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
//! Low-level Rust lexer.
//!
//! The idea with `rustc_lexer` is to make a reusable library,
//! by separating out pure lexing and rustc-specific concerns, like spans,
//! error reporting, and interning.  So, rustc_lexer operates directly on `&str`,
//! produces simple tokens which are a pair of type-tag and a bit of original text,
//! and does not report errors, instead storing them as flags on the token.
//!
//! Tokens produced by this lexer are not yet ready for parsing the Rust syntax.
//! For that see [`rustc_parse::lexer`], which converts this basic token stream
//! into wide tokens used by actual parser.
//!
//! The purpose of this crate is to convert raw sources into a labeled sequence
//! of well-known token types, so building an actual Rust token stream will
//! be easier.
//!
//! The main entity of this crate is the [`TokenKind`] enum which represents common
//! lexeme types.
//!
//! [`rustc_parse::lexer`]: ../rustc_parse/lexer/index.html
// We want to be able to build this crate with a stable compiler, so no
// `#![feature]` attributes should be added.

mod cursor;
pub mod unescape;

#[cfg(test)]
mod tests;

use self::LiteralKind::*;
use self::TokenKind::*;
use crate::cursor::{Cursor, EOF_CHAR};
use std::convert::TryFrom;

/// Parsed token.
/// It doesn't contain information about data that has been parsed,
/// only the type of the token and its size.
#[derive(Debug)]
pub struct Token {
    pub kind: TokenKind,
    pub len: usize,
}

impl Token {
    fn new(kind: TokenKind, len: usize) -> Token {
        Token { kind, len }
    }
}

/// Enum representing common lexeme types.
// perf note: Changing all `usize` to `u32` doesn't change performance. See #77629
#[derive(Clone, Copy, Debug, PartialEq, Eq, PartialOrd, Ord)]
pub enum TokenKind {
    // Multi-char tokens:
    /// "// comment"
    LineComment { doc_style: Option<DocStyle> },
    /// `/* block comment */`
    ///
    /// Block comments can be recursive, so the sequence like `/* /* */`
    /// will not be considered terminated and will result in a parsing error.
    BlockComment { doc_style: Option<DocStyle>, terminated: bool },
    /// Any whitespace characters sequence.
    Whitespace,
    /// "ident" or "continue"
    /// At this step keywords are also considered identifiers.
    Ident,
    /// Like the above, but containing invalid unicode codepoints.
    InvalidIdent,
    /// "r#ident"
    RawIdent,
    /// An unknown prefix like `foo#`, `foo'`, `foo"`. Note that only the
    /// prefix (`foo`) is included in the token, not the separator (which is
    /// lexed as its own distinct token). In Rust 2021 and later, reserved
    /// prefixes are reported as errors; in earlier editions, they result in a
    /// (allowed by default) lint, and are treated as regular identifier
    /// tokens.
    UnknownPrefix,
    /// "12_u8", "1.0e-40", "b"123"". See `LiteralKind` for more details.
    Literal { kind: LiteralKind, suffix_start: usize },
    /// "'a"
    Lifetime { starts_with_number: bool },

    // One-char tokens:
    /// ";"
    Semi,
    /// ","
    Comma,
    /// "."
    Dot,
    /// "("
    OpenParen,
    /// ")"
    CloseParen,
    /// "{"
    OpenBrace,
    /// "}"
    CloseBrace,
    /// "["
    OpenBracket,
    /// "]"
    CloseBracket,
    /// "@"
    At,
    /// "#"
    Pound,
    /// "~"
    Tilde,
    /// "?"
    Question,
    /// ":"
    Colon,
    /// "$"
    Dollar,
    /// "="
    Eq,
    /// "!"
    Bang,
    /// "<"
    Lt,
    /// ">"
    Gt,
    /// "-"
    Minus,
    /// "&"
    And,
    /// "|"
    Or,
    /// "+"
    Plus,
    /// "*"
    Star,
    /// "/"
    Slash,
    /// "^"
    Caret,
    /// "%"
    Percent,

    /// Unknown token, not expected by the lexer, e.g. "№"
    Unknown,
}

#[derive(Clone, Copy, Debug, PartialEq, Eq, PartialOrd, Ord)]
pub enum DocStyle {
    Outer,
    Inner,
}

#[derive(Clone, Copy, Debug, PartialEq, Eq, PartialOrd, Ord)]
pub enum LiteralKind {
    /// "12_u8", "0o100", "0b120i99"
    Int { base: Base, empty_int: bool },
    /// "12.34f32", "0b100.100"
    Float { base: Base, empty_exponent: bool },
    /// "'a'", "'\\'", "'''", "';"
    Char { terminated: bool },
    /// "b'a'", "b'\\'", "b'''", "b';"
    Byte { terminated: bool },
    /// ""abc"", ""abc"
    Str { terminated: bool },
    /// "b"abc"", "b"abc"
    ByteStr { terminated: bool },
    /// "r"abc"", "r#"abc"#", "r####"ab"###"c"####", "r#"a"
    RawStr { n_hashes: u8, err: Option<RawStrError> },
    /// "br"abc"", "br#"abc"#", "br####"ab"###"c"####", "br#"a"
    RawByteStr { n_hashes: u8, err: Option<RawStrError> },
}

/// Error produced validating a raw string. Represents cases like:
/// - `r##~"abcde"##`: `InvalidStarter`
/// - `r###"abcde"##`: `NoTerminator { expected: 3, found: 2, possible_terminator_offset: Some(11)`
/// - Too many `#`s (>255): `TooManyDelimiters`
// perf note: It doesn't matter that this makes `Token` 36 bytes bigger. See #77629
#[derive(Clone, Copy, Debug, PartialEq, Eq, PartialOrd, Ord)]
pub enum RawStrError {
    /// Non `#` characters exist between `r` and `"` eg. `r#~"..`
    InvalidStarter { bad_char: char },
    /// The string was never terminated. `possible_terminator_offset` is the number of characters after `r` or `br` where they
    /// may have intended to terminate it.
    NoTerminator { expected: usize, found: usize, possible_terminator_offset: Option<usize> },
    /// More than 255 `#`s exist.
    TooManyDelimiters { found: usize },
}

/// Base of numeric literal encoding according to its prefix.
#[derive(Clone, Copy, Debug, PartialEq, Eq, PartialOrd, Ord)]
pub enum Base {
    /// Literal starts with "0b".
    Binary,
    /// Literal starts with "0o".
    Octal,
    /// Literal starts with "0x".
    Hexadecimal,
    /// Literal doesn't contain a prefix.
    Decimal,
}

/// `rustc` allows files to have a shebang, e.g. "#!/usr/bin/rustrun",
/// but shebang isn't a part of rust syntax.
pub fn strip_shebang(input: &str) -> Option<usize> {
    // Shebang must start with `#!` literally, without any preceding whitespace.
    // For simplicity we consider any line starting with `#!` a shebang,
    // regardless of restrictions put on shebangs by specific platforms.
    if let Some(input_tail) = input.strip_prefix("#!") {
        // Ok, this is a shebang but if the next non-whitespace token is `[`,
        // then it may be valid Rust code, so consider it Rust code.
        let next_non_whitespace_token = tokenize(input_tail).map(|tok| tok.kind).find(|tok| {
            !matches!(
                tok,
                TokenKind::Whitespace
                    | TokenKind::LineComment { doc_style: None }
                    | TokenKind::BlockComment { doc_style: None, .. }
            )
        });
        if next_non_whitespace_token != Some(TokenKind::OpenBracket) {
            // No other choice than to consider this a shebang.
            return Some(2 + input_tail.lines().next().unwrap_or_default().len());
        }
    }
    None
}

/// Parses the first token from the provided input string.
pub fn first_token(input: &str) -> Token {
    debug_assert!(!input.is_empty());
    Cursor::new(input).advance_token()
}

/// Creates an iterator that produces tokens from the input string.
pub fn tokenize(input: &str) -> impl Iterator<Item = Token> + '_ {
    let mut cursor = Cursor::new(input);
    std::iter::from_fn(move || {
        if cursor.is_eof() {
            None
        } else {
            cursor.reset_len_consumed();
            Some(cursor.advance_token())
        }
    })
}

/// True if `c` is considered a whitespace according to Rust language definition.
/// See [Rust language reference](https://doc.rust-lang.org/reference/whitespace.html)
/// for definitions of these classes.
pub fn is_whitespace(c: char) -> bool {
    // This is Pattern_White_Space.
    //
    // Note that this set is stable (ie, it doesn't change with different
    // Unicode versions), so it's ok to just hard-code the values.

    matches!(
        c,
        // Usual ASCII suspects
        '\u{0009}'   // \t
        | '\u{000A}' // \n
        | '\u{000B}' // vertical tab
        | '\u{000C}' // form feed
        | '\u{000D}' // \r
        | '\u{0020}' // space

        // NEXT LINE from latin1
        | '\u{0085}'

        // Bidi markers
        | '\u{200E}' // LEFT-TO-RIGHT MARK
        | '\u{200F}' // RIGHT-TO-LEFT MARK

        // Dedicated whitespace characters from Unicode
        | '\u{2028}' // LINE SEPARATOR
        | '\u{2029}' // PARAGRAPH SEPARATOR
    )
}

/// True if `c` is valid as a first character of an identifier.
/// See [Rust language reference](https://doc.rust-lang.org/reference/identifiers.html) for
/// a formal definition of valid identifier name.
pub fn is_id_start(c: char) -> bool {
    // This is XID_Start OR '_' (which formally is not a XID_Start).
    c == '_' || unicode_xid::UnicodeXID::is_xid_start(c)
}

/// True if `c` is valid as a non-first character of an identifier.
/// See [Rust language reference](https://doc.rust-lang.org/reference/identifiers.html) for
/// a formal definition of valid identifier name.
pub fn is_id_continue(c: char) -> bool {
    unicode_xid::UnicodeXID::is_xid_continue(c)
}

/// The passed string is lexically an identifier.
pub fn is_ident(string: &str) -> bool {
    let mut chars = string.chars();
    if let Some(start) = chars.next() {
        is_id_start(start) && chars.all(is_id_continue)
    } else {
        false
    }
}

impl Cursor<'_> {
    /// Parses a token from the input string.
    fn advance_token(&mut self) -> Token {
        let first_char = self.bump().unwrap();
        let token_kind = match first_char {
            // Slash, comment or block comment.
            '/' => match self.first() {
                '/' => self.line_comment(),
                '*' => self.block_comment(),
                _ => Slash,
            },

            // Whitespace sequence.
            c if is_whitespace(c) => self.whitespace(),

            // Raw identifier, raw string literal or identifier.
            'r' => match (self.first(), self.second()) {
                ('#', c1) if is_id_start(c1) => self.raw_ident(),
                ('#', _) | ('"', _) => {
                    let (n_hashes, err) = self.raw_double_quoted_string(1);
                    let suffix_start = self.len_consumed();
                    if err.is_none() {
                        self.eat_literal_suffix();
                    }
                    let kind = RawStr { n_hashes, err };
                    Literal { kind, suffix_start }
                }
                _ => self.ident_or_unknown_prefix(),
            },

            // Byte literal, byte string literal, raw byte string literal or identifier.
            'b' => match (self.first(), self.second()) {
                ('\'', _) => {
                    self.bump();
                    let terminated = self.single_quoted_string();
                    let suffix_start = self.len_consumed();
                    if terminated {
                        self.eat_literal_suffix();
                    }
                    let kind = Byte { terminated };
                    Literal { kind, suffix_start }
                }
                ('"', _) => {
                    self.bump();
                    let terminated = self.double_quoted_string();
                    let suffix_start = self.len_consumed();
                    if terminated {
                        self.eat_literal_suffix();
                    }
                    let kind = ByteStr { terminated };
                    Literal { kind, suffix_start }
                }
                ('r', '"') | ('r', '#') => {
                    self.bump();
                    let (n_hashes, err) = self.raw_double_quoted_string(2);
                    let suffix_start = self.len_consumed();
                    if err.is_none() {
                        self.eat_literal_suffix();
                    }
                    let kind = RawByteStr { n_hashes, err };
                    Literal { kind, suffix_start }
                }
                _ => self.ident_or_unknown_prefix(),
            },

            // Identifier (this should be checked after other variant that can
            // start as identifier).
            c if is_id_start(c) => self.ident_or_unknown_prefix(),

            // Numeric literal.
            c @ '0'..='9' => {
                let literal_kind = self.number(c);
                let suffix_start = self.len_consumed();
                self.eat_literal_suffix();
                TokenKind::Literal { kind: literal_kind, suffix_start }
            }

            // One-symbol tokens.
            ';' => Semi,
            ',' => Comma,
            '.' => Dot,
            '(' => OpenParen,
            ')' => CloseParen,
            '{' => OpenBrace,
            '}' => CloseBrace,
            '[' => OpenBracket,
            ']' => CloseBracket,
            '@' => At,
            '#' => Pound,
            '~' => Tilde,
            '?' => Question,
            ':' => Colon,
            '$' => Dollar,
            '=' => Eq,
            '!' => Bang,
            '<' => Lt,
            '>' => Gt,
            '-' => Minus,
            '&' => And,
            '|' => Or,
            '+' => Plus,
            '*' => Star,
            '^' => Caret,
            '%' => Percent,

            // Lifetime or character literal.
            '\'' => self.lifetime_or_char(),

            // String literal.
            '"' => {
                let terminated = self.double_quoted_string();
                let suffix_start = self.len_consumed();
                if terminated {
                    self.eat_literal_suffix();
                }
                let kind = Str { terminated };
                Literal { kind, suffix_start }
            }
            // Identifier starting with an emoji. Only lexed for graceful error recovery.
            c if !c.is_ascii() && unic_emoji_char::is_emoji(c) => {
                self.fake_ident_or_unknown_prefix()
            }
            _ => Unknown,
        };
        Token::new(token_kind, self.len_consumed())
    }

    fn line_comment(&mut self) -> TokenKind {
        debug_assert!(self.prev() == '/' && self.first() == '/');
        self.bump();

        let doc_style = match self.first() {
            // `//!` is an inner line doc comment.
            '!' => Some(DocStyle::Inner),
            // `////` (more than 3 slashes) is not considered a doc comment.
            '/' if self.second() != '/' => Some(DocStyle::Outer),
            _ => None,
        };

        self.eat_while(|c| c != '\n');
        LineComment { doc_style }
    }

    fn block_comment(&mut self) -> TokenKind {
        debug_assert!(self.prev() == '/' && self.first() == '*');
        self.bump();

        let doc_style = match self.first() {
            // `/*!` is an inner block doc comment.
            '!' => Some(DocStyle::Inner),
            // `/***` (more than 2 stars) is not considered a doc comment.
            // `/**/` is not considered a doc comment.
            '*' if !matches!(self.second(), '*' | '/') => Some(DocStyle::Outer),
            _ => None,
        };

        let mut depth = 1usize;
        while let Some(c) = self.bump() {
            match c {
                '/' if self.first() == '*' => {
                    self.bump();
                    depth += 1;
                }
                '*' if self.first() == '/' => {
                    self.bump();
                    depth -= 1;
                    if depth == 0 {
                        // This block comment is closed, so for a construction like "/* */ */"
                        // there will be a successfully parsed block comment "/* */"
                        // and " */" will be processed separately.
                        break;
                    }
                }
                _ => (),
            }
        }

        BlockComment { doc_style, terminated: depth == 0 }
    }

    fn whitespace(&mut self) -> TokenKind {
        debug_assert!(is_whitespace(self.prev()));
        self.eat_while(is_whitespace);
        Whitespace
    }

    fn raw_ident(&mut self) -> TokenKind {
        debug_assert!(self.prev() == 'r' && self.first() == '#' && is_id_start(self.second()));
        // Eat "#" symbol.
        self.bump();
        // Eat the identifier part of RawIdent.
        self.eat_identifier();
        RawIdent
    }

    fn ident_or_unknown_prefix(&mut self) -> TokenKind {
        debug_assert!(is_id_start(self.prev()));
        // Start is already eaten, eat the rest of identifier.
        self.eat_while(is_id_continue);
        // Known prefixes must have been handled earlier. So if
        // we see a prefix here, it is definitely an unknown prefix.
        match self.first() {
            '#' | '"' | '\'' => UnknownPrefix,
            c if !c.is_ascii() && unic_emoji_char::is_emoji(c) => {
                self.fake_ident_or_unknown_prefix()
            }
            _ => Ident,
        }
    }

    fn fake_ident_or_unknown_prefix(&mut self) -> TokenKind {
        // Start is already eaten, eat the rest of identifier.
        self.eat_while(|c| {
            unicode_xid::UnicodeXID::is_xid_continue(c)
                || (!c.is_ascii() && unic_emoji_char::is_emoji(c))
                || c == '\u{200d}'
        });
        // Known prefixes must have been handled earlier. So if
        // we see a prefix here, it is definitely an unknown prefix.
        match self.first() {
            '#' | '"' | '\'' => UnknownPrefix,
            _ => InvalidIdent,
        }
    }

    fn number(&mut self, first_digit: char) -> LiteralKind {
        debug_assert!('0' <= self.prev() && self.prev() <= '9');
        let mut base = Base::Decimal;
        if first_digit == '0' {
            // Attempt to parse encoding base.
            let has_digits = match self.first() {
                'b' => {
                    base = Base::Binary;
                    self.bump();
                    self.eat_decimal_digits()
                }
                'o' => {
                    base = Base::Octal;
                    self.bump();
                    self.eat_decimal_digits()
                }
                'x' => {
                    base = Base::Hexadecimal;
                    self.bump();
                    self.eat_hexadecimal_digits()
                }
                // Not a base prefix.
                '0'..='9' | '_' | '.' | 'e' | 'E' => {
                    self.eat_decimal_digits();
                    true
                }
                // Just a 0.
                _ => return Int { base, empty_int: false },
            };
            // Base prefix was provided, but there were no digits
            // after it, e.g. "0x".
            if !has_digits {
                return Int { base, empty_int: true };
            }
        } else {
            // No base prefix, parse number in the usual way.
            self.eat_decimal_digits();
        };

        match self.first() {
            // Don't be greedy if this is actually an
            // integer literal followed by field/method access or a range pattern
            // (`0..2` and `12.foo()`)
            '.' if self.second() != '.' && !is_id_start(self.second()) => {
                // might have stuff after the ., and if it does, it needs to start
                // with a number
                self.bump();
                let mut empty_exponent = false;
                if self.first().is_digit(10) {
                    self.eat_decimal_digits();
                    match self.first() {
                        'e' | 'E' => {
                            self.bump();
                            empty_exponent = !self.eat_float_exponent();
                        }
                        _ => (),
                    }
                }
                Float { base, empty_exponent }
            }
            'e' | 'E' => {
                self.bump();
                let empty_exponent = !self.eat_float_exponent();
                Float { base, empty_exponent }
            }
            _ => Int { base, empty_int: false },
        }
    }

    fn lifetime_or_char(&mut self) -> TokenKind {
        debug_assert!(self.prev() == '\'');

        let can_be_a_lifetime = if self.second() == '\'' {
            // It's surely not a lifetime.
            false
        } else {
            // If the first symbol is valid for identifier, it can be a lifetime.
            // Also check if it's a number for a better error reporting (so '0 will
            // be reported as invalid lifetime and not as unterminated char literal).
            is_id_start(self.first()) || self.first().is_digit(10)
        };

        if !can_be_a_lifetime {
            let terminated = self.single_quoted_string();
            let suffix_start = self.len_consumed();
            if terminated {
                self.eat_literal_suffix();
            }
            let kind = Char { terminated };
            return Literal { kind, suffix_start };
        }

        // Either a lifetime or a character literal with
        // length greater than 1.

        let starts_with_number = self.first().is_digit(10);

        // Skip the literal contents.
        // First symbol can be a number (which isn't a valid identifier start),
        // so skip it without any checks.
        self.bump();
        self.eat_while(is_id_continue);

        // Check if after skipping literal contents we've met a closing
        // single quote (which means that user attempted to create a
        // string with single quotes).
        if self.first() == '\'' {
            self.bump();
            let kind = Char { terminated: true };
            Literal { kind, suffix_start: self.len_consumed() }
        } else {
            Lifetime { starts_with_number }
        }
    }

    fn single_quoted_string(&mut self) -> bool {
        debug_assert!(self.prev() == '\'');
        // Check if it's a one-symbol literal.
        if self.second() == '\'' && self.first() != '\\' {
            self.bump();
            self.bump();
            return true;
        }

        // Literal has more than one symbol.

        // Parse until either quotes are terminated or error is detected.
        loop {
            match self.first() {
                // Quotes are terminated, finish parsing.
                '\'' => {
                    self.bump();
                    return true;
                }
                // Probably beginning of the comment, which we don't want to include
                // to the error report.
                '/' => break,
                // Newline without following '\'' means unclosed quote, stop parsing.
                '\n' if self.second() != '\'' => break,
                // End of file, stop parsing.
                EOF_CHAR if self.is_eof() => break,
                // Escaped slash is considered one character, so bump twice.
                '\\' => {
                    self.bump();
                    self.bump();
                }
                // Skip the character.
                _ => {
                    self.bump();
                }
            }
        }
        // String was not terminated.
        false
    }

    /// Eats double-quoted string and returns true
    /// if string is terminated.
    fn double_quoted_string(&mut self) -> bool {
        debug_assert!(self.prev() == '"');
        while let Some(c) = self.bump() {
            match c {
                '"' => {
                    return true;
                }
                '\\' if self.first() == '\\' || self.first() == '"' => {
                    // Bump again to skip escaped character.
                    self.bump();
                }
                _ => (),
            }
        }
        // End of file reached.
        false
    }

    /// Eats the double-quoted string and returns `n_hashes` and an error if encountered.
    fn raw_double_quoted_string(&mut self, prefix_len: usize) -> (u8, Option<RawStrError>) {
        // Wrap the actual function to handle the error with too many hashes.
        // This way, it eats the whole raw string.
        let (n_hashes, err) = self.raw_string_unvalidated(prefix_len);
        // Only up to 255 `#`s are allowed in raw strings
        match u8::try_from(n_hashes) {
            Ok(num) => (num, err),
            // We lie about the number of hashes here :P
            Err(_) => (0, Some(RawStrError::TooManyDelimiters { found: n_hashes })),
        }
    }

    fn raw_string_unvalidated(&mut self, prefix_len: usize) -> (usize, Option<RawStrError>) {
        debug_assert!(self.prev() == 'r');
        let start_pos = self.len_consumed();
        let mut possible_terminator_offset = None;
        let mut max_hashes = 0;

        // Count opening '#' symbols.
        let mut eaten = 0;
        while self.first() == '#' {
            eaten += 1;
            self.bump();
        }
        let n_start_hashes = eaten;

        // Check that string is started.
        match self.bump() {
            Some('"') => (),
            c => {
                let c = c.unwrap_or(EOF_CHAR);
                return (n_start_hashes, Some(RawStrError::InvalidStarter { bad_char: c }));
            }
        }

        // Skip the string contents and on each '#' character met, check if this is
        // a raw string termination.
        loop {
            self.eat_while(|c| c != '"');

            if self.is_eof() {
                return (
                    n_start_hashes,
                    Some(RawStrError::NoTerminator {
                        expected: n_start_hashes,
                        found: max_hashes,
                        possible_terminator_offset,
                    }),
                );
            }

            // Eat closing double quote.
            self.bump();

            // Check that amount of closing '#' symbols
            // is equal to the amount of opening ones.
            // Note that this will not consume extra trailing `#` characters:
            // `r###"abcde"####` is lexed as a `RawStr { n_hashes: 3 }`
            // followed by a `#` token.
            let mut n_end_hashes = 0;
            while self.first() == '#' && n_end_hashes < n_start_hashes {
                n_end_hashes += 1;
                self.bump();
            }

            if n_end_hashes == n_start_hashes {
                return (n_start_hashes, None);
            } else if n_end_hashes > max_hashes {
                // Keep track of possible terminators to give a hint about
                // where there might be a missing terminator
                possible_terminator_offset =
                    Some(self.len_consumed() - start_pos - n_end_hashes + prefix_len);
                max_hashes = n_end_hashes;
            }
        }
    }

    fn eat_decimal_digits(&mut self) -> bool {
        let mut has_digits = false;
        loop {
            match self.first() {
                '_' => {
                    self.bump();
                }
                '0'..='9' => {
                    has_digits = true;
                    self.bump();
                }
                _ => break,
            }
        }
        has_digits
    }

    fn eat_hexadecimal_digits(&mut self) -> bool {
        let mut has_digits = false;
        loop {
            match self.first() {
                '_' => {
                    self.bump();
                }
                '0'..='9' | 'a'..='f' | 'A'..='F' => {
                    has_digits = true;
                    self.bump();
                }
                _ => break,
            }
        }
        has_digits
    }

    /// Eats the float exponent. Returns true if at least one digit was met,
    /// and returns false otherwise.
    fn eat_float_exponent(&mut self) -> bool {
        debug_assert!(self.prev() == 'e' || self.prev() == 'E');
        if self.first() == '-' || self.first() == '+' {
            self.bump();
        }
        self.eat_decimal_digits()
    }

    // Eats the suffix of the literal, e.g. "_u8".
    fn eat_literal_suffix(&mut self) {
        self.eat_identifier();
    }

    // Eats the identifier.
    fn eat_identifier(&mut self) {
        if !is_id_start(self.first()) {
            return;
        }
        self.bump();

        self.eat_while(is_id_continue);
    }
}