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
/*!
Defines a high-level intermediate representation for regular expressions.
*/
use std::char;
use std::cmp;
use std::error;
use std::fmt;
use std::result;
use std::u8;
use crate::ast::Span;
use crate::hir::interval::{Interval, IntervalSet, IntervalSetIter};
use crate::unicode;
pub use crate::hir::visitor::{visit, Visitor};
pub use crate::unicode::CaseFoldError;
mod interval;
pub mod literal;
pub mod print;
pub mod translate;
mod visitor;
/// An error that can occur while translating an `Ast` to a `Hir`.
#[derive(Clone, Debug, Eq, PartialEq)]
pub struct Error {
/// The kind of error.
kind: ErrorKind,
/// The original pattern that the translator's Ast was parsed from. Every
/// span in an error is a valid range into this string.
pattern: String,
/// The span of this error, derived from the Ast given to the translator.
span: Span,
}
impl Error {
/// Return the type of this error.
pub fn kind(&self) -> &ErrorKind {
&self.kind
}
/// The original pattern string in which this error occurred.
///
/// Every span reported by this error is reported in terms of this string.
pub fn pattern(&self) -> &str {
&self.pattern
}
/// Return the span at which this error occurred.
pub fn span(&self) -> &Span {
&self.span
}
}
/// The type of an error that occurred while building an `Hir`.
#[derive(Clone, Debug, Eq, PartialEq)]
pub enum ErrorKind {
/// This error occurs when a Unicode feature is used when Unicode
/// support is disabled. For example `(?-u:\pL)` would trigger this error.
UnicodeNotAllowed,
/// This error occurs when translating a pattern that could match a byte
/// sequence that isn't UTF-8 and `allow_invalid_utf8` was disabled.
InvalidUtf8,
/// This occurs when an unrecognized Unicode property name could not
/// be found.
UnicodePropertyNotFound,
/// This occurs when an unrecognized Unicode property value could not
/// be found.
UnicodePropertyValueNotFound,
/// This occurs when a Unicode-aware Perl character class (`\w`, `\s` or
/// `\d`) could not be found. This can occur when the `unicode-perl`
/// crate feature is not enabled.
UnicodePerlClassNotFound,
/// This occurs when the Unicode simple case mapping tables are not
/// available, and the regular expression required Unicode aware case
/// insensitivity.
UnicodeCaseUnavailable,
/// This occurs when the translator attempts to construct a character class
/// that is empty.
///
/// Note that this restriction in the translator may be removed in the
/// future.
EmptyClassNotAllowed,
/// Hints that destructuring should not be exhaustive.
///
/// This enum may grow additional variants, so this makes sure clients
/// don't count on exhaustive matching. (Otherwise, adding a new variant
/// could break existing code.)
#[doc(hidden)]
__Nonexhaustive,
}
impl ErrorKind {
// TODO: Remove this method entirely on the next breaking semver release.
#[allow(deprecated)]
fn description(&self) -> &str {
use self::ErrorKind::*;
match *self {
UnicodeNotAllowed => "Unicode not allowed here",
InvalidUtf8 => "pattern can match invalid UTF-8",
UnicodePropertyNotFound => "Unicode property not found",
UnicodePropertyValueNotFound => "Unicode property value not found",
UnicodePerlClassNotFound => {
"Unicode-aware Perl class not found \
(make sure the unicode-perl feature is enabled)"
}
UnicodeCaseUnavailable => {
"Unicode-aware case insensitivity matching is not available \
(make sure the unicode-case feature is enabled)"
}
EmptyClassNotAllowed => "empty character classes are not allowed",
__Nonexhaustive => unreachable!(),
}
}
}
impl error::Error for Error {
// TODO: Remove this method entirely on the next breaking semver release.
#[allow(deprecated)]
fn description(&self) -> &str {
self.kind.description()
}
}
impl fmt::Display for Error {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
crate::error::Formatter::from(self).fmt(f)
}
}
impl fmt::Display for ErrorKind {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
// TODO: Remove this on the next breaking semver release.
#[allow(deprecated)]
f.write_str(self.description())
}
}
/// A high-level intermediate representation (HIR) for a regular expression.
///
/// The HIR of a regular expression represents an intermediate step between its
/// abstract syntax (a structured description of the concrete syntax) and
/// compiled byte codes. The purpose of HIR is to make regular expressions
/// easier to analyze. In particular, the AST is much more complex than the
/// HIR. For example, while an AST supports arbitrarily nested character
/// classes, the HIR will flatten all nested classes into a single set. The HIR
/// will also "compile away" every flag present in the concrete syntax. For
/// example, users of HIR expressions never need to worry about case folding;
/// it is handled automatically by the translator (e.g., by translating `(?i)A`
/// to `[aA]`).
///
/// If the HIR was produced by a translator that disallows invalid UTF-8, then
/// the HIR is guaranteed to match UTF-8 exclusively.
///
/// This type defines its own destructor that uses constant stack space and
/// heap space proportional to the size of the HIR.
///
/// The specific type of an HIR expression can be accessed via its `kind`
/// or `into_kind` methods. This extra level of indirection exists for two
/// reasons:
///
/// 1. Construction of an HIR expression *must* use the constructor methods
/// on this `Hir` type instead of building the `HirKind` values directly.
/// This permits construction to enforce invariants like "concatenations
/// always consist of two or more sub-expressions."
/// 2. Every HIR expression contains attributes that are defined inductively,
/// and can be computed cheaply during the construction process. For
/// example, one such attribute is whether the expression must match at the
/// beginning of the text.
///
/// Also, an `Hir`'s `fmt::Display` implementation prints an HIR as a regular
/// expression pattern string, and uses constant stack space and heap space
/// proportional to the size of the `Hir`.
#[derive(Clone, Debug, Eq, PartialEq)]
pub struct Hir {
/// The underlying HIR kind.
kind: HirKind,
/// Analysis info about this HIR, computed during construction.
info: HirInfo,
}
/// The kind of an arbitrary `Hir` expression.
#[derive(Clone, Debug, Eq, PartialEq)]
pub enum HirKind {
/// The empty regular expression, which matches everything, including the
/// empty string.
Empty,
/// A single literal character that matches exactly this character.
Literal(Literal),
/// A single character class that matches any of the characters in the
/// class. A class can either consist of Unicode scalar values as
/// characters, or it can use bytes.
Class(Class),
/// An anchor assertion. An anchor assertion match always has zero length.
Anchor(Anchor),
/// A word boundary assertion, which may or may not be Unicode aware. A
/// word boundary assertion match always has zero length.
WordBoundary(WordBoundary),
/// A repetition operation applied to a child expression.
Repetition(Repetition),
/// A possibly capturing group, which contains a child expression.
Group(Group),
/// A concatenation of expressions. A concatenation always has at least two
/// child expressions.
///
/// A concatenation matches only if each of its child expression matches
/// one after the other.
Concat(Vec<Hir>),
/// An alternation of expressions. An alternation always has at least two
/// child expressions.
///
/// An alternation matches only if at least one of its child expression
/// matches. If multiple expressions match, then the leftmost is preferred.
Alternation(Vec<Hir>),
}
impl Hir {
/// Returns a reference to the underlying HIR kind.
pub fn kind(&self) -> &HirKind {
&self.kind
}
/// Consumes ownership of this HIR expression and returns its underlying
/// `HirKind`.
pub fn into_kind(mut self) -> HirKind {
use std::mem;
mem::replace(&mut self.kind, HirKind::Empty)
}
/// Returns an empty HIR expression.
///
/// An empty HIR expression always matches, including the empty string.
pub fn empty() -> Hir {
let mut info = HirInfo::new();
info.set_always_utf8(true);
info.set_all_assertions(true);
info.set_anchored_start(false);
info.set_anchored_end(false);
info.set_line_anchored_start(false);
info.set_line_anchored_end(false);
info.set_any_anchored_start(false);
info.set_any_anchored_end(false);
info.set_match_empty(true);
info.set_literal(false);
info.set_alternation_literal(false);
Hir { kind: HirKind::Empty, info }
}
/// Creates a literal HIR expression.
///
/// If the given literal has a `Byte` variant with an ASCII byte, then this
/// method panics. This enforces the invariant that `Byte` variants are
/// only used to express matching of invalid UTF-8.
pub fn literal(lit: Literal) -> Hir {
if let Literal::Byte(b) = lit {
assert!(b > 0x7F);
}
let mut info = HirInfo::new();
info.set_always_utf8(lit.is_unicode());
info.set_all_assertions(false);
info.set_anchored_start(false);
info.set_anchored_end(false);
info.set_line_anchored_start(false);
info.set_line_anchored_end(false);
info.set_any_anchored_start(false);
info.set_any_anchored_end(false);
info.set_match_empty(false);
info.set_literal(true);
info.set_alternation_literal(true);
Hir { kind: HirKind::Literal(lit), info }
}
/// Creates a class HIR expression.
pub fn class(class: Class) -> Hir {
let mut info = HirInfo::new();
info.set_always_utf8(class.is_always_utf8());
info.set_all_assertions(false);
info.set_anchored_start(false);
info.set_anchored_end(false);
info.set_line_anchored_start(false);
info.set_line_anchored_end(false);
info.set_any_anchored_start(false);
info.set_any_anchored_end(false);
info.set_match_empty(false);
info.set_literal(false);
info.set_alternation_literal(false);
Hir { kind: HirKind::Class(class), info }
}
/// Creates an anchor assertion HIR expression.
pub fn anchor(anchor: Anchor) -> Hir {
let mut info = HirInfo::new();
info.set_always_utf8(true);
info.set_all_assertions(true);
info.set_anchored_start(false);
info.set_anchored_end(false);
info.set_line_anchored_start(false);
info.set_line_anchored_end(false);
info.set_any_anchored_start(false);
info.set_any_anchored_end(false);
info.set_match_empty(true);
info.set_literal(false);
info.set_alternation_literal(false);
if let Anchor::StartText = anchor {
info.set_anchored_start(true);
info.set_line_anchored_start(true);
info.set_any_anchored_start(true);
}
if let Anchor::EndText = anchor {
info.set_anchored_end(true);
info.set_line_anchored_end(true);
info.set_any_anchored_end(true);
}
if let Anchor::StartLine = anchor {
info.set_line_anchored_start(true);
}
if let Anchor::EndLine = anchor {
info.set_line_anchored_end(true);
}
Hir { kind: HirKind::Anchor(anchor), info }
}
/// Creates a word boundary assertion HIR expression.
pub fn word_boundary(word_boundary: WordBoundary) -> Hir {
let mut info = HirInfo::new();
info.set_always_utf8(true);
info.set_all_assertions(true);
info.set_anchored_start(false);
info.set_anchored_end(false);
info.set_line_anchored_start(false);
info.set_line_anchored_end(false);
info.set_any_anchored_start(false);
info.set_any_anchored_end(false);
info.set_literal(false);
info.set_alternation_literal(false);
// A negated word boundary matches '', so that's fine. But \b does not
// match \b, so why do we say it can match the empty string? Well,
// because, if you search for \b against 'a', it will report [0, 0) and
// [1, 1) as matches, and both of those matches correspond to the empty
// string. Thus, only *certain* empty strings match \b, which similarly
// applies to \B.
info.set_match_empty(true);
// Negated ASCII word boundaries can match invalid UTF-8.
if let WordBoundary::AsciiNegate = word_boundary {
info.set_always_utf8(false);
}
Hir { kind: HirKind::WordBoundary(word_boundary), info }
}
/// Creates a repetition HIR expression.
pub fn repetition(rep: Repetition) -> Hir {
let mut info = HirInfo::new();
info.set_always_utf8(rep.hir.is_always_utf8());
info.set_all_assertions(rep.hir.is_all_assertions());
// If this operator can match the empty string, then it can never
// be anchored.
info.set_anchored_start(
!rep.is_match_empty() && rep.hir.is_anchored_start(),
);
info.set_anchored_end(
!rep.is_match_empty() && rep.hir.is_anchored_end(),
);
info.set_line_anchored_start(
!rep.is_match_empty() && rep.hir.is_anchored_start(),
);
info.set_line_anchored_end(
!rep.is_match_empty() && rep.hir.is_anchored_end(),
);
info.set_any_anchored_start(rep.hir.is_any_anchored_start());
info.set_any_anchored_end(rep.hir.is_any_anchored_end());
info.set_match_empty(rep.is_match_empty() || rep.hir.is_match_empty());
info.set_literal(false);
info.set_alternation_literal(false);
Hir { kind: HirKind::Repetition(rep), info }
}
/// Creates a group HIR expression.
pub fn group(group: Group) -> Hir {
let mut info = HirInfo::new();
info.set_always_utf8(group.hir.is_always_utf8());
info.set_all_assertions(group.hir.is_all_assertions());
info.set_anchored_start(group.hir.is_anchored_start());
info.set_anchored_end(group.hir.is_anchored_end());
info.set_line_anchored_start(group.hir.is_line_anchored_start());
info.set_line_anchored_end(group.hir.is_line_anchored_end());
info.set_any_anchored_start(group.hir.is_any_anchored_start());
info.set_any_anchored_end(group.hir.is_any_anchored_end());
info.set_match_empty(group.hir.is_match_empty());
info.set_literal(false);
info.set_alternation_literal(false);
Hir { kind: HirKind::Group(group), info }
}
/// Returns the concatenation of the given expressions.
///
/// This flattens the concatenation as appropriate.
pub fn concat(mut exprs: Vec<Hir>) -> Hir {
match exprs.len() {
0 => Hir::empty(),
1 => exprs.pop().unwrap(),
_ => {
let mut info = HirInfo::new();
info.set_always_utf8(true);
info.set_all_assertions(true);
info.set_any_anchored_start(false);
info.set_any_anchored_end(false);
info.set_match_empty(true);
info.set_literal(true);
info.set_alternation_literal(true);
// Some attributes require analyzing all sub-expressions.
for e in &exprs {
let x = info.is_always_utf8() && e.is_always_utf8();
info.set_always_utf8(x);
let x = info.is_all_assertions() && e.is_all_assertions();
info.set_all_assertions(x);
let x = info.is_any_anchored_start()
|| e.is_any_anchored_start();
info.set_any_anchored_start(x);
let x =
info.is_any_anchored_end() || e.is_any_anchored_end();
info.set_any_anchored_end(x);
let x = info.is_match_empty() && e.is_match_empty();
info.set_match_empty(x);
let x = info.is_literal() && e.is_literal();
info.set_literal(x);
let x = info.is_alternation_literal()
&& e.is_alternation_literal();
info.set_alternation_literal(x);
}
// Anchored attributes require something slightly more
// sophisticated. Normally, WLOG, to determine whether an
// expression is anchored to the start, we'd only need to check
// the first expression of a concatenation. However,
// expressions like `$\b^` are still anchored to the start,
// but the first expression in the concatenation *isn't*
// anchored to the start. So the "first" expression to look at
// is actually one that is either not an assertion or is
// specifically the StartText assertion.
info.set_anchored_start(
exprs
.iter()
.take_while(|e| {
e.is_anchored_start() || e.is_all_assertions()
})
.any(|e| e.is_anchored_start()),
);
// Similarly for the end anchor, but in reverse.
info.set_anchored_end(
exprs
.iter()
.rev()
.take_while(|e| {
e.is_anchored_end() || e.is_all_assertions()
})
.any(|e| e.is_anchored_end()),
);
// Repeat the process for line anchors.
info.set_line_anchored_start(
exprs
.iter()
.take_while(|e| {
e.is_line_anchored_start() || e.is_all_assertions()
})
.any(|e| e.is_line_anchored_start()),
);
info.set_line_anchored_end(
exprs
.iter()
.rev()
.take_while(|e| {
e.is_line_anchored_end() || e.is_all_assertions()
})
.any(|e| e.is_line_anchored_end()),
);
Hir { kind: HirKind::Concat(exprs), info }
}
}
}
/// Returns the alternation of the given expressions.
///
/// This flattens the alternation as appropriate.
pub fn alternation(mut exprs: Vec<Hir>) -> Hir {
match exprs.len() {
0 => Hir::empty(),
1 => exprs.pop().unwrap(),
_ => {
let mut info = HirInfo::new();
info.set_always_utf8(true);
info.set_all_assertions(true);
info.set_anchored_start(true);
info.set_anchored_end(true);
info.set_line_anchored_start(true);
info.set_line_anchored_end(true);
info.set_any_anchored_start(false);
info.set_any_anchored_end(false);
info.set_match_empty(false);
info.set_literal(false);
info.set_alternation_literal(true);
// Some attributes require analyzing all sub-expressions.
for e in &exprs {
let x = info.is_always_utf8() && e.is_always_utf8();
info.set_always_utf8(x);
let x = info.is_all_assertions() && e.is_all_assertions();
info.set_all_assertions(x);
let x = info.is_anchored_start() && e.is_anchored_start();
info.set_anchored_start(x);
let x = info.is_anchored_end() && e.is_anchored_end();
info.set_anchored_end(x);
let x = info.is_line_anchored_start()
&& e.is_line_anchored_start();
info.set_line_anchored_start(x);
let x = info.is_line_anchored_end()
&& e.is_line_anchored_end();
info.set_line_anchored_end(x);
let x = info.is_any_anchored_start()
|| e.is_any_anchored_start();
info.set_any_anchored_start(x);
let x =
info.is_any_anchored_end() || e.is_any_anchored_end();
info.set_any_anchored_end(x);
let x = info.is_match_empty() || e.is_match_empty();
info.set_match_empty(x);
let x = info.is_alternation_literal() && e.is_literal();
info.set_alternation_literal(x);
}
Hir { kind: HirKind::Alternation(exprs), info }
}
}
}
/// Build an HIR expression for `.`.
///
/// A `.` expression matches any character except for `\n`. To build an
/// expression that matches any character, including `\n`, use the `any`
/// method.
///
/// If `bytes` is `true`, then this assumes characters are limited to a
/// single byte.
pub fn dot(bytes: bool) -> Hir {
if bytes {
let mut cls = ClassBytes::empty();
cls.push(ClassBytesRange::new(b'\0', b'\x09'));
cls.push(ClassBytesRange::new(b'\x0B', b'\xFF'));
Hir::class(Class::Bytes(cls))
} else {
let mut cls = ClassUnicode::empty();
cls.push(ClassUnicodeRange::new('\0', '\x09'));
cls.push(ClassUnicodeRange::new('\x0B', '\u{10FFFF}'));
Hir::class(Class::Unicode(cls))
}
}
/// Build an HIR expression for `(?s).`.
///
/// A `(?s).` expression matches any character, including `\n`. To build an
/// expression that matches any character except for `\n`, then use the
/// `dot` method.
///
/// If `bytes` is `true`, then this assumes characters are limited to a
/// single byte.
pub fn any(bytes: bool) -> Hir {
if bytes {
let mut cls = ClassBytes::empty();
cls.push(ClassBytesRange::new(b'\0', b'\xFF'));
Hir::class(Class::Bytes(cls))
} else {
let mut cls = ClassUnicode::empty();
cls.push(ClassUnicodeRange::new('\0', '\u{10FFFF}'));
Hir::class(Class::Unicode(cls))
}
}
/// Return true if and only if this HIR will always match valid UTF-8.
///
/// When this returns false, then it is possible for this HIR expression
/// to match invalid UTF-8.
pub fn is_always_utf8(&self) -> bool {
self.info.is_always_utf8()
}
/// Returns true if and only if this entire HIR expression is made up of
/// zero-width assertions.
///
/// This includes expressions like `^$\b\A\z` and even `((\b)+())*^`, but
/// not `^a`.
pub fn is_all_assertions(&self) -> bool {
self.info.is_all_assertions()
}
/// Return true if and only if this HIR is required to match from the
/// beginning of text. This includes expressions like `^foo`, `^(foo|bar)`,
/// `^foo|^bar` but not `^foo|bar`.
pub fn is_anchored_start(&self) -> bool {
self.info.is_anchored_start()
}
/// Return true if and only if this HIR is required to match at the end
/// of text. This includes expressions like `foo$`, `(foo|bar)$`,
/// `foo$|bar$` but not `foo$|bar`.
pub fn is_anchored_end(&self) -> bool {
self.info.is_anchored_end()
}
/// Return true if and only if this HIR is required to match from the
/// beginning of text or the beginning of a line. This includes expressions
/// like `^foo`, `(?m)^foo`, `^(foo|bar)`, `^(foo|bar)`, `(?m)^foo|^bar`
/// but not `^foo|bar` or `(?m)^foo|bar`.
///
/// Note that if `is_anchored_start` is `true`, then
/// `is_line_anchored_start` will also be `true`. The reverse implication
/// is not true. For example, `(?m)^foo` is line anchored, but not
/// `is_anchored_start`.
pub fn is_line_anchored_start(&self) -> bool {
self.info.is_line_anchored_start()
}
/// Return true if and only if this HIR is required to match at the
/// end of text or the end of a line. This includes expressions like
/// `foo$`, `(?m)foo$`, `(foo|bar)$`, `(?m)(foo|bar)$`, `foo$|bar$`,
/// `(?m)(foo|bar)$`, but not `foo$|bar` or `(?m)foo$|bar`.
///
/// Note that if `is_anchored_end` is `true`, then
/// `is_line_anchored_end` will also be `true`. The reverse implication
/// is not true. For example, `(?m)foo$` is line anchored, but not
/// `is_anchored_end`.
pub fn is_line_anchored_end(&self) -> bool {
self.info.is_line_anchored_end()
}
/// Return true if and only if this HIR contains any sub-expression that
/// is required to match at the beginning of text. Specifically, this
/// returns true if the `^` symbol (when multiline mode is disabled) or the
/// `\A` escape appear anywhere in the regex.
pub fn is_any_anchored_start(&self) -> bool {
self.info.is_any_anchored_start()
}
/// Return true if and only if this HIR contains any sub-expression that is
/// required to match at the end of text. Specifically, this returns true
/// if the `$` symbol (when multiline mode is disabled) or the `\z` escape
/// appear anywhere in the regex.
pub fn is_any_anchored_end(&self) -> bool {
self.info.is_any_anchored_end()
}
/// Return true if and only if the empty string is part of the language
/// matched by this regular expression.
///
/// This includes `a*`, `a?b*`, `a{0}`, `()`, `()+`, `^$`, `a|b?`, `\b`
/// and `\B`, but not `a` or `a+`.
pub fn is_match_empty(&self) -> bool {
self.info.is_match_empty()
}
/// Return true if and only if this HIR is a simple literal. This is only
/// true when this HIR expression is either itself a `Literal` or a
/// concatenation of only `Literal`s.
///
/// For example, `f` and `foo` are literals, but `f+`, `(foo)`, `foo()`,
/// `` are not (even though that contain sub-expressions that are literals).
pub fn is_literal(&self) -> bool {
self.info.is_literal()
}
/// Return true if and only if this HIR is either a simple literal or an
/// alternation of simple literals. This is only
/// true when this HIR expression is either itself a `Literal` or a
/// concatenation of only `Literal`s or an alternation of only `Literal`s.
///
/// For example, `f`, `foo`, `a|b|c`, and `foo|bar|baz` are alternation
/// literals, but `f+`, `(foo)`, `foo()`, ``
/// are not (even though that contain sub-expressions that are literals).
pub fn is_alternation_literal(&self) -> bool {
self.info.is_alternation_literal()
}
}
impl HirKind {
/// Return true if and only if this HIR is the empty regular expression.
///
/// Note that this is not defined inductively. That is, it only tests if
/// this kind is the `Empty` variant. To get the inductive definition,
/// use the `is_match_empty` method on [`Hir`](struct.Hir.html).
pub fn is_empty(&self) -> bool {
match *self {
HirKind::Empty => true,
_ => false,
}
}
/// Returns true if and only if this kind has any (including possibly
/// empty) subexpressions.
pub fn has_subexprs(&self) -> bool {
match *self {
HirKind::Empty
| HirKind::Literal(_)
| HirKind::Class(_)
| HirKind::Anchor(_)
| HirKind::WordBoundary(_) => false,
HirKind::Group(_)
| HirKind::Repetition(_)
| HirKind::Concat(_)
| HirKind::Alternation(_) => true,
}
}
}
/// Print a display representation of this Hir.
///
/// The result of this is a valid regular expression pattern string.
///
/// This implementation uses constant stack space and heap space proportional
/// to the size of the `Hir`.
impl fmt::Display for Hir {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
use crate::hir::print::Printer;
Printer::new().print(self, f)
}
}
/// The high-level intermediate representation of a literal.
///
/// A literal corresponds to a single character, where a character is either
/// defined by a Unicode scalar value or an arbitrary byte. Unicode characters
/// are preferred whenever possible. In particular, a `Byte` variant is only
/// ever produced when it could match invalid UTF-8.
#[derive(Clone, Debug, Eq, PartialEq)]
pub enum Literal {
/// A single character represented by a Unicode scalar value.
Unicode(char),
/// A single character represented by an arbitrary byte.
Byte(u8),
}
impl Literal {
/// Returns true if and only if this literal corresponds to a Unicode
/// scalar value.
pub fn is_unicode(&self) -> bool {
match *self {
Literal::Unicode(_) => true,
Literal::Byte(b) if b <= 0x7F => true,
Literal::Byte(_) => false,
}
}
}
/// The high-level intermediate representation of a character class.
///
/// A character class corresponds to a set of characters. A character is either
/// defined by a Unicode scalar value or a byte. Unicode characters are used
/// by default, while bytes are used when Unicode mode (via the `u` flag) is
/// disabled.
///
/// A character class, regardless of its character type, is represented by a
/// sequence of non-overlapping non-adjacent ranges of characters.
///
/// Note that unlike [`Literal`](enum.Literal.html), a `Bytes` variant may
/// be produced even when it exclusively matches valid UTF-8. This is because
/// a `Bytes` variant represents an intention by the author of the regular
/// expression to disable Unicode mode, which in turn impacts the semantics of
/// case insensitive matching. For example, `(?i)k` and `(?i-u)k` will not
/// match the same set of strings.
#[derive(Clone, Debug, Eq, PartialEq)]
pub enum Class {
/// A set of characters represented by Unicode scalar values.
Unicode(ClassUnicode),
/// A set of characters represented by arbitrary bytes (one byte per
/// character).
Bytes(ClassBytes),
}
impl Class {
/// Apply Unicode simple case folding to this character class, in place.
/// The character class will be expanded to include all simple case folded
/// character variants.
///
/// If this is a byte oriented character class, then this will be limited
/// to the ASCII ranges `A-Z` and `a-z`.
pub fn case_fold_simple(&mut self) {
match *self {
Class::Unicode(ref mut x) => x.case_fold_simple(),
Class::Bytes(ref mut x) => x.case_fold_simple(),
}
}
/// Negate this character class in place.
///
/// After completion, this character class will contain precisely the
/// characters that weren't previously in the class.
pub fn negate(&mut self) {
match *self {
Class::Unicode(ref mut x) => x.negate(),
Class::Bytes(ref mut x) => x.negate(),
}
}
/// Returns true if and only if this character class will only ever match
/// valid UTF-8.
///
/// A character class can match invalid UTF-8 only when the following
/// conditions are met:
///
/// 1. The translator was configured to permit generating an expression
/// that can match invalid UTF-8. (By default, this is disabled.)
/// 2. Unicode mode (via the `u` flag) was disabled either in the concrete
/// syntax or in the parser builder. By default, Unicode mode is
/// enabled.
pub fn is_always_utf8(&self) -> bool {
match *self {
Class::Unicode(_) => true,
Class::Bytes(ref x) => x.is_all_ascii(),
}
}
}
/// A set of characters represented by Unicode scalar values.
#[derive(Clone, Debug, Eq, PartialEq)]
pub struct ClassUnicode {
set: IntervalSet<ClassUnicodeRange>,
}
impl ClassUnicode {
/// Create a new class from a sequence of ranges.
///
/// The given ranges do not need to be in any specific order, and ranges
/// may overlap.
pub fn new<I>(ranges: I) -> ClassUnicode
where
I: IntoIterator<Item = ClassUnicodeRange>,
{
ClassUnicode { set: IntervalSet::new(ranges) }
}
/// Create a new class with no ranges.
pub fn empty() -> ClassUnicode {
ClassUnicode::new(vec![])
}
/// Add a new range to this set.
pub fn push(&mut self, range: ClassUnicodeRange) {
self.set.push(range);
}
/// Return an iterator over all ranges in this class.
///
/// The iterator yields ranges in ascending order.
pub fn iter(&self) -> ClassUnicodeIter<'_> {
ClassUnicodeIter(self.set.iter())
}
/// Return the underlying ranges as a slice.
pub fn ranges(&self) -> &[ClassUnicodeRange] {
self.set.intervals()
}
/// Expand this character class such that it contains all case folded
/// characters, according to Unicode's "simple" mapping. For example, if
/// this class consists of the range `a-z`, then applying case folding will
/// result in the class containing both the ranges `a-z` and `A-Z`.
///
/// # Panics
///
/// This routine panics when the case mapping data necessary for this
/// routine to complete is unavailable. This occurs when the `unicode-case`
/// feature is not enabled.
///
/// Callers should prefer using `try_case_fold_simple` instead, which will
/// return an error instead of panicking.
pub fn case_fold_simple(&mut self) {
self.set
.case_fold_simple()
.expect("unicode-case feature must be enabled");
}
/// Expand this character class such that it contains all case folded
/// characters, according to Unicode's "simple" mapping. For example, if
/// this class consists of the range `a-z`, then applying case folding will
/// result in the class containing both the ranges `a-z` and `A-Z`.
///
/// # Error
///
/// This routine returns an error when the case mapping data necessary
/// for this routine to complete is unavailable. This occurs when the
/// `unicode-case` feature is not enabled.
pub fn try_case_fold_simple(
&mut self,
) -> result::Result<(), CaseFoldError> {
self.set.case_fold_simple()
}
/// Negate this character class.
///
/// For all `c` where `c` is a Unicode scalar value, if `c` was in this
/// set, then it will not be in this set after negation.
pub fn negate(&mut self) {
self.set.negate();
}
/// Union this character class with the given character class, in place.
pub fn union(&mut self, other: &ClassUnicode) {
self.set.union(&other.set);
}
/// Intersect this character class with the given character class, in
/// place.
pub fn intersect(&mut self, other: &ClassUnicode) {
self.set.intersect(&other.set);
}
/// Subtract the given character class from this character class, in place.
pub fn difference(&mut self, other: &ClassUnicode) {
self.set.difference(&other.set);
}
/// Compute the symmetric difference of the given character classes, in
/// place.
///
/// This computes the symmetric difference of two character classes. This
/// removes all elements in this class that are also in the given class,
/// but all adds all elements from the given class that aren't in this
/// class. That is, the class will contain all elements in either class,
/// but will not contain any elements that are in both classes.
pub fn symmetric_difference(&mut self, other: &ClassUnicode) {
self.set.symmetric_difference(&other.set);
}
/// Returns true if and only if this character class will either match
/// nothing or only ASCII bytes. Stated differently, this returns false
/// if and only if this class contains a non-ASCII codepoint.
pub fn is_all_ascii(&self) -> bool {
self.set.intervals().last().map_or(true, |r| r.end <= '\x7F')
}
}
/// An iterator over all ranges in a Unicode character class.
///
/// The lifetime `'a` refers to the lifetime of the underlying class.
#[derive(Debug)]
pub struct ClassUnicodeIter<'a>(IntervalSetIter<'a, ClassUnicodeRange>);
impl<'a> Iterator for ClassUnicodeIter<'a> {
type Item = &'a ClassUnicodeRange;
fn next(&mut self) -> Option<&'a ClassUnicodeRange> {
self.0.next()
}
}
/// A single range of characters represented by Unicode scalar values.
///
/// The range is closed. That is, the start and end of the range are included
/// in the range.
#[derive(Clone, Copy, Default, Eq, PartialEq, PartialOrd, Ord)]
pub struct ClassUnicodeRange {
start: char,
end: char,
}
impl fmt::Debug for ClassUnicodeRange {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
let start = if !self.start.is_whitespace() && !self.start.is_control()
{
self.start.to_string()
} else {
format!("0x{:X}", self.start as u32)
};
let end = if !self.end.is_whitespace() && !self.end.is_control() {
self.end.to_string()
} else {
format!("0x{:X}", self.end as u32)
};
f.debug_struct("ClassUnicodeRange")
.field("start", &start)
.field("end", &end)
.finish()
}
}
impl Interval for ClassUnicodeRange {
type Bound = char;
#[inline]
fn lower(&self) -> char {
self.start
}
#[inline]
fn upper(&self) -> char {
self.end
}
#[inline]
fn set_lower(&mut self, bound: char) {
self.start = bound;
}
#[inline]
fn set_upper(&mut self, bound: char) {
self.end = bound;
}
/// Apply simple case folding to this Unicode scalar value range.
///
/// Additional ranges are appended to the given vector. Canonical ordering
/// is *not* maintained in the given vector.
fn case_fold_simple(
&self,
ranges: &mut Vec<ClassUnicodeRange>,
) -> Result<(), unicode::CaseFoldError> {
if !unicode::contains_simple_case_mapping(self.start, self.end)? {
return Ok(());
}
let start = self.start as u32;
let end = (self.end as u32).saturating_add(1);
let mut next_simple_cp = None;
for cp in (start..end).filter_map(char::from_u32) {
if next_simple_cp.map_or(false, |next| cp < next) {
continue;
}
let it = match unicode::simple_fold(cp)? {
Ok(it) => it,
Err(next) => {
next_simple_cp = next;
continue;
}
};
for cp_folded in it {
ranges.push(ClassUnicodeRange::new(cp_folded, cp_folded));
}
}
Ok(())
}
}
impl ClassUnicodeRange {
/// Create a new Unicode scalar value range for a character class.
///
/// The returned range is always in a canonical form. That is, the range
/// returned always satisfies the invariant that `start <= end`.
pub fn new(start: char, end: char) -> ClassUnicodeRange {
ClassUnicodeRange::create(start, end)
}
/// Return the start of this range.
///
/// The start of a range is always less than or equal to the end of the
/// range.
pub fn start(&self) -> char {
self.start
}
/// Return the end of this range.
///
/// The end of a range is always greater than or equal to the start of the
/// range.
pub fn end(&self) -> char {
self.end
}
}
/// A set of characters represented by arbitrary bytes (where one byte
/// corresponds to one character).
#[derive(Clone, Debug, Eq, PartialEq)]
pub struct ClassBytes {
set: IntervalSet<ClassBytesRange>,
}
impl ClassBytes {
/// Create a new class from a sequence of ranges.
///
/// The given ranges do not need to be in any specific order, and ranges
/// may overlap.
pub fn new<I>(ranges: I) -> ClassBytes
where
I: IntoIterator<Item = ClassBytesRange>,
{
ClassBytes { set: IntervalSet::new(ranges) }
}
/// Create a new class with no ranges.
pub fn empty() -> ClassBytes {
ClassBytes::new(vec![])
}
/// Add a new range to this set.
pub fn push(&mut self, range: ClassBytesRange) {
self.set.push(range);
}
/// Return an iterator over all ranges in this class.
///
/// The iterator yields ranges in ascending order.
pub fn iter(&self) -> ClassBytesIter<'_> {
ClassBytesIter(self.set.iter())
}
/// Return the underlying ranges as a slice.
pub fn ranges(&self) -> &[ClassBytesRange] {
self.set.intervals()
}
/// Expand this character class such that it contains all case folded
/// characters. For example, if this class consists of the range `a-z`,
/// then applying case folding will result in the class containing both the
/// ranges `a-z` and `A-Z`.
///
/// Note that this only applies ASCII case folding, which is limited to the
/// characters `a-z` and `A-Z`.
pub fn case_fold_simple(&mut self) {
self.set.case_fold_simple().expect("ASCII case folding never fails");
}
/// Negate this byte class.
///
/// For all `b` where `b` is a any byte, if `b` was in this set, then it
/// will not be in this set after negation.
pub fn negate(&mut self) {
self.set.negate();
}
/// Union this byte class with the given byte class, in place.
pub fn union(&mut self, other: &ClassBytes) {
self.set.union(&other.set);
}
/// Intersect this byte class with the given byte class, in place.
pub fn intersect(&mut self, other: &ClassBytes) {
self.set.intersect(&other.set);
}
/// Subtract the given byte class from this byte class, in place.
pub fn difference(&mut self, other: &ClassBytes) {
self.set.difference(&other.set);
}
/// Compute the symmetric difference of the given byte classes, in place.
///
/// This computes the symmetric difference of two byte classes. This
/// removes all elements in this class that are also in the given class,
/// but all adds all elements from the given class that aren't in this
/// class. That is, the class will contain all elements in either class,
/// but will not contain any elements that are in both classes.
pub fn symmetric_difference(&mut self, other: &ClassBytes) {
self.set.symmetric_difference(&other.set);
}
/// Returns true if and only if this character class will either match
/// nothing or only ASCII bytes. Stated differently, this returns false
/// if and only if this class contains a non-ASCII byte.
pub fn is_all_ascii(&self) -> bool {
self.set.intervals().last().map_or(true, |r| r.end <= 0x7F)
}
}
/// An iterator over all ranges in a byte character class.
///
/// The lifetime `'a` refers to the lifetime of the underlying class.
#[derive(Debug)]
pub struct ClassBytesIter<'a>(IntervalSetIter<'a, ClassBytesRange>);
impl<'a> Iterator for ClassBytesIter<'a> {
type Item = &'a ClassBytesRange;
fn next(&mut self) -> Option<&'a ClassBytesRange> {
self.0.next()
}
}
/// A single range of characters represented by arbitrary bytes.
///
/// The range is closed. That is, the start and end of the range are included
/// in the range.
#[derive(Clone, Copy, Default, Eq, PartialEq, PartialOrd, Ord)]
pub struct ClassBytesRange {
start: u8,
end: u8,
}
impl Interval for ClassBytesRange {
type Bound = u8;
#[inline]
fn lower(&self) -> u8 {
self.start
}
#[inline]
fn upper(&self) -> u8 {
self.end
}
#[inline]
fn set_lower(&mut self, bound: u8) {
self.start = bound;
}
#[inline]
fn set_upper(&mut self, bound: u8) {
self.end = bound;
}
/// Apply simple case folding to this byte range. Only ASCII case mappings
/// (for a-z) are applied.
///
/// Additional ranges are appended to the given vector. Canonical ordering
/// is *not* maintained in the given vector.
fn case_fold_simple(
&self,
ranges: &mut Vec<ClassBytesRange>,
) -> Result<(), unicode::CaseFoldError> {
if !ClassBytesRange::new(b'a', b'z').is_intersection_empty(self) {
let lower = cmp::max(self.start, b'a');
let upper = cmp::min(self.end, b'z');
ranges.push(ClassBytesRange::new(lower - 32, upper - 32));
}
if !ClassBytesRange::new(b'A', b'Z').is_intersection_empty(self) {
let lower = cmp::max(self.start, b'A');
let upper = cmp::min(self.end, b'Z');
ranges.push(ClassBytesRange::new(lower + 32, upper + 32));
}
Ok(())
}
}
impl ClassBytesRange {
/// Create a new byte range for a character class.
///
/// The returned range is always in a canonical form. That is, the range
/// returned always satisfies the invariant that `start <= end`.
pub fn new(start: u8, end: u8) -> ClassBytesRange {
ClassBytesRange::create(start, end)
}
/// Return the start of this range.
///
/// The start of a range is always less than or equal to the end of the
/// range.
pub fn start(&self) -> u8 {
self.start
}
/// Return the end of this range.
///
/// The end of a range is always greater than or equal to the start of the
/// range.
pub fn end(&self) -> u8 {
self.end
}
}
impl fmt::Debug for ClassBytesRange {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
let mut debug = f.debug_struct("ClassBytesRange");
if self.start <= 0x7F {
debug.field("start", &(self.start as char));
} else {
debug.field("start", &self.start);
}
if self.end <= 0x7F {
debug.field("end", &(self.end as char));
} else {
debug.field("end", &self.end);
}
debug.finish()
}
}
/// The high-level intermediate representation for an anchor assertion.
///
/// A matching anchor assertion is always zero-length.
#[derive(Clone, Debug, Eq, PartialEq)]
pub enum Anchor {
/// Match the beginning of a line or the beginning of text. Specifically,
/// this matches at the starting position of the input, or at the position
/// immediately following a `\n` character.
StartLine,
/// Match the end of a line or the end of text. Specifically,
/// this matches at the end position of the input, or at the position
/// immediately preceding a `\n` character.
EndLine,
/// Match the beginning of text. Specifically, this matches at the starting
/// position of the input.
StartText,
/// Match the end of text. Specifically, this matches at the ending
/// position of the input.
EndText,
}
/// The high-level intermediate representation for a word-boundary assertion.
///
/// A matching word boundary assertion is always zero-length.
#[derive(Clone, Debug, Eq, PartialEq)]
pub enum WordBoundary {
/// Match a Unicode-aware word boundary. That is, this matches a position
/// where the left adjacent character and right adjacent character
/// correspond to a word and non-word or a non-word and word character.
Unicode,
/// Match a Unicode-aware negation of a word boundary.
UnicodeNegate,
/// Match an ASCII-only word boundary. That is, this matches a position
/// where the left adjacent character and right adjacent character
/// correspond to a word and non-word or a non-word and word character.
Ascii,
/// Match an ASCII-only negation of a word boundary.
AsciiNegate,
}
impl WordBoundary {
/// Returns true if and only if this word boundary assertion is negated.
pub fn is_negated(&self) -> bool {
match *self {
WordBoundary::Unicode | WordBoundary::Ascii => false,
WordBoundary::UnicodeNegate | WordBoundary::AsciiNegate => true,
}
}
}
/// The high-level intermediate representation for a group.
///
/// This represents one of three possible group types:
///
/// 1. A non-capturing group (e.g., `(?:expr)`).
/// 2. A capturing group (e.g., `(expr)`).
/// 3. A named capturing group (e.g., `(?P<name>expr)`).
#[derive(Clone, Debug, Eq, PartialEq)]
pub struct Group {
/// The kind of this group. If it is a capturing group, then the kind
/// contains the capture group index (and the name, if it is a named
/// group).
pub kind: GroupKind,
/// The expression inside the capturing group, which may be empty.
pub hir: Box<Hir>,
}
/// The kind of group.
#[derive(Clone, Debug, Eq, PartialEq)]
pub enum GroupKind {
/// A normal unnamed capturing group.
///
/// The value is the capture index of the group.
CaptureIndex(u32),
/// A named capturing group.
CaptureName {
/// The name of the group.
name: String,
/// The capture index of the group.
index: u32,
},
/// A non-capturing group.
NonCapturing,
}
/// The high-level intermediate representation of a repetition operator.
///
/// A repetition operator permits the repetition of an arbitrary
/// sub-expression.
#[derive(Clone, Debug, Eq, PartialEq)]
pub struct Repetition {
/// The kind of this repetition operator.
pub kind: RepetitionKind,
/// Whether this repetition operator is greedy or not. A greedy operator
/// will match as much as it can. A non-greedy operator will match as
/// little as it can.
///
/// Typically, operators are greedy by default and are only non-greedy when
/// a `?` suffix is used, e.g., `(expr)*` is greedy while `(expr)*?` is
/// not. However, this can be inverted via the `U` "ungreedy" flag.
pub greedy: bool,
/// The expression being repeated.
pub hir: Box<Hir>,
}
impl Repetition {
/// Returns true if and only if this repetition operator makes it possible
/// to match the empty string.
///
/// Note that this is not defined inductively. For example, while `a*`
/// will report `true`, `()+` will not, even though `()` matches the empty
/// string and one or more occurrences of something that matches the empty
/// string will always match the empty string. In order to get the
/// inductive definition, see the corresponding method on
/// [`Hir`](struct.Hir.html).
pub fn is_match_empty(&self) -> bool {
match self.kind {
RepetitionKind::ZeroOrOne => true,
RepetitionKind::ZeroOrMore => true,
RepetitionKind::OneOrMore => false,
RepetitionKind::Range(RepetitionRange::Exactly(m)) => m == 0,
RepetitionKind::Range(RepetitionRange::AtLeast(m)) => m == 0,
RepetitionKind::Range(RepetitionRange::Bounded(m, _)) => m == 0,
}
}
}
/// The kind of a repetition operator.
#[derive(Clone, Debug, Eq, PartialEq)]
pub enum RepetitionKind {
/// Matches a sub-expression zero or one times.
ZeroOrOne,
/// Matches a sub-expression zero or more times.
ZeroOrMore,
/// Matches a sub-expression one or more times.
OneOrMore,
/// Matches a sub-expression within a bounded range of times.
Range(RepetitionRange),
}
/// The kind of a counted repetition operator.
#[derive(Clone, Debug, Eq, PartialEq)]
pub enum RepetitionRange {
/// Matches a sub-expression exactly this many times.
Exactly(u32),
/// Matches a sub-expression at least this many times.
AtLeast(u32),
/// Matches a sub-expression at least `m` times and at most `n` times.
Bounded(u32, u32),
}
/// A custom `Drop` impl is used for `HirKind` such that it uses constant stack
/// space but heap space proportional to the depth of the total `Hir`.
impl Drop for Hir {
fn drop(&mut self) {
use std::mem;
match *self.kind() {
HirKind::Empty
| HirKind::Literal(_)
| HirKind::Class(_)
| HirKind::Anchor(_)
| HirKind::WordBoundary(_) => return,
HirKind::Group(ref x) if !x.hir.kind.has_subexprs() => return,
HirKind::Repetition(ref x) if !x.hir.kind.has_subexprs() => return,
HirKind::Concat(ref x) if x.is_empty() => return,
HirKind::Alternation(ref x) if x.is_empty() => return,
_ => {}
}
let mut stack = vec![mem::replace(self, Hir::empty())];
while let Some(mut expr) = stack.pop() {
match expr.kind {
HirKind::Empty
| HirKind::Literal(_)
| HirKind::Class(_)
| HirKind::Anchor(_)
| HirKind::WordBoundary(_) => {}
HirKind::Group(ref mut x) => {
stack.push(mem::replace(&mut x.hir, Hir::empty()));
}
HirKind::Repetition(ref mut x) => {
stack.push(mem::replace(&mut x.hir, Hir::empty()));
}
HirKind::Concat(ref mut x) => {
stack.extend(x.drain(..));
}
HirKind::Alternation(ref mut x) => {
stack.extend(x.drain(..));
}
}
}
}
}
/// A type that documents various attributes of an HIR expression.
///
/// These attributes are typically defined inductively on the HIR.
#[derive(Clone, Debug, Eq, PartialEq)]
struct HirInfo {
/// Represent yes/no questions by a bitfield to conserve space, since
/// this is included in every HIR expression.
///
/// If more attributes need to be added, it is OK to increase the size of
/// this as appropriate.
bools: u16,
}
// A simple macro for defining bitfield accessors/mutators.
macro_rules! define_bool {
($bit:expr, $is_fn_name:ident, $set_fn_name:ident) => {
fn $is_fn_name(&self) -> bool {
self.bools & (0b1 << $bit) > 0
}
fn $set_fn_name(&mut self, yes: bool) {
if yes {
self.bools |= 1 << $bit;
} else {
self.bools &= !(1 << $bit);
}
}
};
}
impl HirInfo {
fn new() -> HirInfo {
HirInfo { bools: 0 }
}
define_bool!(0, is_always_utf8, set_always_utf8);
define_bool!(1, is_all_assertions, set_all_assertions);
define_bool!(2, is_anchored_start, set_anchored_start);
define_bool!(3, is_anchored_end, set_anchored_end);
define_bool!(4, is_line_anchored_start, set_line_anchored_start);
define_bool!(5, is_line_anchored_end, set_line_anchored_end);
define_bool!(6, is_any_anchored_start, set_any_anchored_start);
define_bool!(7, is_any_anchored_end, set_any_anchored_end);
define_bool!(8, is_match_empty, set_match_empty);
define_bool!(9, is_literal, set_literal);
define_bool!(10, is_alternation_literal, set_alternation_literal);
}
#[cfg(test)]
mod tests {
use super::*;
fn uclass(ranges: &[(char, char)]) -> ClassUnicode {
let ranges: Vec<ClassUnicodeRange> = ranges
.iter()
.map(|&(s, e)| ClassUnicodeRange::new(s, e))
.collect();
ClassUnicode::new(ranges)
}
fn bclass(ranges: &[(u8, u8)]) -> ClassBytes {
let ranges: Vec<ClassBytesRange> =
ranges.iter().map(|&(s, e)| ClassBytesRange::new(s, e)).collect();
ClassBytes::new(ranges)
}
fn uranges(cls: &ClassUnicode) -> Vec<(char, char)> {
cls.iter().map(|x| (x.start(), x.end())).collect()
}
#[cfg(feature = "unicode-case")]
fn ucasefold(cls: &ClassUnicode) -> ClassUnicode {
let mut cls_ = cls.clone();
cls_.case_fold_simple();
cls_
}
fn uunion(cls1: &ClassUnicode, cls2: &ClassUnicode) -> ClassUnicode {
let mut cls_ = cls1.clone();
cls_.union(cls2);
cls_
}
fn uintersect(cls1: &ClassUnicode, cls2: &ClassUnicode) -> ClassUnicode {
let mut cls_ = cls1.clone();
cls_.intersect(cls2);
cls_
}
fn udifference(cls1: &ClassUnicode, cls2: &ClassUnicode) -> ClassUnicode {
let mut cls_ = cls1.clone();
cls_.difference(cls2);
cls_
}
fn usymdifference(
cls1: &ClassUnicode,
cls2: &ClassUnicode,
) -> ClassUnicode {
let mut cls_ = cls1.clone();
cls_.symmetric_difference(cls2);
cls_
}
fn unegate(cls: &ClassUnicode) -> ClassUnicode {
let mut cls_ = cls.clone();
cls_.negate();
cls_
}
fn branges(cls: &ClassBytes) -> Vec<(u8, u8)> {
cls.iter().map(|x| (x.start(), x.end())).collect()
}
fn bcasefold(cls: &ClassBytes) -> ClassBytes {
let mut cls_ = cls.clone();
cls_.case_fold_simple();
cls_
}
fn bunion(cls1: &ClassBytes, cls2: &ClassBytes) -> ClassBytes {
let mut cls_ = cls1.clone();
cls_.union(cls2);
cls_
}
fn bintersect(cls1: &ClassBytes, cls2: &ClassBytes) -> ClassBytes {
let mut cls_ = cls1.clone();
cls_.intersect(cls2);
cls_
}
fn bdifference(cls1: &ClassBytes, cls2: &ClassBytes) -> ClassBytes {
let mut cls_ = cls1.clone();
cls_.difference(cls2);
cls_
}
fn bsymdifference(cls1: &ClassBytes, cls2: &ClassBytes) -> ClassBytes {
let mut cls_ = cls1.clone();
cls_.symmetric_difference(cls2);
cls_
}
fn bnegate(cls: &ClassBytes) -> ClassBytes {
let mut cls_ = cls.clone();
cls_.negate();
cls_
}
#[test]
fn class_range_canonical_unicode() {
let range = ClassUnicodeRange::new('\u{00FF}', '\0');
assert_eq!('\0', range.start());
assert_eq!('\u{00FF}', range.end());
}
#[test]
fn class_range_canonical_bytes() {
let range = ClassBytesRange::new(b'\xFF', b'\0');
assert_eq!(b'\0', range.start());
assert_eq!(b'\xFF', range.end());
}
#[test]
fn class_canonicalize_unicode() {
let cls = uclass(&[('a', 'c'), ('x', 'z')]);
let expected = vec![('a', 'c'), ('x', 'z')];
assert_eq!(expected, uranges(&cls));
let cls = uclass(&[('x', 'z'), ('a', 'c')]);
let expected = vec![('a', 'c'), ('x', 'z')];
assert_eq!(expected, uranges(&cls));
let cls = uclass(&[('x', 'z'), ('w', 'y')]);
let expected = vec![('w', 'z')];
assert_eq!(expected, uranges(&cls));
let cls = uclass(&[
('c', 'f'),
('a', 'g'),
('d', 'j'),
('a', 'c'),
('m', 'p'),
('l', 's'),
]);
let expected = vec![('a', 'j'), ('l', 's')];
assert_eq!(expected, uranges(&cls));
let cls = uclass(&[('x', 'z'), ('u', 'w')]);
let expected = vec![('u', 'z')];
assert_eq!(expected, uranges(&cls));
let cls = uclass(&[('\x00', '\u{10FFFF}'), ('\x00', '\u{10FFFF}')]);
let expected = vec![('\x00', '\u{10FFFF}')];
assert_eq!(expected, uranges(&cls));
let cls = uclass(&[('a', 'a'), ('b', 'b')]);
let expected = vec![('a', 'b')];
assert_eq!(expected, uranges(&cls));
}
#[test]
fn class_canonicalize_bytes() {
let cls = bclass(&[(b'a', b'c'), (b'x', b'z')]);
let expected = vec![(b'a', b'c'), (b'x', b'z')];
assert_eq!(expected, branges(&cls));
let cls = bclass(&[(b'x', b'z'), (b'a', b'c')]);
let expected = vec![(b'a', b'c'), (b'x', b'z')];
assert_eq!(expected, branges(&cls));
let cls = bclass(&[(b'x', b'z'), (b'w', b'y')]);
let expected = vec![(b'w', b'z')];
assert_eq!(expected, branges(&cls));
let cls = bclass(&[
(b'c', b'f'),
(b'a', b'g'),
(b'd', b'j'),
(b'a', b'c'),
(b'm', b'p'),
(b'l', b's'),
]);
let expected = vec![(b'a', b'j'), (b'l', b's')];
assert_eq!(expected, branges(&cls));
let cls = bclass(&[(b'x', b'z'), (b'u', b'w')]);
let expected = vec![(b'u', b'z')];
assert_eq!(expected, branges(&cls));
let cls = bclass(&[(b'\x00', b'\xFF'), (b'\x00', b'\xFF')]);
let expected = vec![(b'\x00', b'\xFF')];
assert_eq!(expected, branges(&cls));
let cls = bclass(&[(b'a', b'a'), (b'b', b'b')]);
let expected = vec![(b'a', b'b')];
assert_eq!(expected, branges(&cls));
}
#[test]
#[cfg(feature = "unicode-case")]
fn class_case_fold_unicode() {
let cls = uclass(&[
('C', 'F'),
('A', 'G'),
('D', 'J'),
('A', 'C'),
('M', 'P'),
('L', 'S'),
('c', 'f'),
]);
let expected = uclass(&[
('A', 'J'),
('L', 'S'),
('a', 'j'),
('l', 's'),
('\u{17F}', '\u{17F}'),
]);
assert_eq!(expected, ucasefold(&cls));
let cls = uclass(&[('A', 'Z')]);
let expected = uclass(&[
('A', 'Z'),
('a', 'z'),
('\u{17F}', '\u{17F}'),
('\u{212A}', '\u{212A}'),
]);
assert_eq!(expected, ucasefold(&cls));
let cls = uclass(&[('a', 'z')]);
let expected = uclass(&[
('A', 'Z'),
('a', 'z'),
('\u{17F}', '\u{17F}'),
('\u{212A}', '\u{212A}'),
]);
assert_eq!(expected, ucasefold(&cls));
let cls = uclass(&[('A', 'A'), ('_', '_')]);
let expected = uclass(&[('A', 'A'), ('_', '_'), ('a', 'a')]);
assert_eq!(expected, ucasefold(&cls));
let cls = uclass(&[('A', 'A'), ('=', '=')]);
let expected = uclass(&[('=', '='), ('A', 'A'), ('a', 'a')]);
assert_eq!(expected, ucasefold(&cls));
let cls = uclass(&[('\x00', '\x10')]);
assert_eq!(cls, ucasefold(&cls));
let cls = uclass(&[('k', 'k')]);
let expected =
uclass(&[('K', 'K'), ('k', 'k'), ('\u{212A}', '\u{212A}')]);
assert_eq!(expected, ucasefold(&cls));
let cls = uclass(&[('@', '@')]);
assert_eq!(cls, ucasefold(&cls));
}
#[test]
#[cfg(not(feature = "unicode-case"))]
fn class_case_fold_unicode_disabled() {
let mut cls = uclass(&[
('C', 'F'),
('A', 'G'),
('D', 'J'),
('A', 'C'),
('M', 'P'),
('L', 'S'),
('c', 'f'),
]);
assert!(cls.try_case_fold_simple().is_err());
}
#[test]
#[should_panic]
#[cfg(not(feature = "unicode-case"))]
fn class_case_fold_unicode_disabled_panics() {
let mut cls = uclass(&[
('C', 'F'),
('A', 'G'),
('D', 'J'),
('A', 'C'),
('M', 'P'),
('L', 'S'),
('c', 'f'),
]);
cls.case_fold_simple();
}
#[test]
fn class_case_fold_bytes() {
let cls = bclass(&[
(b'C', b'F'),
(b'A', b'G'),
(b'D', b'J'),
(b'A', b'C'),
(b'M', b'P'),
(b'L', b'S'),
(b'c', b'f'),
]);
let expected =
bclass(&[(b'A', b'J'), (b'L', b'S'), (b'a', b'j'), (b'l', b's')]);
assert_eq!(expected, bcasefold(&cls));
let cls = bclass(&[(b'A', b'Z')]);
let expected = bclass(&[(b'A', b'Z'), (b'a', b'z')]);
assert_eq!(expected, bcasefold(&cls));
let cls = bclass(&[(b'a', b'z')]);
let expected = bclass(&[(b'A', b'Z'), (b'a', b'z')]);
assert_eq!(expected, bcasefold(&cls));
let cls = bclass(&[(b'A', b'A'), (b'_', b'_')]);
let expected = bclass(&[(b'A', b'A'), (b'_', b'_'), (b'a', b'a')]);
assert_eq!(expected, bcasefold(&cls));
let cls = bclass(&[(b'A', b'A'), (b'=', b'=')]);
let expected = bclass(&[(b'=', b'='), (b'A', b'A'), (b'a', b'a')]);
assert_eq!(expected, bcasefold(&cls));
let cls = bclass(&[(b'\x00', b'\x10')]);
assert_eq!(cls, bcasefold(&cls));
let cls = bclass(&[(b'k', b'k')]);
let expected = bclass(&[(b'K', b'K'), (b'k', b'k')]);
assert_eq!(expected, bcasefold(&cls));
let cls = bclass(&[(b'@', b'@')]);
assert_eq!(cls, bcasefold(&cls));
}
#[test]
fn class_negate_unicode() {
let cls = uclass(&[('a', 'a')]);
let expected = uclass(&[('\x00', '\x60'), ('\x62', '\u{10FFFF}')]);
assert_eq!(expected, unegate(&cls));
let cls = uclass(&[('a', 'a'), ('b', 'b')]);
let expected = uclass(&[('\x00', '\x60'), ('\x63', '\u{10FFFF}')]);
assert_eq!(expected, unegate(&cls));
let cls = uclass(&[('a', 'c'), ('x', 'z')]);
let expected = uclass(&[
('\x00', '\x60'),
('\x64', '\x77'),
('\x7B', '\u{10FFFF}'),
]);
assert_eq!(expected, unegate(&cls));
let cls = uclass(&[('\x00', 'a')]);
let expected = uclass(&[('\x62', '\u{10FFFF}')]);
assert_eq!(expected, unegate(&cls));
let cls = uclass(&[('a', '\u{10FFFF}')]);
let expected = uclass(&[('\x00', '\x60')]);
assert_eq!(expected, unegate(&cls));
let cls = uclass(&[('\x00', '\u{10FFFF}')]);
let expected = uclass(&[]);
assert_eq!(expected, unegate(&cls));
let cls = uclass(&[]);
let expected = uclass(&[('\x00', '\u{10FFFF}')]);
assert_eq!(expected, unegate(&cls));
let cls =
uclass(&[('\x00', '\u{10FFFD}'), ('\u{10FFFF}', '\u{10FFFF}')]);
let expected = uclass(&[('\u{10FFFE}', '\u{10FFFE}')]);
assert_eq!(expected, unegate(&cls));
let cls = uclass(&[('\x00', '\u{D7FF}')]);
let expected = uclass(&[('\u{E000}', '\u{10FFFF}')]);
assert_eq!(expected, unegate(&cls));
let cls = uclass(&[('\x00', '\u{D7FE}')]);
let expected = uclass(&[('\u{D7FF}', '\u{10FFFF}')]);
assert_eq!(expected, unegate(&cls));
let cls = uclass(&[('\u{E000}', '\u{10FFFF}')]);
let expected = uclass(&[('\x00', '\u{D7FF}')]);
assert_eq!(expected, unegate(&cls));
let cls = uclass(&[('\u{E001}', '\u{10FFFF}')]);
let expected = uclass(&[('\x00', '\u{E000}')]);
assert_eq!(expected, unegate(&cls));
}
#[test]
fn class_negate_bytes() {
let cls = bclass(&[(b'a', b'a')]);
let expected = bclass(&[(b'\x00', b'\x60'), (b'\x62', b'\xFF')]);
assert_eq!(expected, bnegate(&cls));
let cls = bclass(&[(b'a', b'a'), (b'b', b'b')]);
let expected = bclass(&[(b'\x00', b'\x60'), (b'\x63', b'\xFF')]);
assert_eq!(expected, bnegate(&cls));
let cls = bclass(&[(b'a', b'c'), (b'x', b'z')]);
let expected = bclass(&[
(b'\x00', b'\x60'),
(b'\x64', b'\x77'),
(b'\x7B', b'\xFF'),
]);
assert_eq!(expected, bnegate(&cls));
let cls = bclass(&[(b'\x00', b'a')]);
let expected = bclass(&[(b'\x62', b'\xFF')]);
assert_eq!(expected, bnegate(&cls));
let cls = bclass(&[(b'a', b'\xFF')]);
let expected = bclass(&[(b'\x00', b'\x60')]);
assert_eq!(expected, bnegate(&cls));
let cls = bclass(&[(b'\x00', b'\xFF')]);
let expected = bclass(&[]);
assert_eq!(expected, bnegate(&cls));
let cls = bclass(&[]);
let expected = bclass(&[(b'\x00', b'\xFF')]);
assert_eq!(expected, bnegate(&cls));
let cls = bclass(&[(b'\x00', b'\xFD'), (b'\xFF', b'\xFF')]);
let expected = bclass(&[(b'\xFE', b'\xFE')]);
assert_eq!(expected, bnegate(&cls));
}
#[test]
fn class_union_unicode() {
let cls1 = uclass(&[('a', 'g'), ('m', 't'), ('A', 'C')]);
let cls2 = uclass(&[('a', 'z')]);
let expected = uclass(&[('a', 'z'), ('A', 'C')]);
assert_eq!(expected, uunion(&cls1, &cls2));
}
#[test]
fn class_union_bytes() {
let cls1 = bclass(&[(b'a', b'g'), (b'm', b't'), (b'A', b'C')]);
let cls2 = bclass(&[(b'a', b'z')]);
let expected = bclass(&[(b'a', b'z'), (b'A', b'C')]);
assert_eq!(expected, bunion(&cls1, &cls2));
}
#[test]
fn class_intersect_unicode() {
let cls1 = uclass(&[]);
let cls2 = uclass(&[('a', 'a')]);
let expected = uclass(&[]);
assert_eq!(expected, uintersect(&cls1, &cls2));
let cls1 = uclass(&[('a', 'a')]);
let cls2 = uclass(&[('a', 'a')]);
let expected = uclass(&[('a', 'a')]);
assert_eq!(expected, uintersect(&cls1, &cls2));
let cls1 = uclass(&[('a', 'a')]);
let cls2 = uclass(&[('b', 'b')]);
let expected = uclass(&[]);
assert_eq!(expected, uintersect(&cls1, &cls2));
let cls1 = uclass(&[('a', 'a')]);
let cls2 = uclass(&[('a', 'c')]);
let expected = uclass(&[('a', 'a')]);
assert_eq!(expected, uintersect(&cls1, &cls2));
let cls1 = uclass(&[('a', 'b')]);
let cls2 = uclass(&[('a', 'c')]);
let expected = uclass(&[('a', 'b')]);
assert_eq!(expected, uintersect(&cls1, &cls2));
let cls1 = uclass(&[('a', 'b')]);
let cls2 = uclass(&[('b', 'c')]);
let expected = uclass(&[('b', 'b')]);
assert_eq!(expected, uintersect(&cls1, &cls2));
let cls1 = uclass(&[('a', 'b')]);
let cls2 = uclass(&[('c', 'd')]);
let expected = uclass(&[]);
assert_eq!(expected, uintersect(&cls1, &cls2));
let cls1 = uclass(&[('b', 'c')]);
let cls2 = uclass(&[('a', 'd')]);
let expected = uclass(&[('b', 'c')]);
assert_eq!(expected, uintersect(&cls1, &cls2));
let cls1 = uclass(&[('a', 'b'), ('d', 'e'), ('g', 'h')]);
let cls2 = uclass(&[('a', 'h')]);
let expected = uclass(&[('a', 'b'), ('d', 'e'), ('g', 'h')]);
assert_eq!(expected, uintersect(&cls1, &cls2));
let cls1 = uclass(&[('a', 'b'), ('d', 'e'), ('g', 'h')]);
let cls2 = uclass(&[('a', 'b'), ('d', 'e'), ('g', 'h')]);
let expected = uclass(&[('a', 'b'), ('d', 'e'), ('g', 'h')]);
assert_eq!(expected, uintersect(&cls1, &cls2));
let cls1 = uclass(&[('a', 'b'), ('g', 'h')]);
let cls2 = uclass(&[('d', 'e'), ('k', 'l')]);
let expected = uclass(&[]);
assert_eq!(expected, uintersect(&cls1, &cls2));
let cls1 = uclass(&[('a', 'b'), ('d', 'e'), ('g', 'h')]);
let cls2 = uclass(&[('h', 'h')]);
let expected = uclass(&[('h', 'h')]);
assert_eq!(expected, uintersect(&cls1, &cls2));
let cls1 = uclass(&[('a', 'b'), ('e', 'f'), ('i', 'j')]);
let cls2 = uclass(&[('c', 'd'), ('g', 'h'), ('k', 'l')]);
let expected = uclass(&[]);
assert_eq!(expected, uintersect(&cls1, &cls2));
let cls1 = uclass(&[('a', 'b'), ('c', 'd'), ('e', 'f')]);
let cls2 = uclass(&[('b', 'c'), ('d', 'e'), ('f', 'g')]);
let expected = uclass(&[('b', 'f')]);
assert_eq!(expected, uintersect(&cls1, &cls2));
}
#[test]
fn class_intersect_bytes() {
let cls1 = bclass(&[]);
let cls2 = bclass(&[(b'a', b'a')]);
let expected = bclass(&[]);
assert_eq!(expected, bintersect(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'a')]);
let cls2 = bclass(&[(b'a', b'a')]);
let expected = bclass(&[(b'a', b'a')]);
assert_eq!(expected, bintersect(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'a')]);
let cls2 = bclass(&[(b'b', b'b')]);
let expected = bclass(&[]);
assert_eq!(expected, bintersect(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'a')]);
let cls2 = bclass(&[(b'a', b'c')]);
let expected = bclass(&[(b'a', b'a')]);
assert_eq!(expected, bintersect(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'b')]);
let cls2 = bclass(&[(b'a', b'c')]);
let expected = bclass(&[(b'a', b'b')]);
assert_eq!(expected, bintersect(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'b')]);
let cls2 = bclass(&[(b'b', b'c')]);
let expected = bclass(&[(b'b', b'b')]);
assert_eq!(expected, bintersect(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'b')]);
let cls2 = bclass(&[(b'c', b'd')]);
let expected = bclass(&[]);
assert_eq!(expected, bintersect(&cls1, &cls2));
let cls1 = bclass(&[(b'b', b'c')]);
let cls2 = bclass(&[(b'a', b'd')]);
let expected = bclass(&[(b'b', b'c')]);
assert_eq!(expected, bintersect(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'b'), (b'd', b'e'), (b'g', b'h')]);
let cls2 = bclass(&[(b'a', b'h')]);
let expected = bclass(&[(b'a', b'b'), (b'd', b'e'), (b'g', b'h')]);
assert_eq!(expected, bintersect(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'b'), (b'd', b'e'), (b'g', b'h')]);
let cls2 = bclass(&[(b'a', b'b'), (b'd', b'e'), (b'g', b'h')]);
let expected = bclass(&[(b'a', b'b'), (b'd', b'e'), (b'g', b'h')]);
assert_eq!(expected, bintersect(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'b'), (b'g', b'h')]);
let cls2 = bclass(&[(b'd', b'e'), (b'k', b'l')]);
let expected = bclass(&[]);
assert_eq!(expected, bintersect(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'b'), (b'd', b'e'), (b'g', b'h')]);
let cls2 = bclass(&[(b'h', b'h')]);
let expected = bclass(&[(b'h', b'h')]);
assert_eq!(expected, bintersect(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'b'), (b'e', b'f'), (b'i', b'j')]);
let cls2 = bclass(&[(b'c', b'd'), (b'g', b'h'), (b'k', b'l')]);
let expected = bclass(&[]);
assert_eq!(expected, bintersect(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'b'), (b'c', b'd'), (b'e', b'f')]);
let cls2 = bclass(&[(b'b', b'c'), (b'd', b'e'), (b'f', b'g')]);
let expected = bclass(&[(b'b', b'f')]);
assert_eq!(expected, bintersect(&cls1, &cls2));
}
#[test]
fn class_difference_unicode() {
let cls1 = uclass(&[('a', 'a')]);
let cls2 = uclass(&[('a', 'a')]);
let expected = uclass(&[]);
assert_eq!(expected, udifference(&cls1, &cls2));
let cls1 = uclass(&[('a', 'a')]);
let cls2 = uclass(&[]);
let expected = uclass(&[('a', 'a')]);
assert_eq!(expected, udifference(&cls1, &cls2));
let cls1 = uclass(&[]);
let cls2 = uclass(&[('a', 'a')]);
let expected = uclass(&[]);
assert_eq!(expected, udifference(&cls1, &cls2));
let cls1 = uclass(&[('a', 'z')]);
let cls2 = uclass(&[('a', 'a')]);
let expected = uclass(&[('b', 'z')]);
assert_eq!(expected, udifference(&cls1, &cls2));
let cls1 = uclass(&[('a', 'z')]);
let cls2 = uclass(&[('z', 'z')]);
let expected = uclass(&[('a', 'y')]);
assert_eq!(expected, udifference(&cls1, &cls2));
let cls1 = uclass(&[('a', 'z')]);
let cls2 = uclass(&[('m', 'm')]);
let expected = uclass(&[('a', 'l'), ('n', 'z')]);
assert_eq!(expected, udifference(&cls1, &cls2));
let cls1 = uclass(&[('a', 'c'), ('g', 'i'), ('r', 't')]);
let cls2 = uclass(&[('a', 'z')]);
let expected = uclass(&[]);
assert_eq!(expected, udifference(&cls1, &cls2));
let cls1 = uclass(&[('a', 'c'), ('g', 'i'), ('r', 't')]);
let cls2 = uclass(&[('d', 'v')]);
let expected = uclass(&[('a', 'c')]);
assert_eq!(expected, udifference(&cls1, &cls2));
let cls1 = uclass(&[('a', 'c'), ('g', 'i'), ('r', 't')]);
let cls2 = uclass(&[('b', 'g'), ('s', 'u')]);
let expected = uclass(&[('a', 'a'), ('h', 'i'), ('r', 'r')]);
assert_eq!(expected, udifference(&cls1, &cls2));
let cls1 = uclass(&[('a', 'c'), ('g', 'i'), ('r', 't')]);
let cls2 = uclass(&[('b', 'd'), ('e', 'g'), ('s', 'u')]);
let expected = uclass(&[('a', 'a'), ('h', 'i'), ('r', 'r')]);
assert_eq!(expected, udifference(&cls1, &cls2));
let cls1 = uclass(&[('x', 'z')]);
let cls2 = uclass(&[('a', 'c'), ('e', 'g'), ('s', 'u')]);
let expected = uclass(&[('x', 'z')]);
assert_eq!(expected, udifference(&cls1, &cls2));
let cls1 = uclass(&[('a', 'z')]);
let cls2 = uclass(&[('a', 'c'), ('e', 'g'), ('s', 'u')]);
let expected = uclass(&[('d', 'd'), ('h', 'r'), ('v', 'z')]);
assert_eq!(expected, udifference(&cls1, &cls2));
}
#[test]
fn class_difference_bytes() {
let cls1 = bclass(&[(b'a', b'a')]);
let cls2 = bclass(&[(b'a', b'a')]);
let expected = bclass(&[]);
assert_eq!(expected, bdifference(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'a')]);
let cls2 = bclass(&[]);
let expected = bclass(&[(b'a', b'a')]);
assert_eq!(expected, bdifference(&cls1, &cls2));
let cls1 = bclass(&[]);
let cls2 = bclass(&[(b'a', b'a')]);
let expected = bclass(&[]);
assert_eq!(expected, bdifference(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'z')]);
let cls2 = bclass(&[(b'a', b'a')]);
let expected = bclass(&[(b'b', b'z')]);
assert_eq!(expected, bdifference(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'z')]);
let cls2 = bclass(&[(b'z', b'z')]);
let expected = bclass(&[(b'a', b'y')]);
assert_eq!(expected, bdifference(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'z')]);
let cls2 = bclass(&[(b'm', b'm')]);
let expected = bclass(&[(b'a', b'l'), (b'n', b'z')]);
assert_eq!(expected, bdifference(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'c'), (b'g', b'i'), (b'r', b't')]);
let cls2 = bclass(&[(b'a', b'z')]);
let expected = bclass(&[]);
assert_eq!(expected, bdifference(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'c'), (b'g', b'i'), (b'r', b't')]);
let cls2 = bclass(&[(b'd', b'v')]);
let expected = bclass(&[(b'a', b'c')]);
assert_eq!(expected, bdifference(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'c'), (b'g', b'i'), (b'r', b't')]);
let cls2 = bclass(&[(b'b', b'g'), (b's', b'u')]);
let expected = bclass(&[(b'a', b'a'), (b'h', b'i'), (b'r', b'r')]);
assert_eq!(expected, bdifference(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'c'), (b'g', b'i'), (b'r', b't')]);
let cls2 = bclass(&[(b'b', b'd'), (b'e', b'g'), (b's', b'u')]);
let expected = bclass(&[(b'a', b'a'), (b'h', b'i'), (b'r', b'r')]);
assert_eq!(expected, bdifference(&cls1, &cls2));
let cls1 = bclass(&[(b'x', b'z')]);
let cls2 = bclass(&[(b'a', b'c'), (b'e', b'g'), (b's', b'u')]);
let expected = bclass(&[(b'x', b'z')]);
assert_eq!(expected, bdifference(&cls1, &cls2));
let cls1 = bclass(&[(b'a', b'z')]);
let cls2 = bclass(&[(b'a', b'c'), (b'e', b'g'), (b's', b'u')]);
let expected = bclass(&[(b'd', b'd'), (b'h', b'r'), (b'v', b'z')]);
assert_eq!(expected, bdifference(&cls1, &cls2));
}
#[test]
fn class_symmetric_difference_unicode() {
let cls1 = uclass(&[('a', 'm')]);
let cls2 = uclass(&[('g', 't')]);
let expected = uclass(&[('a', 'f'), ('n', 't')]);
assert_eq!(expected, usymdifference(&cls1, &cls2));
}
#[test]
fn class_symmetric_difference_bytes() {
let cls1 = bclass(&[(b'a', b'm')]);
let cls2 = bclass(&[(b'g', b't')]);
let expected = bclass(&[(b'a', b'f'), (b'n', b't')]);
assert_eq!(expected, bsymdifference(&cls1, &cls2));
}
#[test]
#[should_panic]
fn hir_byte_literal_non_ascii() {
Hir::literal(Literal::Byte(b'a'));
}
// We use a thread with an explicit stack size to test that our destructor
// for Hir can handle arbitrarily sized expressions in constant stack
// space. In case we run on a platform without threads (WASM?), we limit
// this test to Windows/Unix.
#[test]
#[cfg(any(unix, windows))]
fn no_stack_overflow_on_drop() {
use std::thread;
let run = || {
let mut expr = Hir::empty();
for _ in 0..100 {
expr = Hir::group(Group {
kind: GroupKind::NonCapturing,
hir: Box::new(expr),
});
expr = Hir::repetition(Repetition {
kind: RepetitionKind::ZeroOrOne,
greedy: true,
hir: Box::new(expr),
});
expr = Hir {
kind: HirKind::Concat(vec![expr]),
info: HirInfo::new(),
};
expr = Hir {
kind: HirKind::Alternation(vec![expr]),
info: HirInfo::new(),
};
}
assert!(!expr.kind.is_empty());
};
// We run our test on a thread with a small stack size so we can
// force the issue more easily.
//
// NOTE(2023-03-21): See the corresponding test in 'crate::ast::tests'
// for context on the specific stack size chosen here.
thread::Builder::new()
.stack_size(16 << 10)
.spawn(run)
.unwrap()
.join()
.unwrap();
}
}