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
/*!
This crate provides an interface for regular expressions, with a focus on line
oriented search. The purpose of this crate is to provide a low level matching
interface that permits any kind of substring or regex implementation to power
the search routines provided by the
[`grep-searcher`](https://docs.rs/grep-searcher)
crate.
The primary thing provided by this crate is the [`Matcher`] trait. The trait
defines an abstract interface for text search. It is robust enough to support
everything from basic substring search all the way to arbitrarily complex
regular expression implementations without sacrificing performance.
A key design decision made in this crate is the use of *internal iteration*,
or otherwise known as the "push" model of searching. In this paradigm,
implementations of the `Matcher` trait will drive search and execute callbacks
provided by the caller when a match is found. This is in contrast to the
usual style of *external iteration* (the "pull" model) found throughout the
Rust ecosystem. There are two primary reasons why internal iteration was
chosen:
* Some search implementations may themselves require internal iteration.
Converting an internal iterator to an external iterator can be non-trivial
and sometimes even practically impossible.
* Rust's type system isn't quite expressive enough to write a generic interface
using external iteration without giving something else up (namely, ease of
use and/or performance).
In other words, internal iteration was chosen because it is the lowest common
denominator and because it is probably the least bad way of expressing the
interface in today's Rust. As a result, this trait isn't specifically intended
for everyday use, although, you might find it to be a happy price to pay if you
want to write code that is generic over multiple different regex
implementations.
*/
#![deny(missing_docs)]
use crate::interpolate::interpolate;
mod interpolate;
/// The type of a match.
///
/// The type of a match is a possibly empty range pointing to a contiguous
/// block of addressable memory.
///
/// Every `Match` is guaranteed to satisfy the invariant that `start <= end`.
///
/// # Indexing
///
/// This type is structurally identical to `std::ops::Range<usize>`, but
/// is a bit more ergonomic for dealing with match indices. In particular,
/// this type implements `Copy` and provides methods for building new `Match`
/// values based on old `Match` values. Finally, the invariant that `start`
/// is always less than or equal to `end` is enforced.
///
/// A `Match` can be used to slice a `&[u8]`, `&mut [u8]` or `&str` using
/// range notation. e.g.,
///
/// ```
/// use grep_matcher::Match;
///
/// let m = Match::new(2, 5);
/// let bytes = b"abcdefghi";
/// assert_eq!(b"cde", &bytes[m]);
/// ```
#[derive(Clone, Copy, Debug, Eq, Hash, PartialEq)]
pub struct Match {
start: usize,
end: usize,
}
impl Match {
/// Create a new match.
///
/// # Panics
///
/// This function panics if `start > end`.
#[inline]
pub fn new(start: usize, end: usize) -> Match {
assert!(start <= end);
Match { start, end }
}
/// Creates a zero width match at the given offset.
#[inline]
pub fn zero(offset: usize) -> Match {
Match { start: offset, end: offset }
}
/// Return the start offset of this match.
#[inline]
pub fn start(&self) -> usize {
self.start
}
/// Return the end offset of this match.
#[inline]
pub fn end(&self) -> usize {
self.end
}
/// Return a new match with the start offset replaced with the given
/// value.
///
/// # Panics
///
/// This method panics if `start > self.end`.
#[inline]
pub fn with_start(&self, start: usize) -> Match {
assert!(start <= self.end, "{} is not <= {}", start, self.end);
Match { start, ..*self }
}
/// Return a new match with the end offset replaced with the given
/// value.
///
/// # Panics
///
/// This method panics if `self.start > end`.
#[inline]
pub fn with_end(&self, end: usize) -> Match {
assert!(self.start <= end, "{} is not <= {}", self.start, end);
Match { end, ..*self }
}
/// Offset this match by the given amount and return a new match.
///
/// This adds the given offset to the start and end of this match, and
/// returns the resulting match.
///
/// # Panics
///
/// This panics if adding the given amount to either the start or end
/// offset would result in an overflow.
#[inline]
pub fn offset(&self, amount: usize) -> Match {
Match {
start: self.start.checked_add(amount).unwrap(),
end: self.end.checked_add(amount).unwrap(),
}
}
/// Returns the number of bytes in this match.
#[inline]
pub fn len(&self) -> usize {
self.end - self.start
}
/// Returns true if and only if this match is empty.
#[inline]
pub fn is_empty(&self) -> bool {
self.len() == 0
}
}
impl std::ops::Index<Match> for [u8] {
type Output = [u8];
#[inline]
fn index(&self, index: Match) -> &[u8] {
&self[index.start..index.end]
}
}
impl std::ops::IndexMut<Match> for [u8] {
#[inline]
fn index_mut(&mut self, index: Match) -> &mut [u8] {
&mut self[index.start..index.end]
}
}
impl std::ops::Index<Match> for str {
type Output = str;
#[inline]
fn index(&self, index: Match) -> &str {
&self[index.start..index.end]
}
}
/// A line terminator.
///
/// A line terminator represents the end of a line. Generally, every line is
/// either "terminated" by the end of a stream or a specific byte (or sequence
/// of bytes).
///
/// Generally, a line terminator is a single byte, specifically, `\n`, on
/// Unix-like systems. On Windows, a line terminator is `\r\n` (referred to
/// as `CRLF` for `Carriage Return; Line Feed`).
///
/// The default line terminator is `\n` on all platforms.
#[derive(Clone, Copy, Debug, Eq, Hash, PartialEq)]
pub struct LineTerminator(LineTerminatorImp);
#[derive(Clone, Copy, Debug, Eq, Hash, PartialEq)]
enum LineTerminatorImp {
/// Any single byte representing a line terminator.
Byte(u8),
/// A line terminator represented by `\r\n`.
///
/// When this option is used, consumers may generally treat a lone `\n` as
/// a line terminator in addition to `\r\n`.
CRLF,
}
impl LineTerminator {
/// Return a new single-byte line terminator. Any byte is valid.
#[inline]
pub fn byte(byte: u8) -> LineTerminator {
LineTerminator(LineTerminatorImp::Byte(byte))
}
/// Return a new line terminator represented by `\r\n`.
///
/// When this option is used, consumers may generally treat a lone `\n` as
/// a line terminator in addition to `\r\n`.
#[inline]
pub fn crlf() -> LineTerminator {
LineTerminator(LineTerminatorImp::CRLF)
}
/// Returns true if and only if this line terminator is CRLF.
#[inline]
pub fn is_crlf(&self) -> bool {
self.0 == LineTerminatorImp::CRLF
}
/// Returns this line terminator as a single byte.
///
/// If the line terminator is CRLF, then this returns `\n`. This is
/// useful for routines that, for example, find line boundaries by treating
/// `\n` as a line terminator even when it isn't preceded by `\r`.
#[inline]
pub fn as_byte(&self) -> u8 {
match self.0 {
LineTerminatorImp::Byte(byte) => byte,
LineTerminatorImp::CRLF => b'\n',
}
}
/// Returns this line terminator as a sequence of bytes.
///
/// This returns a singleton sequence for all line terminators except for
/// `CRLF`, in which case, it returns `\r\n`.
///
/// The slice returned is guaranteed to have length at least `1`.
#[inline]
pub fn as_bytes(&self) -> &[u8] {
match self.0 {
LineTerminatorImp::Byte(ref byte) => std::slice::from_ref(byte),
LineTerminatorImp::CRLF => &[b'\r', b'\n'],
}
}
/// Returns true if and only if the given slice ends with this line
/// terminator.
///
/// If this line terminator is `CRLF`, then this only checks whether the
/// last byte is `\n`.
#[inline]
pub fn is_suffix(&self, slice: &[u8]) -> bool {
slice.last().map_or(false, |&b| b == self.as_byte())
}
}
impl Default for LineTerminator {
#[inline]
fn default() -> LineTerminator {
LineTerminator::byte(b'\n')
}
}
/// A set of bytes.
///
/// In this crate, byte sets are used to express bytes that can never appear
/// anywhere in a match for a particular implementation of the `Matcher` trait.
/// Specifically, if such a set can be determined, then it's possible for
/// callers to perform additional operations on the basis that certain bytes
/// may never match.
///
/// For example, if a search is configured to possibly produce results that
/// span multiple lines but a caller provided pattern can never match across
/// multiple lines, then it may make sense to divert to more optimized line
/// oriented routines that don't need to handle the multi-line match case.
#[derive(Clone, Debug)]
pub struct ByteSet(BitSet);
#[derive(Clone, Copy)]
struct BitSet([u64; 4]);
impl std::fmt::Debug for BitSet {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
let mut fmtd = f.debug_set();
for b in 0..=255 {
if ByteSet(*self).contains(b) {
fmtd.entry(&b);
}
}
fmtd.finish()
}
}
impl ByteSet {
/// Create an empty set of bytes.
#[inline]
pub fn empty() -> ByteSet {
ByteSet(BitSet([0; 4]))
}
/// Create a full set of bytes such that every possible byte is in the set
/// returned.
#[inline]
pub fn full() -> ByteSet {
ByteSet(BitSet([u64::MAX; 4]))
}
/// Add a byte to this set.
///
/// If the given byte already belongs to this set, then this is a no-op.
#[inline]
pub fn add(&mut self, byte: u8) {
let bucket = byte / 64;
let bit = byte % 64;
(self.0).0[usize::from(bucket)] |= 1 << bit;
}
/// Add an inclusive range of bytes.
#[inline]
pub fn add_all(&mut self, start: u8, end: u8) {
for b in start..=end {
self.add(b);
}
}
/// Remove a byte from this set.
///
/// If the given byte is not in this set, then this is a no-op.
#[inline]
pub fn remove(&mut self, byte: u8) {
let bucket = byte / 64;
let bit = byte % 64;
(self.0).0[usize::from(bucket)] &= !(1 << bit);
}
/// Remove an inclusive range of bytes.
#[inline]
pub fn remove_all(&mut self, start: u8, end: u8) {
for b in start..=end {
self.remove(b);
}
}
/// Return true if and only if the given byte is in this set.
#[inline]
pub fn contains(&self, byte: u8) -> bool {
let bucket = byte / 64;
let bit = byte % 64;
(self.0).0[usize::from(bucket)] & (1 << bit) > 0
}
}
/// A trait that describes implementations of capturing groups.
///
/// When a matcher supports capturing group extraction, then it is the
/// matcher's responsibility to provide an implementation of this trait.
///
/// Principally, this trait provides a way to access capturing groups
/// in a uniform way that does not require any specific representation.
/// Namely, different matcher implementations may require different in-memory
/// representations of capturing groups. This trait permits matchers to
/// maintain their specific in-memory representation.
///
/// Note that this trait explicitly does not provide a way to construct a new
/// capture value. Instead, it is the responsibility of a `Matcher` to build
/// one, which might require knowledge of the matcher's internal implementation
/// details.
pub trait Captures {
/// Return the total number of capturing groups. This includes capturing
/// groups that have not matched anything.
fn len(&self) -> usize;
/// Return the capturing group match at the given index. If no match of
/// that capturing group exists, then this returns `None`.
///
/// When a matcher reports a match with capturing groups, then the first
/// capturing group (at index `0`) must always correspond to the offsets
/// for the overall match.
fn get(&self, i: usize) -> Option<Match>;
/// Returns true if and only if these captures are empty. This occurs
/// when `len` is `0`.
///
/// Note that capturing groups that have non-zero length but otherwise
/// contain no matching groups are *not* empty.
#[inline]
fn is_empty(&self) -> bool {
self.len() == 0
}
/// Expands all instances of `$name` in `replacement` to the corresponding
/// capture group `name`, and writes them to the `dst` buffer given.
///
/// (Note: If you're looking for a convenient way to perform replacements
/// with interpolation, then you'll want to use the `replace_with_captures`
/// method on the `Matcher` trait.)
///
/// `name` may be an integer corresponding to the index of the
/// capture group (counted by order of opening parenthesis where `0` is the
/// entire match) or it can be a name (consisting of letters, digits or
/// underscores) corresponding to a named capture group.
///
/// A `name` is translated to a capture group index via the given
/// `name_to_index` function. If `name` isn't a valid capture group
/// (whether the name doesn't exist or isn't a valid index), then it is
/// replaced with the empty string.
///
/// The longest possible name is used. e.g., `$1a` looks up the capture
/// group named `1a` and not the capture group at index `1`. To exert
/// more precise control over the name, use braces, e.g., `${1}a`. In all
/// cases, capture group names are limited to ASCII letters, numbers and
/// underscores.
///
/// To write a literal `$` use `$$`.
///
/// Note that the capture group match indices are resolved by slicing
/// the given `haystack`. Generally, this means that `haystack` should be
/// the same slice that was searched to get the current capture group
/// matches.
#[inline]
fn interpolate<F>(
&self,
name_to_index: F,
haystack: &[u8],
replacement: &[u8],
dst: &mut Vec<u8>,
) where
F: FnMut(&str) -> Option<usize>,
{
interpolate(
replacement,
|i, dst| {
if let Some(range) = self.get(i) {
dst.extend(&haystack[range]);
}
},
name_to_index,
dst,
)
}
}
/// NoCaptures provides an always-empty implementation of the `Captures` trait.
///
/// This type is useful for implementations of `Matcher` that don't support
/// capturing groups.
#[derive(Clone, Debug)]
pub struct NoCaptures(());
impl NoCaptures {
/// Create an empty set of capturing groups.
#[inline]
pub fn new() -> NoCaptures {
NoCaptures(())
}
}
impl Captures for NoCaptures {
#[inline]
fn len(&self) -> usize {
0
}
#[inline]
fn get(&self, _: usize) -> Option<Match> {
None
}
}
/// NoError provides an error type for matchers that never produce errors.
///
/// This error type implements the `std::error::Error` and `std::fmt::Display`
/// traits for use in matcher implementations that can never produce errors.
///
/// The `std::fmt::Debug` and `std::fmt::Display` impls for this type panics.
#[derive(Debug, Eq, PartialEq)]
pub struct NoError(());
impl std::error::Error for NoError {
fn description(&self) -> &str {
"no error"
}
}
impl std::fmt::Display for NoError {
fn fmt(&self, _: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
panic!("BUG for NoError: an impossible error occurred")
}
}
impl From<NoError> for std::io::Error {
fn from(_: NoError) -> std::io::Error {
panic!("BUG for NoError: an impossible error occurred")
}
}
/// The type of match for a line oriented matcher.
#[derive(Clone, Copy, Debug)]
pub enum LineMatchKind {
/// A position inside a line that is known to contain a match.
///
/// This position can be anywhere in the line. It does not need to point
/// at the location of the match.
Confirmed(usize),
/// A position inside a line that may contain a match, and must be searched
/// for verification.
///
/// This position can be anywhere in the line. It does not need to point
/// at the location of the match.
Candidate(usize),
}
/// A matcher defines an interface for regular expression implementations.
///
/// While this trait is large, there are only two required methods that
/// implementors must provide: `find_at` and `new_captures`. If captures aren't
/// supported by your implementation, then `new_captures` can be implemented
/// with [`NoCaptures`]. If your implementation does support capture groups,
/// then you should also implement the other capture related methods, as
/// dictated by the documentation. Crucially, this includes `captures_at`.
///
/// The rest of the methods on this trait provide default implementations on
/// top of `find_at` and `new_captures`. It is not uncommon for implementations
/// to be able to provide faster variants of some methods; in those cases,
/// simply override the default implementation.
pub trait Matcher {
/// The concrete type of capturing groups used for this matcher.
///
/// If this implementation does not support capturing groups, then set
/// this to `NoCaptures`.
type Captures: Captures;
/// The error type used by this matcher.
///
/// For matchers in which an error is not possible, they are encouraged to
/// use the `NoError` type in this crate. In the future, when the "never"
/// (spelled `!`) type is stabilized, then it should probably be used
/// instead.
type Error: std::fmt::Display;
/// Returns the start and end byte range of the first match in `haystack`
/// after `at`, where the byte offsets are relative to that start of
/// `haystack` (and not `at`). If no match exists, then `None` is returned.
///
/// The text encoding of `haystack` is not strictly specified. Matchers are
/// advised to assume UTF-8, or at worst, some ASCII compatible encoding.
///
/// The significance of the starting point is that it takes the surrounding
/// context into consideration. For example, the `\A` anchor can only
/// match when `at == 0`.
fn find_at(
&self,
haystack: &[u8],
at: usize,
) -> Result<Option<Match>, Self::Error>;
/// Creates an empty group of captures suitable for use with the capturing
/// APIs of this trait.
///
/// Implementations that don't support capturing groups should use
/// the `NoCaptures` type and implement this method by calling
/// `NoCaptures::new()`.
fn new_captures(&self) -> Result<Self::Captures, Self::Error>;
/// Returns the total number of capturing groups in this matcher.
///
/// If a matcher supports capturing groups, then this value must always be
/// at least 1, where the first capturing group always corresponds to the
/// overall match.
///
/// If a matcher does not support capturing groups, then this should
/// always return 0.
///
/// By default, capturing groups are not supported, so this always
/// returns 0.
#[inline]
fn capture_count(&self) -> usize {
0
}
/// Maps the given capture group name to its corresponding capture group
/// index, if one exists. If one does not exist, then `None` is returned.
///
/// If the given capture group name maps to multiple indices, then it is
/// not specified which one is returned. However, it is guaranteed that
/// one of them is returned.
///
/// By default, capturing groups are not supported, so this always returns
/// `None`.
#[inline]
fn capture_index(&self, _name: &str) -> Option<usize> {
None
}
/// Returns the start and end byte range of the first match in `haystack`.
/// If no match exists, then `None` is returned.
///
/// The text encoding of `haystack` is not strictly specified. Matchers are
/// advised to assume UTF-8, or at worst, some ASCII compatible encoding.
#[inline]
fn find(&self, haystack: &[u8]) -> Result<Option<Match>, Self::Error> {
self.find_at(haystack, 0)
}
/// Executes the given function over successive non-overlapping matches
/// in `haystack`. If no match exists, then the given function is never
/// called. If the function returns `false`, then iteration stops.
#[inline]
fn find_iter<F>(
&self,
haystack: &[u8],
matched: F,
) -> Result<(), Self::Error>
where
F: FnMut(Match) -> bool,
{
self.find_iter_at(haystack, 0, matched)
}
/// Executes the given function over successive non-overlapping matches
/// in `haystack`. If no match exists, then the given function is never
/// called. If the function returns `false`, then iteration stops.
///
/// The significance of the starting point is that it takes the surrounding
/// context into consideration. For example, the `\A` anchor can only
/// match when `at == 0`.
#[inline]
fn find_iter_at<F>(
&self,
haystack: &[u8],
at: usize,
mut matched: F,
) -> Result<(), Self::Error>
where
F: FnMut(Match) -> bool,
{
self.try_find_iter_at(haystack, at, |m| Ok(matched(m)))
.map(|r: Result<(), ()>| r.unwrap())
}
/// Executes the given function over successive non-overlapping matches
/// in `haystack`. If no match exists, then the given function is never
/// called. If the function returns `false`, then iteration stops.
/// Similarly, if the function returns an error then iteration stops and
/// the error is yielded. If an error occurs while executing the search,
/// then it is converted to
/// `E`.
#[inline]
fn try_find_iter<F, E>(
&self,
haystack: &[u8],
matched: F,
) -> Result<Result<(), E>, Self::Error>
where
F: FnMut(Match) -> Result<bool, E>,
{
self.try_find_iter_at(haystack, 0, matched)
}
/// Executes the given function over successive non-overlapping matches
/// in `haystack`. If no match exists, then the given function is never
/// called. If the function returns `false`, then iteration stops.
/// Similarly, if the function returns an error then iteration stops and
/// the error is yielded. If an error occurs while executing the search,
/// then it is converted to
/// `E`.
///
/// The significance of the starting point is that it takes the surrounding
/// context into consideration. For example, the `\A` anchor can only
/// match when `at == 0`.
#[inline]
fn try_find_iter_at<F, E>(
&self,
haystack: &[u8],
at: usize,
mut matched: F,
) -> Result<Result<(), E>, Self::Error>
where
F: FnMut(Match) -> Result<bool, E>,
{
let mut last_end = at;
let mut last_match = None;
loop {
if last_end > haystack.len() {
return Ok(Ok(()));
}
let m = match self.find_at(haystack, last_end)? {
None => return Ok(Ok(())),
Some(m) => m,
};
if m.start == m.end {
// This is an empty match. To ensure we make progress, start
// the next search at the smallest possible starting position
// of the next match following this one.
last_end = m.end + 1;
// Don't accept empty matches immediately following a match.
// Just move on to the next match.
if Some(m.end) == last_match {
continue;
}
} else {
last_end = m.end;
}
last_match = Some(m.end);
match matched(m) {
Ok(true) => continue,
Ok(false) => return Ok(Ok(())),
Err(err) => return Ok(Err(err)),
}
}
}
/// Populates the first set of capture group matches from `haystack` into
/// `caps`. If no match exists, then `false` is returned.
///
/// The text encoding of `haystack` is not strictly specified. Matchers are
/// advised to assume UTF-8, or at worst, some ASCII compatible encoding.
#[inline]
fn captures(
&self,
haystack: &[u8],
caps: &mut Self::Captures,
) -> Result<bool, Self::Error> {
self.captures_at(haystack, 0, caps)
}
/// Executes the given function over successive non-overlapping matches
/// in `haystack` with capture groups extracted from each match. If no
/// match exists, then the given function is never called. If the function
/// returns `false`, then iteration stops.
#[inline]
fn captures_iter<F>(
&self,
haystack: &[u8],
caps: &mut Self::Captures,
matched: F,
) -> Result<(), Self::Error>
where
F: FnMut(&Self::Captures) -> bool,
{
self.captures_iter_at(haystack, 0, caps, matched)
}
/// Executes the given function over successive non-overlapping matches
/// in `haystack` with capture groups extracted from each match. If no
/// match exists, then the given function is never called. If the function
/// returns `false`, then iteration stops.
///
/// The significance of the starting point is that it takes the surrounding
/// context into consideration. For example, the `\A` anchor can only
/// match when `at == 0`.
#[inline]
fn captures_iter_at<F>(
&self,
haystack: &[u8],
at: usize,
caps: &mut Self::Captures,
mut matched: F,
) -> Result<(), Self::Error>
where
F: FnMut(&Self::Captures) -> bool,
{
self.try_captures_iter_at(haystack, at, caps, |caps| Ok(matched(caps)))
.map(|r: Result<(), ()>| r.unwrap())
}
/// Executes the given function over successive non-overlapping matches
/// in `haystack` with capture groups extracted from each match. If no
/// match exists, then the given function is never called. If the function
/// returns `false`, then iteration stops. Similarly, if the function
/// returns an error then iteration stops and the error is yielded. If
/// an error occurs while executing the search, then it is converted to
/// `E`.
#[inline]
fn try_captures_iter<F, E>(
&self,
haystack: &[u8],
caps: &mut Self::Captures,
matched: F,
) -> Result<Result<(), E>, Self::Error>
where
F: FnMut(&Self::Captures) -> Result<bool, E>,
{
self.try_captures_iter_at(haystack, 0, caps, matched)
}
/// Executes the given function over successive non-overlapping matches
/// in `haystack` with capture groups extracted from each match. If no
/// match exists, then the given function is never called. If the function
/// returns `false`, then iteration stops. Similarly, if the function
/// returns an error then iteration stops and the error is yielded. If
/// an error occurs while executing the search, then it is converted to
/// `E`.
///
/// The significance of the starting point is that it takes the surrounding
/// context into consideration. For example, the `\A` anchor can only
/// match when `at == 0`.
#[inline]
fn try_captures_iter_at<F, E>(
&self,
haystack: &[u8],
at: usize,
caps: &mut Self::Captures,
mut matched: F,
) -> Result<Result<(), E>, Self::Error>
where
F: FnMut(&Self::Captures) -> Result<bool, E>,
{
let mut last_end = at;
let mut last_match = None;
loop {
if last_end > haystack.len() {
return Ok(Ok(()));
}
if !self.captures_at(haystack, last_end, caps)? {
return Ok(Ok(()));
}
let m = caps.get(0).unwrap();
if m.start == m.end {
// This is an empty match. To ensure we make progress, start
// the next search at the smallest possible starting position
// of the next match following this one.
last_end = m.end + 1;
// Don't accept empty matches immediately following a match.
// Just move on to the next match.
if Some(m.end) == last_match {
continue;
}
} else {
last_end = m.end;
}
last_match = Some(m.end);
match matched(caps) {
Ok(true) => continue,
Ok(false) => return Ok(Ok(())),
Err(err) => return Ok(Err(err)),
}
}
}
/// Populates the first set of capture group matches from `haystack`
/// into `matches` after `at`, where the byte offsets in each capturing
/// group are relative to the start of `haystack` (and not `at`). If no
/// match exists, then `false` is returned and the contents of the given
/// capturing groups are unspecified.
///
/// The text encoding of `haystack` is not strictly specified. Matchers are
/// advised to assume UTF-8, or at worst, some ASCII compatible encoding.
///
/// The significance of the starting point is that it takes the surrounding
/// context into consideration. For example, the `\A` anchor can only
/// match when `at == 0`.
///
/// By default, capturing groups aren't supported, and this implementation
/// will always behave as if a match were impossible.
///
/// Implementors that provide support for capturing groups must guarantee
/// that when a match occurs, the first capture match (at index `0`) is
/// always set to the overall match offsets.
///
/// Note that if implementors seek to support capturing groups, then they
/// should implement this method. Other methods that match based on
/// captures will then work automatically.
#[inline]
fn captures_at(
&self,
_haystack: &[u8],
_at: usize,
_caps: &mut Self::Captures,
) -> Result<bool, Self::Error> {
Ok(false)
}
/// Replaces every match in the given haystack with the result of calling
/// `append`. `append` is given the start and end of a match, along with
/// a handle to the `dst` buffer provided.
///
/// If the given `append` function returns `false`, then replacement stops.
#[inline]
fn replace<F>(
&self,
haystack: &[u8],
dst: &mut Vec<u8>,
mut append: F,
) -> Result<(), Self::Error>
where
F: FnMut(Match, &mut Vec<u8>) -> bool,
{
let mut last_match = 0;
self.find_iter(haystack, |m| {
dst.extend(&haystack[last_match..m.start]);
last_match = m.end;
append(m, dst)
})?;
dst.extend(&haystack[last_match..]);
Ok(())
}
/// Replaces every match in the given haystack with the result of calling
/// `append` with the matching capture groups.
///
/// If the given `append` function returns `false`, then replacement stops.
#[inline]
fn replace_with_captures<F>(
&self,
haystack: &[u8],
caps: &mut Self::Captures,
dst: &mut Vec<u8>,
append: F,
) -> Result<(), Self::Error>
where
F: FnMut(&Self::Captures, &mut Vec<u8>) -> bool,
{
self.replace_with_captures_at(haystack, 0, caps, dst, append)
}
/// Replaces every match in the given haystack with the result of calling
/// `append` with the matching capture groups.
///
/// If the given `append` function returns `false`, then replacement stops.
///
/// The significance of the starting point is that it takes the surrounding
/// context into consideration. For example, the `\A` anchor can only
/// match when `at == 0`.
#[inline]
fn replace_with_captures_at<F>(
&self,
haystack: &[u8],
at: usize,
caps: &mut Self::Captures,
dst: &mut Vec<u8>,
mut append: F,
) -> Result<(), Self::Error>
where
F: FnMut(&Self::Captures, &mut Vec<u8>) -> bool,
{
let mut last_match = at;
self.captures_iter_at(haystack, at, caps, |caps| {
let m = caps.get(0).unwrap();
dst.extend(&haystack[last_match..m.start]);
last_match = m.end;
append(caps, dst)
})?;
dst.extend(&haystack[last_match..]);
Ok(())
}
/// Returns true if and only if the matcher matches the given haystack.
///
/// By default, this method is implemented by calling `shortest_match`.
#[inline]
fn is_match(&self, haystack: &[u8]) -> Result<bool, Self::Error> {
self.is_match_at(haystack, 0)
}
/// Returns true if and only if the matcher matches the given haystack
/// starting at the given position.
///
/// By default, this method is implemented by calling `shortest_match_at`.
///
/// The significance of the starting point is that it takes the surrounding
/// context into consideration. For example, the `\A` anchor can only
/// match when `at == 0`.
#[inline]
fn is_match_at(
&self,
haystack: &[u8],
at: usize,
) -> Result<bool, Self::Error> {
Ok(self.shortest_match_at(haystack, at)?.is_some())
}
/// Returns an end location of the first match in `haystack`. If no match
/// exists, then `None` is returned.
///
/// Note that the end location reported by this method may be less than the
/// same end location reported by `find`. For example, running `find` with
/// the pattern `a+` on the haystack `aaa` should report a range of `[0,
/// 3)`, but `shortest_match` may report `1` as the ending location since
/// that is the place at which a match is guaranteed to occur.
///
/// This method should never report false positives or false negatives. The
/// point of this method is that some implementors may be able to provide
/// a faster implementation of this than what `find` does.
///
/// By default, this method is implemented by calling `find`.
#[inline]
fn shortest_match(
&self,
haystack: &[u8],
) -> Result<Option<usize>, Self::Error> {
self.shortest_match_at(haystack, 0)
}
/// Returns an end location of the first match in `haystack` starting at
/// the given position. If no match exists, then `None` is returned.
///
/// Note that the end location reported by this method may be less than the
/// same end location reported by `find`. For example, running `find` with
/// the pattern `a+` on the haystack `aaa` should report a range of `[0,
/// 3)`, but `shortest_match` may report `1` as the ending location since
/// that is the place at which a match is guaranteed to occur.
///
/// This method should never report false positives or false negatives. The
/// point of this method is that some implementors may be able to provide
/// a faster implementation of this than what `find` does.
///
/// By default, this method is implemented by calling `find_at`.
///
/// The significance of the starting point is that it takes the surrounding
/// context into consideration. For example, the `\A` anchor can only
/// match when `at == 0`.
#[inline]
fn shortest_match_at(
&self,
haystack: &[u8],
at: usize,
) -> Result<Option<usize>, Self::Error> {
Ok(self.find_at(haystack, at)?.map(|m| m.end))
}
/// If available, return a set of bytes that will never appear in a match
/// produced by an implementation.
///
/// Specifically, if such a set can be determined, then it's possible for
/// callers to perform additional operations on the basis that certain
/// bytes may never match.
///
/// For example, if a search is configured to possibly produce results
/// that span multiple lines but a caller provided pattern can never
/// match across multiple lines, then it may make sense to divert to
/// more optimized line oriented routines that don't need to handle the
/// multi-line match case.
///
/// Implementations that produce this set must never report false
/// positives, but may produce false negatives. That is, is a byte is in
/// this set then it must be guaranteed that it is never in a match. But,
/// if a byte is not in this set, then callers cannot assume that a match
/// exists with that byte.
///
/// By default, this returns `None`.
#[inline]
fn non_matching_bytes(&self) -> Option<&ByteSet> {
None
}
/// If this matcher was compiled as a line oriented matcher, then this
/// method returns the line terminator if and only if the line terminator
/// never appears in any match produced by this matcher. If this wasn't
/// compiled as a line oriented matcher, or if the aforementioned guarantee
/// cannot be made, then this must return `None`, which is the default.
/// It is **never wrong** to return `None`, but returning a line terminator
/// when it can appear in a match results in unspecified behavior.
///
/// The line terminator is typically `b'\n'`, but can be any single byte or
/// `CRLF`.
///
/// By default, this returns `None`.
#[inline]
fn line_terminator(&self) -> Option<LineTerminator> {
None
}
/// Return one of the following: a confirmed line match, a candidate line
/// match (which may be a false positive) or no match at all (which **must
/// not** be a false negative). When reporting a confirmed or candidate
/// match, the position returned can be any position in the line.
///
/// By default, this never returns a candidate match, and always either
/// returns a confirmed match or no match at all.
///
/// When a matcher can match spans over multiple lines, then the behavior
/// of this method is unspecified. Namely, use of this method only
/// makes sense in a context where the caller is looking for the next
/// matching line. That is, callers should only use this method when
/// `line_terminator` does not return `None`.
///
/// # Design rationale
///
/// A line matcher is, fundamentally, a normal matcher with the addition
/// of one optional method: finding a line. By default, this routine
/// is implemented via the matcher's `shortest_match` method, which
/// always yields either no match or a `LineMatchKind::Confirmed`. However,
/// implementors may provide a routine for this that can return candidate
/// lines that need subsequent verification to be confirmed as a match.
/// This can be useful in cases where it may be quicker to find candidate
/// lines via some other means instead of relying on the more general
/// implementations for `find` and `shortest_match`.
///
/// For example, consider the regex `\w+foo\s+`. Both `find` and
/// `shortest_match` must consider the entire regex, including the `\w+`
/// and `\s+`, while searching. However, this method could look for lines
/// containing `foo` and return them as candidates. Finding `foo` might
/// be implemented as a highly optimized substring search routine (like
/// `memmem`), which is likely to be faster than whatever more generalized
/// routine is required for resolving `\w+foo\s+`. The caller is then
/// responsible for confirming whether a match exists or not.
///
/// Note that while this method may report false positives, it must never
/// report false negatives. That is, it can never skip over lines that
/// contain a match.
#[inline]
fn find_candidate_line(
&self,
haystack: &[u8],
) -> Result<Option<LineMatchKind>, Self::Error> {
Ok(self.shortest_match(haystack)?.map(LineMatchKind::Confirmed))
}
}
impl<'a, M: Matcher> Matcher for &'a M {
type Captures = M::Captures;
type Error = M::Error;
#[inline]
fn find_at(
&self,
haystack: &[u8],
at: usize,
) -> Result<Option<Match>, Self::Error> {
(*self).find_at(haystack, at)
}
#[inline]
fn new_captures(&self) -> Result<Self::Captures, Self::Error> {
(*self).new_captures()
}
#[inline]
fn captures_at(
&self,
haystack: &[u8],
at: usize,
caps: &mut Self::Captures,
) -> Result<bool, Self::Error> {
(*self).captures_at(haystack, at, caps)
}
#[inline]
fn capture_index(&self, name: &str) -> Option<usize> {
(*self).capture_index(name)
}
#[inline]
fn capture_count(&self) -> usize {
(*self).capture_count()
}
#[inline]
fn find(&self, haystack: &[u8]) -> Result<Option<Match>, Self::Error> {
(*self).find(haystack)
}
#[inline]
fn find_iter<F>(
&self,
haystack: &[u8],
matched: F,
) -> Result<(), Self::Error>
where
F: FnMut(Match) -> bool,
{
(*self).find_iter(haystack, matched)
}
#[inline]
fn find_iter_at<F>(
&self,
haystack: &[u8],
at: usize,
matched: F,
) -> Result<(), Self::Error>
where
F: FnMut(Match) -> bool,
{
(*self).find_iter_at(haystack, at, matched)
}
#[inline]
fn try_find_iter<F, E>(
&self,
haystack: &[u8],
matched: F,
) -> Result<Result<(), E>, Self::Error>
where
F: FnMut(Match) -> Result<bool, E>,
{
(*self).try_find_iter(haystack, matched)
}
#[inline]
fn try_find_iter_at<F, E>(
&self,
haystack: &[u8],
at: usize,
matched: F,
) -> Result<Result<(), E>, Self::Error>
where
F: FnMut(Match) -> Result<bool, E>,
{
(*self).try_find_iter_at(haystack, at, matched)
}
#[inline]
fn captures(
&self,
haystack: &[u8],
caps: &mut Self::Captures,
) -> Result<bool, Self::Error> {
(*self).captures(haystack, caps)
}
#[inline]
fn captures_iter<F>(
&self,
haystack: &[u8],
caps: &mut Self::Captures,
matched: F,
) -> Result<(), Self::Error>
where
F: FnMut(&Self::Captures) -> bool,
{
(*self).captures_iter(haystack, caps, matched)
}
#[inline]
fn captures_iter_at<F>(
&self,
haystack: &[u8],
at: usize,
caps: &mut Self::Captures,
matched: F,
) -> Result<(), Self::Error>
where
F: FnMut(&Self::Captures) -> bool,
{
(*self).captures_iter_at(haystack, at, caps, matched)
}
#[inline]
fn try_captures_iter<F, E>(
&self,
haystack: &[u8],
caps: &mut Self::Captures,
matched: F,
) -> Result<Result<(), E>, Self::Error>
where
F: FnMut(&Self::Captures) -> Result<bool, E>,
{
(*self).try_captures_iter(haystack, caps, matched)
}
#[inline]
fn try_captures_iter_at<F, E>(
&self,
haystack: &[u8],
at: usize,
caps: &mut Self::Captures,
matched: F,
) -> Result<Result<(), E>, Self::Error>
where
F: FnMut(&Self::Captures) -> Result<bool, E>,
{
(*self).try_captures_iter_at(haystack, at, caps, matched)
}
#[inline]
fn replace<F>(
&self,
haystack: &[u8],
dst: &mut Vec<u8>,
append: F,
) -> Result<(), Self::Error>
where
F: FnMut(Match, &mut Vec<u8>) -> bool,
{
(*self).replace(haystack, dst, append)
}
#[inline]
fn replace_with_captures<F>(
&self,
haystack: &[u8],
caps: &mut Self::Captures,
dst: &mut Vec<u8>,
append: F,
) -> Result<(), Self::Error>
where
F: FnMut(&Self::Captures, &mut Vec<u8>) -> bool,
{
(*self).replace_with_captures(haystack, caps, dst, append)
}
#[inline]
fn replace_with_captures_at<F>(
&self,
haystack: &[u8],
at: usize,
caps: &mut Self::Captures,
dst: &mut Vec<u8>,
append: F,
) -> Result<(), Self::Error>
where
F: FnMut(&Self::Captures, &mut Vec<u8>) -> bool,
{
(*self).replace_with_captures_at(haystack, at, caps, dst, append)
}
#[inline]
fn is_match(&self, haystack: &[u8]) -> Result<bool, Self::Error> {
(*self).is_match(haystack)
}
#[inline]
fn is_match_at(
&self,
haystack: &[u8],
at: usize,
) -> Result<bool, Self::Error> {
(*self).is_match_at(haystack, at)
}
#[inline]
fn shortest_match(
&self,
haystack: &[u8],
) -> Result<Option<usize>, Self::Error> {
(*self).shortest_match(haystack)
}
#[inline]
fn shortest_match_at(
&self,
haystack: &[u8],
at: usize,
) -> Result<Option<usize>, Self::Error> {
(*self).shortest_match_at(haystack, at)
}
#[inline]
fn non_matching_bytes(&self) -> Option<&ByteSet> {
(*self).non_matching_bytes()
}
#[inline]
fn line_terminator(&self) -> Option<LineTerminator> {
(*self).line_terminator()
}
#[inline]
fn find_candidate_line(
&self,
haystack: &[u8],
) -> Result<Option<LineMatchKind>, Self::Error> {
(*self).find_candidate_line(haystack)
}
}