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
// 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.

use self::NamesIter::*;
use self::Regex::*;

use std::borrow::IntoCow;
use std::collections::HashMap;
use std::fmt;
use std::string::CowString;

use compile::Program;
use parse;
use vm;
use vm::CaptureLocs;
use vm::MatchKind::{self, Exists, Location, Submatches};

/// Escapes all regular expression meta characters in `text`.
///
/// The string returned may be safely used as a literal in a regular
/// expression.
pub fn quote(text: &str) -> String {
    let mut quoted = String::with_capacity(text.len());
    for c in text.chars() {
        if parse::is_punct(c) {
            quoted.push('\\')
        }
        quoted.push(c);
    }
    quoted
}

/// Tests if the given regular expression matches somewhere in the text given.
///
/// If there was a problem compiling the regular expression, an error is
/// returned.
///
/// To find submatches, split or replace text, you'll need to compile an
/// expression first.
///
/// Note that you should prefer the `regex!` macro when possible. For example,
/// `regex!("...").is_match("...")`.
pub fn is_match(regex: &str, text: &str) -> Result<bool, parse::Error> {
    Regex::new(regex).map(|r| r.is_match(text))
}

/// A compiled regular expression
///
/// It is represented as either a sequence of bytecode instructions (dynamic)
/// or as a specialized Rust function (native). It can be used to search, split
/// or replace text. All searching is done with an implicit `.*?` at the
/// beginning and end of an expression. To force an expression to match the
/// whole string (or a prefix or a suffix), you must use an anchor like `^` or
/// `$` (or `\A` and `\z`).
///
/// While this crate will handle Unicode strings (whether in the regular
/// expression or in the search text), all positions returned are **byte
/// indices**. Every byte index is guaranteed to be at a Unicode code point
/// boundary.
///
/// The lifetimes `'r` and `'t` in this crate correspond to the lifetime of a
/// compiled regular expression and text to search, respectively.
///
/// The only methods that allocate new strings are the string replacement
/// methods. All other methods (searching and splitting) return borrowed
/// pointers into the string given.
///
/// # Examples
///
/// Find the location of a US phone number:
///
/// ```rust
/// # use regex::Regex;
/// let re = match Regex::new("[0-9]{3}-[0-9]{3}-[0-9]{4}") {
///     Ok(re) => re,
///     Err(err) => panic!("{}", err),
/// };
/// assert_eq!(re.find("phone: 111-222-3333"), Some((7, 19)));
/// ```
///
/// You can also use the `regex!` macro to compile a regular expression when
/// you compile your program:
///
/// ```rust
/// #![feature(plugin)]
/// extern crate regex;
/// #[plugin] extern crate regex_macros;
///
/// fn main() {
///     let re = regex!(r"\d+");
///     assert_eq!(re.find("123 abc"), Some((0, 3)));
/// }
/// ```
///
/// Given an incorrect regular expression, `regex!` will cause the Rust
/// compiler to produce a compile time error.
/// Note that `regex!` will compile the expression to native Rust code, which
/// makes it much faster when searching text.
/// More details about the `regex!` macro can be found in the `regex` crate
/// documentation.
#[derive(Clone)]
pub enum Regex {
    // The representation of `Regex` is exported to support the `regex!`
    // syntax extension. Do not rely on it.
    //
    // See the comments for the `program` module in `lib.rs` for a more
    // detailed explanation for what `regex!` requires.
    #[doc(hidden)]
    Dynamic(ExDynamic),
    #[doc(hidden)]
    Native(ExNative),
}

#[derive(Clone)]
#[doc(hidden)]
pub struct ExDynamic {
    original: String,
    names: Vec<Option<String>>,
    #[doc(hidden)]
    pub prog: Program
}

#[doc(hidden)]
pub struct ExNative {
    #[doc(hidden)]
    pub original: &'static str,
    #[doc(hidden)]
    pub names: &'static &'static [Option<&'static str>],
    #[doc(hidden)]
    pub prog: fn(MatchKind, &str, usize, usize) -> Vec<Option<usize>>
}

impl Copy for ExNative {}

impl Clone for ExNative {
    fn clone(&self) -> ExNative {
        *self
    }
}

impl fmt::String for Regex {
    /// Shows the original regular expression.
    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
        write!(f, "{}", self.as_str())
    }
}

impl fmt::Show for Regex {
    /// Shows the original regular expression.
    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
        fmt::String::fmt(self, f)
    }
}

impl Regex {
    /// Compiles a dynamic regular expression. Once compiled, it can be
    /// used repeatedly to search, split or replace text in a string.
    ///
    /// When possible, you should prefer the `regex!` macro since it is
    /// safer and always faster.
    ///
    /// If an invalid expression is given, then an error is returned.
    pub fn new(re: &str) -> Result<Regex, parse::Error> {
        let ast = try!(parse::parse(re));
        let (prog, names) = Program::new(ast);
        Ok(Dynamic(ExDynamic {
            original: re.to_string(),
            names: names,
            prog: prog,
        }))
    }

    /// Returns true if and only if the regex matches the string given.
    ///
    /// # Example
    ///
    /// Test if some text contains at least one word with exactly 13
    /// characters:
    ///
    /// ```rust
    /// # #![feature(plugin)]
    /// # extern crate regex; #[plugin] extern crate regex_macros;
    /// # fn main() {
    /// let text = "I categorically deny having triskaidekaphobia.";
    /// let matched = regex!(r"\b\w{13}\b").is_match(text);
    /// assert!(matched);
    /// # }
    /// ```
    pub fn is_match(&self, text: &str) -> bool {
        has_match(&exec(self, Exists, text))
    }

    /// Returns the start and end byte range of the leftmost-first match in
    /// `text`. If no match exists, then `None` is returned.
    ///
    /// Note that this should only be used if you want to discover the position
    /// of the match. Testing the existence of a match is faster if you use
    /// `is_match`.
    ///
    /// # Example
    ///
    /// Find the start and end location of the first word with exactly 13
    /// characters:
    ///
    /// ```rust
    /// # #![feature(plugin)]
    /// # extern crate regex; #[plugin] extern crate regex_macros;
    /// # fn main() {
    /// let text = "I categorically deny having triskaidekaphobia.";
    /// let pos = regex!(r"\b\w{13}\b").find(text);
    /// assert_eq!(pos, Some((2, 15)));
    /// # }
    /// ```
    pub fn find(&self, text: &str) -> Option<(usize, usize)> {
        let caps = exec(self, Location, text);
        if has_match(&caps) {
            Some((caps[0].unwrap(), caps[1].unwrap()))
        } else {
            None
        }
    }

    /// Returns an iterator for each successive non-overlapping match in
    /// `text`, returning the start and end byte indices with respect to
    /// `text`.
    ///
    /// # Example
    ///
    /// Find the start and end location of every word with exactly 13
    /// characters:
    ///
    /// ```rust
    /// # #![feature(plugin)]
    /// # extern crate regex; #[plugin] extern crate regex_macros;
    /// # fn main() {
    /// let text = "Retroactively relinquishing remunerations is reprehensible.";
    /// for pos in regex!(r"\b\w{13}\b").find_iter(text) {
    ///     println!("{:?}", pos);
    /// }
    /// // Output:
    /// // (0, 13)
    /// // (14, 27)
    /// // (28, 41)
    /// // (45, 58)
    /// # }
    /// ```
    pub fn find_iter<'r, 't>(&'r self, text: &'t str) -> FindMatches<'r, 't> {
        FindMatches {
            re: self,
            search: text,
            last_end: 0,
            last_match: None,
        }
    }

    /// Returns the capture groups corresponding to the leftmost-first
    /// match in `text`. Capture group `0` always corresponds to the entire
    /// match. If no match is found, then `None` is returned.
    ///
    /// You should only use `captures` if you need access to submatches.
    /// Otherwise, `find` is faster for discovering the location of the overall
    /// match.
    ///
    /// # Examples
    ///
    /// Say you have some text with movie names and their release years,
    /// like "'Citizen Kane' (1941)". It'd be nice if we could search for text
    /// looking like that, while also extracting the movie name and its release
    /// year separately.
    ///
    /// ```rust
    /// # #![feature(plugin)]
    /// # extern crate regex; #[plugin] extern crate regex_macros;
    /// # fn main() {
    /// let re = regex!(r"'([^']+)'\s+\((\d{4})\)");
    /// let text = "Not my favorite movie: 'Citizen Kane' (1941).";
    /// let caps = re.captures(text).unwrap();
    /// assert_eq!(caps.at(1), Some("Citizen Kane"));
    /// assert_eq!(caps.at(2), Some("1941"));
    /// assert_eq!(caps.at(0), Some("'Citizen Kane' (1941)"));
    /// # }
    /// ```
    ///
    /// Note that the full match is at capture group `0`. Each subsequent
    /// capture group is indexed by the order of its opening `(`.
    ///
    /// We can make this example a bit clearer by using *named* capture groups:
    ///
    /// ```rust
    /// # #![feature(plugin)]
    /// # extern crate regex; #[plugin] extern crate regex_macros;
    /// # fn main() {
    /// let re = regex!(r"'(?P<title>[^']+)'\s+\((?P<year>\d{4})\)");
    /// let text = "Not my favorite movie: 'Citizen Kane' (1941).";
    /// let caps = re.captures(text).unwrap();
    /// assert_eq!(caps.name("title"), Some("Citizen Kane"));
    /// assert_eq!(caps.name("year"), Some("1941"));
    /// assert_eq!(caps.at(0), Some("'Citizen Kane' (1941)"));
    /// # }
    /// ```
    ///
    /// Here we name the capture groups, which we can access with the `name`
    /// method. Note that the named capture groups are still accessible with
    /// `at`.
    ///
    /// The `0`th capture group is always unnamed, so it must always be
    /// accessed with `at(0)`.
    pub fn captures<'t>(&self, text: &'t str) -> Option<Captures<'t>> {
        let caps = exec(self, Submatches, text);
        Captures::new(self, text, caps)
    }

    /// Returns an iterator over all the non-overlapping capture groups matched
    /// in `text`. This is operationally the same as `find_iter` (except it
    /// yields information about submatches).
    ///
    /// # Example
    ///
    /// We can use this to find all movie titles and their release years in
    /// some text, where the movie is formatted like "'Title' (xxxx)":
    ///
    /// ```rust
    /// # #![feature(plugin)]
    /// # extern crate regex; #[plugin] extern crate regex_macros;
    /// # fn main() {
    /// let re = regex!(r"'(?P<title>[^']+)'\s+\((?P<year>\d{4})\)");
    /// let text = "'Citizen Kane' (1941), 'The Wizard of Oz' (1939), 'M' (1931).";
    /// for caps in re.captures_iter(text) {
    ///     println!("Movie: {:?}, Released: {:?}", caps.name("title"), caps.name("year"));
    /// }
    /// // Output:
    /// // Movie: Citizen Kane, Released: 1941
    /// // Movie: The Wizard of Oz, Released: 1939
    /// // Movie: M, Released: 1931
    /// # }
    /// ```
    pub fn captures_iter<'r, 't>(&'r self, text: &'t str)
                                -> FindCaptures<'r, 't> {
        FindCaptures {
            re: self,
            search: text,
            last_match: None,
            last_end: 0,
        }
    }

    /// Returns an iterator of substrings of `text` delimited by a match
    /// of the regular expression.
    /// Namely, each element of the iterator corresponds to text that *isn't*
    /// matched by the regular expression.
    ///
    /// This method will *not* copy the text given.
    ///
    /// # Example
    ///
    /// To split a string delimited by arbitrary amounts of spaces or tabs:
    ///
    /// ```rust
    /// # #![feature(plugin)]
    /// # extern crate regex; #[plugin] extern crate regex_macros;
    /// # fn main() {
    /// let re = regex!(r"[ \t]+");
    /// let fields: Vec<&str> = re.split("a b \t  c\td    e").collect();
    /// assert_eq!(fields, vec!("a", "b", "c", "d", "e"));
    /// # }
    /// ```
    pub fn split<'r, 't>(&'r self, text: &'t str) -> RegexSplits<'r, 't> {
        RegexSplits {
            finder: self.find_iter(text),
            last: 0,
        }
    }

    /// Returns an iterator of at most `limit` substrings of `text` delimited
    /// by a match of the regular expression. (A `limit` of `0` will return no
    /// substrings.)
    /// Namely, each element of the iterator corresponds to text that *isn't*
    /// matched by the regular expression.
    /// The remainder of the string that is not split will be the last element
    /// in the iterator.
    ///
    /// This method will *not* copy the text given.
    ///
    /// # Example
    ///
    /// Get the first two words in some text:
    ///
    /// ```rust
    /// # #![feature(plugin)]
    /// # extern crate regex; #[plugin] extern crate regex_macros;
    /// # fn main() {
    /// let re = regex!(r"\W+");
    /// let fields: Vec<&str> = re.splitn("Hey! How are you?", 3).collect();
    /// assert_eq!(fields, vec!("Hey", "How", "are you?"));
    /// # }
    /// ```
    pub fn splitn<'r, 't>(&'r self, text: &'t str, limit: usize)
                         -> RegexSplitsN<'r, 't> {
        RegexSplitsN {
            splits: self.split(text),
            cur: 0,
            limit: limit,
        }
    }

    /// Replaces the leftmost-first match with the replacement provided.
    /// The replacement can be a regular string (where `$N` and `$name` are
    /// expanded to match capture groups) or a function that takes the matches'
    /// `Captures` and returns the replaced string.
    ///
    /// If no match is found, then a copy of the string is returned unchanged.
    ///
    /// # Examples
    ///
    /// Note that this function is polymorphic with respect to the replacement.
    /// In typical usage, this can just be a normal string:
    ///
    /// ```rust
    /// # #![feature(plugin)]
    /// # extern crate regex; #[plugin] extern crate regex_macros;
    /// # fn main() {
    /// let re = regex!("[^01]+");
    /// assert_eq!(re.replace("1078910", ""), "1010");
    /// # }
    /// ```
    ///
    /// But anything satisfying the `Replacer` trait will work. For example,
    /// a closure of type `|&Captures| -> String` provides direct access to the
    /// captures corresponding to a match. This allows one to access
    /// submatches easily:
    ///
    /// ```rust
    /// # #![feature(plugin)]
    /// # #![feature(unboxed_closures)]
    /// # extern crate regex; #[plugin] extern crate regex_macros;
    /// # use regex::Captures; fn main() {
    /// let re = regex!(r"([^,\s]+),\s+(\S+)");
    /// let result = re.replace("Springsteen, Bruce", |&: caps: &Captures| {
    ///     format!("{} {}", caps.at(2).unwrap_or(""), caps.at(1).unwrap_or(""))
    /// });
    /// assert_eq!(result, "Bruce Springsteen");
    /// # }
    /// ```
    ///
    /// But this is a bit cumbersome to use all the time. Instead, a simple
    /// syntax is supported that expands `$name` into the corresponding capture
    /// group. Here's the last example, but using this expansion technique
    /// with named capture groups:
    ///
    /// ```rust
    /// # #![feature(plugin)]
    /// # extern crate regex; #[plugin] extern crate regex_macros;
    /// # fn main() {
    /// let re = regex!(r"(?P<last>[^,\s]+),\s+(?P<first>\S+)");
    /// let result = re.replace("Springsteen, Bruce", "$first $last");
    /// assert_eq!(result, "Bruce Springsteen");
    /// # }
    /// ```
    ///
    /// Note that using `$2` instead of `$first` or `$1` instead of `$last`
    /// would produce the same result. To write a literal `$` use `$$`.
    ///
    /// Finally, sometimes you just want to replace a literal string with no
    /// submatch expansion. This can be done by wrapping a string with
    /// `NoExpand`:
    ///
    /// ```rust
    /// # #![feature(plugin)]
    /// # extern crate regex; #[plugin] extern crate regex_macros;
    /// # fn main() {
    /// use regex::NoExpand;
    ///
    /// let re = regex!(r"(?P<last>[^,\s]+),\s+(\S+)");
    /// let result = re.replace("Springsteen, Bruce", NoExpand("$2 $last"));
    /// assert_eq!(result, "$2 $last");
    /// # }
    /// ```
    pub fn replace<R: Replacer>(&self, text: &str, rep: R) -> String {
        self.replacen(text, 1, rep)
    }

    /// Replaces all non-overlapping matches in `text` with the
    /// replacement provided. This is the same as calling `replacen` with
    /// `limit` set to `0`.
    ///
    /// See the documentation for `replace` for details on how to access
    /// submatches in the replacement string.
    pub fn replace_all<R: Replacer>(&self, text: &str, rep: R) -> String {
        self.replacen(text, 0, rep)
    }

    /// Replaces at most `limit` non-overlapping matches in `text` with the
    /// replacement provided. If `limit` is 0, then all non-overlapping matches
    /// are replaced.
    ///
    /// See the documentation for `replace` for details on how to access
    /// submatches in the replacement string.
    pub fn replacen<R: Replacer>
                   (&self, text: &str, limit: usize, mut rep: R) -> String {
        let mut new = String::with_capacity(text.len());
        let mut last_match = 0;

        for (i, cap) in self.captures_iter(text).enumerate() {
            // It'd be nicer to use the 'take' iterator instead, but it seemed
            // awkward given that '0' => no limit.
            if limit > 0 && i >= limit {
                break
            }

            let (s, e) = cap.pos(0).unwrap(); // captures only reports matches
            new.push_str(text.slice(last_match, s));
            new.push_str(rep.reg_replace(&cap).as_slice());
            last_match = e;
        }
        new.push_str(text.slice(last_match, text.len()));
        return new;
    }

    /// Returns the original string of this regex.
    pub fn as_str<'a>(&'a self) -> &'a str {
        match *self {
            Dynamic(ExDynamic { ref original, .. }) => original.as_slice(),
            Native(ExNative { ref original, .. }) => original.as_slice(),
        }
    }

    #[doc(hidden)]
    pub fn names_iter<'a>(&'a self) -> NamesIter<'a> {
        match *self {
            Native(ref n) => NamesIterNative(n.names.iter()),
            Dynamic(ref d) => NamesIterDynamic(d.names.iter())
        }
    }

    fn names_len(&self) -> usize {
        match *self {
            Native(ref n) => n.names.len(),
            Dynamic(ref d) => d.names.len()
        }
    }

}

pub enum NamesIter<'a> {
    NamesIterNative(::std::slice::Iter<'a, Option<&'static str>>),
    NamesIterDynamic(::std::slice::Iter<'a, Option<String>>)
}

impl<'a> Iterator for NamesIter<'a> {
    type Item=Option<String>;

    fn next(&mut self) -> Option<Option<String>> {
        match *self {
            NamesIterNative(ref mut i) => i.next().map(|x| x.map(|s| s.to_string())),
            NamesIterDynamic(ref mut i) => i.next().map(|x| x.as_ref().map(|s| s.to_string())),
        }
    }
}

/// NoExpand indicates literal string replacement.
///
/// It can be used with `replace` and `replace_all` to do a literal
/// string replacement without expanding `$name` to their corresponding
/// capture groups.
///
/// `'r` is the lifetime of the literal text.
pub struct NoExpand<'t>(pub &'t str);

/// Replacer describes types that can be used to replace matches in a string.
pub trait Replacer {
    /// Returns a possibly owned string that is used to replace the match
    /// corresponding to the `caps` capture group.
    ///
    /// The `'a` lifetime refers to the lifetime of a borrowed string when
    /// a new owned string isn't needed (e.g., for `NoExpand`).
    fn reg_replace<'a>(&'a mut self, caps: &Captures) -> CowString<'a>;
}

impl<'t> Replacer for NoExpand<'t> {
    fn reg_replace<'a>(&'a mut self, _: &Captures) -> CowString<'a> {
        let NoExpand(s) = *self;
        s.into_cow()
    }
}

impl<'t> Replacer for &'t str {
    fn reg_replace<'a>(&'a mut self, caps: &Captures) -> CowString<'a> {
        caps.expand(*self).into_cow()
    }
}

impl<F> Replacer for F where F: FnMut(&Captures) -> String {
    fn reg_replace<'a>(&'a mut self, caps: &Captures) -> CowString<'a> {
        (*self)(caps).into_cow()
    }
}

/// Yields all substrings delimited by a regular expression match.
///
/// `'r` is the lifetime of the compiled expression and `'t` is the lifetime
/// of the string being split.
pub struct RegexSplits<'r, 't> {
    finder: FindMatches<'r, 't>,
    last: usize,
}

impl<'r, 't> Iterator for RegexSplits<'r, 't> {
    type Item = &'t str;

    fn next(&mut self) -> Option<&'t str> {
        let text = self.finder.search;
        match self.finder.next() {
            None => {
                if self.last >= text.len() {
                    None
                } else {
                    let s = text.slice(self.last, text.len());
                    self.last = text.len();
                    Some(s)
                }
            }
            Some((s, e)) => {
                let matched = text.slice(self.last, s);
                self.last = e;
                Some(matched)
            }
        }
    }
}

/// Yields at most `N` substrings delimited by a regular expression match.
///
/// The last substring will be whatever remains after splitting.
///
/// `'r` is the lifetime of the compiled expression and `'t` is the lifetime
/// of the string being split.
pub struct RegexSplitsN<'r, 't> {
    splits: RegexSplits<'r, 't>,
    cur: usize,
    limit: usize,
}

impl<'r, 't> Iterator for RegexSplitsN<'r, 't> {
    type Item = &'t str;

    fn next(&mut self) -> Option<&'t str> {
        let text = self.splits.finder.search;
        if self.cur >= self.limit {
            None
        } else {
            self.cur += 1;
            if self.cur >= self.limit {
                Some(text.slice(self.splits.last, text.len()))
            } else {
                self.splits.next()
            }
        }
    }
}

/// Captures represents a group of captured strings for a single match.
///
/// The 0th capture always corresponds to the entire match. Each subsequent
/// index corresponds to the next capture group in the regex.
/// If a capture group is named, then the matched string is *also* available
/// via the `name` method. (Note that the 0th capture is always unnamed and so
/// must be accessed with the `at` method.)
///
/// Positions returned from a capture group are always byte indices.
///
/// `'t` is the lifetime of the matched text.
pub struct Captures<'t> {
    text: &'t str,
    locs: CaptureLocs,
    named: Option<HashMap<String, usize>>,
}

impl<'t> Captures<'t> {
    fn new(re: &Regex, search: &'t str, locs: CaptureLocs)
          -> Option<Captures<'t>> {
        if !has_match(&locs) {
            return None
        }

        let named =
            if re.names_len() == 0 {
                None
            } else {
                let mut named = HashMap::new();
                for (i, name) in re.names_iter().enumerate() {
                    match name {
                        None => {},
                        Some(name) => {
                            named.insert(name, i);
                        }
                    }
                }
                Some(named)
            };
        Some(Captures {
            text: search,
            locs: locs,
            named: named,
        })
    }

    /// Returns the start and end positions of the Nth capture group.
    /// Returns `None` if `i` is not a valid capture group or if the capture
    /// group did not match anything.
    /// The positions returned are *always* byte indices with respect to the
    /// original string matched.
    pub fn pos(&self, i: usize) -> Option<(usize, usize)> {
        let (s, e) = (i * 2, i * 2 + 1);
        if e >= self.locs.len() || self.locs[s].is_none() {
            // VM guarantees that each pair of locations are both Some or None.
            return None
        }
        Some((self.locs[s].unwrap(), self.locs[e].unwrap()))
    }

    /// Returns the matched string for the capture group `i`.  If `i` isn't
    /// a valid capture group or didn't match anything, then `None` is
    /// returned.
    pub fn at(&self, i: usize) -> Option<&'t str> {
        match self.pos(i) {
            None => None,
            Some((s, e)) => Some(self.text.slice(s, e))
        }
    }

    /// Returns the matched string for the capture group named `name`.  If
    /// `name` isn't a valid capture group or didn't match anything, then
    /// `None` is returned.
    pub fn name(&self, name: &str) -> Option<&'t str> {
        match self.named {
            None => None,
            Some(ref h) => {
                match h.get(name) {
                    None => None,
                    Some(i) => self.at(*i),
                }
            }
        }
    }

    /// Creates an iterator of all the capture groups in order of appearance
    /// in the regular expression.
    pub fn iter(&'t self) -> SubCaptures<'t> {
        SubCaptures { idx: 0, caps: self, }
    }

    /// Creates an iterator of all the capture group positions in order of
    /// appearance in the regular expression. Positions are byte indices
    /// in terms of the original string matched.
    pub fn iter_pos(&'t self) -> SubCapturesPos<'t> {
        SubCapturesPos { idx: 0, caps: self, }
    }

    /// Expands all instances of `$name` in `text` to the corresponding capture
    /// group `name`.
    ///
    /// `name` may be an integer corresponding to the index of the
    /// capture group (counted by order of opening parenthesis where `0` is the
    /// entire match) or it can be a name (consisting of letters, digits or
    /// underscores) corresponding to a named capture group.
    ///
    /// If `name` isn't a valid capture group (whether the name doesn't exist or
    /// isn't a valid index), then it is replaced with the empty string.
    ///
    /// To write a literal `$` use `$$`.
    pub fn expand(&self, text: &str) -> String {
        // How evil can you get?
        // FIXME: Don't use regexes for this. It's completely unnecessary.
        let re = Regex::new(r"(^|[^$]|\b)\$(\w+)").unwrap();
        let text = re.replace_all(text, |&mut: refs: &Captures| -> String {
            let pre = refs.at(1).unwrap_or("");
            let name = refs.at(2).unwrap_or("");
            format!("{}{}", pre,
                    match name.parse::<usize>() {
                None => self.name(name).unwrap_or("").to_string(),
                Some(i) => self.at(i).unwrap_or("").to_string(),
            })
        });
        let re = Regex::new(r"\$\$").unwrap();
        re.replace_all(text.as_slice(), NoExpand("$"))
    }

    /// Returns the number of captured groups.
    #[inline]
    pub fn len(&self) -> usize { self.locs.len() / 2 }

    /// Returns if there are no captured groups.
    #[inline]
    pub fn is_empty(&self) -> bool { self.len() == 0 }
}

/// An iterator over capture groups for a particular match of a regular
/// expression.
///
/// `'t` is the lifetime of the matched text.
pub struct SubCaptures<'t> {
    idx: usize,
    caps: &'t Captures<'t>,
}

impl<'t> Iterator for SubCaptures<'t> {
    type Item = &'t str;

    fn next(&mut self) -> Option<&'t str> {
        if self.idx < self.caps.len() {
            self.idx += 1;
            Some(self.caps.at(self.idx - 1).unwrap_or(""))
        } else {
            None
        }
    }
}

/// An iterator over capture group positions for a particular match of a
/// regular expression.
///
/// Positions are byte indices in terms of the original string matched.
///
/// `'t` is the lifetime of the matched text.
pub struct SubCapturesPos<'t> {
    idx: usize,
    caps: &'t Captures<'t>,
}

impl<'t> Iterator for SubCapturesPos<'t> {
    type Item = Option<(usize, usize)>;

    fn next(&mut self) -> Option<Option<(usize, usize)>> {
        if self.idx < self.caps.len() {
            self.idx += 1;
            Some(self.caps.pos(self.idx - 1))
        } else {
            None
        }
    }
}

/// An iterator that yields all non-overlapping capture groups matching a
/// particular regular expression.
///
/// The iterator stops when no more matches can be found.
///
/// `'r` is the lifetime of the compiled expression and `'t` is the lifetime
/// of the matched string.
pub struct FindCaptures<'r, 't> {
    re: &'r Regex,
    search: &'t str,
    last_match: Option<usize>,
    last_end: usize,
}

impl<'r, 't> Iterator for FindCaptures<'r, 't> {
    type Item = Captures<'t>;

    fn next(&mut self) -> Option<Captures<'t>> {
        if self.last_end > self.search.len() {
            return None
        }

        let caps = exec_slice(self.re, Submatches, self.search,
                              self.last_end, self.search.len());
        let (s, e) =
            if !has_match(&caps) {
                return None
            } else {
                (caps[0].unwrap(), caps[1].unwrap())
            };

        // Don't accept empty matches immediately following a match.
        // i.e., no infinite loops please.
        if e == s && Some(self.last_end) == self.last_match {
            self.last_end += 1;
            return self.next()
        }
        self.last_end = e;
        self.last_match = Some(self.last_end);
        Captures::new(self.re, self.search, caps)
    }
}

/// An iterator over all non-overlapping matches for a particular string.
///
/// The iterator yields a tuple of integers corresponding to the start and end
/// of the match. The indices are byte offsets. The iterator stops when no more
/// matches can be found.
///
/// `'r` is the lifetime of the compiled expression and `'t` is the lifetime
/// of the matched string.
pub struct FindMatches<'r, 't> {
    re: &'r Regex,
    search: &'t str,
    last_match: Option<usize>,
    last_end: usize,
}

impl<'r, 't> Iterator for FindMatches<'r, 't> {
    type Item = (usize, usize);

    fn next(&mut self) -> Option<(usize, usize)> {
        if self.last_end > self.search.len() {
            return None
        }

        let caps = exec_slice(self.re, Location, self.search,
                              self.last_end, self.search.len());
        let (s, e) =
            if !has_match(&caps) {
                return None
            } else {
                (caps[0].unwrap(), caps[1].unwrap())
            };

        // Don't accept empty matches immediately following a match.
        // i.e., no infinite loops please.
        if e == s && Some(self.last_end) == self.last_match {
            self.last_end += 1;
            return self.next()
        }
        self.last_end = e;
        self.last_match = Some(self.last_end);
        Some((s, e))
    }
}

fn exec(re: &Regex, which: MatchKind, input: &str) -> CaptureLocs {
    exec_slice(re, which, input, 0, input.len())
}

fn exec_slice(re: &Regex, which: MatchKind,
              input: &str, s: usize, e: usize) -> CaptureLocs {
    match *re {
        Dynamic(ExDynamic { ref prog, .. }) => vm::run(which, prog, input, s, e),
        Native(ExNative { ref prog, .. }) => (*prog)(which, input, s, e),
    }
}

#[inline]
fn has_match(caps: &CaptureLocs) -> bool {
    caps.len() >= 2 && caps[0].is_some() && caps[1].is_some()
}