-
-
Notifications
You must be signed in to change notification settings - Fork 384
/
cursor.rs
511 lines (456 loc) · 16.2 KB
/
cursor.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
//! Module implementing the lexer cursor. This is used for managing the input byte stream.
use crate::{profiler::BoaProfiler, syntax::ast::Position};
use std::io::{self, Bytes, Error, ErrorKind, Read};
/// Cursor over the source code.
#[derive(Debug)]
pub(super) struct Cursor<R> {
iter: InnerIter<R>,
pos: Position,
strict_mode: bool,
}
impl<R> Cursor<R> {
/// Gets the current position of the cursor in the source code.
#[inline]
pub(super) fn pos(&self) -> Position {
self.pos
}
/// Advances the position to the next column.
#[inline]
pub(super) fn next_column(&mut self) {
let current_line = self.pos.line_number();
let next_column = self.pos.column_number() + 1;
self.pos = Position::new(current_line, next_column);
}
/// Advances the position to the next line.
#[inline]
fn next_line(&mut self) {
let next_line = self.pos.line_number() + 1;
self.pos = Position::new(next_line, 1);
}
#[inline]
pub(super) fn strict_mode(&self) -> bool {
self.strict_mode
}
#[inline]
pub(super) fn set_strict_mode(&mut self, strict_mode: bool) {
self.strict_mode = strict_mode
}
}
impl<R> Cursor<R>
where
R: Read,
{
/// Creates a new Lexer cursor.
#[inline]
pub(super) fn new(inner: R) -> Self {
Self {
iter: InnerIter::new(inner.bytes()),
pos: Position::new(1, 1),
strict_mode: false,
}
}
/// Peeks the next byte.
#[inline]
pub(super) fn peek(&mut self) -> Result<Option<u8>, Error> {
let _timer = BoaProfiler::global().start_event("cursor::peek()", "Lexing");
self.iter.peek_byte()
}
/// Peeks the next n bytes, the maximum number of peeked bytes is 4 (n <= 4).
#[inline]
pub(super) fn peek_n(&mut self, n: u8) -> Result<u32, Error> {
let _timer = BoaProfiler::global().start_event("cursor::peek_n()", "Lexing");
self.iter.peek_n_bytes(n)
}
/// Peeks the next UTF-8 character in u32 code point.
#[inline]
pub(super) fn peek_char(&mut self) -> Result<Option<u32>, Error> {
let _timer = BoaProfiler::global().start_event("cursor::peek_char()", "Lexing");
self.iter.peek_char()
}
/// Compares the byte passed in to the next byte, if they match true is returned and the buffer is incremented
#[inline]
pub(super) fn next_is(&mut self, byte: u8) -> io::Result<bool> {
let _timer = BoaProfiler::global().start_event("cursor::next_is()", "Lexing");
Ok(match self.peek()? {
Some(next) if next == byte => {
let _ = self.next_byte()?;
true
}
_ => false,
})
}
/// Applies the predicate to the next character and returns the result.
/// Returns false if the next character is not a valid ascii or there is no next character.
/// Otherwise returns the result from the predicate on the ascii in char
///
/// The buffer is not incremented.
#[inline]
pub(super) fn next_is_ascii_pred<F>(&mut self, pred: &F) -> io::Result<bool>
where
F: Fn(char) -> bool,
{
let _timer = BoaProfiler::global().start_event("cursor::next_is_ascii_pred()", "Lexing");
Ok(match self.peek()? {
Some(byte) => match byte {
0..=0x7F => pred(char::from(byte)),
_ => false,
},
None => false,
})
}
/// Applies the predicate to the next UTF-8 character and returns the result.
/// Returns false if there is no next character, otherwise returns the result from the
/// predicate on the ascii char
///
/// The buffer is not incremented.
#[inline]
pub(super) fn next_is_char_pred<F>(&mut self, pred: &F) -> io::Result<bool>
where
F: Fn(u32) -> bool,
{
let _timer = BoaProfiler::global().start_event("cursor::next_is_char_pred()", "Lexing");
Ok(if let Some(peek) = self.peek_char()? {
pred(peek)
} else {
false
})
}
/// Fills the buffer with all bytes until the stop byte is found.
/// Returns error when reaching the end of the buffer.
///
/// Note that all bytes up until the stop byte are added to the buffer, including the byte right before.
pub(super) fn take_until(&mut self, stop: u8, buf: &mut Vec<u8>) -> io::Result<()> {
let _timer = BoaProfiler::global().start_event("cursor::take_until()", "Lexing");
loop {
if self.next_is(stop)? {
return Ok(());
} else if let Some(byte) = self.next_byte()? {
buf.push(byte);
} else {
return Err(io::Error::new(
ErrorKind::UnexpectedEof,
format!("Unexpected end of file when looking for character {}", stop),
));
}
}
}
/// Fills the buffer with characters until the first ascii character for which the predicate (pred) is false.
/// It also stops when the next character is not an ascii or there is no next character.
///
/// Note that all characters up until the stop character are added to the buffer, including the character right before.
pub(super) fn take_while_ascii_pred<F>(&mut self, buf: &mut Vec<u8>, pred: &F) -> io::Result<()>
where
F: Fn(char) -> bool,
{
let _timer = BoaProfiler::global().start_event("cursor::take_while_ascii_pred()", "Lexing");
loop {
if !self.next_is_ascii_pred(pred)? {
return Ok(());
} else if let Some(byte) = self.next_byte()? {
buf.push(byte);
} else {
// next_is_pred will return false if the next value is None so the None case should already be handled.
unreachable!();
}
}
}
/// Fills the buffer with characters until the first character for which the predicate (pred) is false.
/// It also stops when there is no next character.
///
/// Note that all characters up until the stop character are added to the buffer, including the character right before.
pub(super) fn take_while_char_pred<F>(&mut self, buf: &mut Vec<u8>, pred: &F) -> io::Result<()>
where
F: Fn(u32) -> bool,
{
let _timer = BoaProfiler::global().start_event("cursor::take_while_char_pred()", "Lexing");
loop {
if !self.next_is_char_pred(pred)? {
return Ok(());
} else if let Some(ch) = self.peek_char()? {
for _ in 0..utf8_len(ch) {
buf.push(self.next_byte()?.unwrap());
}
} else {
// next_is_pred will return false if the next value is None so the None case should already be handled.
unreachable!();
}
}
}
/// It will fill the buffer with bytes.
///
/// This expects for the buffer to be fully filled. If it's not, it will fail with an
/// `UnexpectedEof` I/O error.
#[inline]
pub(super) fn fill_bytes(&mut self, buf: &mut [u8]) -> io::Result<()> {
let _timer = BoaProfiler::global().start_event("cursor::fill_bytes()", "Lexing");
self.iter.fill_bytes(buf)
}
/// Retrieves the next byte.
#[inline]
pub(crate) fn next_byte(&mut self) -> Result<Option<u8>, Error> {
let _timer = BoaProfiler::global().start_event("cursor::next_byte()", "Lexing");
let byte = self.iter.next_byte()?;
match byte {
Some(b'\r') => {
// Try to take a newline if it's next, for windows "\r\n" newlines
// Otherwise, treat as a Mac OS9 bare '\r' newline
if self.peek()? == Some(b'\n') {
let _ = self.iter.next_byte();
}
self.next_line();
}
Some(b'\n') => self.next_line(),
Some(0xE2) => {
// Try to match '\u{2028}' (e2 80 a8) and '\u{2029}' (e2 80 a9)
let next_bytes = self.peek_n(2)?;
if next_bytes == 0xA8_80 || next_bytes == 0xA9_80 {
self.next_line();
} else {
// 0xE2 is a utf8 first byte
self.next_column();
}
}
Some(b) if utf8_is_first_byte(b) => self.next_column(),
_ => {}
}
Ok(byte)
}
/// Retrieves the next UTF-8 character.
#[inline]
pub(crate) fn next_char(&mut self) -> Result<Option<u32>, Error> {
let _timer = BoaProfiler::global().start_event("cursor::next_char()", "Lexing");
let ch = self.iter.next_char()?;
match ch {
Some(0xD) => {
// Try to take a newline if it's next, for windows "\r\n" newlines
// Otherwise, treat as a Mac OS9 bare '\r' newline
if self.peek()? == Some(0xA) {
let _ = self.iter.next_byte();
}
self.next_line();
}
// '\n' | '\u{2028}' | '\u{2029}'
Some(0xA) | Some(0x2028) | Some(0x2029) => self.next_line(),
Some(_) => self.next_column(),
_ => {}
}
Ok(ch)
}
}
/// Inner iterator for a cursor.
#[derive(Debug)]
struct InnerIter<R> {
iter: Bytes<R>,
num_peeked_bytes: u8,
peeked_bytes: u32,
peeked_char: Option<Option<u32>>,
}
impl<R> InnerIter<R> {
/// Creates a new inner iterator.
#[inline]
fn new(iter: Bytes<R>) -> Self {
Self {
iter,
num_peeked_bytes: 0,
peeked_bytes: 0,
peeked_char: None,
}
}
}
impl<R> InnerIter<R>
where
R: Read,
{
/// It will fill the buffer with checked ascii bytes.
///
/// This expects for the buffer to be fully filled. If it's not, it will fail with an
/// `UnexpectedEof` I/O error.
#[inline]
fn fill_bytes(&mut self, buf: &mut [u8]) -> io::Result<()> {
for byte in buf.iter_mut() {
*byte = self.next_byte()?.ok_or_else(|| {
io::Error::new(
io::ErrorKind::UnexpectedEof,
"unexpected EOF when filling buffer",
)
})?;
}
Ok(())
}
/// Increments the iter by n bytes.
#[inline]
fn increment(&mut self, n: u32) -> Result<(), Error> {
for _ in 0..n {
if None == self.next_byte()? {
break;
}
}
Ok(())
}
/// Peeks the next byte.
#[inline]
pub(super) fn peek_byte(&mut self) -> Result<Option<u8>, Error> {
if self.num_peeked_bytes > 0 {
let byte = self.peeked_bytes as u8;
Ok(Some(byte))
} else {
match self.iter.next().transpose()? {
Some(byte) => {
self.num_peeked_bytes = 1;
self.peeked_bytes = byte as u32;
Ok(Some(byte))
}
None => Ok(None),
}
}
}
/// Peeks the next n bytes, the maximum number of peeked bytes is 4 (n <= 4).
#[inline]
pub(super) fn peek_n_bytes(&mut self, n: u8) -> Result<u32, Error> {
while self.num_peeked_bytes < n && self.num_peeked_bytes < 4 {
match self.iter.next().transpose()? {
Some(byte) => {
self.peeked_bytes |= (byte as u32) << (self.num_peeked_bytes * 8);
self.num_peeked_bytes += 1;
}
None => break,
};
}
match n {
0 => Ok(0),
1 => Ok(self.peeked_bytes & 0xFF),
2 => Ok(self.peeked_bytes & 0xFFFF),
3 => Ok(self.peeked_bytes & 0xFFFFFF),
_ => Ok(self.peeked_bytes),
}
}
/// Peeks the next unchecked character in u32 code point.
#[inline]
pub(super) fn peek_char(&mut self) -> Result<Option<u32>, Error> {
if let Some(ch) = self.peeked_char {
Ok(ch)
} else {
// Decode UTF-8
let x = match self.peek_byte()? {
Some(b) if b < 128 => {
self.peeked_char = Some(Some(b as u32));
return Ok(Some(b as u32));
}
Some(b) => b,
None => {
self.peeked_char = None;
return Ok(None);
}
};
// Multibyte case follows
// Decode from a byte combination out of: [[[x y] z] w]
// NOTE: Performance is sensitive to the exact formulation here
let init = utf8_first_byte(x, 2);
let y = (self.peek_n_bytes(2)? >> 8) as u8;
let mut ch = utf8_acc_cont_byte(init, y);
if x >= 0xE0 {
// [[x y z] w] case
// 5th bit in 0xE0 .. 0xEF is always clear, so `init` is still valid
let z = (self.peek_n_bytes(3)? >> 16) as u8;
let y_z = utf8_acc_cont_byte((y & CONT_MASK) as u32, z);
ch = init << 12 | y_z;
if x >= 0xF0 {
// [x y z w] case
// use only the lower 3 bits of `init`
let w = (self.peek_n_bytes(4)? >> 24) as u8;
ch = (init & 7) << 18 | utf8_acc_cont_byte(y_z, w);
}
};
self.peeked_char = Some(Some(ch));
Ok(Some(ch))
}
}
/// Retrieves the next byte
#[inline]
fn next_byte(&mut self) -> io::Result<Option<u8>> {
self.peeked_char = None;
if self.num_peeked_bytes > 0 {
let byte = (self.peeked_bytes & 0xFF) as u8;
self.num_peeked_bytes -= 1;
self.peeked_bytes >>= 8;
Ok(Some(byte))
} else {
self.iter.next().transpose()
}
}
/// Retrieves the next unchecked char in u32 code point.
#[inline]
fn next_char(&mut self) -> io::Result<Option<u32>> {
if let Some(ch) = self.peeked_char.take() {
if let Some(c) = ch {
self.increment(utf8_len(c))?;
}
return Ok(ch);
}
// Decode UTF-8
let x = match self.next_byte()? {
Some(b) if b < 128 => return Ok(Some(b as u32)),
Some(b) => b,
None => return Ok(None),
};
// Multibyte case follows
// Decode from a byte combination out of: [[[x y] z] w]
// NOTE: Performance is sensitive to the exact formulation here
let init = utf8_first_byte(x, 2);
let y = unwrap_or_0(self.next_byte()?);
let mut ch = utf8_acc_cont_byte(init, y);
if x >= 0xE0 {
// [[x y z] w] case
// 5th bit in 0xE0 .. 0xEF is always clear, so `init` is still valid
let z = unwrap_or_0(self.next_byte()?);
let y_z = utf8_acc_cont_byte((y & CONT_MASK) as u32, z);
ch = init << 12 | y_z;
if x >= 0xF0 {
// [x y z w] case
// use only the lower 3 bits of `init`
let w = unwrap_or_0(self.next_byte()?);
ch = (init & 7) << 18 | utf8_acc_cont_byte(y_z, w);
}
};
Ok(Some(ch))
}
}
/// Mask of the value bits of a continuation byte.
const CONT_MASK: u8 = 0b0011_1111;
/// Returns the initial codepoint accumulator for the first byte.
/// The first byte is special, only want bottom 5 bits for width 2, 4 bits
/// for width 3, and 3 bits for width 4.
#[inline]
fn utf8_first_byte(byte: u8, width: u32) -> u32 {
(byte & (0x7F >> width)) as u32
}
/// Returns the value of `ch` updated with continuation byte `byte`.
#[inline]
fn utf8_acc_cont_byte(ch: u32, byte: u8) -> u32 {
(ch << 6) | (byte & CONT_MASK) as u32
}
/// Checks whether the byte is a UTF-8 first byte (i.e., ascii byte or starts with the
/// bits `11`).
#[inline]
fn utf8_is_first_byte(byte: u8) -> bool {
byte <= 0x7F || (byte >> 6) == 0x11
}
#[inline]
fn unwrap_or_0(opt: Option<u8>) -> u8 {
match opt {
Some(byte) => byte,
None => 0,
}
}
#[inline]
fn utf8_len(ch: u32) -> u32 {
if ch <= 0x7F {
1
} else if ch <= 0x7FF {
2
} else if ch <= 0xFFFF {
3
} else {
4
}
}