCATION_TOKEN_VALUE(operator),
.expression = expression
};
return node;
}
/**
* Allocate and initialize a new StatementsNode node.
*/
static pm_statements_node_t *
pm_statements_node_create(pm_parser_t *parser) {
pm_statements_node_t *node = PM_ALLOC_NODE(parser, pm_statements_node_t);
*node = (pm_statements_node_t) {
{
.type = PM_STATEMENTS_NODE,
.location = PM_LOCATION_NULL_VALUE(parser)
},
.body = { 0 }
};
return node;
}
/**
* Get the length of the given StatementsNode node's body.
*/
static size_t
pm_statements_node_body_length(pm_statements_node_t *node) {
return node && node->body.size;
}
/**
* Set the location of the given StatementsNode.
*/
static void
pm_statements_node_location_set(pm_statements_node_t *node, const uint8_t *start, const uint8_t *end) {
node->base.location = (pm_location_t) { .start = start, .end = end };
}
/**
* Append a new node to the given StatementsNode node's body.
*/
static void
pm_statements_node_body_append(pm_statements_node_t *node, pm_node_t *statement) {
if (pm_statements_node_body_length(node) == 0 || statement->location.start < node->base.location.start) {
node->base.location.start = statement->location.start;
}
if (statement->location.end > node->base.location.end) {
node->base.location.end = statement->location.end;
}
pm_node_list_append(&node->body, statement);
// Every statement gets marked as a place where a newline can occur.
pm_node_flag_set(statement, PM_NODE_FLAG_NEWLINE);
}
/**
* Allocate a new StringNode node with the current string on the parser.
*/
static inline pm_string_node_t *
pm_string_node_create_unescaped(pm_parser_t *parser, const pm_token_t *opening, const pm_token_t *content, const pm_token_t *closing, const pm_string_t *string) {
pm_string_node_t *node = PM_ALLOC_NODE(parser, pm_string_node_t);
pm_node_flags_t flags = 0;
if (parser->frozen_string_literal) {
flags = PM_NODE_FLAG_STATIC_LITERAL | PM_STRING_FLAGS_FROZEN;
}
*node = (pm_string_node_t) {
{
.type = PM_STRING_NODE,
.flags = flags,
.location = {
.start = (opening->type == PM_TOKEN_NOT_PROVIDED ? content->start : opening->start),
.end = (closing->type == PM_TOKEN_NOT_PROVIDED ? content->end : closing->end)
}
},
.opening_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(opening),
.content_loc = PM_LOCATION_TOKEN_VALUE(content),
.closing_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(closing),
.unescaped = *string
};
return node;
}
/**
* Allocate a new StringNode node.
*/
static pm_string_node_t *
pm_string_node_create(pm_parser_t *parser, const pm_token_t *opening, const pm_token_t *content, const pm_token_t *closing) {
return pm_string_node_create_unescaped(parser, opening, content, closing, &PM_STRING_EMPTY);
}
/**
* Allocate a new StringNode node and create it using the current string on the
* parser.
*/
static pm_string_node_t *
pm_string_node_create_current_string(pm_parser_t *parser, const pm_token_t *opening, const pm_token_t *content, const pm_token_t *closing) {
pm_string_node_t *node = pm_string_node_create_unescaped(parser, opening, content, closing, &parser->current_string);
parser->current_string = PM_STRING_EMPTY;
return node;
}
/**
* Allocate and initialize a new SuperNode node.
*/
static pm_super_node_t *
pm_super_node_create(pm_parser_t *parser, const pm_token_t *keyword, pm_arguments_t *arguments) {
assert(keyword->type == PM_TOKEN_KEYWORD_SUPER);
pm_super_node_t *node = PM_ALLOC_NODE(parser, pm_super_node_t);
const uint8_t *end = pm_arguments_end(arguments);
if (end == NULL) {
assert(false && "unreachable");
}
*node = (pm_super_node_t) {
{
.type = PM_SUPER_NODE,
.location = {
.start = keyword->start,
.end = end,
}
},
.keyword_loc = PM_LOCATION_TOKEN_VALUE(keyword),
.lparen_loc = arguments->opening_loc,
.arguments = arguments->arguments,
.rparen_loc = arguments->closing_loc,
.block = arguments->block
};
return node;
}
/**
* Allocate and initialize a new SymbolNode node with the given unescaped
* string.
*/
static pm_symbol_node_t *
pm_symbol_node_create_unescaped(pm_parser_t *parser, const pm_token_t *opening, const pm_token_t *value, const pm_token_t *closing, const pm_string_t *unescaped) {
pm_symbol_node_t *node = PM_ALLOC_NODE(parser, pm_symbol_node_t);
*node = (pm_symbol_node_t) {
{
.type = PM_SYMBOL_NODE,
.flags = PM_NODE_FLAG_STATIC_LITERAL,
.location = {
.start = (opening->type == PM_TOKEN_NOT_PROVIDED ? value->start : opening->start),
.end = (closing->type == PM_TOKEN_NOT_PROVIDED ? value->end : closing->end)
}
},
.opening_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(opening),
.value_loc = PM_LOCATION_TOKEN_VALUE(value),
.closing_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(closing),
.unescaped = *unescaped
};
return node;
}
/**
* Allocate and initialize a new SymbolNode node.
*/
static inline pm_symbol_node_t *
pm_symbol_node_create(pm_parser_t *parser, const pm_token_t *opening, const pm_token_t *value, const pm_token_t *closing) {
return pm_symbol_node_create_unescaped(parser, opening, value, closing, &PM_STRING_EMPTY);
}
/**
* Allocate and initialize a new SymbolNode node with the current string.
*/
static pm_symbol_node_t *
pm_symbol_node_create_current_string(pm_parser_t *parser, const pm_token_t *opening, const pm_token_t *value, const pm_token_t *closing) {
pm_symbol_node_t *node = pm_symbol_node_create_unescaped(parser, opening, value, closing, &parser->current_string);
parser->current_string = PM_STRING_EMPTY;
return node;
}
/**
* Allocate and initialize a new SymbolNode node from a label.
*/
static pm_symbol_node_t *
pm_symbol_node_label_create(pm_parser_t *parser, const pm_token_t *token) {
pm_symbol_node_t *node;
switch (token->type) {
case PM_TOKEN_LABEL: {
pm_token_t opening = not_provided(parser);
pm_token_t closing = { .type = PM_TOKEN_LABEL_END, .start = token->end - 1, .end = token->end };
pm_token_t label = { .type = PM_TOKEN_LABEL, .start = token->start, .end = token->end - 1 };
node = pm_symbol_node_create(parser, &opening, &label, &closing);
assert((label.end - label.start) >= 0);
pm_string_shared_init(&node->unescaped, label.start, label.end);
break;
}
case PM_TOKEN_MISSING: {
pm_token_t opening = not_provided(parser);
pm_token_t closing = not_provided(parser);
pm_token_t label = { .type = PM_TOKEN_LABEL, .start = token->start, .end = token->end };
node = pm_symbol_node_create(parser, &opening, &label, &closing);
break;
}
default:
assert(false && "unreachable");
node = NULL;
break;
}
return node;
}
/**
* Check if the given node is a label in a hash.
*/
static bool
pm_symbol_node_label_p(pm_node_t *node) {
const uint8_t *end = NULL;
switch (PM_NODE_TYPE(node)) {
case PM_SYMBOL_NODE:
end = ((pm_symbol_node_t *) node)->closing_loc.end;
break;
case PM_INTERPOLATED_SYMBOL_NODE:
end = ((pm_interpolated_symbol_node_t *) node)->closing_loc.end;
break;
default:
return false;
}
return (end != NULL) && (end[-1] == ':');
}
/**
* Convert the given StringNode node to a SymbolNode node.
*/
static pm_symbol_node_t *
pm_string_node_to_symbol_node(pm_parser_t *parser, pm_string_node_t *node, const pm_token_t *opening, const pm_token_t *closing) {
pm_symbol_node_t *new_node = PM_ALLOC_NODE(parser, pm_symbol_node_t);
*new_node = (pm_symbol_node_t) {
{
.type = PM_SYMBOL_NODE,
.flags = PM_NODE_FLAG_STATIC_LITERAL,
.location = {
.start = opening->start,
.end = closing->end
}
},
.opening_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(opening),
.value_loc = node->content_loc,
.closing_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(closing),
.unescaped = node->unescaped
};
// We are explicitly _not_ using pm_node_destroy here because we don't want
// to trash the unescaped string. We could instead copy the string if we
// know that it is owned, but we're taking the fast path for now.
free(node);
return new_node;
}
/**
* Convert the given SymbolNode node to a StringNode node.
*/
static pm_string_node_t *
pm_symbol_node_to_string_node(pm_parser_t *parser, pm_symbol_node_t *node) {
pm_string_node_t *new_node = PM_ALLOC_NODE(parser, pm_string_node_t);
pm_node_flags_t flags = 0;
if (parser->frozen_string_literal) {
flags = PM_NODE_FLAG_STATIC_LITERAL | PM_STRING_FLAGS_FROZEN;
}
*new_node = (pm_string_node_t) {
{
.type = PM_STRING_NODE,
.flags = flags,
.location = node->base.location
},
.opening_loc = node->opening_loc,
.content_loc = node->value_loc,
.closing_loc = node->closing_loc,
.unescaped = node->unescaped
};
// We are explicitly _not_ using pm_node_destroy here because we don't want
// to trash the unescaped string. We could instead copy the string if we
// know that it is owned, but we're taking the fast path for now.
free(node);
return new_node;
}
/**
* Allocate and initialize a new TrueNode node.
*/
static pm_true_node_t *
pm_true_node_create(pm_parser_t *parser, const pm_token_t *token) {
assert(token->type == PM_TOKEN_KEYWORD_TRUE);
pm_true_node_t *node = PM_ALLOC_NODE(parser, pm_true_node_t);
*node = (pm_true_node_t) {{
.type = PM_TRUE_NODE,
.flags = PM_NODE_FLAG_STATIC_LITERAL,
.location = PM_LOCATION_TOKEN_VALUE(token)
}};
return node;
}
/**
* Allocate and initialize a new UndefNode node.
*/
static pm_undef_node_t *
pm_undef_node_create(pm_parser_t *parser, const pm_token_t *token) {
assert(token->type == PM_TOKEN_KEYWORD_UNDEF);
pm_undef_node_t *node = PM_ALLOC_NODE(parser, pm_undef_node_t);
*node = (pm_undef_node_t) {
{
.type = PM_UNDEF_NODE,
.location = PM_LOCATION_TOKEN_VALUE(token),
},
.keyword_loc = PM_LOCATION_TOKEN_VALUE(token),
.names = { 0 }
};
return node;
}
/**
* Append a name to an undef node.
*/
static void
pm_undef_node_append(pm_undef_node_t *node, pm_node_t *name) {
node->base.location.end = name->location.end;
pm_node_list_append(&node->names, name);
}
/**
* Allocate a new UnlessNode node.
*/
static pm_unless_node_t *
pm_unless_node_create(pm_parser_t *parser, const pm_token_t *keyword, pm_node_t *predicate, const pm_token_t *then_keyword, pm_statements_node_t *statements) {
pm_conditional_predicate(predicate);
pm_unless_node_t *node = PM_ALLOC_NODE(parser, pm_unless_node_t);
const uint8_t *end;
if (statements != NULL) {
end = statements->base.location.end;
} else {
end = predicate->location.end;
}
*node = (pm_unless_node_t) {
{
.type = PM_UNLESS_NODE,
.flags = PM_NODE_FLAG_NEWLINE,
.location = {
.start = keyword->start,
.end = end
},
},
.keyword_loc = PM_LOCATION_TOKEN_VALUE(keyword),
.predicate = predicate,
.then_keyword_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(then_keyword),
.statements = statements,
.consequent = NULL,
.end_keyword_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE
};
return node;
}
/**
* Allocate and initialize new UnlessNode node in the modifier form.
*/
static pm_unless_node_t *
pm_unless_node_modifier_create(pm_parser_t *parser, pm_node_t *statement, const pm_token_t *unless_keyword, pm_node_t *predicate) {
pm_conditional_predicate(predicate);
pm_unless_node_t *node = PM_ALLOC_NODE(parser, pm_unless_node_t);
pm_statements_node_t *statements = pm_statements_node_create(parser);
pm_statements_node_body_append(statements, statement);
*node = (pm_unless_node_t) {
{
.type = PM_UNLESS_NODE,
.flags = PM_NODE_FLAG_NEWLINE,
.location = {
.start = statement->location.start,
.end = predicate->location.end
},
},
.keyword_loc = PM_LOCATION_TOKEN_VALUE(unless_keyword),
.predicate = predicate,
.then_keyword_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE,
.statements = statements,
.consequent = NULL,
.end_keyword_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE
};
return node;
}
static inline void
pm_unless_node_end_keyword_loc_set(pm_unless_node_t *node, const pm_token_t *end_keyword) {
node->end_keyword_loc = PM_LOCATION_TOKEN_VALUE(end_keyword);
node->base.location.end = end_keyword->end;
}
/**
* Allocate a new UntilNode node.
*/
static pm_until_node_t *
pm_until_node_create(pm_parser_t *parser, const pm_token_t *keyword, const pm_token_t *closing, pm_node_t *predicate, pm_statements_node_t *statements, pm_node_flags_t flags) {
pm_until_node_t *node = PM_ALLOC_NODE(parser, pm_until_node_t);
*node = (pm_until_node_t) {
{
.type = PM_UNTIL_NODE,
.flags = flags,
.location = {
.start = keyword->start,
.end = closing->end,
},
},
.keyword_loc = PM_LOCATION_TOKEN_VALUE(keyword),
.closing_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(closing),
.predicate = predicate,
.statements = statements
};
return node;
}
/**
* Allocate a new UntilNode node.
*/
static pm_until_node_t *
pm_until_node_modifier_create(pm_parser_t *parser, const pm_token_t *keyword, pm_node_t *predicate, pm_statements_node_t *statements, pm_node_flags_t flags) {
pm_until_node_t *node = PM_ALLOC_NODE(parser, pm_until_node_t);
*node = (pm_until_node_t) {
{
.type = PM_UNTIL_NODE,
.flags = flags,
.location = {
.start = statements->base.location.start,
.end = predicate->location.end,
},
},
.keyword_loc = PM_LOCATION_TOKEN_VALUE(keyword),
.closing_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE,
.predicate = predicate,
.statements = statements
};
return node;
}
/**
* Allocate and initialize a new WhenNode node.
*/
static pm_when_node_t *
pm_when_node_create(pm_parser_t *parser, const pm_token_t *keyword) {
pm_when_node_t *node = PM_ALLOC_NODE(parser, pm_when_node_t);
*node = (pm_when_node_t) {
{
.type = PM_WHEN_NODE,
.location = {
.start = keyword->start,
.end = NULL
}
},
.keyword_loc = PM_LOCATION_TOKEN_VALUE(keyword),
.statements = NULL,
.conditions = { 0 }
};
return node;
}
/**
* Append a new condition to a when node.
*/
static void
pm_when_node_conditions_append(pm_when_node_t *node, pm_node_t *condition) {
node->base.location.end = condition->location.end;
pm_node_list_append(&node->conditions, condition);
}
/**
* Set the statements list of a when node.
*/
static void
pm_when_node_statements_set(pm_when_node_t *node, pm_statements_node_t *statements) {
if (statements->base.location.end > node->base.location.end) {
node->base.location.end = statements->base.location.end;
}
node->statements = statements;
}
/**
* Allocate a new WhileNode node.
*/
static pm_while_node_t *
pm_while_node_create(pm_parser_t *parser, const pm_token_t *keyword, const pm_token_t *closing, pm_node_t *predicate, pm_statements_node_t *statements, pm_node_flags_t flags) {
pm_while_node_t *node = PM_ALLOC_NODE(parser, pm_while_node_t);
*node = (pm_while_node_t) {
{
.type = PM_WHILE_NODE,
.flags = flags,
.location = {
.start = keyword->start,
.end = closing->end
},
},
.keyword_loc = PM_LOCATION_TOKEN_VALUE(keyword),
.closing_loc = PM_OPTIONAL_LOCATION_TOKEN_VALUE(closing),
.predicate = predicate,
.statements = statements
};
return node;
}
/**
* Allocate a new WhileNode node.
*/
static pm_while_node_t *
pm_while_node_modifier_create(pm_parser_t *parser, const pm_token_t *keyword, pm_node_t *predicate, pm_statements_node_t *statements, pm_node_flags_t flags) {
pm_while_node_t *node = PM_ALLOC_NODE(parser, pm_while_node_t);
*node = (pm_while_node_t) {
{
.type = PM_WHILE_NODE,
.flags = flags,
.location = {
.start = statements->base.location.start,
.end = predicate->location.end
},
},
.keyword_loc = PM_LOCATION_TOKEN_VALUE(keyword),
.closing_loc = PM_OPTIONAL_LOCATION_NOT_PROVIDED_VALUE,
.predicate = predicate,
.statements = statements
};
return node;
}
/**
* Allocate and initialize a new XStringNode node with the given unescaped
* string.
*/
static pm_x_string_node_t *
pm_xstring_node_create_unescaped(pm_parser_t *parser, const pm_token_t *opening, const pm_token_t *content, const pm_token_t *closing, const pm_string_t *unescaped) {
pm_x_string_node_t *node = PM_ALLOC_NODE(parser, pm_x_string_node_t);
*node = (pm_x_string_node_t) {
{
.type = PM_X_STRING_NODE,
.flags = PM_STRING_FLAGS_FROZEN,
.location = {
.start = opening->start,
.end = closing->end
},
},
.opening_loc = PM_LOCATION_TOKEN_VALUE(opening),
.content_loc = PM_LOCATION_TOKEN_VALUE(content),
.closing_loc = PM_LOCATION_TOKEN_VALUE(closing),
.unescaped = *unescaped
};
return node;
}
/**
* Allocate and initialize a new XStringNode node.
*/
static inline pm_x_string_node_t *
pm_xstring_node_create(pm_parser_t *parser, const pm_token_t *opening, const pm_token_t *content, const pm_token_t *closing) {
return pm_xstring_node_create_unescaped(parser, opening, content, closing, &PM_STRING_EMPTY);
}
/**
* Allocate a new YieldNode node.
*/
static pm_yield_node_t *
pm_yield_node_create(pm_parser_t *parser, const pm_token_t *keyword, const pm_location_t *lparen_loc, pm_arguments_node_t *arguments, const pm_location_t *rparen_loc) {
pm_yield_node_t *node = PM_ALLOC_NODE(parser, pm_yield_node_t);
const uint8_t *end;
if (rparen_loc->start != NULL) {
end = rparen_loc->end;
} else if (arguments != NULL) {
end = arguments->base.location.end;
} else if (lparen_loc->start != NULL) {
end = lparen_loc->end;
} else {
end = keyword->end;
}
*node = (pm_yield_node_t) {
{
.type = PM_YIELD_NODE,
.location = {
.start = keyword->start,
.end = end
},
},
.keyword_loc = PM_LOCATION_TOKEN_VALUE(keyword),
.lparen_loc = *lparen_loc,
.arguments = arguments,
.rparen_loc = *rparen_loc
};
return node;
}
#undef PM_ALLOC_NODE
/******************************************************************************/
/* Scope-related functions */
/******************************************************************************/
/**
* Allocate and initialize a new scope. Push it onto the scope stack.
*/
static bool
pm_parser_scope_push(pm_parser_t *parser, bool closed) {
pm_scope_t *scope = (pm_scope_t *) malloc(sizeof(pm_scope_t));
if (scope == NULL) return false;
*scope = (pm_scope_t) {
.previous = parser->current_scope,
.closed = closed,
.explicit_params = false,
.numbered_parameters = 0,
};
pm_constant_id_list_init(&scope->locals);
parser->current_scope = scope;
return true;
}
/**
* Check if any of the currently visible scopes contain a local variable
* described by the given constant id.
*/
static int
pm_parser_local_depth_constant_id(pm_parser_t *parser, pm_constant_id_t constant_id) {
pm_scope_t *scope = parser->current_scope;
int depth = 0;
while (scope != NULL) {
if (pm_constant_id_list_includes(&scope->locals, constant_id)) return depth;
if (scope->closed) break;
scope = scope->previous;
depth++;
}
return -1;
}
/**
* Check if any of the currently visible scopes contain a local variable
* described by the given token. This function implicitly inserts a constant
* into the constant pool.
*/
static inline int
pm_parser_local_depth(pm_parser_t *parser, pm_token_t *token) {
return pm_parser_local_depth_constant_id(parser, pm_parser_constant_id_token(parser, token));
}
/**
* Add a constant id to the local table of the current scope.
*/
static inline void
pm_parser_local_add(pm_parser_t *parser, pm_constant_id_t constant_id) {
if (!pm_constant_id_list_includes(&parser->current_scope->locals, constant_id)) {
pm_constant_id_list_append(&parser->current_scope->locals, constant_id);
}
}
/**
* Set the numbered_parameters value of the current scope.
*/
static inline void
pm_parser_numbered_parameters_set(pm_parser_t *parser, uint8_t numbered_parameters) {
parser->current_scope->numbered_parameters = numbered_parameters;
}
/**
* Add a local variable from a location to the current scope.
*/
static pm_constant_id_t
pm_parser_local_add_location(pm_parser_t *parser, const uint8_t *start, const uint8_t *end) {
pm_constant_id_t constant_id = pm_parser_constant_id_location(parser, start, end);
if (constant_id != 0) pm_parser_local_add(parser, constant_id);
return constant_id;
}
/**
* Add a local variable from a token to the current scope.
*/
static inline void
pm_parser_local_add_token(pm_parser_t *parser, pm_token_t *token) {
pm_parser_local_add_location(parser, token->start, token->end);
}
/**
* Add a local variable from an owned string to the current scope.
*/
static pm_constant_id_t
pm_parser_local_add_owned(pm_parser_t *parser, const uint8_t *start, size_t length) {
pm_constant_id_t constant_id = pm_parser_constant_id_owned(parser, start, length);
if (constant_id != 0) pm_parser_local_add(parser, constant_id);
return constant_id;
}
/**
* Add a parameter name to the current scope and check whether the name of the
* parameter is unique or not.
*/
static void
pm_parser_parameter_name_check(pm_parser_t *parser, const pm_token_t *name) {
// We want to check whether the parameter name is a numbered parameter or
// not.
pm_refute_numbered_parameter(parser, name->start, name->end);
// We want to ignore any parameter name that starts with an underscore.
if ((name->start < name->end) && (*name->start == '_')) return;
// Otherwise we'll fetch the constant id for the parameter name and check
// whether it's already in the current scope.
pm_constant_id_t constant_id = pm_parser_constant_id_token(parser, name);
if (pm_constant_id_list_includes(&parser->current_scope->locals, constant_id)) {
pm_parser_err_token(parser, name, PM_ERR_PARAMETER_NAME_REPEAT);
}
}
/**
* Pop the current scope off the scope stack. Note that we specifically do not
* free the associated constant list because we assume that we have already
* transferred ownership of the list to the AST somewhere.
*/
static void
pm_parser_scope_pop(pm_parser_t *parser) {
pm_scope_t *scope = parser->current_scope;
parser->current_scope = scope->previous;
free(scope);
}
/******************************************************************************/
/* Basic character checks */
/******************************************************************************/
/**
* This function is used extremely frequently to lex all of the identifiers in a
* source file, so it's important that it be as fast as possible. For this
* reason we have the encoding_changed boolean to check if we need to go through
* the function pointer or can just directly use the UTF-8 functions.
*/
static inline size_t
char_is_identifier_start(pm_parser_t *parser, const uint8_t *b) {
if (parser->encoding_changed) {
size_t width;
if ((width = parser->encoding->alpha_char(b, parser->end - b)) != 0) {
return width;
} else if (*b == '_') {
return 1;
} else if (*b >= 0x80) {
return parser->encoding->char_width(b, parser->end - b);
} else {
return 0;
}
} else if (*b < 0x80) {
return (pm_encoding_unicode_table[*b] & PRISM_ENCODING_ALPHABETIC_BIT ? 1 : 0) || (*b == '_');
} else {
return (size_t) (pm_encoding_utf_8_alpha_char(b, parser->end - b) || 1u);
}
}
/**
* Similar to char_is_identifier but this function assumes that the encoding
* has not been changed.
*/
static inline size_t
char_is_identifier_utf8(const uint8_t *b, const uint8_t *end) {
if (*b < 0x80) {
return (*b == '_') || (pm_encoding_unicode_table[*b] & PRISM_ENCODING_ALPHANUMERIC_BIT ? 1 : 0);
} else {
return (size_t) (pm_encoding_utf_8_alnum_char(b, end - b) || 1u);
}
}
/**
* Like the above, this function is also used extremely frequently to lex all of
* the identifiers in a source file once the first character has been found. So
* it's important that it be as fast as possible.
*/
static inline size_t
char_is_identifier(pm_parser_t *parser, const uint8_t *b) {
if (parser->encoding_changed) {
size_t width;
if ((width = parser->encoding->alnum_char(b, parser->end - b)) != 0) {
return width;
} else if (*b == '_') {
return 1;
} else if (*b >= 0x80) {
return parser->encoding->char_width(b, parser->end - b);
} else {
return 0;
}
}
return char_is_identifier_utf8(b, parser->end);
}
// Here we're defining a perfect hash for the characters that are allowed in
// global names. This is used to quickly check the next character after a $ to
// see if it's a valid character for a global name.
#define BIT(c, idx) (((c) / 32 - 1 == idx) ? (1U << ((c) % 32)) : 0)
#define PUNCT(idx) ( \
BIT('~', idx) | BIT('*', idx) | BIT('$', idx) | BIT('?', idx) | \
BIT('!', idx) | BIT('@', idx) | BIT('/', idx) | BIT('\\', idx) | \
BIT(';', idx) | BIT(',', idx) | BIT('.', idx) | BIT('=', idx) | \
BIT(':', idx) | BIT('<', idx) | BIT('>', idx) | BIT('\"', idx) | \
BIT('&', idx) | BIT('`', idx) | BIT('\'', idx) | BIT('+', idx) | \
BIT('0', idx))
const unsigned int pm_global_name_punctuation_hash[(0x7e - 0x20 + 31) / 32] = { PUNCT(0), PUNCT(1), PUNCT(2) };
#undef BIT
#undef PUNCT
static inline bool
char_is_global_name_punctuation(const uint8_t b) {
const unsigned int i = (const unsigned int) b;
if (i <= 0x20 || 0x7e < i) return false;
return (pm_global_name_punctuation_hash[(i - 0x20) / 32] >> (i % 32)) & 1;
}
static inline bool
token_is_setter_name(pm_token_t *token) {
return (
(token->type == PM_TOKEN_IDENTIFIER) &&
(token->end - token->start >= 2) &&
(token->end[-1] == '=')
);
}
/******************************************************************************/
/* Stack helpers */
/******************************************************************************/
static inline void
pm_accepts_block_stack_push(pm_parser_t *parser, bool value) {
// Use the negation of the value to prevent stack overflow.
pm_state_stack_push(&parser->accepts_block_stack, !value);
}
static inline void
pm_accepts_block_stack_pop(pm_parser_t *parser) {
pm_state_stack_pop(&parser->accepts_block_stack);
}
static inline bool
pm_accepts_block_stack_p(pm_parser_t *parser) {
return !pm_state_stack_p(&parser->accepts_block_stack);
}
static inline void
pm_do_loop_stack_push(pm_parser_t *parser, bool value) {
pm_state_stack_push(&parser->do_loop_stack, value);
}
static inline void
pm_do_loop_stack_pop(pm_parser_t *parser) {
pm_state_stack_pop(&parser->do_loop_stack);
}
static inline bool
pm_do_loop_stack_p(pm_parser_t *parser) {
return pm_state_stack_p(&parser->do_loop_stack);
}
/******************************************************************************/
/* Lexer check helpers */
/******************************************************************************/
/**
* Get the next character in the source starting from +cursor+. If that position
* is beyond the end of the source then return '\0'.
*/
static inline uint8_t
peek_at(pm_parser_t *parser, const uint8_t *cursor) {
if (cursor < parser->end) {
return *cursor;
} else {
return '\0';
}
}
/**
* Get the next character in the source starting from parser->current.end and
* adding the given offset. If that position is beyond the end of the source
* then return '\0'.
*/
static inline uint8_t
peek_offset(pm_parser_t *parser, ptrdiff_t offset) {
return peek_at(parser, parser->current.end + offset);
}
/**
* Get the next character in the source starting from parser->current.end. If
* that position is beyond the end of the source then return '\0'.
*/
static inline uint8_t
peek(pm_parser_t *parser) {
return peek_at(parser, parser->current.end);
}
/**
* If the character to be read matches the given value, then returns true and
* advanced the current pointer.
*/
static inline bool
match(pm_parser_t *parser, uint8_t value) {
if (peek(parser) == value) {
parser->current.end++;
return true;
}
return false;
}
/**
* Return the length of the line ending string starting at +cursor+, or 0 if it
* is not a line ending. This function is intended to be CRLF/LF agnostic.
*/
static inline size_t
match_eol_at(pm_parser_t *parser, const uint8_t *cursor) {
if (peek_at(parser, cursor) == '\n') {
return 1;
}
if (peek_at(parser, cursor) == '\r' && peek_at(parser, cursor + 1) == '\n') {
return 2;
}
return 0;
}
/**
* Return the length of the line ending string starting at
* `parser->current.end + offset`, or 0 if it is not a line ending. This
* function is intended to be CRLF/LF agnostic.
*/
static inline size_t
match_eol_offset(pm_parser_t *parser, ptrdiff_t offset) {
return match_eol_at(parser, parser->current.end + offset);
}
/**
* Return the length of the line ending string starting at parser->current.end,
* or 0 if it is not a line ending. This function is intended to be CRLF/LF
* agnostic.
*/
static inline size_t
match_eol(pm_parser_t *parser) {
return match_eol_at(parser, parser->current.end);
}
/**
* Skip to the next newline character or NUL byte.
*/
static inline const uint8_t *
next_newline(const uint8_t *cursor, ptrdiff_t length) {
assert(length >= 0);
// Note that it's okay for us to use memchr here to look for \n because none
// of the encodings that we support have \n as a component of a multi-byte
// character.
return memchr(cursor, '\n', (size_t) length);
}
/**
* Here we're going to check if this is a "magic" comment, and perform whatever
* actions are necessary for it here.
*/
static bool
parser_lex_magic_comment_encoding_value(pm_parser_t *parser, const uint8_t *start, const uint8_t *end) {
const pm_encoding_t *encoding = pm_encoding_find(start, end);
if (encoding != NULL) {
if (encoding != PM_ENCODING_UTF_8_ENTRY) {
parser->encoding = encoding;
parser->encoding_changed = true;
if (parser->encoding_changed_callback != NULL) parser->encoding_changed_callback(parser);
}
return true;
}
return false;
}
/**
* Look for a specific pattern of "coding" and potentially set the encoding on
* the parser.
*/
static void
parser_lex_magic_comment_encoding(pm_parser_t *parser) {
const uint8_t *cursor = parser->current.start + 1;
const uint8_t *end = parser->current.end;
bool separator = false;
while (true) {
if (end - cursor <= 6) return;
switch (cursor[6]) {
case 'C': case 'c': cursor += 6; continue;
case 'O': case 'o': cursor += 5; continue;
case 'D': case 'd': cursor += 4; continue;
case 'I': case 'i': cursor += 3; continue;
case 'N': case 'n': cursor += 2; continue;
case 'G': case 'g': cursor += 1; continue;
case '=': case ':':
separator = true;
cursor += 6;
break;
default:
cursor += 6;
if (pm_char_is_whitespace(*cursor)) break;
continue;
}
if (pm_strncasecmp(cursor - 6, (const uint8_t *) "coding", 6) == 0) break;
separator = false;
}
while (true) {
do {
if (++cursor >= end) return;
} while (pm_char_is_whitespace(*cursor));
if (separator) break;
if (*cursor != '=' && *cursor != ':') return;
separator = true;
cursor++;
}
const uint8_t *value_start = cursor;
while ((*cursor == '-' || *cursor == '_' || parser->encoding->alnum_char(cursor, 1)) && ++cursor < end);
if (!parser_lex_magic_comment_encoding_value(parser, value_start, cursor)) {
// If we were unable to parse the encoding value, then we've got an
// issue because we didn't understand the encoding that the user was
// trying to use. In this case we'll keep using the default encoding but
// add an error to the parser to indicate an unsuccessful parse.
pm_parser_err(parser, value_start, cursor, PM_ERR_INVALID_ENCODING_MAGIC_COMMENT);
}
}
/**
* Check if this is a magic comment that includes the frozen_string_literal
* pragma. If it does, set that field on the parser.
*/
static void
parser_lex_magic_comment_frozen_string_literal_value(pm_parser_t *parser, const uint8_t *start, const uint8_t *end) {
if (start + 4 <= end && pm_strncasecmp(start, (const uint8_t *) "true", 4) == 0) {
parser->frozen_string_literal = true;
}
}
static inline bool
pm_char_is_magic_comment_key_delimiter(const uint8_t b) {
return b == '\'' || b == '"' || b == ':' || b == ';';
}
/**
* Find an emacs magic comment marker (-*-) within the given bounds. If one is
* found, it returns a pointer to the start of the marker. Otherwise it returns
* NULL.
*/
static inline const uint8_t *
parser_lex_magic_comment_emacs_marker(pm_parser_t *parser, const uint8_t *cursor, const uint8_t *end) {
while ((cursor + 3 <= end) && (cursor = pm_memchr(cursor, '-', (size_t) (end - cursor), parser->encoding_changed, parser->encoding)) != NULL) {
if (cursor + 3 <= end && cursor[1] == '*' && cursor[2] == '-') {
return cursor;
}
cursor++;
}
return NULL;
}
/**
* Parse the current token on the parser to see if it's a magic comment and
* potentially perform some action based on that. A regular expression that this
* function is effectively matching is:
*
* %r"([^\\s\'\":;]+)\\s*:\\s*(\"(?:\\\\.|[^\"])*\"|[^\"\\s;]+)[\\s;]*"
*
* It returns true if it consumes the entire comment. Otherwise it returns
* false.
*/
static inline bool
parser_lex_magic_comment(pm_parser_t *parser, bool semantic_token_seen) {
bool result = true;
const uint8_t *start = parser->current.start + 1;
const uint8_t *end = parser->current.end;
if (end - start <= 7) return false;
const uint8_t *cursor;
bool indicator = false;
if ((cursor = parser_lex_magic_comment_emacs_marker(parser, start, end)) != NULL) {
start = cursor + 3;
if ((cursor = parser_lex_magic_comment_emacs_marker(parser, start, end)) != NULL) {
end = cursor;
indicator = true;
} else {
// If we have a start marker but not an end marker, then we cannot
// have a magic comment.
return false;
}
}
cursor = start;
while (cursor < end) {
while (cursor < end && (pm_char_is_magic_comment_key_delimiter(*cursor) || pm_char_is_whitespace(*cursor))) cursor++;
const uint8_t *key_start = cursor;
while (cursor < end && (!pm_char_is_magic_comment_key_delimiter(*cursor) && !pm_char_is_whitespace(*cursor))) cursor++;
const uint8_t *key_end = cursor;
while (cursor < end && pm_char_is_whitespace(*cursor)) cursor++;
if (cursor == end) break;
if (*cursor == ':') {
cursor++;
} else {
if (!indicator) return false;
continue;
}
while (cursor < end && pm_char_is_whitespace(*cursor)) cursor++;
if (cursor == end) break;
const uint8_t *value_start;
const uint8_t *value_end;
if (*cursor == '"') {
value_start = ++cursor;
for (; cursor < end && *cursor != '"'; cursor++) {
if (*cursor == '\\' && (cursor + 1 < end)) cursor++;
}
value_end = cursor;
} else {
value_start = cursor;
while (cursor < end && *cursor != '"' && *cursor != ';' && !pm_char_is_whitespace(*cursor)) cursor++;
value_end = cursor;
}
if (indicator) {
while (cursor < end && (*cursor == ';' || pm_char_is_whitespace(*cursor))) cursor++;
} else {
while (cursor < end && pm_char_is_whitespace(*cursor)) cursor++;
if (cursor != end) return false;
}
// Here, we need to do some processing on the key to swap out dashes for
// underscores. We only need to do this if there _is_ a dash in the key.
pm_string_t key;
const size_t key_length = (size_t) (key_end - key_start);
const uint8_t *dash = pm_memchr(key_start, '-', (size_t) key_length, parser->encoding_changed, parser->encoding);
if (dash == NULL) {
pm_string_shared_init(&key, key_start, key_end);
} else {
size_t width = (size_t) (key_end - key_start);
uint8_t *buffer = malloc(width);
if (buffer == NULL) break;
memcpy(buffer, key_start, width);
buffer[dash - key_start] = '_';
while ((dash = pm_memchr(dash + 1, '-', (size_t) (key_end - dash - 1), parser->encoding_changed, parser->encoding)) != NULL) {
buffer[dash - key_start] = '_';
}
pm_string_owned_init(&key, buffer, width);
}
// Finally, we can start checking the key against the list of known
// magic comment keys, and potentially change state based on that.
const uint8_t *key_source = pm_string_source(&key);
// We only want to attempt to compare against encoding comments if it's
// the first line in the file (or the second in the case of a shebang).
if (parser->current.start == parser->encoding_comment_start) {
if (
(key_length == 8 && pm_strncasecmp(key_source, (const uint8_t *) "encoding", 8) == 0) ||
(key_length == 6 && pm_strncasecmp(key_source, (const uint8_t *) "coding", 6) == 0)
) {
result = parser_lex_magic_comment_encoding_value(parser, value_start, value_end);
}
}
// We only want to handle frozen string literal comments if it's before
// any semantic tokens have been seen.
if (!semantic_token_seen) {
if (key_length == 21 && pm_strncasecmp(key_source, (const uint8_t *) "frozen_string_literal", 21) == 0) {
parser_lex_magic_comment_frozen_string_literal_value(parser, value_start, value_end);
}
}
// When we're done, we want to free the string in case we had to
// allocate memory for it.
pm_string_free(&key);
// Allocate a new magic comment node to append to the parser's list.
pm_magic_comment_t *magic_comment;
if ((magic_comment = (pm_magic_comment_t *) calloc(sizeof(pm_magic_comment_t), 1)) != NULL) {
magic_comment->key_start = key_start;
magic_comment->value_start = value_start;
magic_comment->key_length = (uint32_t) key_length;
magic_comment->value_length = (uint32_t) (value_end - value_start);
pm_list_append(&parser->magic_comment_list, (pm_list_node_t *) magic_comment);
}
}
return result;
}
/******************************************************************************/
/* Context manipulations */
/******************************************************************************/
static bool
context_terminator(pm_context_t context, pm_token_t *token) {
switch (context) {
case PM_CONTEXT_MAIN:
case PM_CONTEXT_DEF_PARAMS:
return token->type == PM_TOKEN_EOF;
case PM_CONTEXT_DEFAULT_PARAMS:
return token->type == PM_TOKEN_COMMA || token->type == PM_TOKEN_PARENTHESIS_RIGHT;
case PM_CONTEXT_PREEXE:
case PM_CONTEXT_POSTEXE:
return token->type == PM_TOKEN_BRACE_RIGHT;
case PM_CONTEXT_MODULE:
case PM_CONTEXT_CLASS:
case PM_CONTEXT_SCLASS:
case PM_CONTEXT_LAMBDA_DO_END:
case PM_CONTEXT_DEF:
case PM_CONTEXT_BLOCK_KEYWORDS:
return token->type == PM_TOKEN_KEYWORD_END || token->type == PM_TOKEN_KEYWORD_RESCUE || token->type == PM_TOKEN_KEYWORD_ENSURE;
case PM_CONTEXT_WHILE:
case PM_CONTEXT_UNTIL:
case PM_CONTEXT_ELSE:
case PM_CONTEXT_FOR:
case PM_CONTEXT_ENSURE:
case PM_CONTEXT_ENSURE_DEF:
return token->type == PM_TOKEN_KEYWORD_END;
case PM_CONTEXT_FOR_INDEX:
return token->type == PM_TOKEN_KEYWORD_IN;
case PM_CONTEXT_CASE_WHEN:
return token->type == PM_TOKEN_KEYWORD_WHEN || token->type == PM_TOKEN_KEYWORD_END || token->type == PM_TOKEN_KEYWORD_ELSE;
case PM_CONTEXT_CASE_IN:
return token->type == PM_TOKEN_KEYWORD_IN || token->type == PM_TOKEN_KEYWORD_END || token->type == PM_TOKEN_KEYWORD_ELSE;
case PM_CONTEXT_IF:
case PM_CONTEXT_ELSIF:
return token->type == PM_TOKEN_KEYWORD_ELSE || token->type == PM_TOKEN_KEYWORD_ELSIF || token->type == PM_TOKEN_KEYWORD_END;
case PM_CONTEXT_UNLESS:
return token->type == PM_TOKEN_KEYWORD_ELSE || token->type == PM_TOKEN_KEYWORD_END;
case PM_CONTEXT_EMBEXPR:
return token->type == PM_TOKEN_EMBEXPR_END;
case PM_CONTEXT_BLOCK_BRACES:
return token->type == PM_TOKEN_BRACE_RIGHT;
case PM_CONTEXT_PARENS:
return token->type == PM_TOKEN_PARENTHESIS_RIGHT;
case PM_CONTEXT_BEGIN:
case PM_CONTEXT_RESCUE:
case PM_CONTEXT_RESCUE_DEF:
return token->type == PM_TOKEN_KEYWORD_ENSURE || token->type == PM_TOKEN_KEYWORD_RESCUE || token->type == PM_TOKEN_KEYWORD_ELSE || token->type == PM_TOKEN_KEYWORD_END;
case PM_CONTEXT_RESCUE_ELSE:
case PM_CONTEXT_RESCUE_ELSE_DEF:
return token->type == PM_TOKEN_KEYWORD_ENSURE || token->type == PM_TOKEN_KEYWORD_END;
case PM_CONTEXT_LAMBDA_BRACES:
return token->type == PM_TOKEN_BRACE_RIGHT;
case PM_CONTEXT_PREDICATE:
return token->type == PM_TOKEN_KEYWORD_THEN || token->type == PM_TOKEN_NEWLINE || token->type == PM_TOKEN_SEMICOLON;
}
return false;
}
static bool
context_recoverable(pm_parser_t *parser, pm_token_t *token) {
pm_context_node_t *context_node = parser->current_context;
while (context_node != NULL) {
if (context_terminator(context_node->context, token)) return true;
context_node = context_node->prev;
}
return false;
}
static bool
context_push(pm_parser_t *parser, pm_context_t context) {
pm_context_node_t *context_node = (pm_context_node_t *) malloc(sizeof(pm_context_node_t));
if (context_node == NULL) return false;
*context_node = (pm_context_node_t) { .context = context, .prev = NULL };
if (parser->current_context == NULL) {
parser->current_context = context_node;
} else {
context_node->prev = parser->current_context;
parser->current_context = context_node;
}
return true;
}
static void
context_pop(pm_parser_t *parser) {
pm_context_node_t *prev = parser->current_context->prev;
free(parser->current_context);
parser->current_context = prev;
}
static bool
context_p(pm_parser_t *parser, pm_context_t context) {
pm_context_node_t *context_node = parser->current_context;
while (context_node != NULL) {
if (context_node->context == context) return true;
context_node = context_node->prev;
}
return false;
}
static bool
context_def_p(pm_parser_t *parser) {
pm_context_node_t *context_node = parser->current_context;
while (context_node != NULL) {
switch (context_node->context) {
case PM_CONTEXT_DEF:
case PM_CONTEXT_DEF_PARAMS:
case PM_CONTEXT_ENSURE_DEF:
case PM_CONTEXT_RESCUE_DEF:
case PM_CONTEXT_RESCUE_ELSE_DEF:
return true;
case PM_CONTEXT_CLASS:
case PM_CONTEXT_MODULE:
case PM_CONTEXT_SCLASS:
return false;
default:
context_node = context_node->prev;
}
}
return false;
}
/******************************************************************************/
/* Specific token lexers */
/******************************************************************************/
static void
pm_strspn_number_validate(pm_parser_t *parser, const uint8_t *invalid) {
if (invalid != NULL) {
pm_parser_err(parser, invalid, invalid + 1, PM_ERR_INVALID_NUMBER_UNDERSCORE);
}
}
static size_t
pm_strspn_binary_number_validate(pm_parser_t *parser, const uint8_t *string) {
const uint8_t *invalid = NULL;
size_t length = pm_strspn_binary_number(string, parser->end - string, &invalid);
pm_strspn_number_validate(parser, invalid);
return length;
}
static size_t
pm_strspn_octal_number_validate(pm_parser_t *parser, const uint8_t *string) {
const uint8_t *invalid = NULL;
size_t length = pm_strspn_octal_number(string, parser->end - string, &invalid);
pm_strspn_number_validate(parser, invalid);
return length;
}
static size_t
pm_strspn_decimal_number_validate(pm_parser_t *parser, const uint8_t *string) {
const uint8_t *invalid = NULL;
size_t length = pm_strspn_decimal_number(string, parser->end - string, &invalid);
pm_strspn_number_validate(parser, invalid);
return length;
}
static size_t
pm_strspn_hexadecimal_number_validate(pm_parser_t *parser, const uint8_t *string) {
const uint8_t *invalid = NULL;
size_t length = pm_strspn_hexadecimal_number(string, parser->end - string, &invalid);
pm_strspn_number_validate(parser, invalid);
return length;
}
static pm_token_type_t
lex_optional_float_suffix(pm_parser_t *parser, bool* seen_e) {
pm_token_type_t type = PM_TOKEN_INTEGER;
// Here we're going to attempt to parse the optional decimal portion of a
// float. If it's not there, then it's okay and we'll just continue on.
if (peek(parser) == '.') {
if (pm_char_is_decimal_digit(peek_offset(parser, 1))) {
parser->current.end += 2;
parser->current.end += pm_strspn_decimal_number_validate(parser, parser->current.end);
type = PM_TOKEN_FLOAT;
} else {
// If we had a . and then something else, then it's not a float suffix on
// a number it's a method call or something else.
return type;
}
}
// Here we're going to attempt to parse the optional exponent portion of a
// float. If it's not there, it's okay and we'll just continue on.
if (match(parser, 'e') || match(parser, 'E')) {
(void) (match(parser, '+') || match(parser, '-'));
*seen_e = true;
if (pm_char_is_decimal_digit(peek(parser))) {
parser->current.end++;
parser->current.end += pm_strspn_decimal_number_validate(parser, parser->current.end);
type = PM_TOKEN_FLOAT;
} else {
pm_parser_err_current(parser, PM_ERR_INVALID_FLOAT_EXPONENT);
type = PM_TOKEN_FLOAT;
}
}
return type;
}
static pm_token_type_t
lex_numeric_prefix(pm_parser_t *parser, bool* seen_e) {
pm_token_type_t type = PM_TOKEN_INTEGER;
*seen_e = false;
if (peek_offset(parser, -1) == '0') {
switch (*parser->current.end) {
// 0d1111 is a decimal number
case 'd':
case 'D':
parser->current.end++;
if (pm_char_is_decimal_digit(peek(parser))) {
parser->current.end += pm_strspn_decimal_number_validate(parser, parser->current.end);
} else {
pm_parser_err_current(parser, PM_ERR_INVALID_NUMBER_DECIMAL);
}
break;
// 0b1111 is a binary number
case 'b':
case 'B':
parser->current.end++;
if (pm_char_is_binary_digit(peek(parser))) {
parser->current.end += pm_strspn_binary_number_validate(parser, parser->current.end);
} else {
pm_parser_err_current(parser, PM_ERR_INVALID_NUMBER_BINARY);
}
parser->integer_base = PM_INTEGER_BASE_FLAGS_BINARY;
break;
// 0o1111 is an octal number
case 'o':
case 'O':
parser->current.end++;
if (pm_char_is_octal_digit(peek(parser))) {
parser->current.end += pm_strspn_octal_number_validate(parser, parser->current.end);
} else {
pm_parser_err_current(parser, PM_ERR_INVALID_NUMBER_OCTAL);
}
parser->integer_base = PM_INTEGER_BASE_FLAGS_OCTAL;
break;
// 01111 is an octal number
case '_':
case '0':
case '1':
case '2':
case '3':
case '4':
case '5':
case '6':
case '7':
parser->current.end += pm_strspn_octal_number_validate(parser, parser->current.end);
parser->integer_base = PM_INTEGER_BASE_FLAGS_OCTAL;
break;
// 0x1111 is a hexadecimal number
case 'x':
case 'X':
parser->current.end++;
if (pm_char_is_hexadecimal_digit(peek(parser))) {
parser->current.end += pm_strspn_hexadecimal_number_validate(parser, parser->current.end);
} else {
pm_parser_err_current(parser, PM_ERR_INVALID_NUMBER_HEXADECIMAL);
}
parser->integer_base = PM_INTEGER_BASE_FLAGS_HEXADECIMAL;
break;
// 0.xxx is a float
case '.': {
type = lex_optional_float_suffix(parser, seen_e);
break;
}
// 0exxx is a float
case 'e':
case 'E': {
type = lex_optional_float_suffix(parser, seen_e);
break;
}
}
} else {
// If it didn't start with a 0, then we'll lex as far as we can into a
// decimal number.
parser->current.end += pm_strspn_decimal_number_validate(parser, parser->current.end);
// Afterward, we'll lex as far as we can into an optional float suffix.
type = lex_optional_float_suffix(parser, seen_e);
}
return type;
}
static pm_token_type_t
lex_numeric(pm_parser_t *parser) {
pm_token_type_t type = PM_TOKEN_INTEGER;
parser->integer_base = PM_INTEGER_BASE_FLAGS_DECIMAL;
if (parser->current.end < parser->end) {
bool seen_e = false;
type = lex_numeric_prefix(parser, &seen_e);
const uint8_t *end = parser->current.end;
pm_token_type_t suffix_type = type;
if (type == PM_TOKEN_INTEGER) {
if (match(parser, 'r')) {
suffix_type = PM_TOKEN_INTEGER_RATIONAL;
if (match(parser, 'i')) {
suffix_type = PM_TOKEN_INTEGER_RATIONAL_IMAGINARY;
}
} else if (match(parser, 'i')) {
suffix_type = PM_TOKEN_INTEGER_IMAGINARY;
}
} else {
if (!seen_e && match(parser, 'r')) {
suffix_type = PM_TOKEN_FLOAT_RATIONAL;
if (match(parser, 'i')) {
suffix_type = PM_TOKEN_FLOAT_RATIONAL_IMAGINARY;
}
} else if (match(parser, 'i')) {
suffix_type = PM_TOKEN_FLOAT_IMAGINARY;
}
}
const uint8_t b = peek(parser);
if (b != '\0' && (b >= 0x80 || ((b >= 'a' && b <= 'z') || (b >= 'A' && b <= 'Z')) || b == '_')) {
parser->current.end = end;
} else {
type = suffix_type;
}
}
return type;
}
static pm_token_type_t
lex_global_variable(pm_parser_t *parser) {
if (parser->current.end >= parser->end) {
pm_parser_err_current(parser, PM_ERR_INVALID_VARIABLE_GLOBAL);
return PM_TOKEN_GLOBAL_VARIABLE;
}
switch (*parser->current.end) {
case '~': // $~: match-data
case '*': // $*: argv
case '$': // $$: pid
case '?': // $?: last status
case '!': // $!: error string
case '@': // $@: error position
case '/': // $/: input record separator
case '\\': // $\: output record separator
case ';': // $;: field separator
case ',': // $,: output field separator
case '.': // $.: last read line number
case '=': // $=: ignorecase
case ':': // $:: load path
case '<': // $<: reading filename
case '>': // $>: default output handle
case '\"': // $": already loaded files
parser->current.end++;
return PM_TOKEN_GLOBAL_VARIABLE;
case '&': // $&: last match
case '`': // $`: string before last match
case '\'': // $': string after last match
case '+': // $+: string matches last paren.
parser->current.end++;
return lex_state_p(parser, PM_LEX_STATE_FNAME) ? PM_TOKEN_GLOBAL_VARIABLE : PM_TOKEN_BACK_REFERENCE;
case '0': {
parser->current.end++;
size_t width;
if (parser->current.end < parser->end && (width = char_is_identifier(parser, parser->current.end)) > 0) {
do {
parser->current.end += width;
} while (parser->current.end < parser->end && (width = char_is_identifier(parser, parser->current.end)) > 0);
// $0 isn't allowed to be followed by anything.
pm_parser_err_current(parser, PM_ERR_INVALID_VARIABLE_GLOBAL);
}
return PM_TOKEN_GLOBAL_VARIABLE;
}
case '1':
case '2':
case '3':
case '4':
case '5':
case '6':
case '7':
case '8':
case '9':
parser->current.end += pm_strspn_decimal_digit(parser->current.end, parser->end - parser->current.end);
return lex_state_p(parser, PM_LEX_STATE_FNAME) ? PM_TOKEN_GLOBAL_VARIABLE : PM_TOKEN_NUMBERED_REFERENCE;
case '-':
parser->current.end++;
/* fallthrough */
default: {
size_t width;
if ((width = char_is_identifier(parser, parser->current.end)) > 0) {
do {
parser->current.end += width;
} while (parser->current.end < parser->end && (width = char_is_identifier(parser, parser->current.end)) > 0);
} else {
// If we get here, then we have a $ followed by something that isn't
// recognized as a global variable.
pm_parser_err_current(parser, PM_ERR_INVALID_VARIABLE_GLOBAL);
}
return PM_TOKEN_GLOBAL_VARIABLE;
}
}
}
/**
* This function checks if the current token matches a keyword. If it does, it
* returns the token type. Otherwise, it returns PM_TOKEN_EOF. The arguments are as follows:
*
* * `parser` - the parser object
* * `current_start` - pointer to the start of the current token
* * `value` - the literal string that we're checking for
* * `vlen` - the length of the token
* * `state` - the state that we should transition to if the token matches
* * `type` - the expected token type
* * `modifier_type` - the expected modifier token type
*/
static inline pm_token_type_t
lex_keyword(pm_parser_t *parser, const uint8_t *current_start, const char *value, size_t vlen, pm_lex_state_t state, pm_token_type_t type, pm_token_type_t modifier_type) {
if (memcmp(current_start, value, vlen) == 0) {
pm_lex_state_t last_state = parser->lex_state;
if (parser->lex_state & PM_LEX_STATE_FNAME) {
lex_state_set(parser, PM_LEX_STATE_ENDFN);
} else {
lex_state_set(parser, state);
if (state == PM_LEX_STATE_BEG) {
parser->command_start = true;
}
if ((modifier_type != PM_TOKEN_EOF) && !(last_state & (PM_LEX_STATE_BEG | PM_LEX_STATE_LABELED | PM_LEX_STATE_CLASS))) {
lex_state_set(parser, PM_LEX_STATE_BEG | PM_LEX_STATE_LABEL);
return modifier_type;
}
}
return type;
}
return PM_TOKEN_EOF;
}
static pm_token_type_t
lex_identifier(pm_parser_t *parser, bool previous_command_start) {
// Lex as far as we can into the current identifier.
size_t width;
const uint8_t *end = parser->end;
const uint8_t *current_start = parser->current.start;
const uint8_t *current_end = parser->current.end;
bool encoding_changed = parser->encoding_changed;
if (encoding_changed) {
while (current_end < end && (width = char_is_identifier(parser, current_end)) > 0) {
current_end += width;
}
} else {
while (current_end < end && (width = char_is_identifier_utf8(current_end, end)) > 0) {
current_end += width;
}
}
parser->current.end = current_end;
// Now cache the length of the identifier so that we can quickly compare it
// against known keywords.
width = (size_t) (current_end - current_start);
if (current_end < end) {
if (((current_end + 1 >= end) || (current_end[1] != '=')) && (match(parser, '!') || match(parser, '?'))) {
// First we'll attempt to extend the identifier by a ! or ?. Then we'll
// check if we're returning the defined? keyword or just an identifier.
width++;
if (
((lex_state_p(parser, PM_LEX_STATE_LABEL | PM_LEX_STATE_ENDFN) && !previous_command_start) || lex_state_arg_p(parser)) &&
(peek(parser) == ':') && (peek_offset(parser, 1) != ':')
) {
// If we're in a position where we can accept a : at the end of an
// identifier, then we'll optionally accept it.
lex_state_set(parser, PM_LEX_STATE_ARG | PM_LEX_STATE_LABELED);
(void) match(parser, ':');
return PM_TOKEN_LABEL;
}
if (parser->lex_state != PM_LEX_STATE_DOT) {
if (width == 8 && (lex_keyword(parser, current_start, "defined?", width, PM_LEX_STATE_ARG, PM_TOKEN_KEYWORD_DEFINED, PM_TOKEN_EOF) != PM_TOKEN_EOF)) {
return PM_TOKEN_KEYWORD_DEFINED;
}
}
return PM_TOKEN_METHOD_NAME;
}
if (lex_state_p(parser, PM_LEX_STATE_FNAME) && peek_offset(parser, 1) != '~' && peek_offset(parser, 1) != '>' && (peek_offset(parser, 1) != '=' || peek_offset(parser, 2) == '>') && match(parser, '=')) {
// If we're in a position where we can accept a = at the end of an
// identifier, then we'll optionally accept it.
return PM_TOKEN_IDENTIFIER;
}
if (
((lex_state_p(parser, PM_LEX_STATE_LABEL | PM_LEX_STATE_ENDFN) && !previous_command_start) || lex_state_arg_p(parser)) &&
peek(parser) == ':' && peek_offset(parser, 1) != ':'
) {
// If we're in a position where we can accept a : at the end of an
// identifier, then we'll optionally accept it.
lex_state_set(parser, PM_LEX_STATE_ARG | PM_LEX_STATE_LABELED);
(void) match(parser, ':');
return PM_TOKEN_LABEL;
}
}
if (parser->lex_state != PM_LEX_STATE_DOT) {
pm_token_type_t type;
switch (width) {
case 2:
if (lex_keyword(parser, current_start, "do", width, PM_LEX_STATE_BEG, PM_TOKEN_KEYWORD_DO, PM_TOKEN_EOF) != PM_TOKEN_EOF) {
if (pm_do_loop_stack_p(parser)) {
return PM_TOKEN_KEYWORD_DO_LOOP;
}
return PM_TOKEN_KEYWORD_DO;
}
if ((type = lex_keyword(parser, current_start, "if", width, PM_LEX_STATE_BEG, PM_TOKEN_KEYWORD_IF, PM_TOKEN_KEYWORD_IF_MODIFIER)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "in", width, PM_LEX_STATE_BEG, PM_TOKEN_KEYWORD_IN, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "or", width, PM_LEX_STATE_BEG, PM_TOKEN_KEYWORD_OR, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
break;
case 3:
if ((type = lex_keyword(parser, current_start, "and", width, PM_LEX_STATE_BEG, PM_TOKEN_KEYWORD_AND, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "def", width, PM_LEX_STATE_FNAME, PM_TOKEN_KEYWORD_DEF, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "end", width, PM_LEX_STATE_END, PM_TOKEN_KEYWORD_END, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "END", width, PM_LEX_STATE_END, PM_TOKEN_KEYWORD_END_UPCASE, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "for", width, PM_LEX_STATE_BEG, PM_TOKEN_KEYWORD_FOR, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "nil", width, PM_LEX_STATE_END, PM_TOKEN_KEYWORD_NIL, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "not", width, PM_LEX_STATE_ARG, PM_TOKEN_KEYWORD_NOT, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
break;
case 4:
if ((type = lex_keyword(parser, current_start, "case", width, PM_LEX_STATE_BEG, PM_TOKEN_KEYWORD_CASE, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "else", width, PM_LEX_STATE_BEG, PM_TOKEN_KEYWORD_ELSE, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "next", width, PM_LEX_STATE_MID, PM_TOKEN_KEYWORD_NEXT, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "redo", width, PM_LEX_STATE_END, PM_TOKEN_KEYWORD_REDO, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "self", width, PM_LEX_STATE_END, PM_TOKEN_KEYWORD_SELF, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "then", width, PM_LEX_STATE_BEG, PM_TOKEN_KEYWORD_THEN, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "true", width, PM_LEX_STATE_END, PM_TOKEN_KEYWORD_TRUE, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "when", width, PM_LEX_STATE_BEG, PM_TOKEN_KEYWORD_WHEN, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
break;
case 5:
if ((type = lex_keyword(parser, current_start, "alias", width, PM_LEX_STATE_FNAME | PM_LEX_STATE_FITEM, PM_TOKEN_KEYWORD_ALIAS, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "begin", width, PM_LEX_STATE_BEG, PM_TOKEN_KEYWORD_BEGIN, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "BEGIN", width, PM_LEX_STATE_END, PM_TOKEN_KEYWORD_BEGIN_UPCASE, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "break", width, PM_LEX_STATE_MID, PM_TOKEN_KEYWORD_BREAK, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "class", width, PM_LEX_STATE_CLASS, PM_TOKEN_KEYWORD_CLASS, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "elsif", width, PM_LEX_STATE_BEG, PM_TOKEN_KEYWORD_ELSIF, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "false", width, PM_LEX_STATE_END, PM_TOKEN_KEYWORD_FALSE, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "retry", width, PM_LEX_STATE_END, PM_TOKEN_KEYWORD_RETRY, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "super", width, PM_LEX_STATE_ARG, PM_TOKEN_KEYWORD_SUPER, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "undef", width, PM_LEX_STATE_FNAME | PM_LEX_STATE_FITEM, PM_TOKEN_KEYWORD_UNDEF, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "until", width, PM_LEX_STATE_BEG, PM_TOKEN_KEYWORD_UNTIL, PM_TOKEN_KEYWORD_UNTIL_MODIFIER)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "while", width, PM_LEX_STATE_BEG, PM_TOKEN_KEYWORD_WHILE, PM_TOKEN_KEYWORD_WHILE_MODIFIER)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "yield", width, PM_LEX_STATE_ARG, PM_TOKEN_KEYWORD_YIELD, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
break;
case 6:
if ((type = lex_keyword(parser, current_start, "ensure", width, PM_LEX_STATE_BEG, PM_TOKEN_KEYWORD_ENSURE, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "module", width, PM_LEX_STATE_BEG, PM_TOKEN_KEYWORD_MODULE, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "rescue", width, PM_LEX_STATE_MID, PM_TOKEN_KEYWORD_RESCUE, PM_TOKEN_KEYWORD_RESCUE_MODIFIER)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "return", width, PM_LEX_STATE_MID, PM_TOKEN_KEYWORD_RETURN, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "unless", width, PM_LEX_STATE_BEG, PM_TOKEN_KEYWORD_UNLESS, PM_TOKEN_KEYWORD_UNLESS_MODIFIER)) != PM_TOKEN_EOF) return type;
break;
case 8:
if ((type = lex_keyword(parser, current_start, "__LINE__", width, PM_LEX_STATE_END, PM_TOKEN_KEYWORD___LINE__, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
if ((type = lex_keyword(parser, current_start, "__FILE__", width, PM_LEX_STATE_END, PM_TOKEN_KEYWORD___FILE__, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
break;
case 12:
if ((type = lex_keyword(parser, current_start, "__ENCODING__", width, PM_LEX_STATE_END, PM_TOKEN_KEYWORD___ENCODING__, PM_TOKEN_EOF)) != PM_TOKEN_EOF) return type;
break;
}
}
if (encoding_changed) {
return parser->encoding->isupper_char(current_start, end - current_start) ? PM_TOKEN_CONSTANT : PM_TOKEN_IDENTIFIER;
}
return pm_encoding_utf_8_isupper_char(current_start, end - current_start) ? PM_TOKEN_CONSTANT : PM_TOKEN_IDENTIFIER;
}
/**
* Returns true if the current token that the parser is considering is at the
* beginning of a line or the beginning of the source.
*/
static bool
current_token_starts_line(pm_parser_t *parser) {
return (parser->current.start == parser->start) || (parser->current.start[-1] == '\n');
}
/**
* When we hit a # while lexing something like a string, we need to potentially
* handle interpolation. This function performs that check. It returns a token
* type representing what it found. Those cases are:
*
* * PM_TOKEN_NOT_PROVIDED - No interpolation was found at this point. The
* caller should keep lexing.
* * PM_TOKEN_STRING_CONTENT - No interpolation was found at this point. The
* caller should return this token type.
* * PM_TOKEN_EMBEXPR_BEGIN - An embedded expression was found. The caller
* should return this token type.
* * PM_TOKEN_EMBVAR - An embedded variable was found. The caller should return
* this token type.
*/
static pm_token_type_t
lex_interpolation(pm_parser_t *parser, const uint8_t *pound) {
// If there is no content following this #, then we're at the end of
// the string and we can safely return string content.
if (pound + 1 >= parser->end) {
parser->current.end = pound + 1;
return PM_TOKEN_STRING_CONTENT;
}
// Now we'll check against the character the follows the #. If it constitutes
// valid interplation, we'll handle that, otherwise we'll return
// PM_TOKEN_NOT_PROVIDED.
switch (pound[1]) {
case '@': {
// In this case we may have hit an embedded instance or class variable.
if (pound + 2 >= parser->end) {
parser->current.end = pound + 1;
return PM_TOKEN_STRING_CONTENT;
}
// If we're looking at a @ and there's another @, then we'll skip past the
// second @.
const uint8_t *variable = pound + 2;
if (*variable == '@' && pound + 3 < parser->end) variable++;
if (char_is_identifier_start(parser, variable)) {
// At this point we're sure that we've either hit an embedded instance
// or class variable. In this case we'll first need to check if we've
// already consumed content.
if (pound > parser->current.start) {
parser->current.end = pound;
return PM_TOKEN_STRING_CONTENT;
}
// Otherwise we need to return the embedded variable token
// and then switch to the embedded variable lex mode.
lex_mode_push(parser, (pm_lex_mode_t) { .mode = PM_LEX_EMBVAR });
parser->current.end = pound + 1;
return PM_TOKEN_EMBVAR;
}
// If we didn't get an valid interpolation, then this is just regular
// string content. This is like if we get "#@-". In this case the caller
// should keep lexing.
parser->current.end = pound + 1;
return PM_TOKEN_NOT_PROVIDED;
}
case '$':
// In this case we may have hit an embedded global variable. If there's
// not enough room, then we'll just return string content.
if (pound + 2 >= parser->end) {
parser->current.end = pound + 1;
return PM_TOKEN_STRING_CONTENT;
}
// This is the character that we're going to check to see if it is the
// start of an identifier that would indicate that this is a global
// variable.
const uint8_t *check = pound + 2;
if (pound[2] == '-') {
if (pound + 3 >= parser->end) {
parser->current.end = pound + 2;
return PM_TOKEN_STRING_CONTENT;
}
check++;
}
// If the character that we're going to check is the start of an
// identifier, or we don't have a - and the character is a decimal number
// or a global name punctuation character, then we've hit an embedded
// global variable.
if (
char_is_identifier_start(parser, check) ||
(pound[2] != '-' && (pm_char_is_decimal_digit(pound[2]) || char_is_global_name_punctuation(pound[2])))
) {
// In this case we've hit an embedded global variable. First check to
// see if we've already consumed content. If we have, then we need to
// return that content as string content first.
if (pound > parser->current.start) {
parser->current.end = pound;
return PM_TOKEN_STRING_CONTENT;
}
// Otherwise, we need to return the embedded variable token and switch
// to the embedded variable lex mode.
lex_mode_push(parser, (pm_lex_mode_t) { .mode = PM_LEX_EMBVAR });
parser->current.end = pound + 1;
return PM_TOKEN_EMBVAR;
}
// In this case we've hit a #$ that does not indicate a global variable.
// In this case we'll continue lexing past it.
parser->current.end = pound + 1;
return PM_TOKEN_NOT_PROVIDED;
case '{':
// In this case it's the start of an embedded expression. If we have
// already consumed content, then we need to return that content as string
// content first.
if (pound > parser->current.start) {
parser->current.end = pound;
return PM_TOKEN_STRING_CONTENT;
}
parser->enclosure_nesting++;
// Otherwise we'll skip past the #{ and begin lexing the embedded
// expression.
lex_mode_push(parser, (pm_lex_mode_t) { .mode = PM_LEX_EMBEXPR });
parser->current.end = pound + 2;
parser->command_start = true;
pm_do_loop_stack_push(parser, false);
return PM_TOKEN_EMBEXPR_BEGIN;
default:
// In this case we've hit a # that doesn't constitute interpolation. We'll
// mark that by returning the not provided token type. This tells the
// consumer to keep lexing forward.
parser->current.end = pound + 1;
return PM_TOKEN_NOT_PROVIDED;
}
}
static const uint8_t PM_ESCAPE_FLAG_NONE = 0x0;
static const uint8_t PM_ESCAPE_FLAG_CONTROL = 0x1;
static const uint8_t PM_ESCAPE_FLAG_META = 0x2;
static const uint8_t PM_ESCAPE_FLAG_SINGLE = 0x4;
static const uint8_t PM_ESCAPE_FLAG_REGEXP = 0x8;
/**
* This is a lookup table for whether or not an ASCII character is printable.
*/
static const bool ascii_printable_chars[] = {
0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0
};
static inline bool
char_is_ascii_printable(const uint8_t b) {
return (b < 0x80) && ascii_printable_chars[b];
}
/**
* Return the value that a hexadecimal digit character represents. For example,
* transform 'a' into 10, 'b' into 11, etc.
*/
static inline uint8_t
escape_hexadecimal_digit(const uint8_t value) {
return (uint8_t) ((value <= '9') ? (value - '0') : (value & 0x7) + 9);
}
/**
* Scan the 4 digits of a Unicode escape into the value. Returns the number of
* digits scanned. This function assumes that the characters have already been
* validated.
*/
static inline uint32_t
escape_unicode(const uint8_t *string, size_t length) {
uint32_t value = 0;
for (size_t index = 0; index < length; index++) {
if (index != 0) value <<= 4;
value |= escape_hexadecimal_digit(string[index]);
}
return value;
}
/**
* Escape a single character value based on the given flags.
*/
static inline uint8_t
escape_byte(uint8_t value, const uint8_t flags) {
if (flags & PM_ESCAPE_FLAG_CONTROL) value &= 0x1f;
if (flags & PM_ESCAPE_FLAG_META) value |= 0x80;
return value;
}
/**
* Write a unicode codepoint to the given buffer.
*/
static inline void
escape_write_unicode(pm_parser_t *parser, pm_buffer_t *buffer, const uint8_t flags, const uint8_t *start, const uint8_t *end, uint32_t value) {
// \u escape sequences in string-like structures implicitly change the
// encoding to UTF-8 if they are >= 0x80 or if they are used in a character
// literal.
if (value >= 0x80 || flags & PM_ESCAPE_FLAG_SINGLE) {
if (parser->explicit_encoding != NULL && parser->explicit_encoding != PM_ENCODING_UTF_8_ENTRY) {
PM_PARSER_ERR_FORMAT(parser, start, end, PM_ERR_MIXED_ENCODING, parser->explicit_encoding->name);
}
parser->explicit_encoding = PM_ENCODING_UTF_8_ENTRY;
}
if (value <= 0x7F) { // 0xxxxxxx
pm_buffer_append_byte(buffer, (uint8_t) value);
} else if (value <= 0x7FF) { // 110xxxxx 10xxxxxx
pm_buffer_append_byte(buffer, (uint8_t) (0xC0 | (value >> 6)));
pm_buffer_append_byte(buffer, (uint8_t) (0x80 | (value & 0x3F)));
} else if (value <= 0xFFFF) { // 1110xxxx 10xxxxxx 10xxxxxx
pm_buffer_append_byte(buffer, (uint8_t) (0xE0 | (value >> 12)));
pm_buffer_append_byte(buffer, (uint8_t) (0x80 | ((value >> 6) & 0x3F)));
pm_buffer_append_byte(buffer, (uint8_t) (0x80 | (value & 0x3F)));
} else if (value <= 0x10FFFF) { // 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx
pm_buffer_append_byte(buffer, (uint8_t) (0xF0 | (value >> 18)));
pm_buffer_append_byte(buffer, (uint8_t) (0x80 | ((value >> 12) & 0x3F)));
pm_buffer_append_byte(buffer, (uint8_t) (0x80 | ((value >> 6) & 0x3F)));
pm_buffer_append_byte(buffer, (uint8_t) (0x80 | (value & 0x3F)));
} else {
pm_parser_err(parser, start, end, PM_ERR_ESCAPE_INVALID_UNICODE);
pm_buffer_append_byte(buffer, 0xEF);
pm_buffer_append_byte(buffer, 0xBF);
pm_buffer_append_byte(buffer, 0xBD);
}
}
/**
* When you're writing a byte to the unescape buffer, if the byte is non-ASCII
* (i.e., the top bit is set) then it locks in the encoding.
*/
static inline void
escape_write_byte_encoded(pm_parser_t *parser, pm_buffer_t *buffer, uint8_t byte) {
if (byte >= 0x80) {
if (parser->explicit_encoding != NULL && parser->explicit_encoding == PM_ENCODING_UTF_8_ENTRY && parser->encoding != PM_ENCODING_UTF_8_ENTRY) {
PM_PARSER_ERR_TOKEN_FORMAT(parser, parser->current, PM_ERR_MIXED_ENCODING, parser->encoding->name);
}
parser->explicit_encoding = parser->encoding;
}
pm_buffer_append_byte(buffer, byte);
}
/**
* The regular expression engine doesn't support the same escape sequences as
* Ruby does. So first we have to read the escape sequence, and then we have to
* format it like the regular expression engine expects it. For example, in Ruby
* if we have:
*
* /\M-\C-?/
*
* then the first byte is actually 255, so we have to rewrite this as:
*
* /\xFF/
*
* Note that in this case there is a literal \ byte in the regular expression
* source so that the regular expression engine will perform its own unescaping.
*/
static inline void
escape_write_byte(pm_parser_t *parser, pm_buffer_t *buffer, uint8_t flags, uint8_t byte) {
if (flags & PM_ESCAPE_FLAG_REGEXP) {
pm_buffer_append_bytes(buffer, (const uint8_t *) "\\x", 2);
uint8_t byte1 = (uint8_t) ((byte >> 4) & 0xF);
uint8_t byte2 = (uint8_t) (byte & 0xF);
if (byte1 >= 0xA) {
pm_buffer_append_byte(buffer, (uint8_t) ((byte1 - 0xA) + 'A'));
} else {
pm_buffer_append_byte(buffer, (uint8_t) (byte1 + '0'));
}
if (byte2 >= 0xA) {
pm_buffer_append_byte(buffer, (uint8_t) (byte2 - 0xA + 'A'));
} else {
pm_buffer_append_byte(buffer, (uint8_t) (byte2 + '0'));
}
} else {
escape_write_byte_encoded(parser, buffer, byte);
}
}
/**
* Read the value of an escape into the buffer.
*/
static void
escape_read(pm_parser_t *parser, pm_buffer_t *buffer, uint8_t flags) {
switch (peek(parser)) {
case '\\': {
parser->current.end++;
escape_write_byte_encoded(parser, buffer, escape_byte('\\', flags));
return;
}
case '\'': {
parser->current.end++;
escape_write_byte_encoded(parser, buffer, escape_byte('\'', flags));
return;
}
case 'a': {
parser->current.end++;
escape_write_byte_encoded(parser, buffer, escape_byte('\a', flags));
return;
}
case 'b': {
parser->current.end++;
escape_write_byte_encoded(parser, buffer, escape_byte('\b', flags));
return;
}
case 'e': {
parser->current.end++;
escape_write_byte_encoded(parser, buffer, escape_byte('\033', flags));
return;
}
case 'f': {
parser->current.end++;
escape_write_byte_encoded(parser, buffer, escape_byte('\f', flags));
return;
}
case 'n': {
parser->current.end++;
escape_write_byte_encoded(parser, buffer, escape_byte('\n', flags));
return;
}
case 'r': {
parser->current.end++;
escape_write_byte_encoded(parser, buffer, escape_byte('\r', flags));
return;
}
case 's': {
parser->current.end++;
escape_write_byte_encoded(parser, buffer, escape_byte(' ', flags));
return;
}
case 't': {
parser->current.end++;
escape_write_byte_encoded(parser, buffer, escape_byte('\t', flags));
return;
}
case 'v': {
parser->current.end++;
escape_write_byte_encoded(parser, buffer, escape_byte('\v', flags));
return;
}
case '0': case '1': case '2': case '3': case '4': case '5': case '6': case '7': {
uint8_t value = (uint8_t) (*parser->current.end - '0');
parser->current.end++;
if (pm_char_is_octal_digit(peek(parser))) {
value = ((uint8_t) (value << 3)) | ((uint8_t) (*parser->current.end - '0'));
parser->current.end++;
if (pm_char_is_octal_digit(peek(parser))) {
value = ((uint8_t) (value << 3)) | ((uint8_t) (*parser->current.end - '0'));
parser->current.end++;
}
}
escape_write_byte_encoded(parser, buffer, value);
return;
}
case 'x': {
const uint8_t *start = parser->current.end - 1;
parser->current.end++;
uint8_t byte = peek(parser);
if (pm_char_is_hexadecimal_digit(byte)) {
uint8_t value = escape_hexadecimal_digit(byte);
parser->current.end++;
byte = peek(parser);
if (pm_char_is_hexadecimal_digit(byte)) {
value = (uint8_t) ((value << 4) | escape_hexadecimal_digit(byte));
parser->current.end++;
}
if (flags & PM_ESCAPE_FLAG_REGEXP) {
pm_buffer_append_bytes(buffer, start, (size_t) (parser->current.end - start));
} else {
escape_write_byte_encoded(parser, buffer, value);
}
} else {
pm_parser_err_current(parser, PM_ERR_ESCAPE_INVALID_HEXADECIMAL);
}
return;
}
case 'u': {
const uint8_t *start = parser->current.end - 1;
parser->current.end++;
if (
(parser->current.end + 4 <= parser->end) &&
pm_char_is_hexadecimal_digit(parser->current.end[0]) &&
pm_char_is_hexadecimal_digit(parser->current.end[1]) &&
pm_char_is_hexadecimal_digit(parser->current.end[2]) &&
pm_char_is_hexadecimal_digit(parser->current.end[3])
) {
uint32_t value = escape_unicode(parser->current.end, 4);
if (flags & PM_ESCAPE_FLAG_REGEXP) {
pm_buffer_append_bytes(buffer, start, (size_t) (parser->current.end + 4 - start));
} else {
escape_write_unicode(parser, buffer, flags, start, parser->current.end + 4, value);
}
parser->current.end += 4;
} else if (peek(parser) == '{') {
const uint8_t *unicode_codepoints_start = parser->current.end - 2;
parser->current.end++;
parser->current.end += pm_strspn_whitespace(parser->current.end, parser->end - parser->current.end);
const uint8_t *extra_codepoints_start = NULL;
int codepoints_count = 0;
while ((parser->current.end < parser->end) && (*parser->current.end != '}')) {
const uint8_t *unicode_start = parser->current.end;
size_t hexadecimal_length = pm_strspn_hexadecimal_digit(parser->current.end, parser->end - parser->current.end);
if (hexadecimal_length > 6) {
// \u{nnnn} character literal allows only 1-6 hexadecimal digits
pm_parser_err(parser, unicode_start, unicode_start + hexadecimal_length, PM_ERR_ESCAPE_INVALID_UNICODE_LONG);
} else if (hexadecimal_length == 0) {
// there are not hexadecimal characters
pm_parser_err(parser, unicode_start, unicode_start + hexadecimal_length, PM_ERR_ESCAPE_INVALID_UNICODE);
return;
}
parser->current.end += hexadecimal_length;
codepoints_count++;
if (flags & PM_ESCAPE_FLAG_SINGLE && codepoints_count == 2) {
extra_codepoints_start = unicode_start;
}
if (!(flags & PM_ESCAPE_FLAG_REGEXP)) {
uint32_t value = escape_unicode(unicode_start, hexadecimal_length);
escape_write_unicode(parser, buffer, flags, unicode_start, parser->current.end, value);
}
parser->current.end += pm_strspn_whitespace(parser->current.end, parser->end - parser->current.end);
}
// ?\u{nnnn} character literal should contain only one codepoint
// and cannot be like ?\u{nnnn mmmm}.
if (flags & PM_ESCAPE_FLAG_SINGLE && codepoints_count > 1) {
pm_parser_err(parser, extra_codepoints_start, parser->current.end - 1, PM_ERR_ESCAPE_INVALID_UNICODE_LITERAL);
}
if (peek(parser) == '}') {
parser->current.end++;
} else {
pm_parser_err(parser, unicode_codepoints_start, parser->current.end, PM_ERR_ESCAPE_INVALID_UNICODE_TERM);
}
if (flags & PM_ESCAPE_FLAG_REGEXP) {
pm_buffer_append_bytes(buffer, unicode_codepoints_start, (size_t) (parser->current.end - unicode_codepoints_start));
}
} else {
pm_parser_err_current(parser, PM_ERR_ESCAPE_INVALID_UNICODE);
}
return;
}
case 'c': {
parser->current.end++;
if (parser->current.end == parser->end) {
pm_parser_err_current(parser, PM_ERR_ESCAPE_INVALID_CONTROL);
return;
}
uint8_t peeked = peek(parser);
switch (peeked) {
case '?': {
parser->current.end++;
escape_write_byte(parser, buffer, flags, escape_byte(0x7f, flags));
return;
}
case '\\':
if (flags & PM_ESCAPE_FLAG_CONTROL) {
pm_parser_err_current(parser, PM_ERR_ESCAPE_INVALID_CONTROL_REPEAT);
return;
}
parser->current.end++;
escape_read(parser, buffer, flags | PM_ESCAPE_FLAG_CONTROL);
return;
default: {
if (!char_is_ascii_printable(peeked)) {
pm_parser_err_current(parser, PM_ERR_ESCAPE_INVALID_CONTROL);
return;
}
parser->current.end++;
escape_write_byte(parser, buffer, flags, escape_byte(peeked, flags | PM_ESCAPE_FLAG_CONTROL));
return;
}
}
}
case 'C': {
parser->current.end++;
if (peek(parser) != '-') {
pm_parser_err_current(parser, PM_ERR_ESCAPE_INVALID_CONTROL);
return;
}
parser->current.end++;
if (parser->current.end == parser->end) {
pm_parser_err_current(parser, PM_ERR_ESCAPE_INVALID_CONTROL);
return;
}
uint8_t peeked = peek(parser);
switch (peeked) {
case '?': {
parser->current.end++;
escape_write_byte(parser, buffer, flags, escape_byte(0x7f, flags));
return;
}
case '\\':
if (flags & PM_ESCAPE_FLAG_CONTROL) {
pm_parser_err_current(parser, PM_ERR_ESCAPE_INVALID_CONTROL_REPEAT);
return;
}
parser->current.end++;
escape_read(parser, buffer, flags | PM_ESCAPE_FLAG_CONTROL);
return;
default: {
if (!char_is_ascii_printable(peeked)) {
pm_parser_err_current(parser, PM_ERR_ESCAPE_INVALID_CONTROL);
return;
}
parser->current.end++;
escape_write_byte(parser, buffer, flags, escape_byte(peeked, flags | PM_ESCAPE_FLAG_CONTROL));
return;
}
}
}
case 'M': {
parser->current.end++;
if (peek(parser) != '-') {
pm_parser_err_current(parser, PM_ERR_ESCAPE_INVALID_META);
return;
}
parser->current.end++;
if (parser->current.end == parser->end) {
pm_parser_err_current(parser, PM_ERR_ESCAPE_INVALID_META);
return;
}
uint8_t peeked = peek(parser);
if (peeked == '\\') {
if (flags & PM_ESCAPE_FLAG_META) {
pm_parser_err_current(parser, PM_ERR_ESCAPE_INVALID_META_REPEAT);
return;
}
parser->current.end++;
escape_read(parser, buffer, flags | PM_ESCAPE_FLAG_META);
return;
}
if (!char_is_ascii_printable(peeked)) {
pm_parser_err_current(parser, PM_ERR_ESCAPE_INVALID_META);
return;
}
parser->current.end++;
escape_write_byte(parser, buffer, flags, escape_byte(peeked, flags | PM_ESCAPE_FLAG_META));
return;
}
case '\r': {
if (peek_offset(parser, 1) == '\n') {
parser->current.end += 2;
escape_write_byte_encoded(parser, buffer, escape_byte('\n', flags));
return;
}
}
/* fallthrough */
default: {
if (parser->current.end < parser->end) {
escape_write_byte_encoded(parser, buffer, *parser->current.end++);
}
return;
}
}
}
/**
* This function is responsible for lexing either a character literal or the ?
* operator. The supported character literals are described below.
*
* \\a bell, ASCII 07h (BEL)
* \\b backspace, ASCII 08h (BS)
* \t horizontal tab, ASCII 09h (TAB)
* \\n newline (line feed), ASCII 0Ah (LF)
* \v vertical tab, ASCII 0Bh (VT)
* \f form feed, ASCII 0Ch (FF)
* \r carriage return, ASCII 0Dh (CR)
* \\e escape, ASCII 1Bh (ESC)
* \s space, ASCII 20h (SPC)
* \\ backslash
* \nnn octal bit pattern, where nnn is 1-3 octal digits ([0-7])
* \xnn hexadecimal bit pattern, where nn is 1-2 hexadecimal digits ([0-9a-fA-F])
* \unnnn Unicode character, where nnnn is exactly 4 hexadecimal digits ([0-9a-fA-F])
* \u{nnnn ...} Unicode character(s), where each nnnn is 1-6 hexadecimal digits ([0-9a-fA-F])
* \cx or \C-x control character, where x is an ASCII printable character
* \M-x meta character, where x is an ASCII printable character
* \M-\C-x meta control character, where x is an ASCII printable character
* \M-\cx same as above
* \\c\M-x same as above
* \\c? or \C-? delete, ASCII 7Fh (DEL)
*/
static pm_token_type_t
lex_question_mark(pm_parser_t *parser) {
if (lex_state_end_p(parser)) {
lex_state_set(parser, PM_LEX_STATE_BEG);
return PM_TOKEN_QUESTION_MARK;
}
if (parser->current.end >= parser->end) {
pm_parser_err_current(parser, PM_ERR_INCOMPLETE_QUESTION_MARK);
pm_string_shared_init(&parser->current_string, parser->current.start + 1, parser->current.end);
return PM_TOKEN_CHARACTER_LITERAL;
}
if (pm_char_is_whitespace(*parser->current.end)) {
lex_state_set(parser, PM_LEX_STATE_BEG);
return PM_TOKEN_QUESTION_MARK;
}
lex_state_set(parser, PM_LEX_STATE_BEG);
if (match(parser, '\\')) {
lex_state_set(parser, PM_LEX_STATE_END);
pm_buffer_t buffer;
pm_buffer_init_capacity(&buffer, 3);
escape_read(parser, &buffer, PM_ESCAPE_FLAG_SINGLE);
pm_string_owned_init(&parser->current_string, (uint8_t *) buffer.value, buffer.length);
return PM_TOKEN_CHARACTER_LITERAL;
} else {
size_t encoding_width = parser->encoding->char_width(parser->current.end, parser->end - parser->current.end);
// Ternary operators can have a ? immediately followed by an identifier
// which starts with an underscore. We check for this case here.
if (
!(parser->encoding->alnum_char(parser->current.end, parser->end - parser->current.end) || peek(parser) == '_') ||
(
(parser->current.end + encoding_width >= parser->end) ||
!char_is_identifier(parser, parser->current.end + encoding_width)
)
) {
lex_state_set(parser, PM_LEX_STATE_END);
parser->current.end += encoding_width;
pm_string_shared_init(&parser->current_string, parser->current.start + 1, parser->current.end);
return PM_TOKEN_CHARACTER_LITERAL;
}
}
return PM_TOKEN_QUESTION_MARK;
}
/**
* Lex a variable that starts with an @ sign (either an instance or class
* variable).
*/
static pm_token_type_t
lex_at_variable(pm_parser_t *parser) {
pm_token_type_t type = match(parser, '@') ? PM_TOKEN_CLASS_VARIABLE : PM_TOKEN_INSTANCE_VARIABLE;
size_t width;
if (parser->current.end < parser->end && (width = char_is_identifier_start(parser, parser->current.end)) > 0) {
parser->current.end += width;
while (parser->current.end < parser->end && (width = char_is_identifier(parser, parser->current.end)) > 0) {
parser->current.end += width;
}
} else if (type == PM_TOKEN_CLASS_VARIABLE) {
pm_parser_err_current(parser, PM_ERR_INCOMPLETE_VARIABLE_CLASS);
} else {
pm_parser_err_current(parser, PM_ERR_INCOMPLETE_VARIABLE_INSTANCE);
}
// If we're lexing an embedded variable, then we need to pop back into the
// parent lex context.
if (parser->lex_modes.current->mode == PM_LEX_EMBVAR) {
lex_mode_pop(parser);
}
return type;
}
/**
* Optionally call out to the lex callback if one is provided.
*/
static inline void
parser_lex_callback(pm_parser_t *parser) {
if (parser->lex_callback) {
parser->lex_callback->callback(parser->lex_callback->data, parser, &parser->current);
}
}
/**
* Return a new comment node of the specified type.
*/
static inline pm_comment_t *
parser_comment(pm_parser_t *parser, pm_comment_type_t type) {
pm_comment_t *comment = (pm_comment_t *) calloc(sizeof(pm_comment_t), 1);
if (comment == NULL) return NULL;
*comment = (pm_comment_t) {
.type = type,
.location = { parser->current.start, parser->current.end }
};
return comment;
}
/**
* Lex out embedded documentation, and return when we have either hit the end of
* the file or the end of the embedded documentation. This calls the callback
* manually because only the lexer should see these tokens, not the parser.
*/
static pm_token_type_t
lex_embdoc(pm_parser_t *parser) {
// First, lex out the EMBDOC_BEGIN token.
const uint8_t *newline = next_newline(parser->current.end, parser->end - parser->current.end);
if (newline == NULL) {
parser->current.end = parser->end;
} else {
pm_newline_list_append(&parser->newline_list, newline);
parser->current.end = newline + 1;
}
parser->current.type = PM_TOKEN_EMBDOC_BEGIN;
parser_lex_callback(parser);
// Now, create a comment that is going to be attached to the parser.
pm_comment_t *comment = parser_comment(parser, PM_COMMENT_EMBDOC);
if (comment == NULL) return PM_TOKEN_EOF;
// Now, loop until we find the end of the embedded documentation or the end of
// the file.
while (parser->current.end + 4 <= parser->end) {
parser->current.start = parser->current.end;
// If we've hit the end of the embedded documentation then we'll return that
// token here.
if (memcmp(parser->current.end, "=end", 4) == 0 &&
(parser->current.end + 4 == parser->end || pm_char_is_whitespace(parser->current.end[4]))) {
const uint8_t *newline = next_newline(parser->current.end, parser->end - parser->current.end);
if (newline == NULL) {
parser->current.end = parser->end;
} else {
pm_newline_list_append(&parser->newline_list, newline);
parser->current.end = newline + 1;
}
parser->current.type = PM_TOKEN_EMBDOC_END;
parser_lex_callback(parser);
comment->location.end = parser->current.end;
pm_list_append(&parser->comment_list, (pm_list_node_t *) comment);
return PM_TOKEN_EMBDOC_END;
}
// Otherwise, we'll parse until the end of the line and return a line of
// embedded documentation.
const uint8_t *newline = next_newline(parser->current.end, parser->end - parser->current.end);
if (newline == NULL) {
parser->current.end = parser->end;
} else {
pm_newline_list_append(&parser->newline_list, newline);
parser->current.end = newline + 1;
}
parser->current.type = PM_TOKEN_EMBDOC_LINE;
parser_lex_callback(parser);
}
pm_parser_err_current(parser, PM_ERR_EMBDOC_TERM);
comment->location.end = parser->current.end;
pm_list_append(&parser->comment_list, (pm_list_node_t *) comment);
return PM_TOKEN_EOF;
}
/**
* Set the current type to an ignored newline and then call the lex callback.
* This happens in a couple places depending on whether or not we have already
* lexed a comment.
*/
static inline void
parser_lex_ignored_newline(pm_parser_t *parser) {
parser->current.type = PM_TOKEN_IGNORED_NEWLINE;
parser_lex_callback(parser);
}
/**
* This function will be called when a newline is encountered. In some newlines,
* we need to check if there is a heredoc or heredocs that we have already lexed
* the body of that we need to now skip past. That will be indicated by the
* heredoc_end field on the parser.
*
* If it is set, then we need to skip past the heredoc body and then clear the
* heredoc_end field.
*/
static inline void
parser_flush_heredoc_end(pm_parser_t *parser) {
assert(parser->heredoc_end <= parser->end);
parser->next_start = parser->heredoc_end;
parser->heredoc_end = NULL;
}
/**
* When we're lexing certain types (strings, symbols, lists, etc.) we have
* string content associated with the tokens. For example:
*
* "foo"
*
* In this case, the string content is foo. Since there is no escaping, there's
* no need to track additional information and the token can be returned as
* normal. However, if we have escape sequences:
*
* "foo\n"
*
* then the bytes in the string are "f", "o", "o", "\", "n", but we want to
* provide out consumers with the string content "f", "o", "o", "\n". In these
* cases, when we find the first escape sequence, we initialize a pm_buffer_t
* to keep track of the string content. Then in the parser, it will
* automatically attach the string content to the node that it belongs to.
*/
typedef struct {
/**
* The buffer that we're using to keep track of the string content. It will
* only be initialized if we receive an escape sequence.
*/
pm_buffer_t buffer;
/**
* The cursor into the source string that points to how far we have
* currently copied into the buffer.
*/
const uint8_t *cursor;
} pm_token_buffer_t;
/**
* Push the given byte into the token buffer.
*/
static inline void
pm_token_buffer_push(pm_token_buffer_t *token_buffer, uint8_t byte) {
pm_buffer_append_byte(&token_buffer->buffer, byte);
}
/**
* When we're about to return from lexing the current token and we know for sure
* that we have found an escape sequence, this function is called to copy the
*
* contents of the token buffer into the current string on the parser so that it
* can be attached to the correct node.
*/
static inline void
pm_token_buffer_copy(pm_parser_t *parser, pm_token_buffer_t *token_buffer) {
pm_string_owned_init(&parser->current_string, (uint8_t *) token_buffer->buffer.value, token_buffer->buffer.length);
}
/**
* When we're about to return from lexing the current token, we need to flush
* all of the content that we have pushed into the buffer into the current
* string. If we haven't pushed anything into the buffer, this means that we
* never found an escape sequence, so we can directly reference the bounds of
* the current string. Either way, at the return of this function it is expected
*
* that parser->current_string is established in such a way that it can be
* attached to a node.
*/
static void
pm_token_buffer_flush(pm_parser_t *parser, pm_token_buffer_t *token_buffer) {
if (token_buffer->cursor == NULL) {
pm_string_shared_init(&parser->current_string, parser->current.start, parser->current.end);
} else {
pm_buffer_append_bytes(&token_buffer->buffer, token_buffer->cursor, (size_t) (parser->current.end - token_buffer->cursor));
pm_token_buffer_copy(parser, token_buffer);
}
}
/**
* When we've found an escape sequence, we need to copy everything up to this
* point into the buffer because we're about to provide a string that has
* different content than a direct slice of the source.
*
*
* It is expected that the parser's current token end will be pointing at one
* byte past the backslash that starts the escape sequence.
*/
static void
pm_token_buffer_escape(pm_parser_t *parser, pm_token_buffer_t *token_buffer) {
const uint8_t *start;
if (token_buffer->cursor == NULL) {
pm_buffer_init_capacity(&token_buffer->buffer, 16);
start = parser->current.start;
} else {
start = token_buffer->cursor;
}
const uint8_t *end = parser->current.end - 1;
pm_buffer_append_bytes(&token_buffer->buffer, start, (size_t) (end - start));
}
/**
* Effectively the same thing as pm_strspn_inline_whitespace, but in the case of
* a tilde heredoc expands out tab characters to the nearest tab boundaries.
*/
static inline size_t
pm_heredoc_strspn_inline_whitespace(pm_parser_t *parser, const uint8_t **cursor, pm_heredoc_indent_t indent) {
size_t whitespace = 0;
switch (indent) {
case PM_HEREDOC_INDENT_NONE:
// Do nothing, we can't match a terminator with
// indentation and there's no need to calculate common
// whitespace.
break;
case PM_HEREDOC_INDENT_DASH:
// Skip past inline whitespace.
*cursor += pm_strspn_inline_whitespace(*cursor, parser->end - *cursor);
break;
case PM_HEREDOC_INDENT_TILDE:
// Skip past inline whitespace and calculate common
// whitespace.
while (*cursor < parser->end && pm_char_is_inline_whitespace(**cursor)) {
if (**cursor == '\t') {
whitespace = (whitespace / PM_TAB_WHITESPACE_SIZE + 1) * PM_TAB_WHITESPACE_SIZE;
} else {
whitespace++;
}
(*cursor)++;
}
break;
}
return whitespace;
}
/**
* This is a convenience macro that will set the current token type, call the
* lex callback, and then return from the parser_lex function.
*/
#define LEX(token_type) parser->current.type = token_type; parser_lex_callback(parser); return
/**
* Called when the parser requires a new token. The parser maintains a moving
* window of two tokens at a time: parser.previous and parser.current. This
* function will move the current token into the previous token and then
* lex a new token into the current token.
*/
static void
parser_lex(pm_parser_t *parser) {
assert(parser->current.end <= parser->end);
parser->previous = parser->current;
// This value mirrors cmd_state from CRuby.
bool previous_command_start = parser->command_start;
parser->command_start = false;
// This is used to communicate to the newline lexing function that we've
// already seen a comment.
bool lexed_comment = false;
// Here we cache the current value of the semantic token seen flag. This is
// used to reset it in case we find a token that shouldn't flip this flag.
unsigned int semantic_token_seen = parser->semantic_token_seen;
parser->semantic_token_seen = true;
switch (parser->lex_modes.current->mode) {
case PM_LEX_DEFAULT:
case PM_LEX_EMBEXPR:
case PM_LEX_EMBVAR:
// We have a specific named label here because we are going to jump back to
// this location in the event that we have lexed a token that should not be
// returned to the parser. This includes comments, ignored newlines, and
// invalid tokens of some form.
lex_next_token: {
// If we have the special next_start pointer set, then we're going to jump
// to that location and start lexing from there.
if (parser->next_start != NULL) {
parser->current.end = parser->next_start;
parser->next_start = NULL;
}
// This value mirrors space_seen from CRuby. It tracks whether or not
// space has been eaten before the start of the next token.
bool space_seen = false;
// First, we're going to skip past any whitespace at the front of the next
// token.
bool chomping = true;
while (parser->current.end < parser->end && chomping) {
switch (*parser->current.end) {
case ' ':
case '\t':
case '\f':
case '\v':
parser->current.end++;
space_seen = true;
break;
case '\r':
if (match_eol_offset(parser, 1)) {
chomping = false;
} else {
parser->current.end++;
space_seen = true;
}
break;
case '\\': {
size_t eol_length = match_eol_offset(parser, 1);
if (eol_length) {
if (parser->heredoc_end) {
parser->current.end = parser->heredoc_end;
parser->heredoc_end = NULL;
} else {
parser->current.end += eol_length + 1;
pm_newline_list_append(&parser->newline_list, parser->current.end - 1);
space_seen = true;
}
} else if (pm_char_is_inline_whitespace(*parser->current.end)) {
parser->current.end += 2;
} else {
chomping = false;
}
break;
}
default:
chomping = false;
break;
}
}
// Next, we'll set to start of this token to be the current end.
parser->current.start = parser->current.end;
// We'll check if we're at the end of the file. If we are, then we
// need to return the EOF token.
if (parser->current.end >= parser->end) {
LEX(PM_TOKEN_EOF);
}
// Finally, we'll check the current character to determine the next
// token.
switch (*parser->current.end++) {
case '\0': // NUL or end of script
case '\004': // ^D
case '\032': // ^Z
parser->current.end--;
LEX(PM_TOKEN_EOF);
case '#': { // comments
const uint8_t *ending = next_newline(parser->current.end, parser->end - parser->current.end);
parser->current.end = ending == NULL ? parser->end : ending;
// If we found a comment while lexing, then we're going to
// add it to the list of comments in the file and keep
// lexing.
pm_comment_t *comment = parser_comment(parser, PM_COMMENT_INLINE);
pm_list_append(&parser->comment_list, (pm_list_node_t *) comment);
if (ending) parser->current.end++;
parser->current.type = PM_TOKEN_COMMENT;
parser_lex_callback(parser);
// Here, parse the comment to see if it's a magic comment
// and potentially change state on the parser.
if (!parser_lex_magic_comment(parser, semantic_token_seen) && (parser->current.start == parser->encoding_comment_start)) {
ptrdiff_t length = parser->current.end - parser->current.start;
// If we didn't find a magic comment within the first
// pass and we're at the start of the file, then we need
// to do another pass to potentially find other patterns
// for encoding comments.
if (length >= 10) parser_lex_magic_comment_encoding(parser);
}
lexed_comment = true;
}
/* fallthrough */
case '\r':
case '\n': {
parser->semantic_token_seen = semantic_token_seen & 0x1;
size_t eol_length = match_eol_at(parser, parser->current.end - 1);
if (eol_length) {
// The only way you can have carriage returns in this
// particular loop is if you have a carriage return
// followed by a newline. In that case we'll just skip
// over the carriage return and continue lexing, in
// order to make it so that the newline token
// encapsulates both the carriage return and the
// newline. Note that we need to check that we haven't
// already lexed a comment here because that falls
// through into here as well.
if (!lexed_comment) {
parser->current.end += eol_length - 1; // skip CR
}
if (parser->heredoc_end == NULL) {
pm_newline_list_append(&parser->newline_list, parser->current.end - 1);
}
}
if (parser->heredoc_end) {
parser_flush_heredoc_end(parser);
}
// If this is an ignored newline, then we can continue lexing after
// calling the callback with the ignored newline token.
switch (lex_state_ignored_p(parser)) {
case PM_IGNORED_NEWLINE_NONE:
break;
case PM_IGNORED_NEWLINE_PATTERN:
if (parser->pattern_matching_newlines || parser->in_keyword_arg) {
if (!lexed_comment) parser_lex_ignored_newline(parser);
lex_state_set(parser, PM_LEX_STATE_BEG);
parser->command_start = true;
parser->current.type = PM_TOKEN_NEWLINE;
return;
}
/* fallthrough */
case PM_IGNORED_NEWLINE_ALL:
if (!lexed_comment) parser_lex_ignored_newline(parser);
lexed_comment = false;
goto lex_next_token;
}
// Here we need to look ahead and see if there is a call operator
// (either . or &.) that starts the next line. If there is, then this
// is going to become an ignored newline and we're going to instead
// return the call operator.
const uint8_t *next_content = parser->next_start == NULL ? parser->current.end : parser->next_start;
next_content += pm_strspn_inline_whitespace(next_content, parser->end - next_content);
if (next_content < parser->end) {
// If we hit a comment after a newline, then we're going to check
// if it's ignored or if it's followed by a method call ('.').
// If it is, then we're going to call the
// callback with an ignored newline and then continue lexing.
// Otherwise we'll return a regular newline.
if (next_content[0] == '#') {
// Here we look for a "." or "&." following a "\n".
const uint8_t *following = next_newline(next_content, parser->end - next_content);
while (following && (following + 1 < parser->end)) {
following++;
following += pm_strspn_inline_whitespace(following, parser->end - following);
// If this is not followed by a comment, then we can break out
// of this loop.
if (peek_at(parser, following) != '#') break;
// If there is a comment, then we need to find the end of the
// comment and continue searching from there.
following = next_newline(following, parser->end - following);
}
// If the lex state was ignored, or we hit a '.' or a '&.',
// we will lex the ignored newline
if (
lex_state_ignored_p(parser) ||
(following && (
(peek_at(parser, following) == '.') ||
(peek_at(parser, following) == '&' && peek_at(parser, following + 1) == '.')
))
) {
if (!lexed_comment) parser_lex_ignored_newline(parser);
lexed_comment = false;
goto lex_next_token;
}
}
// If we hit a . after a newline, then we're in a call chain and
// we need to return the call operator.
if (next_content[0] == '.') {
// To match ripper, we need to emit an ignored newline even though
// its a real newline in the case that we have a beginless range
// on a subsequent line.
if (peek_at(parser, next_content + 1) == '.') {
if (!lexed_comment) parser_lex_ignored_newline(parser);
lex_state_set(parser, PM_LEX_STATE_BEG);
parser->command_start = true;
parser->current.type = PM_TOKEN_NEWLINE;
return;
}
if (!lexed_comment) parser_lex_ignored_newline(parser);
lex_state_set(parser, PM_LEX_STATE_DOT);
parser->current.start = next_content;
parser->current.end = next_content + 1;
parser->next_start = NULL;
LEX(PM_TOKEN_DOT);
}
// If we hit a &. after a newline, then we're in a call chain and
// we need to return the call operator.
if (peek_at(parser, next_content) == '&' && peek_at(parser, next_content + 1) == '.') {
if (!lexed_comment) parser_lex_ignored_newline(parser);
lex_state_set(parser, PM_LEX_STATE_DOT);
parser->current.start = next_content;
parser->current.end = next_content + 2;
parser->next_start = NULL;
LEX(PM_TOKEN_AMPERSAND_DOT);
}
}
// At this point we know this is a regular newline, and we can set the
// necessary state and return the token.
lex_state_set(parser, PM_LEX_STATE_BEG);
parser->command_start = true;
parser->current.type = PM_TOKEN_NEWLINE;
if (!lexed_comment) parser_lex_callback(parser);
return;
}
// ,
case ',':
lex_state_set(parser, PM_LEX_STATE_BEG | PM_LEX_STATE_LABEL);
LEX(PM_TOKEN_COMMA);
// (
case '(': {
pm_token_type_t type = PM_TOKEN_PARENTHESIS_LEFT;
if (space_seen && (lex_state_arg_p(parser) || parser->lex_state == (PM_LEX_STATE_END | PM_LEX_STATE_LABEL))) {
type = PM_TOKEN_PARENTHESIS_LEFT_PARENTHESES;
}
parser->enclosure_nesting++;
lex_state_set(parser, PM_LEX_STATE_BEG | PM_LEX_STATE_LABEL);
pm_do_loop_stack_push(parser, false);
LEX(type);
}
// )
case ')':
parser->enclosure_nesting--;
lex_state_set(parser, PM_LEX_STATE_ENDFN);
pm_do_loop_stack_pop(parser);
LEX(PM_TOKEN_PARENTHESIS_RIGHT);
// ;
case ';':
lex_state_set(parser, PM_LEX_STATE_BEG);
parser->command_start = true;
LEX(PM_TOKEN_SEMICOLON);
// [ [] []=
case '[':
parser->enclosure_nesting++;
pm_token_type_t type = PM_TOKEN_BRACKET_LEFT;
if (lex_state_operator_p(parser)) {
if (match(parser, ']')) {
parser->enclosure_nesting--;
lex_state_set(parser, PM_LEX_STATE_ARG);
LEX(match(parser, '=') ? PM_TOKEN_BRACKET_LEFT_RIGHT_EQUAL : PM_TOKEN_BRACKET_LEFT_RIGHT);
}
lex_state_set(parser, PM_LEX_STATE_ARG | PM_LEX_STATE_LABEL);
LEX(type);
}
if (lex_state_beg_p(parser) || (lex_state_arg_p(parser) && (space_seen || lex_state_p(parser, PM_LEX_STATE_LABELED)))) {
type = PM_TOKEN_BRACKET_LEFT_ARRAY;
}
lex_state_set(parser, PM_LEX_STATE_BEG | PM_LEX_STATE_LABEL);
pm_do_loop_stack_push(parser, false);
LEX(type);
// ]
case ']':
parser->enclosure_nesting--;
lex_state_set(parser, PM_LEX_STATE_END);
pm_do_loop_stack_pop(parser);
LEX(PM_TOKEN_BRACKET_RIGHT);
// {
case '{': {
pm_token_type_t type = PM_TOKEN_BRACE_LEFT;
if (parser->enclosure_nesting == parser->lambda_enclosure_nesting) {
// This { begins a lambda
parser->command_start = true;
lex_state_set(parser, PM_LEX_STATE_BEG);
type = PM_TOKEN_LAMBDA_BEGIN;
} else if (lex_state_p(parser, PM_LEX_STATE_LABELED)) {
// This { begins a hash literal
lex_state_set(parser, PM_LEX_STATE_BEG | PM_LEX_STATE_LABEL);
} else if (lex_state_p(parser, PM_LEX_STATE_ARG_ANY | PM_LEX_STATE_END | PM_LEX_STATE_ENDFN)) {
// This { begins a block
parser->command_start = true;
lex_state_set(parser, PM_LEX_STATE_BEG);
} else if (lex_state_p(parser, PM_LEX_STATE_ENDARG)) {
// This { begins a block on a command
parser->command_start = true;
lex_state_set(parser, PM_LEX_STATE_BEG);
} else {
// This { begins a hash literal
lex_state_set(parser, PM_LEX_STATE_BEG | PM_LEX_STATE_LABEL);
}
parser->enclosure_nesting++;
parser->brace_nesting++;
pm_do_loop_stack_push(parser, false);
LEX(type);
}
// }
case '}':
parser->enclosure_nesting--;
pm_do_loop_stack_pop(parser);
if ((parser->lex_modes.current->mode == PM_LEX_EMBEXPR) && (parser->brace_nesting == 0)) {
lex_mode_pop(parser);
LEX(PM_TOKEN_EMBEXPR_END);
}
parser->brace_nesting--;
lex_state_set(parser, PM_LEX_STATE_END);
LEX(PM_TOKEN_BRACE_RIGHT);
// * ** **= *=
case '*': {
if (match(parser, '*')) {
if (match(parser, '=')) {
lex_state_set(parser, PM_LEX_STATE_BEG);
LEX(PM_TOKEN_STAR_STAR_EQUAL);
}
pm_token_type_t type = PM_TOKEN_STAR_STAR;
if (lex_state_spcarg_p(parser, space_seen) || lex_state_beg_p(parser)) {
type = PM_TOKEN_USTAR_STAR;
}
if (lex_state_operator_p(parser)) {
lex_state_set(parser, PM_LEX_STATE_ARG);
} else {
lex_state_set(parser, PM_LEX_STATE_BEG);
}
LEX(type);
}
if (match(parser, '=')) {
lex_state_set(parser, PM_LEX_STATE_BEG);
LEX(PM_TOKEN_STAR_EQUAL);
}
pm_token_type_t type = PM_TOKEN_STAR;
if (lex_state_spcarg_p(parser, space_seen)) {
pm_parser_warn_token(parser, &parser->current, PM_WARN_AMBIGUOUS_PREFIX_STAR);
type = PM_TOKEN_USTAR;
} else if (lex_state_beg_p(parser)) {
type = PM_TOKEN_USTAR;
}
if (lex_state_operator_p(parser)) {
lex_state_set(parser, PM_LEX_STATE_ARG);
} else {
lex_state_set(parser, PM_LEX_STATE_BEG);
}
LEX(type);
}
// ! != !~ !@
case '!':
if (lex_state_operator_p(parser)) {
lex_state_set(parser, PM_LEX_STATE_ARG);
if (match(parser, '@')) {
LEX(PM_TOKEN_BANG);
}
} else {
lex_state_set(parser, PM_LEX_STATE_BEG);
}
if (match(parser, '=')) {
LEX(PM_TOKEN_BANG_EQUAL);
}
if (match(parser, '~')) {
LEX(PM_TOKEN_BANG_TILDE);
}
LEX(PM_TOKEN_BANG);
// = => =~ == === =begin
case '=':
if (current_token_starts_line(parser) && (parser->current.end + 5 <= parser->end) && memcmp(parser->current.end, "begin", 5) == 0 && pm_char_is_whitespace(peek_offset(parser, 5))) {
pm_token_type_t type = lex_embdoc(parser);
if (type == PM_TOKEN_EOF) {
LEX(type);
}
goto lex_next_token;
}
if (lex_state_operator_p(parser)) {
lex_state_set(parser, PM_LEX_STATE_ARG);
} else {
lex_state_set(parser, PM_LEX_STATE_BEG);
}
if (match(parser, '>')) {
LEX(PM_TOKEN_EQUAL_GREATER);
}
if (match(parser, '~')) {
LEX(PM_TOKEN_EQUAL_TILDE);
}
if (match(parser, '=')) {
LEX(match(parser, '=') ? PM_TOKEN_EQUAL_EQUAL_EQUAL : PM_TOKEN_EQUAL_EQUAL);
}
LEX(PM_TOKEN_EQUAL);
// < << <<= <= <=>
case '<':
if (match(parser, '<')) {
if (
!lex_state_p(parser, PM_LEX_STATE_DOT | PM_LEX_STATE_CLASS) &&
!lex_state_end_p(parser) &&
(!lex_state_p(parser, PM_LEX_STATE_ARG_ANY) || lex_state_p(parser, PM_LEX_STATE_LABELED) || space_seen)
) {
const uint8_t *end = parser->current.end;
pm_heredoc_quote_t quote = PM_HEREDOC_QUOTE_NONE;
pm_heredoc_indent_t indent = PM_HEREDOC_INDENT_NONE;
if (match(parser, '-')) {
indent = PM_HEREDOC_INDENT_DASH;
}
else if (match(parser, '~')) {
indent = PM_HEREDOC_INDENT_TILDE;
}
if (match(parser, '`')) {
quote = PM_HEREDOC_QUOTE_BACKTICK;
}
else if (match(parser, '"')) {
quote = PM_HEREDOC_QUOTE_DOUBLE;
}
else if (match(parser, '\'')) {
quote = PM_HEREDOC_QUOTE_SINGLE;
}
const uint8_t *ident_start = parser->current.end;
size_t width = 0;
if (parser->current.end >= parser->end) {
parser->current.end = end;
} else if (quote == PM_HEREDOC_QUOTE_NONE && (width = char_is_identifier(parser, parser->current.end)) == 0) {
parser->current.end = end;
} else {
if (quote == PM_HEREDOC_QUOTE_NONE) {
parser->current.end += width;
while ((parser->current.end < parser->end) && (width = char_is_identifier(parser, parser->current.end))) {
parser->current.end += width;
}
} else {
// If we have quotes, then we're going to go until we find the
// end quote.
while ((parser->current.end < parser->end) && quote != (pm_heredoc_quote_t) (*parser->current.end)) {
parser->current.end++;
}
}
size_t ident_length = (size_t) (parser->current.end - ident_start);
if (quote != PM_HEREDOC_QUOTE_NONE && !match(parser, (uint8_t) quote)) {
// TODO: handle unterminated heredoc
}
parser->explicit_encoding = NULL;
lex_mode_push(parser, (pm_lex_mode_t) {
.mode = PM_LEX_HEREDOC,
.as.heredoc = {
.ident_start = ident_start,
.ident_length = ident_length,
.next_start = parser->current.end,
.quote = quote,
.indent = indent,
.common_whitespace = (size_t) -1
}
});
if (parser->heredoc_end == NULL) {
const uint8_t *body_start = next_newline(parser->current.end, parser->end - parser->current.end);
if (body_start == NULL) {
// If there is no newline after the heredoc identifier, then
// this is not a valid heredoc declaration. In this case we
// will add an error, but we will still return a heredoc
// start.
pm_parser_err_current(parser, PM_ERR_EMBDOC_TERM);
body_start = parser->end;
} else {
// Otherwise, we want to indicate that the body of the
// heredoc starts on the character after the next newline.
pm_newline_list_append(&parser->newline_list, body_start);
body_start++;
}
parser->next_start = body_start;
} else {
parser->next_start = parser->heredoc_end;
}
LEX(PM_TOKEN_HEREDOC_START);
}
}
if (match(parser, '=')) {
lex_state_set(parser, PM_LEX_STATE_BEG);
LEX(PM_TOKEN_LESS_LESS_EQUAL);
}
if (lex_state_operator_p(parser)) {
lex_state_set(parser, PM_LEX_STATE_ARG);
} else {
if (lex_state_p(parser, PM_LEX_STATE_CLASS)) parser->command_start = true;
lex_state_set(parser, PM_LEX_STATE_BEG);
}
LEX(PM_TOKEN_LESS_LESS);
}
if (lex_state_operator_p(parser)) {
lex_state_set(parser, PM_LEX_STATE_ARG);
} else {
if (lex_state_p(parser, PM_LEX_STATE_CLASS)) parser->command_start = true;
lex_state_set(parser, PM_LEX_STATE_BEG);
}
if (match(parser, '=')) {
if (match(parser, '>')) {
LEX(PM_TOKEN_LESS_EQUAL_GREATER);
}
LEX(PM_TOKEN_LESS_EQUAL);
}
LEX(PM_TOKEN_LESS);
// > >> >>= >=
case '>':
if (match(parser, '>')) {
if (lex_state_operator_p(parser)) {
lex_state_set(parser, PM_LEX_STATE_ARG);
} else {
lex_state_set(parser, PM_LEX_STATE_BEG);
}
LEX(match(parser, '=') ? PM_TOKEN_GREATER_GREATER_EQUAL : PM_TOKEN_GREATER_GREATER);
}
if (lex_state_operator_p(parser)) {
lex_state_set(parser, PM_LEX_STATE_ARG);
} else {
lex_state_set(parser, PM_LEX_STATE_BEG);
}
LEX(match(parser, '=') ? PM_TOKEN_GREATER_EQUAL : PM_TOKEN_GREATER);
// double-quoted string literal
case '"': {
bool label_allowed = (lex_state_p(parser, PM_LEX_STATE_LABEL | PM_LEX_STATE_ENDFN) && !previous_command_start) || lex_state_arg_p(parser);
lex_mode_push_string(parser, true, label_allowed, '\0', '"');
LEX(PM_TOKEN_STRING_BEGIN);
}
// xstring literal
case '`': {
if (lex_state_p(parser, PM_LEX_STATE_FNAME)) {
lex_state_set(parser, PM_LEX_STATE_ENDFN);
LEX(PM_TOKEN_BACKTICK);
}
if (lex_state_p(parser, PM_LEX_STATE_DOT)) {
if (previous_command_start) {
lex_state_set(parser, PM_LEX_STATE_CMDARG);
} else {
lex_state_set(parser, PM_LEX_STATE_ARG);
}
LEX(PM_TOKEN_BACKTICK);
}
lex_mode_push_string(parser, true, false, '\0', '`');
LEX(PM_TOKEN_BACKTICK);
}
// single-quoted string literal
case '\'': {
bool label_allowed = (lex_state_p(parser, PM_LEX_STATE_LABEL | PM_LEX_STATE_ENDFN) && !previous_command_start) || lex_state_arg_p(parser);
lex_mode_push_string(parser, false, label_allowed, '\0', '\'');
LEX(PM_TOKEN_STRING_BEGIN);
}
// ? character literal
case '?':
LEX(lex_question_mark(parser));
// & && &&= &=
case '&': {
if (match(parser, '&')) {
lex_state_set(parser, PM_LEX_STATE_BEG);
if (match(parser, '=')) {
LEX(PM_TOKEN_AMPERSAND_AMPERSAND_EQUAL);
}
LEX(PM_TOKEN_AMPERSAND_AMPERSAND);
}
if (match(parser, '=')) {
lex_state_set(parser, PM_LEX_STATE_BEG);
LEX(PM_TOKEN_AMPERSAND_EQUAL);
}
if (match(parser, '.')) {
lex_state_set(parser, PM_LEX_STATE_DOT);
LEX(PM_TOKEN_AMPERSAND_DOT);
}
pm_token_type_t type = PM_TOKEN_AMPERSAND;
if (lex_state_spcarg_p(parser, space_seen) || lex_state_beg_p(parser)) {
type = PM_TOKEN_UAMPERSAND;
}
if (lex_state_operator_p(parser)) {
lex_state_set(parser, PM_LEX_STATE_ARG);
} else {
lex_state_set(parser, PM_LEX_STATE_BEG);
}
LEX(type);
}
// | || ||= |=
case '|':
if (match(parser, '|')) {
if (match(parser, '=')) {
lex_state_set(parser, PM_LEX_STATE_BEG);
LEX(PM_TOKEN_PIPE_PIPE_EQUAL);
}
if (lex_state_p(parser, PM_LEX_STATE_BEG)) {
parser->current.end--;
LEX(PM_TOKEN_PIPE);
}
lex_state_set(parser, PM_LEX_STATE_BEG);
LEX(PM_TOKEN_PIPE_PIPE);
}
if (match(parser, '=')) {
lex_state_set(parser, PM_LEX_STATE_BEG);
LEX(PM_TOKEN_PIPE_EQUAL);
}
if (lex_state_operator_p(parser)) {
lex_state_set(parser, PM_LEX_STATE_ARG);
} else {
lex_state_set(parser, PM_LEX_STATE_BEG | PM_LEX_STATE_LABEL);
}
LEX(PM_TOKEN_PIPE);
// + += +@
case '+': {
if (lex_state_operator_p(parser)) {
lex_state_set(parser, PM_LEX_STATE_ARG);
if (match(parser, '@')) {
LEX(PM_TOKEN_UPLUS);
}
LEX(PM_TOKEN_PLUS);
}
if (match(parser, '=')) {
lex_state_set(parser, PM_LEX_STATE_BEG);
LEX(PM_TOKEN_PLUS_EQUAL);
}
bool spcarg = lex_state_spcarg_p(parser, space_seen);
if (spcarg) {
pm_parser_warn_token(parser, &parser->current, PM_WARN_AMBIGUOUS_FIRST_ARGUMENT_PLUS);
}
if (lex_state_beg_p(parser) || spcarg) {
lex_state_set(parser, PM_LEX_STATE_BEG);
if (pm_char_is_decimal_digit(peek(parser))) {
parser->current.end++;
pm_token_type_t type = lex_numeric(parser);
lex_state_set(parser, PM_LEX_STATE_END);
LEX(type);
}
LEX(PM_TOKEN_UPLUS);
}
lex_state_set(parser, PM_LEX_STATE_BEG);
LEX(PM_TOKEN_PLUS);
}
// - -= -@
case '-': {
if (lex_state_operator_p(parser)) {
lex_state_set(parser, PM_LEX_STATE_ARG);
if (match(parser, '@')) {
LEX(PM_TOKEN_UMINUS);
}
LEX(PM_TOKEN_MINUS);
}
if (match(parser, '=')) {
lex_state_set(parser, PM_LEX_STATE_BEG);
LEX(PM_TOKEN_MINUS_EQUAL);
}
if (match(parser, '>')) {
lex_state_set(parser, PM_LEX_STATE_ENDFN);
LEX(PM_TOKEN_MINUS_GREATER);
}
bool spcarg = lex_state_spcarg_p(parser, space_seen);
if (spcarg) {
pm_parser_warn_token(parser, &parser->current, PM_WARN_AMBIGUOUS_FIRST_ARGUMENT_MINUS);
}
if (lex_state_beg_p(parser) || spcarg) {
lex_state_set(parser, PM_LEX_STATE_BEG);
LEX(pm_char_is_decimal_digit(peek(parser)) ? PM_TOKEN_UMINUS_NUM : PM_TOKEN_UMINUS);
}
lex_state_set(parser, PM_LEX_STATE_BEG);
LEX(PM_TOKEN_MINUS);
}
// . .. ...
case '.': {
bool beg_p = lex_state_beg_p(parser);
if (match(parser, '.')) {
if (match(parser, '.')) {
// If we're _not_ inside a range within default parameters
if (
!context_p(parser, PM_CONTEXT_DEFAULT_PARAMS) &&
context_p(parser, PM_CONTEXT_DEF_PARAMS)
) {
if (lex_state_p(parser, PM_LEX_STATE_END)) {
lex_state_set(parser, PM_LEX_STATE_BEG);
} else {
lex_state_set(parser, PM_LEX_STATE_ENDARG);
}
LEX(PM_TOKEN_UDOT_DOT_DOT);
}
lex_state_set(parser, PM_LEX_STATE_BEG);
LEX(beg_p ? PM_TOKEN_UDOT_DOT_DOT : PM_TOKEN_DOT_DOT_DOT);
}
lex_state_set(parser, PM_LEX_STATE_BEG);
LEX(beg_p ? PM_TOKEN_UDOT_DOT : PM_TOKEN_DOT_DOT);
}
lex_state_set(parser, PM_LEX_STATE_DOT);
LEX(PM_TOKEN_DOT);
}
// integer
case '0':
case '1':
case '2':
case '3':
case '4':
case '5':
case '6':
case '7':
case '8':
case '9': {
pm_token_type_t type = lex_numeric(parser);
lex_state_set(parser, PM_LEX_STATE_END);
LEX(type);
}
// :: symbol
case ':':
if (match(parser, ':')) {
if (lex_state_beg_p(parser) || lex_state_p(parser, PM_LEX_STATE_CLASS) || (lex_state_p(parser, PM_LEX_STATE_ARG_ANY) && space_seen)) {
lex_state_set(parser, PM_LEX_STATE_BEG);
LEX(PM_TOKEN_UCOLON_COLON);
}
lex_state_set(parser, PM_LEX_STATE_DOT);
LEX(PM_TOKEN_COLON_COLON);
}
if (lex_state_end_p(parser) || pm_char_is_whitespace(peek(parser)) || peek(parser) == '#') {
lex_state_set(parser, PM_LEX_STATE_BEG);
LEX(PM_TOKEN_COLON);
}
if (peek(parser) == '"' || peek(parser) == '\'') {
lex_mode_push_string(parser, peek(parser) == '"', false, '\0', *parser->current.end);
parser->current.end++;
}
lex_state_set(parser, PM_LEX_STATE_FNAME);
LEX(PM_TOKEN_SYMBOL_BEGIN);
// / /=
case '/':
if (lex_state_beg_p(parser)) {
lex_mode_push_regexp(parser, '\0', '/');
LEX(PM_TOKEN_REGEXP_BEGIN);
}
if (match(parser, '=')) {
lex_state_set(parser, PM_LEX_STATE_BEG);
LEX(PM_TOKEN_SLASH_EQUAL);
}
if (lex_state_spcarg_p(parser, space_seen)) {
pm_parser_warn_token(parser, &parser->current, PM_WARN_AMBIGUOUS_SLASH);
lex_mode_push_regexp(parser, '\0', '/');
LEX(PM_TOKEN_REGEXP_BEGIN);
}
if (lex_state_operator_p(parser)) {
lex_state_set(parser, PM_LEX_STATE_ARG);
} else {
lex_state_set(parser, PM_LEX_STATE_BEG);
}
LEX(PM_TOKEN_SLASH);
// ^ ^=
case '^':
if (lex_state_operator_p(parser)) {
lex_state_set(parser, PM_LEX_STATE_ARG);
} else {
lex_state_set(parser, PM_LEX_STATE_BEG);
}
LEX(match(parser, '=') ? PM_TOKEN_CARET_EQUAL : PM_TOKEN_CARET);
// ~ ~@
case '~':
if (lex_state_operator_p(parser)) {
(void) match(parser, '@');
lex_state_set(parser, PM_LEX_STATE_ARG);
} else {
lex_state_set(parser, PM_LEX_STATE_BEG);
}
LEX(PM_TOKEN_TILDE);
// % %= %i %I %q %Q %w %W
case '%': {
// If there is no subsequent character then we have an
// invalid token. We're going to say it's the percent
// operator because we don't want to move into the string
// lex mode unnecessarily.
if ((lex_state_beg_p(parser) || lex_state_arg_p(parser)) && (parser->current.end >= parser->end)) {
pm_parser_err_current(parser, PM_ERR_INVALID_PERCENT);
LEX(PM_TOKEN_PERCENT);
}
if (!lex_state_beg_p(parser) && match(parser, '=')) {
lex_state_set(parser, PM_LEX_STATE_BEG);
LEX(PM_TOKEN_PERCENT_EQUAL);
} else if (
lex_state_beg_p(parser) ||
(lex_state_p(parser, PM_LEX_STATE_FITEM) && (peek(parser) == 's')) ||
lex_state_spcarg_p(parser, space_seen)
) {
if (!parser->encoding->alnum_char(parser->current.end, parser->end - parser->current.end)) {
if (*parser->current.end >= 0x80) {
pm_parser_err_current(parser, PM_ERR_INVALID_PERCENT);
}
lex_mode_push_string(parser, true, false, lex_mode_incrementor(*parser->current.end), lex_mode_terminator(*parser->current.end));
size_t eol_length = match_eol(parser);
if (eol_length) {
parser->current.end += eol_length;
pm_newline_list_append(&parser->newline_list, parser->current.end - 1);
} else {
parser->current.end++;
}
if (parser->current.end < parser->end) {
LEX(PM_TOKEN_STRING_BEGIN);
}
}
// Delimiters for %-literals cannot be alphanumeric. We
// validate that here.
uint8_t delimiter = peek_offset(parser, 1);
if (delimiter >= 0x80 || parser->encoding->alnum_char(&delimiter, 1)) {
pm_parser_err_current(parser, PM_ERR_INVALID_PERCENT);
goto lex_next_token;
}
switch (peek(parser)) {
case 'i': {
parser->current.end++;
if (parser->current.end < parser->end) {
lex_mode_push_list(parser, false, *parser->current.end++);
} else {
lex_mode_push_list_eof(parser);
}
LEX(PM_TOKEN_PERCENT_LOWER_I);
}
case 'I': {
parser->current.end++;
if (parser->current.end < parser->end) {
lex_mode_push_list(parser, true, *parser->current.end++);
} else {
lex_mode_push_list_eof(parser);
}
LEX(PM_TOKEN_PERCENT_UPPER_I);
}
case 'r': {
parser->current.end++;
if (parser->current.end < parser->end) {
lex_mode_push_regexp(parser, lex_mode_incrementor(*parser->current.end), lex_mode_terminator(*parser->current.end));
pm_newline_list_check_append(&parser->newline_list, parser->current.end);
parser->current.end++;
} else {
lex_mode_push_regexp(parser, '\0', '\0');
}
LEX(PM_TOKEN_REGEXP_BEGIN);
}
case 'q': {
parser->current.end++;
if (parser->current.end < parser->end) {
lex_mode_push_string(parser, false, false, lex_mode_incrementor(*parser->current.end), lex_mode_terminator(*parser->current.end));
pm_newline_list_check_append(&parser->newline_list, parser->current.end);
parser->current.end++;
} else {
lex_mode_push_string_eof(parser);
}
LEX(PM_TOKEN_STRING_BEGIN);
}
case 'Q': {
parser->current.end++;
if (parser->current.end < parser->end) {
lex_mode_push_string(parser, true, false, lex_mode_incrementor(*parser->current.end), lex_mode_terminator(*parser->current.end));
pm_newline_list_check_append(&parser->newline_list, parser->current.end);
parser->current.end++;
} else {
lex_mode_push_string_eof(parser);
}
LEX(PM_TOKEN_STRING_BEGIN);
}
case 's': {
parser->current.end++;
if (parser->current.end < parser->end) {
lex_mode_push_string(parser, false, false, lex_mode_incrementor(*parser->current.end), lex_mode_terminator(*parser->current.end));
lex_state_set(parser, PM_LEX_STATE_FNAME | PM_LEX_STATE_FITEM);
parser->current.end++;
} else {
lex_mode_push_string_eof(parser);
}
LEX(PM_TOKEN_SYMBOL_BEGIN);
}
case 'w': {
parser->current.end++;
if (parser->current.end < parser->end) {
lex_mode_push_list(parser, false, *parser->current.end++);
} else {
lex_mode_push_list_eof(parser);
}
LEX(PM_TOKEN_PERCENT_LOWER_W);
}
case 'W': {
parser->current.end++;
if (parser->current.end < parser->end) {
lex_mode_push_list(parser, true, *parser->current.end++);
} else {
lex_mode_push_list_eof(parser);
}
LEX(PM_TOKEN_PERCENT_UPPER_W);
}
case 'x': {
parser->current.end++;
if (parser->current.end < parser->end) {
lex_mode_push_string(parser, true, false, lex_mode_incrementor(*parser->current.end), lex_mode_terminator(*parser->current.end));
parser->current.end++;
} else {
lex_mode_push_string_eof(parser);
}
LEX(PM_TOKEN_PERCENT_LOWER_X);
}
default:
// If we get to this point, then we have a % that is completely
// unparseable. In this case we'll just drop it from the parser
// and skip past it and hope that the next token is something
// that we can parse.
pm_parser_err_current(parser, PM_ERR_INVALID_PERCENT);
goto lex_next_token;
}
}
lex_state_set(parser, lex_state_operator_p(parser) ? PM_LEX_STATE_ARG : PM_LEX_STATE_BEG);
LEX(PM_TOKEN_PERCENT);
}
// global variable
case '$': {
pm_token_type_t type = lex_global_variable(parser);
// If we're lexing an embedded variable, then we need to pop back into
// the parent lex context.
if (parser->lex_modes.current->mode == PM_LEX_EMBVAR) {
lex_mode_pop(parser);
}
lex_state_set(parser, PM_LEX_STATE_END);
LEX(type);
}
// instance variable, class variable
case '@':
lex_state_set(parser, parser->lex_state & PM_LEX_STATE_FNAME ? PM_LEX_STATE_ENDFN : PM_LEX_STATE_END);
LEX(lex_at_variable(parser));
default: {
if (*parser->current.start != '_') {
size_t width = char_is_identifier_start(parser, parser->current.start);
// If this isn't the beginning of an identifier, then it's an invalid
// token as we've exhausted all of the other options. We'll skip past
// it and return the next token.
if (!width) {
pm_parser_err_current(parser, PM_ERR_INVALID_TOKEN);
goto lex_next_token;
}
parser->current.end = parser->current.start + width;
}
pm_token_type_t type = lex_identifier(parser, previous_command_start);
// If we've hit a __END__ and it was at the start of the line or the
// start of the file and it is followed by either a \n or a \r\n, then
// this is the last token of the file.
if (
((parser->current.end - parser->current.start) == 7) &&
current_token_starts_line(parser) &&
(memcmp(parser->current.start, "__END__", 7) == 0) &&
(parser->current.end == parser->end || match_eol(parser))
)
{
// Since we know we're about to add an __END__ comment, we know we
// need at add all of the newlines to get the correct column
// information for it.
const uint8_t *cursor = parser->current.end;
while ((cursor = next_newline(cursor, parser->end - cursor)) != NULL) {
pm_newline_list_append(&parser->newline_list, cursor++);
}
parser->current.end = parser->end;
parser->current.type = PM_TOKEN___END__;
parser_lex_callback(parser);
parser->data_loc.start = parser->current.start;
parser->data_loc.end = parser->current.end;
LEX(PM_TOKEN_EOF);
}
pm_lex_state_t last_state = parser->lex_state;
if (type == PM_TOKEN_IDENTIFIER || type == PM_TOKEN_CONSTANT || type == PM_TOKEN_METHOD_NAME) {
if (lex_state_p(parser, PM_LEX_STATE_BEG_ANY | PM_LEX_STATE_ARG_ANY | PM_LEX_STATE_DOT)) {
if (previous_command_start) {
lex_state_set(parser, PM_LEX_STATE_CMDARG);
} else {
lex_state_set(parser, PM_LEX_STATE_ARG);
}
} else if (parser->lex_state == PM_LEX_STATE_FNAME) {
lex_state_set(parser, PM_LEX_STATE_ENDFN);
} else {
lex_state_set(parser, PM_LEX_STATE_END);
}
}
if (
!(last_state & (PM_LEX_STATE_DOT | PM_LEX_STATE_FNAME)) &&
(type == PM_TOKEN_IDENTIFIER) &&
((pm_parser_local_depth(parser, &parser->current) != -1) ||
pm_token_is_numbered_parameter(parser->current.start, parser->current.end))
) {
lex_state_set(parser, PM_LEX_STATE_END | PM_LEX_STATE_LABEL);
}
LEX(type);
}
}
}
case PM_LEX_LIST: {
if (parser->next_start != NULL) {
parser->current.end = parser->next_start;
parser->next_start = NULL;
}
// First we'll set the beginning of the token.
parser->current.start = parser->current.end;
// If there's any whitespace at the start of the list, then we're
// going to trim it off the beginning and create a new token.
size_t whitespace;
if (parser->heredoc_end) {
whitespace = pm_strspn_inline_whitespace(parser->current.end, parser->end - parser->current.end);
if (peek_offset(parser, (ptrdiff_t)whitespace) == '\n') {
whitespace += 1;
}
} else {
whitespace = pm_strspn_whitespace_newlines(parser->current.end, parser->end - parser->current.end, &parser->newline_list);
}
if (whitespace > 0) {
parser->current.end += whitespace;
if (peek_offset(parser, -1) == '\n') {
// mutates next_start
parser_flush_heredoc_end(parser);
}
LEX(PM_TOKEN_WORDS_SEP);
}
// We'll check if we're at the end of the file. If we are, then we
// need to return the EOF token.
if (parser->current.end >= parser->end) {
LEX(PM_TOKEN_EOF);
}
// Here we'll get a list of the places where strpbrk should break,
// and then find the first one.
pm_lex_mode_t *lex_mode = parser->lex_modes.current;
const uint8_t *breakpoints = lex_mode->as.list.breakpoints;
const uint8_t *breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
// If we haven't found an escape yet, then this buffer will be
// unallocated since we can refer directly to the source string.
pm_token_buffer_t token_buffer = { { 0 }, 0 };
while (breakpoint != NULL) {
// If we hit a null byte, skip directly past it.
if (*breakpoint == '\0') {
breakpoint = pm_strpbrk(parser, breakpoint + 1, breakpoints, parser->end - (breakpoint + 1));
continue;
}
// If we hit whitespace, then we must have received content by
// now, so we can return an element of the list.
if (pm_char_is_whitespace(*breakpoint)) {
parser->current.end = breakpoint;
pm_token_buffer_flush(parser, &token_buffer);
LEX(PM_TOKEN_STRING_CONTENT);
}
// If we hit the terminator, we need to check which token to
// return.
if (*breakpoint == lex_mode->as.list.terminator) {
// If this terminator doesn't actually close the list, then
// we need to continue on past it.
if (lex_mode->as.list.nesting > 0) {
parser->current.end = breakpoint + 1;
breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
lex_mode->as.list.nesting--;
continue;
}
// If we've hit the terminator and we've already skipped
// past content, then we can return a list node.
if (breakpoint > parser->current.start) {
parser->current.end = breakpoint;
pm_token_buffer_flush(parser, &token_buffer);
LEX(PM_TOKEN_STRING_CONTENT);
}
// Otherwise, switch back to the default state and return
// the end of the list.
parser->current.end = breakpoint + 1;
lex_mode_pop(parser);
lex_state_set(parser, PM_LEX_STATE_END);
LEX(PM_TOKEN_STRING_END);
}
// If we hit escapes, then we need to treat the next token
// literally. In this case we'll skip past the next character
// and find the next breakpoint.
if (*breakpoint == '\\') {
parser->current.end = breakpoint + 1;
// If we've hit the end of the file, then break out of the
// loop by setting the breakpoint to NULL.
if (parser->current.end == parser->end) {
breakpoint = NULL;
continue;
}
pm_token_buffer_escape(parser, &token_buffer);
uint8_t peeked = peek(parser);
switch (peeked) {
case ' ':
case '\f':
case '\t':
case '\v':
case '\\':
pm_token_buffer_push(&token_buffer, peeked);
parser->current.end++;
break;
case '\r':
parser->current.end++;
if (peek(parser) != '\n') {
pm_token_buffer_push(&token_buffer, '\r');
break;
}
/* fallthrough */
case '\n':
pm_token_buffer_push(&token_buffer, '\n');
if (parser->heredoc_end) {
// ... if we are on the same line as a heredoc,
// flush the heredoc and continue parsing after
// heredoc_end.
parser_flush_heredoc_end(parser);
pm_token_buffer_copy(parser, &token_buffer);
LEX(PM_TOKEN_STRING_CONTENT);
} else {
// ... else track the newline.
pm_newline_list_append(&parser->newline_list, parser->current.end);
}
parser->current.end++;
break;
default:
if (peeked == lex_mode->as.list.incrementor || peeked == lex_mode->as.list.terminator) {
pm_token_buffer_push(&token_buffer, peeked);
parser->current.end++;
} else if (lex_mode->as.list.interpolation) {
escape_read(parser, &token_buffer.buffer, PM_ESCAPE_FLAG_NONE);
} else {
pm_token_buffer_push(&token_buffer, '\\');
pm_token_buffer_push(&token_buffer, peeked);
parser->current.end++;
}
break;
}
token_buffer.cursor = parser->current.end;
breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
continue;
}
// If we hit a #, then we will attempt to lex interpolation.
if (*breakpoint == '#') {
pm_token_type_t type = lex_interpolation(parser, breakpoint);
if (type == PM_TOKEN_NOT_PROVIDED) {
// If we haven't returned at this point then we had something
// that looked like an interpolated class or instance variable
// like "#@" but wasn't actually. In this case we'll just skip
// to the next breakpoint.
breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
continue;
}
if (type == PM_TOKEN_STRING_CONTENT) {
pm_token_buffer_flush(parser, &token_buffer);
}
LEX(type);
}
// If we've hit the incrementor, then we need to skip past it
// and find the next breakpoint.
assert(*breakpoint == lex_mode->as.list.incrementor);
parser->current.end = breakpoint + 1;
breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
lex_mode->as.list.nesting++;
continue;
}
if (parser->current.end > parser->current.start) {
pm_token_buffer_flush(parser, &token_buffer);
LEX(PM_TOKEN_STRING_CONTENT);
}
// If we were unable to find a breakpoint, then this token hits the
// end of the file.
parser->current.end = parser->end;
pm_token_buffer_flush(parser, &token_buffer);
LEX(PM_TOKEN_STRING_CONTENT);
}
case PM_LEX_REGEXP: {
// First, we'll set to start of this token to be the current end.
if (parser->next_start == NULL) {
parser->current.start = parser->current.end;
} else {
parser->current.start = parser->next_start;
parser->current.end = parser->next_start;
parser->next_start = NULL;
}
// We'll check if we're at the end of the file. If we are, then we need to
// return the EOF token.
if (parser->current.end >= parser->end) {
LEX(PM_TOKEN_EOF);
}
// Get a reference to the current mode.
pm_lex_mode_t *lex_mode = parser->lex_modes.current;
// These are the places where we need to split up the content of the
// regular expression. We'll use strpbrk to find the first of these
// characters.
const uint8_t *breakpoints = lex_mode->as.regexp.breakpoints;
const uint8_t *breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
pm_token_buffer_t token_buffer = { { 0 }, 0 };
while (breakpoint != NULL) {
// If we hit a null byte, skip directly past it.
if (*breakpoint == '\0') {
parser->current.end = breakpoint + 1;
breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
continue;
}
// If we've hit a newline, then we need to track that in the
// list of newlines.
if (*breakpoint == '\n') {
// For the special case of a newline-terminated regular expression, we will pass
// through this branch twice -- once with PM_TOKEN_REGEXP_BEGIN and then again
// with PM_TOKEN_STRING_CONTENT. Let's avoid tracking the newline twice, by
// tracking it only in the REGEXP_BEGIN case.
if (
!(lex_mode->as.regexp.terminator == '\n' && parser->current.type != PM_TOKEN_REGEXP_BEGIN)
&& parser->heredoc_end == NULL
) {
pm_newline_list_append(&parser->newline_list, breakpoint);
}
if (lex_mode->as.regexp.terminator != '\n') {
// If the terminator is not a newline, then we can set
// the next breakpoint and continue.
parser->current.end = breakpoint + 1;
breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
continue;
}
}
// If we hit the terminator, we need to determine what kind of
// token to return.
if (*breakpoint == lex_mode->as.regexp.terminator) {
if (lex_mode->as.regexp.nesting > 0) {
parser->current.end = breakpoint + 1;
breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
lex_mode->as.regexp.nesting--;
continue;
}
// Here we've hit the terminator. If we have already consumed
// content then we need to return that content as string content
// first.
if (breakpoint > parser->current.start) {
parser->current.end = breakpoint;
pm_token_buffer_flush(parser, &token_buffer);
LEX(PM_TOKEN_STRING_CONTENT);
}
// Since we've hit the terminator of the regular expression,
// we now need to parse the options.
parser->current.end = breakpoint + 1;
parser->current.end += pm_strspn_regexp_option(parser->current.end, parser->end - parser->current.end);
lex_mode_pop(parser);
lex_state_set(parser, PM_LEX_STATE_END);
LEX(PM_TOKEN_REGEXP_END);
}
// If we hit escapes, then we need to treat the next token
// literally. In this case we'll skip past the next character
// and find the next breakpoint.
if (*breakpoint == '\\') {
parser->current.end = breakpoint + 1;
// If we've hit the end of the file, then break out of the
// loop by setting the breakpoint to NULL.
if (parser->current.end == parser->end) {
breakpoint = NULL;
continue;
}
pm_token_buffer_escape(parser, &token_buffer);
uint8_t peeked = peek(parser);
switch (peeked) {
case '\r':
parser->current.end++;
if (peek(parser) != '\n') {
if (lex_mode->as.regexp.terminator != '\r') {
pm_token_buffer_push(&token_buffer, '\\');
}
pm_token_buffer_push(&token_buffer, '\r');
break;
}
/* fallthrough */
case '\n':
if (parser->heredoc_end) {
// ... if we are on the same line as a heredoc,
// flush the heredoc and continue parsing after
// heredoc_end.
parser_flush_heredoc_end(parser);
pm_token_buffer_copy(parser, &token_buffer);
LEX(PM_TOKEN_STRING_CONTENT);
} else {
// ... else track the newline.
pm_newline_list_append(&parser->newline_list, parser->current.end);
}
parser->current.end++;
break;
case 'c':
case 'C':
case 'M':
case 'u':
case 'x':
escape_read(parser, &token_buffer.buffer, PM_ESCAPE_FLAG_REGEXP);
break;
default:
if (lex_mode->as.regexp.terminator == peeked) {
// Some characters when they are used as the
// terminator also receive an escape. They are
// enumerated here.
switch (peeked) {
case '$': case ')': case '*': case '+':
case '.': case '>': case '?': case ']':
case '^': case '|': case '}':
pm_token_buffer_push(&token_buffer, '\\');
break;
default:
break;
}
pm_token_buffer_push(&token_buffer, peeked);
parser->current.end++;
break;
}
if (peeked < 0x80) pm_token_buffer_push(&token_buffer, '\\');
pm_token_buffer_push(&token_buffer, peeked);
parser->current.end++;
break;
}
token_buffer.cursor = parser->current.end;
breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
continue;
}
// If we hit a #, then we will attempt to lex interpolation.
if (*breakpoint == '#') {
pm_token_type_t type = lex_interpolation(parser, breakpoint);
if (type == PM_TOKEN_NOT_PROVIDED) {
// If we haven't returned at this point then we had
// something that looked like an interpolated class or
// instance variable like "#@" but wasn't actually. In
// this case we'll just skip to the next breakpoint.
breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
continue;
}
if (type == PM_TOKEN_STRING_CONTENT) {
pm_token_buffer_flush(parser, &token_buffer);
}
LEX(type);
}
// If we've hit the incrementor, then we need to skip past it
// and find the next breakpoint.
assert(*breakpoint == lex_mode->as.regexp.incrementor);
parser->current.end = breakpoint + 1;
breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
lex_mode->as.regexp.nesting++;
continue;
}
if (parser->current.end > parser->current.start) {
pm_token_buffer_flush(parser, &token_buffer);
LEX(PM_TOKEN_STRING_CONTENT);
}
// If we were unable to find a breakpoint, then this token hits the
// end of the file.
parser->current.end = parser->end;
pm_token_buffer_flush(parser, &token_buffer);
LEX(PM_TOKEN_STRING_CONTENT);
}
case PM_LEX_STRING: {
// First, we'll set to start of this token to be the current end.
if (parser->next_start == NULL) {
parser->current.start = parser->current.end;
} else {
parser->current.start = parser->next_start;
parser->current.end = parser->next_start;
parser->next_start = NULL;
}
// We'll check if we're at the end of the file. If we are, then we need to
// return the EOF token.
if (parser->current.end >= parser->end) {
LEX(PM_TOKEN_EOF);
}
// These are the places where we need to split up the content of the
// string. We'll use strpbrk to find the first of these characters.
pm_lex_mode_t *lex_mode = parser->lex_modes.current;
const uint8_t *breakpoints = lex_mode->as.string.breakpoints;
const uint8_t *breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
// If we haven't found an escape yet, then this buffer will be
// unallocated since we can refer directly to the source string.
pm_token_buffer_t token_buffer = { { 0 }, 0 };
while (breakpoint != NULL) {
// If we hit the incrementor, then we'll increment then nesting and
// continue lexing.
if (lex_mode->as.string.incrementor != '\0' && *breakpoint == lex_mode->as.string.incrementor) {
lex_mode->as.string.nesting++;
parser->current.end = breakpoint + 1;
breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
continue;
}
// Note that we have to check the terminator here first because we could
// potentially be parsing a % string that has a # character as the
// terminator.
if (*breakpoint == lex_mode->as.string.terminator) {
// If this terminator doesn't actually close the string, then we need
// to continue on past it.
if (lex_mode->as.string.nesting > 0) {
parser->current.end = breakpoint + 1;
breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
lex_mode->as.string.nesting--;
continue;
}
// Here we've hit the terminator. If we have already consumed content
// then we need to return that content as string content first.
if (breakpoint > parser->current.start) {
parser->current.end = breakpoint;
pm_token_buffer_flush(parser, &token_buffer);
LEX(PM_TOKEN_STRING_CONTENT);
}
// Otherwise we need to switch back to the parent lex mode and
// return the end of the string.
size_t eol_length = match_eol_at(parser, breakpoint);
if (eol_length) {
parser->current.end = breakpoint + eol_length;
pm_newline_list_append(&parser->newline_list, parser->current.end - 1);
} else {
parser->current.end = breakpoint + 1;
}
if (lex_mode->as.string.label_allowed && (peek(parser) == ':') && (peek_offset(parser, 1) != ':')) {
parser->current.end++;
lex_state_set(parser, PM_LEX_STATE_ARG | PM_LEX_STATE_LABELED);
lex_mode_pop(parser);
LEX(PM_TOKEN_LABEL_END);
}
lex_state_set(parser, PM_LEX_STATE_END);
lex_mode_pop(parser);
LEX(PM_TOKEN_STRING_END);
}
// When we hit a newline, we need to flush any potential heredocs. Note
// that this has to happen after we check for the terminator in case the
// terminator is a newline character.
if (*breakpoint == '\n') {
if (parser->heredoc_end == NULL) {
pm_newline_list_append(&parser->newline_list, breakpoint);
parser->current.end = breakpoint + 1;
breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
continue;
} else {
parser->current.end = breakpoint + 1;
parser_flush_heredoc_end(parser);
pm_token_buffer_flush(parser, &token_buffer);
LEX(PM_TOKEN_STRING_CONTENT);
}
}
switch (*breakpoint) {
case '\0':
// Skip directly past the null character.
parser->current.end = breakpoint + 1;
breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
break;
case '\\': {
// Here we hit escapes.
parser->current.end = breakpoint + 1;
// If we've hit the end of the file, then break out of
// the loop by setting the breakpoint to NULL.
if (parser->current.end == parser->end) {
breakpoint = NULL;
continue;
}
pm_token_buffer_escape(parser, &token_buffer);
uint8_t peeked = peek(parser);
switch (peeked) {
case '\\':
pm_token_buffer_push(&token_buffer, '\\');
parser->current.end++;
break;
case '\r':
parser->current.end++;
if (peek(parser) != '\n') {
if (!lex_mode->as.string.interpolation) {
pm_token_buffer_push(&token_buffer, '\\');
}
pm_token_buffer_push(&token_buffer, '\r');
break;
}
/* fallthrough */
case '\n':
if (!lex_mode->as.string.interpolation) {
pm_token_buffer_push(&token_buffer, '\\');
pm_token_buffer_push(&token_buffer, '\n');
}
if (parser->heredoc_end) {
// ... if we are on the same line as a heredoc,
// flush the heredoc and continue parsing after
// heredoc_end.
parser_flush_heredoc_end(parser);
pm_token_buffer_copy(parser, &token_buffer);
LEX(PM_TOKEN_STRING_CONTENT);
} else {
// ... else track the newline.
pm_newline_list_append(&parser->newline_list, parser->current.end);
}
parser->current.end++;
break;
default:
if (lex_mode->as.string.incrementor != '\0' && peeked == lex_mode->as.string.incrementor) {
pm_token_buffer_push(&token_buffer, peeked);
parser->current.end++;
} else if (lex_mode->as.string.terminator != '\0' && peeked == lex_mode->as.string.terminator) {
pm_token_buffer_push(&token_buffer, peeked);
parser->current.end++;
} else if (lex_mode->as.string.interpolation) {
escape_read(parser, &token_buffer.buffer, PM_ESCAPE_FLAG_NONE);
} else {
pm_token_buffer_push(&token_buffer, '\\');
pm_token_buffer_push(&token_buffer, peeked);
parser->current.end++;
}
break;
}
token_buffer.cursor = parser->current.end;
breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
break;
}
case '#': {
pm_token_type_t type = lex_interpolation(parser, breakpoint);
if (type == PM_TOKEN_NOT_PROVIDED) {
// If we haven't returned at this point then we had something that
// looked like an interpolated class or instance variable like "#@"
// but wasn't actually. In this case we'll just skip to the next
// breakpoint.
breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
break;
}
if (type == PM_TOKEN_STRING_CONTENT) {
pm_token_buffer_flush(parser, &token_buffer);
}
LEX(type);
}
default:
assert(false && "unreachable");
}
}
if (parser->current.end > parser->current.start) {
pm_token_buffer_flush(parser, &token_buffer);
LEX(PM_TOKEN_STRING_CONTENT);
}
// If we've hit the end of the string, then this is an unterminated
// string. In that case we'll return a string content token.
parser->current.end = parser->end;
pm_token_buffer_flush(parser, &token_buffer);
LEX(PM_TOKEN_STRING_CONTENT);
}
case PM_LEX_HEREDOC: {
// First, we'll set to start of this token.
if (parser->next_start == NULL) {
parser->current.start = parser->current.end;
} else {
parser->current.start = parser->next_start;
parser->current.end = parser->next_start;
parser->heredoc_end = NULL;
parser->next_start = NULL;
}
// We'll check if we're at the end of the file. If we are, then we need to
// return the EOF token.
if (parser->current.end >= parser->end) {
LEX(PM_TOKEN_EOF);
}
// Now let's grab the information about the identifier off of the current
// lex mode.
pm_lex_mode_t *lex_mode = parser->lex_modes.current;
const uint8_t *ident_start = lex_mode->as.heredoc.ident_start;
size_t ident_length = lex_mode->as.heredoc.ident_length;
// If we are immediately following a newline and we have hit the
// terminator, then we need to return the ending of the heredoc.
if (current_token_starts_line(parser)) {
const uint8_t *start = parser->current.start;
if (start + ident_length <= parser->end) {
const uint8_t *newline = next_newline(start, parser->end - start);
const uint8_t *ident_end = newline;
const uint8_t *terminator_end = newline;
if (newline == NULL) {
terminator_end = parser->end;
ident_end = parser->end;
} else {
terminator_end++;
if (newline[-1] == '\r') {
ident_end--; // Remove \r
}
}
const uint8_t *terminator_start = ident_end - ident_length;
const uint8_t *cursor = start;
if (
lex_mode->as.heredoc.indent == PM_HEREDOC_INDENT_DASH ||
lex_mode->as.heredoc.indent == PM_HEREDOC_INDENT_TILDE
) {
while (cursor < terminator_start && pm_char_is_inline_whitespace(*cursor)) {
cursor++;
}
}
if (
(cursor == terminator_start) &&
(memcmp(terminator_start, ident_start, ident_length) == 0)
) {
if (newline != NULL) {
pm_newline_list_append(&parser->newline_list, newline);
}
parser->current.end = terminator_end;
if (*lex_mode->as.heredoc.next_start == '\\') {
parser->next_start = NULL;
} else {
parser->next_start = lex_mode->as.heredoc.next_start;
parser->heredoc_end = parser->current.end;
}
lex_state_set(parser, PM_LEX_STATE_END);
LEX(PM_TOKEN_HEREDOC_END);
}
}
size_t whitespace = pm_heredoc_strspn_inline_whitespace(parser, &start, lex_mode->as.heredoc.indent);
if (
lex_mode->as.heredoc.indent == PM_HEREDOC_INDENT_TILDE &&
(lex_mode->as.heredoc.common_whitespace > whitespace) &&
peek_at(parser, start) != '\n'
) {
lex_mode->as.heredoc.common_whitespace = whitespace;
}
}
// Otherwise we'll be parsing string content. These are the places
// where we need to split up the content of the heredoc. We'll use
// strpbrk to find the first of these characters.
uint8_t breakpoints[] = "\n\\#";
pm_heredoc_quote_t quote = lex_mode->as.heredoc.quote;
if (quote == PM_HEREDOC_QUOTE_SINGLE) {
breakpoints[2] = '\0';
}
const uint8_t *breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
pm_token_buffer_t token_buffer = { { 0 }, 0 };
bool was_escaped_newline = false;
while (breakpoint != NULL) {
switch (*breakpoint) {
case '\0':
// Skip directly past the null character.
parser->current.end = breakpoint + 1;
breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
break;
case '\n': {
if (parser->heredoc_end != NULL && (parser->heredoc_end > breakpoint)) {
parser_flush_heredoc_end(parser);
parser->current.end = breakpoint + 1;
pm_token_buffer_flush(parser, &token_buffer);
LEX(PM_TOKEN_STRING_CONTENT);
}
pm_newline_list_append(&parser->newline_list, breakpoint);
// If we have a - or ~ heredoc, then we can match after
// some leading whitespace.
const uint8_t *start = breakpoint + 1;
if (!was_escaped_newline && (start + ident_length <= parser->end)) {
// We want to match the terminator starting from the end of the line in case
// there is whitespace in the ident such as <<-' DOC' or <<~' DOC'.
const uint8_t *newline = next_newline(start, parser->end - start);
if (newline == NULL) {
newline = parser->end;
} else if (newline[-1] == '\r') {
newline--; // Remove \r
}
// Start of a possible terminator.
const uint8_t *terminator_start = newline - ident_length;
// Cursor to check for the leading whitespace. We skip the
// leading whitespace if we have a - or ~ heredoc.
const uint8_t *cursor = start;
if (lex_mode->as.heredoc.indent == PM_HEREDOC_INDENT_DASH ||
lex_mode->as.heredoc.indent == PM_HEREDOC_INDENT_TILDE) {
while (cursor < terminator_start && pm_char_is_inline_whitespace(*cursor)) {
cursor++;
}
}
if (
cursor == terminator_start &&
(memcmp(terminator_start, ident_start, ident_length) == 0)
) {
parser->current.end = breakpoint + 1;
pm_token_buffer_flush(parser, &token_buffer);
LEX(PM_TOKEN_STRING_CONTENT);
}
}
size_t whitespace = pm_heredoc_strspn_inline_whitespace(parser, &start, lex_mode->as.heredoc.indent);
// If we have hit a newline that is followed by a valid
// terminator, then we need to return the content of the
// heredoc here as string content. Then, the next time a
// token is lexed, it will match again and return the
// end of the heredoc.
if (lex_mode->as.heredoc.indent == PM_HEREDOC_INDENT_TILDE) {
if ((lex_mode->as.heredoc.common_whitespace > whitespace) && peek_at(parser, start) != '\n') {
lex_mode->as.heredoc.common_whitespace = whitespace;
}
parser->current.end = breakpoint + 1;
if (!was_escaped_newline) {
pm_token_buffer_flush(parser, &token_buffer);
LEX(PM_TOKEN_STRING_CONTENT);
}
}
// Otherwise we hit a newline and it wasn't followed by
// a terminator, so we can continue parsing.
parser->current.end = breakpoint + 1;
breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
break;
}
case '\\': {
// If we hit an escape, then we need to skip past
// however many characters the escape takes up. However
// it's important that if \n or \r\n are escaped that we
// stop looping before the newline and not after the
// newline so that we can still potentially find the
// terminator of the heredoc.
parser->current.end = breakpoint + 1;
// If we've hit the end of the file, then break out of
// the loop by setting the breakpoint to NULL.
if (parser->current.end == parser->end) {
breakpoint = NULL;
continue;
}
pm_token_buffer_escape(parser, &token_buffer);
uint8_t peeked = peek(parser);
if (quote == PM_HEREDOC_QUOTE_SINGLE) {
switch (peeked) {
case '\r':
parser->current.end++;
if (peek(parser) != '\n') {
pm_token_buffer_push(&token_buffer, '\\');
pm_token_buffer_push(&token_buffer, '\r');
break;
}
/* fallthrough */
case '\n':
pm_token_buffer_push(&token_buffer, '\\');
pm_token_buffer_push(&token_buffer, '\n');
token_buffer.cursor = parser->current.end + 1;
breakpoint = parser->current.end;
continue;
default:
parser->current.end++;
pm_token_buffer_push(&token_buffer, '\\');
pm_token_buffer_push(&token_buffer, peeked);
break;
}
} else {
switch (peeked) {
case '\r':
parser->current.end++;
if (peek(parser) != '\n') {
pm_token_buffer_push(&token_buffer, '\r');
break;
}
/* fallthrough */
case '\n':
was_escaped_newline = true;
token_buffer.cursor = parser->current.end + 1;
breakpoint = parser->current.end;
continue;
default:
escape_read(parser, &token_buffer.buffer, PM_ESCAPE_FLAG_NONE);
break;
}
}
token_buffer.cursor = parser->current.end;
breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
break;
}
case '#': {
pm_token_type_t type = lex_interpolation(parser, breakpoint);
if (type == PM_TOKEN_NOT_PROVIDED) {
// If we haven't returned at this point then we had
// something that looked like an interpolated class
// or instance variable like "#@" but wasn't
// actually. In this case we'll just skip to the
// next breakpoint.
breakpoint = pm_strpbrk(parser, parser->current.end, breakpoints, parser->end - parser->current.end);
break;
}
if (type == PM_TOKEN_STRING_CONTENT) {
pm_token_buffer_flush(parser, &token_buffer);
}
LEX(type);
}
default:
assert(false && "unreachable");
}
was_escaped_newline = false;
}
if (parser->current.end > parser->current.start) {
parser->current.end = parser->end;
pm_token_buffer_flush(parser, &token_buffer);
LEX(PM_TOKEN_STRING_CONTENT);
}
// If we've hit the end of the string, then this is an unterminated
// heredoc. In that case we'll return a string content token.
parser->current.end = parser->end;
pm_token_buffer_flush(parser, &token_buffer);
LEX(PM_TOKEN_STRING_CONTENT);
}
}
assert(false && "unreachable");
}
#undef LEX
/******************************************************************************/
/* Parse functions */
/******************************************************************************/
/**
* These are the various precedence rules. Because we are using a Pratt parser,
* they are named binding power to represent the manner in which nodes are bound
* together in the stack.
*
* We increment by 2 because we want to leave room for the infix operators to
* specify their associativity by adding or subtracting one.
*/
typedef enum {
PM_BINDING_POWER_UNSET = 0, // used to indicate this token cannot be used as an infix operator
PM_BINDING_POWER_STATEMENT = 2,
PM_BINDING_POWER_MODIFIER = 4, // if unless until while
PM_BINDING_POWER_MODIFIER_RESCUE = 6, // rescue
PM_BINDING_POWER_COMPOSITION = 8, // and or
PM_BINDING_POWER_NOT = 10, // not
PM_BINDING_POWER_MATCH = 12, // => in
PM_BINDING_POWER_DEFINED = 14, // defined?
PM_BINDING_POWER_MULTI_ASSIGNMENT = 16, // =
PM_BINDING_POWER_ASSIGNMENT = 18, // = += -= *= /= %= &= |= ^= &&= ||= <<= >>= **=
PM_BINDING_POWER_TERNARY = 20, // ?:
PM_BINDING_POWER_RANGE = 22, // .. ...
PM_BINDING_POWER_LOGICAL_OR = 24, // ||
PM_BINDING_POWER_LOGICAL_AND = 26, // &&
PM_BINDING_POWER_EQUALITY = 28, // <=> == === != =~ !~
PM_BINDING_POWER_COMPARISON = 30, // > >= < <=
PM_BINDING_POWER_BITWISE_OR = 32, // | ^
PM_BINDING_POWER_BITWISE_AND = 34, // &
PM_BINDING_POWER_SHIFT = 36, // << >>
PM_BINDING_POWER_TERM = 38, // + -
PM_BINDING_POWER_FACTOR = 40, // * / %
PM_BINDING_POWER_UMINUS = 42, // -@
PM_BINDING_POWER_EXPONENT = 44, // **
PM_BINDING_POWER_UNARY = 46, // ! ~ +@
PM_BINDING_POWER_INDEX = 48, // [] []=
PM_BINDING_POWER_CALL = 50, // :: .
PM_BINDING_POWER_MAX = 52
} pm_binding_power_t;
/**
* This struct represents a set of binding powers used for a given token. They
* are combined in this way to make it easier to represent associativity.
*/
typedef struct {
/** The left binding power. */
pm_binding_power_t left;
/** The right binding power. */
pm_binding_power_t right;
/** Whether or not this token can be used as a binary operator. */
bool binary;
/**
* Whether or not this token can be used as non-associative binary operator.
* Non-associative operators (e.g. in and =>) need special treatment in parse_expression.
*/
bool nonassoc;
} pm_binding_powers_t;
#define BINDING_POWER_ASSIGNMENT { PM_BINDING_POWER_UNARY, PM_BINDING_POWER_ASSIGNMENT, true, false }
#define LEFT_ASSOCIATIVE(precedence) { precedence, precedence + 1, true, false }
#define RIGHT_ASSOCIATIVE(precedence) { precedence, precedence, true, false }
#define NON_ASSOCIATIVE(precedence) { precedence, precedence + 1, true, true }
#define RIGHT_ASSOCIATIVE_UNARY(precedence) { precedence, precedence, false, false }
pm_binding_powers_t pm_binding_powers[PM_TOKEN_MAXIMUM] = {
// if unless until while
[PM_TOKEN_KEYWORD_IF_MODIFIER] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_MODIFIER),
[PM_TOKEN_KEYWORD_UNLESS_MODIFIER] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_MODIFIER),
[PM_TOKEN_KEYWORD_UNTIL_MODIFIER] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_MODIFIER),
[PM_TOKEN_KEYWORD_WHILE_MODIFIER] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_MODIFIER),
// rescue
[PM_TOKEN_KEYWORD_RESCUE_MODIFIER] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_MODIFIER_RESCUE),
// and or
[PM_TOKEN_KEYWORD_AND] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_COMPOSITION),
[PM_TOKEN_KEYWORD_OR] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_COMPOSITION),
// => in
[PM_TOKEN_EQUAL_GREATER] = NON_ASSOCIATIVE(PM_BINDING_POWER_MATCH),
[PM_TOKEN_KEYWORD_IN] = NON_ASSOCIATIVE(PM_BINDING_POWER_MATCH),
// &&= &= ^= = >>= <<= -= %= |= += /= *= **=
[PM_TOKEN_AMPERSAND_AMPERSAND_EQUAL] = BINDING_POWER_ASSIGNMENT,
[PM_TOKEN_AMPERSAND_EQUAL] = BINDING_POWER_ASSIGNMENT,
[PM_TOKEN_CARET_EQUAL] = BINDING_POWER_ASSIGNMENT,
[PM_TOKEN_EQUAL] = BINDING_POWER_ASSIGNMENT,
[PM_TOKEN_GREATER_GREATER_EQUAL] = BINDING_POWER_ASSIGNMENT,
[PM_TOKEN_LESS_LESS_EQUAL] = BINDING_POWER_ASSIGNMENT,
[PM_TOKEN_MINUS_EQUAL] = BINDING_POWER_ASSIGNMENT,
[PM_TOKEN_PERCENT_EQUAL] = BINDING_POWER_ASSIGNMENT,
[PM_TOKEN_PIPE_EQUAL] = BINDING_POWER_ASSIGNMENT,
[PM_TOKEN_PIPE_PIPE_EQUAL] = BINDING_POWER_ASSIGNMENT,
[PM_TOKEN_PLUS_EQUAL] = BINDING_POWER_ASSIGNMENT,
[PM_TOKEN_SLASH_EQUAL] = BINDING_POWER_ASSIGNMENT,
[PM_TOKEN_STAR_EQUAL] = BINDING_POWER_ASSIGNMENT,
[PM_TOKEN_STAR_STAR_EQUAL] = BINDING_POWER_ASSIGNMENT,
// ?:
[PM_TOKEN_QUESTION_MARK] = RIGHT_ASSOCIATIVE(PM_BINDING_POWER_TERNARY),
// .. ...
[PM_TOKEN_DOT_DOT] = NON_ASSOCIATIVE(PM_BINDING_POWER_RANGE),
[PM_TOKEN_DOT_DOT_DOT] = NON_ASSOCIATIVE(PM_BINDING_POWER_RANGE),
[PM_TOKEN_UDOT_DOT] = RIGHT_ASSOCIATIVE_UNARY(PM_BINDING_POWER_LOGICAL_OR),
[PM_TOKEN_UDOT_DOT_DOT] = RIGHT_ASSOCIATIVE_UNARY(PM_BINDING_POWER_LOGICAL_OR),
// ||
[PM_TOKEN_PIPE_PIPE] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_LOGICAL_OR),
// &&
[PM_TOKEN_AMPERSAND_AMPERSAND] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_LOGICAL_AND),
// != !~ == === =~ <=>
[PM_TOKEN_BANG_EQUAL] = NON_ASSOCIATIVE(PM_BINDING_POWER_EQUALITY),
[PM_TOKEN_BANG_TILDE] = NON_ASSOCIATIVE(PM_BINDING_POWER_EQUALITY),
[PM_TOKEN_EQUAL_EQUAL] = NON_ASSOCIATIVE(PM_BINDING_POWER_EQUALITY),
[PM_TOKEN_EQUAL_EQUAL_EQUAL] = NON_ASSOCIATIVE(PM_BINDING_POWER_EQUALITY),
[PM_TOKEN_EQUAL_TILDE] = NON_ASSOCIATIVE(PM_BINDING_POWER_EQUALITY),
[PM_TOKEN_LESS_EQUAL_GREATER] = NON_ASSOCIATIVE(PM_BINDING_POWER_EQUALITY),
// > >= < <=
[PM_TOKEN_GREATER] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_COMPARISON),
[PM_TOKEN_GREATER_EQUAL] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_COMPARISON),
[PM_TOKEN_LESS] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_COMPARISON),
[PM_TOKEN_LESS_EQUAL] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_COMPARISON),
// ^ |
[PM_TOKEN_CARET] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_BITWISE_OR),
[PM_TOKEN_PIPE] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_BITWISE_OR),
// &
[PM_TOKEN_AMPERSAND] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_BITWISE_AND),
// >> <<
[PM_TOKEN_GREATER_GREATER] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_SHIFT),
[PM_TOKEN_LESS_LESS] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_SHIFT),
// - +
[PM_TOKEN_MINUS] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_TERM),
[PM_TOKEN_PLUS] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_TERM),
// % / *
[PM_TOKEN_PERCENT] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_FACTOR),
[PM_TOKEN_SLASH] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_FACTOR),
[PM_TOKEN_STAR] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_FACTOR),
[PM_TOKEN_USTAR] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_FACTOR),
// -@
[PM_TOKEN_UMINUS] = RIGHT_ASSOCIATIVE_UNARY(PM_BINDING_POWER_UMINUS),
[PM_TOKEN_UMINUS_NUM] = { PM_BINDING_POWER_UMINUS, PM_BINDING_POWER_MAX, false, false },
// **
[PM_TOKEN_STAR_STAR] = RIGHT_ASSOCIATIVE(PM_BINDING_POWER_EXPONENT),
[PM_TOKEN_USTAR_STAR] = RIGHT_ASSOCIATIVE_UNARY(PM_BINDING_POWER_UNARY),
// ! ~ +@
[PM_TOKEN_BANG] = RIGHT_ASSOCIATIVE_UNARY(PM_BINDING_POWER_UNARY),
[PM_TOKEN_TILDE] = RIGHT_ASSOCIATIVE_UNARY(PM_BINDING_POWER_UNARY),
[PM_TOKEN_UPLUS] = RIGHT_ASSOCIATIVE_UNARY(PM_BINDING_POWER_UNARY),
// [
[PM_TOKEN_BRACKET_LEFT] = LEFT_ASSOCIATIVE(PM_BINDING_POWER_INDEX),
// :: . &.
[PM_TOKEN_COLON_COLON] = RIGHT_ASSOCIATIVE(PM_BINDING_POWER_CALL),
[PM_TOKEN_DOT] = RIGHT_ASSOCIATIVE(PM_BINDING_POWER_CALL),
[PM_TOKEN_AMPERSAND_DOT] = RIGHT_ASSOCIATIVE(PM_BINDING_POWER_CALL)
};
#undef BINDING_POWER_ASSIGNMENT
#undef LEFT_ASSOCIATIVE
#undef RIGHT_ASSOCIATIVE
#undef RIGHT_ASSOCIATIVE_UNARY
/**
* Returns true if the current token is of the given type.
*/
static inline bool
match1(const pm_parser_t *parser, pm_token_type_t type) {
return parser->current.type == type;
}
/**
* Returns true if the current token is of either of the given types.
*/
static inline bool
match2(const pm_parser_t *parser, pm_token_type_t type1, pm_token_type_t type2) {
return match1(parser, type1) || match1(parser, type2);
}
/**
* Returns true if the current token is any of the three given types.
*/
static inline bool
match3(const pm_parser_t *parser, pm_token_type_t type1, pm_token_type_t type2, pm_token_type_t type3) {
return match1(parser, type1) || match1(parser, type2) || match1(parser, type3);
}
/**
* Returns true if the current token is any of the four given types.
*/
static inline bool
match4(const pm_parser_t *parser, pm_token_type_t type1, pm_token_type_t type2, pm_token_type_t type3, pm_token_type_t type4) {
return match1(parser, type1) || match1(parser, type2) || match1(parser, type3) || match1(parser, type4);
}
/**
* Returns true if the current token is any of the five given types.
*/
static inline bool
match5(const pm_parser_t *parser, pm_token_type_t type1, pm_token_type_t type2, pm_token_type_t type3, pm_token_type_t type4, pm_token_type_t type5) {
return match1(parser, type1) || match1(parser, type2) || match1(parser, type3) || match1(parser, type4) || match1(parser, type5);
}
/**
* Returns true if the current token is any of the six given types.
*/
static inline bool
match6(const pm_parser_t *parser, pm_token_type_t type1, pm_token_type_t type2, pm_token_type_t type3, pm_token_type_t type4, pm_token_type_t type5, pm_token_type_t type6) {
return match1(parser, type1) || match1(parser, type2) || match1(parser, type3) || match1(parser, type4) || match1(parser, type5) || match1(parser, type6);
}
/**
* Returns true if the current token is any of the seven given types.
*/
static inline bool
match7(const pm_parser_t *parser, pm_token_type_t type1, pm_token_type_t type2, pm_token_type_t type3, pm_token_type_t type4, pm_token_type_t type5, pm_token_type_t type6, pm_token_type_t type7) {
return match1(parser, type1) || match1(parser, type2) || match1(parser, type3) || match1(parser, type4) || match1(parser, type5) || match1(parser, type6) || match1(parser, type7);
}
/**
* Returns true if the current token is any of the eight given types.
*/
static inline bool
match8(const pm_parser_t *parser, pm_token_type_t type1, pm_token_type_t type2, pm_token_type_t type3, pm_token_type_t type4, pm_token_type_t type5, pm_token_type_t type6, pm_token_type_t type7, pm_token_type_t type8) {
return match1(parser, type1) || match1(parser, type2) || match1(parser, type3) || match1(parser, type4) || match1(parser, type5) || match1(parser, type6) || match1(parser, type7) || match1(parser, type8);
}
/**
* If the current token is of the specified type, lex forward by one token and
* return true. Otherwise, return false. For example:
*
* if (accept1(parser, PM_TOKEN_COLON)) { ... }
*/
static bool
accept1(pm_parser_t *parser, pm_token_type_t type) {
if (match1(parser, type)) {
parser_lex(parser);
return true;
}
return false;
}
/**
* If the current token is either of the two given types, lex forward by one
* token and return true. Otherwise return false.
*/
static inline bool
accept2(pm_parser_t *parser, pm_token_type_t type1, pm_token_type_t type2) {
if (match2(parser, type1, type2)) {
parser_lex(parser);
return true;
}
return false;
}
/**
* If the current token is any of the three given types, lex forward by one
* token and return true. Otherwise return false.
*/
static inline bool
accept3(pm_parser_t *parser, pm_token_type_t type1, pm_token_type_t type2, pm_token_type_t type3) {
if (match3(parser, type1, type2, type3)) {
parser_lex(parser);
return true;
}
return false;
}
/**
* This function indicates that the parser expects a token in a specific
* position. For example, if you're parsing a BEGIN block, you know that a { is
* expected immediately after the keyword. In that case you would call this
* function to indicate that that token should be found.
*
* If we didn't find the token that we were expecting, then we're going to add
* an error to the parser's list of errors (to indicate that the tree is not
* valid) and create an artificial token instead. This allows us to recover from
* the fact that the token isn't present and continue parsing.
*/
static void
expect1(pm_parser_t *parser, pm_token_type_t type, pm_diagnostic_id_t diag_id) {
if (accept1(parser, type)) return;
const uint8_t *location = parser->previous.end;
pm_parser_err(parser, location, location, diag_id);
parser->previous.start = location;
parser->previous.type = PM_TOKEN_MISSING;
}
/**
* This function is the same as expect1, but it expects either of two token
* types.
*/
static void
expect2(pm_parser_t *parser, pm_token_type_t type1, pm_token_type_t type2, pm_diagnostic_id_t diag_id) {
if (accept2(parser, type1, type2)) return;
const uint8_t *location = parser->previous.end;
pm_parser_err(parser, location, location, diag_id);
parser->previous.start = location;
parser->previous.type = PM_TOKEN_MISSING;
}
/**
* This function is the same as expect2, but it expects one of three token types.
*/
static void
expect3(pm_parser_t *parser, pm_token_type_t type1, pm_token_type_t type2, pm_token_type_t type3, pm_diagnostic_id_t diag_id) {
if (accept3(parser, type1, type2, type3)) return;
const uint8_t *location = parser->previous.end;
pm_parser_err(parser, location, location, diag_id);
parser->previous.start = location;
parser->previous.type = PM_TOKEN_MISSING;
}
static pm_node_t *
parse_expression(pm_parser_t *parser, pm_binding_power_t binding_power, bool accepts_command_call, pm_diagnostic_id_t diag_id);
/**
* This is a wrapper of parse_expression, which also checks whether the resulting node is value expression.
*/
static pm_node_t *
parse_value_expression(pm_parser_t *parser, pm_binding_power_t binding_power, bool accepts_command_call, pm_diagnostic_id_t diag_id) {
pm_node_t *node = parse_expression(parser, binding_power, accepts_command_call, diag_id);
pm_assert_value_expression(parser, node);
return node;
}
/**
* This function controls whether or not we will attempt to parse an expression
* beginning at the subsequent token. It is used when we are in a context where
* an expression is optional.
*
* For example, looking at a range object when we've already lexed the operator,
* we need to know if we should attempt to parse an expression on the right.
*
* For another example, if we've parsed an identifier or a method call and we do
* not have parentheses, then the next token may be the start of an argument or
* it may not.
*
* CRuby parsers that are generated would resolve this by using a lookahead and
* potentially backtracking. We attempt to do this by just looking at the next
* token and making a decision based on that. I am not sure if this is going to
*
* work in all cases, it may need to be refactored later. But it appears to work
* for now.
*/
static inline bool
token_begins_expression_p(pm_token_type_t type) {
switch (type) {
case PM_TOKEN_EQUAL_GREATER:
case PM_TOKEN_KEYWORD_IN:
// We need to special case this because it is a binary operator that
// should not be marked as beginning an expression.
return false;
case PM_TOKEN_BRACE_RIGHT:
case PM_TOKEN_BRACKET_RIGHT:
case PM_TOKEN_COLON:
case PM_TOKEN_COMMA:
case PM_TOKEN_EMBEXPR_END:
case PM_TOKEN_EOF:
case PM_TOKEN_LAMBDA_BEGIN:
case PM_TOKEN_KEYWORD_DO:
case PM_TOKEN_KEYWORD_DO_LOOP:
case PM_TOKEN_KEYWORD_END:
case PM_TOKEN_KEYWORD_ELSE:
case PM_TOKEN_KEYWORD_ELSIF:
case PM_TOKEN_KEYWORD_ENSURE:
case PM_TOKEN_KEYWORD_THEN:
case PM_TOKEN_KEYWORD_RESCUE:
case PM_TOKEN_KEYWORD_WHEN:
case PM_TOKEN_NEWLINE:
case PM_TOKEN_PARENTHESIS_RIGHT:
case PM_TOKEN_SEMICOLON:
// The reason we need this short-circuit is because we're using the
// binding powers table to tell us if the subsequent token could
// potentially be the start of an expression . If there _is_ a binding
// power for one of these tokens, then we should remove it from this list
// and let it be handled by the default case below.
assert(pm_binding_powers[type].left == PM_BINDING_POWER_UNSET);
return false;
case PM_TOKEN_UAMPERSAND:
// This is a special case because this unary operator cannot appear
// as a general operator, it only appears in certain circumstances.
return false;
case PM_TOKEN_UCOLON_COLON:
case PM_TOKEN_UMINUS:
case PM_TOKEN_UMINUS_NUM:
case PM_TOKEN_UPLUS:
case PM_TOKEN_BANG:
case PM_TOKEN_TILDE:
case PM_TOKEN_UDOT_DOT:
case PM_TOKEN_UDOT_DOT_DOT:
// These unary tokens actually do have binding power associated with them
// so that we can correctly place them into the precedence order. But we
// want them to be marked as beginning an expression, so we need to
// special case them here.
return true;
default:
return pm_binding_powers[type].left == PM_BINDING_POWER_UNSET;
}
}
/**
* Parse an expression with the given binding power that may be optionally
* prefixed by the * operator.
*/
static pm_node_t *
parse_starred_expression(pm_parser_t *parser, pm_binding_power_t binding_power, bool accepts_command_call, pm_diagnostic_id_t diag_id) {
if (accept1(parser, PM_TOKEN_USTAR)) {
pm_token_t operator = parser->previous;
pm_node_t *expression = parse_value_expression(parser, binding_power, false, PM_ERR_EXPECT_EXPRESSION_AFTER_STAR);
return (pm_node_t *) pm_splat_node_create(parser, &operator, expression);
}
return parse_value_expression(parser, binding_power, accepts_command_call, diag_id);
}
/**
* Convert the name of a method into the corresponding write method name. For
* example, foo would be turned into foo=.
*/
static void
parse_write_name(pm_parser_t *parser, pm_constant_id_t *name_field) {
// The method name needs to change. If we previously had
// foo, we now need foo=. In this case we'll allocate a new
// owned string, copy the previous method name in, and
// append an =.
pm_constant_t *constant = pm_constant_pool_id_to_constant(&parser->constant_pool, *name_field);
size_t length = constant->length;
uint8_t *name = calloc(length + 1, sizeof(uint8_t));
if (name == NULL) return;
memcpy(name, constant->start, length);
name[length] = '=';
// Now switch the name to the new string.
// This silences clang analyzer warning about leak of memory pointed by `name`.
// NOLINTNEXTLINE(clang-analyzer-*)
*name_field = pm_constant_pool_insert_owned(&parser->constant_pool, name, length + 1);
}
/**
* Convert the given node into a valid target node.
*/
static pm_node_t *
parse_target(pm_parser_t *parser, pm_node_t *target) {
switch (PM_NODE_TYPE(target)) {
case PM_MISSING_NODE:
return target;
case PM_CLASS_VARIABLE_READ_NODE:
assert(sizeof(pm_class_variable_target_node_t) == sizeof(pm_class_variable_read_node_t));
target->type = PM_CLASS_VARIABLE_TARGET_NODE;
return target;
case PM_CONSTANT_PATH_NODE:
assert(sizeof(pm_constant_path_target_node_t) == sizeof(pm_constant_path_node_t));
target->type = PM_CONSTANT_PATH_TARGET_NODE;
return target;
case PM_CONSTANT_READ_NODE:
assert(sizeof(pm_constant_target_node_t) == sizeof(pm_constant_read_node_t));
target->type = PM_CONSTANT_TARGET_NODE;
return target;
case PM_BACK_REFERENCE_READ_NODE:
case PM_NUMBERED_REFERENCE_READ_NODE:
pm_parser_err_node(parser, target, PM_ERR_WRITE_TARGET_READONLY);
return target;
case PM_GLOBAL_VARIABLE_READ_NODE:
assert(sizeof(pm_global_variable_target_node_t) == sizeof(pm_global_variable_read_node_t));
target->type = PM_GLOBAL_VARIABLE_TARGET_NODE;
return target;
case PM_LOCAL_VARIABLE_READ_NODE:
if (pm_token_is_numbered_parameter(target->location.start, target->location.end)) {
PM_PARSER_ERR_NODE_FORMAT(parser, target, PM_ERR_PARAMETER_NUMBERED_RESERVED, target->location.start);
} else {
assert(sizeof(pm_local_variable_target_node_t) == sizeof(pm_local_variable_read_node_t));
target->type = PM_LOCAL_VARIABLE_TARGET_NODE;
}
return target;
case PM_INSTANCE_VARIABLE_READ_NODE:
assert(sizeof(pm_instance_variable_target_node_t) == sizeof(pm_instance_variable_read_node_t));
target->type = PM_INSTANCE_VARIABLE_TARGET_NODE;
return target;
case PM_MULTI_TARGET_NODE:
return target;
case PM_SPLAT_NODE: {
pm_splat_node_t *splat = (pm_splat_node_t *) target;
if (splat->expression != NULL) {
splat->expression = parse_target(parser, splat->expression);
}
return (pm_node_t *) splat;
}
case PM_CALL_NODE: {
pm_call_node_t *call = (pm_call_node_t *) target;
// If we have no arguments to the call node and we need this to be a
// target then this is either a method call or a local variable write.
if (
(call->message_loc.start != NULL) &&
(call->message_loc.end[-1] != '!') &&
(call->message_loc.end[-1] != '?') &&
(call->opening_loc.start == NULL) &&
(call->arguments == NULL) &&
(call->block == NULL)
) {
if (call->receiver == NULL) {
// When we get here, we have a local variable write, because it
// was previously marked as a method call but now we have an =.
// This looks like:
//
// foo = 1
//
// When it was parsed in the prefix position, foo was seen as a
// method call with no receiver and no arguments. Now we have an
// =, so we know it's a local variable write.
const pm_location_t message = call->message_loc;
pm_parser_local_add_location(parser, message.start, message.end);
pm_node_destroy(parser, target);
uint32_t depth = 0;
const pm_token_t name = { .type = PM_TOKEN_IDENTIFIER, .start = message.start, .end = message.end };
target = (pm_node_t *) pm_local_variable_read_node_create(parser, &name, depth);
assert(sizeof(pm_local_variable_target_node_t) == sizeof(pm_local_variable_read_node_t));
target->type = PM_LOCAL_VARIABLE_TARGET_NODE;
pm_refute_numbered_parameter(parser, message.start, message.end);
return target;
}
if (*call->message_loc.start == '_' || parser->encoding->alnum_char(call->message_loc.start, call->message_loc.end - call->message_loc.start)) {
parse_write_name(parser, &call->name);
return (pm_node_t *) pm_call_target_node_create(parser, call);
}
}
// If there is no call operator and the message is "[]" then this is
// an aref expression, and we can transform it into an aset
// expression.
if (pm_call_node_index_p(call)) {
return (pm_node_t *) pm_index_target_node_create(parser, call);
}
}
/* fallthrough */
default:
// In this case we have a node that we don't know how to convert
// into a target. We need to treat it as an error. For now, we'll
// mark it as an error and just skip right past it.
pm_parser_err_node(parser, target, PM_ERR_WRITE_TARGET_UNEXPECTED);
return target;
}
}
/**
* Parse a write target and validate that it is in a valid position for
* assignment.
*/
static pm_node_t *
parse_target_validate(pm_parser_t *parser, pm_node_t *target) {
pm_node_t *result = parse_target(parser, target);
// Ensure that we have one of an =, an 'in' in for indexes, and a ')' in parens after the targets.
if (
!match1(parser, PM_TOKEN_EQUAL) &&
!(context_p(parser, PM_CONTEXT_FOR_INDEX) && match1(parser, PM_TOKEN_KEYWORD_IN)) &&
!(context_p(parser, PM_CONTEXT_PARENS) && match1(parser, PM_TOKEN_PARENTHESIS_RIGHT))
) {
pm_parser_err_node(parser, result, PM_ERR_WRITE_TARGET_UNEXPECTED);
}
return result;
}
/**
* Convert the given node into a valid write node.
*/
static pm_node_t *
parse_write(pm_parser_t *parser, pm_node_t *target, pm_token_t *operator, pm_node_t *value) {
switch (PM_NODE_TYPE(target)) {
case PM_MISSING_NODE:
pm_node_destroy(parser, value);
return target;
case PM_CLASS_VARIABLE_READ_NODE: {
pm_class_variable_write_node_t *node = pm_class_variable_write_node_create(parser, (pm_class_variable_read_node_t *) target, operator, value);
pm_node_destroy(parser, target);
return (pm_node_t *) node;
}
case PM_CONSTANT_PATH_NODE:
return (pm_node_t *) pm_constant_path_write_node_create(parser, (pm_constant_path_node_t *) target, operator, value);
case PM_CONSTANT_READ_NODE: {
pm_constant_write_node_t *node = pm_constant_write_node_create(parser, (pm_constant_read_node_t *) target, operator, value);
if (context_def_p(parser)) {
pm_parser_err_node(parser, (pm_node_t *) node, PM_ERR_WRITE_TARGET_IN_METHOD);
}
pm_node_destroy(parser, target);
return (pm_node_t *) node;
}
case PM_BACK_REFERENCE_READ_NODE:
case PM_NUMBERED_REFERENCE_READ_NODE:
pm_parser_err_node(parser, target, PM_ERR_WRITE_TARGET_READONLY);
/* fallthrough */
case PM_GLOBAL_VARIABLE_READ_NODE: {
pm_global_variable_write_node_t *node = pm_global_variable_write_node_create(parser, target, operator, value);
pm_node_destroy(parser, target);
return (pm_node_t *) node;
}
case PM_LOCAL_VARIABLE_READ_NODE: {
pm_refute_numbered_parameter(parser, target->location.start, target->location.end);
pm_local_variable_read_node_t *local_read = (pm_local_variable_read_node_t *) target;
pm_constant_id_t constant_id = local_read->name;
uint32_t depth = local_read->depth;
pm_location_t name_loc = target->location;
pm_node_destroy(parser, target);
return (pm_node_t *) pm_local_variable_write_node_create(parser, constant_id, depth, value, &name_loc, operator);
}
case PM_INSTANCE_VARIABLE_READ_NODE: {
pm_node_t *write_node = (pm_node_t *) pm_instance_variable_write_node_create(parser, (pm_instance_variable_read_node_t *) target, operator, value);
pm_node_destroy(parser, target);
return write_node;
}
case PM_MULTI_TARGET_NODE:
return (pm_node_t *) pm_multi_write_node_create(parser, (pm_multi_target_node_t *) target, operator, value);
case PM_SPLAT_NODE: {
pm_splat_node_t *splat = (pm_splat_node_t *) target;
if (splat->expression != NULL) {
splat->expression = parse_write(parser, splat->expression, operator, value);
}
pm_multi_target_node_t *multi_target = pm_multi_target_node_create(parser);
pm_multi_target_node_targets_append(parser, multi_target, (pm_node_t *) splat);
return (pm_node_t *) pm_multi_write_node_create(parser, multi_target, operator, value);
}
case PM_CALL_NODE: {
pm_call_node_t *call = (pm_call_node_t *) target;
// If we have no arguments to the call node and we need this to be a
// target then this is either a method call or a local variable
// write.
if (
(call->message_loc.start != NULL) &&
(call->message_loc.end[-1] != '!') &&
(call->message_loc.end[-1] != '?') &&
(call->opening_loc.start == NULL) &&
(call->arguments == NULL) &&
(call->block == NULL)
) {
if (call->receiver == NULL) {
// When we get here, we have a local variable write, because it
// was previously marked as a method call but now we have an =.
// This looks like:
//
// foo = 1
//
// When it was parsed in the prefix position, foo was seen as a
// method call with no receiver and no arguments. Now we have an
// =, so we know it's a local variable write.
const pm_location_t message = call->message_loc;
pm_parser_local_add_location(parser, message.start, message.end);
pm_node_destroy(parser, target);
pm_constant_id_t constant_id = pm_parser_constant_id_location(parser, message.start, message.end);
target = (pm_node_t *) pm_local_variable_write_node_create(parser, constant_id, 0, value, &message, operator);
pm_refute_numbered_parameter(parser, message.start, message.end);
return target;
}
if (*call->message_loc.start == '_' || parser->encoding->alnum_char(call->message_loc.start, call->message_loc.end - call->message_loc.start)) {
// When we get here, we have a method call, because it was
// previously marked as a method call but now we have an =. This
// looks like:
//
// foo.bar = 1
//
// When it was parsed in the prefix position, foo.bar was seen as a
// method call with no arguments. Now we have an =, so we know it's
// a method call with an argument. In this case we will create the
// arguments node, parse the argument, and add it to the list.
pm_arguments_node_t *arguments = pm_arguments_node_create(parser);
call->arguments = arguments;
pm_arguments_node_arguments_append(arguments, value);
call->base.location.end = arguments->base.location.end;
parse_write_name(parser, &call->name);
pm_node_flag_set((pm_node_t *) call, PM_CALL_NODE_FLAGS_ATTRIBUTE_WRITE);
return (pm_node_t *) call;
}
}
// If there is no call operator and the message is "[]" then this is
// an aref expression, and we can transform it into an aset
// expression.
if (pm_call_node_index_p(call)) {
if (call->arguments == NULL) {
call->arguments = pm_arguments_node_create(parser);
}
pm_arguments_node_arguments_append(call->arguments, value);
target->location.end = value->location.end;
// Replace the name with "[]=".
call->name = pm_parser_constant_id_constant(parser, "[]=", 3);
pm_node_flag_set((pm_node_t *) call, PM_CALL_NODE_FLAGS_ATTRIBUTE_WRITE);
return target;
}
// If there are arguments on the call node, then it can't be a method
// call ending with = or a local variable write, so it must be a
// syntax error. In this case we'll fall through to our default
// handling. We need to free the value that we parsed because there
// is no way for us to attach it to the tree at this point.
pm_node_destroy(parser, value);
}
/* fallthrough */
default:
// In this case we have a node that we don't know how to convert into a
// target. We need to treat it as an error. For now, we'll mark it as an
// error and just skip right past it.
pm_parser_err_token(parser, operator, PM_ERR_WRITE_TARGET_UNEXPECTED);
return target;
}
}
/**
* Parse a list of targets for assignment. This is used in the case of a for
* loop or a multi-assignment. For example, in the following code:
*
* for foo, bar in baz
* ^^^^^^^^
*
* The targets are `foo` and `bar`. This function will either return a single
* target node or a multi-target node.
*/
static pm_node_t *
parse_targets(pm_parser_t *parser, pm_node_t *first_target, pm_binding_power_t binding_power) {
bool has_rest = PM_NODE_TYPE_P(first_target, PM_SPLAT_NODE);
pm_multi_target_node_t *result = pm_multi_target_node_create(parser);
pm_multi_target_node_targets_append(parser, result, parse_target(parser, first_target));
while (accept1(parser, PM_TOKEN_COMMA)) {
if (accept1(parser, PM_TOKEN_USTAR)) {
// Here we have a splat operator. It can have a name or be
// anonymous. It can be the final target or be in the middle if
// there haven't been any others yet.
if (has_rest) {
pm_parser_err_previous(parser, PM_ERR_MULTI_ASSIGN_MULTI_SPLATS);
}
pm_token_t star_operator = parser->previous;
pm_node_t *name = NULL;
if (token_begins_expression_p(parser->current.type)) {
name = parse_expression(parser, binding_power, false, PM_ERR_EXPECT_EXPRESSION_AFTER_STAR);
name = parse_target(parser, name);
}
pm_node_t *splat = (pm_node_t *) pm_splat_node_create(parser, &star_operator, name);
pm_multi_target_node_targets_append(parser, result, splat);
has_rest = true;
} else if (token_begins_expression_p(parser->current.type)) {
pm_node_t *target = parse_expression(parser, binding_power, false, PM_ERR_EXPECT_EXPRESSION_AFTER_COMMA);
target = parse_target(parser, target);
pm_multi_target_node_targets_append(parser, result, target);
} else if (!match1(parser, PM_TOKEN_EOF)) {
// If we get here, then we have a trailing , in a multi target node.
// We'll set the implicit rest flag to indicate this.
pm_node_t *rest = (pm_node_t *) pm_implicit_rest_node_create(parser, &parser->previous);
pm_multi_target_node_targets_append(parser, result, rest);
break;
}
}
return (pm_node_t *) result;
}
/**
* Parse a list of targets and validate that it is in a valid position for
* assignment.
*/
static pm_node_t *
parse_targets_validate(pm_parser_t *parser, pm_node_t *first_target, pm_binding_power_t binding_power) {
pm_node_t *result = parse_targets(parser, first_target, binding_power);
// Ensure that we have either an = or a ) after the targets.
if (!match2(parser, PM_TOKEN_EQUAL, PM_TOKEN_PARENTHESIS_RIGHT)) {
pm_parser_err_node(parser, result, PM_ERR_WRITE_TARGET_UNEXPECTED);
}
return result;
}
/**
* Parse a list of statements separated by newlines or semicolons.
*/
static pm_statements_node_t *
parse_statements(pm_parser_t *parser, pm_context_t context) {
// First, skip past any optional terminators that might be at the beginning of
// the statements.
while (accept2(parser, PM_TOKEN_SEMICOLON, PM_TOKEN_NEWLINE));
// If we have a terminator, then we can just return NULL.
if (context_terminator(context, &parser->current)) return NULL;
pm_statements_node_t *statements = pm_statements_node_create(parser);
// At this point we know we have at least one statement, and that it
// immediately follows the current token.
context_push(parser, context);
while (true) {
pm_node_t *node = parse_expression(parser, PM_BINDING_POWER_STATEMENT, true, PM_ERR_CANNOT_PARSE_EXPRESSION);
pm_statements_node_body_append(statements, node);
// If we're recovering from a syntax error, then we need to stop parsing the
// statements now.
if (parser->recovering) {
// If this is the level of context where the recovery has happened, then
// we can mark the parser as done recovering.
if (context_terminator(context, &parser->current)) parser->recovering = false;
break;
}
// If we have a terminator, then we will parse all consequtive terminators
// and then continue parsing the statements list.
if (accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON)) {
// If we have a terminator, then we will continue parsing the statements
// list.
while (accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON));
if (context_terminator(context, &parser->current)) break;
// Now we can continue parsing the list of statements.
continue;
}
// At this point we have a list of statements that are not terminated by a
// newline or semicolon. At this point we need to check if we're at the end
// of the statements list. If we are, then we should break out of the loop.
if (context_terminator(context, &parser->current)) break;
// At this point, we have a syntax error, because the statement was not
// terminated by a newline or semicolon, and we're not at the end of the
// statements list. Ideally we should scan forward to determine if we should
// insert a missing terminator or break out of parsing the statements list
// at this point.
//
// We don't have that yet, so instead we'll do a more naive approach. If we
// were unable to parse an expression, then we will skip past this token and
// continue parsing the statements list. Otherwise we'll add an error and
// continue parsing the statements list.
if (PM_NODE_TYPE_P(node, PM_MISSING_NODE)) {
parser_lex(parser);
while (accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON));
if (context_terminator(context, &parser->current)) break;
} else {
expect1(parser, PM_TOKEN_NEWLINE, PM_ERR_EXPECT_EOL_AFTER_STATEMENT);
}
}
context_pop(parser);
return statements;
}
/**
* Parse all of the elements of a hash. returns true if a double splat was found.
*/
static bool
parse_assocs(pm_parser_t *parser, pm_node_t *node) {
assert(PM_NODE_TYPE_P(node, PM_HASH_NODE) || PM_NODE_TYPE_P(node, PM_KEYWORD_HASH_NODE));
bool contains_keyword_splat = false;
while (true) {
pm_node_t *element;
switch (parser->current.type) {
case PM_TOKEN_USTAR_STAR: {
parser_lex(parser);
pm_token_t operator = parser->previous;
pm_node_t *value = NULL;
if (token_begins_expression_p(parser->current.type)) {
value = parse_value_expression(parser, PM_BINDING_POWER_DEFINED, false, PM_ERR_EXPECT_EXPRESSION_AFTER_SPLAT_HASH);
} else if (pm_parser_local_depth(parser, &operator) == -1) {
pm_parser_err_token(parser, &operator, PM_ERR_EXPECT_EXPRESSION_AFTER_SPLAT_HASH);
}
element = (pm_node_t *) pm_assoc_splat_node_create(parser, value, &operator);
contains_keyword_splat = true;
break;
}
case PM_TOKEN_LABEL: {
pm_token_t label = parser->current;
parser_lex(parser);
pm_node_t *key = (pm_node_t *) pm_symbol_node_label_create(parser, &label);
pm_token_t operator = not_provided(parser);
pm_node_t *value = NULL;
if (token_begins_expression_p(parser->current.type)) {
value = parse_value_expression(parser, PM_BINDING_POWER_DEFINED, false, PM_ERR_HASH_EXPRESSION_AFTER_LABEL);
} else {
if (parser->encoding->isupper_char(label.start, (label.end - 1) - label.start)) {
pm_token_t constant = { .type = PM_TOKEN_CONSTANT, .start = label.start, .end = label.end - 1 };
value = (pm_node_t *) pm_constant_read_node_create(parser, &constant);
} else {
int depth = pm_parser_local_depth(parser, &((pm_token_t) { .type = PM_TOKEN_IDENTIFIER, .start = label.start, .end = label.end - 1 }));
pm_token_t identifier = { .type = PM_TOKEN_IDENTIFIER, .start = label.start, .end = label.end - 1 };
if (depth == -1) {
value = (pm_node_t *) pm_call_node_variable_call_create(parser, &identifier);
} else {
value = (pm_node_t *) pm_local_variable_read_node_create(parser, &identifier, (uint32_t) depth);
}
}
value->location.end++;
value = (pm_node_t *) pm_implicit_node_create(parser, value);
}
element = (pm_node_t *) pm_assoc_node_create(parser, key, &operator, value);
break;
}
default: {
pm_node_t *key = parse_value_expression(parser, PM_BINDING_POWER_DEFINED, false, PM_ERR_HASH_KEY);
pm_token_t operator;
if (pm_symbol_node_label_p(key)) {
operator = not_provided(parser);
} else {
expect1(parser, PM_TOKEN_EQUAL_GREATER, PM_ERR_HASH_ROCKET);
operator = parser->previous;
}
pm_node_t *value = parse_value_expression(parser, PM_BINDING_POWER_DEFINED, false, PM_ERR_HASH_VALUE);
element = (pm_node_t *) pm_assoc_node_create(parser, key, &operator, value);
break;
}
}
if (PM_NODE_TYPE_P(node, PM_HASH_NODE)) {
pm_hash_node_elements_append((pm_hash_node_t *) node, element);
} else {
pm_keyword_hash_node_elements_append((pm_keyword_hash_node_t *) node, element);
}
// If there's no comma after the element, then we're done.
if (!accept1(parser, PM_TOKEN_COMMA)) break;
// If the next element starts with a label or a **, then we know we have
// another element in the hash, so we'll continue parsing.
if (match2(parser, PM_TOKEN_USTAR_STAR, PM_TOKEN_LABEL)) continue;
// Otherwise we need to check if the subsequent token begins an expression.
// If it does, then we'll continue parsing.
if (token_begins_expression_p(parser->current.type)) continue;
// Otherwise by default we will exit out of this loop.
break;
}
return contains_keyword_splat;
}
/**
* Append an argument to a list of arguments.
*/
static inline void
parse_arguments_append(pm_parser_t *parser, pm_arguments_t *arguments, pm_node_t *argument) {
if (arguments->arguments == NULL) {
arguments->arguments = pm_arguments_node_create(parser);
}
pm_arguments_node_arguments_append(arguments->arguments, argument);
}
/**
* Parse a list of arguments.
*/
static void
parse_arguments(pm_parser_t *parser, pm_arguments_t *arguments, bool accepts_forwarding, pm_token_type_t terminator) {
pm_binding_power_t binding_power = pm_binding_powers[parser->current.type].left;
// First we need to check if the next token is one that could be the start of
// an argument. If it's not, then we can just return.
if (
match2(parser, terminator, PM_TOKEN_EOF) ||
(binding_power != PM_BINDING_POWER_UNSET && binding_power < PM_BINDING_POWER_RANGE) ||
context_terminator(parser->current_context->context, &parser->current)
) {
return;
}
bool parsed_first_argument = false;
bool parsed_bare_hash = false;
bool parsed_block_argument = false;
bool parsed_forwarding_arguments = false;
while (!match1(parser, PM_TOKEN_EOF)) {
if (parsed_block_argument) {
pm_parser_err_current(parser, PM_ERR_ARGUMENT_AFTER_BLOCK);
}
if (parsed_forwarding_arguments) {
pm_parser_err_current(parser, PM_ERR_ARGUMENT_AFTER_FORWARDING_ELLIPSES);
}
pm_node_t *argument = NULL;
switch (parser->current.type) {
case PM_TOKEN_USTAR_STAR:
case PM_TOKEN_LABEL: {
if (parsed_bare_hash) {
pm_parser_err_current(parser, PM_ERR_ARGUMENT_BARE_HASH);
}
pm_keyword_hash_node_t *hash = pm_keyword_hash_node_create(parser);
argument = (pm_node_t *) hash;
bool contains_keyword_splat = parse_assocs(parser, (pm_node_t *) hash);
parsed_bare_hash = true;
parse_arguments_append(parser, arguments, argument);
if (contains_keyword_splat) {
pm_node_flag_set((pm_node_t *)arguments->arguments, PM_ARGUMENTS_NODE_FLAGS_CONTAINS_KEYWORD_SPLAT);
}
break;
}
case PM_TOKEN_UAMPERSAND: {
parser_lex(parser);
pm_token_t operator = parser->previous;
pm_node_t *expression = NULL;
if (token_begins_expression_p(parser->current.type)) {
expression = parse_value_expression(parser, PM_BINDING_POWER_DEFINED, false, PM_ERR_EXPECT_ARGUMENT);
} else {
if (pm_parser_local_depth(parser, &operator) == -1) {
// A block forwarding in a method having `...` parameter (e.g. `def foo(...); bar(&); end`) is available.
pm_constant_id_t ellipsis_id = pm_parser_constant_id_constant(parser, "...", 3);
if (pm_parser_local_depth_constant_id(parser, ellipsis_id) == -1) {
pm_parser_err_token(parser, &operator, PM_ERR_ARGUMENT_NO_FORWARDING_AMP);
}
}
}
argument = (pm_node_t *) pm_block_argument_node_create(parser, &operator, expression);
if (parsed_block_argument) {
parse_arguments_append(parser, arguments, argument);
} else {
arguments->block = argument;
}
parsed_block_argument = true;
break;
}
case PM_TOKEN_USTAR: {
parser_lex(parser);
pm_token_t operator = parser->previous;
if (match4(parser, PM_TOKEN_PARENTHESIS_RIGHT, PM_TOKEN_COMMA, PM_TOKEN_SEMICOLON, PM_TOKEN_BRACKET_RIGHT)) {
if (pm_parser_local_depth(parser, &parser->previous) == -1) {
pm_parser_err_token(parser, &operator, PM_ERR_ARGUMENT_NO_FORWARDING_STAR);
}
argument = (pm_node_t *) pm_splat_node_create(parser, &operator, NULL);
} else {
pm_node_t *expression = parse_value_expression(parser, PM_BINDING_POWER_DEFINED, false, PM_ERR_EXPECT_EXPRESSION_AFTER_SPLAT);
if (parsed_bare_hash) {
pm_parser_err(parser, operator.start, expression->location.end, PM_ERR_ARGUMENT_SPLAT_AFTER_ASSOC_SPLAT);
}
argument = (pm_node_t *) pm_splat_node_create(parser, &operator, expression);
}
parse_arguments_append(parser, arguments, argument);
break;
}
case PM_TOKEN_UDOT_DOT_DOT: {
if (accepts_forwarding) {
parser_lex(parser);
if (token_begins_expression_p(parser->current.type)) {
// If the token begins an expression then this ... was not actually
// argument forwarding but was instead a range.
pm_token_t operator = parser->previous;
pm_node_t *right = parse_expression(parser, PM_BINDING_POWER_RANGE, false, PM_ERR_EXPECT_EXPRESSION_AFTER_OPERATOR);
argument = (pm_node_t *) pm_range_node_create(parser, NULL, &operator, right);
} else {
if (pm_parser_local_depth(parser, &parser->previous) == -1) {
pm_parser_err_previous(parser, PM_ERR_ARGUMENT_NO_FORWARDING_ELLIPSES);
}
if (parsed_first_argument && terminator == PM_TOKEN_EOF) {
pm_parser_err_previous(parser, PM_ERR_ARGUMENT_FORWARDING_UNBOUND);
}
argument = (pm_node_t *) pm_forwarding_arguments_node_create(parser, &parser->previous);
parse_arguments_append(parser, arguments, argument);
parsed_forwarding_arguments = true;
break;
}
}
}
/* fallthrough */
default: {
if (argument == NULL) {
argument = parse_value_expression(parser, PM_BINDING_POWER_DEFINED, !parsed_first_argument, PM_ERR_EXPECT_ARGUMENT);
}
bool contains_keyword_splat = false;
if (pm_symbol_node_label_p(argument) || accept1(parser, PM_TOKEN_EQUAL_GREATER)) {
if (parsed_bare_hash) {
pm_parser_err_previous(parser, PM_ERR_ARGUMENT_BARE_HASH);
}
pm_token_t operator;
if (parser->previous.type == PM_TOKEN_EQUAL_GREATER) {
operator = parser->previous;
} else {
operator = not_provided(parser);
}
pm_keyword_hash_node_t *bare_hash = pm_keyword_hash_node_create(parser);
// Finish parsing the one we are part way through
pm_node_t *value = parse_value_expression(parser, PM_BINDING_POWER_DEFINED, false, PM_ERR_HASH_VALUE);
argument = (pm_node_t *) pm_assoc_node_create(parser, argument, &operator, value);
pm_keyword_hash_node_elements_append(bare_hash, argument);
argument = (pm_node_t *) bare_hash;
// Then parse more if we have a comma
if (accept1(parser, PM_TOKEN_COMMA) && (
token_begins_expression_p(parser->current.type) ||
match2(parser, PM_TOKEN_USTAR_STAR, PM_TOKEN_LABEL)
)) {
contains_keyword_splat = parse_assocs(parser, (pm_node_t *) bare_hash);
}
parsed_bare_hash = true;
}
parse_arguments_append(parser, arguments, argument);
if (contains_keyword_splat) {
pm_node_flag_set((pm_node_t *)arguments->arguments, PM_ARGUMENTS_NODE_FLAGS_CONTAINS_KEYWORD_SPLAT);
}
break;
}
}
parsed_first_argument = true;
// If parsing the argument failed, we need to stop parsing arguments.
if (PM_NODE_TYPE_P(argument, PM_MISSING_NODE) || parser->recovering) break;
// If the terminator of these arguments is not EOF, then we have a specific
// token we're looking for. In that case we can accept a newline here
// because it is not functioning as a statement terminator.
if (terminator != PM_TOKEN_EOF) accept1(parser, PM_TOKEN_NEWLINE);
if (parser->previous.type == PM_TOKEN_COMMA && parsed_bare_hash) {
// If we previously were on a comma and we just parsed a bare hash, then
// we want to continue parsing arguments. This is because the comma was
// grabbed up by the hash parser.
} else {
// If there is no comma at the end of the argument list then we're done
// parsing arguments and can break out of this loop.
if (!accept1(parser, PM_TOKEN_COMMA)) break;
}
// If we hit the terminator, then that means we have a trailing comma so we
// can accept that output as well.
if (match1(parser, terminator)) break;
}
}
/**
* Required parameters on method, block, and lambda declarations can be
* destructured using parentheses. This looks like:
*
* def foo((bar, baz))
* end
*
*
* It can recurse infinitely down, and splats are allowed to group arguments.
*/
static pm_multi_target_node_t *
parse_required_destructured_parameter(pm_parser_t *parser) {
expect1(parser, PM_TOKEN_PARENTHESIS_LEFT, PM_ERR_EXPECT_LPAREN_REQ_PARAMETER);
pm_multi_target_node_t *node = pm_multi_target_node_create(parser);
pm_multi_target_node_opening_set(node, &parser->previous);
do {
pm_node_t *param;
// If we get here then we have a trailing comma, which isn't allowed in
// the grammar. In other places, multi targets _do_ allow trailing
// commas, so here we'll assume this is a mistake of the user not
// knowing it's not allowed here.
if (node->lefts.size > 0 && match1(parser, PM_TOKEN_PARENTHESIS_RIGHT)) {
param = (pm_node_t *) pm_implicit_rest_node_create(parser, &parser->previous);
pm_multi_target_node_targets_append(parser, node, param);
pm_parser_err_current(parser, PM_ERR_PARAMETER_WILD_LOOSE_COMMA);
break;
}
if (match1(parser, PM_TOKEN_PARENTHESIS_LEFT)) {
param = (pm_node_t *) parse_required_destructured_parameter(parser);
} else if (accept1(parser, PM_TOKEN_USTAR)) {
pm_token_t star = parser->previous;
pm_node_t *value = NULL;
if (accept1(parser, PM_TOKEN_IDENTIFIER)) {
pm_token_t name = parser->previous;
value = (pm_node_t *) pm_required_parameter_node_create(parser, &name);
pm_parser_parameter_name_check(parser, &name);
pm_parser_local_add_token(parser, &name);
}
param = (pm_node_t *) pm_splat_node_create(parser, &star, value);
} else {
expect1(parser, PM_TOKEN_IDENTIFIER, PM_ERR_EXPECT_IDENT_REQ_PARAMETER);
pm_token_t name = parser->previous;
param = (pm_node_t *) pm_required_parameter_node_create(parser, &name);
pm_parser_parameter_name_check(parser, &name);
pm_parser_local_add_token(parser, &name);
}
pm_multi_target_node_targets_append(parser, node, param);
} while (accept1(parser, PM_TOKEN_COMMA));
expect1(parser, PM_TOKEN_PARENTHESIS_RIGHT, PM_ERR_EXPECT_RPAREN_REQ_PARAMETER);
pm_multi_target_node_closing_set(node, &parser->previous);
return node;
}
/**
* This represents the different order states we can be in when parsing
* method parameters.
*/
typedef enum {
PM_PARAMETERS_NO_CHANGE = 0, // Extra state for tokens that should not change the state
PM_PARAMETERS_ORDER_NOTHING_AFTER = 1,
PM_PARAMETERS_ORDER_KEYWORDS_REST,
PM_PARAMETERS_ORDER_KEYWORDS,
PM_PARAMETERS_ORDER_REST,
PM_PARAMETERS_ORDER_AFTER_OPTIONAL,
PM_PARAMETERS_ORDER_OPTIONAL,
PM_PARAMETERS_ORDER_NAMED,
PM_PARAMETERS_ORDER_NONE,
} pm_parameters_order_t;
/**
* This matches parameters tokens with parameters state.
*/
static pm_parameters_order_t parameters_ordering[PM_TOKEN_MAXIMUM] = {
[0] = PM_PARAMETERS_NO_CHANGE,
[PM_TOKEN_UAMPERSAND] = PM_PARAMETERS_ORDER_NOTHING_AFTER,
[PM_TOKEN_AMPERSAND] = PM_PARAMETERS_ORDER_NOTHING_AFTER,
[PM_TOKEN_UDOT_DOT_DOT] = PM_PARAMETERS_ORDER_NOTHING_AFTER,
[PM_TOKEN_IDENTIFIER] = PM_PARAMETERS_ORDER_NAMED,
[PM_TOKEN_PARENTHESIS_LEFT] = PM_PARAMETERS_ORDER_NAMED,
[PM_TOKEN_EQUAL] = PM_PARAMETERS_ORDER_OPTIONAL,
[PM_TOKEN_LABEL] = PM_PARAMETERS_ORDER_KEYWORDS,
[PM_TOKEN_USTAR] = PM_PARAMETERS_ORDER_AFTER_OPTIONAL,
[PM_TOKEN_STAR] = PM_PARAMETERS_ORDER_AFTER_OPTIONAL,
[PM_TOKEN_USTAR_STAR] = PM_PARAMETERS_ORDER_KEYWORDS_REST,
[PM_TOKEN_STAR_STAR] = PM_PARAMETERS_ORDER_KEYWORDS_REST
};
/**
* Check if current parameter follows valid parameters ordering. If not it adds
* an error to the list without stopping the parsing, otherwise sets the
* parameters state to the one corresponding to the current parameter.
*/
static void
update_parameter_state(pm_parser_t *parser, pm_token_t *token, pm_parameters_order_t *current) {
pm_parameters_order_t state = parameters_ordering[token->type];
if (state == PM_PARAMETERS_NO_CHANGE) return;
// If we see another ordered argument after a optional argument
// we only continue parsing ordered arguments until we stop seeing ordered arguments
if (*current == PM_PARAMETERS_ORDER_OPTIONAL && state == PM_PARAMETERS_ORDER_NAMED) {
*current = PM_PARAMETERS_ORDER_AFTER_OPTIONAL;
return;
} else if (*current == PM_PARAMETERS_ORDER_AFTER_OPTIONAL && state == PM_PARAMETERS_ORDER_NAMED) {
return;
}
if (token->type == PM_TOKEN_USTAR && *current == PM_PARAMETERS_ORDER_AFTER_OPTIONAL) {
pm_parser_err_token(parser, token, PM_ERR_PARAMETER_STAR);
}
if (*current == PM_PARAMETERS_ORDER_NOTHING_AFTER || state > *current) {
// We know what transition we failed on, so we can provide a better error here.
pm_parser_err_token(parser, token, PM_ERR_PARAMETER_ORDER);
} else if (state < *current) {
*current = state;
}
}
/**
* Parse a list of parameters on a method definition.
*/
static pm_parameters_node_t *
parse_parameters(
pm_parser_t *parser,
pm_binding_power_t binding_power,
bool uses_parentheses,
bool allows_trailing_comma,
bool allows_forwarding_parameters
) {
pm_parameters_node_t *params = pm_parameters_node_create(parser);
bool looping = true;
pm_do_loop_stack_push(parser, false);
pm_parameters_order_t order = PM_PARAMETERS_ORDER_NONE;
do {
switch (parser->current.type) {
case PM_TOKEN_PARENTHESIS_LEFT: {
update_parameter_state(parser, &parser->current, &order);
pm_node_t *param = (pm_node_t *) parse_required_destructured_parameter(parser);
if (order > PM_PARAMETERS_ORDER_AFTER_OPTIONAL) {
pm_parameters_node_requireds_append(params, param);
} else {
pm_parameters_node_posts_append(params, param);
}
break;
}
case PM_TOKEN_UAMPERSAND:
case PM_TOKEN_AMPERSAND: {
update_parameter_state(parser, &parser->current, &order);
parser_lex(parser);
pm_token_t operator = parser->previous;
pm_token_t name;
if (accept1(parser, PM_TOKEN_IDENTIFIER)) {
name = parser->previous;
pm_parser_parameter_name_check(parser, &name);
pm_parser_local_add_token(parser, &name);
} else {
name = not_provided(parser);
if (allows_forwarding_parameters) {
pm_parser_local_add_token(parser, &operator);
}
}
pm_block_parameter_node_t *param = pm_block_parameter_node_create(parser, &name, &operator);
if (params->block == NULL) {
pm_parameters_node_block_set(params, param);
} else {
pm_parser_err_node(parser, (pm_node_t *) param, PM_ERR_PARAMETER_BLOCK_MULTI);
pm_parameters_node_posts_append(params, (pm_node_t *) param);
}
break;
}
case PM_TOKEN_UDOT_DOT_DOT: {
if (!allows_forwarding_parameters) {
pm_parser_err_current(parser, PM_ERR_ARGUMENT_NO_FORWARDING_ELLIPSES);
}
if (order > PM_PARAMETERS_ORDER_NOTHING_AFTER) {
update_parameter_state(parser, &parser->current, &order);
parser_lex(parser);
if (allows_forwarding_parameters) {
pm_parser_local_add_token(parser, &parser->previous);
}
pm_forwarding_parameter_node_t *param = pm_forwarding_parameter_node_create(parser, &parser->previous);
if (params->keyword_rest != NULL) {
// If we already have a keyword rest parameter, then we replace it with the
// forwarding parameter and move the keyword rest parameter to the posts list.
pm_node_t *keyword_rest = params->keyword_rest;
pm_parameters_node_posts_append(params, keyword_rest);
pm_parser_err_previous(parser, PM_ERR_PARAMETER_UNEXPECTED_FWD);
params->keyword_rest = NULL;
}
pm_parameters_node_keyword_rest_set(params, (pm_node_t *)param);
} else {
update_parameter_state(parser, &parser->current, &order);
parser_lex(parser);
}
break;
}
case PM_TOKEN_CLASS_VARIABLE:
case PM_TOKEN_IDENTIFIER:
case PM_TOKEN_CONSTANT:
case PM_TOKEN_INSTANCE_VARIABLE:
case PM_TOKEN_GLOBAL_VARIABLE:
case PM_TOKEN_METHOD_NAME: {
parser_lex(parser);
switch (parser->previous.type) {
case PM_TOKEN_CONSTANT:
pm_parser_err_previous(parser, PM_ERR_ARGUMENT_FORMAL_CONSTANT);
break;
case PM_TOKEN_INSTANCE_VARIABLE:
pm_parser_err_previous(parser, PM_ERR_ARGUMENT_FORMAL_IVAR);
break;
case PM_TOKEN_GLOBAL_VARIABLE:
pm_parser_err_previous(parser, PM_ERR_ARGUMENT_FORMAL_GLOBAL);
break;
case PM_TOKEN_CLASS_VARIABLE:
pm_parser_err_previous(parser, PM_ERR_ARGUMENT_FORMAL_CLASS);
break;
case PM_TOKEN_METHOD_NAME:
pm_parser_err_previous(parser, PM_ERR_PARAMETER_METHOD_NAME);
break;
default: break;
}
if (parser->current.type == PM_TOKEN_EQUAL) {
update_parameter_state(parser, &parser->current, &order);
} else {
update_parameter_state(parser, &parser->previous, &order);
}
pm_token_t name = parser->previous;
pm_parser_parameter_name_check(parser, &name);
pm_parser_local_add_token(parser, &name);
if (accept1(parser, PM_TOKEN_EQUAL)) {
pm_token_t operator = parser->previous;
context_push(parser, PM_CONTEXT_DEFAULT_PARAMS);
pm_constant_id_t old_param_name = parser->current_param_name;
parser->current_param_name = pm_parser_constant_id_token(parser, &name);
pm_node_t *value = parse_value_expression(parser, binding_power, false, PM_ERR_PARAMETER_NO_DEFAULT);
pm_optional_parameter_node_t *param = pm_optional_parameter_node_create(parser, &name, &operator, value);
pm_parameters_node_optionals_append(params, param);
parser->current_param_name = old_param_name;
context_pop(parser);
// If parsing the value of the parameter resulted in error recovery,
// then we can put a missing node in its place and stop parsing the
// parameters entirely now.
if (parser->recovering) {
looping = false;
break;
}
} else if (order > PM_PARAMETERS_ORDER_AFTER_OPTIONAL) {
pm_required_parameter_node_t *param = pm_required_parameter_node_create(parser, &name);
pm_parameters_node_requireds_append(params, (pm_node_t *) param);
} else {
pm_required_parameter_node_t *param = pm_required_parameter_node_create(parser, &name);
pm_parameters_node_posts_append(params, (pm_node_t *) param);
}
break;
}
case PM_TOKEN_LABEL: {
if (!uses_parentheses) parser->in_keyword_arg = true;
update_parameter_state(parser, &parser->current, &order);
parser_lex(parser);
pm_token_t name = parser->previous;
pm_token_t local = name;
local.end -= 1;
pm_parser_parameter_name_check(parser, &local);
pm_parser_local_add_token(parser, &local);
switch (parser->current.type) {
case PM_TOKEN_COMMA:
case PM_TOKEN_PARENTHESIS_RIGHT:
case PM_TOKEN_PIPE: {
pm_node_t *param = (pm_node_t *) pm_required_keyword_parameter_node_create(parser, &name);
pm_parameters_node_keywords_append(params, param);
break;
}
case PM_TOKEN_SEMICOLON:
case PM_TOKEN_NEWLINE: {
if (uses_parentheses) {
looping = false;
break;
}
pm_node_t *param = (pm_node_t *) pm_required_keyword_parameter_node_create(parser, &name);
pm_parameters_node_keywords_append(params, param);
break;
}
default: {
pm_node_t *param;
if (token_begins_expression_p(parser->current.type)) {
context_push(parser, PM_CONTEXT_DEFAULT_PARAMS);
pm_constant_id_t old_param_name = parser->current_param_name;
parser->current_param_name = pm_parser_constant_id_token(parser, &local);
pm_node_t *value = parse_value_expression(parser, binding_power, false, PM_ERR_PARAMETER_NO_DEFAULT_KW);
parser->current_param_name = old_param_name;
context_pop(parser);
param = (pm_node_t *) pm_optional_keyword_parameter_node_create(parser, &name, value);
}
else {
param = (pm_node_t *) pm_required_keyword_parameter_node_create(parser, &name);
}
pm_parameters_node_keywords_append(params, param);
// If parsing the value of the parameter resulted in error recovery,
// then we can put a missing node in its place and stop parsing the
// parameters entirely now.
if (parser->recovering) {
looping = false;
break;
}
}
}
parser->in_keyword_arg = false;
break;
}
case PM_TOKEN_USTAR:
case PM_TOKEN_STAR: {
update_parameter_state(parser, &parser->current, &order);
parser_lex(parser);
pm_token_t operator = parser->previous;
pm_token_t name;
if (accept1(parser, PM_TOKEN_IDENTIFIER)) {
name = parser->previous;
pm_parser_parameter_name_check(parser, &name);
pm_parser_local_add_token(parser, &name);
} else {
name = not_provided(parser);
if (allows_forwarding_parameters) {
pm_parser_local_add_token(parser, &operator);
}
}
pm_node_t *param = (pm_node_t *) pm_rest_parameter_node_create(parser, &operator, &name);
if (params->rest == NULL) {
pm_parameters_node_rest_set(params, param);
} else {
pm_parser_err_node(parser, param, PM_ERR_PARAMETER_SPLAT_MULTI);
pm_parameters_node_posts_append(params, param);
}
break;
}
case PM_TOKEN_STAR_STAR:
case PM_TOKEN_USTAR_STAR: {
update_parameter_state(parser, &parser->current, &order);
parser_lex(parser);
pm_token_t operator = parser->previous;
pm_node_t *param;
if (accept1(parser, PM_TOKEN_KEYWORD_NIL)) {
param = (pm_node_t *) pm_no_keywords_parameter_node_create(parser, &operator, &parser->previous);
} else {
pm_token_t name;
if (accept1(parser, PM_TOKEN_IDENTIFIER)) {
name = parser->previous;
pm_parser_parameter_name_check(parser, &name);
pm_parser_local_add_token(parser, &name);
} else {
name = not_provided(parser);
if (allows_forwarding_parameters) {
pm_parser_local_add_token(parser, &operator);
}
}
param = (pm_node_t *) pm_keyword_rest_parameter_node_create(parser, &operator, &name);
}
if (params->keyword_rest == NULL) {
pm_parameters_node_keyword_rest_set(params, param);
} else {
pm_parser_err_node(parser, param, PM_ERR_PARAMETER_ASSOC_SPLAT_MULTI);
pm_parameters_node_posts_append(params, param);
}
break;
}
default:
if (parser->previous.type == PM_TOKEN_COMMA) {
if (allows_trailing_comma) {
// If we get here, then we have a trailing comma in a
// block parameter list.
pm_node_t *param = (pm_node_t *) pm_implicit_rest_node_create(parser, &parser->previous);
if (params->rest == NULL) {
pm_parameters_node_rest_set(params, param);
} else {
pm_parser_err_node(parser, (pm_node_t *) param, PM_ERR_PARAMETER_SPLAT_MULTI);
pm_parameters_node_posts_append(params, (pm_node_t *) param);
}
} else {
pm_parser_err_previous(parser, PM_ERR_PARAMETER_WILD_LOOSE_COMMA);
}
}
looping = false;
break;
}
if (looping && uses_parentheses) {
accept1(parser, PM_TOKEN_NEWLINE);
}
} while (looping && accept1(parser, PM_TOKEN_COMMA));
pm_do_loop_stack_pop(parser);
// If we don't have any parameters, return `NULL` instead of an empty `ParametersNode`.
if (params->base.location.start == params->base.location.end) {
pm_node_destroy(parser, (pm_node_t *) params);
return NULL;
}
return params;
}
/**
* Parse any number of rescue clauses. This will form a linked list of if
* nodes pointing to each other from the top.
*/
static inline void
parse_rescues(pm_parser_t *parser, pm_begin_node_t *parent_node, bool def_p) {
pm_rescue_node_t *current = NULL;
while (accept1(parser, PM_TOKEN_KEYWORD_RESCUE)) {
pm_rescue_node_t *rescue = pm_rescue_node_create(parser, &parser->previous);
switch (parser->current.type) {
case PM_TOKEN_EQUAL_GREATER: {
// Here we have an immediate => after the rescue keyword, in which case
// we're going to have an empty list of exceptions to rescue (which
// implies StandardError).
parser_lex(parser);
pm_rescue_node_operator_set(rescue, &parser->previous);
pm_node_t *reference = parse_expression(parser, PM_BINDING_POWER_INDEX, false, PM_ERR_RESCUE_VARIABLE);
reference = parse_target(parser, reference);
pm_rescue_node_reference_set(rescue, reference);
break;
}
case PM_TOKEN_NEWLINE:
case PM_TOKEN_SEMICOLON:
case PM_TOKEN_KEYWORD_THEN:
// Here we have a terminator for the rescue keyword, in which case we're
// going to just continue on.
break;
default: {
if (token_begins_expression_p(parser->current.type) || match1(parser, PM_TOKEN_USTAR)) {
// Here we have something that could be an exception expression, so
// we'll attempt to parse it here and any others delimited by commas.
do {
pm_node_t *expression = parse_starred_expression(parser, PM_BINDING_POWER_DEFINED, false, PM_ERR_RESCUE_EXPRESSION);
pm_rescue_node_exceptions_append(rescue, expression);
// If we hit a newline, then this is the end of the rescue expression. We
// can continue on to parse the statements.
if (match3(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON, PM_TOKEN_KEYWORD_THEN)) break;
// If we hit a `=>` then we're going to parse the exception variable. Once
// we've done that, we'll break out of the loop and parse the statements.
if (accept1(parser, PM_TOKEN_EQUAL_GREATER)) {
pm_rescue_node_operator_set(rescue, &parser->previous);
pm_node_t *reference = parse_expression(parser, PM_BINDING_POWER_INDEX, false, PM_ERR_RESCUE_VARIABLE);
reference = parse_target(parser, reference);
pm_rescue_node_reference_set(rescue, reference);
break;
}
} while (accept1(parser, PM_TOKEN_COMMA));
}
}
}
if (accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON)) {
accept1(parser, PM_TOKEN_KEYWORD_THEN);
} else {
expect1(parser, PM_TOKEN_KEYWORD_THEN, PM_ERR_RESCUE_TERM);
}
if (!match3(parser, PM_TOKEN_KEYWORD_ELSE, PM_TOKEN_KEYWORD_ENSURE, PM_TOKEN_KEYWORD_END)) {
pm_accepts_block_stack_push(parser, true);
pm_statements_node_t *statements = parse_statements(parser, def_p ? PM_CONTEXT_RESCUE_DEF : PM_CONTEXT_RESCUE);
if (statements) {
pm_rescue_node_statements_set(rescue, statements);
}
pm_accepts_block_stack_pop(parser);
accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON);
}
if (current == NULL) {
pm_begin_node_rescue_clause_set(parent_node, rescue);
} else {
pm_rescue_node_consequent_set(current, rescue);
}
current = rescue;
}
// The end node locations on rescue nodes will not be set correctly
// since we won't know the end until we've found all consequent
// clauses. This sets the end location on all rescues once we know it
if (current) {
const uint8_t *end_to_set = current->base.location.end;
current = parent_node->rescue_clause;
while (current) {
current->base.location.end = end_to_set;
current = current->consequent;
}
}
if (accept1(parser, PM_TOKEN_KEYWORD_ELSE)) {
pm_token_t else_keyword = parser->previous;
accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON);
pm_statements_node_t *else_statements = NULL;
if (!match2(parser, PM_TOKEN_KEYWORD_END, PM_TOKEN_KEYWORD_ENSURE)) {
pm_accepts_block_stack_push(parser, true);
else_statements = parse_statements(parser, def_p ? PM_CONTEXT_RESCUE_ELSE_DEF : PM_CONTEXT_RESCUE_ELSE);
pm_accepts_block_stack_pop(parser);
accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON);
}
pm_else_node_t *else_clause = pm_else_node_create(parser, &else_keyword, else_statements, &parser->current);
pm_begin_node_else_clause_set(parent_node, else_clause);
}
if (accept1(parser, PM_TOKEN_KEYWORD_ENSURE)) {
pm_token_t ensure_keyword = parser->previous;
accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON);
pm_statements_node_t *ensure_statements = NULL;
if (!match1(parser, PM_TOKEN_KEYWORD_END)) {
pm_accepts_block_stack_push(parser, true);
ensure_statements = parse_statements(parser, def_p ? PM_CONTEXT_ENSURE_DEF : PM_CONTEXT_ENSURE);
pm_accepts_block_stack_pop(parser);
accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON);
}
pm_ensure_node_t *ensure_clause = pm_ensure_node_create(parser, &ensure_keyword, ensure_statements, &parser->current);
pm_begin_node_ensure_clause_set(parent_node, ensure_clause);
}
if (parser->current.type == PM_TOKEN_KEYWORD_END) {
pm_begin_node_end_keyword_set(parent_node, &parser->current);
} else {
pm_token_t end_keyword = (pm_token_t) { .type = PM_TOKEN_MISSING, .start = parser->previous.end, .end = parser->previous.end };
pm_begin_node_end_keyword_set(parent_node, &end_keyword);
}
}
static inline pm_begin_node_t *
parse_rescues_as_begin(pm_parser_t *parser, pm_statements_node_t *statements, bool def_p) {
pm_token_t no_begin_token = not_provided(parser);
pm_begin_node_t *begin_node = pm_begin_node_create(parser, &no_begin_token, statements);
parse_rescues(parser, begin_node, def_p);
// All nodes within a begin node are optional, so we look
// for the earliest possible node that we can use to set
// the BeginNode's start location
const uint8_t *start = begin_node->base.location.start;
if (begin_node->statements) {
start = begin_node->statements->base.location.start;
} else if (begin_node->rescue_clause) {
start = begin_node->rescue_clause->base.location.start;
} else if (begin_node->else_clause) {
start = begin_node->else_clause->base.location.start;
} else if (begin_node->ensure_clause) {
start = begin_node->ensure_clause->base.location.start;
}
begin_node->base.location.start = start;
return begin_node;
}
/**
* Parse a list of parameters and local on a block definition.
*/
static pm_block_parameters_node_t *
parse_block_parameters(
pm_parser_t *parser,
bool allows_trailing_comma,
const pm_token_t *opening,
bool is_lambda_literal
) {
pm_parameters_node_t *parameters = NULL;
if (!match1(parser, PM_TOKEN_SEMICOLON)) {
parameters = parse_parameters(
parser,
is_lambda_literal ? PM_BINDING_POWER_DEFINED : PM_BINDING_POWER_INDEX,
false,
allows_trailing_comma,
false
);
}
pm_block_parameters_node_t *block_parameters = pm_block_parameters_node_create(parser, parameters, opening);
if ((opening->type != PM_TOKEN_NOT_PROVIDED) && accept1(parser, PM_TOKEN_SEMICOLON)) {
do {
expect1(parser, PM_TOKEN_IDENTIFIER, PM_ERR_BLOCK_PARAM_LOCAL_VARIABLE);
pm_parser_parameter_name_check(parser, &parser->previous);
pm_parser_local_add_token(parser, &parser->previous);
pm_block_local_variable_node_t *local = pm_block_local_variable_node_create(parser, &parser->previous);
pm_block_parameters_node_append_local(block_parameters, local);
} while (accept1(parser, PM_TOKEN_COMMA));
}
return block_parameters;
}
/**
* Parse a block.
*/
static pm_block_node_t *
parse_block(pm_parser_t *parser) {
pm_token_t opening = parser->previous;
accept1(parser, PM_TOKEN_NEWLINE);
pm_accepts_block_stack_push(parser, true);
pm_parser_scope_push(parser, false);
pm_block_parameters_node_t *block_parameters = NULL;
if (accept1(parser, PM_TOKEN_PIPE)) {
parser->current_scope->explicit_params = true;
pm_token_t block_parameters_opening = parser->previous;
if (match1(parser, PM_TOKEN_PIPE)) {
block_parameters = pm_block_parameters_node_create(parser, NULL, &block_parameters_opening);
parser->command_start = true;
parser_lex(parser);
} else {
block_parameters = parse_block_parameters(parser, true, &block_parameters_opening, false);
accept1(parser, PM_TOKEN_NEWLINE);
parser->command_start = true;
expect1(parser, PM_TOKEN_PIPE, PM_ERR_BLOCK_PARAM_PIPE_TERM);
}
pm_block_parameters_node_closing_set(block_parameters, &parser->previous);
}
uint32_t locals_body_index = 0;
if (block_parameters) {
locals_body_index = (uint32_t) parser->current_scope->locals.size;
}
accept1(parser, PM_TOKEN_NEWLINE);
pm_node_t *statements = NULL;
if (opening.type == PM_TOKEN_BRACE_LEFT) {
if (!match1(parser, PM_TOKEN_BRACE_RIGHT)) {
statements = (pm_node_t *) parse_statements(parser, PM_CONTEXT_BLOCK_BRACES);
}
expect1(parser, PM_TOKEN_BRACE_RIGHT, PM_ERR_BLOCK_TERM_BRACE);
} else {
if (!match1(parser, PM_TOKEN_KEYWORD_END)) {
if (!match3(parser, PM_TOKEN_KEYWORD_RESCUE, PM_TOKEN_KEYWORD_ELSE, PM_TOKEN_KEYWORD_ENSURE)) {
pm_accepts_block_stack_push(parser, true);
statements = (pm_node_t *) parse_statements(parser, PM_CONTEXT_BLOCK_KEYWORDS);
pm_accepts_block_stack_pop(parser);
}
if (match2(parser, PM_TOKEN_KEYWORD_RESCUE, PM_TOKEN_KEYWORD_ENSURE)) {
assert(statements == NULL || PM_NODE_TYPE_P(statements, PM_STATEMENTS_NODE));
statements = (pm_node_t *) parse_rescues_as_begin(parser, (pm_statements_node_t *) statements, false);
}
}
expect1(parser, PM_TOKEN_KEYWORD_END, PM_ERR_BLOCK_TERM_END);
}
pm_node_t *parameters = (pm_node_t *) block_parameters;
uint8_t maximum = parser->current_scope->numbered_parameters;
if (parameters == NULL && (maximum > 0)) {
parameters = (pm_node_t *) pm_numbered_parameters_node_create(parser, &(pm_location_t) { .start = opening.start, .end = parser->previous.end }, maximum);
locals_body_index = maximum;
}
pm_constant_id_list_t locals = parser->current_scope->locals;
pm_parser_scope_pop(parser);
pm_accepts_block_stack_pop(parser);
return pm_block_node_create(parser, &locals, locals_body_index, &opening, parameters, statements, &parser->previous);
}
/**
* Parse a list of arguments and their surrounding parentheses if they are
* present. It returns true if it found any pieces of arguments (parentheses,
* arguments, or blocks).
*/
static bool
parse_arguments_list(pm_parser_t *parser, pm_arguments_t *arguments, bool accepts_block, bool accepts_command_call) {
bool found = false;
if (accept1(parser, PM_TOKEN_PARENTHESIS_LEFT)) {
found |= true;
arguments->opening_loc = PM_LOCATION_TOKEN_VALUE(&parser->previous);
if (accept1(parser, PM_TOKEN_PARENTHESIS_RIGHT)) {
arguments->closing_loc = PM_LOCATION_TOKEN_VALUE(&parser->previous);
} else {
pm_accepts_block_stack_push(parser, true);
parse_arguments(parser, arguments, true, PM_TOKEN_PARENTHESIS_RIGHT);
expect1(parser, PM_TOKEN_PARENTHESIS_RIGHT, PM_ERR_ARGUMENT_TERM_PAREN);
pm_accepts_block_stack_pop(parser);
arguments->closing_loc = PM_LOCATION_TOKEN_VALUE(&parser->previous);
}
} else if (accepts_command_call && (token_begins_expression_p(parser->current.type) || match3(parser, PM_TOKEN_USTAR, PM_TOKEN_USTAR_STAR, PM_TOKEN_UAMPERSAND)) && !match1(parser, PM_TOKEN_BRACE_LEFT)) {
found |= true;
pm_accepts_block_stack_push(parser, false);
// If we get here, then the subsequent token cannot be used as an infix
// operator. In this case we assume the subsequent token is part of an
// argument to this method call.
parse_arguments(parser, arguments, true, PM_TOKEN_EOF);
// If we have done with the arguments and still not consumed the comma,
// then we have a trailing comma where we need to check whether it is
// allowed or not.
if (parser->previous.type == PM_TOKEN_COMMA && !match1(parser, PM_TOKEN_SEMICOLON)) {
pm_parser_err_previous(parser, PM_ERR_EXPECT_ARGUMENT);
}
pm_accepts_block_stack_pop(parser);
}
// If we're at the end of the arguments, we can now check if there is a block
// node that starts with a {. If there is, then we can parse it and add it to
// the arguments.
if (accepts_block) {
pm_block_node_t *block = NULL;
if (accept1(parser, PM_TOKEN_BRACE_LEFT)) {
found |= true;
block = parse_block(parser);
pm_arguments_validate_block(parser, arguments, block);
} else if (pm_accepts_block_stack_p(parser) && accept1(parser, PM_TOKEN_KEYWORD_DO)) {
found |= true;
block = parse_block(parser);
}
if (block != NULL) {
if (arguments->block == NULL) {
arguments->block = (pm_node_t *) block;
} else {
pm_parser_err_node(parser, (pm_node_t *) block, PM_ERR_ARGUMENT_BLOCK_MULTI);
if (arguments->arguments == NULL) {
arguments->arguments = pm_arguments_node_create(parser);
}
pm_arguments_node_arguments_append(arguments->arguments, arguments->block);
arguments->block = (pm_node_t *) block;
}
}
}
return found;
}
static inline pm_node_t *
parse_predicate(pm_parser_t *parser, pm_binding_power_t binding_power, pm_context_t context, pm_token_t *then_keyword) {
context_push(parser, PM_CONTEXT_PREDICATE);
pm_diagnostic_id_t error_id = context == PM_CONTEXT_IF ? PM_ERR_CONDITIONAL_IF_PREDICATE : PM_ERR_CONDITIONAL_UNLESS_PREDICATE;
pm_node_t *predicate = parse_value_expression(parser, binding_power, true, error_id);
// Predicates are closed by a term, a "then", or a term and then a "then".
bool predicate_closed = accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON);
if (accept1(parser, PM_TOKEN_KEYWORD_THEN)) {
predicate_closed = true;
*then_keyword = parser->previous;
}
if (!predicate_closed) {
pm_parser_err_current(parser, PM_ERR_CONDITIONAL_PREDICATE_TERM);
}
context_pop(parser);
return predicate;
}
static inline pm_node_t *
parse_conditional(pm_parser_t *parser, pm_context_t context) {
pm_token_t keyword = parser->previous;
pm_token_t then_keyword = not_provided(parser);
pm_node_t *predicate = parse_predicate(parser, PM_BINDING_POWER_MODIFIER, context, &then_keyword);
pm_statements_node_t *statements = NULL;
if (!match3(parser, PM_TOKEN_KEYWORD_ELSIF, PM_TOKEN_KEYWORD_ELSE, PM_TOKEN_KEYWORD_END)) {
pm_accepts_block_stack_push(parser, true);
statements = parse_statements(parser, context);
pm_accepts_block_stack_pop(parser);
accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON);
}
pm_token_t end_keyword = not_provided(parser);
pm_node_t *parent = NULL;
switch (context) {
case PM_CONTEXT_IF:
parent = (pm_node_t *) pm_if_node_create(parser, &keyword, predicate, &then_keyword, statements, NULL, &end_keyword);
break;
case PM_CONTEXT_UNLESS:
parent = (pm_node_t *) pm_unless_node_create(parser, &keyword, predicate, &then_keyword, statements);
break;
default:
assert(false && "unreachable");
break;
}
pm_node_t *current = parent;
// Parse any number of elsif clauses. This will form a linked list of if
// nodes pointing to each other from the top.
if (context == PM_CONTEXT_IF) {
while (accept1(parser, PM_TOKEN_KEYWORD_ELSIF)) {
pm_token_t elsif_keyword = parser->previous;
pm_node_t *predicate = parse_predicate(parser, PM_BINDING_POWER_MODIFIER, PM_CONTEXT_ELSIF, &then_keyword);
pm_accepts_block_stack_push(parser, true);
pm_statements_node_t *statements = parse_statements(parser, PM_CONTEXT_ELSIF);
pm_accepts_block_stack_pop(parser);
accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON);
pm_node_t *elsif = (pm_node_t *) pm_if_node_create(parser, &elsif_keyword, predicate, &then_keyword, statements, NULL, &end_keyword);
((pm_if_node_t *) current)->consequent = elsif;
current = elsif;
}
}
if (match1(parser, PM_TOKEN_KEYWORD_ELSE)) {
parser_lex(parser);
pm_token_t else_keyword = parser->previous;
pm_accepts_block_stack_push(parser, true);
pm_statements_node_t *else_statements = parse_statements(parser, PM_CONTEXT_ELSE);
pm_accepts_block_stack_pop(parser);
accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON);
expect1(parser, PM_TOKEN_KEYWORD_END, PM_ERR_CONDITIONAL_TERM_ELSE);
pm_else_node_t *else_node = pm_else_node_create(parser, &else_keyword, else_statements, &parser->previous);
switch (context) {
case PM_CONTEXT_IF:
((pm_if_node_t *) current)->consequent = (pm_node_t *) else_node;
break;
case PM_CONTEXT_UNLESS:
((pm_unless_node_t *) parent)->consequent = else_node;
break;
default:
assert(false && "unreachable");
break;
}
} else {
// We should specialize this error message to refer to 'if' or 'unless' explicitly.
expect1(parser, PM_TOKEN_KEYWORD_END, PM_ERR_CONDITIONAL_TERM);
}
// Set the appropriate end location for all of the nodes in the subtree.
switch (context) {
case PM_CONTEXT_IF: {
pm_node_t *current = parent;
bool recursing = true;
while (recursing) {
switch (PM_NODE_TYPE(current)) {
case PM_IF_NODE:
pm_if_node_end_keyword_loc_set((pm_if_node_t *) current, &parser->previous);
current = ((pm_if_node_t *) current)->consequent;
recursing = current != NULL;
break;
case PM_ELSE_NODE:
pm_else_node_end_keyword_loc_set((pm_else_node_t *) current, &parser->previous);
recursing = false;
break;
default: {
recursing = false;
break;
}
}
}
break;
}
case PM_CONTEXT_UNLESS:
pm_unless_node_end_keyword_loc_set((pm_unless_node_t *) parent, &parser->previous);
break;
default:
assert(false && "unreachable");
break;
}
return parent;
}
/**
* This macro allows you to define a case statement for all of the keywords.
* It's meant to be used in a switch statement.
*/
#define PM_CASE_KEYWORD PM_TOKEN_KEYWORD___ENCODING__: case PM_TOKEN_KEYWORD___FILE__: case PM_TOKEN_KEYWORD___LINE__: \
case PM_TOKEN_KEYWORD_ALIAS: case PM_TOKEN_KEYWORD_AND: case PM_TOKEN_KEYWORD_BEGIN: case PM_TOKEN_KEYWORD_BEGIN_UPCASE: \
case PM_TOKEN_KEYWORD_BREAK: case PM_TOKEN_KEYWORD_CASE: case PM_TOKEN_KEYWORD_CLASS: case PM_TOKEN_KEYWORD_DEF: \
case PM_TOKEN_KEYWORD_DEFINED: case PM_TOKEN_KEYWORD_DO: case PM_TOKEN_KEYWORD_DO_LOOP: case PM_TOKEN_KEYWORD_ELSE: \
case PM_TOKEN_KEYWORD_ELSIF: case PM_TOKEN_KEYWORD_END: case PM_TOKEN_KEYWORD_END_UPCASE: case PM_TOKEN_KEYWORD_ENSURE: \
case PM_TOKEN_KEYWORD_FALSE: case PM_TOKEN_KEYWORD_FOR: case PM_TOKEN_KEYWORD_IF: case PM_TOKEN_KEYWORD_IN: \
case PM_TOKEN_KEYWORD_MODULE: case PM_TOKEN_KEYWORD_NEXT: case PM_TOKEN_KEYWORD_NIL: case PM_TOKEN_KEYWORD_NOT: \
case PM_TOKEN_KEYWORD_OR: case PM_TOKEN_KEYWORD_REDO: case PM_TOKEN_KEYWORD_RESCUE: case PM_TOKEN_KEYWORD_RETRY: \
case PM_TOKEN_KEYWORD_RETURN: case PM_TOKEN_KEYWORD_SELF: case PM_TOKEN_KEYWORD_SUPER: case PM_TOKEN_KEYWORD_THEN: \
case PM_TOKEN_KEYWORD_TRUE: case PM_TOKEN_KEYWORD_UNDEF: case PM_TOKEN_KEYWORD_UNLESS: case PM_TOKEN_KEYWORD_UNTIL: \
case PM_TOKEN_KEYWORD_WHEN: case PM_TOKEN_KEYWORD_WHILE: case PM_TOKEN_KEYWORD_YIELD
/**
* This macro allows you to define a case statement for all of the operators.
* It's meant to be used in a switch statement.
*/
#define PM_CASE_OPERATOR PM_TOKEN_AMPERSAND: case PM_TOKEN_BACKTICK: case PM_TOKEN_BANG_EQUAL: \
case PM_TOKEN_BANG_TILDE: case PM_TOKEN_BANG: case PM_TOKEN_BRACKET_LEFT_RIGHT_EQUAL: \
case PM_TOKEN_BRACKET_LEFT_RIGHT: case PM_TOKEN_CARET: case PM_TOKEN_EQUAL_EQUAL_EQUAL: case PM_TOKEN_EQUAL_EQUAL: \
case PM_TOKEN_EQUAL_TILDE: case PM_TOKEN_GREATER_EQUAL: case PM_TOKEN_GREATER_GREATER: case PM_TOKEN_GREATER: \
case PM_TOKEN_LESS_EQUAL_GREATER: case PM_TOKEN_LESS_EQUAL: case PM_TOKEN_LESS_LESS: case PM_TOKEN_LESS: \
case PM_TOKEN_MINUS: case PM_TOKEN_PERCENT: case PM_TOKEN_PIPE: case PM_TOKEN_PLUS: case PM_TOKEN_SLASH: \
case PM_TOKEN_STAR_STAR: case PM_TOKEN_STAR: case PM_TOKEN_TILDE: case PM_TOKEN_UAMPERSAND: case PM_TOKEN_UMINUS: \
case PM_TOKEN_UMINUS_NUM: case PM_TOKEN_UPLUS: case PM_TOKEN_USTAR: case PM_TOKEN_USTAR_STAR
/**
* This macro allows you to define a case statement for all of the token types
* that represent the beginning of nodes that are "primitives" in a pattern
* matching expression.
*/
#define PM_CASE_PRIMITIVE PM_TOKEN_INTEGER: case PM_TOKEN_INTEGER_IMAGINARY: case PM_TOKEN_INTEGER_RATIONAL: \
case PM_TOKEN_INTEGER_RATIONAL_IMAGINARY: case PM_TOKEN_FLOAT: case PM_TOKEN_FLOAT_IMAGINARY: \
case PM_TOKEN_FLOAT_RATIONAL: case PM_TOKEN_FLOAT_RATIONAL_IMAGINARY: case PM_TOKEN_SYMBOL_BEGIN: \
case PM_TOKEN_REGEXP_BEGIN: case PM_TOKEN_BACKTICK: case PM_TOKEN_PERCENT_LOWER_X: case PM_TOKEN_PERCENT_LOWER_I: \
case PM_TOKEN_PERCENT_LOWER_W: case PM_TOKEN_PERCENT_UPPER_I: case PM_TOKEN_PERCENT_UPPER_W: \
case PM_TOKEN_STRING_BEGIN: case PM_TOKEN_KEYWORD_NIL: case PM_TOKEN_KEYWORD_SELF: case PM_TOKEN_KEYWORD_TRUE: \
case PM_TOKEN_KEYWORD_FALSE: case PM_TOKEN_KEYWORD___FILE__: case PM_TOKEN_KEYWORD___LINE__: \
case PM_TOKEN_KEYWORD___ENCODING__: case PM_TOKEN_MINUS_GREATER: case PM_TOKEN_HEREDOC_START: \
case PM_TOKEN_UMINUS_NUM: case PM_TOKEN_CHARACTER_LITERAL
/**
* This macro allows you to define a case statement for all of the token types
* that could begin a parameter.
*/
#define PM_CASE_PARAMETER PM_TOKEN_UAMPERSAND: case PM_TOKEN_AMPERSAND: case PM_TOKEN_UDOT_DOT_DOT: \
case PM_TOKEN_IDENTIFIER: case PM_TOKEN_LABEL: case PM_TOKEN_USTAR: case PM_TOKEN_STAR: case PM_TOKEN_STAR_STAR: \
case PM_TOKEN_USTAR_STAR: case PM_TOKEN_CONSTANT: case PM_TOKEN_INSTANCE_VARIABLE: case PM_TOKEN_GLOBAL_VARIABLE: \
case PM_TOKEN_CLASS_VARIABLE
/**
* This macro allows you to define a case statement for all of the nodes that
* can be transformed into write targets.
*/
#define PM_CASE_WRITABLE PM_CLASS_VARIABLE_READ_NODE: case PM_CONSTANT_PATH_NODE: \
case PM_CONSTANT_READ_NODE: case PM_GLOBAL_VARIABLE_READ_NODE: case PM_LOCAL_VARIABLE_READ_NODE: \
case PM_INSTANCE_VARIABLE_READ_NODE: case PM_MULTI_TARGET_NODE: case PM_BACK_REFERENCE_READ_NODE: \
case PM_NUMBERED_REFERENCE_READ_NODE
// Assert here that the flags are the same so that we can safely switch the type
// of the node without having to move the flags.
PM_STATIC_ASSERT(__LINE__, ((int) PM_STRING_FLAGS_FORCED_UTF8_ENCODING) == ((int) PM_ENCODING_FLAGS_FORCED_UTF8_ENCODING), "Expected the flags to match.");
/**
* If the encoding was explicitly set through the lexing process, then we need
* to potentially mark the string's flags to indicate how to encode it.
*/
static inline pm_node_flags_t
parse_unescaped_encoding(const pm_parser_t *parser) {
if (parser->explicit_encoding != NULL) {
if (parser->explicit_encoding == PM_ENCODING_UTF_8_ENTRY) {
return PM_STRING_FLAGS_FORCED_UTF8_ENCODING;
} else if (parser->encoding == PM_ENCODING_US_ASCII_ENTRY) {
return PM_STRING_FLAGS_FORCED_BINARY_ENCODING;
}
}
return 0;
}
/**
* Parse a node that is part of a string. If the subsequent tokens cannot be
* parsed as a string part, then NULL is returned.
*/
static pm_node_t *
parse_string_part(pm_parser_t *parser) {
switch (parser->current.type) {
// Here the lexer has returned to us plain string content. In this case
// we'll create a string node that has no opening or closing and return that
// as the part. These kinds of parts look like:
//
// "aaa #{bbb} #@ccc ddd"
// ^^^^ ^ ^^^^
case PM_TOKEN_STRING_CONTENT: {
pm_token_t opening = not_provided(parser);
pm_token_t closing = not_provided(parser);
pm_node_t *node = (pm_node_t *) pm_string_node_create_current_string(parser, &opening, &parser->current, &closing);
pm_node_flag_set(node, parse_unescaped_encoding(parser));
parser_lex(parser);
return node;
}
// Here the lexer has returned the beginning of an embedded expression. In
// that case we'll parse the inner statements and return that as the part.
// These kinds of parts look like:
//
// "aaa #{bbb} #@ccc ddd"
// ^^^^^^
case PM_TOKEN_EMBEXPR_BEGIN: {
pm_lex_state_t state = parser->lex_state;
int brace_nesting = parser->brace_nesting;
parser->brace_nesting = 0;
lex_state_set(parser, PM_LEX_STATE_BEG);
parser_lex(parser);
pm_token_t opening = parser->previous;
pm_statements_node_t *statements = NULL;
if (!match1(parser, PM_TOKEN_EMBEXPR_END)) {
pm_accepts_block_stack_push(parser, true);
statements = parse_statements(parser, PM_CONTEXT_EMBEXPR);
pm_accepts_block_stack_pop(parser);
}
parser->brace_nesting = brace_nesting;
lex_state_set(parser, state);
expect1(parser, PM_TOKEN_EMBEXPR_END, PM_ERR_EMBEXPR_END);
pm_token_t closing = parser->previous;
return (pm_node_t *) pm_embedded_statements_node_create(parser, &opening, statements, &closing);
}
// Here the lexer has returned the beginning of an embedded variable.
// In that case we'll parse the variable and create an appropriate node
// for it and then return that node. These kinds of parts look like:
//
// "aaa #{bbb} #@ccc ddd"
// ^^^^^
case PM_TOKEN_EMBVAR: {
lex_state_set(parser, PM_LEX_STATE_BEG);
parser_lex(parser);
pm_token_t operator = parser->previous;
pm_node_t *variable;
switch (parser->current.type) {
// In this case a back reference is being interpolated. We'll
// create a global variable read node.
case PM_TOKEN_BACK_REFERENCE:
parser_lex(parser);
variable = (pm_node_t *) pm_back_reference_read_node_create(parser, &parser->previous);
break;
// In this case an nth reference is being interpolated. We'll
// create a global variable read node.
case PM_TOKEN_NUMBERED_REFERENCE:
parser_lex(parser);
variable = (pm_node_t *) pm_numbered_reference_read_node_create(parser, &parser->previous);
break;
// In this case a global variable is being interpolated. We'll
// create a global variable read node.
case PM_TOKEN_GLOBAL_VARIABLE:
parser_lex(parser);
variable = (pm_node_t *) pm_global_variable_read_node_create(parser, &parser->previous);
break;
// In this case an instance variable is being interpolated.
// We'll create an instance variable read node.
case PM_TOKEN_INSTANCE_VARIABLE:
parser_lex(parser);
variable = (pm_node_t *) pm_instance_variable_read_node_create(parser, &parser->previous);
break;
// In this case a class variable is being interpolated. We'll
// create a class variable read node.
case PM_TOKEN_CLASS_VARIABLE:
parser_lex(parser);
variable = (pm_node_t *) pm_class_variable_read_node_create(parser, &parser->previous);
break;
// We can hit here if we got an invalid token. In that case
// we'll not attempt to lex this token and instead just return a
// missing node.
default:
expect1(parser, PM_TOKEN_IDENTIFIER, PM_ERR_EMBVAR_INVALID);
variable = (pm_node_t *) pm_missing_node_create(parser, parser->current.start, parser->current.end);
break;
}
return (pm_node_t *) pm_embedded_variable_node_create(parser, &operator, variable);
}
default:
parser_lex(parser);
pm_parser_err_previous(parser, PM_ERR_CANNOT_PARSE_STRING_PART);
return NULL;
}
}
static pm_node_t *
parse_symbol(pm_parser_t *parser, pm_lex_mode_t *lex_mode, pm_lex_state_t next_state) {
pm_token_t opening = parser->previous;
if (lex_mode->mode != PM_LEX_STRING) {
if (next_state != PM_LEX_STATE_NONE) lex_state_set(parser, next_state);
switch (parser->current.type) {
case PM_TOKEN_IDENTIFIER:
case PM_TOKEN_CONSTANT:
case PM_TOKEN_INSTANCE_VARIABLE:
case PM_TOKEN_METHOD_NAME:
case PM_TOKEN_CLASS_VARIABLE:
case PM_TOKEN_GLOBAL_VARIABLE:
case PM_TOKEN_NUMBERED_REFERENCE:
case PM_TOKEN_BACK_REFERENCE:
case PM_CASE_KEYWORD:
parser_lex(parser);
break;
case PM_CASE_OPERATOR:
lex_state_set(parser, next_state == PM_LEX_STATE_NONE ? PM_LEX_STATE_ENDFN : next_state);
parser_lex(parser);
break;
default:
expect2(parser, PM_TOKEN_IDENTIFIER, PM_TOKEN_METHOD_NAME, PM_ERR_SYMBOL_INVALID);
break;
}
pm_token_t closing = not_provided(parser);
pm_symbol_node_t *symbol = pm_symbol_node_create(parser, &opening, &parser->previous, &closing);
pm_string_shared_init(&symbol->unescaped, parser->previous.start, parser->previous.end);
return (pm_node_t *) symbol;
}
if (lex_mode->as.string.interpolation) {
// If we have the end of the symbol, then we can return an empty symbol.
if (match1(parser, PM_TOKEN_STRING_END)) {
if (next_state != PM_LEX_STATE_NONE) lex_state_set(parser, next_state);
parser_lex(parser);
pm_token_t content = not_provided(parser);
pm_token_t closing = parser->previous;
return (pm_node_t *) pm_symbol_node_create(parser, &opening, &content, &closing);
}
// Now we can parse the first part of the symbol.
pm_node_t *part = parse_string_part(parser);
// If we got a string part, then it's possible that we could transform
// what looks like an interpolated symbol into a regular symbol.
if (part && PM_NODE_TYPE_P(part, PM_STRING_NODE) && match2(parser, PM_TOKEN_STRING_END, PM_TOKEN_EOF)) {
if (next_state != PM_LEX_STATE_NONE) lex_state_set(parser, next_state);
expect1(parser, PM_TOKEN_STRING_END, PM_ERR_SYMBOL_TERM_INTERPOLATED);
return (pm_node_t *) pm_string_node_to_symbol_node(parser, (pm_string_node_t *) part, &opening, &parser->previous);
}
// Create a node_list first. We'll use this to check if it should be an
// InterpolatedSymbolNode or a SymbolNode.
pm_node_list_t node_list = { 0 };
if (part) pm_node_list_append(&node_list, part);
while (!match2(parser, PM_TOKEN_STRING_END, PM_TOKEN_EOF)) {
if ((part = parse_string_part(parser)) != NULL) {
pm_node_list_append(&node_list, part);
}
}
if (next_state != PM_LEX_STATE_NONE) lex_state_set(parser, next_state);
if (match1(parser, PM_TOKEN_EOF)) {
pm_parser_err_token(parser, &opening, PM_ERR_SYMBOL_TERM_INTERPOLATED);
} else {
expect1(parser, PM_TOKEN_STRING_END, PM_ERR_SYMBOL_TERM_INTERPOLATED);
}
return (pm_node_t *) pm_interpolated_symbol_node_create(parser, &opening, &node_list, &parser->previous);
}
pm_token_t content;
pm_string_t unescaped;
if (match1(parser, PM_TOKEN_STRING_CONTENT)) {
content = parser->current;
unescaped = parser->current_string;
parser_lex(parser);
// If we have two string contents in a row, then the content of this
// symbol is split because of heredoc contents. This looks like:
//
// <<A; :'a
// A
// b'
//
// In this case, the best way we have to represent this is as an
// interpolated string node, so that's what we'll do here.
if (match1(parser, PM_TOKEN_STRING_CONTENT)) {
pm_node_list_t parts = { 0 };
pm_token_t bounds = not_provided(parser);
pm_node_t *part = (pm_node_t *) pm_string_node_create_unescaped(parser, &bounds, &content, &bounds, &unescaped);
pm_node_list_append(&parts, part);
part = (pm_node_t *) pm_string_node_create_unescaped(parser, &bounds, &parser->current, &bounds, &parser->current_string);
pm_node_list_append(&parts, part);
if (next_state != PM_LEX_STATE_NONE) {
lex_state_set(parser, next_state);
}
parser_lex(parser);
expect1(parser, PM_TOKEN_STRING_END, PM_ERR_SYMBOL_TERM_DYNAMIC);
return (pm_node_t *) pm_interpolated_symbol_node_create(parser, &opening, &parts, &parser->previous);
}
} else {
content = (pm_token_t) { .type = PM_TOKEN_STRING_CONTENT, .start = parser->previous.end, .end = parser->previous.end };
pm_string_shared_init(&unescaped, content.start, content.end);
}
if (next_state != PM_LEX_STATE_NONE) {
lex_state_set(parser, next_state);
}
if (match1(parser, PM_TOKEN_EOF)) {
pm_parser_err_token(parser, &opening, PM_ERR_SYMBOL_TERM_DYNAMIC);
} else {
expect1(parser, PM_TOKEN_STRING_END, PM_ERR_SYMBOL_TERM_DYNAMIC);
}
return (pm_node_t *) pm_symbol_node_create_unescaped(parser, &opening, &content, &parser->previous, &unescaped);
}
/**
* Parse an argument to undef which can either be a bare word, a symbol, a
* constant, or an interpolated symbol.
*/
static inline pm_node_t *
parse_undef_argument(pm_parser_t *parser) {
switch (parser->current.type) {
case PM_CASE_KEYWORD:
case PM_CASE_OPERATOR:
case PM_TOKEN_CONSTANT:
case PM_TOKEN_IDENTIFIER:
case PM_TOKEN_METHOD_NAME: {
parser_lex(parser);
pm_token_t opening = not_provided(parser);
pm_token_t closing = not_provided(parser);
pm_symbol_node_t *symbol = pm_symbol_node_create(parser, &opening, &parser->previous, &closing);
pm_string_shared_init(&symbol->unescaped, parser->previous.start, parser->previous.end);
return (pm_node_t *) symbol;
}
case PM_TOKEN_SYMBOL_BEGIN: {
pm_lex_mode_t lex_mode = *parser->lex_modes.current;
parser_lex(parser);
return parse_symbol(parser, &lex_mode, PM_LEX_STATE_NONE);
}
default:
pm_parser_err_current(parser, PM_ERR_UNDEF_ARGUMENT);
return (pm_node_t *) pm_missing_node_create(parser, parser->current.start, parser->current.end);
}
}
/**
* Parse an argument to alias which can either be a bare word, a symbol, an
* interpolated symbol or a global variable. If this is the first argument, then
* we need to set the lex state to PM_LEX_STATE_FNAME | PM_LEX_STATE_FITEM
* between the first and second arguments.
*/
static inline pm_node_t *
parse_alias_argument(pm_parser_t *parser, bool first) {
switch (parser->current.type) {
case PM_CASE_OPERATOR:
case PM_CASE_KEYWORD:
case PM_TOKEN_CONSTANT:
case PM_TOKEN_IDENTIFIER:
case PM_TOKEN_METHOD_NAME: {
if (first) {
lex_state_set(parser, PM_LEX_STATE_FNAME | PM_LEX_STATE_FITEM);
}
parser_lex(parser);
pm_token_t opening = not_provided(parser);
pm_token_t closing = not_provided(parser);
pm_symbol_node_t *symbol = pm_symbol_node_create(parser, &opening, &parser->previous, &closing);
pm_string_shared_init(&symbol->unescaped, parser->previous.start, parser->previous.end);
return (pm_node_t *) symbol;
}
case PM_TOKEN_SYMBOL_BEGIN: {
pm_lex_mode_t lex_mode = *parser->lex_modes.current;
parser_lex(parser);
return parse_symbol(parser, &lex_mode, first ? PM_LEX_STATE_FNAME | PM_LEX_STATE_FITEM : PM_LEX_STATE_NONE);
}
case PM_TOKEN_BACK_REFERENCE:
parser_lex(parser);
return (pm_node_t *) pm_back_reference_read_node_create(parser, &parser->previous);
case PM_TOKEN_NUMBERED_REFERENCE:
parser_lex(parser);
return (pm_node_t *) pm_numbered_reference_read_node_create(parser, &parser->previous);
case PM_TOKEN_GLOBAL_VARIABLE:
parser_lex(parser);
return (pm_node_t *) pm_global_variable_read_node_create(parser, &parser->previous);
default:
pm_parser_err_current(parser, PM_ERR_ALIAS_ARGUMENT);
return (pm_node_t *) pm_missing_node_create(parser, parser->current.start, parser->current.end);
}
}
/**
* Return true if any of the visible scopes to the current context are using
* numbered parameters.
*/
static bool
outer_scope_using_numbered_parameters_p(pm_parser_t *parser) {
for (pm_scope_t *scope = parser->current_scope->previous; scope != NULL && !scope->closed; scope = scope->previous) {
if (scope->numbered_parameters) return true;
}
return false;
}
/**
* Parse an identifier into either a local variable read or a call.
*/
static pm_node_t *
parse_variable_call(pm_parser_t *parser) {
pm_node_flags_t flags = 0;
if (!match1(parser, PM_TOKEN_PARENTHESIS_LEFT) && (parser->previous.end[-1] != '!') && (parser->previous.end[-1] != '?')) {
int depth;
if ((depth = pm_parser_local_depth(parser, &parser->previous)) != -1) {
return (pm_node_t *) pm_local_variable_read_node_create(parser, &parser->previous, (uint32_t) depth);
}
if (!parser->current_scope->closed && pm_token_is_numbered_parameter(parser->previous.start, parser->previous.end)) {
// Now that we know we have a numbered parameter, we need to check
// if it's allowed in this context. If it is, then we will create a
// local variable read. If it's not, then we'll create a normal call
// node but add an error.
if (parser->current_scope->explicit_params) {
pm_parser_err_previous(parser, PM_ERR_NUMBERED_PARAMETER_NOT_ALLOWED);
} else if (outer_scope_using_numbered_parameters_p(parser)) {
pm_parser_err_previous(parser, PM_ERR_NUMBERED_PARAMETER_OUTER_SCOPE);
} else {
// Indicate that this scope is using numbered params so that child
// scopes cannot.
uint8_t number = parser->previous.start[1];
// We subtract the value for the character '0' to get the actual
// integer value of the number (only _1 through _9 are valid)
uint8_t numbered_parameters = (uint8_t) (number - '0');
if (numbered_parameters > parser->current_scope->numbered_parameters) {
parser->current_scope->numbered_parameters = numbered_parameters;
pm_parser_numbered_parameters_set(parser, numbered_parameters);
}
// When you use a numbered parameter, it implies the existence
// of all of the locals that exist before it. For example,
// referencing _2 means that _1 must exist. Therefore here we
// loop through all of the possibilities and add them into the
// constant pool.
uint8_t current = '1';
uint8_t *value;
while (current < number) {
value = malloc(2);
value[0] = '_';
value[1] = current++;
pm_parser_local_add_owned(parser, value, 2);
}
// Now we can add the actual token that is being used. For
// this one we can add a shared version since it is directly
// referenced in the source.
pm_parser_local_add_token(parser, &parser->previous);
return (pm_node_t *) pm_local_variable_read_node_create(parser, &parser->previous, 0);
}
}
flags |= PM_CALL_NODE_FLAGS_VARIABLE_CALL;
}
pm_call_node_t *node = pm_call_node_variable_call_create(parser, &parser->previous);
pm_node_flag_set((pm_node_t *)node, flags);
return (pm_node_t *) node;
}
/**
* Parse the method definition name based on the current token available on the
* parser. If it does not match a valid method definition name, then a missing
* token is returned.
*/
static inline pm_token_t
parse_method_definition_name(pm_parser_t *parser) {
switch (parser->current.type) {
case PM_CASE_KEYWORD:
case PM_TOKEN_CONSTANT:
case PM_TOKEN_METHOD_NAME:
parser_lex(parser);
return parser->previous;
case PM_TOKEN_IDENTIFIER:
pm_refute_numbered_parameter(parser, parser->current.start, parser->current.end);
parser_lex(parser);
return parser->previous;
case PM_CASE_OPERATOR:
lex_state_set(parser, PM_LEX_STATE_ENDFN);
parser_lex(parser);
return parser->previous;
default:
return (pm_token_t) { .type = PM_TOKEN_MISSING, .start = parser->current.start, .end = parser->current.end };
}
}
static void
parse_heredoc_dedent_string(pm_string_t *string, size_t common_whitespace) {
// Get a reference to the string struct that is being held by the string
// node. This is the value we're going to actually manipulate.
pm_string_ensure_owned(string);
// Now get the bounds of the existing string. We'll use this as a
// destination to move bytes into. We'll also use it for bounds checking
// since we don't require that these strings be null terminated.
size_t dest_length = pm_string_length(string);
const uint8_t *source_cursor = (uint8_t *) string->source;
const uint8_t *source_end = source_cursor + dest_length;
// We're going to move bytes backward in the string when we get leading
// whitespace, so we'll maintain a pointer to the current position in the
// string that we're writing to.
size_t trimmed_whitespace = 0;
// While we haven't reached the amount of common whitespace that we need to
// trim and we haven't reached the end of the string, we'll keep trimming
// whitespace. Trimming in this context means skipping over these bytes such
// that they aren't copied into the new string.
while ((source_cursor < source_end) && pm_char_is_inline_whitespace(*source_cursor) && trimmed_whitespace < common_whitespace) {
if (*source_cursor == '\t') {
trimmed_whitespace = (trimmed_whitespace / PM_TAB_WHITESPACE_SIZE + 1) * PM_TAB_WHITESPACE_SIZE;
if (trimmed_whitespace > common_whitespace) break;
} else {
trimmed_whitespace++;
}
source_cursor++;
dest_length--;
}
memmove((uint8_t *) string->source, source_cursor, (size_t) (source_end - source_cursor));
string->length = dest_length;
}
/**
* Take a heredoc node that is indented by a ~ and trim the leading whitespace.
*/
static void
parse_heredoc_dedent(pm_parser_t *parser, pm_node_list_t *nodes, size_t common_whitespace) {
// The next node should be dedented if it's the first node in the list or if
// if follows a string node.
bool dedent_next = true;
// Iterate over all nodes, and trim whitespace accordingly. We're going to
// keep around two indices: a read and a write. If we end up trimming all of
// the whitespace from a node, then we'll drop it from the list entirely.
size_t write_index = 0;
for (size_t read_index = 0; read_index < nodes->size; read_index++) {
pm_node_t *node = nodes->nodes[read_index];
// We're not manipulating child nodes that aren't strings. In this case
// we'll skip past it and indicate that the subsequent node should not
// be dedented.
if (!PM_NODE_TYPE_P(node, PM_STRING_NODE)) {
nodes->nodes[write_index++] = node;
dedent_next = false;
continue;
}
pm_string_node_t *string_node = ((pm_string_node_t *) node);
if (dedent_next) {
parse_heredoc_dedent_string(&string_node->unescaped, common_whitespace);
}
if (string_node->unescaped.length == 0) {
pm_node_destroy(parser, node);
} else {
nodes->nodes[write_index++] = node;
}
// We always dedent the next node if it follows a string node.
dedent_next = true;
}
nodes->size = write_index;
}
static pm_node_t *
parse_pattern(pm_parser_t *parser, bool top_pattern, pm_diagnostic_id_t diag_id);
/**
* Accept any number of constants joined by :: delimiters.
*/
static pm_node_t *
parse_pattern_constant_path(pm_parser_t *parser, pm_node_t *node) {
// Now, if there are any :: operators that follow, parse them as constant
// path nodes.
while (accept1(parser, PM_TOKEN_COLON_COLON)) {
pm_token_t delimiter = parser->previous;
expect1(parser, PM_TOKEN_CONSTANT, PM_ERR_CONSTANT_PATH_COLON_COLON_CONSTANT);
pm_node_t *child = (pm_node_t *) pm_constant_read_node_create(parser, &parser->previous);
node = (pm_node_t *)pm_constant_path_node_create(parser, node, &delimiter, child);
}
// If there is a [ or ( that follows, then this is part of a larger pattern
// expression. We'll parse the inner pattern here, then modify the returned
// inner pattern with our constant path attached.
if (!match2(parser, PM_TOKEN_BRACKET_LEFT, PM_TOKEN_PARENTHESIS_LEFT)) {
return node;
}
pm_token_t opening;
pm_token_t closing;
pm_node_t *inner = NULL;
if (accept1(parser, PM_TOKEN_BRACKET_LEFT)) {
opening = parser->previous;
accept1(parser, PM_TOKEN_NEWLINE);
if (!accept1(parser, PM_TOKEN_BRACKET_RIGHT)) {
inner = parse_pattern(parser, true, PM_ERR_PATTERN_EXPRESSION_AFTER_BRACKET);
accept1(parser, PM_TOKEN_NEWLINE);
expect1(parser, PM_TOKEN_BRACKET_RIGHT, PM_ERR_PATTERN_TERM_BRACKET);
}
closing = parser->previous;
} else {
parser_lex(parser);
opening = parser->previous;
if (!accept1(parser, PM_TOKEN_PARENTHESIS_RIGHT)) {
inner = parse_pattern(parser, true, PM_ERR_PATTERN_EXPRESSION_AFTER_PAREN);
expect1(parser, PM_TOKEN_PARENTHESIS_RIGHT, PM_ERR_PATTERN_TERM_PAREN);
}
closing = parser->previous;
}
if (!inner) {
// If there was no inner pattern, then we have something like Foo() or
// Foo[]. In that case we'll create an array pattern with no requireds.
return (pm_node_t *) pm_array_pattern_node_constant_create(parser, node, &opening, &closing);
}
// Now that we have the inner pattern, check to see if it's an array, find,
// or hash pattern. If it is, then we'll attach our constant path to it if
// it doesn't already have a constant. If it's not one of those node types
// or it does have a constant, then we'll create an array pattern.
switch (PM_NODE_TYPE(inner)) {
case PM_ARRAY_PATTERN_NODE: {
pm_array_pattern_node_t *pattern_node = (pm_array_pattern_node_t *) inner;
if (pattern_node->constant == NULL && pattern_node->opening_loc.start == NULL) {
pattern_node->base.location.start = node->location.start;
pattern_node->base.location.end = closing.end;
pattern_node->constant = node;
pattern_node->opening_loc = PM_LOCATION_TOKEN_VALUE(&opening);
pattern_node->closing_loc = PM_LOCATION_TOKEN_VALUE(&closing);
return (pm_node_t *) pattern_node;
}
break;
}
case PM_FIND_PATTERN_NODE: {
pm_find_pattern_node_t *pattern_node = (pm_find_pattern_node_t *) inner;
if (pattern_node->constant == NULL && pattern_node->opening_loc.start == NULL) {
pattern_node->base.location.start = node->location.start;
pattern_node->base.location.end = closing.end;
pattern_node->constant = node;
pattern_node->opening_loc = PM_LOCATION_TOKEN_VALUE(&opening);
pattern_node->closing_loc = PM_LOCATION_TOKEN_VALUE(&closing);
return (pm_node_t *) pattern_node;
}
break;
}
case PM_HASH_PATTERN_NODE: {
pm_hash_pattern_node_t *pattern_node = (pm_hash_pattern_node_t *) inner;
if (pattern_node->constant == NULL && pattern_node->opening_loc.start == NULL) {
pattern_node->base.location.start = node->location.start;
pattern_node->base.location.end = closing.end;
pattern_node->constant = node;
pattern_node->opening_loc = PM_LOCATION_TOKEN_VALUE(&opening);
pattern_node->closing_loc = PM_LOCATION_TOKEN_VALUE(&closing);
return (pm_node_t *) pattern_node;
}
break;
}
default:
break;
}
// If we got here, then we didn't return one of the inner patterns by
// attaching its constant. In this case we'll create an array pattern and
// attach our constant to it.
pm_array_pattern_node_t *pattern_node = pm_array_pattern_node_constant_create(parser, node, &opening, &closing);
pm_array_pattern_node_requireds_append(pattern_node, inner);
return (pm_node_t *) pattern_node;
}
/**
* Parse a rest pattern.
*/
static pm_splat_node_t *
parse_pattern_rest(pm_parser_t *parser) {
assert(parser->previous.type == PM_TOKEN_USTAR);
pm_token_t operator = parser->previous;
pm_node_t *name = NULL;
// Rest patterns don't necessarily have a name associated with them. So we
// will check for that here. If they do, then we'll add it to the local table
// since this pattern will cause it to become a local variable.
if (accept1(parser, PM_TOKEN_IDENTIFIER)) {
pm_token_t identifier = parser->previous;
pm_parser_local_add_token(parser, &identifier);
name = (pm_node_t *) pm_local_variable_target_node_create(parser, &identifier);
}
// Finally we can return the created node.
return pm_splat_node_create(parser, &operator, name);
}
/**
* Parse a keyword rest node.
*/
static pm_node_t *
parse_pattern_keyword_rest(pm_parser_t *parser) {
assert(parser->current.type == PM_TOKEN_USTAR_STAR);
parser_lex(parser);
pm_token_t operator = parser->previous;
pm_node_t *value = NULL;
if (accept1(parser, PM_TOKEN_KEYWORD_NIL)) {
return (pm_node_t *) pm_no_keywords_parameter_node_create(parser, &operator, &parser->previous);
}
if (accept1(parser, PM_TOKEN_IDENTIFIER)) {
pm_parser_local_add_token(parser, &parser->previous);
value = (pm_node_t *) pm_local_variable_target_node_create(parser, &parser->previous);
}
return (pm_node_t *) pm_assoc_splat_node_create(parser, value, &operator);
}
/**
* Parse a hash pattern.
*/
static pm_hash_pattern_node_t *
parse_pattern_hash(pm_parser_t *parser, pm_node_t *first_assoc) {
pm_node_list_t assocs = { 0 };
pm_node_t *rest = NULL;
switch (PM_NODE_TYPE(first_assoc)) {
case PM_ASSOC_NODE: {
if (!match7(parser, PM_TOKEN_COMMA, PM_TOKEN_KEYWORD_THEN, PM_TOKEN_BRACE_RIGHT, PM_TOKEN_BRACKET_RIGHT, PM_TOKEN_PARENTHESIS_RIGHT, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON)) {
// Here we have a value for the first assoc in the list, so we will
// parse it now and update the first assoc.
pm_node_t *value = parse_pattern(parser, false, PM_ERR_PATTERN_EXPRESSION_AFTER_KEY);
pm_assoc_node_t *assoc = (pm_assoc_node_t *) first_assoc;
assoc->base.location.end = value->location.end;
assoc->value = value;
} else {
pm_node_t *key = ((pm_assoc_node_t *) first_assoc)->key;
if (PM_NODE_TYPE_P(key, PM_SYMBOL_NODE)) {
const pm_location_t *value_loc = &((pm_symbol_node_t *) key)->value_loc;
pm_parser_local_add_location(parser, value_loc->start, value_loc->end);
}
}
pm_node_list_append(&assocs, first_assoc);
break;
}
case PM_ASSOC_SPLAT_NODE:
case PM_NO_KEYWORDS_PARAMETER_NODE:
rest = first_assoc;
break;
default:
assert(false);
break;
}
// If there are any other assocs, then we'll parse them now.
while (accept1(parser, PM_TOKEN_COMMA)) {
// Here we need to break to support trailing commas.
if (match6(parser, PM_TOKEN_KEYWORD_THEN, PM_TOKEN_BRACE_RIGHT, PM_TOKEN_BRACKET_RIGHT, PM_TOKEN_PARENTHESIS_RIGHT, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON)) {
break;
}
if (match1(parser, PM_TOKEN_USTAR_STAR)) {
pm_node_t *assoc = parse_pattern_keyword_rest(parser);
if (rest == NULL) {
rest = assoc;
} else {
pm_parser_err_node(parser, assoc, PM_ERR_PATTERN_EXPRESSION_AFTER_REST);
pm_node_list_append(&assocs, assoc);
}
} else {
expect1(parser, PM_TOKEN_LABEL, PM_ERR_PATTERN_LABEL_AFTER_COMMA);
pm_node_t *key = (pm_node_t *) pm_symbol_node_label_create(parser, &parser->previous);
pm_node_t *value = NULL;
if (!match7(parser, PM_TOKEN_COMMA, PM_TOKEN_KEYWORD_THEN, PM_TOKEN_BRACE_RIGHT, PM_TOKEN_BRACKET_RIGHT, PM_TOKEN_PARENTHESIS_RIGHT, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON)) {
value = parse_pattern(parser, false, PM_ERR_PATTERN_EXPRESSION_AFTER_KEY);
} else {
const pm_location_t *value_loc = &((pm_symbol_node_t *) key)->value_loc;
pm_parser_local_add_location(parser, value_loc->start, value_loc->end);
}
pm_token_t operator = not_provided(parser);
pm_node_t *assoc = (pm_node_t *) pm_assoc_node_create(parser, key, &operator, value);
if (rest != NULL) {
pm_parser_err_node(parser, assoc, PM_ERR_PATTERN_EXPRESSION_AFTER_REST);
}
pm_node_list_append(&assocs, assoc);
}
}
pm_hash_pattern_node_t *node = pm_hash_pattern_node_node_list_create(parser, &assocs, rest);
free(assocs.nodes);
return node;
}
/**
* Parse a pattern expression primitive.
*/
static pm_node_t *
parse_pattern_primitive(pm_parser_t *parser, pm_diagnostic_id_t diag_id) {
switch (parser->current.type) {
case PM_TOKEN_IDENTIFIER:
case PM_TOKEN_METHOD_NAME: {
parser_lex(parser);
pm_token_t name = parser->previous;
int depth = pm_parser_local_depth(parser, &name);
if (depth < 0) {
depth = 0;
pm_parser_local_add_token(parser, &name);
}
return (pm_node_t *) pm_local_variable_target_node_create_depth(parser, &name, (uint32_t) depth);
}
case PM_TOKEN_BRACKET_LEFT_ARRAY: {
pm_token_t opening = parser->current;
parser_lex(parser);
if (accept1(parser, PM_TOKEN_BRACKET_RIGHT)) {
// If we have an empty array pattern, then we'll just return a new
// array pattern node.
return (pm_node_t *)pm_array_pattern_node_empty_create(parser, &opening, &parser->previous);
}
// Otherwise, we'll parse the inner pattern, then deal with it depending
// on the type it returns.
pm_node_t *inner = parse_pattern(parser, true, PM_ERR_PATTERN_EXPRESSION_AFTER_BRACKET);
accept1(parser, PM_TOKEN_NEWLINE);
expect1(parser, PM_TOKEN_BRACKET_RIGHT, PM_ERR_PATTERN_TERM_BRACKET);
pm_token_t closing = parser->previous;
switch (PM_NODE_TYPE(inner)) {
case PM_ARRAY_PATTERN_NODE: {
pm_array_pattern_node_t *pattern_node = (pm_array_pattern_node_t *) inner;
if (pattern_node->opening_loc.start == NULL) {
pattern_node->base.location.start = opening.start;
pattern_node->base.location.end = closing.end;
pattern_node->opening_loc = PM_LOCATION_TOKEN_VALUE(&opening);
pattern_node->closing_loc = PM_LOCATION_TOKEN_VALUE(&closing);
return (pm_node_t *) pattern_node;
}
break;
}
case PM_FIND_PATTERN_NODE: {
pm_find_pattern_node_t *pattern_node = (pm_find_pattern_node_t *) inner;
if (pattern_node->opening_loc.start == NULL) {
pattern_node->base.location.start = opening.start;
pattern_node->base.location.end = closing.end;
pattern_node->opening_loc = PM_LOCATION_TOKEN_VALUE(&opening);
pattern_node->closing_loc = PM_LOCATION_TOKEN_VALUE(&closing);
return (pm_node_t *) pattern_node;
}
break;
}
default:
break;
}
pm_array_pattern_node_t *node = pm_array_pattern_node_empty_create(parser, &opening, &closing);
pm_array_pattern_node_requireds_append(node, inner);
return (pm_node_t *) node;
}
case PM_TOKEN_BRACE_LEFT: {
bool previous_pattern_matching_newlines = parser->pattern_matching_newlines;
parser->pattern_matching_newlines = false;
pm_hash_pattern_node_t *node;
pm_token_t opening = parser->current;
parser_lex(parser);
if (accept1(parser, PM_TOKEN_BRACE_RIGHT)) {
// If we have an empty hash pattern, then we'll just return a new hash
// pattern node.
node = pm_hash_pattern_node_empty_create(parser, &opening, &parser->previous);
} else {
pm_node_t *first_assoc;
switch (parser->current.type) {
case PM_TOKEN_LABEL: {
parser_lex(parser);
pm_symbol_node_t *key = pm_symbol_node_label_create(parser, &parser->previous);
pm_token_t operator = not_provided(parser);
first_assoc = (pm_node_t *) pm_assoc_node_create(parser, (pm_node_t *) key, &operator, NULL);
break;
}
case PM_TOKEN_USTAR_STAR:
first_assoc = parse_pattern_keyword_rest(parser);
break;
case PM_TOKEN_STRING_BEGIN: {
pm_node_t *key = parse_expression(parser, PM_BINDING_POWER_MAX, false, PM_ERR_PATTERN_HASH_KEY);
pm_token_t operator = not_provided(parser);
if (!pm_symbol_node_label_p(key)) {
pm_parser_err_node(parser, key, PM_ERR_PATTERN_HASH_KEY_LABEL);
}
first_assoc = (pm_node_t *) pm_assoc_node_create(parser, key, &operator, NULL);
break;
}
default: {
parser_lex(parser);
pm_parser_err_previous(parser, PM_ERR_PATTERN_HASH_KEY);
pm_missing_node_t *key = pm_missing_node_create(parser, parser->previous.start, parser->previous.end);
pm_token_t operator = not_provided(parser);
first_assoc = (pm_node_t *) pm_assoc_node_create(parser, (pm_node_t *) key, &operator, NULL);
break;
}
}
node = parse_pattern_hash(parser, first_assoc);
accept1(parser, PM_TOKEN_NEWLINE);
expect1(parser, PM_TOKEN_BRACE_RIGHT, PM_ERR_PATTERN_TERM_BRACE);
pm_token_t closing = parser->previous;
node->base.location.start = opening.start;
node->base.location.end = closing.end;
node->opening_loc = PM_LOCATION_TOKEN_VALUE(&opening);
node->closing_loc = PM_LOCATION_TOKEN_VALUE(&closing);
}
parser->pattern_matching_newlines = previous_pattern_matching_newlines;
return (pm_node_t *) node;
}
case PM_TOKEN_UDOT_DOT:
case PM_TOKEN_UDOT_DOT_DOT: {
pm_token_t operator = parser->current;
parser_lex(parser);
// Since we have a unary range operator, we need to parse the subsequent
// expression as the right side of the range.
switch (parser->current.type) {
case PM_CASE_PRIMITIVE: {
pm_node_t *right = parse_expression(parser, PM_BINDING_POWER_MAX, false, PM_ERR_PATTERN_EXPRESSION_AFTER_RANGE);
return (pm_node_t *) pm_range_node_create(parser, NULL, &operator, right);
}
default: {
pm_parser_err_token(parser, &operator, PM_ERR_PATTERN_EXPRESSION_AFTER_RANGE);
pm_node_t *right = (pm_node_t *) pm_missing_node_create(parser, operator.start, operator.end);
return (pm_node_t *) pm_range_node_create(parser, NULL, &operator, right);
}
}
}
case PM_CASE_PRIMITIVE: {
pm_node_t *node = parse_expression(parser, PM_BINDING_POWER_MAX, false, diag_id);
// Now that we have a primitive, we need to check if it's part of a range.
if (accept2(parser, PM_TOKEN_DOT_DOT, PM_TOKEN_DOT_DOT_DOT)) {
pm_token_t operator = parser->previous;
// Now that we have the operator, we need to check if this is followed
// by another expression. If it is, then we will create a full range
// node. Otherwise, we'll create an endless range.
switch (parser->current.type) {
case PM_CASE_PRIMITIVE: {
pm_node_t *right = parse_expression(parser, PM_BINDING_POWER_MAX, false, PM_ERR_PATTERN_EXPRESSION_AFTER_RANGE);
return (pm_node_t *) pm_range_node_create(parser, node, &operator, right);
}
default:
return (pm_node_t *) pm_range_node_create(parser, node, &operator, NULL);
}
}
return node;
}
case PM_TOKEN_CARET: {
parser_lex(parser);
pm_token_t operator = parser->previous;
// At this point we have a pin operator. We need to check the subsequent
// expression to determine if it's a variable or an expression.
switch (parser->current.type) {
case PM_TOKEN_IDENTIFIER: {
parser_lex(parser);
pm_node_t *variable = (pm_node_t *) pm_local_variable_read_node_create(parser, &parser->previous, 0);
return (pm_node_t *) pm_pinned_variable_node_create(parser, &operator, variable);
}
case PM_TOKEN_INSTANCE_VARIABLE: {
parser_lex(parser);
pm_node_t *variable = (pm_node_t *) pm_instance_variable_read_node_create(parser, &parser->previous);
return (pm_node_t *) pm_pinned_variable_node_create(parser, &operator, variable);
}
case PM_TOKEN_CLASS_VARIABLE: {
parser_lex(parser);
pm_node_t *variable = (pm_node_t *) pm_class_variable_read_node_create(parser, &parser->previous);
return (pm_node_t *) pm_pinned_variable_node_create(parser, &operator, variable);
}
case PM_TOKEN_GLOBAL_VARIABLE: {
parser_lex(parser);
pm_node_t *variable = (pm_node_t *) pm_global_variable_read_node_create(parser, &parser->previous);
return (pm_node_t *) pm_pinned_variable_node_create(parser, &operator, variable);
}
case PM_TOKEN_NUMBERED_REFERENCE: {
parser_lex(parser);
pm_node_t *variable = (pm_node_t *) pm_numbered_reference_read_node_create(parser, &parser->previous);
return (pm_node_t *) pm_pinned_variable_node_create(parser, &operator, variable);
}
case PM_TOKEN_BACK_REFERENCE: {
parser_lex(parser);
pm_node_t *variable = (pm_node_t *) pm_back_reference_read_node_create(parser, &parser->previous);
return (pm_node_t *) pm_pinned_variable_node_create(parser, &operator, variable);
}
case PM_TOKEN_PARENTHESIS_LEFT: {
bool previous_pattern_matching_newlines = parser->pattern_matching_newlines;
parser->pattern_matching_newlines = false;
pm_token_t lparen = parser->current;
parser_lex(parser);
pm_node_t *expression = parse_value_expression(parser, PM_BINDING_POWER_STATEMENT, true, PM_ERR_PATTERN_EXPRESSION_AFTER_PIN);
parser->pattern_matching_newlines = previous_pattern_matching_newlines;
accept1(parser, PM_TOKEN_NEWLINE);
expect1(parser, PM_TOKEN_PARENTHESIS_RIGHT, PM_ERR_PATTERN_TERM_PAREN);
return (pm_node_t *) pm_pinned_expression_node_create(parser, expression, &operator, &lparen, &parser->previous);
}
default: {
// If we get here, then we have a pin operator followed by something
// not understood. We'll create a missing node and return that.
pm_parser_err_token(parser, &operator, PM_ERR_PATTERN_EXPRESSION_AFTER_PIN);
pm_node_t *variable = (pm_node_t *) pm_missing_node_create(parser, operator.start, operator.end);
return (pm_node_t *) pm_pinned_variable_node_create(parser, &operator, variable);
}
}
}
case PM_TOKEN_UCOLON_COLON: {
pm_token_t delimiter = parser->current;
parser_lex(parser);
expect1(parser, PM_TOKEN_CONSTANT, PM_ERR_CONSTANT_PATH_COLON_COLON_CONSTANT);
pm_node_t *child = (pm_node_t *) pm_constant_read_node_create(parser, &parser->previous);
pm_constant_path_node_t *node = pm_constant_path_node_create(parser, NULL, &delimiter, child);
return parse_pattern_constant_path(parser, (pm_node_t *)node);
}
case PM_TOKEN_CONSTANT: {
pm_token_t constant = parser->current;
parser_lex(parser);
pm_node_t *node = (pm_node_t *) pm_constant_read_node_create(parser, &constant);
return parse_pattern_constant_path(parser, node);
}
default:
pm_parser_err_current(parser, diag_id);
return (pm_node_t *) pm_missing_node_create(parser, parser->current.start, parser->current.end);
}
}
/**
* Parse any number of primitives joined by alternation and ended optionally by
* assignment.
*/
static pm_node_t *
parse_pattern_primitives(pm_parser_t *parser, pm_diagnostic_id_t diag_id) {
pm_node_t *node = NULL;
do {
pm_token_t operator = parser->previous;
switch (parser->current.type) {
case PM_TOKEN_IDENTIFIER:
case PM_TOKEN_BRACKET_LEFT_ARRAY:
case PM_TOKEN_BRACE_LEFT:
case PM_TOKEN_CARET:
case PM_TOKEN_CONSTANT:
case PM_TOKEN_UCOLON_COLON:
case PM_TOKEN_UDOT_DOT:
case PM_TOKEN_UDOT_DOT_DOT:
case PM_CASE_PRIMITIVE: {
if (node == NULL) {
node = parse_pattern_primitive(parser, diag_id);
} else {
pm_node_t *right = parse_pattern_primitive(parser, PM_ERR_PATTERN_EXPRESSION_AFTER_PIPE);
node = (pm_node_t *) pm_alternation_pattern_node_create(parser, node, right, &operator);
}
break;
}
case PM_TOKEN_PARENTHESIS_LEFT: {
parser_lex(parser);
if (node != NULL) {
pm_node_destroy(parser, node);
}
node = parse_pattern(parser, false, PM_ERR_PATTERN_EXPRESSION_AFTER_PAREN);
expect1(parser, PM_TOKEN_PARENTHESIS_RIGHT, PM_ERR_PATTERN_TERM_PAREN);
break;
}
default: {
pm_parser_err_current(parser, diag_id);
pm_node_t *right = (pm_node_t *) pm_missing_node_create(parser, parser->current.start, parser->current.end);
if (node == NULL) {
node = right;
} else {
node = (pm_node_t *) pm_alternation_pattern_node_create(parser, node, right, &operator);
}
break;
}
}
} while (accept1(parser, PM_TOKEN_PIPE));
// If we have an =>, then we are assigning this pattern to a variable.
// In this case we should create an assignment node.
while (accept1(parser, PM_TOKEN_EQUAL_GREATER)) {
pm_token_t operator = parser->previous;
expect1(parser, PM_TOKEN_IDENTIFIER, PM_ERR_PATTERN_IDENT_AFTER_HROCKET);
pm_token_t identifier = parser->previous;
int depth = pm_parser_local_depth(parser, &identifier);
if (depth < 0) {
depth = 0;
pm_parser_local_add_token(parser, &identifier);
}
pm_node_t *target = (pm_node_t *) pm_local_variable_target_node_create_depth(parser, &identifier, (uint32_t) depth);
node = (pm_node_t *) pm_capture_pattern_node_create(parser, node, target, &operator);
}
return node;
}
/**
* Parse a pattern matching expression.
*/
static pm_node_t *
parse_pattern(pm_parser_t *parser, bool top_pattern, pm_diagnostic_id_t diag_id) {
pm_node_t *node = NULL;
bool leading_rest = false;
bool trailing_rest = false;
switch (parser->current.type) {
case PM_TOKEN_LABEL: {
parser_lex(parser);
pm_node_t *key = (pm_node_t *) pm_symbol_node_label_create(parser, &parser->previous);
pm_token_t operator = not_provided(parser);
return (pm_node_t *) parse_pattern_hash(parser, (pm_node_t *) pm_assoc_node_create(parser, key, &operator, NULL));
}
case PM_TOKEN_USTAR_STAR: {
node = parse_pattern_keyword_rest(parser);
return (pm_node_t *) parse_pattern_hash(parser, node);
}
case PM_TOKEN_USTAR: {
if (top_pattern) {
parser_lex(parser);
node = (pm_node_t *) parse_pattern_rest(parser);
leading_rest = true;
break;
}
}
/* fallthrough */
default:
node = parse_pattern_primitives(parser, diag_id);
break;
}
// If we got a dynamic label symbol, then we need to treat it like the
// beginning of a hash pattern.
if (pm_symbol_node_label_p(node)) {
pm_token_t operator = not_provided(parser);
return (pm_node_t *) parse_pattern_hash(parser, (pm_node_t *) pm_assoc_node_create(parser, node, &operator, NULL));
}
if (top_pattern && match1(parser, PM_TOKEN_COMMA)) {
// If we have a comma, then we are now parsing either an array pattern or a
// find pattern. We need to parse all of the patterns, put them into a big
// list, and then determine which type of node we have.
pm_node_list_t nodes = { 0 };
pm_node_list_append(&nodes, node);
// Gather up all of the patterns into the list.
while (accept1(parser, PM_TOKEN_COMMA)) {
// Break early here in case we have a trailing comma.
if (match5(parser, PM_TOKEN_KEYWORD_THEN, PM_TOKEN_BRACE_RIGHT, PM_TOKEN_BRACKET_RIGHT, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON)) {
node = (pm_node_t *) pm_implicit_rest_node_create(parser, &parser->previous);
pm_node_list_append(&nodes, node);
break;
}
if (accept1(parser, PM_TOKEN_USTAR)) {
node = (pm_node_t *) parse_pattern_rest(parser);
// If we have already parsed a splat pattern, then this is an error. We
// will continue to parse the rest of the patterns, but we will indicate
// it as an error.
if (trailing_rest) {
pm_parser_err_previous(parser, PM_ERR_PATTERN_REST);
}
trailing_rest = true;
} else {
node = parse_pattern_primitives(parser, PM_ERR_PATTERN_EXPRESSION_AFTER_COMMA);
}
pm_node_list_append(&nodes, node);
}
// If the first pattern and the last pattern are rest patterns, then we will
// call this a find pattern, regardless of how many rest patterns are in
// between because we know we already added the appropriate errors.
// Otherwise we will create an array pattern.
if (PM_NODE_TYPE_P(nodes.nodes[0], PM_SPLAT_NODE) && PM_NODE_TYPE_P(nodes.nodes[nodes.size - 1], PM_SPLAT_NODE)) {
node = (pm_node_t *) pm_find_pattern_node_create(parser, &nodes);
} else {
node = (pm_node_t *) pm_array_pattern_node_node_list_create(parser, &nodes);
}
free(nodes.nodes);
} else if (leading_rest) {
// Otherwise, if we parsed a single splat pattern, then we know we have an
// array pattern, so we can go ahead and create that node.
node = (pm_node_t *) pm_array_pattern_node_rest_create(parser, node);
}
return node;
}
/**
* Incorporate a negative sign into a numeric node by subtracting 1 character
* from its start bounds. If it's a compound node, then we will recursively
* apply this function to its value.
*/
static inline void
parse_negative_numeric(pm_node_t *node) {
switch (PM_NODE_TYPE(node)) {
case PM_INTEGER_NODE:
case PM_FLOAT_NODE:
node->location.start--;
break;
case PM_RATIONAL_NODE:
node->location.start--;
parse_negative_numeric(((pm_rational_node_t *) node)->numeric);
break;
case PM_IMAGINARY_NODE:
node->location.start--;
parse_negative_numeric(((pm_imaginary_node_t *) node)->numeric);
break;
default:
assert(false && "unreachable");
break;
}
}
/**
* Returns a string content token at a particular location that is empty.
*/
static pm_token_t
parse_strings_empty_content(const uint8_t *location) {
return (pm_token_t) { .type = PM_TOKEN_STRING_CONTENT, .start = location, .end = location };
}
/**
* Parse a set of strings that could be concatenated together.
*/
static inline pm_node_t *
parse_strings(pm_parser_t *parser, pm_node_t *current) {
assert(parser->current.type == PM_TOKEN_STRING_BEGIN);
bool concating = false;
bool state_is_arg_labeled = lex_state_p(parser, PM_LEX_STATE_ARG | PM_LEX_STATE_LABELED);
while (match1(parser, PM_TOKEN_STRING_BEGIN)) {
pm_node_t *node = NULL;
// Here we have found a string literal. We'll parse it and add it to
// the list of strings.
const pm_lex_mode_t *lex_mode = parser->lex_modes.current;
assert(lex_mode->mode == PM_LEX_STRING);
bool lex_interpolation = lex_mode->as.string.interpolation;
pm_token_t opening = parser->current;
parser_lex(parser);
if (match2(parser, PM_TOKEN_STRING_END, PM_TOKEN_EOF)) {
expect1(parser, PM_TOKEN_STRING_END, PM_ERR_STRING_LITERAL_TERM);
// If we get here, then we have an end immediately after a
// start. In that case we'll create an empty content token and
// return an uninterpolated string.
pm_token_t content = parse_strings_empty_content(parser->previous.start);
pm_string_node_t *string = pm_string_node_create(parser, &opening, &content, &parser->previous);
pm_string_shared_init(&string->unescaped, content.start, content.end);
node = (pm_node_t *) string;
} else if (accept1(parser, PM_TOKEN_LABEL_END)) {
// If we get here, then we have an end of a label immediately
// after a start. In that case we'll create an empty symbol
// node.
pm_token_t opening = not_provided(parser);
pm_token_t content = parse_strings_empty_content(parser->previous.start);
pm_symbol_node_t *symbol = pm_symbol_node_create(parser, &opening, &content, &parser->previous);
pm_string_shared_init(&symbol->unescaped, content.start, content.end);
node = (pm_node_t *) symbol;
} else if (!lex_interpolation) {
// If we don't accept interpolation then we expect the string to
// start with a single string content node.
pm_string_t unescaped;
pm_token_t content;
if (match1(parser, PM_TOKEN_EOF)) {
unescaped = PM_STRING_EMPTY;
content = not_provided(parser);
} else {
unescaped = parser->current_string;
expect1(parser, PM_TOKEN_STRING_CONTENT, PM_ERR_EXPECT_STRING_CONTENT);
content = parser->previous;
}
// It is unfortunately possible to have multiple string content
// nodes in a row in the case that there's heredoc content in
// the middle of the string, like this cursed example:
//
// <<-END+'b
// a
// END
// c'+'d'
//
// In that case we need to switch to an interpolated string to
// be able to contain all of the parts.
if (match1(parser, PM_TOKEN_STRING_CONTENT)) {
pm_node_list_t parts = { 0 };
pm_token_t delimiters = not_provided(parser);
pm_node_t *part = (pm_node_t *) pm_string_node_create_unescaped(parser, &delimiters, &content, &delimiters, &unescaped);
pm_node_list_append(&parts, part);
do {
part = (pm_node_t *) pm_string_node_create_current_string(parser, &delimiters, &parser->current, &delimiters);
pm_node_list_append(&parts, part);
parser_lex(parser);
} while (match1(parser, PM_TOKEN_STRING_CONTENT));
expect1(parser, PM_TOKEN_STRING_END, PM_ERR_STRING_LITERAL_TERM);
node = (pm_node_t *) pm_interpolated_string_node_create(parser, &opening, &parts, &parser->previous);
} else if (accept1(parser, PM_TOKEN_LABEL_END) && !state_is_arg_labeled) {
node = (pm_node_t *) pm_symbol_node_create_unescaped(parser, &opening, &content, &parser->previous, &unescaped);
} else if (match1(parser, PM_TOKEN_EOF)) {
pm_parser_err_token(parser, &opening, PM_ERR_STRING_LITERAL_TERM);
node = (pm_node_t *) pm_string_node_create_unescaped(parser, &opening, &content, &parser->current, &unescaped);
} else {
expect1(parser, PM_TOKEN_STRING_END, PM_ERR_STRING_LITERAL_TERM);
node = (pm_node_t *) pm_string_node_create_unescaped(parser, &opening, &content, &parser->previous, &unescaped);
}
} else if (match1(parser, PM_TOKEN_STRING_CONTENT)) {
// In this case we've hit string content so we know the string
// at least has something in it. We'll need to check if the
// following token is the end (in which case we can return a
// plain string) or if it's not then it has interpolation.
pm_token_t content = parser->current;
pm_string_t unescaped = parser->current_string;
parser_lex(parser);
if (match2(parser, PM_TOKEN_STRING_END, PM_TOKEN_EOF)) {
node = (pm_node_t *) pm_string_node_create_unescaped(parser, &opening, &content, &parser->current, &unescaped);
pm_node_flag_set(node, parse_unescaped_encoding(parser));
expect1(parser, PM_TOKEN_STRING_END, PM_ERR_STRING_LITERAL_TERM);
} else if (accept1(parser, PM_TOKEN_LABEL_END)) {
node = (pm_node_t *) pm_symbol_node_create_unescaped(parser, &opening, &content, &parser->previous, &unescaped);
} else {
// If we get here, then we have interpolation so we'll need
// to create a string or symbol node with interpolation.
pm_node_list_t parts = { 0 };
pm_token_t string_opening = not_provided(parser);
pm_token_t string_closing = not_provided(parser);
pm_node_t *part = (pm_node_t *) pm_string_node_create_unescaped(parser, &string_opening, &parser->previous, &string_closing, &unescaped);
pm_node_flag_set(part, parse_unescaped_encoding(parser));
pm_node_list_append(&parts, part);
while (!match3(parser, PM_TOKEN_STRING_END, PM_TOKEN_LABEL_END, PM_TOKEN_EOF)) {
if ((part = parse_string_part(parser)) != NULL) {
pm_node_list_append(&parts, part);
}
}
if (accept1(parser, PM_TOKEN_LABEL_END) && !state_is_arg_labeled) {
node = (pm_node_t *) pm_interpolated_symbol_node_create(parser, &opening, &parts, &parser->previous);
} else if (match1(parser, PM_TOKEN_EOF)) {
pm_parser_err_token(parser, &opening, PM_ERR_STRING_INTERPOLATED_TERM);
node = (pm_node_t *) pm_interpolated_string_node_create(parser, &opening, &parts, &parser->current);
} else {
expect1(parser, PM_TOKEN_STRING_END, PM_ERR_STRING_INTERPOLATED_TERM);
node = (pm_node_t *) pm_interpolated_string_node_create(parser, &opening, &parts, &parser->previous);
}
}
} else {
// If we get here, then the first part of the string is not plain
// string content, in which case we need to parse the string as an
// interpolated string.
pm_node_list_t parts = { 0 };
pm_node_t *part;
while (!match3(parser, PM_TOKEN_STRING_END, PM_TOKEN_LABEL_END, PM_TOKEN_EOF)) {
if ((part = parse_string_part(parser)) != NULL) {
pm_node_list_append(&parts, part);
}
}
if (accept1(parser, PM_TOKEN_LABEL_END)) {
node = (pm_node_t *) pm_interpolated_symbol_node_create(parser, &opening, &parts, &parser->previous);
} else if (match1(parser, PM_TOKEN_EOF)) {
pm_parser_err_token(parser, &opening, PM_ERR_STRING_INTERPOLATED_TERM);
node = (pm_node_t *) pm_interpolated_string_node_create(parser, &opening, &parts, &parser->current);
} else {
expect1(parser, PM_TOKEN_STRING_END, PM_ERR_STRING_INTERPOLATED_TERM);
node = (pm_node_t *) pm_interpolated_string_node_create(parser, &opening, &parts, &parser->previous);
}
}
if (current == NULL) {
// If the node we just parsed is a symbol node, then we can't
// concatenate it with anything else, so we can now return that
// node.
if (PM_NODE_TYPE_P(node, PM_SYMBOL_NODE) || PM_NODE_TYPE_P(node, PM_INTERPOLATED_SYMBOL_NODE)) {
return node;
}
// If we don't already have a node, then it's fine and we can just
// set the result to be the node we just parsed.
current = node;
} else {
// Otherwise we need to check the type of the node we just parsed.
// If it cannot be concatenated with the previous node, then we'll
// need to add a syntax error.
if (!PM_NODE_TYPE_P(node, PM_STRING_NODE) && !PM_NODE_TYPE_P(node, PM_INTERPOLATED_STRING_NODE)) {
pm_parser_err_node(parser, node, PM_ERR_STRING_CONCATENATION);
}
// If we haven't already created our container for concatenation,
// we'll do that now.
if (!concating) {
concating = true;
pm_token_t bounds = not_provided(parser);
pm_interpolated_string_node_t *container = pm_interpolated_string_node_create(parser, &bounds, NULL, &bounds);
pm_interpolated_string_node_append(container, current);
current = (pm_node_t *) container;
}
pm_interpolated_string_node_append((pm_interpolated_string_node_t *) current, node);
}
}
return current;
}
/**
* Parse an expression that begins with the previous node that we just lexed.
*/
static inline pm_node_t *
parse_expression_prefix(pm_parser_t *parser, pm_binding_power_t binding_power, bool accepts_command_call) {
switch (parser->current.type) {
case PM_TOKEN_BRACKET_LEFT_ARRAY: {
parser_lex(parser);
pm_array_node_t *array = pm_array_node_create(parser, &parser->previous);
pm_accepts_block_stack_push(parser, true);
bool parsed_bare_hash = false;
while (!match2(parser, PM_TOKEN_BRACKET_RIGHT, PM_TOKEN_EOF)) {
// Handle the case where we don't have a comma and we have a
// newline followed by a right bracket.
if (accept1(parser, PM_TOKEN_NEWLINE) && match1(parser, PM_TOKEN_BRACKET_RIGHT)) {
break;
}
if (pm_array_node_size(array) != 0) {
expect1(parser, PM_TOKEN_COMMA, PM_ERR_ARRAY_SEPARATOR);
}
// If we have a right bracket immediately following a comma,
// this is allowed since it's a trailing comma. In this case we
// can break out of the loop.
if (match1(parser, PM_TOKEN_BRACKET_RIGHT)) break;
pm_node_t *element;
if (accept1(parser, PM_TOKEN_USTAR)) {
pm_token_t operator = parser->previous;
pm_node_t *expression = NULL;
if (match3(parser, PM_TOKEN_BRACKET_RIGHT, PM_TOKEN_COMMA, PM_TOKEN_EOF)) {
if (pm_parser_local_depth(parser, &parser->previous) == -1) {
pm_parser_err_token(parser, &operator, PM_ERR_ARGUMENT_NO_FORWARDING_STAR);
}
} else {
expression = parse_value_expression(parser, PM_BINDING_POWER_DEFINED, false, PM_ERR_ARRAY_EXPRESSION_AFTER_STAR);
}
element = (pm_node_t *) pm_splat_node_create(parser, &operator, expression);
} else if (match2(parser, PM_TOKEN_LABEL, PM_TOKEN_USTAR_STAR)) {
if (parsed_bare_hash) {
pm_parser_err_current(parser, PM_ERR_EXPRESSION_BARE_HASH);
}
pm_keyword_hash_node_t *hash = pm_keyword_hash_node_create(parser);
element = (pm_node_t *)hash;
if (!match8(parser, PM_TOKEN_EOF, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON, PM_TOKEN_EOF, PM_TOKEN_BRACE_RIGHT, PM_TOKEN_BRACKET_RIGHT, PM_TOKEN_KEYWORD_DO, PM_TOKEN_PARENTHESIS_RIGHT)) {
parse_assocs(parser, (pm_node_t *) hash);
}
parsed_bare_hash = true;
} else {
element = parse_value_expression(parser, PM_BINDING_POWER_DEFINED, false, PM_ERR_ARRAY_EXPRESSION);
if (pm_symbol_node_label_p(element) || accept1(parser, PM_TOKEN_EQUAL_GREATER)) {
if (parsed_bare_hash) {
pm_parser_err_previous(parser, PM_ERR_EXPRESSION_BARE_HASH);
}
pm_keyword_hash_node_t *hash = pm_keyword_hash_node_create(parser);
pm_token_t operator;
if (parser->previous.type == PM_TOKEN_EQUAL_GREATER) {
operator = parser->previous;
} else {
operator = not_provided(parser);
}
pm_node_t *value = parse_value_expression(parser, PM_BINDING_POWER_DEFINED, false, PM_ERR_HASH_VALUE);
pm_node_t *assoc = (pm_node_t *) pm_assoc_node_create(parser, element, &operator, value);
pm_keyword_hash_node_elements_append(hash, assoc);
element = (pm_node_t *)hash;
if (accept1(parser, PM_TOKEN_COMMA) && !match1(parser, PM_TOKEN_BRACKET_RIGHT)) {
parse_assocs(parser, (pm_node_t *) hash);
}
parsed_bare_hash = true;
}
}
pm_array_node_elements_append(array, element);
if (PM_NODE_TYPE_P(element, PM_MISSING_NODE)) break;
}
accept1(parser, PM_TOKEN_NEWLINE);
expect1(parser, PM_TOKEN_BRACKET_RIGHT, PM_ERR_ARRAY_TERM);
pm_array_node_close_set(array, &parser->previous);
pm_accepts_block_stack_pop(parser);
return (pm_node_t *) array;
}
case PM_TOKEN_PARENTHESIS_LEFT:
case PM_TOKEN_PARENTHESIS_LEFT_PARENTHESES: {
pm_token_t opening = parser->current;
parser_lex(parser);
while (accept2(parser, PM_TOKEN_SEMICOLON, PM_TOKEN_NEWLINE));
// If this is the end of the file or we match a right parenthesis, then
// we have an empty parentheses node, and we can immediately return.
if (match2(parser, PM_TOKEN_PARENTHESIS_RIGHT, PM_TOKEN_EOF)) {
expect1(parser, PM_TOKEN_PARENTHESIS_RIGHT, PM_ERR_EXPECT_RPAREN);
return (pm_node_t *) pm_parentheses_node_create(parser, &opening, NULL, &parser->previous);
}
// Otherwise, we're going to parse the first statement in the list
// of statements within the parentheses.
pm_accepts_block_stack_push(parser, true);
context_push(parser, PM_CONTEXT_PARENS);
pm_node_t *statement = parse_expression(parser, PM_BINDING_POWER_STATEMENT, true, PM_ERR_CANNOT_PARSE_EXPRESSION);
context_pop(parser);
// Determine if this statement is followed by a terminator. In the
// case of a single statement, this is fine. But in the case of
// multiple statements it's required.
bool terminator_found = accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON);
if (terminator_found) {
while (accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON));
}
// If we hit a right parenthesis, then we're done parsing the
// parentheses node, and we can check which kind of node we should
// return.
if (match1(parser, PM_TOKEN_PARENTHESIS_RIGHT)) {
if (opening.type == PM_TOKEN_PARENTHESIS_LEFT_PARENTHESES) {
lex_state_set(parser, PM_LEX_STATE_ENDARG);
}
parser_lex(parser);
pm_accepts_block_stack_pop(parser);
if (PM_NODE_TYPE_P(statement, PM_MULTI_TARGET_NODE) || PM_NODE_TYPE_P(statement, PM_SPLAT_NODE)) {
// If we have a single statement and are ending on a right
// parenthesis, then we need to check if this is possibly a
// multiple target node.
pm_multi_target_node_t *multi_target;
if (PM_NODE_TYPE_P(statement, PM_MULTI_TARGET_NODE) && ((pm_multi_target_node_t *) statement)->lparen_loc.start == NULL) {
multi_target = (pm_multi_target_node_t *) statement;
} else {
multi_target = pm_multi_target_node_create(parser);
pm_multi_target_node_targets_append(parser, multi_target, statement);
}
pm_location_t lparen_loc = PM_LOCATION_TOKEN_VALUE(&opening);
pm_location_t rparen_loc = PM_LOCATION_TOKEN_VALUE(&parser->previous);
multi_target->lparen_loc = lparen_loc;
multi_target->rparen_loc = rparen_loc;
multi_target->base.location.start = lparen_loc.start;
multi_target->base.location.end = rparen_loc.end;
if (match1(parser, PM_TOKEN_COMMA)) {
if (binding_power == PM_BINDING_POWER_STATEMENT) {
return parse_targets_validate(parser, (pm_node_t *) multi_target, PM_BINDING_POWER_INDEX);
}
return (pm_node_t *) multi_target;
}
return parse_target_validate(parser, (pm_node_t *) multi_target);
}
// If we have a single statement and are ending on a right parenthesis
// and we didn't return a multiple assignment node, then we can return a
// regular parentheses node now.
pm_statements_node_t *statements = pm_statements_node_create(parser);
pm_statements_node_body_append(statements, statement);
return (pm_node_t *) pm_parentheses_node_create(parser, &opening, (pm_node_t *) statements, &parser->previous);
}
// If we have more than one statement in the set of parentheses,
// then we are going to parse all of them as a list of statements.
// We'll do that here.
context_push(parser, PM_CONTEXT_PARENS);
pm_statements_node_t *statements = pm_statements_node_create(parser);
pm_statements_node_body_append(statements, statement);
// If we didn't find a terminator and we didn't find a right
// parenthesis, then this is a syntax error.
if (!terminator_found) {
pm_parser_err(parser, parser->current.start, parser->current.start, PM_ERR_EXPECT_EOL_AFTER_STATEMENT);
}
// Parse each statement within the parentheses.
while (true) {
pm_node_t *node = parse_expression(parser, PM_BINDING_POWER_STATEMENT, true, PM_ERR_CANNOT_PARSE_EXPRESSION);
pm_statements_node_body_append(statements, node);
// If we're recovering from a syntax error, then we need to stop
// parsing the statements now.
if (parser->recovering) {
// If this is the level of context where the recovery has
// happened, then we can mark the parser as done recovering.
if (match1(parser, PM_TOKEN_PARENTHESIS_RIGHT)) parser->recovering = false;
break;
}
// If we couldn't parse an expression at all, then we need to
// bail out of the loop.
if (PM_NODE_TYPE_P(node, PM_MISSING_NODE)) break;
// If we successfully parsed a statement, then we are going to
// need terminator to delimit them.
if (accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON)) {
while (accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON));
if (match1(parser, PM_TOKEN_PARENTHESIS_RIGHT)) break;
} else if (match1(parser, PM_TOKEN_PARENTHESIS_RIGHT)) {
break;
} else {
pm_parser_err(parser, parser->current.start, parser->current.start, PM_ERR_EXPECT_EOL_AFTER_STATEMENT);
}
}
context_pop(parser);
pm_accepts_block_stack_pop(parser);
expect1(parser, PM_TOKEN_PARENTHESIS_RIGHT, PM_ERR_EXPECT_RPAREN);
return (pm_node_t *) pm_parentheses_node_create(parser, &opening, (pm_node_t *) statements, &parser->previous);
}
case PM_TOKEN_BRACE_LEFT: {
pm_accepts_block_stack_push(parser, true);
parser_lex(parser);
pm_hash_node_t *node = pm_hash_node_create(parser, &parser->previous);
if (!match2(parser, PM_TOKEN_BRACE_RIGHT, PM_TOKEN_EOF)) {
parse_assocs(parser, (pm_node_t *) node);
accept1(parser, PM_TOKEN_NEWLINE);
}
pm_accepts_block_stack_pop(parser);
expect1(parser, PM_TOKEN_BRACE_RIGHT, PM_ERR_HASH_TERM);
pm_hash_node_closing_loc_set(node, &parser->previous);
return (pm_node_t *) node;
}
case PM_TOKEN_CHARACTER_LITERAL: {
parser_lex(parser);
pm_token_t opening = parser->previous;
opening.type = PM_TOKEN_STRING_BEGIN;
opening.end = opening.start + 1;
pm_token_t content = parser->previous;
content.type = PM_TOKEN_STRING_CONTENT;
content.start = content.start + 1;
pm_token_t closing = not_provided(parser);
pm_node_t *node = (pm_node_t *) pm_string_node_create_current_string(parser, &opening, &content, &closing);
pm_node_flag_set(node, parse_unescaped_encoding(parser));
// Characters can be followed by strings in which case they are
// automatically concatenated.
if (match1(parser, PM_TOKEN_STRING_BEGIN)) {
return parse_strings(parser, node);
}
return node;
}
case PM_TOKEN_CLASS_VARIABLE: {
parser_lex(parser);
pm_node_t *node = (pm_node_t *) pm_class_variable_read_node_create(parser, &parser->previous);
if (binding_power == PM_BINDING_POWER_STATEMENT && match1(parser, PM_TOKEN_COMMA)) {
node = parse_targets_validate(parser, node, PM_BINDING_POWER_INDEX);
}
return node;
}
case PM_TOKEN_CONSTANT: {
parser_lex(parser);
pm_token_t constant = parser->previous;
// If a constant is immediately followed by parentheses, then this is in
// fact a method call, not a constant read.
if (
match1(parser, PM_TOKEN_PARENTHESIS_LEFT) ||
(accepts_command_call && (token_begins_expression_p(parser->current.type) || match3(parser, PM_TOKEN_UAMPERSAND, PM_TOKEN_USTAR, PM_TOKEN_USTAR_STAR))) ||
(pm_accepts_block_stack_p(parser) && match2(parser, PM_TOKEN_KEYWORD_DO, PM_TOKEN_BRACE_LEFT))
) {
pm_arguments_t arguments = { 0 };
parse_arguments_list(parser, &arguments, true, accepts_command_call);
return (pm_node_t *) pm_call_node_fcall_create(parser, &constant, &arguments);
}
pm_node_t *node = (pm_node_t *) pm_constant_read_node_create(parser, &parser->previous);
if ((binding_power == PM_BINDING_POWER_STATEMENT) && match1(parser, PM_TOKEN_COMMA)) {
// If we get here, then we have a comma immediately following a
// constant, so we're going to parse this as a multiple assignment.
node = parse_targets_validate(parser, node, PM_BINDING_POWER_INDEX);
}
return node;
}
case PM_TOKEN_UCOLON_COLON: {
parser_lex(parser);
pm_token_t delimiter = parser->previous;
expect1(parser, PM_TOKEN_CONSTANT, PM_ERR_CONSTANT_PATH_COLON_COLON_CONSTANT);
pm_node_t *constant = (pm_node_t *) pm_constant_read_node_create(parser, &parser->previous);
pm_node_t *node = (pm_node_t *)pm_constant_path_node_create(parser, NULL, &delimiter, constant);
if ((binding_power == PM_BINDING_POWER_STATEMENT) && match1(parser, PM_TOKEN_COMMA)) {
node = parse_targets_validate(parser, node, PM_BINDING_POWER_INDEX);
}
return node;
}
case PM_TOKEN_UDOT_DOT:
case PM_TOKEN_UDOT_DOT_DOT: {
pm_token_t operator = parser->current;
parser_lex(parser);
pm_node_t *right = parse_expression(parser, pm_binding_powers[operator.type].left, false, PM_ERR_EXPECT_EXPRESSION_AFTER_OPERATOR);
return (pm_node_t *) pm_range_node_create(parser, NULL, &operator, right);
}
case PM_TOKEN_FLOAT:
parser_lex(parser);
return (pm_node_t *) pm_float_node_create(parser, &parser->previous);
case PM_TOKEN_FLOAT_IMAGINARY:
parser_lex(parser);
return (pm_node_t *) pm_float_node_imaginary_create(parser, &parser->previous);
case PM_TOKEN_FLOAT_RATIONAL:
parser_lex(parser);
return (pm_node_t *) pm_float_node_rational_create(parser, &parser->previous);
case PM_TOKEN_FLOAT_RATIONAL_IMAGINARY:
parser_lex(parser);
return (pm_node_t *) pm_float_node_rational_imaginary_create(parser, &parser->previous);
case PM_TOKEN_NUMBERED_REFERENCE: {
parser_lex(parser);
pm_node_t *node = (pm_node_t *) pm_numbered_reference_read_node_create(parser, &parser->previous);
if (binding_power == PM_BINDING_POWER_STATEMENT && match1(parser, PM_TOKEN_COMMA)) {
node = parse_targets_validate(parser, node, PM_BINDING_POWER_INDEX);
}
return node;
}
case PM_TOKEN_GLOBAL_VARIABLE: {
parser_lex(parser);
pm_node_t *node = (pm_node_t *) pm_global_variable_read_node_create(parser, &parser->previous);
if (binding_power == PM_BINDING_POWER_STATEMENT && match1(parser, PM_TOKEN_COMMA)) {
node = parse_targets_validate(parser, node, PM_BINDING_POWER_INDEX);
}
return node;
}
case PM_TOKEN_BACK_REFERENCE: {
parser_lex(parser);
pm_node_t *node = (pm_node_t *) pm_back_reference_read_node_create(parser, &parser->previous);
if (binding_power == PM_BINDING_POWER_STATEMENT && match1(parser, PM_TOKEN_COMMA)) {
node = parse_targets_validate(parser, node, PM_BINDING_POWER_INDEX);
}
return node;
}
case PM_TOKEN_IDENTIFIER:
case PM_TOKEN_METHOD_NAME: {
parser_lex(parser);
pm_token_t identifier = parser->previous;
pm_node_t *node = parse_variable_call(parser);
if (PM_NODE_TYPE_P(node, PM_CALL_NODE)) {
// If parse_variable_call returned with a call node, then we
// know the identifier is not in the local table. In that case
// we need to check if there are arguments following the
// identifier.
pm_call_node_t *call = (pm_call_node_t *) node;
pm_arguments_t arguments = { 0 };
if (parse_arguments_list(parser, &arguments, true, accepts_command_call)) {
// Since we found arguments, we need to turn off the
// variable call bit in the flags.
pm_node_flag_unset((pm_node_t *)call, PM_CALL_NODE_FLAGS_VARIABLE_CALL);
call->opening_loc = arguments.opening_loc;
call->arguments = arguments.arguments;
call->closing_loc = arguments.closing_loc;
call->block = arguments.block;
if (arguments.block != NULL) {
call->base.location.end = arguments.block->location.end;
} else if (arguments.closing_loc.start == NULL) {
if (arguments.arguments != NULL) {
call->base.location.end = arguments.arguments->base.location.end;
} else {
call->base.location.end = call->message_loc.end;
}
} else {
call->base.location.end = arguments.closing_loc.end;
}
}
} else {
// Otherwise, we know the identifier is in the local table. This
// can still be a method call if it is followed by arguments or
// a block, so we need to check for that here.
if (
(accepts_command_call && (token_begins_expression_p(parser->current.type) || match3(parser, PM_TOKEN_UAMPERSAND, PM_TOKEN_USTAR, PM_TOKEN_USTAR_STAR))) ||
(pm_accepts_block_stack_p(parser) && match2(parser, PM_TOKEN_KEYWORD_DO, PM_TOKEN_BRACE_LEFT))
) {
pm_arguments_t arguments = { 0 };
parse_arguments_list(parser, &arguments, true, accepts_command_call);
pm_call_node_t *fcall = pm_call_node_fcall_create(parser, &identifier, &arguments);
pm_node_destroy(parser, node);
return (pm_node_t *) fcall;
}
}
if ((binding_power == PM_BINDING_POWER_STATEMENT) && match1(parser, PM_TOKEN_COMMA)) {
node = parse_targets_validate(parser, node, PM_BINDING_POWER_INDEX);
}
return node;
}
case PM_TOKEN_HEREDOC_START: {
// Here we have found a heredoc. We'll parse it and add it to the
// list of strings.
pm_lex_mode_t *lex_mode = parser->lex_modes.current;
assert(lex_mode->mode == PM_LEX_HEREDOC);
pm_heredoc_quote_t quote = lex_mode->as.heredoc.quote;
pm_heredoc_indent_t indent = lex_mode->as.heredoc.indent;
parser_lex(parser);
pm_token_t opening = parser->previous;
pm_node_t *node;
pm_node_t *part;
if (match2(parser, PM_TOKEN_HEREDOC_END, PM_TOKEN_EOF)) {
// If we get here, then we have an empty heredoc. We'll create
// an empty content token and return an empty string node.
lex_mode_pop(parser);
expect1(parser, PM_TOKEN_HEREDOC_END, PM_ERR_HEREDOC_TERM);
pm_token_t content = parse_strings_empty_content(parser->previous.start);
if (quote == PM_HEREDOC_QUOTE_BACKTICK) {
node = (pm_node_t *) pm_xstring_node_create_unescaped(parser, &opening, &content, &parser->previous, &PM_STRING_EMPTY);
} else {
node = (pm_node_t *) pm_string_node_create_unescaped(parser, &opening, &content, &parser->previous, &PM_STRING_EMPTY);
}
node->location.end = opening.end;
} else if ((part = parse_string_part(parser)) == NULL) {
// If we get here, then we tried to find something in the
// heredoc but couldn't actually parse anything, so we'll just
// return a missing node.
node = (pm_node_t *) pm_missing_node_create(parser, parser->previous.start, parser->previous.end);
} else if (PM_NODE_TYPE_P(part, PM_STRING_NODE) && match2(parser, PM_TOKEN_HEREDOC_END, PM_TOKEN_EOF)) {
// If we get here, then the part that we parsed was plain string
// content and we're at the end of the heredoc, so we can return
// just a string node with the heredoc opening and closing as
// its opening and closing.
pm_node_flag_set(part, parse_unescaped_encoding(parser));
pm_string_node_t *cast = (pm_string_node_t *) part;
cast->opening_loc = PM_LOCATION_TOKEN_VALUE(&opening);
cast->closing_loc = PM_LOCATION_TOKEN_VALUE(&parser->current);
cast->base.location = cast->opening_loc;
if (quote == PM_HEREDOC_QUOTE_BACKTICK) {
assert(sizeof(pm_string_node_t) == sizeof(pm_x_string_node_t));
cast->base.type = PM_X_STRING_NODE;
}
size_t common_whitespace = lex_mode->as.heredoc.common_whitespace;
if (indent == PM_HEREDOC_INDENT_TILDE && (common_whitespace != (size_t) -1) && (common_whitespace != 0)) {
parse_heredoc_dedent_string(&cast->unescaped, common_whitespace);
}
node = (pm_node_t *) cast;
lex_mode_pop(parser);
expect1(parser, PM_TOKEN_HEREDOC_END, PM_ERR_HEREDOC_TERM);
} else {
// If we get here, then we have multiple parts in the heredoc,
// so we'll need to create an interpolated string node to hold
// them all.
pm_node_list_t parts = { 0 };
pm_node_list_append(&parts, part);
while (!match2(parser, PM_TOKEN_HEREDOC_END, PM_TOKEN_EOF)) {
if ((part = parse_string_part(parser)) != NULL) {
pm_node_list_append(&parts, part);
}
}
size_t common_whitespace = lex_mode->as.heredoc.common_whitespace;
// Now that we have all of the parts, create the correct type of
// interpolated node.
if (quote == PM_HEREDOC_QUOTE_BACKTICK) {
pm_interpolated_x_string_node_t *cast = pm_interpolated_xstring_node_create(parser, &opening, &opening);
cast->parts = parts;
lex_mode_pop(parser);
expect1(parser, PM_TOKEN_HEREDOC_END, PM_ERR_HEREDOC_TERM);
pm_interpolated_xstring_node_closing_set(cast, &parser->previous);
cast->base.location = cast->opening_loc;
node = (pm_node_t *) cast;
} else {
pm_interpolated_string_node_t *cast = pm_interpolated_string_node_create(parser, &opening, &parts, &opening);
lex_mode_pop(parser);
expect1(parser, PM_TOKEN_HEREDOC_END, PM_ERR_HEREDOC_TERM);
pm_interpolated_string_node_closing_set(cast, &parser->previous);
cast->base.location = cast->opening_loc;
node = (pm_node_t *) cast;
}
// If this is a heredoc that is indented with a ~, then we need
// to dedent each line by the common leading whitespace.
if (indent == PM_HEREDOC_INDENT_TILDE && (common_whitespace != (size_t) -1) && (common_whitespace != 0)) {
pm_node_list_t *nodes;
if (quote == PM_HEREDOC_QUOTE_BACKTICK) {
nodes = &((pm_interpolated_x_string_node_t *) node)->parts;
} else {
nodes = &((pm_interpolated_string_node_t *) node)->parts;
}
parse_heredoc_dedent(parser, nodes, common_whitespace);
}
}
if (match1(parser, PM_TOKEN_STRING_BEGIN)) {
return parse_strings(parser, node);
}
return node;
}
case PM_TOKEN_INSTANCE_VARIABLE: {
parser_lex(parser);
pm_node_t *node = (pm_node_t *) pm_instance_variable_read_node_create(parser, &parser->previous);
if (binding_power == PM_BINDING_POWER_STATEMENT && match1(parser, PM_TOKEN_COMMA)) {
node = parse_targets_validate(parser, node, PM_BINDING_POWER_INDEX);
}
return node;
}
case PM_TOKEN_INTEGER: {
pm_node_flags_t base = parser->integer_base;
parser_lex(parser);
return (pm_node_t *) pm_integer_node_create(parser, base, &parser->previous);
}
case PM_TOKEN_INTEGER_IMAGINARY: {
pm_node_flags_t base = parser->integer_base;
parser_lex(parser);
return (pm_node_t *) pm_integer_node_imaginary_create(parser, base, &parser->previous);
}
case PM_TOKEN_INTEGER_RATIONAL: {
pm_node_flags_t base = parser->integer_base;
parser_lex(parser);
return (pm_node_t *) pm_integer_node_rational_create(parser, base, &parser->previous);
}
case PM_TOKEN_INTEGER_RATIONAL_IMAGINARY: {
pm_node_flags_t base = parser->integer_base;
parser_lex(parser);
return (pm_node_t *) pm_integer_node_rational_imaginary_create(parser, base, &parser->previous);
}
case PM_TOKEN_KEYWORD___ENCODING__:
parser_lex(parser);
return (pm_node_t *) pm_source_encoding_node_create(parser, &parser->previous);
case PM_TOKEN_KEYWORD___FILE__:
parser_lex(parser);
return (pm_node_t *) pm_source_file_node_create(parser, &parser->previous);
case PM_TOKEN_KEYWORD___LINE__:
parser_lex(parser);
return (pm_node_t *) pm_source_line_node_create(parser, &parser->previous);
case PM_TOKEN_KEYWORD_ALIAS: {
if (binding_power != PM_BINDING_POWER_STATEMENT) {
pm_parser_err_current(parser, PM_ERR_STATEMENT_ALIAS);
}
parser_lex(parser);
pm_token_t keyword = parser->previous;
pm_node_t *new_name = parse_alias_argument(parser, true);
pm_node_t *old_name = parse_alias_argument(parser, false);
switch (PM_NODE_TYPE(new_name)) {
case PM_BACK_REFERENCE_READ_NODE:
case PM_NUMBERED_REFERENCE_READ_NODE:
case PM_GLOBAL_VARIABLE_READ_NODE: {
if (PM_NODE_TYPE_P(old_name, PM_BACK_REFERENCE_READ_NODE) || PM_NODE_TYPE_P(old_name, PM_NUMBERED_REFERENCE_READ_NODE) || PM_NODE_TYPE_P(old_name, PM_GLOBAL_VARIABLE_READ_NODE)) {
if (PM_NODE_TYPE_P(old_name, PM_NUMBERED_REFERENCE_READ_NODE)) {
pm_parser_err_node(parser, old_name, PM_ERR_ALIAS_ARGUMENT);
}
} else {
pm_parser_err_node(parser, old_name, PM_ERR_ALIAS_ARGUMENT);
}
return (pm_node_t *) pm_alias_global_variable_node_create(parser, &keyword, new_name, old_name);
}
case PM_SYMBOL_NODE:
case PM_INTERPOLATED_SYMBOL_NODE: {
if (!PM_NODE_TYPE_P(old_name, PM_SYMBOL_NODE) && !PM_NODE_TYPE_P(old_name, PM_INTERPOLATED_SYMBOL_NODE)) {
pm_parser_err_node(parser, old_name, PM_ERR_ALIAS_ARGUMENT);
}
}
/* fallthrough */
default:
return (pm_node_t *) pm_alias_method_node_create(parser, &keyword, new_name, old_name);
}
}
case PM_TOKEN_KEYWORD_CASE: {
parser_lex(parser);
pm_token_t case_keyword = parser->previous;
pm_node_t *predicate = NULL;
if (accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON)) {
while (accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON));
predicate = NULL;
} else if (match3(parser, PM_TOKEN_KEYWORD_WHEN, PM_TOKEN_KEYWORD_IN, PM_TOKEN_KEYWORD_END)) {
predicate = NULL;
} else if (!token_begins_expression_p(parser->current.type)) {
predicate = NULL;
} else {
predicate = parse_value_expression(parser, PM_BINDING_POWER_COMPOSITION, true, PM_ERR_CASE_EXPRESSION_AFTER_CASE);
while (accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON));
}
if (accept1(parser, PM_TOKEN_KEYWORD_END)) {
pm_parser_err_token(parser, &case_keyword, PM_ERR_CASE_MISSING_CONDITIONS);
return (pm_node_t *) pm_case_node_create(parser, &case_keyword, predicate, &parser->previous);
}
// At this point we can create a case node, though we don't yet know if it
// is a case-in or case-when node.
pm_token_t end_keyword = not_provided(parser);
pm_node_t *node;
if (match1(parser, PM_TOKEN_KEYWORD_WHEN)) {
pm_case_node_t *case_node = pm_case_node_create(parser, &case_keyword, predicate, &end_keyword);
// At this point we've seen a when keyword, so we know this is a
// case-when node. We will continue to parse the when nodes until we hit
// the end of the list.
while (accept1(parser, PM_TOKEN_KEYWORD_WHEN)) {
pm_token_t when_keyword = parser->previous;
pm_when_node_t *when_node = pm_when_node_create(parser, &when_keyword);
do {
if (accept1(parser, PM_TOKEN_USTAR)) {
pm_token_t operator = parser->previous;
pm_node_t *expression = parse_value_expression(parser, PM_BINDING_POWER_DEFINED, false, PM_ERR_EXPECT_EXPRESSION_AFTER_STAR);
pm_splat_node_t *splat_node = pm_splat_node_create(parser, &operator, expression);
pm_when_node_conditions_append(when_node, (pm_node_t *) splat_node);
if (PM_NODE_TYPE_P(expression, PM_MISSING_NODE)) break;
} else {
pm_node_t *condition = parse_value_expression(parser, PM_BINDING_POWER_DEFINED, false, PM_ERR_CASE_EXPRESSION_AFTER_WHEN);
pm_when_node_conditions_append(when_node, condition);
if (PM_NODE_TYPE_P(condition, PM_MISSING_NODE)) break;
}
} while (accept1(parser, PM_TOKEN_COMMA));
if (accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON)) {
accept1(parser, PM_TOKEN_KEYWORD_THEN);
} else {
expect1(parser, PM_TOKEN_KEYWORD_THEN, PM_ERR_EXPECT_WHEN_DELIMITER);
}
if (!match3(parser, PM_TOKEN_KEYWORD_WHEN, PM_TOKEN_KEYWORD_ELSE, PM_TOKEN_KEYWORD_END)) {
pm_statements_node_t *statements = parse_statements(parser, PM_CONTEXT_CASE_WHEN);
if (statements != NULL) {
pm_when_node_statements_set(when_node, statements);
}
}
pm_case_node_condition_append(case_node, (pm_node_t *) when_node);
}
// If we didn't parse any conditions (in or when) then we need
// to indicate that we have an error.
if (case_node->conditions.size == 0) {
pm_parser_err_token(parser, &case_keyword, PM_ERR_CASE_MISSING_CONDITIONS);
}
node = (pm_node_t *) case_node;
} else {
pm_case_match_node_t *case_node = pm_case_match_node_create(parser, &case_keyword, predicate, &end_keyword);
// If this is a case-match node (i.e., it is a pattern matching
// case statement) then we must have a predicate.
if (predicate == NULL) {
pm_parser_err_token(parser, &case_keyword, PM_ERR_CASE_MATCH_MISSING_PREDICATE);
}
// At this point we expect that we're parsing a case-in node. We will
// continue to parse the in nodes until we hit the end of the list.
while (match1(parser, PM_TOKEN_KEYWORD_IN)) {
bool previous_pattern_matching_newlines = parser->pattern_matching_newlines;
parser->pattern_matching_newlines = true;
lex_state_set(parser, PM_LEX_STATE_BEG | PM_LEX_STATE_LABEL);
parser->command_start = false;
parser_lex(parser);
pm_token_t in_keyword = parser->previous;
pm_node_t *pattern = parse_pattern(parser, true, PM_ERR_PATTERN_EXPRESSION_AFTER_IN);
parser->pattern_matching_newlines = previous_pattern_matching_newlines;
// Since we're in the top-level of the case-in node we need to check
// for guard clauses in the form of `if` or `unless` statements.
if (accept1(parser, PM_TOKEN_KEYWORD_IF_MODIFIER)) {
pm_token_t keyword = parser->previous;
pm_node_t *predicate = parse_value_expression(parser, PM_BINDING_POWER_DEFINED, true, PM_ERR_CONDITIONAL_IF_PREDICATE);
pattern = (pm_node_t *) pm_if_node_modifier_create(parser, pattern, &keyword, predicate);
} else if (accept1(parser, PM_TOKEN_KEYWORD_UNLESS_MODIFIER)) {
pm_token_t keyword = parser->previous;
pm_node_t *predicate = parse_value_expression(parser, PM_BINDING_POWER_DEFINED, true, PM_ERR_CONDITIONAL_UNLESS_PREDICATE);
pattern = (pm_node_t *) pm_unless_node_modifier_create(parser, pattern, &keyword, predicate);
}
// Now we need to check for the terminator of the in node's pattern.
// It can be a newline or semicolon optionally followed by a `then`
// keyword.
pm_token_t then_keyword;
if (accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON)) {
if (accept1(parser, PM_TOKEN_KEYWORD_THEN)) {
then_keyword = parser->previous;
} else {
then_keyword = not_provided(parser);
}
} else {
expect1(parser, PM_TOKEN_KEYWORD_THEN, PM_ERR_EXPECT_WHEN_DELIMITER);
then_keyword = parser->previous;
}
// Now we can actually parse the statements associated with the in
// node.
pm_statements_node_t *statements;
if (match3(parser, PM_TOKEN_KEYWORD_IN, PM_TOKEN_KEYWORD_ELSE, PM_TOKEN_KEYWORD_END)) {
statements = NULL;
} else {
statements = parse_statements(parser, PM_CONTEXT_CASE_IN);
}
// Now that we have the full pattern and statements, we can create the
// node and attach it to the case node.
pm_node_t *condition = (pm_node_t *) pm_in_node_create(parser, pattern, statements, &in_keyword, &then_keyword);
pm_case_match_node_condition_append(case_node, condition);
}
// If we didn't parse any conditions (in or when) then we need
// to indicate that we have an error.
if (case_node->conditions.size == 0) {
pm_parser_err_token(parser, &case_keyword, PM_ERR_CASE_MISSING_CONDITIONS);
}
node = (pm_node_t *) case_node;
}
accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON);
if (accept1(parser, PM_TOKEN_KEYWORD_ELSE)) {
pm_token_t else_keyword = parser->previous;
pm_else_node_t *else_node;
if (!match1(parser, PM_TOKEN_KEYWORD_END)) {
else_node = pm_else_node_create(parser, &else_keyword, parse_statements(parser, PM_CONTEXT_ELSE), &parser->current);
} else {
else_node = pm_else_node_create(parser, &else_keyword, NULL, &parser->current);
}
if (PM_NODE_TYPE_P(node, PM_CASE_NODE)) {
pm_case_node_consequent_set((pm_case_node_t *) node, else_node);
} else {
pm_case_match_node_consequent_set((pm_case_match_node_t *) node, else_node);
}
}
expect1(parser, PM_TOKEN_KEYWORD_END, PM_ERR_CASE_TERM);
if (PM_NODE_TYPE_P(node, PM_CASE_NODE)) {
pm_case_node_end_keyword_loc_set((pm_case_node_t *) node, &parser->previous);
} else {
pm_case_match_node_end_keyword_loc_set((pm_case_match_node_t *) node, &parser->previous);
}
return node;
}
case PM_TOKEN_KEYWORD_BEGIN: {
parser_lex(parser);
pm_token_t begin_keyword = parser->previous;
accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON);
pm_statements_node_t *begin_statements = NULL;
if (!match3(parser, PM_TOKEN_KEYWORD_RESCUE, PM_TOKEN_KEYWORD_ENSURE, PM_TOKEN_KEYWORD_END)) {
pm_accepts_block_stack_push(parser, true);
begin_statements = parse_statements(parser, PM_CONTEXT_BEGIN);
pm_accepts_block_stack_pop(parser);
accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON);
}
pm_begin_node_t *begin_node = pm_begin_node_create(parser, &begin_keyword, begin_statements);
parse_rescues(parser, begin_node, false);
expect1(parser, PM_TOKEN_KEYWORD_END, PM_ERR_BEGIN_TERM);
begin_node->base.location.end = parser->previous.end;
pm_begin_node_end_keyword_set(begin_node, &parser->previous);
if ((begin_node->else_clause != NULL) && (begin_node->rescue_clause == NULL)) {
pm_parser_err_node(parser, (pm_node_t *) begin_node->else_clause, PM_ERR_BEGIN_LONELY_ELSE);
}
return (pm_node_t *) begin_node;
}
case PM_TOKEN_KEYWORD_BEGIN_UPCASE: {
if (binding_power != PM_BINDING_POWER_STATEMENT) {
pm_parser_err_current(parser, PM_ERR_STATEMENT_PREEXE_BEGIN);
}
parser_lex(parser);
pm_token_t keyword = parser->previous;
expect1(parser, PM_TOKEN_BRACE_LEFT, PM_ERR_BEGIN_UPCASE_BRACE);
pm_token_t opening = parser->previous;
pm_statements_node_t *statements = parse_statements(parser, PM_CONTEXT_PREEXE);
expect1(parser, PM_TOKEN_BRACE_RIGHT, PM_ERR_BEGIN_UPCASE_TERM);
pm_context_t context = parser->current_context->context;
if ((context != PM_CONTEXT_MAIN) && (context != PM_CONTEXT_PREEXE)) {
pm_parser_err_token(parser, &keyword, PM_ERR_BEGIN_UPCASE_TOPLEVEL);
}
return (pm_node_t *) pm_pre_execution_node_create(parser, &keyword, &opening, statements, &parser->previous);
}
case PM_TOKEN_KEYWORD_BREAK:
case PM_TOKEN_KEYWORD_NEXT:
case PM_TOKEN_KEYWORD_RETURN: {
parser_lex(parser);
pm_token_t keyword = parser->previous;
pm_arguments_t arguments = { 0 };
if (
token_begins_expression_p(parser->current.type) ||
match2(parser, PM_TOKEN_USTAR, PM_TOKEN_USTAR_STAR)
) {
pm_binding_power_t binding_power = pm_binding_powers[parser->current.type].left;
if (binding_power == PM_BINDING_POWER_UNSET || binding_power >= PM_BINDING_POWER_RANGE) {
parse_arguments(parser, &arguments, false, PM_TOKEN_EOF);
}
}
switch (keyword.type) {
case PM_TOKEN_KEYWORD_BREAK:
return (pm_node_t *) pm_break_node_create(parser, &keyword, arguments.arguments);
case PM_TOKEN_KEYWORD_NEXT:
return (pm_node_t *) pm_next_node_create(parser, &keyword, arguments.arguments);
case PM_TOKEN_KEYWORD_RETURN: {
if (
(parser->current_context->context == PM_CONTEXT_CLASS) ||
(parser->current_context->context == PM_CONTEXT_MODULE)
) {
pm_parser_err_current(parser, PM_ERR_RETURN_INVALID);
}
return (pm_node_t *) pm_return_node_create(parser, &keyword, arguments.arguments);
}
default:
assert(false && "unreachable");
return (pm_node_t *) pm_missing_node_create(parser, parser->previous.start, parser->previous.end);
}
}
case PM_TOKEN_KEYWORD_SUPER: {
parser_lex(parser);
pm_token_t keyword = parser->previous;
pm_arguments_t arguments = { 0 };
parse_arguments_list(parser, &arguments, true, accepts_command_call);
if (
arguments.opening_loc.start == NULL &&
arguments.arguments == NULL &&
((arguments.block == NULL) || PM_NODE_TYPE_P(arguments.block, PM_BLOCK_NODE))
) {
return (pm_node_t *) pm_forwarding_super_node_create(parser, &keyword, &arguments);
}
return (pm_node_t *) pm_super_node_create(parser, &keyword, &arguments);
}
case PM_TOKEN_KEYWORD_YIELD: {
parser_lex(parser);
pm_token_t keyword = parser->previous;
pm_arguments_t arguments = { 0 };
parse_arguments_list(parser, &arguments, false, accepts_command_call);
return (pm_node_t *) pm_yield_node_create(parser, &keyword, &arguments.opening_loc, arguments.arguments, &arguments.closing_loc);
}
case PM_TOKEN_KEYWORD_CLASS: {
parser_lex(parser);
pm_token_t class_keyword = parser->previous;
pm_do_loop_stack_push(parser, false);
if (accept1(parser, PM_TOKEN_LESS_LESS)) {
pm_token_t operator = parser->previous;
pm_node_t *expression = parse_value_expression(parser, PM_BINDING_POWER_NOT, true, PM_ERR_EXPECT_EXPRESSION_AFTER_LESS_LESS);
pm_constant_id_t old_param_name = parser->current_param_name;
parser->current_param_name = 0;
pm_parser_scope_push(parser, true);
accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON);
pm_node_t *statements = NULL;
if (!match3(parser, PM_TOKEN_KEYWORD_RESCUE, PM_TOKEN_KEYWORD_ENSURE, PM_TOKEN_KEYWORD_END)) {
pm_accepts_block_stack_push(parser, true);
statements = (pm_node_t *) parse_statements(parser, PM_CONTEXT_SCLASS);
pm_accepts_block_stack_pop(parser);
}
if (match2(parser, PM_TOKEN_KEYWORD_RESCUE, PM_TOKEN_KEYWORD_ENSURE)) {
assert(statements == NULL || PM_NODE_TYPE_P(statements, PM_STATEMENTS_NODE));
statements = (pm_node_t *) parse_rescues_as_begin(parser, (pm_statements_node_t *) statements, false);
}
expect1(parser, PM_TOKEN_KEYWORD_END, PM_ERR_CLASS_TERM);
pm_constant_id_list_t locals = parser->current_scope->locals;
pm_parser_scope_pop(parser);
parser->current_param_name = old_param_name;
pm_do_loop_stack_pop(parser);
return (pm_node_t *) pm_singleton_class_node_create(parser, &locals, &class_keyword, &operator, expression, statements, &parser->previous);
}
pm_node_t *constant_path = parse_expression(parser, PM_BINDING_POWER_INDEX, false, PM_ERR_CLASS_NAME);
pm_token_t name = parser->previous;
if (name.type != PM_TOKEN_CONSTANT) {
pm_parser_err_token(parser, &name, PM_ERR_CLASS_NAME);
}
pm_token_t inheritance_operator;
pm_node_t *superclass;
if (match1(parser, PM_TOKEN_LESS)) {
inheritance_operator = parser->current;
lex_state_set(parser, PM_LEX_STATE_BEG);
parser->command_start = true;
parser_lex(parser);
superclass = parse_value_expression(parser, PM_BINDING_POWER_COMPOSITION, true, PM_ERR_CLASS_SUPERCLASS);
} else {
inheritance_operator = not_provided(parser);
superclass = NULL;
}
pm_constant_id_t old_param_name = parser->current_param_name;
parser->current_param_name = 0;
pm_parser_scope_push(parser, true);
if (inheritance_operator.type != PM_TOKEN_NOT_PROVIDED) {
expect2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON, PM_ERR_CLASS_UNEXPECTED_END);
} else {
accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON);
}
pm_node_t *statements = NULL;
if (!match3(parser, PM_TOKEN_KEYWORD_RESCUE, PM_TOKEN_KEYWORD_ENSURE, PM_TOKEN_KEYWORD_END)) {
pm_accepts_block_stack_push(parser, true);
statements = (pm_node_t *) parse_statements(parser, PM_CONTEXT_CLASS);
pm_accepts_block_stack_pop(parser);
}
if (match2(parser, PM_TOKEN_KEYWORD_RESCUE, PM_TOKEN_KEYWORD_ENSURE)) {
assert(statements == NULL || PM_NODE_TYPE_P(statements, PM_STATEMENTS_NODE));
statements = (pm_node_t *) parse_rescues_as_begin(parser, (pm_statements_node_t *) statements, false);
}
expect1(parser, PM_TOKEN_KEYWORD_END, PM_ERR_CLASS_TERM);
if (context_def_p(parser)) {
pm_parser_err_token(parser, &class_keyword, PM_ERR_CLASS_IN_METHOD);
}
pm_constant_id_list_t locals = parser->current_scope->locals;
pm_parser_scope_pop(parser);
parser->current_param_name = old_param_name;
pm_do_loop_stack_pop(parser);
if (!PM_NODE_TYPE_P(constant_path, PM_CONSTANT_PATH_NODE) && !(PM_NODE_TYPE_P(constant_path, PM_CONSTANT_READ_NODE))) {
pm_parser_err_node(parser, constant_path, PM_ERR_CLASS_NAME);
}
return (pm_node_t *) pm_class_node_create(parser, &locals, &class_keyword, constant_path, &name, &inheritance_operator, superclass, statements, &parser->previous);
}
case PM_TOKEN_KEYWORD_DEF: {
pm_token_t def_keyword = parser->current;
pm_node_t *receiver = NULL;
pm_token_t operator = not_provided(parser);
pm_token_t name = (pm_token_t) { .type = PM_TOKEN_MISSING, .start = def_keyword.end, .end = def_keyword.end };
// This context is necessary for lexing `...` in a bare params correctly.
// It must be pushed before lexing the first param, so it is here.
context_push(parser, PM_CONTEXT_DEF_PARAMS);
parser_lex(parser);
pm_constant_id_t old_param_name = parser->current_param_name;
switch (parser->current.type) {
case PM_CASE_OPERATOR:
pm_parser_scope_push(parser, true);
parser->current_param_name = 0;
lex_state_set(parser, PM_LEX_STATE_ENDFN);
parser_lex(parser);
name = parser->previous;
break;
case PM_TOKEN_IDENTIFIER: {
parser_lex(parser);
if (match2(parser, PM_TOKEN_DOT, PM_TOKEN_COLON_COLON)) {
receiver = parse_variable_call(parser);
pm_parser_scope_push(parser, true);
parser->current_param_name = 0;
lex_state_set(parser, PM_LEX_STATE_FNAME);
parser_lex(parser);
operator = parser->previous;
name = parse_method_definition_name(parser);
} else {
pm_refute_numbered_parameter(parser, parser->previous.start, parser->previous.end);
pm_parser_scope_push(parser, true);
parser->current_param_name = 0;
name = parser->previous;
}
break;
}
case PM_TOKEN_CONSTANT:
case PM_TOKEN_INSTANCE_VARIABLE:
case PM_TOKEN_CLASS_VARIABLE:
case PM_TOKEN_GLOBAL_VARIABLE:
case PM_TOKEN_KEYWORD_NIL:
case PM_TOKEN_KEYWORD_SELF:
case PM_TOKEN_KEYWORD_TRUE:
case PM_TOKEN_KEYWORD_FALSE:
case PM_TOKEN_KEYWORD___FILE__:
case PM_TOKEN_KEYWORD___LINE__:
case PM_TOKEN_KEYWORD___ENCODING__: {
pm_parser_scope_push(parser, true);
parser->current_param_name = 0;
parser_lex(parser);
pm_token_t identifier = parser->previous;
if (match2(parser, PM_TOKEN_DOT, PM_TOKEN_COLON_COLON)) {
lex_state_set(parser, PM_LEX_STATE_FNAME);
parser_lex(parser);
operator = parser->previous;
switch (identifier.type) {
case PM_TOKEN_CONSTANT:
receiver = (pm_node_t *) pm_constant_read_node_create(parser, &identifier);
break;
case PM_TOKEN_INSTANCE_VARIABLE:
receiver = (pm_node_t *) pm_instance_variable_read_node_create(parser, &identifier);
break;
case PM_TOKEN_CLASS_VARIABLE:
receiver = (pm_node_t *) pm_class_variable_read_node_create(parser, &identifier);
break;
case PM_TOKEN_GLOBAL_VARIABLE:
receiver = (pm_node_t *) pm_global_variable_read_node_create(parser, &identifier);
break;
case PM_TOKEN_KEYWORD_NIL:
receiver = (pm_node_t *) pm_nil_node_create(parser, &identifier);
break;
case PM_TOKEN_KEYWORD_SELF:
receiver = (pm_node_t *) pm_self_node_create(parser, &identifier);
break;
case PM_TOKEN_KEYWORD_TRUE:
receiver = (pm_node_t *) pm_true_node_create(parser, &identifier);
break;
case PM_TOKEN_KEYWORD_FALSE:
receiver = (pm_node_t *)pm_false_node_create(parser, &identifier);
break;
case PM_TOKEN_KEYWORD___FILE__:
receiver = (pm_node_t *) pm_source_file_node_create(parser, &identifier);
break;
case PM_TOKEN_KEYWORD___LINE__:
receiver = (pm_node_t *) pm_source_line_node_create(parser, &identifier);
break;
case PM_TOKEN_KEYWORD___ENCODING__:
receiver = (pm_node_t *) pm_source_encoding_node_create(parser, &identifier);
break;
default:
break;
}
name = parse_method_definition_name(parser);
} else {
name = identifier;
}
break;
}
case PM_TOKEN_PARENTHESIS_LEFT: {
// The current context is `PM_CONTEXT_DEF_PARAMS`, however the inner expression
// of this parenthesis should not be processed under this context.
// Thus, the context is popped here.
context_pop(parser);
parser_lex(parser);
pm_token_t lparen = parser->previous;
pm_node_t *expression = parse_value_expression(parser, PM_BINDING_POWER_STATEMENT, true, PM_ERR_DEF_RECEIVER);
expect1(parser, PM_TOKEN_PARENTHESIS_RIGHT, PM_ERR_EXPECT_RPAREN);
pm_token_t rparen = parser->previous;
lex_state_set(parser, PM_LEX_STATE_FNAME);
expect2(parser, PM_TOKEN_DOT, PM_TOKEN_COLON_COLON, PM_ERR_DEF_RECEIVER_TERM);
operator = parser->previous;
receiver = (pm_node_t *) pm_parentheses_node_create(parser, &lparen, expression, &rparen);
pm_parser_scope_push(parser, true);
parser->current_param_name = 0;
// To push `PM_CONTEXT_DEF_PARAMS` again is for the same reason as described the above.
context_push(parser, PM_CONTEXT_DEF_PARAMS);
name = parse_method_definition_name(parser);
break;
}
default:
pm_parser_scope_push(parser, true);
parser->current_param_name = 0;
name = parse_method_definition_name(parser);
break;
}
// If, after all that, we were unable to find a method name, add an
// error to the error list.
if (name.type == PM_TOKEN_MISSING) {
pm_parser_err_previous(parser, PM_ERR_DEF_NAME);
}
pm_token_t lparen;
pm_token_t rparen;
pm_parameters_node_t *params;
switch (parser->current.type) {
case PM_TOKEN_PARENTHESIS_LEFT: {
parser_lex(parser);
lparen = parser->previous;
if (match1(parser, PM_TOKEN_PARENTHESIS_RIGHT)) {
params = NULL;
} else {
params = parse_parameters(parser, PM_BINDING_POWER_DEFINED, true, false, true);
}
lex_state_set(parser, PM_LEX_STATE_BEG);
parser->command_start = true;
expect1(parser, PM_TOKEN_PARENTHESIS_RIGHT, PM_ERR_DEF_PARAMS_TERM_PAREN);
rparen = parser->previous;
break;
}
case PM_CASE_PARAMETER: {
// If we're about to lex a label, we need to add the label
// state to make sure the next newline is ignored.
if (parser->current.type == PM_TOKEN_LABEL) {
lex_state_set(parser, parser->lex_state | PM_LEX_STATE_LABEL);
}
lparen = not_provided(parser);
rparen = not_provided(parser);
params = parse_parameters(parser, PM_BINDING_POWER_DEFINED, false, false, true);
break;
}
default: {
lparen = not_provided(parser);
rparen = not_provided(parser);
params = NULL;
break;
}
}
uint32_t locals_body_index = (uint32_t) parser->current_scope->locals.size;
context_pop(parser);
pm_node_t *statements = NULL;
pm_token_t equal;
pm_token_t end_keyword;
if (accept1(parser, PM_TOKEN_EQUAL)) {
if (token_is_setter_name(&name)) {
pm_parser_err_token(parser, &name, PM_ERR_DEF_ENDLESS_SETTER);
}
equal = parser->previous;
context_push(parser, PM_CONTEXT_DEF);
pm_do_loop_stack_push(parser, false);
statements = (pm_node_t *) pm_statements_node_create(parser);
pm_node_t *statement = parse_expression(parser, PM_BINDING_POWER_DEFINED + 1, binding_power < PM_BINDING_POWER_COMPOSITION, PM_ERR_DEF_ENDLESS);
if (accept1(parser, PM_TOKEN_KEYWORD_RESCUE_MODIFIER)) {
pm_token_t rescue_keyword = parser->previous;
pm_node_t *value = parse_expression(parser, binding_power, false, PM_ERR_RESCUE_MODIFIER_VALUE);
pm_rescue_modifier_node_t *rescue_node = pm_rescue_modifier_node_create(parser, statement, &rescue_keyword, value);
statement = (pm_node_t *)rescue_node;
}
pm_statements_node_body_append((pm_statements_node_t *) statements, statement);
pm_do_loop_stack_pop(parser);
context_pop(parser);
end_keyword = not_provided(parser);
} else {
equal = not_provided(parser);
if (lparen.type == PM_TOKEN_NOT_PROVIDED) {
lex_state_set(parser, PM_LEX_STATE_BEG);
parser->command_start = true;
expect2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON, PM_ERR_DEF_PARAMS_TERM);
} else {
accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON);
}
pm_accepts_block_stack_push(parser, true);
pm_do_loop_stack_push(parser, false);
if (!match3(parser, PM_TOKEN_KEYWORD_RESCUE, PM_TOKEN_KEYWORD_ENSURE, PM_TOKEN_KEYWORD_END)) {
pm_accepts_block_stack_push(parser, true);
statements = (pm_node_t *) parse_statements(parser, PM_CONTEXT_DEF);
pm_accepts_block_stack_pop(parser);
}
if (match2(parser, PM_TOKEN_KEYWORD_RESCUE, PM_TOKEN_KEYWORD_ENSURE)) {
assert(statements == NULL || PM_NODE_TYPE_P(statements, PM_STATEMENTS_NODE));
statements = (pm_node_t *) parse_rescues_as_begin(parser, (pm_statements_node_t *) statements, true);
}
pm_accepts_block_stack_pop(parser);
pm_do_loop_stack_pop(parser);
expect1(parser, PM_TOKEN_KEYWORD_END, PM_ERR_DEF_TERM);
end_keyword = parser->previous;
}
pm_constant_id_list_t locals = parser->current_scope->locals;
parser->current_param_name = old_param_name;
pm_parser_scope_pop(parser);
return (pm_node_t *) pm_def_node_create(
parser,
&name,
receiver,
params,
statements,
&locals,
locals_body_index,
&def_keyword,
&operator,
&lparen,
&rparen,
&equal,
&end_keyword
);
}
case PM_TOKEN_KEYWORD_DEFINED: {
parser_lex(parser);
pm_token_t keyword = parser->previous;
pm_token_t lparen;
pm_token_t rparen;
pm_node_t *expression;
if (accept1(parser, PM_TOKEN_PARENTHESIS_LEFT)) {
lparen = parser->previous;
expression = parse_expression(parser, PM_BINDING_POWER_COMPOSITION, true, PM_ERR_DEFINED_EXPRESSION);
if (parser->recovering) {
rparen = not_provided(parser);
} else {
accept1(parser, PM_TOKEN_NEWLINE);
expect1(parser, PM_TOKEN_PARENTHESIS_RIGHT, PM_ERR_EXPECT_RPAREN);
rparen = parser->previous;
}
} else {
lparen = not_provided(parser);
rparen = not_provided(parser);
expression = parse_expression(parser, PM_BINDING_POWER_DEFINED, false, PM_ERR_DEFINED_EXPRESSION);
}
return (pm_node_t *) pm_defined_node_create(
parser,
&lparen,
expression,
&rparen,
&PM_LOCATION_TOKEN_VALUE(&keyword)
);
}
case PM_TOKEN_KEYWORD_END_UPCASE: {
if (binding_power != PM_BINDING_POWER_STATEMENT) {
pm_parser_err_current(parser, PM_ERR_STATEMENT_POSTEXE_END);
}
parser_lex(parser);
pm_token_t keyword = parser->previous;
if (context_def_p(parser)) {
pm_parser_warn_token(parser, &keyword, PM_WARN_END_IN_METHOD);
}
expect1(parser, PM_TOKEN_BRACE_LEFT, PM_ERR_END_UPCASE_BRACE);
pm_token_t opening = parser->previous;
pm_statements_node_t *statements = parse_statements(parser, PM_CONTEXT_POSTEXE);
expect1(parser, PM_TOKEN_BRACE_RIGHT, PM_ERR_END_UPCASE_TERM);
return (pm_node_t *) pm_post_execution_node_create(parser, &keyword, &opening, statements, &parser->previous);
}
case PM_TOKEN_KEYWORD_FALSE:
parser_lex(parser);
return (pm_node_t *)pm_false_node_create(parser, &parser->previous);
case PM_TOKEN_KEYWORD_FOR: {
parser_lex(parser);
pm_token_t for_keyword = parser->previous;
pm_node_t *index;
context_push(parser, PM_CONTEXT_FOR_INDEX);
// First, parse out the first index expression.
if (accept1(parser, PM_TOKEN_USTAR)) {
pm_token_t star_operator = parser->previous;
pm_node_t *name = NULL;
if (token_begins_expression_p(parser->current.type)) {
name = parse_expression(parser, PM_BINDING_POWER_INDEX, false, PM_ERR_EXPECT_EXPRESSION_AFTER_STAR);
}
index = (pm_node_t *) pm_splat_node_create(parser, &star_operator, name);
} else if (token_begins_expression_p(parser->current.type)) {
index = parse_expression(parser, PM_BINDING_POWER_INDEX, false, PM_ERR_EXPECT_EXPRESSION_AFTER_COMMA);
} else {
pm_parser_err_token(parser, &for_keyword, PM_ERR_FOR_INDEX);
index = (pm_node_t *) pm_missing_node_create(parser, for_keyword.start, for_keyword.end);
}
// Now, if there are multiple index expressions, parse them out.
if (match1(parser, PM_TOKEN_COMMA)) {
index = parse_targets(parser, index, PM_BINDING_POWER_INDEX);
} else {
index = parse_target(parser, index);
}
context_pop(parser);
pm_do_loop_stack_push(parser, true);
expect1(parser, PM_TOKEN_KEYWORD_IN, PM_ERR_FOR_IN);
pm_token_t in_keyword = parser->previous;
pm_node_t *collection = parse_value_expression(parser, PM_BINDING_POWER_COMPOSITION, true, PM_ERR_FOR_COLLECTION);
pm_do_loop_stack_pop(parser);
pm_token_t do_keyword;
if (accept1(parser, PM_TOKEN_KEYWORD_DO_LOOP)) {
do_keyword = parser->previous;
} else {
do_keyword = not_provided(parser);
}
accept2(parser, PM_TOKEN_SEMICOLON, PM_TOKEN_NEWLINE);
pm_statements_node_t *statements = NULL;
if (!accept1(parser, PM_TOKEN_KEYWORD_END)) {
statements = parse_statements(parser, PM_CONTEXT_FOR);
expect1(parser, PM_TOKEN_KEYWORD_END, PM_ERR_FOR_TERM);
}
return (pm_node_t *) pm_for_node_create(parser, index, collection, statements, &for_keyword, &in_keyword, &do_keyword, &parser->previous);
}
case PM_TOKEN_KEYWORD_IF:
parser_lex(parser);
return parse_conditional(parser, PM_CONTEXT_IF);
case PM_TOKEN_KEYWORD_UNDEF: {
if (binding_power != PM_BINDING_POWER_STATEMENT) {
pm_parser_err_current(parser, PM_ERR_STATEMENT_UNDEF);
}
parser_lex(parser);
pm_undef_node_t *undef = pm_undef_node_create(parser, &parser->previous);
pm_node_t *name = parse_undef_argument(parser);
if (PM_NODE_TYPE_P(name, PM_MISSING_NODE)) {
pm_node_destroy(parser, name);
} else {
pm_undef_node_append(undef, name);
while (match1(parser, PM_TOKEN_COMMA)) {
lex_state_set(parser, PM_LEX_STATE_FNAME | PM_LEX_STATE_FITEM);
parser_lex(parser);
name = parse_undef_argument(parser);
if (PM_NODE_TYPE_P(name, PM_MISSING_NODE)) {
pm_node_destroy(parser, name);
break;
}
pm_undef_node_append(undef, name);
}
}
return (pm_node_t *) undef;
}
case PM_TOKEN_KEYWORD_NOT: {
parser_lex(parser);
pm_token_t message = parser->previous;
pm_arguments_t arguments = { 0 };
pm_node_t *receiver = NULL;
accept1(parser, PM_TOKEN_NEWLINE);
if (accept1(parser, PM_TOKEN_PARENTHESIS_LEFT)) {
pm_token_t lparen = parser->previous;
if (accept1(parser, PM_TOKEN_PARENTHESIS_RIGHT)) {
receiver = (pm_node_t *) pm_parentheses_node_create(parser, &lparen, NULL, &parser->previous);
} else {
arguments.closing_loc = PM_LOCATION_TOKEN_VALUE(&parser->previous);
receiver = parse_expression(parser, PM_BINDING_POWER_COMPOSITION, true, PM_ERR_NOT_EXPRESSION);
pm_conditional_predicate(receiver);
if (!parser->recovering) {
accept1(parser, PM_TOKEN_NEWLINE);
expect1(parser, PM_TOKEN_PARENTHESIS_RIGHT, PM_ERR_EXPECT_RPAREN);
arguments.closing_loc = PM_LOCATION_TOKEN_VALUE(&parser->previous);
}
}
} else {
receiver = parse_expression(parser, PM_BINDING_POWER_NOT, true, PM_ERR_NOT_EXPRESSION);
pm_conditional_predicate(receiver);
}
return (pm_node_t *) pm_call_node_not_create(parser, receiver, &message, &arguments);
}
case PM_TOKEN_KEYWORD_UNLESS:
parser_lex(parser);
return parse_conditional(parser, PM_CONTEXT_UNLESS);
case PM_TOKEN_KEYWORD_MODULE: {
parser_lex(parser);
pm_token_t module_keyword = parser->previous;
pm_node_t *constant_path = parse_expression(parser, PM_BINDING_POWER_INDEX, false, PM_ERR_MODULE_NAME);
pm_token_t name;
// If we can recover from a syntax error that occurred while parsing
// the name of the module, then we'll handle that here.
if (PM_NODE_TYPE_P(constant_path, PM_MISSING_NODE)) {
pm_token_t missing = (pm_token_t) { .type = PM_TOKEN_MISSING, .start = parser->previous.end, .end = parser->previous.end };
return (pm_node_t *) pm_module_node_create(parser, NULL, &module_keyword, constant_path, &missing, NULL, &missing);
}
while (accept1(parser, PM_TOKEN_COLON_COLON)) {
pm_token_t double_colon = parser->previous;
expect1(parser, PM_TOKEN_CONSTANT, PM_ERR_CONSTANT_PATH_COLON_COLON_CONSTANT);
pm_node_t *constant = (pm_node_t *) pm_constant_read_node_create(parser, &parser->previous);
constant_path = (pm_node_t *) pm_constant_path_node_create(parser, constant_path, &double_colon, constant);
}
// Here we retrieve the name of the module. If it wasn't a constant,
// then it's possible that `module foo` was passed, which is a
// syntax error. We handle that here as well.
name = parser->previous;
if (name.type != PM_TOKEN_CONSTANT) {
pm_parser_err_token(parser, &name, PM_ERR_MODULE_NAME);
}
pm_constant_id_t old_param_name = parser->current_param_name;
parser->current_param_name = 0;
pm_parser_scope_push(parser, true);
accept2(parser, PM_TOKEN_SEMICOLON, PM_TOKEN_NEWLINE);
pm_node_t *statements = NULL;
if (!match3(parser, PM_TOKEN_KEYWORD_RESCUE, PM_TOKEN_KEYWORD_ENSURE, PM_TOKEN_KEYWORD_END)) {
pm_accepts_block_stack_push(parser, true);
statements = (pm_node_t *) parse_statements(parser, PM_CONTEXT_MODULE);
pm_accepts_block_stack_pop(parser);
}
if (match2(parser, PM_TOKEN_KEYWORD_RESCUE, PM_TOKEN_KEYWORD_ENSURE)) {
assert(statements == NULL || PM_NODE_TYPE_P(statements, PM_STATEMENTS_NODE));
statements = (pm_node_t *) parse_rescues_as_begin(parser, (pm_statements_node_t *) statements, false);
}
pm_constant_id_list_t locals = parser->current_scope->locals;
pm_parser_scope_pop(parser);
parser->current_param_name = old_param_name;
expect1(parser, PM_TOKEN_KEYWORD_END, PM_ERR_MODULE_TERM);
if (context_def_p(parser)) {
pm_parser_err_token(parser, &module_keyword, PM_ERR_MODULE_IN_METHOD);
}
return (pm_node_t *) pm_module_node_create(parser, &locals, &module_keyword, constant_path, &name, statements, &parser->previous);
}
case PM_TOKEN_KEYWORD_NIL:
parser_lex(parser);
return (pm_node_t *) pm_nil_node_create(parser, &parser->previous);
case PM_TOKEN_KEYWORD_REDO:
parser_lex(parser);
return (pm_node_t *) pm_redo_node_create(parser, &parser->previous);
case PM_TOKEN_KEYWORD_RETRY:
parser_lex(parser);
return (pm_node_t *) pm_retry_node_create(parser, &parser->previous);
case PM_TOKEN_KEYWORD_SELF:
parser_lex(parser);
return (pm_node_t *) pm_self_node_create(parser, &parser->previous);
case PM_TOKEN_KEYWORD_TRUE:
parser_lex(parser);
return (pm_node_t *) pm_true_node_create(parser, &parser->previous);
case PM_TOKEN_KEYWORD_UNTIL: {
pm_do_loop_stack_push(parser, true);
parser_lex(parser);
pm_token_t keyword = parser->previous;
pm_node_t *predicate = parse_value_expression(parser, PM_BINDING_POWER_COMPOSITION, true, PM_ERR_CONDITIONAL_UNTIL_PREDICATE);
pm_do_loop_stack_pop(parser);
expect3(parser, PM_TOKEN_KEYWORD_DO_LOOP, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON, PM_ERR_CONDITIONAL_UNTIL_PREDICATE);
pm_statements_node_t *statements = NULL;
if (!accept1(parser, PM_TOKEN_KEYWORD_END)) {
pm_accepts_block_stack_push(parser, true);
statements = parse_statements(parser, PM_CONTEXT_UNTIL);
pm_accepts_block_stack_pop(parser);
accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON);
expect1(parser, PM_TOKEN_KEYWORD_END, PM_ERR_UNTIL_TERM);
}
return (pm_node_t *) pm_until_node_create(parser, &keyword, &parser->previous, predicate, statements, 0);
}
case PM_TOKEN_KEYWORD_WHILE: {
pm_do_loop_stack_push(parser, true);
parser_lex(parser);
pm_token_t keyword = parser->previous;
pm_node_t *predicate = parse_value_expression(parser, PM_BINDING_POWER_COMPOSITION, true, PM_ERR_CONDITIONAL_WHILE_PREDICATE);
pm_do_loop_stack_pop(parser);
expect3(parser, PM_TOKEN_KEYWORD_DO_LOOP, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON, PM_ERR_CONDITIONAL_WHILE_PREDICATE);
pm_statements_node_t *statements = NULL;
if (!accept1(parser, PM_TOKEN_KEYWORD_END)) {
pm_accepts_block_stack_push(parser, true);
statements = parse_statements(parser, PM_CONTEXT_WHILE);
pm_accepts_block_stack_pop(parser);
accept2(parser, PM_TOKEN_NEWLINE, PM_TOKEN_SEMICOLON);
expect1(parser, PM_TOKEN_KEYWORD_END, PM_ERR_WHILE_TERM);
}
return (pm_node_t *) pm_while_node_create(parser, &keyword, &parser->previous, predicate, statements, 0);
}
case PM_TOKEN_PERCENT_LOWER_I: {
parser_lex(parser);
pm_token_t opening = parser->previous;
pm_array_node_t *array = pm_array_node_create(parser, &opening);
while (!match2(parser, PM_TOKEN_STRING_END, PM_TOKEN_EOF)) {
accept1(parser, PM_TOKEN_WORDS_SEP);
if (match1(parser, PM_TOKEN_STRING_END)) break;
if (match1(parser, PM_TOKEN_STRING_CONTENT)) {
pm_token_t opening = not_provided(parser);
pm_token_t closing = not_provided(parser);
pm_array_node_elements_append(array, (pm_node_t *) pm_symbol_node_create_current_string(parser, &opening, &parser->current, &closing));
}
expect1(parser, PM_TOKEN_STRING_CONTENT, PM_ERR_LIST_I_LOWER_ELEMENT);
}
pm_token_t closing = parser->current;
if (match1(parser, PM_TOKEN_EOF)) {
pm_parser_err_token(parser, &opening, PM_ERR_LIST_I_LOWER_TERM);
closing = (pm_token_t) { .type = PM_TOKEN_MISSING, .start = parser->previous.end, .end = parser->previous.end };
} else {
expect1(parser, PM_TOKEN_STRING_END, PM_ERR_LIST_I_LOWER_TERM);
}
pm_array_node_close_set(array, &closing);
return (pm_node_t *) array;
}
case PM_TOKEN_PERCENT_UPPER_I: {
parser_lex(parser);
pm_token_t opening = parser->previous;
pm_array_node_t *array = pm_array_node_create(parser, &opening);
// This is the current node that we are parsing that will be added to the
// list of elements.
pm_node_t *current = NULL;
while (!match2(parser, PM_TOKEN_STRING_END, PM_TOKEN_EOF)) {
switch (parser->current.type) {
case PM_TOKEN_WORDS_SEP: {
if (current == NULL) {
// If we hit a separator before we have any content, then we don't
// need to do anything.
} else {
// If we hit a separator after we've hit content, then we need to
// append that content to the list and reset the current node.
pm_array_node_elements_append(array, current);
current = NULL;
}
parser_lex(parser);
break;
}
case PM_TOKEN_STRING_CONTENT: {
pm_token_t opening = not_provided(parser);
pm_token_t closing = not_provided(parser);
if (current == NULL) {
// If we hit content and the current node is NULL, then this is
// the first string content we've seen. In that case we're going
// to create a new string node and set that to the current.
current = (pm_node_t *) pm_symbol_node_create_current_string(parser, &opening, &parser->current, &closing);
parser_lex(parser);
} else if (PM_NODE_TYPE_P(current, PM_INTERPOLATED_SYMBOL_NODE)) {
// If we hit string content and the current node is an
// interpolated string, then we need to append the string content
// to the list of child nodes.
pm_node_t *string = (pm_node_t *) pm_string_node_create_current_string(parser, &opening, &parser->current, &closing);
parser_lex(parser);
pm_interpolated_symbol_node_append((pm_interpolated_symbol_node_t *) current, string);
} else if (PM_NODE_TYPE_P(current, PM_SYMBOL_NODE)) {
// If we hit string content and the current node is a string node,
// then we need to convert the current node into an interpolated
// string and add the string content to the list of child nodes.
pm_node_t *string = (pm_node_t *) pm_string_node_create_current_string(parser, &opening, &parser->previous, &closing);
parser_lex(parser);
pm_interpolated_symbol_node_t *interpolated = pm_interpolated_symbol_node_create(parser, &opening, NULL, &closing);
pm_interpolated_symbol_node_append(interpolated, current);
pm_interpolated_symbol_node_append(interpolated, string);
current = (pm_node_t *) interpolated;
} else {
assert(false && "unreachable");
}
break;
}
case PM_TOKEN_EMBVAR: {
bool start_location_set = false;
if (current == NULL) {
// If we hit an embedded variable and the current node is NULL,
// then this is the start of a new string. We'll set the current
// node to a new interpolated string.
pm_token_t opening = not_provided(parser);
pm_token_t closing = not_provided(parser);
current = (pm_node_t *) pm_interpolated_symbol_node_create(parser, &opening, NULL, &closing);
} else if (PM_NODE_TYPE_P(current, PM_SYMBOL_NODE)) {
// If we hit an embedded variable and the current node is a string
// node, then we'll convert the current into an interpolated
// string and add the string node to the list of parts.
pm_token_t opening = not_provided(parser);
pm_token_t closing = not_provided(parser);
pm_interpolated_symbol_node_t *interpolated = pm_interpolated_symbol_node_create(parser, &opening, NULL, &closing);
current = (pm_node_t *) pm_symbol_node_to_string_node(parser, (pm_symbol_node_t *) current);
pm_interpolated_symbol_node_append(interpolated, current);
interpolated->base.location.start = current->location.start;
start_location_set = true;
current = (pm_node_t *) interpolated;
} else {
// If we hit an embedded variable and the current node is an
// interpolated string, then we'll just add the embedded variable.
}
pm_node_t *part = parse_string_part(parser);
pm_interpolated_symbol_node_append((pm_interpolated_symbol_node_t *) current, part);
if (!start_location_set) {
current->location.start = part->location.start;
}
break;
}
case PM_TOKEN_EMBEXPR_BEGIN: {
bool start_location_set = false;
if (current == NULL) {
// If we hit an embedded expression and the current node is NULL,
// then this is the start of a new string. We'll set the current
// node to a new interpolated string.
pm_token_t opening = not_provided(parser);
pm_token_t closing = not_provided(parser);
current = (pm_node_t *) pm_interpolated_symbol_node_create(parser, &opening, NULL, &closing);
} else if (PM_NODE_TYPE_P(current, PM_SYMBOL_NODE)) {
// If we hit an embedded expression and the current node is a
// string node, then we'll convert the current into an
// interpolated string and add the string node to the list of
// parts.
pm_token_t opening = not_provided(parser);
pm_token_t closing = not_provided(parser);
pm_interpolated_symbol_node_t *interpolated = pm_interpolated_symbol_node_create(parser, &opening, NULL, &closing);
current = (pm_node_t *) pm_symbol_node_to_string_node(parser, (pm_symbol_node_t *) current);
pm_interpolated_symbol_node_append(interpolated, current);
interpolated->base.location.start = current->location.start;
start_location_set = true;
current = (pm_node_t *) interpolated;
} else if (PM_NODE_TYPE_P(current, PM_INTERPOLATED_SYMBOL_NODE)) {
// If we hit an embedded expression and the current node is an
// interpolated string, then we'll just continue on.
} else {
assert(false && "unreachable");
}
pm_node_t *part = parse_string_part(parser);
pm_interpolated_symbol_node_append((pm_interpolated_symbol_node_t *) current, part);
if (!start_location_set) {
current->location.start = part->location.start;
}
break;
}
default:
expect1(parser, PM_TOKEN_STRING_CONTENT, PM_ERR_LIST_I_UPPER_ELEMENT);
parser_lex(parser);
break;
}
}
// If we have a current node, then we need to append it to the list.
if (current) {
pm_array_node_elements_append(array, current);
}
pm_token_t closing = parser->current;
if (match1(parser, PM_TOKEN_EOF)) {
pm_parser_err_token(parser, &opening, PM_ERR_LIST_I_UPPER_TERM);
closing = (pm_token_t) { .type = PM_TOKEN_MISSING, .start = parser->previous.end, .end = parser->previous.end };
} else {
expect1(parser, PM_TOKEN_STRING_END, PM_ERR_LIST_I_UPPER_TERM);
}
pm_array_node_close_set(array, &closing);
return (pm_node_t *) array;
}
case PM_TOKEN_PERCENT_LOWER_W: {
parser_lex(parser);
pm_token_t opening = parser->previous;
pm_array_node_t *array = pm_array_node_create(parser, &opening);
// skip all leading whitespaces
accept1(parser, PM_TOKEN_WORDS_SEP);
while (!match2(parser, PM_TOKEN_STRING_END, PM_TOKEN_EOF)) {
accept1(parser, PM_TOKEN_WORDS_SEP);
if (match1(parser, PM_TOKEN_STRING_END)) break;
if (match1(parser, PM_TOKEN_STRING_CONTENT)) {
pm_token_t opening = not_provided(parser);
pm_token_t closing = not_provided(parser);
pm_node_t *string = (pm_node_t *) pm_string_node_create_current_string(parser, &opening, &parser->current, &closing);
pm_array_node_elements_append(array, string);
}
expect1(parser, PM_TOKEN_STRING_CONTENT, PM_ERR_LIST_W_LOWER_ELEMENT);
}
pm_token_t closing = parser->current;
if (match1(parser, PM_TOKEN_EOF)) {
pm_parser_err_token(parser, &opening, PM_ERR_LIST_W_LOWER_TERM);
closing = (pm_token_t) { .type = PM_TOKEN_MISSING, .start = parser->previous.end, .end = parser->previous.end };
} else {
expect1(parser, PM_TOKEN_STRING_END, PM_ERR_LIST_W_LOWER_TERM);
}
pm_array_node_close_set(array, &closing);
return (pm_node_t *) array;
}
case PM_TOKEN_PERCENT_UPPER_W: {
parser_lex(parser);
pm_token_t opening = parser->previous;
pm_array_node_t *array = pm_array_node_create(parser, &opening);
// This is the current node that we are parsing that will be added
// to the list of elements.
pm_node_t *current = NULL;
while (!match2(parser, PM_TOKEN_STRING_END, PM_TOKEN_EOF)) {
switch (parser->current.type) {
case PM_TOKEN_WORDS_SEP: {
// Reset the explicit encoding if we hit a separator
// since each element can have its own encoding.
parser->explicit_encoding = NULL;
if (current == NULL) {
// If we hit a separator before we have any content,
// then we don't need to do anything.
} else {
// If we hit a separator after we've hit content,
// then we need to append that content to the list
// and reset the current node.
pm_array_node_elements_append(array, current);
current = NULL;
}
parser_lex(parser);
break;
}
case PM_TOKEN_STRING_CONTENT: {
pm_token_t opening = not_provided(parser);
pm_token_t closing = not_provided(parser);
pm_node_t *string = (pm_node_t *) pm_string_node_create_current_string(parser, &opening, &parser->current, &closing);
pm_node_flag_set(string, parse_unescaped_encoding(parser));
parser_lex(parser);
if (current == NULL) {
// If we hit content and the current node is NULL,
// then this is the first string content we've seen.
// In that case we're going to create a new string
// node and set that to the current.
current = string;
} else if (PM_NODE_TYPE_P(current, PM_INTERPOLATED_STRING_NODE)) {
// If we hit string content and the current node is
// an interpolated string, then we need to append
// the string content to the list of child nodes.
pm_interpolated_string_node_append((pm_interpolated_string_node_t *) current, string);
} else if (PM_NODE_TYPE_P(current, PM_STRING_NODE)) {
// If we hit string content and the current node is
// a string node, then we need to convert the
// current node into an interpolated string and add
// the string content to the list of child nodes.
pm_interpolated_string_node_t *interpolated = pm_interpolated_string_node_create(parser, &opening, NULL, &closing);
pm_interpolated_string_node_append(interpolated, current);
pm_interpolated_string_node_append(interpolated, string);
current = (pm_node_t *) interpolated;
} else {
assert(false && "unreachable");
}
break;
}
case PM_TOKEN_EMBVAR: {
if (current == NULL) {
// If we hit an embedded variable and the current
// node is NULL, then this is the start of a new
// string. We'll set the current node to a new
// interpolated string.
pm_token_t opening = not_provided(parser);
pm_token_t closing = not_provided(parser);
current = (pm_node_t *) pm_interpolated_string_node_create(parser, &opening, NULL, &closing);
} else if (PM_NODE_TYPE_P(current, PM_STRING_NODE)) {
// If we hit an embedded variable and the current
// node is a string node, then we'll convert the
// current into an interpolated string and add the
// string node to the list of parts.
pm_token_t opening = not_provided(parser);
pm_token_t closing = not_provided(parser);
pm_interpolated_string_node_t *interpolated = pm_interpolated_string_node_create(parser, &opening, NULL, &closing);
pm_interpolated_string_node_append(interpolated, current);
current = (pm_node_t *) interpolated;
} else {
// If we hit an embedded variable and the current
// node is an interpolated string, then we'll just
// add the embedded variable.
}
pm_node_t *part = parse_string_part(parser);
pm_interpolated_string_node_append((pm_interpolated_string_node_t *) current, part);
break;
}
case PM_TOKEN_EMBEXPR_BEGIN: {
if (current == NULL) {
// If we hit an embedded expression and the current
// node is NULL, then this is the start of a new
// string. We'll set the current node to a new
// interpolated string.
pm_token_t opening = not_provided(parser);
pm_token_t closing = not_provided(parser);
current = (pm_node_t *) pm_interpolated_string_node_create(parser, &opening, NULL, &closing);
} else if (PM_NODE_TYPE_P(current, PM_STRING_NODE)) {
// If we hit an embedded expression and the current
// node is a string node, then we'll convert the
// current into an interpolated string and add the
// string node to the list of parts.
pm_token_t opening = not_provided(parser);
pm_token_t closing = not_provided(parser);
pm_interpolated_string_node_t *interpolated = pm_interpolated_string_node_create(parser, &opening, NULL, &closing);
pm_interpolated_string_node_append(interpolated, current);
current = (pm_node_t *) interpolated;
} else if (PM_NODE_TYPE_P(current, PM_INTERPOLATED_STRING_NODE)) {
// If we hit an embedded expression and the current
// node is an interpolated string, then we'll just
// continue on.
} else {
assert(false && "unreachable");
}
pm_node_t *part = parse_string_part(parser);
pm_interpolated_string_node_append((pm_interpolated_string_node_t *) current, part);
break;
}
default:
expect1(parser, PM_TOKEN_STRING_CONTENT, PM_ERR_LIST_W_UPPER_ELEMENT);
parser_lex(parser);
break;
}
}
// If we have a current node, then we need to append it to the list.
if (current) {
pm_array_node_elements_append(array, current);
}
pm_token_t closing = parser->current;
if (match1(parser, PM_TOKEN_EOF)) {
pm_parser_err_token(parser, &opening, PM_ERR_LIST_W_UPPER_TERM);
closing = (pm_token_t) { .type = PM_TOKEN_MISSING, .start = parser->previous.end, .end = parser->previous.end };
} else {
expect1(parser, PM_TOKEN_STRING_END, PM_ERR_LIST_W_UPPER_TERM);
}
pm_array_node_close_set(array, &closing);
return (pm_node_t *) array;
}
case PM_TOKEN_REGEXP_BEGIN: {
pm_token_t opening = parser->current;
parser_lex(parser);
if (match1(parser, PM_TOKEN_REGEXP_END)) {
// If we get here, then we have an end immediately after a start. In
// that case we'll create an empty content token and return an
// uninterpolated regular expression.
pm_token_t content = (pm_token_t) {
.type = PM_TOKEN_STRING_CONTENT,
.start = parser->previous.end,
.end = parser->previous.end
};
parser_lex(parser);
return (pm_node_t *) pm_regular_expression_node_create(parser, &opening, &content, &parser->previous);
}
pm_interpolated_regular_expression_node_t *node;
if (match1(parser, PM_TOKEN_STRING_CONTENT)) {
// In this case we've hit string content so we know the regular
// expression at least has something in it. We'll need to check if the
// following token is the end (in which case we can return a plain
// regular expression) or if it's not then it has interpolation.
pm_string_t unescaped = parser->current_string;
pm_token_t content = parser->current;
parser_lex(parser);
// If we hit an end, then we can create a regular expression node
// without interpolation, which can be represented more succinctly and
// more easily compiled.
if (accept1(parser, PM_TOKEN_REGEXP_END)) {
return (pm_node_t *) pm_regular_expression_node_create_unescaped(parser, &opening, &content, &parser->previous, &unescaped);
}
// If we get here, then we have interpolation so we'll need to create
// a regular expression node with interpolation.
node = pm_interpolated_regular_expression_node_create(parser, &opening);
pm_token_t opening = not_provided(parser);
pm_token_t closing = not_provided(parser);
pm_node_t *part = (pm_node_t *) pm_string_node_create_unescaped(parser, &opening, &parser->previous, &closing, &unescaped);
pm_interpolated_regular_expression_node_append(node, part);
} else {
// If the first part of the body of the regular expression is not a
// string content, then we have interpolation and we need to create an
// interpolated regular expression node.
node = pm_interpolated_regular_expression_node_create(parser, &opening);
}
// Now that we're here and we have interpolation, we'll parse all of the
// parts into the list.
pm_node_t *part;
while (!match2(parser, PM_TOKEN_REGEXP_END, PM_TOKEN_EOF)) {
if ((part = parse_string_part(parser)) != NULL) {
pm_interpolated_regular_expression_node_append(node, part);
}
}
pm_token_t closing = parser->current;
if (match1(parser, PM_TOKEN_EOF)) {
pm_parser_err_token(parser, &opening, PM_ERR_REGEXP_TERM);
closing = (pm_token_t) { .type = PM_TOKEN_MISSING, .start = parser->previous.end, .end = parser->previous.end };
} else {
expect1(parser, PM_TOKEN_REGEXP_END, PM_ERR_REGEXP_TERM);
}
pm_interpolated_regular_expression_node_closing_set(node, &closing);
return (pm_node_t *) node;
}
case PM_TOKEN_BACKTICK:
case PM_TOKEN_PERCENT_LOWER_X: {
parser_lex(parser);
pm_token_t opening = parser->previous;
// When we get here, we don't know if this string is going to have
// interpolation or not, even though it is allowed. Still, we want to be
// able to return a string node without interpolation if we can since
// it'll be faster.
if (match1(parser, PM_TOKEN_STRING_END)) {
// If we get here, then we have an end immediately after a start. In
// that case we'll create an empty content token and return an
// uninterpolated string.
pm_token_t content = (pm_token_t) {
.type = PM_TOKEN_STRING_CONTENT,
.start = parser->previous.end,
.end = parser->previous.end
};
parser_lex(parser);
return (pm_node_t *) pm_xstring_node_create(parser, &opening, &content, &parser->previous);
}
pm_interpolated_x_string_node_t *node;
if (match1(parser, PM_TOKEN_STRING_CONTENT)) {
// In this case we've hit string content so we know the string
// at least has something in it. We'll need to check if the
// following token is the end (in which case we can return a
// plain string) or if it's not then it has interpolation.
pm_string_t unescaped = parser->current_string;
pm_token_t content = parser->current;
parser_lex(parser);
if (match1(parser, PM_TOKEN_STRING_END)) {
pm_node_t *node = (pm_node_t *) pm_xstring_node_create_unescaped(parser, &opening, &content, &parser->current, &unescaped);
pm_node_flag_set(node, parse_unescaped_encoding(parser));
parser_lex(parser);
return node;
}
// If we get here, then we have interpolation so we'll need to
// create a string node with interpolation.
node = pm_interpolated_xstring_node_create(parser, &opening, &opening);
pm_token_t opening = not_provided(parser);
pm_token_t closing = not_provided(parser);
pm_node_t *part = (pm_node_t *) pm_string_node_create_unescaped(parser, &opening, &parser->previous, &closing, &unescaped);
pm_node_flag_set(part, parse_unescaped_encoding(parser));
pm_interpolated_xstring_node_append(node, part);
} else {
// If the first part of the body of the string is not a string
// content, then we have interpolation and we need to create an
// interpolated string node.
node = pm_interpolated_xstring_node_create(parser, &opening, &opening);
}
pm_node_t *part;
while (!match2(parser, PM_TOKEN_STRING_END, PM_TOKEN_EOF)) {
if ((part = parse_string_part(parser)) != NULL) {
pm_interpolated_xstring_node_append(node, part);
}
}
pm_token_t closing = parser->current;
if (match1(parser, PM_TOKEN_EOF)) {
pm_parser_err_token(parser, &opening, PM_ERR_XSTRING_TERM);
closing = (pm_token_t) { .type = PM_TOKEN_MISSING, .start = parser->previous.end, .end = parser->previous.end };
} else {
expect1(parser, PM_TOKEN_STRING_END, PM_ERR_XSTRING_TERM);
}
pm_interpolated_xstring_node_closing_set(node, &closing);
return (pm_node_t *) node;
}
case PM_TOKEN_USTAR: {
parser_lex(parser);
// * operators at the beginning of expressions are only valid in the
// context of a multiple assignment. We enforce that here. We'll
// still lex past it though and create a missing node place.
if (binding_power != PM_BINDING_POWER_STATEMENT) {
return (pm_node_t *) pm_missing_node_create(parser, parser->previous.start, parser->previous.end);
}
pm_token_t operator = parser->previous;
pm_node_t *name = NULL;
if (token_begins_expression_p(parser->current.type)) {
name = parse_expression(parser, PM_BINDING_POWER_INDEX, false, PM_ERR_EXPECT_EXPRESSION_AFTER_STAR);
}
pm_node_t *splat = (pm_node_t *) pm_splat_node_create(parser, &operator, name);
if (match1(parser, PM_TOKEN_COMMA)) {
return parse_targets_validate(parser, splat, PM_BINDING_POWER_INDEX);
} else {
return parse_target_validate(parser, splat);
}
}
case PM_TOKEN_BANG: {
parser_lex(parser);
pm_token_t operator = parser->previous;
pm_node_t *receiver = parse_expression(parser, pm_binding_powers[parser->previous.type].right, binding_power < PM_BINDING_POWER_MATCH, PM_ERR_UNARY_RECEIVER_BANG);
pm_call_node_t *node = pm_call_node_unary_create(parser, &operator, receiver, "!");
pm_conditional_predicate(receiver);
return (pm_node_t *) node;
}
case PM_TOKEN_TILDE: {
parser_lex(parser);
pm_token_t operator = parser->previous;
pm_node_t *receiver = parse_expression(parser, pm_binding_powers[parser->previous.type].right, false, PM_ERR_UNARY_RECEIVER_TILDE);
pm_call_node_t *node = pm_call_node_unary_create(parser, &operator, receiver, "~");
return (pm_node_t *) node;
}
case PM_TOKEN_UMINUS: {
parser_lex(parser);
pm_token_t operator = parser->previous;
pm_node_t *receiver = parse_expression(parser, pm_binding_powers[parser->previous.type].right, false, PM_ERR_UNARY_RECEIVER_MINUS);
pm_call_node_t *node = pm_call_node_unary_create(parser, &operator, receiver, "-@");
return (pm_node_t *) node;
}
case PM_TOKEN_UMINUS_NUM: {
parser_lex(parser);
pm_token_t operator = parser->previous;
pm_node_t *node = parse_expression(parser, pm_binding_powers[parser->previous.type].right, false, PM_ERR_UNARY_RECEIVER_MINUS);
if (accept1(parser, PM_TOKEN_STAR_STAR)) {
pm_token_t exponent_operator = parser->previous;
pm_node_t *exponent = parse_expression(parser, pm_binding_powers[exponent_operator.type].right, false, PM_ERR_EXPECT_ARGUMENT);
node = (pm_node_t *) pm_call_node_binary_create(parser, node, &exponent_operator, exponent);
node = (pm_node_t *) pm_call_node_unary_create(parser, &operator, node, "-@");
} else {
switch (PM_NODE_TYPE(node)) {
case PM_INTEGER_NODE:
case PM_FLOAT_NODE:
case PM_RATIONAL_NODE:
case PM_IMAGINARY_NODE:
parse_negative_numeric(node);
break;
default:
node = (pm_node_t *) pm_call_node_unary_create(parser, &operator, node, "-@");
break;
}
}
return node;
}
case PM_TOKEN_MINUS_GREATER: {
int previous_lambda_enclosure_nesting = parser->lambda_enclosure_nesting;
parser->lambda_enclosure_nesting = parser->enclosure_nesting;
pm_accepts_block_stack_push(parser, true);
parser_lex(parser);
pm_token_t operator = parser->previous;
pm_parser_scope_push(parser, false);
pm_block_parameters_node_t *block_parameters;
switch (parser->current.type) {
case PM_TOKEN_PARENTHESIS_LEFT: {
parser->current_scope->explicit_params = true;
pm_token_t opening = parser->current;
parser_lex(parser);
if (match1(parser, PM_TOKEN_PARENTHESIS_RIGHT)) {
block_parameters = pm_block_parameters_node_create(parser, NULL, &opening);
} else {
block_parameters = parse_block_parameters(parser, false, &opening, true);
}
accept1(parser, PM_TOKEN_NEWLINE);
expect1(parser, PM_TOKEN_PARENTHESIS_RIGHT, PM_ERR_EXPECT_RPAREN);
pm_block_parameters_node_closing_set(block_parameters, &parser->previous);
break;
}
case PM_CASE_PARAMETER: {
parser->current_scope->explicit_params = true;
pm_accepts_block_stack_push(parser, false);
pm_token_t opening = not_provided(parser);
block_parameters = parse_block_parameters(parser, false, &opening, true);
pm_accepts_block_stack_pop(parser);
break;
}
default: {
block_parameters = NULL;
break;
}
}
uint32_t locals_body_index = 0;
if (block_parameters) {
locals_body_index = (uint32_t) parser->current_scope->locals.size;
}
pm_token_t opening;
pm_node_t *body = NULL;
parser->lambda_enclosure_nesting = previous_lambda_enclosure_nesting;
if (accept1(parser, PM_TOKEN_LAMBDA_BEGIN)) {
opening = parser->previous;
if (!accept1(parser, PM_TOKEN_BRACE_RIGHT)) {
body = (pm_node_t *) parse_statements(parser, PM_CONTEXT_LAMBDA_BRACES);
expect1(parser, PM_TOKEN_BRACE_RIGHT, PM_ERR_LAMBDA_TERM_BRACE);
}
} else {
expect1(parser, PM_TOKEN_KEYWORD_DO, PM_ERR_LAMBDA_OPEN);
opening = parser->previous;
if (!match3(parser, PM_TOKEN_KEYWORD_END, PM_TOKEN_KEYWORD_RESCUE, PM_TOKEN_KEYWORD_ENSURE)) {
pm_accepts_block_stack_push(parser, true);
body = (pm_node_t *) parse_statements(parser, PM_CONTEXT_LAMBDA_DO_END);
pm_accepts_block_stack_pop(parser);
}
if (match2(parser, PM_TOKEN_KEYWORD_RESCUE, PM_TOKEN_KEYWORD_ENSURE)) {
assert(body == NULL || PM_NODE_TYPE_P(body, PM_STATEMENTS_NODE));
body = (pm_node_t *) parse_rescues_as_begin(parser, (pm_statements_node_t *) body, false);
}
expect1(parser, PM_TOKEN_KEYWORD_END, PM_ERR_LAMBDA_TERM_END);
}
pm_node_t *parameters = (pm_node_t *) block_parameters;
uint8_t maximum = parser->current_scope->numbered_parameters;
if (parameters == NULL && (maximum > 0)) {
parameters = (pm_node_t *) pm_numbered_parameters_node_create(parser, &(pm_location_t) { .start = operator.start, .end = parser->previous.end }, maximum);
locals_body_index = maximum;
}
pm_constant_id_list_t locals = parser->current_scope->locals;
pm_parser_scope_pop(parser);
pm_accepts_block_stack_pop(parser);
return (pm_node_t *) pm_lambda_node_create(parser, &locals, locals_body_index, &operator, &opening, &parser->previous, parameters, body);
}
case PM_TOKEN_UPLUS: {
parser_lex(parser);
pm_token_t operator = parser->previous;
pm_node_t *receiver = parse_expression(parser, pm_binding_powers[parser->previous.type].right, false, PM_ERR_UNARY_RECEIVER_PLUS);
pm_call_node_t *node = pm_call_node_unary_create(parser, &operator, receiver, "+@");
return (pm_node_t *) node;
}
case PM_TOKEN_STRING_BEGIN:
return parse_strings(parser, NULL);
case PM_TOKEN_SYMBOL_BEGIN: {
pm_lex_mode_t lex_mode = *parser->lex_modes.current;
parser_lex(parser);
return parse_symbol(parser, &lex_mode, PM_LEX_STATE_END);
}
default:
if (context_recoverable(parser, &parser->current)) {
parser->recovering = true;
}
return (pm_node_t *) pm_missing_node_create(parser, parser->previous.start, parser->previous.end);
}
}
static inline pm_node_t *
parse_assignment_value(pm_parser_t *parser, pm_binding_power_t previous_binding_power, pm_binding_power_t binding_power, bool accepts_command_call, pm_diagnostic_id_t diag_id) {
pm_node_t *value = parse_value_expression(parser, binding_power, previous_binding_power == PM_BINDING_POWER_ASSIGNMENT ? accepts_command_call : previous_binding_power < PM_BINDING_POWER_MATCH, diag_id);
// Contradicting binding powers, the right-hand-side value of rthe assignment allows the `rescue` modifier.
if (match1(parser, PM_TOKEN_KEYWORD_RESCUE_MODIFIER)) {
pm_token_t rescue = parser->current;
parser_lex(parser);
pm_node_t *right = parse_expression(parser, binding_power, false, PM_ERR_RESCUE_MODIFIER_VALUE);
return (pm_node_t *) pm_rescue_modifier_node_create(parser, value, &rescue, right);
}
return value;
}
static inline pm_node_t *
parse_assignment_values(pm_parser_t *parser, pm_binding_power_t previous_binding_power, pm_binding_power_t binding_power, bool accepts_command_call, pm_diagnostic_id_t diag_id) {
pm_node_t *value = parse_starred_expression(parser, binding_power, previous_binding_power == PM_BINDING_POWER_ASSIGNMENT ? accepts_command_call : previous_binding_power < PM_BINDING_POWER_MATCH, diag_id);
bool is_single_value = true;
if (previous_binding_power == PM_BINDING_POWER_STATEMENT && (PM_NODE_TYPE_P(value, PM_SPLAT_NODE) || match1(parser, PM_TOKEN_COMMA))) {
is_single_value = false;
pm_token_t opening = not_provided(parser);
pm_array_node_t *array = pm_array_node_create(parser, &opening);
pm_array_node_elements_append(array, value);
value = (pm_node_t *) array;
while (accept1(parser, PM_TOKEN_COMMA)) {
pm_node_t *element = parse_starred_expression(parser, binding_power, false, PM_ERR_ARRAY_ELEMENT);
pm_array_node_elements_append(array, element);
if (PM_NODE_TYPE_P(element, PM_MISSING_NODE)) break;
}
}
// Contradicting binding powers, the right-hand-side value of the assignment allows the `rescue` modifier.
if (is_single_value && match1(parser, PM_TOKEN_KEYWORD_RESCUE_MODIFIER)) {
pm_token_t rescue = parser->current;
parser_lex(parser);
pm_node_t *right = parse_expression(parser, binding_power, false, PM_ERR_RESCUE_MODIFIER_VALUE);
return (pm_node_t *) pm_rescue_modifier_node_create(parser, value, &rescue, right);
}
return value;
}
/**
* Ensures a call node that is about to become a call operator node does not
* have arguments or a block attached. If it does, then we'll need to add an
* error message and destroy the arguments/block. Ideally we would keep the node
* around so that consumers would still have access to it, but we don't have a
* great structure for that at the moment.
*/
static void
parse_call_operator_write(pm_parser_t *parser, pm_call_node_t *call_node, const pm_token_t *operator) {
if (call_node->arguments != NULL) {
pm_parser_err_token(parser, operator, PM_ERR_OPERATOR_WRITE_ARGUMENTS);
pm_node_destroy(parser, (pm_node_t *) call_node->arguments);
call_node->arguments = NULL;
}
if (call_node->block != NULL) {
pm_parser_err_token(parser, operator, PM_ERR_OPERATOR_WRITE_BLOCK);
pm_node_destroy(parser, (pm_node_t *) call_node->block);
call_node->block = NULL;
}
}
static bool
name_is_identifier(pm_parser_t *parser, const uint8_t *source, size_t length) {
if (length == 0) {
return false;
}
size_t width = char_is_identifier_start(parser, source);
if (!width) {
return false;
}
uint8_t *cursor = ((uint8_t *)source) + width;
while (cursor < source + length && (width = char_is_identifier(parser, cursor))) {
cursor += width;
}
return cursor == source + length;
}
/**
* Potentially change a =~ with a regular expression with named captures into a
* match write node.
*/
static pm_node_t *
parse_regular_expression_named_captures(pm_parser_t *parser, const pm_string_t *content, pm_call_node_t *call) {
pm_string_list_t named_captures = { 0 };
pm_node_t *result;
if (pm_regexp_named_capture_group_names(pm_string_source(content), pm_string_length(content), &named_captures, parser->encoding_changed, parser->encoding) && (named_captures.length > 0)) {
// Since we should not create a MatchWriteNode when all capture names
// are invalid, creating a MatchWriteNode is delayed here.
pm_match_write_node_t *match = NULL;
pm_constant_id_list_t names = { 0 };
for (size_t index = 0; index < named_captures.length; index++) {
pm_string_t *string = &named_captures.strings[index];
const uint8_t *source = pm_string_source(string);
size_t length = pm_string_length(string);
pm_location_t location;
pm_constant_id_t name;
// If the name of the capture group isn't a valid identifier, we do
// not add it to the local table.
if (!name_is_identifier(parser, source, length)) continue;
if (content->type == PM_STRING_SHARED) {
// If the unescaped string is a slice of the source, then we can
// copy the names directly. The pointers will line up.
location = (pm_location_t) { .start = source, .end = source + length };
name = pm_parser_constant_id_location(parser, location.start, location.end);
pm_refute_numbered_parameter(parser, source, source + length);
} else {
// Otherwise, the name is a slice of the malloc-ed owned string,
// in which case we need to copy it out into a new string.
location = call->receiver->location;
void *memory = malloc(length);
if (memory == NULL) abort();
memcpy(memory, source, length);
// This silences clang analyzer warning about leak of memory pointed by `memory`.
// NOLINTNEXTLINE(clang-analyzer-*)
name = pm_parser_constant_id_owned(parser, (const uint8_t *) memory, length);
if (pm_token_is_numbered_parameter(source, source + length)) {
const pm_location_t *location = &call->receiver->location;
PM_PARSER_ERR_LOCATION_FORMAT(parser, location, PM_ERR_PARAMETER_NUMBERED_RESERVED, location->start);
}
}
if (name != 0) {
// We dont want to create duplicate targets if the capture name
// is duplicated.
if (pm_constant_id_list_includes(&names, name)) continue;
pm_constant_id_list_append(&names, name);
// Here we lazily create the MatchWriteNode since we know we're
// about to add a target.
if (match == NULL) match = pm_match_write_node_create(parser, call);
// First, find the depth of the local that is being assigned.
int depth;
if ((depth = pm_parser_local_depth_constant_id(parser, name)) == -1) {
pm_parser_local_add(parser, name);
}
// Next, create the local variable target and add it to the
// list of targets for the match.
pm_node_t *target = (pm_node_t *) pm_local_variable_target_node_create_values(parser, &location, name, depth == -1 ? 0 : (uint32_t) depth);
pm_node_list_append(&match->targets, target);
}
}
if (match != NULL) {
result = (pm_node_t *) match;
} else {
result = (pm_node_t *) call;
}
pm_constant_id_list_free(&names);
} else {
result = (pm_node_t *) call;
}
pm_string_list_free(&named_captures);
return result;
}
static inline pm_node_t *
parse_expression_infix(pm_parser_t *parser, pm_node_t *node, pm_binding_power_t previous_binding_power, pm_binding_power_t binding_power, bool accepts_command_call) {
pm_token_t token = parser->current;
switch (token.type) {
case PM_TOKEN_EQUAL: {
switch (PM_NODE_TYPE(node)) {
case PM_CALL_NODE: {
// If we have no arguments to the call node and we need this
// to be a target then this is either a method call or a
// local variable write. This _must_ happen before the value
// is parsed because it could be referenced in the value.
pm_call_node_t *call_node = (pm_call_node_t *) node;
if (pm_call_node_variable_call_p(call_node)) {
pm_parser_local_add_location(parser, call_node->message_loc.start, call_node->message_loc.end);
}
}
/* fallthrough */
case PM_CASE_WRITABLE: {
parser_lex(parser);
pm_node_t *value = parse_assignment_values(parser, previous_binding_power, PM_NODE_TYPE_P(node, PM_MULTI_TARGET_NODE) ? PM_BINDING_POWER_MULTI_ASSIGNMENT + 1 : binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_EQUAL);
return parse_write(parser, node, &token, value);
}
case PM_SPLAT_NODE: {
pm_multi_target_node_t *multi_target = pm_multi_target_node_create(parser);
pm_multi_target_node_targets_append(parser, multi_target, node);
parser_lex(parser);
pm_node_t *value = parse_assignment_values(parser, previous_binding_power, PM_BINDING_POWER_MULTI_ASSIGNMENT + 1, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_EQUAL);
return parse_write(parser, (pm_node_t *) multi_target, &token, value);
}
default:
parser_lex(parser);
// In this case we have an = sign, but we don't know what it's for. We
// need to treat it as an error. For now, we'll mark it as an error
// and just skip right past it.
pm_parser_err_token(parser, &token, PM_ERR_EXPECT_EXPRESSION_AFTER_EQUAL);
return node;
}
}
case PM_TOKEN_AMPERSAND_AMPERSAND_EQUAL: {
switch (PM_NODE_TYPE(node)) {
case PM_BACK_REFERENCE_READ_NODE:
case PM_NUMBERED_REFERENCE_READ_NODE:
pm_parser_err_node(parser, node, PM_ERR_WRITE_TARGET_READONLY);
/* fallthrough */
case PM_GLOBAL_VARIABLE_READ_NODE: {
parser_lex(parser);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_AMPAMPEQ);
pm_node_t *result = (pm_node_t *) pm_global_variable_and_write_node_create(parser, node, &token, value);
pm_node_destroy(parser, node);
return result;
}
case PM_CLASS_VARIABLE_READ_NODE: {
parser_lex(parser);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_AMPAMPEQ);
pm_node_t *result = (pm_node_t *) pm_class_variable_and_write_node_create(parser, (pm_class_variable_read_node_t *) node, &token, value);
pm_node_destroy(parser, node);
return result;
}
case PM_CONSTANT_PATH_NODE: {
parser_lex(parser);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_AMPAMPEQ);
return (pm_node_t *) pm_constant_path_and_write_node_create(parser, (pm_constant_path_node_t *) node, &token, value);
}
case PM_CONSTANT_READ_NODE: {
parser_lex(parser);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_AMPAMPEQ);
pm_node_t *result = (pm_node_t *) pm_constant_and_write_node_create(parser, (pm_constant_read_node_t *) node, &token, value);
pm_node_destroy(parser, node);
return result;
}
case PM_INSTANCE_VARIABLE_READ_NODE: {
parser_lex(parser);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_AMPAMPEQ);
pm_node_t *result = (pm_node_t *) pm_instance_variable_and_write_node_create(parser, (pm_instance_variable_read_node_t *) node, &token, value);
pm_node_destroy(parser, node);
return result;
}
case PM_LOCAL_VARIABLE_READ_NODE: {
pm_local_variable_read_node_t *cast = (pm_local_variable_read_node_t *) node;
parser_lex(parser);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_AMPAMPEQ);
pm_node_t *result = (pm_node_t *) pm_local_variable_and_write_node_create(parser, node, &token, value, cast->name, cast->depth);
pm_node_destroy(parser, node);
return result;
}
case PM_CALL_NODE: {
parser_lex(parser);
pm_call_node_t *cast = (pm_call_node_t *) node;
// If we have a vcall (a method with no arguments and no
// receiver that could have been a local variable) then we
// will transform it into a local variable write.
if (pm_call_node_variable_call_p(cast)) {
pm_location_t *message_loc = &cast->message_loc;
pm_refute_numbered_parameter(parser, message_loc->start, message_loc->end);
pm_constant_id_t constant_id = pm_parser_local_add_location(parser, message_loc->start, message_loc->end);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_AMPAMPEQ);
pm_node_t *result = (pm_node_t *) pm_local_variable_and_write_node_create(parser, (pm_node_t *) cast, &token, value, constant_id, 0);
pm_node_destroy(parser, (pm_node_t *) cast);
return result;
}
// If there is no call operator and the message is "[]" then
// this is an aref expression, and we can transform it into
// an aset expression.
if (pm_call_node_index_p(cast)) {
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_AMPAMPEQ);
return (pm_node_t *) pm_index_and_write_node_create(parser, cast, &token, value);
}
// If this node cannot be writable, then we have an error.
if (pm_call_node_writable_p(cast)) {
parse_write_name(parser, &cast->name);
} else {
pm_parser_err_node(parser, node, PM_ERR_WRITE_TARGET_UNEXPECTED);
}
parse_call_operator_write(parser, cast, &token);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_AMPAMPEQ);
return (pm_node_t *) pm_call_and_write_node_create(parser, cast, &token, value);
}
case PM_MULTI_WRITE_NODE: {
parser_lex(parser);
pm_parser_err_token(parser, &token, PM_ERR_AMPAMPEQ_MULTI_ASSIGN);
return node;
}
default:
parser_lex(parser);
// In this case we have an &&= sign, but we don't know what it's for.
// We need to treat it as an error. For now, we'll mark it as an error
// and just skip right past it.
pm_parser_err_token(parser, &token, PM_ERR_EXPECT_EXPRESSION_AFTER_AMPAMPEQ);
return node;
}
}
case PM_TOKEN_PIPE_PIPE_EQUAL: {
switch (PM_NODE_TYPE(node)) {
case PM_BACK_REFERENCE_READ_NODE:
case PM_NUMBERED_REFERENCE_READ_NODE:
pm_parser_err_node(parser, node, PM_ERR_WRITE_TARGET_READONLY);
/* fallthrough */
case PM_GLOBAL_VARIABLE_READ_NODE: {
parser_lex(parser);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_PIPEPIPEEQ);
pm_node_t *result = (pm_node_t *) pm_global_variable_or_write_node_create(parser, node, &token, value);
pm_node_destroy(parser, node);
return result;
}
case PM_CLASS_VARIABLE_READ_NODE: {
parser_lex(parser);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_PIPEPIPEEQ);
pm_node_t *result = (pm_node_t *) pm_class_variable_or_write_node_create(parser, (pm_class_variable_read_node_t *) node, &token, value);
pm_node_destroy(parser, node);
return result;
}
case PM_CONSTANT_PATH_NODE: {
parser_lex(parser);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_PIPEPIPEEQ);
return (pm_node_t *) pm_constant_path_or_write_node_create(parser, (pm_constant_path_node_t *) node, &token, value);
}
case PM_CONSTANT_READ_NODE: {
parser_lex(parser);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_PIPEPIPEEQ);
pm_node_t *result = (pm_node_t *) pm_constant_or_write_node_create(parser, (pm_constant_read_node_t *) node, &token, value);
pm_node_destroy(parser, node);
return result;
}
case PM_INSTANCE_VARIABLE_READ_NODE: {
parser_lex(parser);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_PIPEPIPEEQ);
pm_node_t *result = (pm_node_t *) pm_instance_variable_or_write_node_create(parser, (pm_instance_variable_read_node_t *) node, &token, value);
pm_node_destroy(parser, node);
return result;
}
case PM_LOCAL_VARIABLE_READ_NODE: {
pm_local_variable_read_node_t *cast = (pm_local_variable_read_node_t *) node;
parser_lex(parser);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_PIPEPIPEEQ);
pm_node_t *result = (pm_node_t *) pm_local_variable_or_write_node_create(parser, node, &token, value, cast->name, cast->depth);
pm_node_destroy(parser, node);
return result;
}
case PM_CALL_NODE: {
parser_lex(parser);
pm_call_node_t *cast = (pm_call_node_t *) node;
// If we have a vcall (a method with no arguments and no
// receiver that could have been a local variable) then we
// will transform it into a local variable write.
if (pm_call_node_variable_call_p(cast)) {
pm_location_t *message_loc = &cast->message_loc;
pm_refute_numbered_parameter(parser, message_loc->start, message_loc->end);
pm_constant_id_t constant_id = pm_parser_local_add_location(parser, message_loc->start, message_loc->end);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_PIPEPIPEEQ);
pm_node_t *result = (pm_node_t *) pm_local_variable_or_write_node_create(parser, (pm_node_t *) cast, &token, value, constant_id, 0);
pm_node_destroy(parser, (pm_node_t *) cast);
return result;
}
// If there is no call operator and the message is "[]" then
// this is an aref expression, and we can transform it into
// an aset expression.
if (pm_call_node_index_p(cast)) {
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_PIPEPIPEEQ);
return (pm_node_t *) pm_index_or_write_node_create(parser, cast, &token, value);
}
// If this node cannot be writable, then we have an error.
if (pm_call_node_writable_p(cast)) {
parse_write_name(parser, &cast->name);
} else {
pm_parser_err_node(parser, node, PM_ERR_WRITE_TARGET_UNEXPECTED);
}
parse_call_operator_write(parser, cast, &token);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_PIPEPIPEEQ);
return (pm_node_t *) pm_call_or_write_node_create(parser, cast, &token, value);
}
case PM_MULTI_WRITE_NODE: {
parser_lex(parser);
pm_parser_err_token(parser, &token, PM_ERR_PIPEPIPEEQ_MULTI_ASSIGN);
return node;
}
default:
parser_lex(parser);
// In this case we have an ||= sign, but we don't know what it's for.
// We need to treat it as an error. For now, we'll mark it as an error
// and just skip right past it.
pm_parser_err_token(parser, &token, PM_ERR_EXPECT_EXPRESSION_AFTER_PIPEPIPEEQ);
return node;
}
}
case PM_TOKEN_AMPERSAND_EQUAL:
case PM_TOKEN_CARET_EQUAL:
case PM_TOKEN_GREATER_GREATER_EQUAL:
case PM_TOKEN_LESS_LESS_EQUAL:
case PM_TOKEN_MINUS_EQUAL:
case PM_TOKEN_PERCENT_EQUAL:
case PM_TOKEN_PIPE_EQUAL:
case PM_TOKEN_PLUS_EQUAL:
case PM_TOKEN_SLASH_EQUAL:
case PM_TOKEN_STAR_EQUAL:
case PM_TOKEN_STAR_STAR_EQUAL: {
switch (PM_NODE_TYPE(node)) {
case PM_BACK_REFERENCE_READ_NODE:
case PM_NUMBERED_REFERENCE_READ_NODE:
pm_parser_err_node(parser, node, PM_ERR_WRITE_TARGET_READONLY);
/* fallthrough */
case PM_GLOBAL_VARIABLE_READ_NODE: {
parser_lex(parser);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_OPERATOR);
pm_node_t *result = (pm_node_t *) pm_global_variable_operator_write_node_create(parser, node, &token, value);
pm_node_destroy(parser, node);
return result;
}
case PM_CLASS_VARIABLE_READ_NODE: {
parser_lex(parser);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_OPERATOR);
pm_node_t *result = (pm_node_t *) pm_class_variable_operator_write_node_create(parser, (pm_class_variable_read_node_t *) node, &token, value);
pm_node_destroy(parser, node);
return result;
}
case PM_CONSTANT_PATH_NODE: {
parser_lex(parser);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_OPERATOR);
return (pm_node_t *) pm_constant_path_operator_write_node_create(parser, (pm_constant_path_node_t *) node, &token, value);
}
case PM_CONSTANT_READ_NODE: {
parser_lex(parser);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_OPERATOR);
pm_node_t *result = (pm_node_t *) pm_constant_operator_write_node_create(parser, (pm_constant_read_node_t *) node, &token, value);
pm_node_destroy(parser, node);
return result;
}
case PM_INSTANCE_VARIABLE_READ_NODE: {
parser_lex(parser);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_OPERATOR);
pm_node_t *result = (pm_node_t *) pm_instance_variable_operator_write_node_create(parser, (pm_instance_variable_read_node_t *) node, &token, value);
pm_node_destroy(parser, node);
return result;
}
case PM_LOCAL_VARIABLE_READ_NODE: {
pm_local_variable_read_node_t *cast = (pm_local_variable_read_node_t *) node;
parser_lex(parser);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_OPERATOR);
pm_node_t *result = (pm_node_t *) pm_local_variable_operator_write_node_create(parser, node, &token, value, cast->name, cast->depth);
pm_node_destroy(parser, node);
return result;
}
case PM_CALL_NODE: {
parser_lex(parser);
pm_call_node_t *cast = (pm_call_node_t *) node;
// If we have a vcall (a method with no arguments and no
// receiver that could have been a local variable) then we
// will transform it into a local variable write.
if (pm_call_node_variable_call_p(cast)) {
pm_location_t *message_loc = &cast->message_loc;
pm_refute_numbered_parameter(parser, message_loc->start, message_loc->end);
pm_constant_id_t constant_id = pm_parser_local_add_location(parser, message_loc->start, message_loc->end);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_OPERATOR);
pm_node_t *result = (pm_node_t *) pm_local_variable_operator_write_node_create(parser, (pm_node_t *) cast, &token, value, constant_id, 0);
pm_node_destroy(parser, (pm_node_t *) cast);
return result;
}
// If there is no call operator and the message is "[]" then
// this is an aref expression, and we can transform it into
// an aset expression.
if (pm_call_node_index_p(cast)) {
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_OPERATOR);
return (pm_node_t *) pm_index_operator_write_node_create(parser, cast, &token, value);
}
// If this node cannot be writable, then we have an error.
if (pm_call_node_writable_p(cast)) {
parse_write_name(parser, &cast->name);
} else {
pm_parser_err_node(parser, node, PM_ERR_WRITE_TARGET_UNEXPECTED);
}
parse_call_operator_write(parser, cast, &token);
pm_node_t *value = parse_assignment_value(parser, previous_binding_power, binding_power, accepts_command_call, PM_ERR_EXPECT_EXPRESSION_AFTER_OPERATOR);
return (pm_node_t *) pm_call_operator_write_node_create(parser, cast, &token, value);
}
case PM_MULTI_WRITE_NODE: {
parser_lex(parser);
pm_parser_err_token(parser, &token, PM_ERR_OPERATOR_MULTI_ASSIGN);
return node;
}
default:
parser_lex(parser);
// In this case we have an operator but we don't know what it's for.
// We need to treat it as an error. For now, we'll mark it as an error
// and just skip right past it.
pm_parser_err_previous(parser, PM_ERR_EXPECT_EXPRESSION_AFTER_OPERATOR);
return node;
}
}
case PM_TOKEN_AMPERSAND_AMPERSAND:
case PM_TOKEN_KEYWORD_AND: {
parser_lex(parser);
pm_node_t *right = parse_expression(parser, binding_power, parser->previous.type == PM_TOKEN_KEYWORD_AND, PM_ERR_EXPECT_EXPRESSION_AFTER_OPERATOR);
return (pm_node_t *) pm_and_node_create(parser, node, &token, right);
}
case PM_TOKEN_KEYWORD_OR:
case PM_TOKEN_PIPE_PIPE: {
parser_lex(parser);
pm_node_t *right = parse_expression(parser, binding_power, parser->previous.type == PM_TOKEN_KEYWORD_OR, PM_ERR_EXPECT_EXPRESSION_AFTER_OPERATOR);
return (pm_node_t *) pm_or_node_create(parser, node, &token, right);
}
case PM_TOKEN_EQUAL_TILDE: {
// Note that we _must_ parse the value before adding the local
// variables in order to properly mirror the behavior of Ruby. For
// example,
//
// /(?<foo>bar)/ =~ foo
//
// In this case, `foo` should be a method call and not a local yet.
parser_lex(parser);
pm_node_t *argument = parse_expression(parser, binding_power, false, PM_ERR_EXPECT_EXPRESSION_AFTER_OPERATOR);
// By default, we're going to create a call node and then return it.
pm_call_node_t *call = pm_call_node_binary_create(parser, node, &token, argument);
pm_node_t *result = (pm_node_t *) call;
// If the receiver of this =~ is a regular expression node, then we
// need to introduce local variables for it based on its named
// capture groups.
if (PM_NODE_TYPE_P(node, PM_INTERPOLATED_REGULAR_EXPRESSION_NODE)) {
// It's possible to have an interpolated regular expression node
// that only contains strings. This is because it can be split
// up by a heredoc. In this case we need to concat the unescaped
// strings together and then parse them as a regular expression.
pm_node_list_t *parts = &((pm_interpolated_regular_expression_node_t *) node)->parts;
bool interpolated = false;
size_t total_length = 0;
for (size_t index = 0; index < parts->size; index++) {
pm_node_t *part = parts->nodes[index];
if (PM_NODE_TYPE_P(part, PM_STRING_NODE)) {
total_length += pm_string_length(&((pm_string_node_t *) part)->unescaped);
} else {
interpolated = true;
break;
}
}
if (!interpolated && total_length > 0) {
void *memory = malloc(total_length);
if (!memory) abort();
uint8_t *cursor = memory;
for (size_t index = 0; index < parts->size; index++) {
pm_string_t *unescaped = &((pm_string_node_t *) parts->nodes[index])->unescaped;
size_t length = pm_string_length(unescaped);
memcpy(cursor, pm_string_source(unescaped), length);
cursor += length;
}
pm_string_t owned;
pm_string_owned_init(&owned, (uint8_t *) memory, total_length);
result = parse_regular_expression_named_captures(parser, &owned, call);
pm_string_free(&owned);
}
} else if (PM_NODE_TYPE_P(node, PM_REGULAR_EXPRESSION_NODE)) {
// If we have a regular expression node, then we can just parse
// the named captures directly off the unescaped string.
const pm_string_t *content = &((pm_regular_expression_node_t *) node)->unescaped;
result = parse_regular_expression_named_captures(parser, content, call);
}
return result;
}
case PM_TOKEN_UAMPERSAND:
case PM_TOKEN_USTAR:
case PM_TOKEN_USTAR_STAR:
// The only times this will occur are when we are in an error state,
// but we'll put them in here so that errors can propagate.
case PM_TOKEN_BANG_EQUAL:
case PM_TOKEN_BANG_TILDE:
case PM_TOKEN_EQUAL_EQUAL:
case PM_TOKEN_EQUAL_EQUAL_EQUAL:
case PM_TOKEN_LESS_EQUAL_GREATER:
case PM_TOKEN_GREATER:
case PM_TOKEN_GREATER_EQUAL:
case PM_TOKEN_LESS:
case PM_TOKEN_LESS_EQUAL:
case PM_TOKEN_CARET:
case PM_TOKEN_PIPE:
case PM_TOKEN_AMPERSAND:
case PM_TOKEN_GREATER_GREATER:
case PM_TOKEN_LESS_LESS:
case PM_TOKEN_MINUS:
case PM_TOKEN_PLUS:
case PM_TOKEN_PERCENT:
case PM_TOKEN_SLASH:
case PM_TOKEN_STAR:
case PM_TOKEN_STAR_STAR: {
parser_lex(parser);
pm_node_t *argument = parse_expression(parser, binding_power, false, PM_ERR_EXPECT_EXPRESSION_AFTER_OPERATOR);
return (pm_node_t *) pm_call_node_binary_create(parser, node, &token, argument);
}
case PM_TOKEN_AMPERSAND_DOT:
case PM_TOKEN_DOT: {
parser_lex(parser);
pm_token_t operator = parser->previous;
pm_arguments_t arguments = { 0 };
// This if statement handles the foo.() syntax.
if (match1(parser, PM_TOKEN_PARENTHESIS_LEFT)) {
parse_arguments_list(parser, &arguments, true, false);
return (pm_node_t *) pm_call_node_shorthand_create(parser, node, &operator, &arguments);
}
pm_token_t message;
switch (parser->current.type) {
case PM_CASE_OPERATOR:
case PM_CASE_KEYWORD:
case PM_TOKEN_CONSTANT:
case PM_TOKEN_IDENTIFIER:
case PM_TOKEN_METHOD_NAME: {
parser_lex(parser);
message = parser->previous;
break;
}
default: {
pm_parser_err_current(parser, PM_ERR_DEF_NAME);
message = (pm_token_t) { .type = PM_TOKEN_MISSING, .start = parser->previous.end, .end = parser->previous.end };
}
}
parse_arguments_list(parser, &arguments, true, accepts_command_call);
pm_call_node_t *call = pm_call_node_call_create(parser, node, &operator, &message, &arguments);
if (
(previous_binding_power == PM_BINDING_POWER_STATEMENT) &&
arguments.arguments == NULL &&
arguments.opening_loc.start == NULL &&
match1(parser, PM_TOKEN_COMMA)
) {
return parse_targets_validate(parser, (pm_node_t *) call, PM_BINDING_POWER_INDEX);
} else {
return (pm_node_t *) call;
}
}
case PM_TOKEN_DOT_DOT:
case PM_TOKEN_DOT_DOT_DOT: {
parser_lex(parser);
pm_node_t *right = NULL;
if (token_begins_expression_p(parser->current.type)) {
right = parse_expression(parser, binding_power, false, PM_ERR_EXPECT_EXPRESSION_AFTER_OPERATOR);
}
return (pm_node_t *) pm_range_node_create(parser, node, &token, right);
}
case PM_TOKEN_KEYWORD_IF_MODIFIER: {
pm_token_t keyword = parser->current;
parser_lex(parser);
pm_node_t *predicate = parse_value_expression(parser, binding_power, true, PM_ERR_CONDITIONAL_IF_PREDICATE);
return (pm_node_t *) pm_if_node_modifier_create(parser, node, &keyword, predicate);
}
case PM_TOKEN_KEYWORD_UNLESS_MODIFIER: {
pm_token_t keyword = parser->current;
parser_lex(parser);
pm_node_t *predicate = parse_value_expression(parser, binding_power, true, PM_ERR_CONDITIONAL_UNLESS_PREDICATE);
return (pm_node_t *) pm_unless_node_modifier_create(parser, node, &keyword, predicate);
}
case PM_TOKEN_KEYWORD_UNTIL_MODIFIER: {
parser_lex(parser);
pm_statements_node_t *statements = pm_statements_node_create(parser);
pm_statements_node_body_append(statements, node);
pm_node_t *predicate = parse_value_expression(parser, binding_power, true, PM_ERR_CONDITIONAL_UNTIL_PREDICATE);
return (pm_node_t *) pm_until_node_modifier_create(parser, &token, predicate, statements, PM_NODE_TYPE_P(node, PM_BEGIN_NODE) ? PM_LOOP_FLAGS_BEGIN_MODIFIER : 0);
}
case PM_TOKEN_KEYWORD_WHILE_MODIFIER: {
parser_lex(parser);
pm_statements_node_t *statements = pm_statements_node_create(parser);
pm_statements_node_body_append(statements, node);
pm_node_t *predicate = parse_value_expression(parser, binding_power, true, PM_ERR_CONDITIONAL_WHILE_PREDICATE);
return (pm_node_t *) pm_while_node_modifier_create(parser, &token, predicate, statements, PM_NODE_TYPE_P(node, PM_BEGIN_NODE) ? PM_LOOP_FLAGS_BEGIN_MODIFIER : 0);
}
case PM_TOKEN_QUESTION_MARK: {
pm_token_t qmark = parser->current;
parser_lex(parser);
pm_node_t *true_expression = parse_expression(parser, PM_BINDING_POWER_DEFINED, false, PM_ERR_TERNARY_EXPRESSION_TRUE);
if (parser->recovering) {
// If parsing the true expression of this ternary resulted in a syntax
// error that we can recover from, then we're going to put missing nodes
// and tokens into the remaining places. We want to be sure to do this
// before the `expect` function call to make sure it doesn't
// accidentally move past a ':' token that occurs after the syntax
// error.
pm_token_t colon = (pm_token_t) { .type = PM_TOKEN_MISSING, .start = parser->previous.end, .end = parser->previous.end };
pm_node_t *false_expression = (pm_node_t *) pm_missing_node_create(parser, colon.start, colon.end);
return (pm_node_t *) pm_if_node_ternary_create(parser, node, &qmark, true_expression, &colon, false_expression);
}
accept1(parser, PM_TOKEN_NEWLINE);
expect1(parser, PM_TOKEN_COLON, PM_ERR_TERNARY_COLON);
pm_token_t colon = parser->previous;
pm_node_t *false_expression = parse_expression(parser, PM_BINDING_POWER_DEFINED, false, PM_ERR_TERNARY_EXPRESSION_FALSE);
return (pm_node_t *) pm_if_node_ternary_create(parser, node, &qmark, true_expression, &colon, false_expression);
}
case PM_TOKEN_COLON_COLON: {
parser_lex(parser);
pm_token_t delimiter = parser->previous;
switch (parser->current.type) {
case PM_TOKEN_CONSTANT: {
parser_lex(parser);
pm_node_t *path;
if (
(parser->current.type == PM_TOKEN_PARENTHESIS_LEFT) ||
(token_begins_expression_p(parser->current.type) || match3(parser, PM_TOKEN_UAMPERSAND, PM_TOKEN_USTAR, PM_TOKEN_USTAR_STAR))
) {
// If we have a constant immediately following a '::' operator, then
// this can either be a constant path or a method call, depending on
// what follows the constant.
//
// If we have parentheses, then this is a method call. That would
// look like Foo::Bar().
pm_token_t message = parser->previous;
pm_arguments_t arguments = { 0 };
parse_arguments_list(parser, &arguments, true, accepts_command_call);
path = (pm_node_t *) pm_call_node_call_create(parser, node, &delimiter, &message, &arguments);
} else {
// Otherwise, this is a constant path. That would look like Foo::Bar.
pm_node_t *child = (pm_node_t *) pm_constant_read_node_create(parser, &parser->previous);
path = (pm_node_t *)pm_constant_path_node_create(parser, node, &delimiter, child);
}
// If this is followed by a comma then it is a multiple assignment.
if (previous_binding_power == PM_BINDING_POWER_STATEMENT && match1(parser, PM_TOKEN_COMMA)) {
return parse_targets_validate(parser, path, PM_BINDING_POWER_INDEX);
}
return path;
}
case PM_CASE_OPERATOR:
case PM_CASE_KEYWORD:
case PM_TOKEN_IDENTIFIER:
case PM_TOKEN_METHOD_NAME: {
parser_lex(parser);
pm_token_t message = parser->previous;
// If we have an identifier following a '::' operator, then it is for
// sure a method call.
pm_arguments_t arguments = { 0 };
parse_arguments_list(parser, &arguments, true, accepts_command_call);
pm_call_node_t *call = pm_call_node_call_create(parser, node, &delimiter, &message, &arguments);
// If this is followed by a comma then it is a multiple assignment.
if (previous_binding_power == PM_BINDING_POWER_STATEMENT && match1(parser, PM_TOKEN_COMMA)) {
return parse_targets_validate(parser, (pm_node_t *) call, PM_BINDING_POWER_INDEX);
}
return (pm_node_t *) call;
}
case PM_TOKEN_PARENTHESIS_LEFT: {
// If we have a parenthesis following a '::' operator, then it is the
// method call shorthand. That would look like Foo::(bar).
pm_arguments_t arguments = { 0 };
parse_arguments_list(parser, &arguments, true, false);
return (pm_node_t *) pm_call_node_shorthand_create(parser, node, &delimiter, &arguments);
}
default: {
pm_parser_err_token(parser, &delimiter, PM_ERR_CONSTANT_PATH_COLON_COLON_CONSTANT);
pm_node_t *child = (pm_node_t *) pm_missing_node_create(parser, delimiter.start, delimiter.end);
return (pm_node_t *)pm_constant_path_node_create(parser, node, &delimiter, child);
}
}
}
case PM_TOKEN_KEYWORD_RESCUE_MODIFIER: {
parser_lex(parser);
accept1(parser, PM_TOKEN_NEWLINE);
pm_node_t *value = parse_expression(parser, binding_power, true, PM_ERR_RESCUE_MODIFIER_VALUE);
return (pm_node_t *) pm_rescue_modifier_node_create(parser, node, &token, value);
}
case PM_TOKEN_BRACKET_LEFT: {
parser_lex(parser);
pm_arguments_t arguments = { 0 };
arguments.opening_loc = PM_LOCATION_TOKEN_VALUE(&parser->previous);
if (!accept1(parser, PM_TOKEN_BRACKET_RIGHT)) {
pm_accepts_block_stack_push(parser, true);
parse_arguments(parser, &arguments, false, PM_TOKEN_BRACKET_RIGHT);
pm_accepts_block_stack_pop(parser);
expect1(parser, PM_TOKEN_BRACKET_RIGHT, PM_ERR_EXPECT_RBRACKET);
}
arguments.closing_loc = PM_LOCATION_TOKEN_VALUE(&parser->previous);
// If we have a comma after the closing bracket then this is a multiple
// assignment and we should parse the targets.
if (previous_binding_power == PM_BINDING_POWER_STATEMENT && match1(parser, PM_TOKEN_COMMA)) {
pm_call_node_t *aref = pm_call_node_aref_create(parser, node, &arguments);
return parse_targets_validate(parser, (pm_node_t *) aref, PM_BINDING_POWER_INDEX);
}
// If we're at the end of the arguments, we can now check if there is a
// block node that starts with a {. If there is, then we can parse it and
// add it to the arguments.
pm_block_node_t *block = NULL;
if (accept1(parser, PM_TOKEN_BRACE_LEFT)) {
block = parse_block(parser);
pm_arguments_validate_block(parser, &arguments, block);
} else if (pm_accepts_block_stack_p(parser) && accept1(parser, PM_TOKEN_KEYWORD_DO)) {
block = parse_block(parser);
}
if (block != NULL) {
if (arguments.block != NULL) {
pm_parser_err_node(parser, (pm_node_t *) block, PM_ERR_ARGUMENT_AFTER_BLOCK);
if (arguments.arguments == NULL) {
arguments.arguments = pm_arguments_node_create(parser);
}
pm_arguments_node_arguments_append(arguments.arguments, arguments.block);
}
arguments.block = (pm_node_t *) block;
}
return (pm_node_t *) pm_call_node_aref_create(parser, node, &arguments);
}
case PM_TOKEN_KEYWORD_IN: {
bool previous_pattern_matching_newlines = parser->pattern_matching_newlines;
parser->pattern_matching_newlines = true;
pm_token_t operator = parser->current;
parser->command_start = false;
lex_state_set(parser, PM_LEX_STATE_BEG | PM_LEX_STATE_LABEL);
parser_lex(parser);
pm_node_t *pattern = parse_pattern(parser, true, PM_ERR_PATTERN_EXPRESSION_AFTER_IN);
parser->pattern_matching_newlines = previous_pattern_matching_newlines;
return (pm_node_t *) pm_match_predicate_node_create(parser, node, pattern, &operator);
}
case PM_TOKEN_EQUAL_GREATER: {
bool previous_pattern_matching_newlines = parser->pattern_matching_newlines;
parser->pattern_matching_newlines = true;
pm_token_t operator = parser->current;
parser->command_start = false;
lex_state_set(parser, PM_LEX_STATE_BEG | PM_LEX_STATE_LABEL);
parser_lex(parser);
pm_node_t *pattern = parse_pattern(parser, true, PM_ERR_PATTERN_EXPRESSION_AFTER_HROCKET);
parser->pattern_matching_newlines = previous_pattern_matching_newlines;
return (pm_node_t *) pm_match_required_node_create(parser, node, pattern, &operator);
}
default:
assert(false && "unreachable");
return NULL;
}
}
/**
* Parse an expression at the given point of the parser using the given binding
* power to parse subsequent chains. If this function finds a syntax error, it
* will append the error message to the parser's error list.
*
* Consumers of this function should always check parser->recovering to
* determine if they need to perform additional cleanup.
*/
static pm_node_t *
parse_expression(pm_parser_t *parser, pm_binding_power_t binding_power, bool accepts_command_call, pm_diagnostic_id_t diag_id) {
pm_token_t recovery = parser->previous;
pm_node_t *node = parse_expression_prefix(parser, binding_power, accepts_command_call);
switch (PM_NODE_TYPE(node)) {
case PM_MISSING_NODE:
// If we found a syntax error, then the type of node returned by
// parse_expression_prefix is going to be a missing node. In that
// case we need to add the error message to the parser's error list.
pm_parser_err(parser, recovery.end, recovery.end, diag_id);
return node;
case PM_PRE_EXECUTION_NODE:
case PM_POST_EXECUTION_NODE:
case PM_ALIAS_GLOBAL_VARIABLE_NODE:
case PM_ALIAS_METHOD_NODE:
case PM_UNDEF_NODE:
// These expressions are statements, and cannot be followed by
// operators (except modifiers).
if (pm_binding_powers[parser->current.type].left > PM_BINDING_POWER_MODIFIER_RESCUE) {
return node;
}
break;
case PM_RANGE_NODE:
// Range operators are non-associative, so that it does not
// associate with other range operators (i.e. `..1..` should be
// rejected.) For this reason, we check such a case for unary ranges
// here, and if so, it returns the node immediately,
if ((((pm_range_node_t *) node)->left == NULL) && pm_binding_powers[parser->current.type].left >= PM_BINDING_POWER_RANGE) {
return node;
}
break;
default:
break;
}
// Otherwise we'll look and see if the next token can be parsed as an infix
// operator. If it can, then we'll parse it using parse_expression_infix.
pm_binding_powers_t current_binding_powers;
while (
current_binding_powers = pm_binding_powers[parser->current.type],
binding_power <= current_binding_powers.left &&
current_binding_powers.binary
) {
node = parse_expression_infix(parser, node, binding_power, current_binding_powers.right, accepts_command_call);
if (current_binding_powers.nonassoc) {
bool endless_range_p = PM_NODE_TYPE_P(node, PM_RANGE_NODE) && ((pm_range_node_t *) node)->right == NULL;
pm_binding_power_t left = endless_range_p ? PM_BINDING_POWER_TERM : current_binding_powers.left;
if (
left <= pm_binding_powers[parser->current.type].left ||
// Exceptionally to operator precedences, '1.. & 2' is rejected.
// '1.. || 2' is also an exception, but it is handled by the lexer.
// (Here, parser->current is PM_TOKEN_PIPE, not PM_TOKEN_PIPE_PIPE).
(endless_range_p && match1(parser, PM_TOKEN_AMPERSAND))
) {
break;
}
}
if (accepts_command_call) {
// A command-style method call is only accepted on method chains.
// Thus, we check whether the parsed node can continue method chains.
// The method chain can continue if the parsed node is one of the following five kinds:
// (1) index access: foo[1]
// (2) attribute access: foo.bar
// (3) method call with parenthesis: foo.bar(1)
// (4) method call with a block: foo.bar do end
// (5) constant path: foo::Bar
switch (node->type) {
case PM_CALL_NODE: {
pm_call_node_t *cast = (pm_call_node_t *)node;
if (
// (1) foo[1]
!(
cast->call_operator_loc.start == NULL &&
cast->message_loc.start != NULL &&
cast->message_loc.start[0] == '[' &&
cast->message_loc.end[-1] == ']'
) &&
// (2) foo.bar
!(
cast->call_operator_loc.start != NULL &&
cast->arguments == NULL &&
cast->block == NULL &&
cast->opening_loc.start == NULL
) &&
// (3) foo.bar(1)
!(
cast->call_operator_loc.start != NULL &&
cast->opening_loc.start != NULL
) &&
// (4) foo.bar do end
!(
cast->block != NULL && PM_NODE_TYPE_P(cast->block, PM_BLOCK_NODE)
)
) {
accepts_command_call = false;
}
break;
}
// (5) foo::Bar
case PM_CONSTANT_PATH_NODE:
break;
default:
accepts_command_call = false;
break;
}
}
}
return node;
}
static pm_node_t *
parse_program(pm_parser_t *parser) {
pm_parser_scope_push(parser, !parser->current_scope);
parser_lex(parser);
pm_statements_node_t *statements = parse_statements(parser, PM_CONTEXT_MAIN);
if (!statements) {
statements = pm_statements_node_create(parser);
}
pm_constant_id_list_t locals = parser->current_scope->locals;
pm_parser_scope_pop(parser);
// If this is an empty file, then we're still going to parse all of the
// statements in order to gather up all of the comments and such. Here we'll
// correct the location information.
if (pm_statements_node_body_length(statements) == 0) {
pm_statements_node_location_set(statements, parser->start, parser->start);
}
return (pm_node_t *) pm_program_node_create(parser, &locals, statements);
}
/******************************************************************************/
/* External functions */
/******************************************************************************/
/**
* Initialize a parser with the given start and end pointers.
*/
PRISM_EXPORTED_FUNCTION void
pm_parser_init(pm_parser_t *parser, const uint8_t *source, size_t size, const pm_options_t *options) {
assert(source != NULL);
*parser = (pm_parser_t) {
.lex_state = PM_LEX_STATE_BEG,
.enclosure_nesting = 0,
.lambda_enclosure_nesting = -1,
.brace_nesting = 0,
.do_loop_stack = 0,
.accepts_block_stack = 0,
.lex_modes = {
.index = 0,
.stack = {{ .mode = PM_LEX_DEFAULT }},
.current = &parser->lex_modes.stack[0],
},
.start = source,
.end = source + size,
.previous = { .type = PM_TOKEN_EOF, .start = source, .end = source },
.current = { .type = PM_TOKEN_EOF, .start = source, .end = source },
.next_start = NULL,
.heredoc_end = NULL,
.comment_list = { 0 },
.magic_comment_list = { 0 },
.warning_list = { 0 },
.error_list = { 0 },
.current_scope = NULL,
.current_context = NULL,
.encoding = PM_ENCODING_UTF_8_ENTRY,
.encoding_changed_callback = NULL,
.encoding_comment_start = source,
.lex_callback = NULL,
.filepath_string = { 0 },
.constant_pool = { 0 },
.newline_list = { 0 },
.integer_base = 0,
.current_string = PM_STRING_EMPTY,
.start_line = 1,
.explicit_encoding = NULL,
.command_start = true,
.recovering = false,
.encoding_changed = false,
.pattern_matching_newlines = false,
.in_keyword_arg = false,
.current_param_name = 0,
.semantic_token_seen = false,
.frozen_string_literal = false,
.suppress_warnings = false
};
// Initialize the constant pool. We're going to completely guess as to the
// number of constants that we'll need based on the size of the input. The
// ratio we chose here is actually less arbitrary than you might think.
//
// We took ~50K Ruby files and measured the size of the file versus the
// number of constants that were found in those files. Then we found the
// average and standard deviation of the ratios of constants/bytesize. Then
// we added 1.34 standard deviations to the average to get a ratio that
// would fit 75% of the files (for a two-tailed distribution). This works
// because there was about a 0.77 correlation and the distribution was
// roughly normal.
//
// This ratio will need to change if we add more constants to the constant
// pool for another node type.
uint32_t constant_size = ((uint32_t) size) / 95;
pm_constant_pool_init(&parser->constant_pool, constant_size < 4 ? 4 : constant_size);
// Initialize the newline list. Similar to the constant pool, we're going to
// guess at the number of newlines that we'll need based on the size of the
// input.
size_t newline_size = size / 22;
pm_newline_list_init(&parser->newline_list, source, newline_size < 4 ? 4 : newline_size);
// If options were provided to this parse, establish them here.
if (options != NULL) {
// filepath option
parser->filepath_string = options->filepath;
// line option
parser->start_line = options->line;
// encoding option
size_t encoding_length = pm_string_length(&options->encoding);
if (encoding_length > 0) {
const uint8_t *encoding_source = pm_string_source(&options->encoding);
parser_lex_magic_comment_encoding_value(parser, encoding_source, encoding_source + encoding_length);
}
// frozen_string_literal option
if (options->frozen_string_literal) {
parser->frozen_string_literal = true;
}
// suppress_warnings option
if (options->suppress_warnings) {
parser->suppress_warnings = true;
}
// scopes option
for (size_t scope_index = 0; scope_index < options->scopes_count; scope_index++) {
const pm_options_scope_t *scope = pm_options_scope_get(options, scope_index);
pm_parser_scope_push(parser, scope_index == 0);
for (size_t local_index = 0; local_index < scope->locals_count; local_index++) {
const pm_string_t *local = pm_options_scope_local_get(scope, local_index);
const uint8_t *source = pm_string_source(local);
size_t length = pm_string_length(local);
uint8_t *allocated = malloc(length);
if (allocated == NULL) continue;
memcpy((void *) allocated, source, length);
pm_parser_local_add_owned(parser, allocated, length);
}
}
}
pm_accepts_block_stack_push(parser, true);
// Skip past the UTF-8 BOM if it exists.
if (size >= 3 && source[0] == 0xef && source[1] == 0xbb && source[2] == 0xbf) {
parser->current.end += 3;
parser->encoding_comment_start += 3;
}
// If the first two bytes of the source are a shebang, then we'll indicate
// that the encoding comment is at the end of the shebang.
if (peek(parser) == '#' && peek_offset(parser, 1) == '!') {
const uint8_t *encoding_comment_start = next_newline(source, (ptrdiff_t) size);
if (encoding_comment_start) {
parser->encoding_comment_start = encoding_comment_start + 1;
}
}
}
/**
* Register a callback that will be called whenever prism changes the encoding
* it is using to parse based on the magic comment.
*/
PRISM_EXPORTED_FUNCTION void
pm_parser_register_encoding_changed_callback(pm_parser_t *parser, pm_encoding_changed_callback_t callback) {
parser->encoding_changed_callback = callback;
}
/**
* Free all of the memory associated with the comment list.
*/
static inline void
pm_comment_list_free(pm_list_t *list) {
pm_list_node_t *node, *next;
for (node = list->head; node != NULL; node = next) {
next = node->next;
pm_comment_t *comment = (pm_comment_t *) node;
free(comment);
}
}
/**
* Free all of the memory associated with the magic comment list.
*/
static inline void
pm_magic_comment_list_free(pm_list_t *list) {
pm_list_node_t *node, *next;
for (node = list->head; node != NULL; node = next) {
next = node->next;
pm_magic_comment_t *magic_comment = (pm_magic_comment_t *) node;
free(magic_comment);
}
}
/**
* Free any memory associated with the given parser.
*/
PRISM_EXPORTED_FUNCTION void
pm_parser_free(pm_parser_t *parser) {
pm_string_free(&parser->filepath_string);
pm_diagnostic_list_free(&parser->error_list);
pm_diagnostic_list_free(&parser->warning_list);
pm_comment_list_free(&parser->comment_list);
pm_magic_comment_list_free(&parser->magic_comment_list);
pm_constant_pool_free(&parser->constant_pool);
pm_newline_list_free(&parser->newline_list);
while (parser->current_scope != NULL) {
// Normally, popping the scope doesn't free the locals since it is
// assumed that ownership has transferred to the AST. However if we have
// scopes while we're freeing the parser, it's likely they came from
// eval scopes and we need to free them explicitly here.
pm_constant_id_list_free(&parser->current_scope->locals);
pm_parser_scope_pop(parser);
}
while (parser->lex_modes.index >= PM_LEX_STACK_SIZE) {
lex_mode_pop(parser);
}
}
/**
* Parse the Ruby source associated with the given parser and return the tree.
*/
PRISM_EXPORTED_FUNCTION pm_node_t *
pm_parse(pm_parser_t *parser) {
return parse_program(parser);
}
static inline void
pm_serialize_header(pm_buffer_t *buffer) {
pm_buffer_append_string(buffer, "PRISM", 5);
pm_buffer_append_byte(buffer, PRISM_VERSION_MAJOR);
pm_buffer_append_byte(buffer, PRISM_VERSION_MINOR);
pm_buffer_append_byte(buffer, PRISM_VERSION_PATCH);
pm_buffer_append_byte(buffer, PRISM_SERIALIZE_ONLY_SEMANTICS_FIELDS ? 1 : 0);
}
/**
* Serialize the AST represented by the given node to the given buffer.
*/
PRISM_EXPORTED_FUNCTION void
pm_serialize(pm_parser_t *parser, pm_node_t *node, pm_buffer_t *buffer) {
pm_serialize_header(buffer);
pm_serialize_content(parser, node, buffer);
pm_buffer_append_byte(buffer, '\0');
}
/**
* Parse and serialize the AST represented by the given source to the given
* buffer.
*/
PRISM_EXPORTED_FUNCTION void
pm_serialize_parse(pm_buffer_t *buffer, const uint8_t *source, size_t size, const char *data) {
pm_options_t options = { 0 };
pm_options_read(&options, data);
pm_parser_t parser;
pm_parser_init(&parser, source, size, &options);
pm_node_t *node = pm_parse(&parser);
pm_serialize_header(buffer);
pm_serialize_content(&parser, node, buffer);
pm_buffer_append_byte(buffer, '\0');
pm_node_destroy(&parser, node);
pm_parser_free(&parser);
pm_options_free(&options);
}
/**
* Parse and serialize the comments in the given source to the given buffer.
*/
PRISM_EXPORTED_FUNCTION void
pm_serialize_parse_comments(pm_buffer_t *buffer, const uint8_t *source, size_t size, const char *data) {
pm_options_t options = { 0 };
pm_options_read(&options, data);
pm_parser_t parser;
pm_parser_init(&parser, source, size, &options);
pm_node_t *node = pm_parse(&parser);
pm_serialize_header(buffer);
pm_serialize_encoding(parser.encoding, buffer);
pm_buffer_append_varsint(buffer, parser.start_line);
pm_serialize_comment_list(&parser, &parser.comment_list, buffer);
pm_node_destroy(&parser, node);
pm_parser_free(&parser);
pm_options_free(&options);
}
#undef PM_CASE_KEYWORD
#undef PM_CASE_OPERATOR
#undef PM_CASE_WRITABLE
#undef PM_STRING_EMPTY
#undef PM_LOCATION_NODE_BASE_VALUE
#undef PM_LOCATION_NODE_VALUE
#undef PM_LOCATION_NULL_VALUE
#undef PM_LOCATION_TOKEN_VALUE
|