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
use std::char;
use std::cmp::Ordering;
use std::fmt;
use std::ops;
use std::u32;

use crate::literal::LiteralSearcher;
use crate::prog::InstEmptyLook;
use crate::utf8::{decode_last_utf8, decode_utf8};

/// Represents a location in the input.
#[derive(Clone, Copy, Debug)]
pub struct InputAt {
    pos: usize,
    c: Char,
    byte: Option<u8>,
    len: usize,
}

impl InputAt {
    /// Returns true iff this position is at the beginning of the input.
    pub fn is_start(&self) -> bool {
        self.pos == 0
    }

    /// Returns true iff this position is past the end of the input.
    pub fn is_end(&self) -> bool {
        self.c.is_none() && self.byte.is_none()
    }

    /// Returns the character at this position.
    ///
    /// If this position is just before or after the input, then an absent
    /// character is returned.
    pub fn char(&self) -> Char {
        self.c
    }

    /// Returns the byte at this position.
    pub fn byte(&self) -> Option<u8> {
        self.byte
    }

    /// Returns the UTF-8 width of the character at this position.
    pub fn len(&self) -> usize {
        self.len
    }

    /// Returns whether the UTF-8 width of the character at this position
    /// is zero.
    pub fn is_empty(&self) -> bool {
        self.len == 0
    }

    /// Returns the byte offset of this position.
    pub fn pos(&self) -> usize {
        self.pos
    }

    /// Returns the byte offset of the next position in the input.
    pub fn next_pos(&self) -> usize {
        self.pos + self.len
    }
}

/// An abstraction over input used in the matching engines.
pub trait Input: fmt::Debug {
    /// Return an encoding of the position at byte offset `i`.
    fn at(&self, i: usize) -> InputAt;

    /// Return the Unicode character occurring next to `at`.
    ///
    /// If no such character could be decoded, then `Char` is absent.
    fn next_char(&self, at: InputAt) -> Char;

    /// Return the Unicode character occurring previous to `at`.
    ///
    /// If no such character could be decoded, then `Char` is absent.
    fn previous_char(&self, at: InputAt) -> Char;

    /// Return true if the given empty width instruction matches at the
    /// input position given.
    fn is_empty_match(&self, at: InputAt, empty: &InstEmptyLook) -> bool;

    /// Scan the input for a matching prefix.
    fn prefix_at(
        &self,
        prefixes: &LiteralSearcher,
        at: InputAt,
    ) -> Option<InputAt>;

    /// The number of bytes in the input.
    fn len(&self) -> usize;

    /// Whether the input is empty.
    fn is_empty(&self) -> bool {
        self.len() == 0
    }

    /// Return the given input as a sequence of bytes.
    fn as_bytes(&self) -> &[u8];
}

impl<'a, T: Input> Input for &'a T {
    fn at(&self, i: usize) -> InputAt {
        (**self).at(i)
    }

    fn next_char(&self, at: InputAt) -> Char {
        (**self).next_char(at)
    }

    fn previous_char(&self, at: InputAt) -> Char {
        (**self).previous_char(at)
    }

    fn is_empty_match(&self, at: InputAt, empty: &InstEmptyLook) -> bool {
        (**self).is_empty_match(at, empty)
    }

    fn prefix_at(
        &self,
        prefixes: &LiteralSearcher,
        at: InputAt,
    ) -> Option<InputAt> {
        (**self).prefix_at(prefixes, at)
    }

    fn len(&self) -> usize {
        (**self).len()
    }

    fn as_bytes(&self) -> &[u8] {
        (**self).as_bytes()
    }
}

/// An input reader over characters.
#[derive(Clone, Copy, Debug)]
pub struct CharInput<'t>(&'t [u8]);

impl<'t> CharInput<'t> {
    /// Return a new character input reader for the given string.
    pub fn new(s: &'t [u8]) -> CharInput<'t> {
        CharInput(s)
    }
}

impl<'t> ops::Deref for CharInput<'t> {
    type Target = [u8];

    fn deref(&self) -> &[u8] {
        self.0
    }
}

impl<'t> Input for CharInput<'t> {
    fn at(&self, i: usize) -> InputAt {
        if i >= self.len() {
            InputAt { pos: self.len(), c: None.into(), byte: None, len: 0 }
        } else {
            let c = decode_utf8(&self[i..]).map(|(c, _)| c).into();
            InputAt { pos: i, c, byte: None, len: c.len_utf8() }
        }
    }

    fn next_char(&self, at: InputAt) -> Char {
        at.char()
    }

    fn previous_char(&self, at: InputAt) -> Char {
        decode_last_utf8(&self[..at.pos()]).map(|(c, _)| c).into()
    }

    fn is_empty_match(&self, at: InputAt, empty: &InstEmptyLook) -> bool {
        use crate::prog::EmptyLook::*;
        match empty.look {
            StartLine => {
                let c = self.previous_char(at);
                at.pos() == 0 || c == '\n'
            }
            EndLine => {
                let c = self.next_char(at);
                at.pos() == self.len() || c == '\n'
            }
            StartText => at.pos() == 0,
            EndText => at.pos() == self.len(),
            WordBoundary => {
                let (c1, c2) = (self.previous_char(at), self.next_char(at));
                c1.is_word_char() != c2.is_word_char()
            }
            NotWordBoundary => {
                let (c1, c2) = (self.previous_char(at), self.next_char(at));
                c1.is_word_char() == c2.is_word_char()
            }
            WordBoundaryAscii => {
                let (c1, c2) = (self.previous_char(at), self.next_char(at));
                c1.is_word_byte() != c2.is_word_byte()
            }
            NotWordBoundaryAscii => {
                let (c1, c2) = (self.previous_char(at), self.next_char(at));
                c1.is_word_byte() == c2.is_word_byte()
            }
        }
    }

    fn prefix_at(
        &self,
        prefixes: &LiteralSearcher,
        at: InputAt,
    ) -> Option<InputAt> {
        prefixes.find(&self[at.pos()..]).map(|(s, _)| self.at(at.pos() + s))
    }

    fn len(&self) -> usize {
        self.0.len()
    }

    fn as_bytes(&self) -> &[u8] {
        self.0
    }
}

/// An input reader over bytes.
#[derive(Clone, Copy, Debug)]
pub struct ByteInput<'t> {
    text: &'t [u8],
    only_utf8: bool,
}

impl<'t> ByteInput<'t> {
    /// Return a new byte-based input reader for the given string.
    pub fn new(text: &'t [u8], only_utf8: bool) -> ByteInput<'t> {
        ByteInput { text, only_utf8 }
    }
}

impl<'t> ops::Deref for ByteInput<'t> {
    type Target = [u8];

    fn deref(&self) -> &[u8] {
        self.text
    }
}

impl<'t> Input for ByteInput<'t> {
    fn at(&self, i: usize) -> InputAt {
        if i >= self.len() {
            InputAt { pos: self.len(), c: None.into(), byte: None, len: 0 }
        } else {
            InputAt {
                pos: i,
                c: None.into(),
                byte: self.get(i).cloned(),
                len: 1,
            }
        }
    }

    fn next_char(&self, at: InputAt) -> Char {
        decode_utf8(&self[at.pos()..]).map(|(c, _)| c).into()
    }

    fn previous_char(&self, at: InputAt) -> Char {
        decode_last_utf8(&self[..at.pos()]).map(|(c, _)| c).into()
    }

    fn is_empty_match(&self, at: InputAt, empty: &InstEmptyLook) -> bool {
        use crate::prog::EmptyLook::*;
        match empty.look {
            StartLine => {
                let c = self.previous_char(at);
                at.pos() == 0 || c == '\n'
            }
            EndLine => {
                let c = self.next_char(at);
                at.pos() == self.len() || c == '\n'
            }
            StartText => at.pos() == 0,
            EndText => at.pos() == self.len(),
            WordBoundary => {
                let (c1, c2) = (self.previous_char(at), self.next_char(at));
                c1.is_word_char() != c2.is_word_char()
            }
            NotWordBoundary => {
                let (c1, c2) = (self.previous_char(at), self.next_char(at));
                c1.is_word_char() == c2.is_word_char()
            }
            WordBoundaryAscii => {
                let (c1, c2) = (self.previous_char(at), self.next_char(at));
                if self.only_utf8 {
                    // If we must match UTF-8, then we can't match word
                    // boundaries at invalid UTF-8.
                    if c1.is_none() && !at.is_start() {
                        return false;
                    }
                    if c2.is_none() && !at.is_end() {
                        return false;
                    }
                }
                c1.is_word_byte() != c2.is_word_byte()
            }
            NotWordBoundaryAscii => {
                let (c1, c2) = (self.previous_char(at), self.next_char(at));
                if self.only_utf8 {
                    // If we must match UTF-8, then we can't match word
                    // boundaries at invalid UTF-8.
                    if c1.is_none() && !at.is_start() {
                        return false;
                    }
                    if c2.is_none() && !at.is_end() {
                        return false;
                    }
                }
                c1.is_word_byte() == c2.is_word_byte()
            }
        }
    }

    fn prefix_at(
        &self,
        prefixes: &LiteralSearcher,
        at: InputAt,
    ) -> Option<InputAt> {
        prefixes.find(&self[at.pos()..]).map(|(s, _)| self.at(at.pos() + s))
    }

    fn len(&self) -> usize {
        self.text.len()
    }

    fn as_bytes(&self) -> &[u8] {
        self.text
    }
}

/// An inline representation of `Option<char>`.
///
/// This eliminates the need to do case analysis on `Option<char>` to determine
/// ordinality with other characters.
///
/// (The `Option<char>` is not related to encoding. Instead, it is used in the
/// matching engines to represent the beginning and ending boundaries of the
/// search text.)
#[derive(Clone, Copy, Hash, PartialEq, Eq, PartialOrd, Ord)]
pub struct Char(u32);

impl fmt::Debug for Char {
    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
        match char::from_u32(self.0) {
            None => write!(f, "Empty"),
            Some(c) => write!(f, "{:?}", c),
        }
    }
}

impl Char {
    /// Returns true iff the character is absent.
    #[inline]
    pub fn is_none(self) -> bool {
        self.0 == u32::MAX
    }

    /// Returns the length of the character's UTF-8 encoding.
    ///
    /// If the character is absent, then `1` is returned.
    #[inline]
    pub fn len_utf8(self) -> usize {
        char::from_u32(self.0).map_or(1, |c| c.len_utf8())
    }

    /// Returns true iff the character is a word character.
    ///
    /// If the character is absent, then false is returned.
    pub fn is_word_char(self) -> bool {
        // is_word_character can panic if the Unicode data for \w isn't
        // available. However, our compiler ensures that if a Unicode word
        // boundary is used, then the data must also be available. If it isn't,
        // then the compiler returns an error.
        char::from_u32(self.0).map_or(false, regex_syntax::is_word_character)
    }

    /// Returns true iff the byte is a word byte.
    ///
    /// If the byte is absent, then false is returned.
    pub fn is_word_byte(self) -> bool {
        match char::from_u32(self.0) {
            Some(c) if c <= '\u{7F}' => regex_syntax::is_word_byte(c as u8),
            None | Some(_) => false,
        }
    }
}

impl From<char> for Char {
    fn from(c: char) -> Char {
        Char(c as u32)
    }
}

impl From<Option<char>> for Char {
    fn from(c: Option<char>) -> Char {
        c.map_or(Char(u32::MAX), |c| c.into())
    }
}

impl PartialEq<char> for Char {
    #[inline]
    fn eq(&self, other: &char) -> bool {
        self.0 == *other as u32
    }
}

impl PartialEq<Char> for char {
    #[inline]
    fn eq(&self, other: &Char) -> bool {
        *self as u32 == other.0
    }
}

impl PartialOrd<char> for Char {
    #[inline]
    fn partial_cmp(&self, other: &char) -> Option<Ordering> {
        self.0.partial_cmp(&(*other as u32))
    }
}

impl PartialOrd<Char> for char {
    #[inline]
    fn partial_cmp(&self, other: &Char) -> Option<Ordering> {
        (*self as u32).partial_cmp(&other.0)
    }
}