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
// Copyright 2014 The Rust Project Developers. See the COPYRIGHT
// file at the top-level directory of this distribution and at
// http://rust-lang.org/COPYRIGHT.
//
// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
// option. This file may not be copied, modified, or distributed
// except according to those terms.
//! Support for matching file paths against Unix shell style patterns.
//!
//! The `glob` and `glob_with` functions allow querying the filesystem for all
//! files that match a particular pattern (similar to the libc `glob` function).
//! The methods on the `Pattern` type provide functionality for checking if
//! individual paths match a particular pattern (similar to the libc `fnmatch`
//! function).
//!
//! For consistency across platforms, and for Windows support, this module
//! is implemented entirely in Rust rather than deferring to the libc
//! `glob`/`fnmatch` functions.
//!
//! # Examples
//!
//! To print all jpg files in `/media/` and all of its subdirectories.
//!
//! ```rust,no_run
//! use glob::glob;
//!
//! for entry in glob("/media/**/*.jpg").expect("Failed to read glob pattern") {
//! match entry {
//! Ok(path) => println!("{:?}", path.display()),
//! Err(e) => println!("{:?}", e),
//! }
//! }
//! ```
//!
//! To print all files containing the letter "a", case insensitive, in a `local`
//! directory relative to the current working directory. This ignores errors
//! instead of printing them.
//!
//! ```rust,no_run
//! use glob::glob_with;
//! use glob::MatchOptions;
//!
//! let options = MatchOptions {
//! case_sensitive: false,
//! require_literal_separator: false,
//! require_literal_leading_dot: false,
//! };
//! for entry in glob_with("local/*a*", options).unwrap() {
//! if let Ok(path) = entry {
//! println!("{:?}", path.display())
//! }
//! }
//! ```
#![doc(
html_logo_url = "https://www.rust-lang.org/logos/rust-logo-128x128-blk-v2.png",
html_favicon_url = "https://www.rust-lang.org/favicon.ico",
html_root_url = "https://docs.rs/glob/0.3.0"
)]
#![deny(missing_docs)]
#![cfg_attr(all(test, windows), feature(std_misc))]
use std::cmp;
use std::error::Error;
use std::fmt;
use std::fs;
use std::io;
use std::path::{self, Component, Path, PathBuf};
use std::str::FromStr;
use CharSpecifier::{CharRange, SingleChar};
use MatchResult::{EntirePatternDoesntMatch, Match, SubPatternDoesntMatch};
use PatternToken::AnyExcept;
use PatternToken::{AnyChar, AnyRecursiveSequence, AnySequence, AnyWithin, Char};
/// An iterator that yields `Path`s from the filesystem that match a particular
/// pattern.
///
/// Note that it yields `GlobResult` in order to report any `IoErrors` that may
/// arise during iteration. If a directory matches but is unreadable,
/// thereby preventing its contents from being checked for matches, a
/// `GlobError` is returned to express this.
///
/// See the `glob` function for more details.
pub struct Paths {
dir_patterns: Vec<Pattern>,
require_dir: bool,
options: MatchOptions,
todo: Vec<Result<(PathBuf, usize), GlobError>>,
scope: Option<PathBuf>,
}
/// Return an iterator that produces all the `Path`s that match the given
/// pattern using default match options, which may be absolute or relative to
/// the current working directory.
///
/// This may return an error if the pattern is invalid.
///
/// This method uses the default match options and is equivalent to calling
/// `glob_with(pattern, MatchOptions::new())`. Use `glob_with` directly if you
/// want to use non-default match options.
///
/// When iterating, each result is a `GlobResult` which expresses the
/// possibility that there was an `IoError` when attempting to read the contents
/// of the matched path. In other words, each item returned by the iterator
/// will either be an `Ok(Path)` if the path matched, or an `Err(GlobError)` if
/// the path (partially) matched _but_ its contents could not be read in order
/// to determine if its contents matched.
///
/// See the `Paths` documentation for more information.
///
/// # Examples
///
/// Consider a directory `/media/pictures` containing only the files
/// `kittens.jpg`, `puppies.jpg` and `hamsters.gif`:
///
/// ```rust,no_run
/// use glob::glob;
///
/// for entry in glob("/media/pictures/*.jpg").unwrap() {
/// match entry {
/// Ok(path) => println!("{:?}", path.display()),
///
/// // if the path matched but was unreadable,
/// // thereby preventing its contents from matching
/// Err(e) => println!("{:?}", e),
/// }
/// }
/// ```
///
/// The above code will print:
///
/// ```ignore
/// /media/pictures/kittens.jpg
/// /media/pictures/puppies.jpg
/// ```
///
/// If you want to ignore unreadable paths, you can use something like
/// `filter_map`:
///
/// ```rust
/// use glob::glob;
/// use std::result::Result;
///
/// for path in glob("/media/pictures/*.jpg").unwrap().filter_map(Result::ok) {
/// println!("{}", path.display());
/// }
/// ```
/// Paths are yielded in alphabetical order.
pub fn glob(pattern: &str) -> Result<Paths, PatternError> {
glob_with(pattern, MatchOptions::new())
}
/// Return an iterator that produces all the `Path`s that match the given
/// pattern using the specified match options, which may be absolute or relative
/// to the current working directory.
///
/// This may return an error if the pattern is invalid.
///
/// This function accepts Unix shell style patterns as described by
/// `Pattern::new(..)`. The options given are passed through unchanged to
/// `Pattern::matches_with(..)` with the exception that
/// `require_literal_separator` is always set to `true` regardless of the value
/// passed to this function.
///
/// Paths are yielded in alphabetical order.
pub fn glob_with(pattern: &str, options: MatchOptions) -> Result<Paths, PatternError> {
#[cfg(windows)]
fn check_windows_verbatim(p: &Path) -> bool {
use std::path::Prefix;
match p.components().next() {
Some(Component::Prefix(ref p)) => p.kind().is_verbatim(),
_ => false,
}
}
#[cfg(not(windows))]
fn check_windows_verbatim(_: &Path) -> bool {
false
}
#[cfg(windows)]
fn to_scope(p: &Path) -> PathBuf {
// FIXME handle volume relative paths here
p.to_path_buf()
}
#[cfg(not(windows))]
fn to_scope(p: &Path) -> PathBuf {
p.to_path_buf()
}
// make sure that the pattern is valid first, else early return with error
if let Err(err) = Pattern::new(pattern) {
return Err(err);
}
let mut components = Path::new(pattern).components().peekable();
loop {
match components.peek() {
Some(&Component::Prefix(..)) | Some(&Component::RootDir) => {
components.next();
}
_ => break,
}
}
let rest = components.map(|s| s.as_os_str()).collect::<PathBuf>();
let normalized_pattern = Path::new(pattern).iter().collect::<PathBuf>();
let root_len = normalized_pattern.to_str().unwrap().len() - rest.to_str().unwrap().len();
let root = if root_len > 0 {
Some(Path::new(&pattern[..root_len]))
} else {
None
};
if root_len > 0 && check_windows_verbatim(root.unwrap()) {
// FIXME: How do we want to handle verbatim paths? I'm inclined to
// return nothing, since we can't very well find all UNC shares with a
// 1-letter server name.
return Ok(Paths {
dir_patterns: Vec::new(),
require_dir: false,
options,
todo: Vec::new(),
scope: None,
});
}
let scope = root.map_or_else(|| PathBuf::from("."), to_scope);
let mut dir_patterns = Vec::new();
let components =
pattern[cmp::min(root_len, pattern.len())..].split_terminator(path::is_separator);
for component in components {
dir_patterns.push(Pattern::new(component)?);
}
if root_len == pattern.len() {
dir_patterns.push(Pattern {
original: "".to_string(),
tokens: Vec::new(),
is_recursive: false,
});
}
let last_is_separator = pattern.chars().next_back().map(path::is_separator);
let require_dir = last_is_separator == Some(true);
let todo = Vec::new();
Ok(Paths {
dir_patterns,
require_dir,
options,
todo,
scope: Some(scope),
})
}
/// A glob iteration error.
///
/// This is typically returned when a particular path cannot be read
/// to determine if its contents match the glob pattern. This is possible
/// if the program lacks the appropriate permissions, for example.
#[derive(Debug)]
pub struct GlobError {
path: PathBuf,
error: io::Error,
}
impl GlobError {
/// The Path that the error corresponds to.
pub fn path(&self) -> &Path {
&self.path
}
/// The error in question.
pub fn error(&self) -> &io::Error {
&self.error
}
/// Consumes self, returning the _raw_ underlying `io::Error`
pub fn into_error(self) -> io::Error {
self.error
}
}
impl Error for GlobError {
fn description(&self) -> &str {
self.error.description()
}
fn cause(&self) -> Option<&Error> {
Some(&self.error)
}
}
impl fmt::Display for GlobError {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
write!(
f,
"attempting to read `{}` resulted in an error: {}",
self.path.display(),
self.error
)
}
}
fn is_dir(p: &Path) -> bool {
fs::metadata(p).map(|m| m.is_dir()).unwrap_or(false)
}
/// An alias for a glob iteration result.
///
/// This represents either a matched path or a glob iteration error,
/// such as failing to read a particular directory's contents.
pub type GlobResult = Result<PathBuf, GlobError>;
impl Iterator for Paths {
type Item = GlobResult;
fn next(&mut self) -> Option<GlobResult> {
// the todo buffer hasn't been initialized yet, so it's done at this
// point rather than in glob() so that the errors are unified that is,
// failing to fill the buffer is an iteration error construction of the
// iterator (i.e. glob()) only fails if it fails to compile the Pattern
if let Some(scope) = self.scope.take() {
if !self.dir_patterns.is_empty() {
// Shouldn't happen, but we're using -1 as a special index.
assert!(self.dir_patterns.len() < !0 as usize);
fill_todo(&mut self.todo, &self.dir_patterns, 0, &scope, self.options);
}
}
loop {
if self.dir_patterns.is_empty() || self.todo.is_empty() {
return None;
}
let (path, mut idx) = match self.todo.pop().unwrap() {
Ok(pair) => pair,
Err(e) => return Some(Err(e)),
};
// idx -1: was already checked by fill_todo, maybe path was '.' or
// '..' that we can't match here because of normalization.
if idx == !0 as usize {
if self.require_dir && !is_dir(&path) {
continue;
}
return Some(Ok(path));
}
if self.dir_patterns[idx].is_recursive {
let mut next = idx;
// collapse consecutive recursive patterns
while (next + 1) < self.dir_patterns.len()
&& self.dir_patterns[next + 1].is_recursive
{
next += 1;
}
if is_dir(&path) {
// the path is a directory, so it's a match
// push this directory's contents
fill_todo(
&mut self.todo,
&self.dir_patterns,
next,
&path,
self.options,
);
if next == self.dir_patterns.len() - 1 {
// pattern ends in recursive pattern, so return this
// directory as a result
return Some(Ok(path));
} else {
// advanced to the next pattern for this path
idx = next + 1;
}
} else if next == self.dir_patterns.len() - 1 {
// not a directory and it's the last pattern, meaning no
// match
continue;
} else {
// advanced to the next pattern for this path
idx = next + 1;
}
}
// not recursive, so match normally
if self.dir_patterns[idx].matches_with(
{
match path.file_name().and_then(|s| s.to_str()) {
// FIXME (#9639): How do we handle non-utf8 filenames?
// Ignore them for now; ideally we'd still match them
// against a *
None => continue,
Some(x) => x,
}
},
self.options,
) {
if idx == self.dir_patterns.len() - 1 {
// it is not possible for a pattern to match a directory
// *AND* its children so we don't need to check the
// children
if !self.require_dir || is_dir(&path) {
return Some(Ok(path));
}
} else {
fill_todo(
&mut self.todo,
&self.dir_patterns,
idx + 1,
&path,
self.options,
);
}
}
}
}
}
/// A pattern parsing error.
#[derive(Debug)]
#[allow(missing_copy_implementations)]
pub struct PatternError {
/// The approximate character index of where the error occurred.
pub pos: usize,
/// A message describing the error.
pub msg: &'static str,
}
impl Error for PatternError {
fn description(&self) -> &str {
self.msg
}
}
impl fmt::Display for PatternError {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
write!(
f,
"Pattern syntax error near position {}: {}",
self.pos, self.msg
)
}
}
/// A compiled Unix shell style pattern.
///
/// - `?` matches any single character.
///
/// - `*` matches any (possibly empty) sequence of characters.
///
/// - `**` matches the current directory and arbitrary subdirectories. This
/// sequence **must** form a single path component, so both `**a` and `b**`
/// are invalid and will result in an error. A sequence of more than two
/// consecutive `*` characters is also invalid.
///
/// - `[...]` matches any character inside the brackets. Character sequences
/// can also specify ranges of characters, as ordered by Unicode, so e.g.
/// `[0-9]` specifies any character between 0 and 9 inclusive. An unclosed
/// bracket is invalid.
///
/// - `[!...]` is the negation of `[...]`, i.e. it matches any characters
/// **not** in the brackets.
///
/// - The metacharacters `?`, `*`, `[`, `]` can be matched by using brackets
/// (e.g. `[?]`). When a `]` occurs immediately following `[` or `[!` then it
/// is interpreted as being part of, rather then ending, the character set, so
/// `]` and NOT `]` can be matched by `[]]` and `[!]]` respectively. The `-`
/// character can be specified inside a character sequence pattern by placing
/// it at the start or the end, e.g. `[abc-]`.
#[derive(Clone, PartialEq, Eq, PartialOrd, Ord, Hash, Default, Debug)]
pub struct Pattern {
original: String,
tokens: Vec<PatternToken>,
is_recursive: bool,
}
/// Show the original glob pattern.
impl fmt::Display for Pattern {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
self.original.fmt(f)
}
}
impl FromStr for Pattern {
type Err = PatternError;
fn from_str(s: &str) -> Result<Self, PatternError> {
Self::new(s)
}
}
#[derive(Clone, PartialEq, Eq, PartialOrd, Ord, Hash, Debug)]
enum PatternToken {
Char(char),
AnyChar,
AnySequence,
AnyRecursiveSequence,
AnyWithin(Vec<CharSpecifier>),
AnyExcept(Vec<CharSpecifier>),
}
#[derive(Copy, Clone, PartialEq, Eq, PartialOrd, Ord, Hash, Debug)]
enum CharSpecifier {
SingleChar(char),
CharRange(char, char),
}
#[derive(Copy, Clone, PartialEq)]
enum MatchResult {
Match,
SubPatternDoesntMatch,
EntirePatternDoesntMatch,
}
const ERROR_WILDCARDS: &str = "wildcards are either regular `*` or recursive `**`";
const ERROR_RECURSIVE_WILDCARDS: &str = "recursive wildcards must form a single path \
component";
const ERROR_INVALID_RANGE: &str = "invalid range pattern";
impl Pattern {
/// This function compiles Unix shell style patterns.
///
/// An invalid glob pattern will yield a `PatternError`.
pub fn new(pattern: &str) -> Result<Self, PatternError> {
let chars = pattern.chars().collect::<Vec<_>>();
let mut tokens = Vec::new();
let mut is_recursive = false;
let mut i = 0;
while i < chars.len() {
match chars[i] {
'?' => {
tokens.push(AnyChar);
i += 1;
}
'*' => {
let old = i;
while i < chars.len() && chars[i] == '*' {
i += 1;
}
let count = i - old;
if count > 2 {
return Err(PatternError {
pos: old + 2,
msg: ERROR_WILDCARDS,
});
} else if count == 2 {
// ** can only be an entire path component
// i.e. a/**/b is valid, but a**/b or a/**b is not
// invalid matches are treated literally
let is_valid = if i == 2 || path::is_separator(chars[i - count - 1]) {
// it ends in a '/'
if i < chars.len() && path::is_separator(chars[i]) {
i += 1;
true
// or the pattern ends here
// this enables the existing globbing mechanism
} else if i == chars.len() {
true
// `**` ends in non-separator
} else {
return Err(PatternError {
pos: i,
msg: ERROR_RECURSIVE_WILDCARDS,
});
}
// `**` begins with non-separator
} else {
return Err(PatternError {
pos: old - 1,
msg: ERROR_RECURSIVE_WILDCARDS,
});
};
let tokens_len = tokens.len();
if is_valid {
// collapse consecutive AnyRecursiveSequence to a
// single one
if !(tokens_len > 1 && tokens[tokens_len - 1] == AnyRecursiveSequence) {
is_recursive = true;
tokens.push(AnyRecursiveSequence);
}
}
} else {
tokens.push(AnySequence);
}
}
'[' => {
if i + 4 <= chars.len() && chars[i + 1] == '!' {
match chars[i + 3..].iter().position(|x| *x == ']') {
None => (),
Some(j) => {
let chars = &chars[i + 2..i + 3 + j];
let cs = parse_char_specifiers(chars);
tokens.push(AnyExcept(cs));
i += j + 4;
continue;
}
}
} else if i + 3 <= chars.len() && chars[i + 1] != '!' {
match chars[i + 2..].iter().position(|x| *x == ']') {
None => (),
Some(j) => {
let cs = parse_char_specifiers(&chars[i + 1..i + 2 + j]);
tokens.push(AnyWithin(cs));
i += j + 3;
continue;
}
}
}
// if we get here then this is not a valid range pattern
return Err(PatternError {
pos: i,
msg: ERROR_INVALID_RANGE,
});
}
c => {
tokens.push(Char(c));
i += 1;
}
}
}
Ok(Self {
tokens,
original: pattern.to_string(),
is_recursive,
})
}
/// Escape metacharacters within the given string by surrounding them in
/// brackets. The resulting string will, when compiled into a `Pattern`,
/// match the input string and nothing else.
pub fn escape(s: &str) -> String {
let mut escaped = String::new();
for c in s.chars() {
match c {
// note that ! does not need escaping because it is only special
// inside brackets
'?' | '*' | '[' | ']' => {
escaped.push('[');
escaped.push(c);
escaped.push(']');
}
c => {
escaped.push(c);
}
}
}
escaped
}
/// Return if the given `str` matches this `Pattern` using the default
/// match options (i.e. `MatchOptions::new()`).
///
/// # Examples
///
/// ```rust
/// use glob::Pattern;
///
/// assert!(Pattern::new("c?t").unwrap().matches("cat"));
/// assert!(Pattern::new("k[!e]tteh").unwrap().matches("kitteh"));
/// assert!(Pattern::new("d*g").unwrap().matches("doog"));
/// ```
pub fn matches(&self, str: &str) -> bool {
self.matches_with(str, MatchOptions::new())
}
/// Return if the given `Path`, when converted to a `str`, matches this
/// `Pattern` using the default match options (i.e. `MatchOptions::new()`).
pub fn matches_path(&self, path: &Path) -> bool {
// FIXME (#9639): This needs to handle non-utf8 paths
path.to_str().map_or(false, |s| self.matches(s))
}
/// Return if the given `str` matches this `Pattern` using the specified
/// match options.
pub fn matches_with(&self, str: &str, options: MatchOptions) -> bool {
self.matches_from(true, str.chars(), 0, options) == Match
}
/// Return if the given `Path`, when converted to a `str`, matches this
/// `Pattern` using the specified match options.
pub fn matches_path_with(&self, path: &Path, options: MatchOptions) -> bool {
// FIXME (#9639): This needs to handle non-utf8 paths
path.to_str()
.map_or(false, |s| self.matches_with(s, options))
}
/// Access the original glob pattern.
pub fn as_str(&self) -> &str {
&self.original
}
fn matches_from(
&self,
mut follows_separator: bool,
mut file: std::str::Chars,
i: usize,
options: MatchOptions,
) -> MatchResult {
for (ti, token) in self.tokens[i..].iter().enumerate() {
match *token {
AnySequence | AnyRecursiveSequence => {
// ** must be at the start.
debug_assert!(match *token {
AnyRecursiveSequence => follows_separator,
_ => true,
});
// Empty match
match self.matches_from(follows_separator, file.clone(), i + ti + 1, options) {
SubPatternDoesntMatch => (), // keep trying
m => return m,
};
while let Some(c) = file.next() {
if follows_separator && options.require_literal_leading_dot && c == '.' {
return SubPatternDoesntMatch;
}
follows_separator = path::is_separator(c);
match *token {
AnyRecursiveSequence if !follows_separator => continue,
AnySequence
if options.require_literal_separator && follows_separator =>
{
return SubPatternDoesntMatch
}
_ => (),
}
match self.matches_from(
follows_separator,
file.clone(),
i + ti + 1,
options,
) {
SubPatternDoesntMatch => (), // keep trying
m => return m,
}
}
}
_ => {
let c = match file.next() {
Some(c) => c,
None => return EntirePatternDoesntMatch,
};
let is_sep = path::is_separator(c);
if !match *token {
AnyChar | AnyWithin(..) | AnyExcept(..)
if (options.require_literal_separator && is_sep)
|| (follows_separator
&& options.require_literal_leading_dot
&& c == '.') =>
{
false
}
AnyChar => true,
AnyWithin(ref specifiers) => in_char_specifiers(&specifiers, c, options),
AnyExcept(ref specifiers) => !in_char_specifiers(&specifiers, c, options),
Char(c2) => chars_eq(c, c2, options.case_sensitive),
AnySequence | AnyRecursiveSequence => unreachable!(),
} {
return SubPatternDoesntMatch;
}
follows_separator = is_sep;
}
}
}
// Iter is fused.
if file.next().is_none() {
Match
} else {
SubPatternDoesntMatch
}
}
}
// Fills `todo` with paths under `path` to be matched by `patterns[idx]`,
// special-casing patterns to match `.` and `..`, and avoiding `readdir()`
// calls when there are no metacharacters in the pattern.
fn fill_todo(
todo: &mut Vec<Result<(PathBuf, usize), GlobError>>,
patterns: &[Pattern],
idx: usize,
path: &Path,
options: MatchOptions,
) {
// convert a pattern that's just many Char(_) to a string
fn pattern_as_str(pattern: &Pattern) -> Option<String> {
let mut s = String::new();
for token in &pattern.tokens {
match *token {
Char(c) => s.push(c),
_ => return None,
}
}
Some(s)
}
let add = |todo: &mut Vec<_>, next_path: PathBuf| {
if idx + 1 == patterns.len() {
// We know it's good, so don't make the iterator match this path
// against the pattern again. In particular, it can't match
// . or .. globs since these never show up as path components.
todo.push(Ok((next_path, !0 as usize)));
} else {
fill_todo(todo, patterns, idx + 1, &next_path, options);
}
};
let pattern = &patterns[idx];
let is_dir = is_dir(path);
let curdir = path == Path::new(".");
match pattern_as_str(pattern) {
Some(s) => {
// This pattern component doesn't have any metacharacters, so we
// don't need to read the current directory to know where to
// continue. So instead of passing control back to the iterator,
// we can just check for that one entry and potentially recurse
// right away.
let special = "." == s || ".." == s;
let next_path = if curdir {
PathBuf::from(s)
} else {
path.join(&s)
};
if (special && is_dir) || (!special && fs::metadata(&next_path).is_ok()) {
add(todo, next_path);
}
}
None if is_dir => {
let dirs = fs::read_dir(path).and_then(|d| {
d.map(|e| {
e.map(|e| {
if curdir {
PathBuf::from(e.path().file_name().unwrap())
} else {
e.path()
}
})
})
.collect::<Result<Vec<_>, _>>()
});
match dirs {
Ok(mut children) => {
children.sort_by(|p1, p2| p2.file_name().cmp(&p1.file_name()));
todo.extend(children.into_iter().map(|x| Ok((x, idx))));
// Matching the special directory entries . and .. that
// refer to the current and parent directory respectively
// requires that the pattern has a leading dot, even if the
// `MatchOptions` field `require_literal_leading_dot` is not
// set.
if !pattern.tokens.is_empty() && pattern.tokens[0] == Char('.') {
for &special in &[".", ".."] {
if pattern.matches_with(special, options) {
add(todo, path.join(special));
}
}
}
}
Err(e) => {
todo.push(Err(GlobError {
path: path.to_path_buf(),
error: e,
}));
}
}
}
None => {
// not a directory, nothing more to find
}
}
}
fn parse_char_specifiers(s: &[char]) -> Vec<CharSpecifier> {
let mut cs = Vec::new();
let mut i = 0;
while i < s.len() {
if i + 3 <= s.len() && s[i + 1] == '-' {
cs.push(CharRange(s[i], s[i + 2]));
i += 3;
} else {
cs.push(SingleChar(s[i]));
i += 1;
}
}
cs
}
fn in_char_specifiers(specifiers: &[CharSpecifier], c: char, options: MatchOptions) -> bool {
for &specifier in specifiers.iter() {
match specifier {
SingleChar(sc) => {
if chars_eq(c, sc, options.case_sensitive) {
return true;
}
}
CharRange(start, end) => {
// FIXME: work with non-ascii chars properly (issue #1347)
if !options.case_sensitive && c.is_ascii() && start.is_ascii() && end.is_ascii() {
let start = start.to_ascii_lowercase();
let end = end.to_ascii_lowercase();
let start_up = start.to_uppercase().next().unwrap();
let end_up = end.to_uppercase().next().unwrap();
// only allow case insensitive matching when
// both start and end are within a-z or A-Z
if start != start_up && end != end_up {
let c = c.to_ascii_lowercase();
if c >= start && c <= end {
return true;
}
}
}
if c >= start && c <= end {
return true;
}
}
}
}
false
}
/// A helper function to determine if two chars are (possibly case-insensitively) equal.
fn chars_eq(a: char, b: char, case_sensitive: bool) -> bool {
if cfg!(windows) && path::is_separator(a) && path::is_separator(b) {
true
} else if !case_sensitive && a.is_ascii() && b.is_ascii() {
// FIXME: work with non-ascii chars properly (issue #9084)
a.to_ascii_lowercase() == b.to_ascii_lowercase()
} else {
a == b
}
}
/// Configuration options to modify the behaviour of `Pattern::matches_with(..)`.
#[allow(missing_copy_implementations)]
#[derive(Copy, Clone, PartialEq, Eq, PartialOrd, Ord, Hash, Default)]
pub struct MatchOptions {
/// Whether or not patterns should be matched in a case-sensitive manner.
/// This currently only considers upper/lower case relationships between
/// ASCII characters, but in future this might be extended to work with
/// Unicode.
pub case_sensitive: bool,
/// Whether or not path-component separator characters (e.g. `/` on
/// Posix) must be matched by a literal `/`, rather than by `*` or `?` or
/// `[...]`.
pub require_literal_separator: bool,
/// Whether or not paths that contain components that start with a `.`
/// will require that `.` appears literally in the pattern; `*`, `?`, `**`,
/// or `[...]` will not match. This is useful because such files are
/// conventionally considered hidden on Unix systems and it might be
/// desirable to skip them when listing files.
pub require_literal_leading_dot: bool,
}
impl MatchOptions {
/// Constructs a new `MatchOptions` with default field values. This is used
/// when calling functions that do not take an explicit `MatchOptions`
/// parameter.
///
/// This function always returns this value:
///
/// ```rust,ignore
/// MatchOptions {
/// case_sensitive: true,
/// require_literal_separator: false,
/// require_literal_leading_dot: false
/// }
/// ```
pub fn new() -> Self {
Self {
case_sensitive: true,
require_literal_separator: false,
require_literal_leading_dot: false,
}
}
}
#[cfg(test)]
mod test {
use super::{glob, MatchOptions, Pattern};
use std::path::Path;
#[test]
fn test_pattern_from_str() {
assert!("a*b".parse::<Pattern>().unwrap().matches("a_b"));
assert!("a/**b".parse::<Pattern>().unwrap_err().pos == 4);
}
#[test]
fn test_wildcard_errors() {
assert!(Pattern::new("a/**b").unwrap_err().pos == 4);
assert!(Pattern::new("a/bc**").unwrap_err().pos == 3);
assert!(Pattern::new("a/*****").unwrap_err().pos == 4);
assert!(Pattern::new("a/b**c**d").unwrap_err().pos == 2);
assert!(Pattern::new("a**b").unwrap_err().pos == 0);
}
#[test]
fn test_unclosed_bracket_errors() {
assert!(Pattern::new("abc[def").unwrap_err().pos == 3);
assert!(Pattern::new("abc[!def").unwrap_err().pos == 3);
assert!(Pattern::new("abc[").unwrap_err().pos == 3);
assert!(Pattern::new("abc[!").unwrap_err().pos == 3);
assert!(Pattern::new("abc[d").unwrap_err().pos == 3);
assert!(Pattern::new("abc[!d").unwrap_err().pos == 3);
assert!(Pattern::new("abc[]").unwrap_err().pos == 3);
assert!(Pattern::new("abc[!]").unwrap_err().pos == 3);
}
#[test]
fn test_glob_errors() {
assert!(glob("a/**b").err().unwrap().pos == 4);
assert!(glob("abc[def").err().unwrap().pos == 3);
}
// this test assumes that there is a /root directory and that
// the user running this test is not root or otherwise doesn't
// have permission to read its contents
#[cfg(all(unix, not(target_os = "macos")))]
#[test]
fn test_iteration_errors() {
use std::io;
let mut iter = glob("/root/*").unwrap();
// GlobErrors shouldn't halt iteration
let next = iter.next();
assert!(next.is_some());
let err = next.unwrap();
assert!(err.is_err());
let err = err.err().unwrap();
assert!(err.path() == Path::new("/root"));
assert!(err.error().kind() == io::ErrorKind::PermissionDenied);
}
#[test]
fn test_absolute_pattern() {
assert!(glob("/").unwrap().next().is_some());
assert!(glob("//").unwrap().next().is_some());
// assume that the filesystem is not empty!
assert!(glob("/*").unwrap().next().is_some());
#[cfg(not(windows))]
fn win() {}
#[cfg(windows)]
fn win() {
use std::env::current_dir;
use std::ffi::AsOsStr;
// check windows absolute paths with host/device components
let root_with_device = current_dir()
.ok()
.and_then(|p| p.prefix().map(|p| p.join("*")))
.unwrap();
// FIXME (#9639): This needs to handle non-utf8 paths
assert!(glob(root_with_device.as_os_str().to_str().unwrap())
.unwrap()
.next()
.is_some());
}
win()
}
#[test]
fn test_wildcards() {
assert!(Pattern::new("a*b").unwrap().matches("a_b"));
assert!(Pattern::new("a*b*c").unwrap().matches("abc"));
assert!(!Pattern::new("a*b*c").unwrap().matches("abcd"));
assert!(Pattern::new("a*b*c").unwrap().matches("a_b_c"));
assert!(Pattern::new("a*b*c").unwrap().matches("a___b___c"));
assert!(Pattern::new("abc*abc*abc")
.unwrap()
.matches("abcabcabcabcabcabcabc"));
assert!(!Pattern::new("abc*abc*abc")
.unwrap()
.matches("abcabcabcabcabcabcabca"));
assert!(Pattern::new("a*a*a*a*a*a*a*a*a")
.unwrap()
.matches("aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa"));
assert!(Pattern::new("a*b[xyz]c*d").unwrap().matches("abxcdbxcddd"));
}
#[test]
fn test_recursive_wildcards() {
let pat = Pattern::new("some/**/needle.txt").unwrap();
assert!(pat.matches("some/needle.txt"));
assert!(pat.matches("some/one/needle.txt"));
assert!(pat.matches("some/one/two/needle.txt"));
assert!(pat.matches("some/other/needle.txt"));
assert!(!pat.matches("some/other/notthis.txt"));
// a single ** should be valid, for globs
// Should accept anything
let pat = Pattern::new("**").unwrap();
assert!(pat.is_recursive);
assert!(pat.matches("abcde"));
assert!(pat.matches(""));
assert!(pat.matches(".asdf"));
assert!(pat.matches("/x/.asdf"));
// collapse consecutive wildcards
let pat = Pattern::new("some/**/**/needle.txt").unwrap();
assert!(pat.matches("some/needle.txt"));
assert!(pat.matches("some/one/needle.txt"));
assert!(pat.matches("some/one/two/needle.txt"));
assert!(pat.matches("some/other/needle.txt"));
assert!(!pat.matches("some/other/notthis.txt"));
// ** can begin the pattern
let pat = Pattern::new("**/test").unwrap();
assert!(pat.matches("one/two/test"));
assert!(pat.matches("one/test"));
assert!(pat.matches("test"));
// /** can begin the pattern
let pat = Pattern::new("/**/test").unwrap();
assert!(pat.matches("/one/two/test"));
assert!(pat.matches("/one/test"));
assert!(pat.matches("/test"));
assert!(!pat.matches("/one/notthis"));
assert!(!pat.matches("/notthis"));
// Only start sub-patterns on start of path segment.
let pat = Pattern::new("**/.*").unwrap();
assert!(pat.matches(".abc"));
assert!(pat.matches("abc/.abc"));
assert!(!pat.matches("ab.c"));
assert!(!pat.matches("abc/ab.c"));
}
#[test]
fn test_lots_of_files() {
// this is a good test because it touches lots of differently named files
glob("/*/*/*/*").unwrap().skip(10000).next();
}
#[test]
fn test_range_pattern() {
let pat = Pattern::new("a[0-9]b").unwrap();
for i in 0..10 {
assert!(pat.matches(&format!("a{}b", i)));
}
assert!(!pat.matches("a_b"));
let pat = Pattern::new("a[!0-9]b").unwrap();
for i in 0..10 {
assert!(!pat.matches(&format!("a{}b", i)));
}
assert!(pat.matches("a_b"));
let pats = ["[a-z123]", "[1a-z23]", "[123a-z]"];
for &p in pats.iter() {
let pat = Pattern::new(p).unwrap();
for c in "abcdefghijklmnopqrstuvwxyz".chars() {
assert!(pat.matches(&c.to_string()));
}
for c in "ABCDEFGHIJKLMNOPQRSTUVWXYZ".chars() {
let options = MatchOptions {
case_sensitive: false,
..MatchOptions::new()
};
assert!(pat.matches_with(&c.to_string(), options));
}
assert!(pat.matches("1"));
assert!(pat.matches("2"));
assert!(pat.matches("3"));
}
let pats = ["[abc-]", "[-abc]", "[a-c-]"];
for &p in pats.iter() {
let pat = Pattern::new(p).unwrap();
assert!(pat.matches("a"));
assert!(pat.matches("b"));
assert!(pat.matches("c"));
assert!(pat.matches("-"));
assert!(!pat.matches("d"));
}
let pat = Pattern::new("[2-1]").unwrap();
assert!(!pat.matches("1"));
assert!(!pat.matches("2"));
assert!(Pattern::new("[-]").unwrap().matches("-"));
assert!(!Pattern::new("[!-]").unwrap().matches("-"));
}
#[test]
fn test_pattern_matches() {
let txt_pat = Pattern::new("*hello.txt").unwrap();
assert!(txt_pat.matches("hello.txt"));
assert!(txt_pat.matches("gareth_says_hello.txt"));
assert!(txt_pat.matches("some/path/to/hello.txt"));
assert!(txt_pat.matches("some\\path\\to\\hello.txt"));
assert!(txt_pat.matches("/an/absolute/path/to/hello.txt"));
assert!(!txt_pat.matches("hello.txt-and-then-some"));
assert!(!txt_pat.matches("goodbye.txt"));
let dir_pat = Pattern::new("*some/path/to/hello.txt").unwrap();
assert!(dir_pat.matches("some/path/to/hello.txt"));
assert!(dir_pat.matches("a/bigger/some/path/to/hello.txt"));
assert!(!dir_pat.matches("some/path/to/hello.txt-and-then-some"));
assert!(!dir_pat.matches("some/other/path/to/hello.txt"));
}
#[test]
fn test_pattern_escape() {
let s = "_[_]_?_*_!_";
assert_eq!(Pattern::escape(s), "_[[]_[]]_[?]_[*]_!_".to_string());
assert!(Pattern::new(&Pattern::escape(s)).unwrap().matches(s));
}
#[test]
fn test_pattern_matches_case_insensitive() {
let pat = Pattern::new("aBcDeFg").unwrap();
let options = MatchOptions {
case_sensitive: false,
require_literal_separator: false,
require_literal_leading_dot: false,
};
assert!(pat.matches_with("aBcDeFg", options));
assert!(pat.matches_with("abcdefg", options));
assert!(pat.matches_with("ABCDEFG", options));
assert!(pat.matches_with("AbCdEfG", options));
}
#[test]
fn test_pattern_matches_case_insensitive_range() {
let pat_within = Pattern::new("[a]").unwrap();
let pat_except = Pattern::new("[!a]").unwrap();
let options_case_insensitive = MatchOptions {
case_sensitive: false,
require_literal_separator: false,
require_literal_leading_dot: false,
};
let options_case_sensitive = MatchOptions {
case_sensitive: true,
require_literal_separator: false,
require_literal_leading_dot: false,
};
assert!(pat_within.matches_with("a", options_case_insensitive));
assert!(pat_within.matches_with("A", options_case_insensitive));
assert!(!pat_within.matches_with("A", options_case_sensitive));
assert!(!pat_except.matches_with("a", options_case_insensitive));
assert!(!pat_except.matches_with("A", options_case_insensitive));
assert!(pat_except.matches_with("A", options_case_sensitive));
}
#[test]
fn test_pattern_matches_require_literal_separator() {
let options_require_literal = MatchOptions {
case_sensitive: true,
require_literal_separator: true,
require_literal_leading_dot: false,
};
let options_not_require_literal = MatchOptions {
case_sensitive: true,
require_literal_separator: false,
require_literal_leading_dot: false,
};
assert!(Pattern::new("abc/def")
.unwrap()
.matches_with("abc/def", options_require_literal));
assert!(!Pattern::new("abc?def")
.unwrap()
.matches_with("abc/def", options_require_literal));
assert!(!Pattern::new("abc*def")
.unwrap()
.matches_with("abc/def", options_require_literal));
assert!(!Pattern::new("abc[/]def")
.unwrap()
.matches_with("abc/def", options_require_literal));
assert!(Pattern::new("abc/def")
.unwrap()
.matches_with("abc/def", options_not_require_literal));
assert!(Pattern::new("abc?def")
.unwrap()
.matches_with("abc/def", options_not_require_literal));
assert!(Pattern::new("abc*def")
.unwrap()
.matches_with("abc/def", options_not_require_literal));
assert!(Pattern::new("abc[/]def")
.unwrap()
.matches_with("abc/def", options_not_require_literal));
}
#[test]
fn test_pattern_matches_require_literal_leading_dot() {
let options_require_literal_leading_dot = MatchOptions {
case_sensitive: true,
require_literal_separator: false,
require_literal_leading_dot: true,
};
let options_not_require_literal_leading_dot = MatchOptions {
case_sensitive: true,
require_literal_separator: false,
require_literal_leading_dot: false,
};
let f = |options| {
Pattern::new("*.txt")
.unwrap()
.matches_with(".hello.txt", options)
};
assert!(f(options_not_require_literal_leading_dot));
assert!(!f(options_require_literal_leading_dot));
let f = |options| {
Pattern::new(".*.*")
.unwrap()
.matches_with(".hello.txt", options)
};
assert!(f(options_not_require_literal_leading_dot));
assert!(f(options_require_literal_leading_dot));
let f = |options| {
Pattern::new("aaa/bbb/*")
.unwrap()
.matches_with("aaa/bbb/.ccc", options)
};
assert!(f(options_not_require_literal_leading_dot));
assert!(!f(options_require_literal_leading_dot));
let f = |options| {
Pattern::new("aaa/bbb/*")
.unwrap()
.matches_with("aaa/bbb/c.c.c.", options)
};
assert!(f(options_not_require_literal_leading_dot));
assert!(f(options_require_literal_leading_dot));
let f = |options| {
Pattern::new("aaa/bbb/.*")
.unwrap()
.matches_with("aaa/bbb/.ccc", options)
};
assert!(f(options_not_require_literal_leading_dot));
assert!(f(options_require_literal_leading_dot));
let f = |options| {
Pattern::new("aaa/?bbb")
.unwrap()
.matches_with("aaa/.bbb", options)
};
assert!(f(options_not_require_literal_leading_dot));
assert!(!f(options_require_literal_leading_dot));
let f = |options| {
Pattern::new("aaa/[.]bbb")
.unwrap()
.matches_with("aaa/.bbb", options)
};
assert!(f(options_not_require_literal_leading_dot));
assert!(!f(options_require_literal_leading_dot));
let f = |options| Pattern::new("**/*").unwrap().matches_with(".bbb", options);
assert!(f(options_not_require_literal_leading_dot));
assert!(!f(options_require_literal_leading_dot));
}
#[test]
fn test_matches_path() {
// on windows, (Path::new("a/b").as_str().unwrap() == "a\\b"), so this
// tests that / and \ are considered equivalent on windows
assert!(Pattern::new("a/b").unwrap().matches_path(&Path::new("a/b")));
}
#[test]
fn test_path_join() {
let pattern = Path::new("one").join(&Path::new("**/*.rs"));
assert!(Pattern::new(pattern.to_str().unwrap()).is_ok());
}
}