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
use std::{cell::RefCell, mem, mem::take, rc::Rc};

use debug_unreachable::debug_unreachable;
use lexer::TokenContexts;
use swc_common::{BytePos, Span};

use super::Parser;
use crate::{
    error::Error,
    lexer::{self},
    token::*,
    Context, EsVersion, Syntax,
};

/// Clone should be cheap if you are parsing typescript because typescript
/// syntax requires backtracking.
pub trait Tokens: Clone + Iterator<Item = TokenAndSpan> {
    fn set_ctx(&mut self, ctx: Context);
    fn ctx(&self) -> Context;
    fn syntax(&self) -> Syntax;
    fn target(&self) -> EsVersion;

    fn start_pos(&self) -> BytePos {
        BytePos(0)
    }

    fn set_expr_allowed(&mut self, allow: bool);
    fn set_next_regexp(&mut self, start: Option<BytePos>);

    fn token_context(&self) -> &lexer::TokenContexts;
    fn token_context_mut(&mut self) -> &mut lexer::TokenContexts;
    fn set_token_context(&mut self, _c: lexer::TokenContexts);

    /// Implementors should use Rc<RefCell<Vec<Error>>>.
    ///
    /// It is required because parser should backtrack while parsing typescript
    /// code.
    fn add_error(&self, error: Error);

    /// Add an error which is valid syntax in script mode.
    ///
    /// This errors should be dropped if it's not a module.
    ///
    /// Implementor should check for if [Context].module, and buffer errors if
    /// module is false. Also, implementors should move errors to the error
    /// buffer on set_ctx if the parser mode become module mode.
    fn add_module_mode_error(&self, error: Error);

    fn take_errors(&mut self) -> Vec<Error>;
}

#[derive(Clone)]
pub struct TokensInput {
    iter: <Vec<TokenAndSpan> as IntoIterator>::IntoIter,
    ctx: Context,
    syntax: Syntax,
    start_pos: BytePos,
    target: EsVersion,
    token_ctx: TokenContexts,
    errors: Rc<RefCell<Vec<Error>>>,
    module_errors: Rc<RefCell<Vec<Error>>>,
}

impl TokensInput {
    pub fn new(tokens: Vec<TokenAndSpan>, ctx: Context, syntax: Syntax, target: EsVersion) -> Self {
        let start_pos = tokens.first().map(|t| t.span.lo).unwrap_or(BytePos(0));

        TokensInput {
            iter: tokens.into_iter(),
            ctx,
            syntax,
            start_pos,
            target,
            token_ctx: Default::default(),
            errors: Default::default(),
            module_errors: Default::default(),
        }
    }
}

impl Iterator for TokensInput {
    type Item = TokenAndSpan;

    fn next(&mut self) -> Option<Self::Item> {
        self.iter.next()
    }
}

impl Tokens for TokensInput {
    fn set_ctx(&mut self, ctx: Context) {
        if ctx.module && !self.module_errors.borrow().is_empty() {
            let mut module_errors = self.module_errors.borrow_mut();
            self.errors.borrow_mut().append(&mut *module_errors);
        }
        self.ctx = ctx;
    }

    fn ctx(&self) -> Context {
        self.ctx
    }

    fn syntax(&self) -> Syntax {
        self.syntax
    }

    fn target(&self) -> EsVersion {
        self.target
    }

    fn start_pos(&self) -> BytePos {
        self.start_pos
    }

    fn set_expr_allowed(&mut self, _: bool) {}

    fn set_next_regexp(&mut self, _: Option<BytePos>) {}

    fn token_context(&self) -> &TokenContexts {
        &self.token_ctx
    }

    fn token_context_mut(&mut self) -> &mut TokenContexts {
        &mut self.token_ctx
    }

    fn set_token_context(&mut self, c: TokenContexts) {
        self.token_ctx = c;
    }

    fn add_error(&self, error: Error) {
        self.errors.borrow_mut().push(error);
    }

    fn add_module_mode_error(&self, error: Error) {
        if self.ctx.module {
            self.add_error(error);
            return;
        }
        self.module_errors.borrow_mut().push(error);
    }

    fn take_errors(&mut self) -> Vec<Error> {
        take(&mut self.errors.borrow_mut())
    }
}

/// Note: Lexer need access to parser's context to lex correctly.
#[derive(Debug)]
pub struct Capturing<I: Tokens> {
    inner: I,
    captured: Rc<RefCell<Vec<TokenAndSpan>>>,
}

impl<I: Tokens> Clone for Capturing<I> {
    fn clone(&self) -> Self {
        Capturing {
            inner: self.inner.clone(),
            captured: self.captured.clone(),
        }
    }
}

impl<I: Tokens> Capturing<I> {
    pub fn new(input: I) -> Self {
        Capturing {
            inner: input,
            captured: Default::default(),
        }
    }

    /// Take captured tokens
    pub fn take(&mut self) -> Vec<TokenAndSpan> {
        mem::take(&mut *self.captured.borrow_mut())
    }
}

impl<I: Tokens> Iterator for Capturing<I> {
    type Item = TokenAndSpan;

    fn next(&mut self) -> Option<Self::Item> {
        let next = self.inner.next();

        match next {
            Some(ts) => {
                let mut v = self.captured.borrow_mut();

                // remove tokens that could change due to backtracing
                while let Some(last) = v.last() {
                    if last.span.lo >= ts.span.lo {
                        v.pop();
                    } else {
                        break;
                    }
                }

                v.push(ts.clone());

                Some(ts)
            }
            None => None,
        }
    }
}

impl<I: Tokens> Tokens for Capturing<I> {
    fn set_ctx(&mut self, ctx: Context) {
        self.inner.set_ctx(ctx)
    }

    fn ctx(&self) -> Context {
        self.inner.ctx()
    }

    fn syntax(&self) -> Syntax {
        self.inner.syntax()
    }

    fn target(&self) -> EsVersion {
        self.inner.target()
    }

    fn start_pos(&self) -> BytePos {
        self.inner.start_pos()
    }

    fn set_expr_allowed(&mut self, allow: bool) {
        self.inner.set_expr_allowed(allow)
    }

    fn set_next_regexp(&mut self, start: Option<BytePos>) {
        self.inner.set_next_regexp(start);
    }

    fn token_context(&self) -> &TokenContexts {
        self.inner.token_context()
    }

    fn token_context_mut(&mut self) -> &mut TokenContexts {
        self.inner.token_context_mut()
    }

    fn set_token_context(&mut self, c: TokenContexts) {
        self.inner.set_token_context(c)
    }

    fn add_error(&self, error: Error) {
        self.inner.add_error(error);
    }

    fn add_module_mode_error(&self, error: Error) {
        self.inner.add_module_mode_error(error)
    }

    fn take_errors(&mut self) -> Vec<Error> {
        self.inner.take_errors()
    }
}

/// This struct is responsible for managing current token and peeked token.
#[derive(Clone)]
pub(super) struct Buffer<I: Tokens> {
    iter: I,
    /// Span of the previous token.
    prev_span: Span,
    cur: Option<TokenAndSpan>,
    /// Peeked token
    next: Option<TokenAndSpan>,
}

impl<I: Tokens> Parser<I> {
    pub fn input(&mut self) -> &mut I {
        &mut self.input.iter
    }

    pub(crate) fn input_ref(&self) -> &I {
        &self.input.iter
    }
}

impl<I: Tokens> Buffer<I> {
    pub fn new(lexer: I) -> Self {
        let start_pos = lexer.start_pos();
        Buffer {
            iter: lexer,
            cur: None,
            prev_span: Span::new(start_pos, start_pos, Default::default()),
            next: None,
        }
    }

    pub fn store(&mut self, token: Token) {
        debug_assert!(self.next.is_none());
        debug_assert!(self.cur.is_none());
        let span = self.prev_span;

        self.cur = Some(TokenAndSpan {
            span,
            token,
            had_line_break: false,
        });
    }

    #[allow(dead_code)]
    pub fn cur_debug(&self) -> Option<&Token> {
        self.cur.as_ref().map(|it| &it.token)
    }

    #[cold]
    #[inline(never)]
    pub fn dump_cur(&mut self) -> String {
        match self.cur() {
            Some(v) => format!("{:?}", v),
            None => "<eof>".to_string(),
        }
    }

    /// Returns current token.
    pub fn bump(&mut self) -> Token {
        let prev = match self.cur.take() {
            Some(t) => t,
            None => unsafe {
                debug_unreachable!(
                    "Current token is `None`. Parser should not call bump() without knowing \
                     current token"
                )
            },
        };
        self.prev_span = prev.span;

        prev.token
    }

    pub fn knows_cur(&self) -> bool {
        self.cur.is_some()
    }

    pub fn peek(&mut self) -> Option<&Token> {
        debug_assert!(
            self.cur.is_some(),
            "parser should not call peek() without knowing current token"
        );

        if self.next.is_none() {
            self.next = self.iter.next();
        }

        self.next.as_ref().map(|ts| &ts.token)
    }

    /// Returns true on eof.
    pub fn had_line_break_before_cur(&mut self) -> bool {
        self.cur();

        self.cur
            .as_ref()
            .map(|it| it.had_line_break)
            .unwrap_or_else(|| true)
    }

    /// This returns true on eof.
    pub fn has_linebreak_between_cur_and_peeked(&mut self) -> bool {
        let _ = self.peek();
        self.next
            .as_ref()
            .map(|item| item.had_line_break)
            .unwrap_or({
                // return true on eof.
                true
            })
    }

    /// Get current token. Returns `None` only on eof.
    #[inline]
    pub fn cur(&mut self) -> Option<&Token> {
        if self.cur.is_none() {
            // If we have peeked a token, take it instead of calling lexer.next()
            self.cur = self.next.take().or_else(|| self.iter.next());
        }

        match &self.cur {
            Some(v) => Some(&v.token),
            None => None,
        }
    }

    #[inline]
    pub fn cut_lshift(&mut self) {
        debug_assert!(
            self.is(&tok!("<<")),
            "parser should only call cut_lshift when encountering LShift token"
        );
        self.cur = Some(TokenAndSpan {
            token: tok!('<'),
            span: self.cur_span().with_lo(self.cur_span().lo + BytePos(1)),
            had_line_break: false,
        });
    }

    #[inline]
    pub fn is(&mut self, expected: &Token) -> bool {
        match self.cur() {
            Some(t) => *expected == *t,
            _ => false,
        }
    }

    #[inline]
    pub fn eat(&mut self, expected: &Token) -> bool {
        let v = self.is(expected);
        if v {
            self.bump();
        }
        v
    }

    /// Returns start of current token.
    #[inline]
    pub fn cur_pos(&mut self) -> BytePos {
        let _ = self.cur();
        self.cur
            .as_ref()
            .map(|item| item.span.lo)
            .unwrap_or_else(|| {
                // eof
                self.last_pos()
            })
    }

    #[inline]
    pub fn cur_span(&self) -> Span {
        let data = self
            .cur
            .as_ref()
            .map(|item| item.span)
            .unwrap_or(self.prev_span);

        Span::new(data.lo, data.hi, data.ctxt)
    }

    /// Returns last byte position of previous token.
    #[inline]
    pub fn last_pos(&self) -> BytePos {
        self.prev_span.hi
    }

    /// Returns span of the previous token.
    #[inline]
    pub fn prev_span(&self) -> Span {
        self.prev_span
    }

    #[inline]
    pub(crate) fn get_ctx(&self) -> Context {
        self.iter.ctx()
    }

    #[inline]
    pub(crate) fn set_ctx(&mut self, ctx: Context) {
        self.iter.set_ctx(ctx);
    }

    #[inline]
    pub fn syntax(&self) -> Syntax {
        self.iter.syntax()
    }

    #[inline]
    pub fn target(&self) -> EsVersion {
        self.iter.target()
    }

    #[inline]
    pub(crate) fn set_expr_allowed(&mut self, allow: bool) {
        self.iter.set_expr_allowed(allow)
    }

    #[inline]
    pub fn set_next_regexp(&mut self, start: Option<BytePos>) {
        self.iter.set_next_regexp(start);
    }

    #[inline]
    pub(crate) fn token_context(&self) -> &lexer::TokenContexts {
        self.iter.token_context()
    }

    #[inline]
    pub(crate) fn token_context_mut(&mut self) -> &mut lexer::TokenContexts {
        self.iter.token_context_mut()
    }

    #[inline]
    pub(crate) fn set_token_context(&mut self, c: lexer::TokenContexts) {
        self.iter.set_token_context(c)
    }
}