syntax/ast/
token_ext.rs

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
//! There are many AstNodes, but only a few tokens, so we hand-write them here.

use std::{borrow::Cow, num::ParseIntError};

use rustc_lexer::unescape::{
    unescape_byte, unescape_char, unescape_mixed, unescape_unicode, EscapeError, MixedUnit, Mode,
};
use stdx::always;

use crate::{
    ast::{self, AstToken},
    TextRange, TextSize,
};

impl ast::Comment {
    pub fn kind(&self) -> CommentKind {
        CommentKind::from_text(self.text())
    }

    pub fn is_doc(&self) -> bool {
        self.kind().doc.is_some()
    }

    pub fn is_inner(&self) -> bool {
        self.kind().doc == Some(CommentPlacement::Inner)
    }

    pub fn is_outer(&self) -> bool {
        self.kind().doc == Some(CommentPlacement::Outer)
    }

    pub fn prefix(&self) -> &'static str {
        let &(prefix, _kind) = CommentKind::BY_PREFIX
            .iter()
            .find(|&(prefix, kind)| self.kind() == *kind && self.text().starts_with(prefix))
            .unwrap();
        prefix
    }

    /// Returns the textual content of a doc comment node as a single string with prefix and suffix
    /// removed.
    pub fn doc_comment(&self) -> Option<&str> {
        let kind = self.kind();
        match kind {
            CommentKind { shape, doc: Some(_) } => {
                let prefix = kind.prefix();
                let text = &self.text()[prefix.len()..];
                let text = if shape == CommentShape::Block {
                    text.strip_suffix("*/").unwrap_or(text)
                } else {
                    text
                };
                Some(text)
            }
            _ => None,
        }
    }
}

#[derive(Debug, PartialEq, Eq, Clone, Copy)]
pub struct CommentKind {
    pub shape: CommentShape,
    pub doc: Option<CommentPlacement>,
}

#[derive(Debug, PartialEq, Eq, Clone, Copy)]
pub enum CommentShape {
    Line,
    Block,
}

impl CommentShape {
    pub fn is_line(self) -> bool {
        self == CommentShape::Line
    }

    pub fn is_block(self) -> bool {
        self == CommentShape::Block
    }
}

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

impl CommentKind {
    const BY_PREFIX: [(&'static str, CommentKind); 9] = [
        ("/**/", CommentKind { shape: CommentShape::Block, doc: None }),
        ("/***", CommentKind { shape: CommentShape::Block, doc: None }),
        ("////", CommentKind { shape: CommentShape::Line, doc: None }),
        ("///", CommentKind { shape: CommentShape::Line, doc: Some(CommentPlacement::Outer) }),
        ("//!", CommentKind { shape: CommentShape::Line, doc: Some(CommentPlacement::Inner) }),
        ("/**", CommentKind { shape: CommentShape::Block, doc: Some(CommentPlacement::Outer) }),
        ("/*!", CommentKind { shape: CommentShape::Block, doc: Some(CommentPlacement::Inner) }),
        ("//", CommentKind { shape: CommentShape::Line, doc: None }),
        ("/*", CommentKind { shape: CommentShape::Block, doc: None }),
    ];

    pub(crate) fn from_text(text: &str) -> CommentKind {
        let &(_prefix, kind) = CommentKind::BY_PREFIX
            .iter()
            .find(|&(prefix, _kind)| text.starts_with(prefix))
            .unwrap();
        kind
    }

    pub fn prefix(&self) -> &'static str {
        let &(prefix, _) =
            CommentKind::BY_PREFIX.iter().rev().find(|(_, kind)| kind == self).unwrap();
        prefix
    }
}

impl ast::Whitespace {
    pub fn spans_multiple_lines(&self) -> bool {
        let text = self.text();
        text.find('\n').map_or(false, |idx| text[idx + 1..].contains('\n'))
    }
}

#[derive(Debug)]
pub struct QuoteOffsets {
    pub quotes: (TextRange, TextRange),
    pub contents: TextRange,
}

impl QuoteOffsets {
    fn new(literal: &str) -> Option<QuoteOffsets> {
        let left_quote = literal.find('"')?;
        let right_quote = literal.rfind('"')?;
        if left_quote == right_quote {
            // `literal` only contains one quote
            return None;
        }

        let start = TextSize::from(0);
        let left_quote = TextSize::try_from(left_quote).unwrap() + TextSize::of('"');
        let right_quote = TextSize::try_from(right_quote).unwrap();
        let end = TextSize::of(literal);

        let res = QuoteOffsets {
            quotes: (TextRange::new(start, left_quote), TextRange::new(right_quote, end)),
            contents: TextRange::new(left_quote, right_quote),
        };
        Some(res)
    }
}

pub trait IsString: AstToken {
    const RAW_PREFIX: &'static str;
    const MODE: Mode;
    fn is_raw(&self) -> bool {
        self.text().starts_with(Self::RAW_PREFIX)
    }
    fn quote_offsets(&self) -> Option<QuoteOffsets> {
        let text = self.text();
        let offsets = QuoteOffsets::new(text)?;
        let o = self.syntax().text_range().start();
        let offsets = QuoteOffsets {
            quotes: (offsets.quotes.0 + o, offsets.quotes.1 + o),
            contents: offsets.contents + o,
        };
        Some(offsets)
    }
    fn text_range_between_quotes(&self) -> Option<TextRange> {
        self.quote_offsets().map(|it| it.contents)
    }
    fn text_without_quotes(&self) -> &str {
        let text = self.text();
        let Some(offsets) = self.text_range_between_quotes() else { return text };
        &text[offsets - self.syntax().text_range().start()]
    }
    fn open_quote_text_range(&self) -> Option<TextRange> {
        self.quote_offsets().map(|it| it.quotes.0)
    }
    fn close_quote_text_range(&self) -> Option<TextRange> {
        self.quote_offsets().map(|it| it.quotes.1)
    }
    fn escaped_char_ranges(&self, cb: &mut dyn FnMut(TextRange, Result<char, EscapeError>)) {
        let Some(text_range_no_quotes) = self.text_range_between_quotes() else { return };

        let start = self.syntax().text_range().start();
        let text = &self.text()[text_range_no_quotes - start];
        let offset = text_range_no_quotes.start() - start;

        unescape_unicode(text, Self::MODE, &mut |range, unescaped_char| {
            if let Some((s, e)) = range.start.try_into().ok().zip(range.end.try_into().ok()) {
                cb(TextRange::new(s, e) + offset, unescaped_char);
            }
        });
    }
    fn map_range_up(&self, range: TextRange) -> Option<TextRange> {
        let contents_range = self.text_range_between_quotes()?;
        if always!(TextRange::up_to(contents_range.len()).contains_range(range)) {
            Some(range + contents_range.start())
        } else {
            None
        }
    }
}

impl IsString for ast::String {
    const RAW_PREFIX: &'static str = "r";
    const MODE: Mode = Mode::Str;
}

impl ast::String {
    pub fn value(&self) -> Result<Cow<'_, str>, EscapeError> {
        let text = self.text();
        let text_range = self.text_range_between_quotes().ok_or(EscapeError::LoneSlash)?;
        let text = &text[text_range - self.syntax().text_range().start()];
        if self.is_raw() {
            return Ok(Cow::Borrowed(text));
        }

        let mut buf = String::new();
        let mut prev_end = 0;
        let mut has_error = None;
        unescape_unicode(text, Self::MODE, &mut |char_range, unescaped_char| match (
            unescaped_char,
            buf.capacity() == 0,
        ) {
            (Ok(c), false) => buf.push(c),
            (Ok(_), true) if char_range.len() == 1 && char_range.start == prev_end => {
                prev_end = char_range.end
            }
            (Ok(c), true) => {
                buf.reserve_exact(text.len());
                buf.push_str(&text[..prev_end]);
                buf.push(c);
            }
            (Err(e), _) => has_error = Some(e),
        });

        match (has_error, buf.capacity() == 0) {
            (Some(e), _) => Err(e),
            (None, true) => Ok(Cow::Borrowed(text)),
            (None, false) => Ok(Cow::Owned(buf)),
        }
    }
}

impl IsString for ast::ByteString {
    const RAW_PREFIX: &'static str = "br";
    const MODE: Mode = Mode::ByteStr;
}

impl ast::ByteString {
    pub fn value(&self) -> Result<Cow<'_, [u8]>, EscapeError> {
        let text = self.text();
        let text_range = self.text_range_between_quotes().ok_or(EscapeError::LoneSlash)?;
        let text = &text[text_range - self.syntax().text_range().start()];
        if self.is_raw() {
            return Ok(Cow::Borrowed(text.as_bytes()));
        }

        let mut buf: Vec<u8> = Vec::new();
        let mut prev_end = 0;
        let mut has_error = None;
        unescape_unicode(text, Self::MODE, &mut |char_range, unescaped_char| match (
            unescaped_char,
            buf.capacity() == 0,
        ) {
            (Ok(c), false) => buf.push(c as u8),
            (Ok(_), true) if char_range.len() == 1 && char_range.start == prev_end => {
                prev_end = char_range.end
            }
            (Ok(c), true) => {
                buf.reserve_exact(text.len());
                buf.extend_from_slice(text[..prev_end].as_bytes());
                buf.push(c as u8);
            }
            (Err(e), _) => has_error = Some(e),
        });

        match (has_error, buf.capacity() == 0) {
            (Some(e), _) => Err(e),
            (None, true) => Ok(Cow::Borrowed(text.as_bytes())),
            (None, false) => Ok(Cow::Owned(buf)),
        }
    }
}

impl IsString for ast::CString {
    const RAW_PREFIX: &'static str = "cr";
    const MODE: Mode = Mode::CStr;

    fn escaped_char_ranges(&self, cb: &mut dyn FnMut(TextRange, Result<char, EscapeError>)) {
        let text_range_no_quotes = match self.text_range_between_quotes() {
            Some(it) => it,
            None => return,
        };

        let start = self.syntax().text_range().start();
        let text = &self.text()[text_range_no_quotes - start];
        let offset = text_range_no_quotes.start() - start;

        unescape_mixed(text, Self::MODE, &mut |range, unescaped_char| {
            let text_range =
                TextRange::new(range.start.try_into().unwrap(), range.end.try_into().unwrap());
            // XXX: This method should only be used for highlighting ranges. The unescaped
            // char/byte is not used. For simplicity, we return an arbitrary placeholder char.
            cb(text_range + offset, unescaped_char.map(|_| ' '));
        });
    }
}

impl ast::CString {
    pub fn value(&self) -> Result<Cow<'_, [u8]>, EscapeError> {
        let text = self.text();
        let text_range = self.text_range_between_quotes().ok_or(EscapeError::LoneSlash)?;
        let text = &text[text_range - self.syntax().text_range().start()];
        if self.is_raw() {
            return Ok(Cow::Borrowed(text.as_bytes()));
        }

        let mut buf = Vec::new();
        let mut prev_end = 0;
        let mut has_error = None;
        let extend_unit = |buf: &mut Vec<u8>, unit: MixedUnit| match unit {
            MixedUnit::Char(c) => buf.extend(c.encode_utf8(&mut [0; 4]).as_bytes()),
            MixedUnit::HighByte(b) => buf.push(b),
        };
        unescape_mixed(text, Self::MODE, &mut |char_range, unescaped| match (
            unescaped,
            buf.capacity() == 0,
        ) {
            (Ok(u), false) => extend_unit(&mut buf, u),
            (Ok(_), true) if char_range.len() == 1 && char_range.start == prev_end => {
                prev_end = char_range.end
            }
            (Ok(u), true) => {
                buf.reserve_exact(text.len());
                buf.extend(text[..prev_end].as_bytes());
                extend_unit(&mut buf, u);
            }
            (Err(e), _) => has_error = Some(e),
        });

        match (has_error, buf.capacity() == 0) {
            (Some(e), _) => Err(e),
            (None, true) => Ok(Cow::Borrowed(text.as_bytes())),
            (None, false) => Ok(Cow::Owned(buf)),
        }
    }
}

impl ast::IntNumber {
    pub fn radix(&self) -> Radix {
        match self.text().get(..2).unwrap_or_default() {
            "0b" => Radix::Binary,
            "0o" => Radix::Octal,
            "0x" => Radix::Hexadecimal,
            _ => Radix::Decimal,
        }
    }

    pub fn split_into_parts(&self) -> (&str, &str, &str) {
        let radix = self.radix();
        let (prefix, mut text) = self.text().split_at(radix.prefix_len());

        let is_suffix_start: fn(&(usize, char)) -> bool = match radix {
            Radix::Hexadecimal => |(_, c)| matches!(c, 'g'..='z' | 'G'..='Z'),
            _ => |(_, c)| c.is_ascii_alphabetic(),
        };

        let mut suffix = "";
        if let Some((suffix_start, _)) = text.char_indices().find(is_suffix_start) {
            let (text2, suffix2) = text.split_at(suffix_start);
            text = text2;
            suffix = suffix2;
        };

        (prefix, text, suffix)
    }

    pub fn value(&self) -> Result<u128, ParseIntError> {
        let (_, text, _) = self.split_into_parts();
        u128::from_str_radix(&text.replace('_', ""), self.radix() as u32)
    }

    pub fn suffix(&self) -> Option<&str> {
        let (_, _, suffix) = self.split_into_parts();
        if suffix.is_empty() {
            None
        } else {
            Some(suffix)
        }
    }

    pub fn value_string(&self) -> String {
        let (_, text, _) = self.split_into_parts();
        text.replace('_', "")
    }
}

impl ast::FloatNumber {
    pub fn split_into_parts(&self) -> (&str, &str) {
        let text = self.text();
        let mut float_text = self.text();
        let mut suffix = "";
        let mut indices = text.char_indices();
        if let Some((mut suffix_start, c)) = indices.by_ref().find(|(_, c)| c.is_ascii_alphabetic())
        {
            if c == 'e' || c == 'E' {
                if let Some(suffix_start_tuple) = indices.find(|(_, c)| c.is_ascii_alphabetic()) {
                    suffix_start = suffix_start_tuple.0;

                    float_text = &text[..suffix_start];
                    suffix = &text[suffix_start..];
                }
            } else {
                float_text = &text[..suffix_start];
                suffix = &text[suffix_start..];
            }
        }

        (float_text, suffix)
    }

    pub fn suffix(&self) -> Option<&str> {
        let (_, suffix) = self.split_into_parts();
        if suffix.is_empty() {
            None
        } else {
            Some(suffix)
        }
    }

    pub fn value_string(&self) -> String {
        let (text, _) = self.split_into_parts();
        text.replace('_', "")
    }
}

#[derive(Debug, PartialEq, Eq, Copy, Clone)]
pub enum Radix {
    Binary = 2,
    Octal = 8,
    Decimal = 10,
    Hexadecimal = 16,
}

impl Radix {
    pub const ALL: &'static [Radix] =
        &[Radix::Binary, Radix::Octal, Radix::Decimal, Radix::Hexadecimal];

    const fn prefix_len(self) -> usize {
        match self {
            Self::Decimal => 0,
            _ => 2,
        }
    }
}

impl ast::Char {
    pub fn value(&self) -> Result<char, EscapeError> {
        let mut text = self.text();
        if text.starts_with('\'') {
            text = &text[1..];
        } else {
            return Err(EscapeError::ZeroChars);
        }
        if text.ends_with('\'') {
            text = &text[0..text.len() - 1];
        }

        unescape_char(text)
    }
}

impl ast::Byte {
    pub fn value(&self) -> Result<u8, EscapeError> {
        let mut text = self.text();
        if text.starts_with("b\'") {
            text = &text[2..];
        } else {
            return Err(EscapeError::ZeroChars);
        }
        if text.ends_with('\'') {
            text = &text[0..text.len() - 1];
        }

        unescape_byte(text)
    }
}

#[cfg(test)]
mod tests {
    use rustc_apfloat::ieee::Quad as f128;

    use crate::ast::{self, make, FloatNumber, IntNumber};

    fn check_float_suffix<'a>(lit: &str, expected: impl Into<Option<&'a str>>) {
        assert_eq!(FloatNumber { syntax: make::tokens::literal(lit) }.suffix(), expected.into());
    }

    fn check_int_suffix<'a>(lit: &str, expected: impl Into<Option<&'a str>>) {
        assert_eq!(IntNumber { syntax: make::tokens::literal(lit) }.suffix(), expected.into());
    }

    // FIXME(#17451) Use `expected: f128` once `f128` is stabilised.
    fn check_float_value(lit: &str, expected: &str) {
        let expected = Some(expected.parse::<f128>().unwrap());
        assert_eq!(
            FloatNumber { syntax: make::tokens::literal(lit) }.value_string().parse::<f128>().ok(),
            expected
        );
        assert_eq!(
            IntNumber { syntax: make::tokens::literal(lit) }.value_string().parse::<f128>().ok(),
            expected
        );
    }

    fn check_int_value(lit: &str, expected: impl Into<Option<u128>>) {
        assert_eq!(IntNumber { syntax: make::tokens::literal(lit) }.value().ok(), expected.into());
    }

    #[test]
    fn test_float_number_suffix() {
        check_float_suffix("123.0", None);
        check_float_suffix("123f32", "f32");
        check_float_suffix("123.0e", None);
        check_float_suffix("123.0e4", None);
        check_float_suffix("123.0ef16", "f16");
        check_float_suffix("123.0E4f32", "f32");
        check_float_suffix("1_2_3.0_f128", "f128");
    }

    #[test]
    fn test_int_number_suffix() {
        check_int_suffix("123", None);
        check_int_suffix("123i32", "i32");
        check_int_suffix("1_0_1_l_o_l", "l_o_l");
        check_int_suffix("0b11", None);
        check_int_suffix("0o11", None);
        check_int_suffix("0xff", None);
        check_int_suffix("0b11u32", "u32");
        check_int_suffix("0o11u32", "u32");
        check_int_suffix("0xffu32", "u32");
    }

    fn check_string_value<'a>(lit: &str, expected: impl Into<Option<&'a str>>) {
        assert_eq!(
            ast::String { syntax: make::tokens::literal(&format!("\"{lit}\"")) }
                .value()
                .as_deref()
                .ok(),
            expected.into()
        );
    }

    #[test]
    fn test_string_escape() {
        check_string_value(r"foobar", "foobar");
        check_string_value(r"\foobar", None);
        check_string_value(r"\nfoobar", "\nfoobar");
        check_string_value(r"C:\\Windows\\System32\\", "C:\\Windows\\System32\\");
        check_string_value(r"\x61bcde", "abcde");
        check_string_value(
            r"a\
bcde", "abcde",
        );
    }

    fn check_byte_string_value<'a, const N: usize>(
        lit: &str,
        expected: impl Into<Option<&'a [u8; N]>>,
    ) {
        assert_eq!(
            ast::ByteString { syntax: make::tokens::literal(&format!("b\"{lit}\"")) }
                .value()
                .as_deref()
                .ok(),
            expected.into().map(|value| &value[..])
        );
    }

    #[test]
    fn test_byte_string_escape() {
        check_byte_string_value(r"foobar", b"foobar");
        check_byte_string_value(r"\foobar", None::<&[u8; 0]>);
        check_byte_string_value(r"\nfoobar", b"\nfoobar");
        check_byte_string_value(r"C:\\Windows\\System32\\", b"C:\\Windows\\System32\\");
        check_byte_string_value(r"\x61bcde", b"abcde");
        check_byte_string_value(
            r"a\
bcde", b"abcde",
        );
    }

    #[test]
    fn test_value_underscores() {
        check_float_value("1.3_4665449586950493453___6_f128", "1.346654495869504934536");
        check_float_value("1.234567891011121_f64", "1.234567891011121");
        check_float_value("1__0.__0__f32", "10.0");
        check_float_value("3._0_f16", "3.0");
        check_int_value("0b__1_0_", 2);
        check_int_value("1_1_1_1_1_1", 111111);
    }
}