File size: 98,438 Bytes
2c8dc05 |
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 |
# -----------------------------------------------------------------------------
# ply: yacc.py
#
# Copyright (C) 2001-2022
# David M. Beazley (Dabeaz LLC)
# All rights reserved.
#
# Latest version: https://github.com/dabeaz/ply
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are
# met:
#
# * Redistributions of source code must retain the above copyright notice,
# this list of conditions and the following disclaimer.
# * Redistributions in binary form must reproduce the above copyright notice,
# this list of conditions and the following disclaimer in the documentation
# and/or other materials provided with the distribution.
# * Neither the name of David Beazley or Dabeaz LLC may be used to
# endorse or promote products derived from this software without
# specific prior written permission.
#
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
# -----------------------------------------------------------------------------
#
# This implements an LR parser that is constructed from grammar rules defined
# as Python functions. The grammar is specified by supplying the BNF inside
# Python documentation strings. The inspiration for this technique was borrowed
# from John Aycock's Spark parsing system. PLY might be viewed as cross between
# Spark and the GNU bison utility.
#
# The current implementation is only somewhat object-oriented. The
# LR parser itself is defined in terms of an object (which allows multiple
# parsers to co-exist). However, most of the variables used during table
# construction are defined in terms of global variables. Users shouldn't
# notice unless they are trying to define multiple parsers at the same
# time using threads (in which case they should have their head examined).
#
# This implementation supports both SLR and LALR(1) parsing. LALR(1)
# support was originally implemented by Elias Ioup ([email protected]),
# using the algorithm found in Aho, Sethi, and Ullman "Compilers: Principles,
# Techniques, and Tools" (The Dragon Book). LALR(1) has since been replaced
# by the more efficient DeRemer and Pennello algorithm.
#
# :::::::: WARNING :::::::
#
# Construction of LR parsing tables is fairly complicated and expensive.
# To make this module run fast, a *LOT* of work has been put into
# optimization---often at the expensive of readability and what might
# consider to be good Python "coding style." Modify the code at your
# own risk!
# ----------------------------------------------------------------------------
import re
import types
import sys
import inspect
#-----------------------------------------------------------------------------
# === User configurable parameters ===
#
# Change these to modify the default behavior of yacc (if you wish)
#-----------------------------------------------------------------------------
yaccdebug = False # Debugging mode. If set, yacc generates a
# a 'parser.out' file in the current directory
debug_file = 'parser.out' # Default name of the debugging file
error_count = 3 # Number of symbols that must be shifted to leave recovery mode
resultlimit = 40 # Size limit of results when running in debug mode.
MAXINT = sys.maxsize
# This object is a stand-in for a logging object created by the
# logging module. PLY will use this by default to create things
# such as the parser.out file. If a user wants more detailed
# information, they can create their own logging object and pass
# it into PLY.
class PlyLogger(object):
def __init__(self, f):
self.f = f
def debug(self, msg, *args, **kwargs):
self.f.write((msg % args) + '\n')
info = debug
def warning(self, msg, *args, **kwargs):
self.f.write('WARNING: ' + (msg % args) + '\n')
def error(self, msg, *args, **kwargs):
self.f.write('ERROR: ' + (msg % args) + '\n')
critical = debug
# Null logger is used when no output is generated. Does nothing.
class NullLogger(object):
def __getattribute__(self, name):
return self
def __call__(self, *args, **kwargs):
return self
# Exception raised for yacc-related errors
class YaccError(Exception):
pass
# Format the result message that the parser produces when running in debug mode.
def format_result(r):
repr_str = repr(r)
if '\n' in repr_str:
repr_str = repr(repr_str)
if len(repr_str) > resultlimit:
repr_str = repr_str[:resultlimit] + ' ...'
result = '<%s @ 0x%x> (%s)' % (type(r).__name__, id(r), repr_str)
return result
# Format stack entries when the parser is running in debug mode
def format_stack_entry(r):
repr_str = repr(r)
if '\n' in repr_str:
repr_str = repr(repr_str)
if len(repr_str) < 16:
return repr_str
else:
return '<%s @ 0x%x>' % (type(r).__name__, id(r))
#-----------------------------------------------------------------------------
# === LR Parsing Engine ===
#
# The following classes are used for the LR parser itself. These are not
# used during table construction and are independent of the actual LR
# table generation algorithm
#-----------------------------------------------------------------------------
# This class is used to hold non-terminal grammar symbols during parsing.
# It normally has the following attributes set:
# .type = Grammar symbol type
# .value = Symbol value
# .lineno = Starting line number
# .endlineno = Ending line number (optional, set automatically)
# .lexpos = Starting lex position
# .endlexpos = Ending lex position (optional, set automatically)
class YaccSymbol:
def __str__(self):
return self.type
def __repr__(self):
return str(self)
# This class is a wrapper around the objects actually passed to each
# grammar rule. Index lookup and assignment actually assign the
# .value attribute of the underlying YaccSymbol object.
# The lineno() method returns the line number of a given
# item (or 0 if not defined). The linespan() method returns
# a tuple of (startline,endline) representing the range of lines
# for a symbol. The lexspan() method returns a tuple (lexpos,endlexpos)
# representing the range of positional information for a symbol.
class YaccProduction:
def __init__(self, s, stack=None):
self.slice = s
self.stack = stack
self.lexer = None
self.parser = None
def __getitem__(self, n):
if isinstance(n, slice):
return [s.value for s in self.slice[n]]
elif n >= 0:
return self.slice[n].value
else:
return self.stack[n].value
def __setitem__(self, n, v):
self.slice[n].value = v
def __getslice__(self, i, j):
return [s.value for s in self.slice[i:j]]
def __len__(self):
return len(self.slice)
def lineno(self, n):
return getattr(self.slice[n], 'lineno', 0)
def set_lineno(self, n, lineno):
self.slice[n].lineno = lineno
def linespan(self, n):
startline = getattr(self.slice[n], 'lineno', 0)
endline = getattr(self.slice[n], 'endlineno', startline)
return startline, endline
def lexpos(self, n):
return getattr(self.slice[n], 'lexpos', 0)
def set_lexpos(self, n, lexpos):
self.slice[n].lexpos = lexpos
def lexspan(self, n):
startpos = getattr(self.slice[n], 'lexpos', 0)
endpos = getattr(self.slice[n], 'endlexpos', startpos)
return startpos, endpos
def error(self):
raise SyntaxError
# -----------------------------------------------------------------------------
# == LRParser ==
#
# The LR Parsing engine.
# -----------------------------------------------------------------------------
class LRParser:
def __init__(self, lrtab, errorf):
self.productions = lrtab.lr_productions
self.action = lrtab.lr_action
self.goto = lrtab.lr_goto
self.errorfunc = errorf
self.set_defaulted_states()
self.errorok = True
def errok(self):
self.errorok = True
def restart(self):
del self.statestack[:]
del self.symstack[:]
sym = YaccSymbol()
sym.type = '$end'
self.symstack.append(sym)
self.statestack.append(0)
# Defaulted state support.
# This method identifies parser states where there is only one possible reduction action.
# For such states, the parser can make a choose to make a rule reduction without consuming
# the next look-ahead token. This delayed invocation of the tokenizer can be useful in
# certain kinds of advanced parsing situations where the lexer and parser interact with
# each other or change states (i.e., manipulation of scope, lexer states, etc.).
#
# See: http://www.gnu.org/software/bison/manual/html_node/Default-Reductions.html#Default-Reductions
def set_defaulted_states(self):
self.defaulted_states = {}
for state, actions in self.action.items():
rules = list(actions.values())
if len(rules) == 1 and rules[0] < 0:
self.defaulted_states[state] = rules[0]
def disable_defaulted_states(self):
self.defaulted_states = {}
# parse().
#
# This is the core parsing engine. To operate, it requires a lexer object.
# Two options are provided. The debug flag turns on debugging so that you can
# see the various rule reductions and parsing steps. tracking turns on position
# tracking. In this mode, symbols will record the starting/ending line number and
# character index.
def parse(self, input=None, lexer=None, debug=False, tracking=False):
# If debugging has been specified as a flag, turn it into a logging object
if isinstance(debug, int) and debug:
debug = PlyLogger(sys.stderr)
lookahead = None # Current lookahead symbol
lookaheadstack = [] # Stack of lookahead symbols
actions = self.action # Local reference to action table (to avoid lookup on self.)
goto = self.goto # Local reference to goto table (to avoid lookup on self.)
prod = self.productions # Local reference to production list (to avoid lookup on self.)
defaulted_states = self.defaulted_states # Local reference to defaulted states
pslice = YaccProduction(None) # Production object passed to grammar rules
errorcount = 0 # Used during error recovery
if debug:
debug.info('PLY: PARSE DEBUG START')
# If no lexer was given, we will try to use the lex module
if not lexer:
from . import lex
lexer = lex.lexer
# Set up the lexer and parser objects on pslice
pslice.lexer = lexer
pslice.parser = self
# If input was supplied, pass to lexer
if input is not None:
lexer.input(input)
# Set the token function
get_token = self.token = lexer.token
# Set up the state and symbol stacks
statestack = self.statestack = [] # Stack of parsing states
symstack = self.symstack = [] # Stack of grammar symbols
pslice.stack = symstack # Put in the production
errtoken = None # Err token
# The start state is assumed to be (0,$end)
statestack.append(0)
sym = YaccSymbol()
sym.type = '$end'
symstack.append(sym)
state = 0
while True:
# Get the next symbol on the input. If a lookahead symbol
# is already set, we just use that. Otherwise, we'll pull
# the next token off of the lookaheadstack or from the lexer
if debug:
debug.debug('State : %s', state)
if state not in defaulted_states:
if not lookahead:
if not lookaheadstack:
lookahead = get_token() # Get the next token
else:
lookahead = lookaheadstack.pop()
if not lookahead:
lookahead = YaccSymbol()
lookahead.type = '$end'
# Check the action table
ltype = lookahead.type
t = actions[state].get(ltype)
else:
t = defaulted_states[state]
if debug:
debug.debug('Defaulted state %s: Reduce using %d', state, -t)
if debug:
debug.debug('Stack : %s',
('%s . %s' % (' '.join([xx.type for xx in symstack][1:]), str(lookahead))).lstrip())
if t is not None:
if t > 0:
# shift a symbol on the stack
statestack.append(t)
state = t
if debug:
debug.debug('Action : Shift and goto state %s', t)
symstack.append(lookahead)
lookahead = None
# Decrease error count on successful shift
if errorcount:
errorcount -= 1
continue
if t < 0:
# reduce a symbol on the stack, emit a production
p = prod[-t]
pname = p.name
plen = p.len
# Get production function
sym = YaccSymbol()
sym.type = pname # Production name
sym.value = None
if debug:
if plen:
debug.info('Action : Reduce rule [%s] with %s and goto state %d', p.str,
'['+','.join([format_stack_entry(_v.value) for _v in symstack[-plen:]])+']',
goto[statestack[-1-plen]][pname])
else:
debug.info('Action : Reduce rule [%s] with %s and goto state %d', p.str, [],
goto[statestack[-1]][pname])
if plen:
targ = symstack[-plen-1:]
targ[0] = sym
if tracking:
t1 = targ[1]
sym.lineno = t1.lineno
sym.lexpos = t1.lexpos
t1 = targ[-1]
sym.endlineno = getattr(t1, 'endlineno', t1.lineno)
sym.endlexpos = getattr(t1, 'endlexpos', t1.lexpos)
# !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
# The code enclosed in this section is duplicated
# below as a performance optimization. Make sure
# changes get made in both locations.
pslice.slice = targ
try:
# Call the grammar rule with our special slice object
del symstack[-plen:]
self.state = state
p.callable(pslice)
del statestack[-plen:]
if debug:
debug.info('Result : %s', format_result(pslice[0]))
symstack.append(sym)
state = goto[statestack[-1]][pname]
statestack.append(state)
except SyntaxError:
# If an error was set. Enter error recovery state
lookaheadstack.append(lookahead) # Save the current lookahead token
symstack.extend(targ[1:-1]) # Put the production slice back on the stack
statestack.pop() # Pop back one state (before the reduce)
state = statestack[-1]
sym.type = 'error'
sym.value = 'error'
lookahead = sym
errorcount = error_count
self.errorok = False
continue
else:
if tracking:
sym.lineno = lexer.lineno
sym.lexpos = lexer.lexpos
targ = [sym]
# !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
# The code enclosed in this section is duplicated
# above as a performance optimization. Make sure
# changes get made in both locations.
pslice.slice = targ
try:
# Call the grammar rule with our special slice object
self.state = state
p.callable(pslice)
if debug:
debug.info('Result : %s', format_result(pslice[0]))
symstack.append(sym)
state = goto[statestack[-1]][pname]
statestack.append(state)
except SyntaxError:
# If an error was set. Enter error recovery state
lookaheadstack.append(lookahead) # Save the current lookahead token
statestack.pop() # Pop back one state (before the reduce)
state = statestack[-1]
sym.type = 'error'
sym.value = 'error'
lookahead = sym
errorcount = error_count
self.errorok = False
continue
if t == 0:
n = symstack[-1]
result = getattr(n, 'value', None)
if debug:
debug.info('Done : Returning %s', format_result(result))
debug.info('PLY: PARSE DEBUG END')
return result
if t is None:
if debug:
debug.error('Error : %s',
('%s . %s' % (' '.join([xx.type for xx in symstack][1:]), str(lookahead))).lstrip())
# We have some kind of parsing error here. To handle
# this, we are going to push the current token onto
# the tokenstack and replace it with an 'error' token.
# If there are any synchronization rules, they may
# catch it.
#
# In addition to pushing the error token, we call call
# the user defined p_error() function if this is the
# first syntax error. This function is only called if
# errorcount == 0.
if errorcount == 0 or self.errorok:
errorcount = error_count
self.errorok = False
errtoken = lookahead
if errtoken.type == '$end':
errtoken = None # End of file!
if self.errorfunc:
if errtoken and not hasattr(errtoken, 'lexer'):
errtoken.lexer = lexer
self.state = state
tok = self.errorfunc(errtoken)
if self.errorok:
# User must have done some kind of panic
# mode recovery on their own. The
# returned token is the next lookahead
lookahead = tok
errtoken = None
continue
else:
if errtoken:
if hasattr(errtoken, 'lineno'):
lineno = lookahead.lineno
else:
lineno = 0
if lineno:
sys.stderr.write('yacc: Syntax error at line %d, token=%s\n' % (lineno, errtoken.type))
else:
sys.stderr.write('yacc: Syntax error, token=%s' % errtoken.type)
else:
sys.stderr.write('yacc: Parse error in input. EOF\n')
return
else:
errorcount = error_count
# case 1: the statestack only has 1 entry on it. If we're in this state, the
# entire parse has been rolled back and we're completely hosed. The token is
# discarded and we just keep going.
if len(statestack) <= 1 and lookahead.type != '$end':
lookahead = None
errtoken = None
state = 0
# Nuke the pushback stack
del lookaheadstack[:]
continue
# case 2: the statestack has a couple of entries on it, but we're
# at the end of the file. nuke the top entry and generate an error token
# Start nuking entries on the stack
if lookahead.type == '$end':
# Whoa. We're really hosed here. Bail out
return
if lookahead.type != 'error':
sym = symstack[-1]
if sym.type == 'error':
# Hmmm. Error is on top of stack, we'll just nuke input
# symbol and continue
if tracking:
sym.endlineno = getattr(lookahead, 'lineno', sym.lineno)
sym.endlexpos = getattr(lookahead, 'lexpos', sym.lexpos)
lookahead = None
continue
# Create the error symbol for the first time and make it the new lookahead symbol
t = YaccSymbol()
t.type = 'error'
if hasattr(lookahead, 'lineno'):
t.lineno = t.endlineno = lookahead.lineno
if hasattr(lookahead, 'lexpos'):
t.lexpos = t.endlexpos = lookahead.lexpos
t.value = lookahead
lookaheadstack.append(lookahead)
lookahead = t
else:
sym = symstack.pop()
if tracking:
lookahead.lineno = sym.lineno
lookahead.lexpos = sym.lexpos
statestack.pop()
state = statestack[-1]
continue
# If we'r here, something really bad happened
raise RuntimeError('yacc: internal parser error!!!\n')
# -----------------------------------------------------------------------------
# === Grammar Representation ===
#
# The following functions, classes, and variables are used to represent and
# manipulate the rules that make up a grammar.
# -----------------------------------------------------------------------------
# regex matching identifiers
_is_identifier = re.compile(r'^[a-zA-Z0-9_-]+$')
# -----------------------------------------------------------------------------
# class Production:
#
# This class stores the raw information about a single production or grammar rule.
# A grammar rule refers to a specification such as this:
#
# expr : expr PLUS term
#
# Here are the basic attributes defined on all productions
#
# name - Name of the production. For example 'expr'
# prod - A list of symbols on the right side ['expr','PLUS','term']
# prec - Production precedence level
# number - Production number.
# func - Function that executes on reduce
# file - File where production function is defined
# lineno - Line number where production function is defined
#
# The following attributes are defined or optional.
#
# len - Length of the production (number of symbols on right hand side)
# usyms - Set of unique symbols found in the production
# -----------------------------------------------------------------------------
class Production(object):
reduced = 0
def __init__(self, number, name, prod, precedence=('right', 0), func=None, file='', line=0):
self.name = name
self.prod = tuple(prod)
self.number = number
self.func = func
self.callable = None
self.file = file
self.line = line
self.prec = precedence
# Internal settings used during table construction
self.len = len(self.prod) # Length of the production
# Create a list of unique production symbols used in the production
self.usyms = []
for s in self.prod:
if s not in self.usyms:
self.usyms.append(s)
# List of all LR items for the production
self.lr_items = []
self.lr_next = None
# Create a string representation
if self.prod:
self.str = '%s -> %s' % (self.name, ' '.join(self.prod))
else:
self.str = '%s -> <empty>' % self.name
def __str__(self):
return self.str
def __repr__(self):
return 'Production(' + str(self) + ')'
def __len__(self):
return len(self.prod)
def __nonzero__(self):
return 1
def __getitem__(self, index):
return self.prod[index]
# Return the nth lr_item from the production (or None if at the end)
def lr_item(self, n):
if n > len(self.prod):
return None
p = LRItem(self, n)
# Precompute the list of productions immediately following.
try:
p.lr_after = self.Prodnames[p.prod[n+1]]
except (IndexError, KeyError):
p.lr_after = []
try:
p.lr_before = p.prod[n-1]
except IndexError:
p.lr_before = None
return p
# Bind the production function name to a callable
def bind(self, pdict):
if self.func:
self.callable = pdict[self.func]
# -----------------------------------------------------------------------------
# class LRItem
#
# This class represents a specific stage of parsing a production rule. For
# example:
#
# expr : expr . PLUS term
#
# In the above, the "." represents the current location of the parse. Here
# basic attributes:
#
# name - Name of the production. For example 'expr'
# prod - A list of symbols on the right side ['expr','.', 'PLUS','term']
# number - Production number.
#
# lr_next Next LR item. Example, if we are ' expr -> expr . PLUS term'
# then lr_next refers to 'expr -> expr PLUS . term'
# lr_index - LR item index (location of the ".") in the prod list.
# lookaheads - LALR lookahead symbols for this item
# len - Length of the production (number of symbols on right hand side)
# lr_after - List of all productions that immediately follow
# lr_before - Grammar symbol immediately before
# -----------------------------------------------------------------------------
class LRItem(object):
def __init__(self, p, n):
self.name = p.name
self.prod = list(p.prod)
self.number = p.number
self.lr_index = n
self.lookaheads = {}
self.prod.insert(n, '.')
self.prod = tuple(self.prod)
self.len = len(self.prod)
self.usyms = p.usyms
def __str__(self):
if self.prod:
s = '%s -> %s' % (self.name, ' '.join(self.prod))
else:
s = '%s -> <empty>' % self.name
return s
def __repr__(self):
return 'LRItem(' + str(self) + ')'
# -----------------------------------------------------------------------------
# rightmost_terminal()
#
# Return the rightmost terminal from a list of symbols. Used in add_production()
# -----------------------------------------------------------------------------
def rightmost_terminal(symbols, terminals):
i = len(symbols) - 1
while i >= 0:
if symbols[i] in terminals:
return symbols[i]
i -= 1
return None
# -----------------------------------------------------------------------------
# === GRAMMAR CLASS ===
#
# The following class represents the contents of the specified grammar along
# with various computed properties such as first sets, follow sets, LR items, etc.
# This data is used for critical parts of the table generation process later.
# -----------------------------------------------------------------------------
class GrammarError(YaccError):
pass
class Grammar(object):
def __init__(self, terminals):
self.Productions = [None] # A list of all of the productions. The first
# entry is always reserved for the purpose of
# building an augmented grammar
self.Prodnames = {} # A dictionary mapping the names of nonterminals to a list of all
# productions of that nonterminal.
self.Prodmap = {} # A dictionary that is only used to detect duplicate
# productions.
self.Terminals = {} # A dictionary mapping the names of terminal symbols to a
# list of the rules where they are used.
for term in terminals:
self.Terminals[term] = []
self.Terminals['error'] = []
self.Nonterminals = {} # A dictionary mapping names of nonterminals to a list
# of rule numbers where they are used.
self.First = {} # A dictionary of precomputed FIRST(x) symbols
self.Follow = {} # A dictionary of precomputed FOLLOW(x) symbols
self.Precedence = {} # Precedence rules for each terminal. Contains tuples of the
# form ('right',level) or ('nonassoc', level) or ('left',level)
self.UsedPrecedence = set() # Precedence rules that were actually used by the grammer.
# This is only used to provide error checking and to generate
# a warning about unused precedence rules.
self.Start = None # Starting symbol for the grammar
def __len__(self):
return len(self.Productions)
def __getitem__(self, index):
return self.Productions[index]
# -----------------------------------------------------------------------------
# set_precedence()
#
# Sets the precedence for a given terminal. assoc is the associativity such as
# 'left','right', or 'nonassoc'. level is a numeric level.
#
# -----------------------------------------------------------------------------
def set_precedence(self, term, assoc, level):
assert self.Productions == [None], 'Must call set_precedence() before add_production()'
if term in self.Precedence:
raise GrammarError('Precedence already specified for terminal %r' % term)
if assoc not in ['left', 'right', 'nonassoc']:
raise GrammarError("Associativity must be one of 'left','right', or 'nonassoc'")
self.Precedence[term] = (assoc, level)
# -----------------------------------------------------------------------------
# add_production()
#
# Given an action function, this function assembles a production rule and
# computes its precedence level.
#
# The production rule is supplied as a list of symbols. For example,
# a rule such as 'expr : expr PLUS term' has a production name of 'expr' and
# symbols ['expr','PLUS','term'].
#
# Precedence is determined by the precedence of the right-most non-terminal
# or the precedence of a terminal specified by %prec.
#
# A variety of error checks are performed to make sure production symbols
# are valid and that %prec is used correctly.
# -----------------------------------------------------------------------------
def add_production(self, prodname, syms, func=None, file='', line=0):
if prodname in self.Terminals:
raise GrammarError('%s:%d: Illegal rule name %r. Already defined as a token' % (file, line, prodname))
if prodname == 'error':
raise GrammarError('%s:%d: Illegal rule name %r. error is a reserved word' % (file, line, prodname))
if not _is_identifier.match(prodname):
raise GrammarError('%s:%d: Illegal rule name %r' % (file, line, prodname))
# Look for literal tokens
for n, s in enumerate(syms):
if s[0] in "'\"":
try:
c = eval(s)
if (len(c) > 1):
raise GrammarError('%s:%d: Literal token %s in rule %r may only be a single character' %
(file, line, s, prodname))
if c not in self.Terminals:
self.Terminals[c] = []
syms[n] = c
continue
except SyntaxError:
pass
if not _is_identifier.match(s) and s != '%prec':
raise GrammarError('%s:%d: Illegal name %r in rule %r' % (file, line, s, prodname))
# Determine the precedence level
if '%prec' in syms:
if syms[-1] == '%prec':
raise GrammarError('%s:%d: Syntax error. Nothing follows %%prec' % (file, line))
if syms[-2] != '%prec':
raise GrammarError('%s:%d: Syntax error. %%prec can only appear at the end of a grammar rule' %
(file, line))
precname = syms[-1]
prodprec = self.Precedence.get(precname)
if not prodprec:
raise GrammarError('%s:%d: Nothing known about the precedence of %r' % (file, line, precname))
else:
self.UsedPrecedence.add(precname)
del syms[-2:] # Drop %prec from the rule
else:
# If no %prec, precedence is determined by the rightmost terminal symbol
precname = rightmost_terminal(syms, self.Terminals)
prodprec = self.Precedence.get(precname, ('right', 0))
# See if the rule is already in the rulemap
map = '%s -> %s' % (prodname, syms)
if map in self.Prodmap:
m = self.Prodmap[map]
raise GrammarError('%s:%d: Duplicate rule %s. ' % (file, line, m) +
'Previous definition at %s:%d' % (m.file, m.line))
# From this point on, everything is valid. Create a new Production instance
pnumber = len(self.Productions)
if prodname not in self.Nonterminals:
self.Nonterminals[prodname] = []
# Add the production number to Terminals and Nonterminals
for t in syms:
if t in self.Terminals:
self.Terminals[t].append(pnumber)
else:
if t not in self.Nonterminals:
self.Nonterminals[t] = []
self.Nonterminals[t].append(pnumber)
# Create a production and add it to the list of productions
p = Production(pnumber, prodname, syms, prodprec, func, file, line)
self.Productions.append(p)
self.Prodmap[map] = p
# Add to the global productions list
try:
self.Prodnames[prodname].append(p)
except KeyError:
self.Prodnames[prodname] = [p]
# -----------------------------------------------------------------------------
# set_start()
#
# Sets the starting symbol and creates the augmented grammar. Production
# rule 0 is S' -> start where start is the start symbol.
# -----------------------------------------------------------------------------
def set_start(self, start=None):
if not start:
start = self.Productions[1].name
if start not in self.Nonterminals:
raise GrammarError('start symbol %s undefined' % start)
self.Productions[0] = Production(0, "S'", [start])
self.Nonterminals[start].append(0)
self.Start = start
# -----------------------------------------------------------------------------
# find_unreachable()
#
# Find all of the nonterminal symbols that can't be reached from the starting
# symbol. Returns a list of nonterminals that can't be reached.
# -----------------------------------------------------------------------------
def find_unreachable(self):
# Mark all symbols that are reachable from a symbol s
def mark_reachable_from(s):
if s in reachable:
return
reachable.add(s)
for p in self.Prodnames.get(s, []):
for r in p.prod:
mark_reachable_from(r)
reachable = set()
mark_reachable_from(self.Productions[0].prod[0])
return [s for s in self.Nonterminals if s not in reachable]
# -----------------------------------------------------------------------------
# infinite_cycles()
#
# This function looks at the various parsing rules and tries to detect
# infinite recursion cycles (grammar rules where there is no possible way
# to derive a string of only terminals).
# -----------------------------------------------------------------------------
def infinite_cycles(self):
terminates = {}
# Terminals:
for t in self.Terminals:
terminates[t] = True
terminates['$end'] = True
# Nonterminals:
# Initialize to false:
for n in self.Nonterminals:
terminates[n] = False
# Then propagate termination until no change:
while True:
some_change = False
for (n, pl) in self.Prodnames.items():
# Nonterminal n terminates iff any of its productions terminates.
for p in pl:
# Production p terminates iff all of its rhs symbols terminate.
for s in p.prod:
if not terminates[s]:
# The symbol s does not terminate,
# so production p does not terminate.
p_terminates = False
break
else:
# didn't break from the loop,
# so every symbol s terminates
# so production p terminates.
p_terminates = True
if p_terminates:
# symbol n terminates!
if not terminates[n]:
terminates[n] = True
some_change = True
# Don't need to consider any more productions for this n.
break
if not some_change:
break
infinite = []
for (s, term) in terminates.items():
if not term:
if s not in self.Prodnames and s not in self.Terminals and s != 'error':
# s is used-but-not-defined, and we've already warned of that,
# so it would be overkill to say that it's also non-terminating.
pass
else:
infinite.append(s)
return infinite
# -----------------------------------------------------------------------------
# undefined_symbols()
#
# Find all symbols that were used the grammar, but not defined as tokens or
# grammar rules. Returns a list of tuples (sym, prod) where sym in the symbol
# and prod is the production where the symbol was used.
# -----------------------------------------------------------------------------
def undefined_symbols(self):
result = []
for p in self.Productions:
if not p:
continue
for s in p.prod:
if s not in self.Prodnames and s not in self.Terminals and s != 'error':
result.append((s, p))
return result
# -----------------------------------------------------------------------------
# unused_terminals()
#
# Find all terminals that were defined, but not used by the grammar. Returns
# a list of all symbols.
# -----------------------------------------------------------------------------
def unused_terminals(self):
unused_tok = []
for s, v in self.Terminals.items():
if s != 'error' and not v:
unused_tok.append(s)
return unused_tok
# ------------------------------------------------------------------------------
# unused_rules()
#
# Find all grammar rules that were defined, but not used (maybe not reachable)
# Returns a list of productions.
# ------------------------------------------------------------------------------
def unused_rules(self):
unused_prod = []
for s, v in self.Nonterminals.items():
if not v:
p = self.Prodnames[s][0]
unused_prod.append(p)
return unused_prod
# -----------------------------------------------------------------------------
# unused_precedence()
#
# Returns a list of tuples (term,precedence) corresponding to precedence
# rules that were never used by the grammar. term is the name of the terminal
# on which precedence was applied and precedence is a string such as 'left' or
# 'right' corresponding to the type of precedence.
# -----------------------------------------------------------------------------
def unused_precedence(self):
unused = []
for termname in self.Precedence:
if not (termname in self.Terminals or termname in self.UsedPrecedence):
unused.append((termname, self.Precedence[termname][0]))
return unused
# -------------------------------------------------------------------------
# _first()
#
# Compute the value of FIRST1(beta) where beta is a tuple of symbols.
#
# During execution of compute_first1, the result may be incomplete.
# Afterward (e.g., when called from compute_follow()), it will be complete.
# -------------------------------------------------------------------------
def _first(self, beta):
# We are computing First(x1,x2,x3,...,xn)
result = []
for x in beta:
x_produces_empty = False
# Add all the non-<empty> symbols of First[x] to the result.
for f in self.First[x]:
if f == '<empty>':
x_produces_empty = True
else:
if f not in result:
result.append(f)
if x_produces_empty:
# We have to consider the next x in beta,
# i.e. stay in the loop.
pass
else:
# We don't have to consider any further symbols in beta.
break
else:
# There was no 'break' from the loop,
# so x_produces_empty was true for all x in beta,
# so beta produces empty as well.
result.append('<empty>')
return result
# -------------------------------------------------------------------------
# compute_first()
#
# Compute the value of FIRST1(X) for all symbols
# -------------------------------------------------------------------------
def compute_first(self):
if self.First:
return self.First
# Terminals:
for t in self.Terminals:
self.First[t] = [t]
self.First['$end'] = ['$end']
# Nonterminals:
# Initialize to the empty set:
for n in self.Nonterminals:
self.First[n] = []
# Then propagate symbols until no change:
while True:
some_change = False
for n in self.Nonterminals:
for p in self.Prodnames[n]:
for f in self._first(p.prod):
if f not in self.First[n]:
self.First[n].append(f)
some_change = True
if not some_change:
break
return self.First
# ---------------------------------------------------------------------
# compute_follow()
#
# Computes all of the follow sets for every non-terminal symbol. The
# follow set is the set of all symbols that might follow a given
# non-terminal. See the Dragon book, 2nd Ed. p. 189.
# ---------------------------------------------------------------------
def compute_follow(self, start=None):
# If already computed, return the result
if self.Follow:
return self.Follow
# If first sets not computed yet, do that first.
if not self.First:
self.compute_first()
# Add '$end' to the follow list of the start symbol
for k in self.Nonterminals:
self.Follow[k] = []
if not start:
start = self.Productions[1].name
self.Follow[start] = ['$end']
while True:
didadd = False
for p in self.Productions[1:]:
# Here is the production set
for i, B in enumerate(p.prod):
if B in self.Nonterminals:
# Okay. We got a non-terminal in a production
fst = self._first(p.prod[i+1:])
hasempty = False
for f in fst:
if f != '<empty>' and f not in self.Follow[B]:
self.Follow[B].append(f)
didadd = True
if f == '<empty>':
hasempty = True
if hasempty or i == (len(p.prod)-1):
# Add elements of follow(a) to follow(b)
for f in self.Follow[p.name]:
if f not in self.Follow[B]:
self.Follow[B].append(f)
didadd = True
if not didadd:
break
return self.Follow
# -----------------------------------------------------------------------------
# build_lritems()
#
# This function walks the list of productions and builds a complete set of the
# LR items. The LR items are stored in two ways: First, they are uniquely
# numbered and placed in the list _lritems. Second, a linked list of LR items
# is built for each production. For example:
#
# E -> E PLUS E
#
# Creates the list
#
# [E -> . E PLUS E, E -> E . PLUS E, E -> E PLUS . E, E -> E PLUS E . ]
# -----------------------------------------------------------------------------
def build_lritems(self):
for p in self.Productions:
lastlri = p
i = 0
lr_items = []
while True:
if i > len(p):
lri = None
else:
lri = LRItem(p, i)
# Precompute the list of productions immediately following
try:
lri.lr_after = self.Prodnames[lri.prod[i+1]]
except (IndexError, KeyError):
lri.lr_after = []
try:
lri.lr_before = lri.prod[i-1]
except IndexError:
lri.lr_before = None
lastlri.lr_next = lri
if not lri:
break
lr_items.append(lri)
lastlri = lri
i += 1
p.lr_items = lr_items
# -----------------------------------------------------------------------------
# === LR Generator ===
#
# The following classes and functions are used to generate LR parsing tables on
# a grammar.
# -----------------------------------------------------------------------------
# -----------------------------------------------------------------------------
# digraph()
# traverse()
#
# The following two functions are used to compute set valued functions
# of the form:
#
# F(x) = F'(x) U U{F(y) | x R y}
#
# This is used to compute the values of Read() sets as well as FOLLOW sets
# in LALR(1) generation.
#
# Inputs: X - An input set
# R - A relation
# FP - Set-valued function
# ------------------------------------------------------------------------------
def digraph(X, R, FP):
N = {}
for x in X:
N[x] = 0
stack = []
F = {}
for x in X:
if N[x] == 0:
traverse(x, N, stack, F, X, R, FP)
return F
def traverse(x, N, stack, F, X, R, FP):
stack.append(x)
d = len(stack)
N[x] = d
F[x] = FP(x) # F(X) <- F'(x)
rel = R(x) # Get y's related to x
for y in rel:
if N[y] == 0:
traverse(y, N, stack, F, X, R, FP)
N[x] = min(N[x], N[y])
for a in F.get(y, []):
if a not in F[x]:
F[x].append(a)
if N[x] == d:
N[stack[-1]] = MAXINT
F[stack[-1]] = F[x]
element = stack.pop()
while element != x:
N[stack[-1]] = MAXINT
F[stack[-1]] = F[x]
element = stack.pop()
class LALRError(YaccError):
pass
# -----------------------------------------------------------------------------
# == LRTable ==
#
# This class implements the LR table generation algorithm. There are no
# public methods.
# -----------------------------------------------------------------------------
class LRTable:
def __init__(self, grammar, log=None):
self.grammar = grammar
# Set up the logger
if not log:
log = NullLogger()
self.log = log
# Internal attributes
self.lr_action = {} # Action table
self.lr_goto = {} # Goto table
self.lr_productions = grammar.Productions # Copy of grammar Production array
self.lr_goto_cache = {} # Cache of computed gotos
self.lr0_cidhash = {} # Cache of closures
self._add_count = 0 # Internal counter used to detect cycles
# Diagnostic information filled in by the table generator
self.sr_conflict = 0
self.rr_conflict = 0
self.conflicts = [] # List of conflicts
self.sr_conflicts = []
self.rr_conflicts = []
# Build the tables
self.grammar.build_lritems()
self.grammar.compute_first()
self.grammar.compute_follow()
self.lr_parse_table()
# Bind all production function names to callable objects in pdict
def bind_callables(self, pdict):
for p in self.lr_productions:
p.bind(pdict)
# Compute the LR(0) closure operation on I, where I is a set of LR(0) items.
def lr0_closure(self, I):
self._add_count += 1
# Add everything in I to J
J = I[:]
didadd = True
while didadd:
didadd = False
for j in J:
for x in j.lr_after:
if getattr(x, 'lr0_added', 0) == self._add_count:
continue
# Add B --> .G to J
J.append(x.lr_next)
x.lr0_added = self._add_count
didadd = True
return J
# Compute the LR(0) goto function goto(I,X) where I is a set
# of LR(0) items and X is a grammar symbol. This function is written
# in a way that guarantees uniqueness of the generated goto sets
# (i.e. the same goto set will never be returned as two different Python
# objects). With uniqueness, we can later do fast set comparisons using
# id(obj) instead of element-wise comparison.
def lr0_goto(self, I, x):
# First we look for a previously cached entry
g = self.lr_goto_cache.get((id(I), x))
if g:
return g
# Now we generate the goto set in a way that guarantees uniqueness
# of the result
s = self.lr_goto_cache.get(x)
if not s:
s = {}
self.lr_goto_cache[x] = s
gs = []
for p in I:
n = p.lr_next
if n and n.lr_before == x:
s1 = s.get(id(n))
if not s1:
s1 = {}
s[id(n)] = s1
gs.append(n)
s = s1
g = s.get('$end')
if not g:
if gs:
g = self.lr0_closure(gs)
s['$end'] = g
else:
s['$end'] = gs
self.lr_goto_cache[(id(I), x)] = g
return g
# Compute the LR(0) sets of item function
def lr0_items(self):
C = [self.lr0_closure([self.grammar.Productions[0].lr_next])]
i = 0
for I in C:
self.lr0_cidhash[id(I)] = i
i += 1
# Loop over the items in C and each grammar symbols
i = 0
while i < len(C):
I = C[i]
i += 1
# Collect all of the symbols that could possibly be in the goto(I,X) sets
asyms = {}
for ii in I:
for s in ii.usyms:
asyms[s] = None
for x in asyms:
g = self.lr0_goto(I, x)
if not g or id(g) in self.lr0_cidhash:
continue
self.lr0_cidhash[id(g)] = len(C)
C.append(g)
return C
# -----------------------------------------------------------------------------
# ==== LALR(1) Parsing ====
#
# LALR(1) parsing is almost exactly the same as SLR except that instead of
# relying upon Follow() sets when performing reductions, a more selective
# lookahead set that incorporates the state of the LR(0) machine is utilized.
# Thus, we mainly just have to focus on calculating the lookahead sets.
#
# The method used here is due to DeRemer and Pennelo (1982).
#
# DeRemer, F. L., and T. J. Pennelo: "Efficient Computation of LALR(1)
# Lookahead Sets", ACM Transactions on Programming Languages and Systems,
# Vol. 4, No. 4, Oct. 1982, pp. 615-649
#
# Further details can also be found in:
#
# J. Tremblay and P. Sorenson, "The Theory and Practice of Compiler Writing",
# McGraw-Hill Book Company, (1985).
#
# -----------------------------------------------------------------------------
# -----------------------------------------------------------------------------
# compute_nullable_nonterminals()
#
# Creates a dictionary containing all of the non-terminals that might produce
# an empty production.
# -----------------------------------------------------------------------------
def compute_nullable_nonterminals(self):
nullable = set()
num_nullable = 0
while True:
for p in self.grammar.Productions[1:]:
if p.len == 0:
nullable.add(p.name)
continue
for t in p.prod:
if t not in nullable:
break
else:
nullable.add(p.name)
if len(nullable) == num_nullable:
break
num_nullable = len(nullable)
return nullable
# -----------------------------------------------------------------------------
# find_nonterminal_trans(C)
#
# Given a set of LR(0) items, this functions finds all of the non-terminal
# transitions. These are transitions in which a dot appears immediately before
# a non-terminal. Returns a list of tuples of the form (state,N) where state
# is the state number and N is the nonterminal symbol.
#
# The input C is the set of LR(0) items.
# -----------------------------------------------------------------------------
def find_nonterminal_transitions(self, C):
trans = []
for stateno, state in enumerate(C):
for p in state:
if p.lr_index < p.len - 1:
t = (stateno, p.prod[p.lr_index+1])
if t[1] in self.grammar.Nonterminals:
if t not in trans:
trans.append(t)
return trans
# -----------------------------------------------------------------------------
# dr_relation()
#
# Computes the DR(p,A) relationships for non-terminal transitions. The input
# is a tuple (state,N) where state is a number and N is a nonterminal symbol.
#
# Returns a list of terminals.
# -----------------------------------------------------------------------------
def dr_relation(self, C, trans, nullable):
state, N = trans
terms = []
g = self.lr0_goto(C[state], N)
for p in g:
if p.lr_index < p.len - 1:
a = p.prod[p.lr_index+1]
if a in self.grammar.Terminals:
if a not in terms:
terms.append(a)
# This extra bit is to handle the start state
if state == 0 and N == self.grammar.Productions[0].prod[0]:
terms.append('$end')
return terms
# -----------------------------------------------------------------------------
# reads_relation()
#
# Computes the READS() relation (p,A) READS (t,C).
# -----------------------------------------------------------------------------
def reads_relation(self, C, trans, empty):
# Look for empty transitions
rel = []
state, N = trans
g = self.lr0_goto(C[state], N)
j = self.lr0_cidhash.get(id(g), -1)
for p in g:
if p.lr_index < p.len - 1:
a = p.prod[p.lr_index + 1]
if a in empty:
rel.append((j, a))
return rel
# -----------------------------------------------------------------------------
# compute_lookback_includes()
#
# Determines the lookback and includes relations
#
# LOOKBACK:
#
# This relation is determined by running the LR(0) state machine forward.
# For example, starting with a production "N : . A B C", we run it forward
# to obtain "N : A B C ." We then build a relationship between this final
# state and the starting state. These relationships are stored in a dictionary
# lookdict.
#
# INCLUDES:
#
# Computes the INCLUDE() relation (p,A) INCLUDES (p',B).
#
# This relation is used to determine non-terminal transitions that occur
# inside of other non-terminal transition states. (p,A) INCLUDES (p', B)
# if the following holds:
#
# B -> LAT, where T -> epsilon and p' -L-> p
#
# L is essentially a prefix (which may be empty), T is a suffix that must be
# able to derive an empty string. State p' must lead to state p with the string L.
#
# -----------------------------------------------------------------------------
def compute_lookback_includes(self, C, trans, nullable):
lookdict = {} # Dictionary of lookback relations
includedict = {} # Dictionary of include relations
# Make a dictionary of non-terminal transitions
dtrans = {}
for t in trans:
dtrans[t] = 1
# Loop over all transitions and compute lookbacks and includes
for state, N in trans:
lookb = []
includes = []
for p in C[state]:
if p.name != N:
continue
# Okay, we have a name match. We now follow the production all the way
# through the state machine until we get the . on the right hand side
lr_index = p.lr_index
j = state
while lr_index < p.len - 1:
lr_index = lr_index + 1
t = p.prod[lr_index]
# Check to see if this symbol and state are a non-terminal transition
if (j, t) in dtrans:
# Yes. Okay, there is some chance that this is an includes relation
# the only way to know for certain is whether the rest of the
# production derives empty
li = lr_index + 1
while li < p.len:
if p.prod[li] in self.grammar.Terminals:
break # No forget it
if p.prod[li] not in nullable:
break
li = li + 1
else:
# Appears to be a relation between (j,t) and (state,N)
includes.append((j, t))
g = self.lr0_goto(C[j], t) # Go to next set
j = self.lr0_cidhash.get(id(g), -1) # Go to next state
# When we get here, j is the final state, now we have to locate the production
for r in C[j]:
if r.name != p.name:
continue
if r.len != p.len:
continue
i = 0
# This look is comparing a production ". A B C" with "A B C ."
while i < r.lr_index:
if r.prod[i] != p.prod[i+1]:
break
i = i + 1
else:
lookb.append((j, r))
for i in includes:
if i not in includedict:
includedict[i] = []
includedict[i].append((state, N))
lookdict[(state, N)] = lookb
return lookdict, includedict
# -----------------------------------------------------------------------------
# compute_read_sets()
#
# Given a set of LR(0) items, this function computes the read sets.
#
# Inputs: C = Set of LR(0) items
# ntrans = Set of nonterminal transitions
# nullable = Set of empty transitions
#
# Returns a set containing the read sets
# -----------------------------------------------------------------------------
def compute_read_sets(self, C, ntrans, nullable):
FP = lambda x: self.dr_relation(C, x, nullable)
R = lambda x: self.reads_relation(C, x, nullable)
F = digraph(ntrans, R, FP)
return F
# -----------------------------------------------------------------------------
# compute_follow_sets()
#
# Given a set of LR(0) items, a set of non-terminal transitions, a readset,
# and an include set, this function computes the follow sets
#
# Follow(p,A) = Read(p,A) U U {Follow(p',B) | (p,A) INCLUDES (p',B)}
#
# Inputs:
# ntrans = Set of nonterminal transitions
# readsets = Readset (previously computed)
# inclsets = Include sets (previously computed)
#
# Returns a set containing the follow sets
# -----------------------------------------------------------------------------
def compute_follow_sets(self, ntrans, readsets, inclsets):
FP = lambda x: readsets[x]
R = lambda x: inclsets.get(x, [])
F = digraph(ntrans, R, FP)
return F
# -----------------------------------------------------------------------------
# add_lookaheads()
#
# Attaches the lookahead symbols to grammar rules.
#
# Inputs: lookbacks - Set of lookback relations
# followset - Computed follow set
#
# This function directly attaches the lookaheads to productions contained
# in the lookbacks set
# -----------------------------------------------------------------------------
def add_lookaheads(self, lookbacks, followset):
for trans, lb in lookbacks.items():
# Loop over productions in lookback
for state, p in lb:
if state not in p.lookaheads:
p.lookaheads[state] = []
f = followset.get(trans, [])
for a in f:
if a not in p.lookaheads[state]:
p.lookaheads[state].append(a)
# -----------------------------------------------------------------------------
# add_lalr_lookaheads()
#
# This function does all of the work of adding lookahead information for use
# with LALR parsing
# -----------------------------------------------------------------------------
def add_lalr_lookaheads(self, C):
# Determine all of the nullable nonterminals
nullable = self.compute_nullable_nonterminals()
# Find all non-terminal transitions
trans = self.find_nonterminal_transitions(C)
# Compute read sets
readsets = self.compute_read_sets(C, trans, nullable)
# Compute lookback/includes relations
lookd, included = self.compute_lookback_includes(C, trans, nullable)
# Compute LALR FOLLOW sets
followsets = self.compute_follow_sets(trans, readsets, included)
# Add all of the lookaheads
self.add_lookaheads(lookd, followsets)
# -----------------------------------------------------------------------------
# lr_parse_table()
#
# This function constructs the parse tables for SLR or LALR
# -----------------------------------------------------------------------------
def lr_parse_table(self):
Productions = self.grammar.Productions
Precedence = self.grammar.Precedence
goto = self.lr_goto # Goto array
action = self.lr_action # Action array
log = self.log # Logger for output
actionp = {} # Action production array (temporary)
# Step 1: Construct C = { I0, I1, ... IN}, collection of LR(0) items
# This determines the number of states
C = self.lr0_items()
self.add_lalr_lookaheads(C)
# Build the parser table, state by state
st = 0
for I in C:
# Loop over each production in I
actlist = [] # List of actions
st_action = {}
st_actionp = {}
st_goto = {}
log.info('')
log.info('state %d', st)
log.info('')
for p in I:
log.info(' (%d) %s', p.number, p)
log.info('')
for p in I:
if p.len == p.lr_index + 1:
if p.name == "S'":
# Start symbol. Accept!
st_action['$end'] = 0
st_actionp['$end'] = p
else:
# We are at the end of a production. Reduce!
laheads = p.lookaheads[st]
for a in laheads:
actlist.append((a, p, 'reduce using rule %d (%s)' % (p.number, p)))
r = st_action.get(a)
if r is not None:
# Whoa. Have a shift/reduce or reduce/reduce conflict
if r > 0:
# Need to decide on shift or reduce here
# By default we favor shifting. Need to add
# some precedence rules here.
# Shift precedence comes from the token
sprec, slevel = Precedence.get(a, ('right', 0))
# Reduce precedence comes from rule being reduced (p)
rprec, rlevel = Productions[p.number].prec
if (slevel < rlevel) or ((slevel == rlevel) and (rprec == 'left')):
# We really need to reduce here.
st_action[a] = -p.number
st_actionp[a] = p
if not slevel and not rlevel:
log.info(' ! shift/reduce conflict for %s resolved as reduce', a)
self.sr_conflicts.append((st, a, 'reduce'))
Productions[p.number].reduced += 1
elif (slevel == rlevel) and (rprec == 'nonassoc'):
st_action[a] = None
else:
# Hmmm. Guess we'll keep the shift
if not rlevel:
log.info(' ! shift/reduce conflict for %s resolved as shift', a)
self.sr_conflicts.append((st, a, 'shift'))
elif r < 0:
# Reduce/reduce conflict. In this case, we favor the rule
# that was defined first in the grammar file
oldp = Productions[-r]
pp = Productions[p.number]
if oldp.line > pp.line:
st_action[a] = -p.number
st_actionp[a] = p
chosenp, rejectp = pp, oldp
Productions[p.number].reduced += 1
Productions[oldp.number].reduced -= 1
else:
chosenp, rejectp = oldp, pp
self.rr_conflicts.append((st, chosenp, rejectp))
log.info(' ! reduce/reduce conflict for %s resolved using rule %d (%s)',
a, st_actionp[a].number, st_actionp[a])
else:
raise LALRError('Unknown conflict in state %d' % st)
else:
st_action[a] = -p.number
st_actionp[a] = p
Productions[p.number].reduced += 1
else:
i = p.lr_index
a = p.prod[i+1] # Get symbol right after the "."
if a in self.grammar.Terminals:
g = self.lr0_goto(I, a)
j = self.lr0_cidhash.get(id(g), -1)
if j >= 0:
# We are in a shift state
actlist.append((a, p, 'shift and go to state %d' % j))
r = st_action.get(a)
if r is not None:
# Whoa have a shift/reduce or shift/shift conflict
if r > 0:
if r != j:
raise LALRError('Shift/shift conflict in state %d' % st)
elif r < 0:
# Do a precedence check.
# - if precedence of reduce rule is higher, we reduce.
# - if precedence of reduce is same and left assoc, we reduce.
# - otherwise we shift
# Shift precedence comes from the token
sprec, slevel = Precedence.get(a, ('right', 0))
# Reduce precedence comes from the rule that could have been reduced
rprec, rlevel = Productions[st_actionp[a].number].prec
if (slevel > rlevel) or ((slevel == rlevel) and (rprec == 'right')):
# We decide to shift here... highest precedence to shift
Productions[st_actionp[a].number].reduced -= 1
st_action[a] = j
st_actionp[a] = p
if not rlevel:
log.info(' ! shift/reduce conflict for %s resolved as shift', a)
self.sr_conflicts.append((st, a, 'shift'))
elif (slevel == rlevel) and (rprec == 'nonassoc'):
st_action[a] = None
else:
# Hmmm. Guess we'll keep the reduce
if not slevel and not rlevel:
log.info(' ! shift/reduce conflict for %s resolved as reduce', a)
self.sr_conflicts.append((st, a, 'reduce'))
else:
raise LALRError('Unknown conflict in state %d' % st)
else:
st_action[a] = j
st_actionp[a] = p
# Print the actions associated with each terminal
_actprint = {}
for a, p, m in actlist:
if a in st_action:
if p is st_actionp[a]:
log.info(' %-15s %s', a, m)
_actprint[(a, m)] = 1
log.info('')
# Print the actions that were not used. (debugging)
not_used = 0
for a, p, m in actlist:
if a in st_action:
if p is not st_actionp[a]:
if not (a, m) in _actprint:
log.debug(' ! %-15s [ %s ]', a, m)
not_used = 1
_actprint[(a, m)] = 1
if not_used:
log.debug('')
# Construct the goto table for this state
nkeys = {}
for ii in I:
for s in ii.usyms:
if s in self.grammar.Nonterminals:
nkeys[s] = None
for n in nkeys:
g = self.lr0_goto(I, n)
j = self.lr0_cidhash.get(id(g), -1)
if j >= 0:
st_goto[n] = j
log.info(' %-30s shift and go to state %d', n, j)
action[st] = st_action
actionp[st] = st_actionp
goto[st] = st_goto
st += 1
# -----------------------------------------------------------------------------
# === INTROSPECTION ===
#
# The following functions and classes are used to implement the PLY
# introspection features followed by the yacc() function itself.
# -----------------------------------------------------------------------------
# -----------------------------------------------------------------------------
# get_caller_module_dict()
#
# This function returns a dictionary containing all of the symbols defined within
# a caller further down the call stack. This is used to get the environment
# associated with the yacc() call if none was provided.
# -----------------------------------------------------------------------------
def get_caller_module_dict(levels):
f = sys._getframe(levels)
ldict = f.f_globals.copy()
if f.f_globals != f.f_locals:
ldict.update(f.f_locals)
return ldict
# -----------------------------------------------------------------------------
# parse_grammar()
#
# This takes a raw grammar rule string and parses it into production data
# -----------------------------------------------------------------------------
def parse_grammar(doc, file, line):
grammar = []
# Split the doc string into lines
pstrings = doc.splitlines()
lastp = None
dline = line
for ps in pstrings:
dline += 1
p = ps.split()
if not p:
continue
try:
if p[0] == '|':
# This is a continuation of a previous rule
if not lastp:
raise SyntaxError("%s:%d: Misplaced '|'" % (file, dline))
prodname = lastp
syms = p[1:]
else:
prodname = p[0]
lastp = prodname
syms = p[2:]
assign = p[1]
if assign != ':' and assign != '::=':
raise SyntaxError("%s:%d: Syntax error. Expected ':'" % (file, dline))
grammar.append((file, dline, prodname, syms))
except SyntaxError:
raise
except Exception:
raise SyntaxError('%s:%d: Syntax error in rule %r' % (file, dline, ps.strip()))
return grammar
# -----------------------------------------------------------------------------
# ParserReflect()
#
# This class represents information extracted for building a parser including
# start symbol, error function, tokens, precedence list, action functions,
# etc.
# -----------------------------------------------------------------------------
class ParserReflect(object):
def __init__(self, pdict, log=None):
self.pdict = pdict
self.start = None
self.error_func = None
self.tokens = None
self.modules = set()
self.grammar = []
self.error = False
if log is None:
self.log = PlyLogger(sys.stderr)
else:
self.log = log
# Get all of the basic information
def get_all(self):
self.get_start()
self.get_error_func()
self.get_tokens()
self.get_precedence()
self.get_pfunctions()
# Validate all of the information
def validate_all(self):
self.validate_start()
self.validate_error_func()
self.validate_tokens()
self.validate_precedence()
self.validate_pfunctions()
self.validate_modules()
return self.error
# Compute a signature over the grammar
def signature(self):
parts = []
try:
if self.start:
parts.append(self.start)
if self.prec:
parts.append(''.join([''.join(p) for p in self.prec]))
if self.tokens:
parts.append(' '.join(self.tokens))
for f in self.pfuncs:
if f[3]:
parts.append(f[3])
except (TypeError, ValueError):
pass
return ''.join(parts)
# -----------------------------------------------------------------------------
# validate_modules()
#
# This method checks to see if there are duplicated p_rulename() functions
# in the parser module file. Without this function, it is really easy for
# users to make mistakes by cutting and pasting code fragments (and it's a real
# bugger to try and figure out why the resulting parser doesn't work). Therefore,
# we just do a little regular expression pattern matching of def statements
# to try and detect duplicates.
# -----------------------------------------------------------------------------
def validate_modules(self):
# Match def p_funcname(
fre = re.compile(r'\s*def\s+(p_[a-zA-Z_0-9]*)\(')
for module in self.modules:
try:
lines, linen = inspect.getsourcelines(module)
except IOError:
continue
counthash = {}
for linen, line in enumerate(lines):
linen += 1
m = fre.match(line)
if m:
name = m.group(1)
prev = counthash.get(name)
if not prev:
counthash[name] = linen
else:
filename = inspect.getsourcefile(module)
self.log.warning('%s:%d: Function %s redefined. Previously defined on line %d',
filename, linen, name, prev)
# Get the start symbol
def get_start(self):
self.start = self.pdict.get('start')
# Validate the start symbol
def validate_start(self):
if self.start is not None:
if not isinstance(self.start, str):
self.log.error("'start' must be a string")
# Look for error handler
def get_error_func(self):
self.error_func = self.pdict.get('p_error')
# Validate the error function
def validate_error_func(self):
if self.error_func:
if isinstance(self.error_func, types.FunctionType):
ismethod = 0
elif isinstance(self.error_func, types.MethodType):
ismethod = 1
else:
self.log.error("'p_error' defined, but is not a function or method")
self.error = True
return
eline = self.error_func.__code__.co_firstlineno
efile = self.error_func.__code__.co_filename
module = inspect.getmodule(self.error_func)
self.modules.add(module)
argcount = self.error_func.__code__.co_argcount - ismethod
if argcount != 1:
self.log.error('%s:%d: p_error() requires 1 argument', efile, eline)
self.error = True
# Get the tokens map
def get_tokens(self):
tokens = self.pdict.get('tokens')
if not tokens:
self.log.error('No token list is defined')
self.error = True
return
if not isinstance(tokens, (list, tuple)):
self.log.error('tokens must be a list or tuple')
self.error = True
return
if not tokens:
self.log.error('tokens is empty')
self.error = True
return
self.tokens = sorted(tokens)
# Validate the tokens
def validate_tokens(self):
# Validate the tokens.
if 'error' in self.tokens:
self.log.error("Illegal token name 'error'. Is a reserved word")
self.error = True
return
terminals = set()
for n in self.tokens:
if n in terminals:
self.log.warning('Token %r multiply defined', n)
terminals.add(n)
# Get the precedence map (if any)
def get_precedence(self):
self.prec = self.pdict.get('precedence')
# Validate and parse the precedence map
def validate_precedence(self):
preclist = []
if self.prec:
if not isinstance(self.prec, (list, tuple)):
self.log.error('precedence must be a list or tuple')
self.error = True
return
for level, p in enumerate(self.prec):
if not isinstance(p, (list, tuple)):
self.log.error('Bad precedence table')
self.error = True
return
if len(p) < 2:
self.log.error('Malformed precedence entry %s. Must be (assoc, term, ..., term)', p)
self.error = True
return
assoc = p[0]
if not isinstance(assoc, str):
self.log.error('precedence associativity must be a string')
self.error = True
return
for term in p[1:]:
if not isinstance(term, str):
self.log.error('precedence items must be strings')
self.error = True
return
preclist.append((term, assoc, level+1))
self.preclist = preclist
# Get all p_functions from the grammar
def get_pfunctions(self):
p_functions = []
for name, item in self.pdict.items():
if not name.startswith('p_') or name == 'p_error':
continue
if isinstance(item, (types.FunctionType, types.MethodType)):
line = getattr(item, 'co_firstlineno', item.__code__.co_firstlineno)
module = inspect.getmodule(item)
p_functions.append((line, module, name, item.__doc__))
# Sort all of the actions by line number; make sure to stringify
# modules to make them sortable, since `line` may not uniquely sort all
# p functions
p_functions.sort(key=lambda p_function: (
p_function[0],
str(p_function[1]),
p_function[2],
p_function[3]))
self.pfuncs = p_functions
# Validate all of the p_functions
def validate_pfunctions(self):
grammar = []
# Check for non-empty symbols
if len(self.pfuncs) == 0:
self.log.error('no rules of the form p_rulename are defined')
self.error = True
return
for line, module, name, doc in self.pfuncs:
file = inspect.getsourcefile(module)
func = self.pdict[name]
if isinstance(func, types.MethodType):
reqargs = 2
else:
reqargs = 1
if func.__code__.co_argcount > reqargs:
self.log.error('%s:%d: Rule %r has too many arguments', file, line, func.__name__)
self.error = True
elif func.__code__.co_argcount < reqargs:
self.log.error('%s:%d: Rule %r requires an argument', file, line, func.__name__)
self.error = True
elif not func.__doc__:
self.log.warning('%s:%d: No documentation string specified in function %r (ignored)',
file, line, func.__name__)
else:
try:
parsed_g = parse_grammar(doc, file, line)
for g in parsed_g:
grammar.append((name, g))
except SyntaxError as e:
self.log.error(str(e))
self.error = True
# Looks like a valid grammar rule
# Mark the file in which defined.
self.modules.add(module)
# Secondary validation step that looks for p_ definitions that are not functions
# or functions that look like they might be grammar rules.
for n, v in self.pdict.items():
if n.startswith('p_') and isinstance(v, (types.FunctionType, types.MethodType)):
continue
if n.startswith('t_'):
continue
if n.startswith('p_') and n != 'p_error':
self.log.warning('%r not defined as a function', n)
if ((isinstance(v, types.FunctionType) and v.__code__.co_argcount == 1) or
(isinstance(v, types.MethodType) and v.__func__.__code__.co_argcount == 2)):
if v.__doc__:
try:
doc = v.__doc__.split(' ')
if doc[1] == ':':
self.log.warning('%s:%d: Possible grammar rule %r defined without p_ prefix',
v.__code__.co_filename, v.__code__.co_firstlineno, n)
except IndexError:
pass
self.grammar = grammar
# -----------------------------------------------------------------------------
# yacc(module)
#
# Build a parser
# -----------------------------------------------------------------------------
def yacc(*, debug=yaccdebug, module=None, start=None,
check_recursion=True, optimize=False, debugfile=debug_file,
debuglog=None, errorlog=None):
# Reference to the parsing method of the last built parser
global parse
if errorlog is None:
errorlog = PlyLogger(sys.stderr)
# Get the module dictionary used for the parser
if module:
_items = [(k, getattr(module, k)) for k in dir(module)]
pdict = dict(_items)
# If no __file__ or __package__ attributes are available, try to obtain them
# from the __module__ instead
if '__file__' not in pdict:
pdict['__file__'] = sys.modules[pdict['__module__']].__file__
if '__package__' not in pdict and '__module__' in pdict:
if hasattr(sys.modules[pdict['__module__']], '__package__'):
pdict['__package__'] = sys.modules[pdict['__module__']].__package__
else:
pdict = get_caller_module_dict(2)
# Set start symbol if it's specified directly using an argument
if start is not None:
pdict['start'] = start
# Collect parser information from the dictionary
pinfo = ParserReflect(pdict, log=errorlog)
pinfo.get_all()
if pinfo.error:
raise YaccError('Unable to build parser')
if debuglog is None:
if debug:
try:
debuglog = PlyLogger(open(debugfile, 'w'))
except IOError as e:
errorlog.warning("Couldn't open %r. %s" % (debugfile, e))
debuglog = NullLogger()
else:
debuglog = NullLogger()
debuglog.info('Created by PLY (http://www.dabeaz.com/ply)')
errors = False
# Validate the parser information
if pinfo.validate_all():
raise YaccError('Unable to build parser')
if not pinfo.error_func:
errorlog.warning('no p_error() function is defined')
# Create a grammar object
grammar = Grammar(pinfo.tokens)
# Set precedence level for terminals
for term, assoc, level in pinfo.preclist:
try:
grammar.set_precedence(term, assoc, level)
except GrammarError as e:
errorlog.warning('%s', e)
# Add productions to the grammar
for funcname, gram in pinfo.grammar:
file, line, prodname, syms = gram
try:
grammar.add_production(prodname, syms, funcname, file, line)
except GrammarError as e:
errorlog.error('%s', e)
errors = True
# Set the grammar start symbols
try:
if start is None:
grammar.set_start(pinfo.start)
else:
grammar.set_start(start)
except GrammarError as e:
errorlog.error(str(e))
errors = True
if errors:
raise YaccError('Unable to build parser')
# Verify the grammar structure
undefined_symbols = grammar.undefined_symbols()
for sym, prod in undefined_symbols:
errorlog.error('%s:%d: Symbol %r used, but not defined as a token or a rule', prod.file, prod.line, sym)
errors = True
unused_terminals = grammar.unused_terminals()
if unused_terminals:
debuglog.info('')
debuglog.info('Unused terminals:')
debuglog.info('')
for term in unused_terminals:
errorlog.warning('Token %r defined, but not used', term)
debuglog.info(' %s', term)
# Print out all productions to the debug log
if debug:
debuglog.info('')
debuglog.info('Grammar')
debuglog.info('')
for n, p in enumerate(grammar.Productions):
debuglog.info('Rule %-5d %s', n, p)
# Find unused non-terminals
unused_rules = grammar.unused_rules()
for prod in unused_rules:
errorlog.warning('%s:%d: Rule %r defined, but not used', prod.file, prod.line, prod.name)
if len(unused_terminals) == 1:
errorlog.warning('There is 1 unused token')
if len(unused_terminals) > 1:
errorlog.warning('There are %d unused tokens', len(unused_terminals))
if len(unused_rules) == 1:
errorlog.warning('There is 1 unused rule')
if len(unused_rules) > 1:
errorlog.warning('There are %d unused rules', len(unused_rules))
if debug:
debuglog.info('')
debuglog.info('Terminals, with rules where they appear')
debuglog.info('')
terms = list(grammar.Terminals)
terms.sort()
for term in terms:
debuglog.info('%-20s : %s', term, ' '.join([str(s) for s in grammar.Terminals[term]]))
debuglog.info('')
debuglog.info('Nonterminals, with rules where they appear')
debuglog.info('')
nonterms = list(grammar.Nonterminals)
nonterms.sort()
for nonterm in nonterms:
debuglog.info('%-20s : %s', nonterm, ' '.join([str(s) for s in grammar.Nonterminals[nonterm]]))
debuglog.info('')
if check_recursion:
unreachable = grammar.find_unreachable()
for u in unreachable:
errorlog.warning('Symbol %r is unreachable', u)
infinite = grammar.infinite_cycles()
for inf in infinite:
errorlog.error('Infinite recursion detected for symbol %r', inf)
errors = True
unused_prec = grammar.unused_precedence()
for term, assoc in unused_prec:
errorlog.error('Precedence rule %r defined for unknown symbol %r', assoc, term)
errors = True
if errors:
raise YaccError('Unable to build parser')
# Run the LRTable on the grammar
lr = LRTable(grammar, debuglog)
if debug:
num_sr = len(lr.sr_conflicts)
# Report shift/reduce and reduce/reduce conflicts
if num_sr == 1:
errorlog.warning('1 shift/reduce conflict')
elif num_sr > 1:
errorlog.warning('%d shift/reduce conflicts', num_sr)
num_rr = len(lr.rr_conflicts)
if num_rr == 1:
errorlog.warning('1 reduce/reduce conflict')
elif num_rr > 1:
errorlog.warning('%d reduce/reduce conflicts', num_rr)
# Write out conflicts to the output file
if debug and (lr.sr_conflicts or lr.rr_conflicts):
debuglog.warning('')
debuglog.warning('Conflicts:')
debuglog.warning('')
for state, tok, resolution in lr.sr_conflicts:
debuglog.warning('shift/reduce conflict for %s in state %d resolved as %s', tok, state, resolution)
already_reported = set()
for state, rule, rejected in lr.rr_conflicts:
if (state, id(rule), id(rejected)) in already_reported:
continue
debuglog.warning('reduce/reduce conflict in state %d resolved using rule (%s)', state, rule)
debuglog.warning('rejected rule (%s) in state %d', rejected, state)
errorlog.warning('reduce/reduce conflict in state %d resolved using rule (%s)', state, rule)
errorlog.warning('rejected rule (%s) in state %d', rejected, state)
already_reported.add((state, id(rule), id(rejected)))
warned_never = []
for state, rule, rejected in lr.rr_conflicts:
if not rejected.reduced and (rejected not in warned_never):
debuglog.warning('Rule (%s) is never reduced', rejected)
errorlog.warning('Rule (%s) is never reduced', rejected)
warned_never.append(rejected)
# Build the parser
lr.bind_callables(pinfo.pdict)
parser = LRParser(lr, pinfo.error_func)
parse = parser.parse
return parser
|