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
use std::collections::HashMap; use grep_matcher::{ Captures, LineMatchKind, LineTerminator, Match, Matcher, NoError, ByteSet, }; use regex::bytes::{CaptureLocations, Regex}; use config::{Config, ConfiguredHIR}; use crlf::CRLFMatcher; use error::Error; use multi::MultiLiteralMatcher; use word::WordMatcher; /// A builder for constructing a `Matcher` using regular expressions. /// /// This builder re-exports many of the same options found on the regex crate's /// builder, in addition to a few other options such as smart case, word /// matching and the ability to set a line terminator which may enable certain /// types of optimizations. /// /// The syntax supported is documented as part of the regex crate: /// https://docs.rs/regex/*/regex/#syntax #[derive(Clone, Debug)] pub struct RegexMatcherBuilder { config: Config, } impl Default for RegexMatcherBuilder { fn default() -> RegexMatcherBuilder { RegexMatcherBuilder::new() } } impl RegexMatcherBuilder { /// Create a new builder for configuring a regex matcher. pub fn new() -> RegexMatcherBuilder { RegexMatcherBuilder { config: Config::default(), } } /// Build a new matcher using the current configuration for the provided /// pattern. /// /// The syntax supported is documented as part of the regex crate: /// https://docs.rs/regex/*/regex/#syntax pub fn build(&self, pattern: &str) -> Result<RegexMatcher, Error> { let chir = self.config.hir(pattern)?; let fast_line_regex = chir.fast_line_regex()?; let non_matching_bytes = chir.non_matching_bytes(); if let Some(ref re) = fast_line_regex { trace!("extracted fast line regex: {:?}", re); } let matcher = RegexMatcherImpl::new(&chir)?; trace!("final regex: {:?}", matcher.regex()); Ok(RegexMatcher { config: self.config.clone(), matcher: matcher, fast_line_regex: fast_line_regex, non_matching_bytes: non_matching_bytes, }) } /// Build a new matcher from a plain alternation of literals. /// /// Depending on the configuration set by the builder, this may be able to /// build a matcher substantially faster than by joining the patterns with /// a `|` and calling `build`. pub fn build_literals<B: AsRef<str>>( &self, literals: &[B], ) -> Result<RegexMatcher, Error> { let slices: Vec<_> = literals.iter().map(|s| s.as_ref()).collect(); if !self.config.can_plain_aho_corasick() || literals.len() < 40 { return self.build(&slices.join("|")); } let matcher = MultiLiteralMatcher::new(&slices)?; let imp = RegexMatcherImpl::MultiLiteral(matcher); Ok(RegexMatcher { config: self.config.clone(), matcher: imp, fast_line_regex: None, non_matching_bytes: ByteSet::empty(), }) } /// Set the value for the case insensitive (`i`) flag. /// /// When enabled, letters in the pattern will match both upper case and /// lower case variants. pub fn case_insensitive(&mut self, yes: bool) -> &mut RegexMatcherBuilder { self.config.case_insensitive = yes; self } /// Whether to enable "smart case" or not. /// /// When smart case is enabled, the builder will automatically enable /// case insensitive matching based on how the pattern is written. Namely, /// case insensitive mode is enabled when both of the following things /// are true: /// /// 1. The pattern contains at least one literal character. For example, /// `a\w` contains a literal (`a`) but `\w` does not. /// 2. Of the literals in the pattern, none of them are considered to be /// uppercase according to Unicode. For example, `foo\pL` has no /// uppercase literals but `Foo\pL` does. pub fn case_smart(&mut self, yes: bool) -> &mut RegexMatcherBuilder { self.config.case_smart = yes; self } /// Set the value for the multi-line matching (`m`) flag. /// /// When enabled, `^` matches the beginning of lines and `$` matches the /// end of lines. /// /// By default, they match beginning/end of the input. pub fn multi_line(&mut self, yes: bool) -> &mut RegexMatcherBuilder { self.config.multi_line = yes; self } /// Set the value for the any character (`s`) flag, where in `.` matches /// anything when `s` is set and matches anything except for new line when /// it is not set (the default). /// /// N.B. "matches anything" means "any byte" when Unicode is disabled and /// means "any valid UTF-8 encoding of any Unicode scalar value" when /// Unicode is enabled. pub fn dot_matches_new_line( &mut self, yes: bool, ) -> &mut RegexMatcherBuilder { self.config.dot_matches_new_line = yes; self } /// Set the value for the greedy swap (`U`) flag. /// /// When enabled, a pattern like `a*` is lazy (tries to find shortest /// match) and `a*?` is greedy (tries to find longest match). /// /// By default, `a*` is greedy and `a*?` is lazy. pub fn swap_greed(&mut self, yes: bool) -> &mut RegexMatcherBuilder { self.config.swap_greed = yes; self } /// Set the value for the ignore whitespace (`x`) flag. /// /// When enabled, whitespace such as new lines and spaces will be ignored /// between expressions of the pattern, and `#` can be used to start a /// comment until the next new line. pub fn ignore_whitespace( &mut self, yes: bool, ) -> &mut RegexMatcherBuilder { self.config.ignore_whitespace = yes; self } /// Set the value for the Unicode (`u`) flag. /// /// Enabled by default. When disabled, character classes such as `\w` only /// match ASCII word characters instead of all Unicode word characters. pub fn unicode(&mut self, yes: bool) -> &mut RegexMatcherBuilder { self.config.unicode = yes; self } /// Whether to support octal syntax or not. /// /// Octal syntax is a little-known way of uttering Unicode codepoints in /// a regular expression. For example, `a`, `\x61`, `\u0061` and /// `\141` are all equivalent regular expressions, where the last example /// shows octal syntax. /// /// While supporting octal syntax isn't in and of itself a problem, it does /// make good error messages harder. That is, in PCRE based regex engines, /// syntax like `\0` invokes a backreference, which is explicitly /// unsupported in Rust's regex engine. However, many users expect it to /// be supported. Therefore, when octal support is disabled, the error /// message will explicitly mention that backreferences aren't supported. /// /// Octal syntax is disabled by default. pub fn octal(&mut self, yes: bool) -> &mut RegexMatcherBuilder { self.config.octal = yes; self } /// Set the approximate size limit of the compiled regular expression. /// /// This roughly corresponds to the number of bytes occupied by a single /// compiled program. If the program exceeds this number, then a /// compilation error is returned. pub fn size_limit(&mut self, bytes: usize) -> &mut RegexMatcherBuilder { self.config.size_limit = bytes; self } /// Set the approximate size of the cache used by the DFA. /// /// This roughly corresponds to the number of bytes that the DFA will /// use while searching. /// /// Note that this is a *per thread* limit. There is no way to set a global /// limit. In particular, if a regex is used from multiple threads /// simultaneously, then each thread may use up to the number of bytes /// specified here. pub fn dfa_size_limit( &mut self, bytes: usize, ) -> &mut RegexMatcherBuilder { self.config.dfa_size_limit = bytes; self } /// Set the nesting limit for this parser. /// /// The nesting limit controls how deep the abstract syntax tree is allowed /// to be. If the AST exceeds the given limit (e.g., with too many nested /// groups), then an error is returned by the parser. /// /// The purpose of this limit is to act as a heuristic to prevent stack /// overflow for consumers that do structural induction on an `Ast` using /// explicit recursion. While this crate never does this (instead using /// constant stack space and moving the call stack to the heap), other /// crates may. /// /// This limit is not checked until the entire Ast is parsed. Therefore, /// if callers want to put a limit on the amount of heap space used, then /// they should impose a limit on the length, in bytes, of the concrete /// pattern string. In particular, this is viable since this parser /// implementation will limit itself to heap space proportional to the /// lenth of the pattern string. /// /// Note that a nest limit of `0` will return a nest limit error for most /// patterns but not all. For example, a nest limit of `0` permits `a` but /// not `ab`, since `ab` requires a concatenation, which results in a nest /// depth of `1`. In general, a nest limit is not something that manifests /// in an obvious way in the concrete syntax, therefore, it should not be /// used in a granular way. pub fn nest_limit(&mut self, limit: u32) -> &mut RegexMatcherBuilder { self.config.nest_limit = limit; self } /// Set an ASCII line terminator for the matcher. /// /// The purpose of setting a line terminator is to enable a certain class /// of optimizations that can make line oriented searching faster. Namely, /// when a line terminator is enabled, then the builder will guarantee that /// the resulting matcher will never be capable of producing a match that /// contains the line terminator. Because of this guarantee, users of the /// resulting matcher do not need to slowly execute a search line by line /// for line oriented search. /// /// If the aforementioned guarantee about not matching a line terminator /// cannot be made because of how the pattern was written, then the builder /// will return an error when attempting to construct the matcher. For /// example, the pattern `a\sb` will be transformed such that it can never /// match `a\nb` (when `\n` is the line terminator), but the pattern `a\nb` /// will result in an error since the `\n` cannot be easily removed without /// changing the fundamental intent of the pattern. /// /// If the given line terminator isn't an ASCII byte (`<=127`), then the /// builder will return an error when constructing the matcher. pub fn line_terminator( &mut self, line_term: Option<u8>, ) -> &mut RegexMatcherBuilder { self.config.line_terminator = line_term.map(LineTerminator::byte); self } /// Set the line terminator to `\r\n` and enable CRLF matching for `$` in /// regex patterns. /// /// This method sets two distinct settings: /// /// 1. It causes the line terminator for the matcher to be `\r\n`. Namely, /// this prevents the matcher from ever producing a match that contains /// a `\r` or `\n`. /// 2. It translates all instances of `$` in the pattern to `(?:\r??$)`. /// This works around the fact that the regex engine does not support /// matching CRLF as a line terminator when using `$`. /// /// In particular, because of (2), the matches produced by the matcher may /// be slightly different than what one would expect given the pattern. /// This is the trade off made: in many cases, `$` will "just work" in the /// presence of `\r\n` line terminators, but matches may require some /// trimming to faithfully represent the intended match. /// /// Note that if you do not wish to set the line terminator but would still /// like `$` to match `\r\n` line terminators, then it is valid to call /// `crlf(true)` followed by `line_terminator(None)`. Ordering is /// important, since `crlf` and `line_terminator` override each other. pub fn crlf(&mut self, yes: bool) -> &mut RegexMatcherBuilder { if yes { self.config.line_terminator = Some(LineTerminator::crlf()); } else { self.config.line_terminator = None; } self.config.crlf = yes; self } /// Require that all matches occur on word boundaries. /// /// Enabling this option is subtly different than putting `\b` assertions /// on both sides of your pattern. In particular, a `\b` assertion requires /// that one side of it match a word character while the other match a /// non-word character. This option, in contrast, merely requires that /// one side match a non-word character. /// /// For example, `\b-2\b` will not match `foo -2 bar` since `-` is not a /// word character. However, `-2` with this `word` option enabled will /// match the `-2` in `foo -2 bar`. pub fn word(&mut self, yes: bool) -> &mut RegexMatcherBuilder { self.config.word = yes; self } } /// An implementation of the `Matcher` trait using Rust's standard regex /// library. #[derive(Clone, Debug)] pub struct RegexMatcher { /// The configuration specified by the caller. config: Config, /// The underlying matcher implementation. matcher: RegexMatcherImpl, /// A regex that never reports false negatives but may report false /// positives that is believed to be capable of being matched more quickly /// than `regex`. Typically, this is a single literal or an alternation /// of literals. fast_line_regex: Option<Regex>, /// A set of bytes that will never appear in a match. non_matching_bytes: ByteSet, } impl RegexMatcher { /// Create a new matcher from the given pattern using the default /// configuration. pub fn new(pattern: &str) -> Result<RegexMatcher, Error> { RegexMatcherBuilder::new().build(pattern) } /// Create a new matcher from the given pattern using the default /// configuration, but matches lines terminated by `\n`. /// /// This is meant to be a convenience constructor for using a /// `RegexMatcherBuilder` and setting its /// [`line_terminator`](struct.RegexMatcherBuilder.html#method.line_terminator) /// to `\n`. The purpose of using this constructor is to permit special /// optimizations that help speed up line oriented search. These types of /// optimizations are only appropriate when matches span no more than one /// line. For this reason, this constructor will return an error if the /// given pattern contains a literal `\n`. Other uses of `\n` (such as in /// `\s`) are removed transparently. pub fn new_line_matcher(pattern: &str) -> Result<RegexMatcher, Error> { RegexMatcherBuilder::new() .line_terminator(Some(b'\n')) .build(pattern) } } /// An encapsulation of the type of matcher we use in `RegexMatcher`. #[derive(Clone, Debug)] enum RegexMatcherImpl { /// The standard matcher used for all regular expressions. Standard(StandardMatcher), /// A matcher for an alternation of plain literals. MultiLiteral(MultiLiteralMatcher), /// A matcher that strips `\r` from the end of matches. /// /// This is only used when the CRLF hack is enabled and the regex is line /// anchored at the end. CRLF(CRLFMatcher), /// A matcher that only matches at word boundaries. This transforms the /// regex to `(^|\W)(...)($|\W)` instead of the more intuitive `\b(...)\b`. /// Because of this, the WordMatcher provides its own implementation of /// `Matcher` to encapsulate its use of capture groups to make them /// invisible to the caller. Word(WordMatcher), } impl RegexMatcherImpl { /// Based on the configuration, create a new implementation of the /// `Matcher` trait. fn new(expr: &ConfiguredHIR) -> Result<RegexMatcherImpl, Error> { if expr.config().word { Ok(RegexMatcherImpl::Word(WordMatcher::new(expr)?)) } else if expr.needs_crlf_stripped() { Ok(RegexMatcherImpl::CRLF(CRLFMatcher::new(expr)?)) } else { if let Some(lits) = expr.alternation_literals() { if lits.len() >= 40 { let matcher = MultiLiteralMatcher::new(&lits)?; return Ok(RegexMatcherImpl::MultiLiteral(matcher)); } } Ok(RegexMatcherImpl::Standard(StandardMatcher::new(expr)?)) } } /// Return the underlying regex object used. fn regex(&self) -> String { match *self { RegexMatcherImpl::Word(ref x) => x.regex().to_string(), RegexMatcherImpl::CRLF(ref x) => x.regex().to_string(), RegexMatcherImpl::MultiLiteral(_) => "<N/A>".to_string(), RegexMatcherImpl::Standard(ref x) => x.regex.to_string(), } } } // This implementation just dispatches on the internal matcher impl except // for the line terminator optimization, which is possibly executed via // `fast_line_regex`. impl Matcher for RegexMatcher { type Captures = RegexCaptures; type Error = NoError; fn find_at( &self, haystack: &[u8], at: usize, ) -> Result<Option<Match>, NoError> { use self::RegexMatcherImpl::*; match self.matcher { Standard(ref m) => m.find_at(haystack, at), MultiLiteral(ref m) => m.find_at(haystack, at), CRLF(ref m) => m.find_at(haystack, at), Word(ref m) => m.find_at(haystack, at), } } fn new_captures(&self) -> Result<RegexCaptures, NoError> { use self::RegexMatcherImpl::*; match self.matcher { Standard(ref m) => m.new_captures(), MultiLiteral(ref m) => m.new_captures(), CRLF(ref m) => m.new_captures(), Word(ref m) => m.new_captures(), } } fn capture_count(&self) -> usize { use self::RegexMatcherImpl::*; match self.matcher { Standard(ref m) => m.capture_count(), MultiLiteral(ref m) => m.capture_count(), CRLF(ref m) => m.capture_count(), Word(ref m) => m.capture_count(), } } fn capture_index(&self, name: &str) -> Option<usize> { use self::RegexMatcherImpl::*; match self.matcher { Standard(ref m) => m.capture_index(name), MultiLiteral(ref m) => m.capture_index(name), CRLF(ref m) => m.capture_index(name), Word(ref m) => m.capture_index(name), } } fn find(&self, haystack: &[u8]) -> Result<Option<Match>, NoError> { use self::RegexMatcherImpl::*; match self.matcher { Standard(ref m) => m.find(haystack), MultiLiteral(ref m) => m.find(haystack), CRLF(ref m) => m.find(haystack), Word(ref m) => m.find(haystack), } } fn find_iter<F>( &self, haystack: &[u8], matched: F, ) -> Result<(), NoError> where F: FnMut(Match) -> bool { use self::RegexMatcherImpl::*; match self.matcher { Standard(ref m) => m.find_iter(haystack, matched), MultiLiteral(ref m) => m.find_iter(haystack, matched), CRLF(ref m) => m.find_iter(haystack, matched), Word(ref m) => m.find_iter(haystack, matched), } } fn try_find_iter<F, E>( &self, haystack: &[u8], matched: F, ) -> Result<Result<(), E>, NoError> where F: FnMut(Match) -> Result<bool, E> { use self::RegexMatcherImpl::*; match self.matcher { Standard(ref m) => m.try_find_iter(haystack, matched), MultiLiteral(ref m) => m.try_find_iter(haystack, matched), CRLF(ref m) => m.try_find_iter(haystack, matched), Word(ref m) => m.try_find_iter(haystack, matched), } } fn captures( &self, haystack: &[u8], caps: &mut RegexCaptures, ) -> Result<bool, NoError> { use self::RegexMatcherImpl::*; match self.matcher { Standard(ref m) => m.captures(haystack, caps), MultiLiteral(ref m) => m.captures(haystack, caps), CRLF(ref m) => m.captures(haystack, caps), Word(ref m) => m.captures(haystack, caps), } } fn captures_iter<F>( &self, haystack: &[u8], caps: &mut RegexCaptures, matched: F, ) -> Result<(), NoError> where F: FnMut(&RegexCaptures) -> bool { use self::RegexMatcherImpl::*; match self.matcher { Standard(ref m) => m.captures_iter(haystack, caps, matched), MultiLiteral(ref m) => m.captures_iter(haystack, caps, matched), CRLF(ref m) => m.captures_iter(haystack, caps, matched), Word(ref m) => m.captures_iter(haystack, caps, matched), } } fn try_captures_iter<F, E>( &self, haystack: &[u8], caps: &mut RegexCaptures, matched: F, ) -> Result<Result<(), E>, NoError> where F: FnMut(&RegexCaptures) -> Result<bool, E> { use self::RegexMatcherImpl::*; match self.matcher { Standard(ref m) => m.try_captures_iter(haystack, caps, matched), MultiLiteral(ref m) => { m.try_captures_iter(haystack, caps, matched) } CRLF(ref m) => m.try_captures_iter(haystack, caps, matched), Word(ref m) => m.try_captures_iter(haystack, caps, matched), } } fn captures_at( &self, haystack: &[u8], at: usize, caps: &mut RegexCaptures, ) -> Result<bool, NoError> { use self::RegexMatcherImpl::*; match self.matcher { Standard(ref m) => m.captures_at(haystack, at, caps), MultiLiteral(ref m) => m.captures_at(haystack, at, caps), CRLF(ref m) => m.captures_at(haystack, at, caps), Word(ref m) => m.captures_at(haystack, at, caps), } } fn replace<F>( &self, haystack: &[u8], dst: &mut Vec<u8>, append: F, ) -> Result<(), NoError> where F: FnMut(Match, &mut Vec<u8>) -> bool { use self::RegexMatcherImpl::*; match self.matcher { Standard(ref m) => m.replace(haystack, dst, append), MultiLiteral(ref m) => m.replace(haystack, dst, append), CRLF(ref m) => m.replace(haystack, dst, append), Word(ref m) => m.replace(haystack, dst, append), } } fn replace_with_captures<F>( &self, haystack: &[u8], caps: &mut RegexCaptures, dst: &mut Vec<u8>, append: F, ) -> Result<(), NoError> where F: FnMut(&Self::Captures, &mut Vec<u8>) -> bool { use self::RegexMatcherImpl::*; match self.matcher { Standard(ref m) => { m.replace_with_captures(haystack, caps, dst, append) } MultiLiteral(ref m) => { m.replace_with_captures(haystack, caps, dst, append) } CRLF(ref m) => { m.replace_with_captures(haystack, caps, dst, append) } Word(ref m) => { m.replace_with_captures(haystack, caps, dst, append) } } } fn is_match(&self, haystack: &[u8]) -> Result<bool, NoError> { use self::RegexMatcherImpl::*; match self.matcher { Standard(ref m) => m.is_match(haystack), MultiLiteral(ref m) => m.is_match(haystack), CRLF(ref m) => m.is_match(haystack), Word(ref m) => m.is_match(haystack), } } fn is_match_at( &self, haystack: &[u8], at: usize, ) -> Result<bool, NoError> { use self::RegexMatcherImpl::*; match self.matcher { Standard(ref m) => m.is_match_at(haystack, at), MultiLiteral(ref m) => m.is_match_at(haystack, at), CRLF(ref m) => m.is_match_at(haystack, at), Word(ref m) => m.is_match_at(haystack, at), } } fn shortest_match( &self, haystack: &[u8], ) -> Result<Option<usize>, NoError> { use self::RegexMatcherImpl::*; match self.matcher { Standard(ref m) => m.shortest_match(haystack), MultiLiteral(ref m) => m.shortest_match(haystack), CRLF(ref m) => m.shortest_match(haystack), Word(ref m) => m.shortest_match(haystack), } } fn shortest_match_at( &self, haystack: &[u8], at: usize, ) -> Result<Option<usize>, NoError> { use self::RegexMatcherImpl::*; match self.matcher { Standard(ref m) => m.shortest_match_at(haystack, at), MultiLiteral(ref m) => m.shortest_match_at(haystack, at), CRLF(ref m) => m.shortest_match_at(haystack, at), Word(ref m) => m.shortest_match_at(haystack, at), } } fn non_matching_bytes(&self) -> Option<&ByteSet> { Some(&self.non_matching_bytes) } fn line_terminator(&self) -> Option<LineTerminator> { self.config.line_terminator } fn find_candidate_line( &self, haystack: &[u8], ) -> Result<Option<LineMatchKind>, NoError> { Ok(match self.fast_line_regex { Some(ref regex) => { regex.shortest_match(haystack).map(LineMatchKind::Candidate) } None => { self.shortest_match(haystack)?.map(LineMatchKind::Confirmed) } }) } } /// The implementation of the standard regex matcher. #[derive(Clone, Debug)] struct StandardMatcher { /// The regular expression compiled from the pattern provided by the /// caller. regex: Regex, /// A map from capture group name to its corresponding index. names: HashMap<String, usize>, } impl StandardMatcher { fn new(expr: &ConfiguredHIR) -> Result<StandardMatcher, Error> { let regex = expr.regex()?; let mut names = HashMap::new(); for (i, optional_name) in regex.capture_names().enumerate() { if let Some(name) = optional_name { names.insert(name.to_string(), i); } } Ok(StandardMatcher { regex, names }) } } impl Matcher for StandardMatcher { type Captures = RegexCaptures; type Error = NoError; fn find_at( &self, haystack: &[u8], at: usize, ) -> Result<Option<Match>, NoError> { Ok(self.regex .find_at(haystack, at) .map(|m| Match::new(m.start(), m.end()))) } fn new_captures(&self) -> Result<RegexCaptures, NoError> { Ok(RegexCaptures::new(self.regex.capture_locations())) } fn capture_count(&self) -> usize { self.regex.captures_len() } fn capture_index(&self, name: &str) -> Option<usize> { self.names.get(name).map(|i| *i) } fn try_find_iter<F, E>( &self, haystack: &[u8], mut matched: F, ) -> Result<Result<(), E>, NoError> where F: FnMut(Match) -> Result<bool, E> { for m in self.regex.find_iter(haystack) { match matched(Match::new(m.start(), m.end())) { Ok(true) => continue, Ok(false) => return Ok(Ok(())), Err(err) => return Ok(Err(err)), } } Ok(Ok(())) } fn captures_at( &self, haystack: &[u8], at: usize, caps: &mut RegexCaptures, ) -> Result<bool, NoError> { Ok(self.regex.captures_read_at( &mut caps.locations_mut(), haystack, at, ).is_some()) } fn shortest_match_at( &self, haystack: &[u8], at: usize, ) -> Result<Option<usize>, NoError> { Ok(self.regex.shortest_match_at(haystack, at)) } } /// Represents the match offsets of each capturing group in a match. /// /// The first, or `0`th capture group, always corresponds to the entire match /// and is guaranteed to be present when a match occurs. The next capture /// group, at index `1`, corresponds to the first capturing group in the regex, /// ordered by the position at which the left opening parenthesis occurs. /// /// Note that not all capturing groups are guaranteed to be present in a match. /// For example, in the regex, `(?P<foo>\w)|(?P<bar>\W)`, only one of `foo` /// or `bar` will ever be set in any given match. /// /// In order to access a capture group by name, you'll need to first find the /// index of the group using the corresponding matcher's `capture_index` /// method, and then use that index with `RegexCaptures::get`. #[derive(Clone, Debug)] pub struct RegexCaptures(RegexCapturesImp); #[derive(Clone, Debug)] enum RegexCapturesImp { AhoCorasick { /// The start and end of the match, corresponding to capture group 0. mat: Option<Match>, }, Regex { /// Where the locations are stored. locs: CaptureLocations, /// These captures behave as if the capturing groups begin at the given /// offset. When set to `0`, this has no affect and capture groups are /// indexed like normal. /// /// This is useful when building matchers that wrap arbitrary regular /// expressions. For example, `WordMatcher` takes an existing regex /// `re` and creates `(?:^|\W)(re)(?:$|\W)`, but hides the fact that /// the regex has been wrapped from the caller. In order to do this, /// the matcher and the capturing groups must behave as if `(re)` is /// the `0`th capture group. offset: usize, /// When enable, the end of a match has `\r` stripped from it, if one /// exists. strip_crlf: bool, }, } impl Captures for RegexCaptures { fn len(&self) -> usize { match self.0 { RegexCapturesImp::AhoCorasick { .. } => 1, RegexCapturesImp::Regex { ref locs, offset, .. } => { locs.len().checked_sub(offset).unwrap() } } } fn get(&self, i: usize) -> Option<Match> { match self.0 { RegexCapturesImp::AhoCorasick { mat, .. } => { if i == 0 { mat } else { None } } RegexCapturesImp::Regex { ref locs, offset, strip_crlf } => { if !strip_crlf { let actual = i.checked_add(offset).unwrap(); return locs.pos(actual).map(|(s, e)| Match::new(s, e)); } // currently don't support capture offsetting with CRLF // stripping assert_eq!(offset, 0); let m = match locs.pos(i).map(|(s, e)| Match::new(s, e)) { None => return None, Some(m) => m, }; // If the end position of this match corresponds to the end // position of the overall match, then we apply our CRLF // stripping. Otherwise, we cannot assume stripping is correct. if i == 0 || m.end() == locs.pos(0).unwrap().1 { Some(m.with_end(m.end() - 1)) } else { Some(m) } } } } } impl RegexCaptures { pub(crate) fn simple() -> RegexCaptures { RegexCaptures(RegexCapturesImp::AhoCorasick { mat: None }) } pub(crate) fn new(locs: CaptureLocations) -> RegexCaptures { RegexCaptures::with_offset(locs, 0) } pub(crate) fn with_offset( locs: CaptureLocations, offset: usize, ) -> RegexCaptures { RegexCaptures(RegexCapturesImp::Regex { locs, offset, strip_crlf: false, }) } pub(crate) fn locations(&self) -> &CaptureLocations { match self.0 { RegexCapturesImp::AhoCorasick { .. } => { panic!("getting locations for simple captures is invalid") } RegexCapturesImp::Regex { ref locs, .. } => { locs } } } pub(crate) fn locations_mut(&mut self) -> &mut CaptureLocations { match self.0 { RegexCapturesImp::AhoCorasick { .. } => { panic!("getting locations for simple captures is invalid") } RegexCapturesImp::Regex { ref mut locs, .. } => { locs } } } pub(crate) fn strip_crlf(&mut self, yes: bool) { match self.0 { RegexCapturesImp::AhoCorasick { .. } => { panic!("setting strip_crlf for simple captures is invalid") } RegexCapturesImp::Regex { ref mut strip_crlf, .. } => { *strip_crlf = yes; } } } pub(crate) fn set_simple(&mut self, one: Option<Match>) { match self.0 { RegexCapturesImp::AhoCorasick { ref mut mat } => { *mat = one; } RegexCapturesImp::Regex { .. } => { panic!("setting simple captures for regex is invalid") } } } } #[cfg(test)] mod tests { use grep_matcher::{LineMatchKind, Matcher}; use super::*; // Test that enabling word matches does the right thing and demonstrate // the difference between it and surrounding the regex in `\b`. #[test] fn word() { let matcher = RegexMatcherBuilder::new() .word(true) .build(r"-2") .unwrap(); assert!(matcher.is_match(b"abc -2 foo").unwrap()); let matcher = RegexMatcherBuilder::new() .word(false) .build(r"\b-2\b") .unwrap(); assert!(!matcher.is_match(b"abc -2 foo").unwrap()); } // Test that enabling a line terminator prevents it from matching through // said line terminator. #[test] fn line_terminator() { // This works, because there's no line terminator specified. let matcher = RegexMatcherBuilder::new() .build(r"abc\sxyz") .unwrap(); assert!(matcher.is_match(b"abc\nxyz").unwrap()); // This doesn't. let matcher = RegexMatcherBuilder::new() .line_terminator(Some(b'\n')) .build(r"abc\sxyz") .unwrap(); assert!(!matcher.is_match(b"abc\nxyz").unwrap()); } // Ensure that the builder returns an error if a line terminator is set // and the regex could not be modified to remove a line terminator. #[test] fn line_terminator_error() { assert!(RegexMatcherBuilder::new() .line_terminator(Some(b'\n')) .build(r"a\nz") .is_err()) } // Test that enabling CRLF permits `$` to match at the end of a line. #[test] fn line_terminator_crlf() { // Test normal use of `$` with a `\n` line terminator. let matcher = RegexMatcherBuilder::new() .multi_line(true) .build(r"abc$") .unwrap(); assert!(matcher.is_match(b"abc\n").unwrap()); // Test that `$` doesn't match at `\r\n` boundary normally. let matcher = RegexMatcherBuilder::new() .multi_line(true) .build(r"abc$") .unwrap(); assert!(!matcher.is_match(b"abc\r\n").unwrap()); // Now check the CRLF handling. let matcher = RegexMatcherBuilder::new() .multi_line(true) .crlf(true) .build(r"abc$") .unwrap(); assert!(matcher.is_match(b"abc\r\n").unwrap()); } // Test that smart case works. #[test] fn case_smart() { let matcher = RegexMatcherBuilder::new() .case_smart(true) .build(r"abc") .unwrap(); assert!(matcher.is_match(b"ABC").unwrap()); let matcher = RegexMatcherBuilder::new() .case_smart(true) .build(r"aBc") .unwrap(); assert!(!matcher.is_match(b"ABC").unwrap()); } // Test that finding candidate lines works as expected. #[test] fn candidate_lines() { fn is_confirmed(m: LineMatchKind) -> bool { match m { LineMatchKind::Confirmed(_) => true, _ => false, } } fn is_candidate(m: LineMatchKind) -> bool { match m { LineMatchKind::Candidate(_) => true, _ => false, } } // With no line terminator set, we can't employ any optimizations, // so we get a confirmed match. let matcher = RegexMatcherBuilder::new() .build(r"\wfoo\s") .unwrap(); let m = matcher.find_candidate_line(b"afoo ").unwrap().unwrap(); assert!(is_confirmed(m)); // With a line terminator and a regex specially crafted to have an // easy-to-detect inner literal, we can apply an optimization that // quickly finds candidate matches. let matcher = RegexMatcherBuilder::new() .line_terminator(Some(b'\n')) .build(r"\wfoo\s") .unwrap(); let m = matcher.find_candidate_line(b"afoo ").unwrap().unwrap(); assert!(is_candidate(m)); } }