forked from swc-project/swc
-
Notifications
You must be signed in to change notification settings - Fork 4
/
source_map.rs
1813 lines (1572 loc) · 60.7 KB
/
source_map.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
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
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
// Copyright 2012 The Rust Project Developers. See the COPYRIGHT
// file at the top-level directory of this distribution and at
// http://rust-lang.org/COPYRIGHT.
//
// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
// option. This file may not be copied, modified, or distributed
// except according to those terms.
//! The SourceMap tracks all the source code used within a single crate, mapping
//! from integer byte positions to the original source code location. Each bit
//! of source parsed during crate parsing (typically files, in-memory strings,
//! or various bits of macro expansion) cover a continuous range of bytes in the
//! SourceMap and are represented by SourceFiles. Byte positions are stored in
//! `spans` and used pervasively in the compiler. They are absolute positions
//! within the SourceMap, which upon request can be converted to line and column
//! information, source code snippets, etc.
use std::{
cmp, env, fs,
hash::Hash,
io,
path::{Path, PathBuf},
sync::atomic::{AtomicUsize, Ordering::SeqCst},
};
use once_cell::sync::Lazy;
#[cfg(feature = "sourcemap")]
use sourcemap::SourceMapBuilder;
use tracing::debug;
pub use crate::syntax_pos::*;
use crate::{
collections::AHashMap,
errors::SourceMapper,
rustc_data_structures::stable_hasher::StableHasher,
sync::{Lock, LockGuard, Lrc, MappedLockGuard},
};
static CURRENT_DIR: Lazy<Option<PathBuf>> = Lazy::new(|| env::current_dir().ok());
// _____________________________________________________________________________
// SourceFile, MultiByteChar, FileName, FileLines
//
/// An abstraction over the fs operations used by the Parser.
pub trait FileLoader {
/// Query the existence of a file.
fn file_exists(&self, path: &Path) -> bool;
/// Return an absolute path to a file, if possible.
fn abs_path(&self, path: &Path) -> Option<PathBuf>;
/// Read the contents of an UTF-8 file into memory.
fn read_file(&self, path: &Path) -> io::Result<String>;
}
/// A FileLoader that uses std::fs to load real files.
pub struct RealFileLoader;
impl FileLoader for RealFileLoader {
fn file_exists(&self, path: &Path) -> bool {
fs::metadata(path).is_ok()
}
fn abs_path(&self, path: &Path) -> Option<PathBuf> {
if path.is_absolute() {
Some(path.to_path_buf())
} else {
CURRENT_DIR.as_ref().map(|cwd| cwd.join(path))
}
}
fn read_file(&self, path: &Path) -> io::Result<String> {
fs::read_to_string(path)
}
}
// This is a SourceFile identifier that is used to correlate SourceFiles between
// subsequent compilation sessions (which is something we need to do during
// incremental compilation).
#[derive(Copy, Clone, PartialEq, Eq, Hash, Debug)]
pub struct StableSourceFileId(u128);
impl StableSourceFileId {
pub fn new(source_file: &SourceFile) -> StableSourceFileId {
let mut hasher = StableHasher::new();
source_file.name.hash(&mut hasher);
source_file.name_was_remapped.hash(&mut hasher);
source_file.unmapped_path.hash(&mut hasher);
StableSourceFileId(hasher.finish())
}
}
// _____________________________________________________________________________
// SourceMap
//
#[derive(Default)]
pub(super) struct SourceMapFiles {
pub(super) source_files: Vec<Lrc<SourceFile>>,
stable_id_to_source_file: AHashMap<StableSourceFileId, Lrc<SourceFile>>,
}
/// The interner for spans.
///
/// As most spans are simply stored, we store them as interned form.
///
/// - Each ast node only stores pointer to actual data ([BytePos]).
/// - The pointers ([BytePos]) can be converted to file name, line and column
/// using this struct.
///
/// # Note
///
/// This struct should be shared. `swc_common` uses [crate::sync::Lrc], which is
/// [std::rc::Rc] or [std::sync::Arc], depending on the compile option, for this
/// purpose.
///
/// ## Note for bundler authors
///
/// If you are bundling modules, you should share this struct while parsing
/// modules. Otherwise, you have to implement a code generator which accepts
/// multiple [SourceMap].
pub struct SourceMap {
pub(super) files: Lock<SourceMapFiles>,
start_pos: AtomicUsize,
file_loader: Box<dyn FileLoader + Sync + Send>,
// This is used to apply the file path remapping as specified via
// --remap-path-prefix to all SourceFiles allocated within this SourceMap.
path_mapping: FilePathMapping,
/// In case we are in a doctest, replace all file names with the PathBuf,
/// and add the given offsets to the line info
doctest_offset: Option<(FileName, isize)>,
}
impl Default for SourceMap {
fn default() -> Self {
Self::new(FilePathMapping::empty())
}
}
impl SourceMap {
pub fn new(path_mapping: FilePathMapping) -> SourceMap {
SourceMap {
files: Default::default(),
start_pos: AtomicUsize::new(1),
file_loader: Box::new(RealFileLoader),
path_mapping,
doctest_offset: None,
}
}
pub fn with_file_loader(
file_loader: Box<dyn FileLoader + Sync + Send>,
path_mapping: FilePathMapping,
) -> SourceMap {
SourceMap {
files: Default::default(),
start_pos: AtomicUsize::new(1),
file_loader,
path_mapping,
doctest_offset: None,
}
}
pub fn path_mapping(&self) -> &FilePathMapping {
&self.path_mapping
}
pub fn file_exists(&self, path: &Path) -> bool {
self.file_loader.file_exists(path)
}
pub fn load_file(&self, path: &Path) -> io::Result<Lrc<SourceFile>> {
let src = self.file_loader.read_file(path)?;
let filename = path.to_owned().into();
Ok(self.new_source_file(filename, src))
}
pub fn files(&self) -> MappedLockGuard<'_, Vec<Lrc<SourceFile>>> {
LockGuard::map(self.files.borrow(), |files| &mut files.source_files)
}
pub fn source_file_by_stable_id(
&self,
stable_id: StableSourceFileId,
) -> Option<Lrc<SourceFile>> {
self.files
.borrow()
.stable_id_to_source_file
.get(&stable_id)
.cloned()
}
fn next_start_pos(&self, len: usize) -> usize {
// Add one so there is some space between files. This lets us distinguish
// positions in the source_map, even in the presence of zero-length files.
self.start_pos.fetch_add(len + 1, SeqCst)
}
/// Creates a new source_file.
/// This does not ensure that only one SourceFile exists per file name.
pub fn new_source_file(&self, filename: FileName, mut src: String) -> Lrc<SourceFile> {
remove_bom(&mut src);
self.new_source_file_from(filename, Lrc::new(src))
}
/// Creates a new source_file.
/// This does not ensure that only one SourceFile exists per file name.
///
/// `src` should not have UTF8 BOM
pub fn new_source_file_from(&self, filename: FileName, src: Lrc<String>) -> Lrc<SourceFile> {
// The path is used to determine the directory for loading submodules and
// include files, so it must be before remapping.
// Note that filename may not be a valid path, eg it may be `<anon>` etc,
// but this is okay because the directory determined by `path.pop()` will
// be empty, so the working directory will be used.
let unmapped_path = filename.clone();
let (filename, was_remapped) = match filename {
FileName::Real(filename) => {
let (filename, was_remapped) = self.path_mapping.map_prefix(filename);
(FileName::Real(filename), was_remapped)
}
other => (other, false),
};
// We hold lock at here to prevent panic
// If we don't do this, lookup_char_pos and its family **may** panic.
let mut files = self.files.borrow_mut();
let start_pos = self.next_start_pos(src.len());
let source_file = Lrc::new(SourceFile::new_from(
filename,
was_remapped,
unmapped_path,
src,
Pos::from_usize(start_pos),
));
{
files.source_files.push(source_file.clone());
files
.stable_id_to_source_file
.insert(StableSourceFileId::new(&source_file), source_file.clone());
}
source_file
}
pub fn mk_substr_filename(&self, sp: Span) -> String {
let pos = self.lookup_char_pos(sp.lo());
format!(
"<{}:{}:{}>",
pos.file.name,
pos.line,
pos.col.to_usize() + 1
)
}
// If there is a doctest_offset, apply it to the line
pub fn doctest_offset_line(&self, mut orig: usize) -> usize {
if let Some((_, line)) = self.doctest_offset {
if line >= 0 {
orig += line as usize;
} else {
orig -= (-line) as usize;
}
}
orig
}
/// Lookup source information about a BytePos
pub fn lookup_char_pos(&self, pos: BytePos) -> Loc {
let fm = self.lookup_source_file(pos);
self.lookup_char_pos_with(fm, pos)
}
/// Lookup source information about a BytePos
///
///
/// This method exists only for optimization and it's not part of public
/// api.
#[doc(hidden)]
pub fn lookup_char_pos_with(&self, fm: Lrc<SourceFile>, pos: BytePos) -> Loc {
let line_info = self.lookup_line_with(fm, pos);
match line_info {
Ok(SourceFileAndLine { sf: f, line: a }) => {
let chpos = self.bytepos_to_file_charpos_with(&f, pos);
let line = a + 1; // Line numbers start at 1
let linebpos = f.lines[a];
assert!(
pos >= linebpos,
"{}: bpos = {:?}; linebpos = {:?};",
f.name,
pos,
linebpos,
);
let linechpos = self.bytepos_to_file_charpos_with(&f, linebpos);
let col = chpos - linechpos;
let col_display = {
let start_width_idx = f
.non_narrow_chars
.binary_search_by_key(&linebpos, |x| x.pos())
.unwrap_or_else(|x| x);
let end_width_idx = f
.non_narrow_chars
.binary_search_by_key(&pos, |x| x.pos())
.unwrap_or_else(|x| x);
let special_chars = end_width_idx - start_width_idx;
let non_narrow: usize = f.non_narrow_chars[start_width_idx..end_width_idx]
.iter()
.map(|x| x.width())
.sum();
col.0 - special_chars + non_narrow
};
if cfg!(feature = "debug") {
debug!(
"byte pos {:?} is on the line at byte pos {:?}",
pos, linebpos
);
debug!(
"char pos {:?} is on the line at char pos {:?}",
chpos, linechpos
);
debug!("byte is on line: {}", line);
}
// assert!(chpos >= linechpos);
Loc {
file: f,
line,
col,
col_display,
}
}
Err(f) => {
let chpos = self.bytepos_to_file_charpos(pos);
let col_display = {
let end_width_idx = f
.non_narrow_chars
.binary_search_by_key(&pos, |x| x.pos())
.unwrap_or_else(|x| x);
let non_narrow: usize = f.non_narrow_chars[0..end_width_idx]
.iter()
.map(|x| x.width())
.sum();
chpos.0 - end_width_idx + non_narrow
};
Loc {
file: f,
line: 0,
col: chpos,
col_display,
}
}
}
}
/// If the relevant source_file is empty, we don't return a line number.
pub fn lookup_line(&self, pos: BytePos) -> Result<SourceFileAndLine, Lrc<SourceFile>> {
let f = self.lookup_source_file(pos);
self.lookup_line_with(f, pos)
}
/// If the relevant source_file is empty, we don't return a line number.
///
/// This method exists only for optimization and it's not part of public
/// api.
#[doc(hidden)]
pub fn lookup_line_with(
&self,
f: Lrc<SourceFile>,
pos: BytePos,
) -> Result<SourceFileAndLine, Lrc<SourceFile>> {
match f.lookup_line(pos) {
Some(line) => Ok(SourceFileAndLine { sf: f, line }),
None => Err(f),
}
}
pub fn lookup_char_pos_adj(&self, pos: BytePos) -> LocWithOpt {
let loc = self.lookup_char_pos(pos);
LocWithOpt {
filename: loc.file.name.clone(),
line: loc.line,
col: loc.col,
file: Some(loc.file),
}
}
/// Returns `Some(span)`, a union of the lhs and rhs span. The lhs must
/// precede the rhs. If there are gaps between lhs and rhs, the
/// resulting union will cross these gaps. For this to work, the spans
/// have to be:
///
/// * the ctxt of both spans much match
/// * the lhs span needs to end on the same line the rhs span begins
/// * the lhs span must start at or before the rhs span
pub fn merge_spans(&self, sp_lhs: Span, sp_rhs: Span) -> Option<Span> {
// make sure we're at the same expansion id
if sp_lhs.ctxt() != sp_rhs.ctxt() {
return None;
}
let lhs_end = match self.lookup_line(sp_lhs.hi()) {
Ok(x) => x,
Err(_) => return None,
};
let rhs_begin = match self.lookup_line(sp_rhs.lo()) {
Ok(x) => x,
Err(_) => return None,
};
// if we must cross lines to merge, don't merge
if lhs_end.line != rhs_begin.line {
return None;
}
// ensure these follow the expected order and we don't overlap
if (sp_lhs.lo() <= sp_rhs.lo()) && (sp_lhs.hi() <= sp_rhs.lo()) {
Some(sp_lhs.to(sp_rhs))
} else {
None
}
}
pub fn span_to_string(&self, sp: Span) -> String {
if self.files.borrow().source_files.is_empty() && sp.is_dummy() {
return "no-location".to_string();
}
let lo = self.lookup_char_pos_adj(sp.lo());
let hi = self.lookup_char_pos_adj(sp.hi());
format!(
"{}:{}:{}: {}:{}",
lo.filename,
lo.line,
lo.col.to_usize() + 1,
hi.line,
hi.col.to_usize() + 1
)
}
pub fn span_to_filename(&self, sp: Span) -> FileName {
self.lookup_char_pos(sp.lo()).file.name.clone()
}
pub fn span_to_unmapped_path(&self, sp: Span) -> FileName {
self.lookup_char_pos(sp.lo())
.file
.unmapped_path
.clone()
.expect("SourceMap::span_to_unmapped_path called for imported SourceFile?")
}
pub fn is_multiline(&self, sp: Span) -> bool {
let lo = self.lookup_char_pos(sp.lo());
let hi = self.lookup_char_pos(sp.hi());
lo.line != hi.line
}
pub fn span_to_lines(&self, sp: Span) -> FileLinesResult {
if cfg!(feature = "debug") {
debug!("span_to_lines(sp={:?})", sp);
}
if sp.lo() > sp.hi() {
return Err(Box::new(SpanLinesError::IllFormedSpan(sp)));
}
let lo = self.lookup_char_pos(sp.lo());
if cfg!(feature = "debug") {
debug!("span_to_lines: lo={:?}", lo);
}
let hi = self.lookup_char_pos(sp.hi());
if cfg!(feature = "debug") {
debug!("span_to_lines: hi={:?}", hi);
}
if lo.file.start_pos != hi.file.start_pos {
return Err(Box::new(SpanLinesError::DistinctSources(DistinctSources {
begin: FilePos(lo.file.name.clone(), lo.file.start_pos),
end: FilePos(hi.file.name.clone(), hi.file.start_pos),
})));
}
assert!(hi.line >= lo.line);
// Empty file contains no lines
if lo.file.src.is_empty() {
return Ok(FileLines {
file: lo.file,
lines: vec![],
});
}
let mut lines = Vec::with_capacity(hi.line - lo.line + 1);
// The span starts partway through the first line,
// but after that it starts from offset 0.
let mut start_col = lo.col;
// For every line but the last, it extends from `start_col`
// and to the end of the line. Be careful because the line
// numbers in Loc are 1-based, so we subtract 1 to get 0-based
// lines.
for line_index in lo.line - 1..hi.line - 1 {
let line_len = lo
.file
.get_line(line_index)
.map(|s| s.chars().count())
.unwrap_or(0);
lines.push(LineInfo {
line_index,
start_col,
end_col: CharPos::from_usize(line_len),
});
start_col = CharPos::from_usize(0);
}
// For the last line, it extends from `start_col` to `hi.col`:
lines.push(LineInfo {
line_index: hi.line - 1,
start_col,
end_col: hi.col,
});
Ok(FileLines {
file: lo.file,
lines,
})
}
/// Extract the source surrounding the given `Span` using the
/// `extract_source` function. The extract function takes three
/// arguments: a string slice containing the source, an index in
/// the slice for the beginning of the span and an index in the slice for
/// the end of the span.
fn span_to_source<F, Ret>(
&self,
sp: Span,
extract_source: F,
) -> Result<Ret, Box<SpanSnippetError>>
where
F: FnOnce(&str, usize, usize) -> Ret,
{
if sp.lo() > sp.hi() {
return Err(Box::new(SpanSnippetError::IllFormedSpan(sp)));
}
if sp.lo.is_dummy() || sp.hi.is_dummy() {
return Err(Box::new(SpanSnippetError::DummyBytePos));
}
let local_begin = self.lookup_byte_offset(sp.lo());
let local_end = self.lookup_byte_offset(sp.hi());
if local_begin.sf.start_pos != local_end.sf.start_pos {
Err(Box::new(SpanSnippetError::DistinctSources(
DistinctSources {
begin: FilePos(local_begin.sf.name.clone(), local_begin.sf.start_pos),
end: FilePos(local_end.sf.name.clone(), local_end.sf.start_pos),
},
)))
} else {
let start_index = local_begin.pos.to_usize();
let end_index = local_end.pos.to_usize();
let source_len = (local_begin.sf.end_pos - local_begin.sf.start_pos).to_usize();
if start_index > end_index || end_index > source_len {
return Err(Box::new(SpanSnippetError::MalformedForSourcemap(
MalformedSourceMapPositions {
name: local_begin.sf.name.clone(),
source_len,
begin_pos: local_begin.pos,
end_pos: local_end.pos,
},
)));
}
let src = &local_begin.sf.src;
Ok(extract_source(src, start_index, end_index))
}
}
/// Calls `op` with the source code located at `sp`.
pub fn with_snippet_of_span<F, Ret>(
&self,
sp: Span,
op: F,
) -> Result<Ret, Box<SpanSnippetError>>
where
F: FnOnce(&str) -> Ret,
{
self.span_to_source(sp, |src, start_index, end_index| {
op(&src[start_index..end_index])
})
}
pub fn span_to_margin(&self, sp: Span) -> Option<usize> {
match self.span_to_prev_source(sp) {
Err(_) => None,
Ok(source) => source
.split('\n')
.last()
.map(|last_line| last_line.len() - last_line.trim_start().len()),
}
}
/// Calls the given closure with the source snippet before the given `Span`
pub fn with_span_to_prev_source<F, Ret>(
&self,
sp: Span,
op: F,
) -> Result<Ret, Box<SpanSnippetError>>
where
F: FnOnce(&str) -> Ret,
{
self.span_to_source(sp, |src, start_index, _| op(&src[..start_index]))
}
/// Return the source snippet as `String` before the given `Span`
pub fn span_to_prev_source(&self, sp: Span) -> Result<String, Box<SpanSnippetError>> {
self.with_span_to_prev_source(sp, |s| s.to_string())
}
/// Calls the given closure with the source snippet after the given `Span`
pub fn with_span_to_next_source<F, Ret>(
&self,
sp: Span,
op: F,
) -> Result<Ret, Box<SpanSnippetError>>
where
F: FnOnce(&str) -> Ret,
{
self.span_to_source(sp, |src, _, end_index| op(&src[end_index..]))
}
/// Return the source snippet as `String` after the given `Span`
pub fn span_to_next_source(&self, sp: Span) -> Result<String, Box<SpanSnippetError>> {
self.with_span_to_next_source(sp, |s| s.to_string())
}
/// Extend the given `Span` to just after the previous occurrence of `c`.
/// Return the same span if no character could be found or if an error
/// occurred while retrieving the code snippet.
pub fn span_extend_to_prev_char(&self, sp: Span, c: char) -> Span {
if let Ok(prev_source) = self.span_to_prev_source(sp) {
let prev_source = prev_source.rsplit(c).next().unwrap_or("").trim_start();
if !prev_source.is_empty() && !prev_source.contains('\n') {
return sp.with_lo(BytePos(sp.lo().0 - prev_source.len() as u32));
}
}
sp
}
/// Extend the given `Span` to just after the previous occurrence of `pat`
/// when surrounded by whitespace. Return the same span if no character
/// could be found or if an error occurred while retrieving the code
/// snippet.
pub fn span_extend_to_prev_str(&self, sp: Span, pat: &str, accept_newlines: bool) -> Span {
// assure that the pattern is delimited, to avoid the following
// fn my_fn()
// ^^^^ returned span without the check
// ---------- correct span
for ws in &[" ", "\t", "\n"] {
let pat = pat.to_owned() + ws;
if let Ok(prev_source) = self.span_to_prev_source(sp) {
let prev_source = prev_source.rsplit(&pat).next().unwrap_or("").trim_start();
if !prev_source.is_empty() && (!prev_source.contains('\n') || accept_newlines) {
return sp.with_lo(BytePos(sp.lo().0 - prev_source.len() as u32));
}
}
}
sp
}
/// Extend the given `Span` to just after the next occurrence of `c`.
/// Return the same span if no character could be found or if an error
/// occurred while retrieving the code snippet.
pub fn span_extend_to_next_char(&self, sp: Span, c: char) -> Span {
if let Ok(next_source) = self.span_to_next_source(sp) {
let next_source = next_source.split(c).next().unwrap_or("").trim_end();
if !next_source.is_empty() && !next_source.contains('\n') {
return sp.with_hi(BytePos(sp.hi().0 + next_source.len() as u32));
}
}
sp
}
/// Extend the given `Span` to just after the next occurrence of `pat`
/// when surrounded by whitespace. Return the same span if no character
/// could be found or if an error occurred while retrieving the code
/// snippet.
pub fn span_extend_to_next_str(&self, sp: Span, pat: &str, accept_newlines: bool) -> Span {
for ws in &[" ", "\t", "\n"] {
let pat = pat.to_owned() + ws;
if let Ok(next_source) = self.span_to_next_source(sp) {
let next_source = next_source.split(&pat).next().unwrap_or("").trim_end();
if !next_source.is_empty() && (!next_source.contains('\n') || accept_newlines) {
return sp.with_hi(BytePos(sp.hi().0 + next_source.len() as u32));
}
}
}
sp
}
/// Given a `Span`, try to get a shorter span ending before the first
/// occurrence of `c` `char`
///
///
/// # Notes
///
/// This method returns a dummy span for a dummy span.
pub fn span_until_char(&self, sp: Span, c: char) -> Span {
if sp.is_dummy() {
return sp;
}
let v = self.span_to_source(sp, |src, start_index, end_index| {
let snippet = &src[start_index..end_index];
let snippet = snippet.split(c).next().unwrap_or("").trim_end();
if !snippet.is_empty() && !snippet.contains('\n') {
sp.with_hi(BytePos(sp.lo().0 + snippet.len() as u32))
} else {
sp
}
});
match v {
Ok(v) => v,
Err(_) => sp,
}
}
/// Given a `Span`, try to get a shorter span ending just after the first
/// occurrence of `char` `c`.
///
/// # Notes
///
/// This method returns a dummy span for a dummy span.
pub fn span_through_char(&self, sp: Span, c: char) -> Span {
if sp.is_dummy() {
return sp;
}
if let Ok(snippet) = self.span_to_snippet(sp) {
if let Some(offset) = snippet.find(c) {
return sp.with_hi(BytePos(sp.lo().0 + (offset + c.len_utf8()) as u32));
}
}
sp
}
/// Given a `Span`, get a new `Span` covering the first token and all its
/// trailing whitespace or the original `Span`.
///
/// If `sp` points to `"let mut x"`, then a span pointing at `"let "` will
/// be returned.
pub fn span_until_non_whitespace(&self, sp: Span) -> Span {
let mut whitespace_found = false;
self.span_take_while(sp, |c| {
if !whitespace_found && c.is_whitespace() {
whitespace_found = true;
}
!whitespace_found || c.is_whitespace()
})
}
/// Given a `Span`, get a new `Span` covering the first token without its
/// trailing whitespace or the original `Span` in case of error.
///
/// If `sp` points to `"let mut x"`, then a span pointing at `"let"` will be
/// returned.
pub fn span_until_whitespace(&self, sp: Span) -> Span {
self.span_take_while(sp, |c| !c.is_whitespace())
}
/// Given a `Span`, get a shorter one until `predicate` yields false.
pub fn span_take_while<P>(&self, sp: Span, mut predicate: P) -> Span
where
P: for<'r> FnMut(&'r char) -> bool,
{
self.span_to_source(sp, |src, start_index, end_index| {
let snippet = &src[start_index..end_index];
let offset = snippet
.chars()
.take_while(&mut predicate)
.map(|c| c.len_utf8())
.sum::<usize>();
sp.with_hi(BytePos(sp.lo().0 + (offset as u32)))
})
.unwrap_or(sp)
}
pub fn def_span(&self, sp: Span) -> Span {
self.span_until_char(sp, '{')
}
/// Returns a new span representing just the start-point of this span
pub fn start_point(&self, sp: Span) -> Span {
let pos = sp.lo().0;
let width = self.find_width_of_character_at_span(sp, false);
let corrected_start_position = pos.checked_add(width).unwrap_or(pos);
let end_point = BytePos(cmp::max(corrected_start_position, sp.lo().0));
sp.with_hi(end_point)
}
/// Returns a new span representing just the end-point of this span
pub fn end_point(&self, sp: Span) -> Span {
let pos = sp.hi().0;
let width = self.find_width_of_character_at_span(sp, false);
let corrected_end_position = pos.checked_sub(width).unwrap_or(pos);
let end_point = BytePos(cmp::max(corrected_end_position, sp.lo().0));
sp.with_lo(end_point)
}
/// Returns a new span representing the next character after the end-point
/// of this span
pub fn next_point(&self, sp: Span) -> Span {
let start_of_next_point = sp.hi().0;
let width = self.find_width_of_character_at_span(sp, true);
// If the width is 1, then the next span should point to the same `lo` and `hi`.
// However, in the case of a multibyte character, where the width != 1,
// the next span should span multiple bytes to include the whole
// character.
let end_of_next_point = start_of_next_point
.checked_add(width - 1)
.unwrap_or(start_of_next_point);
let end_of_next_point = BytePos(cmp::max(sp.lo().0 + 1, end_of_next_point));
Span::new(BytePos(start_of_next_point), end_of_next_point, sp.ctxt())
}
/// Finds the width of a character, either before or after the provided
/// span.
fn find_width_of_character_at_span(&self, sp: Span, forwards: bool) -> u32 {
// Disregard malformed spans and assume a one-byte wide character.
if sp.lo() >= sp.hi() {
debug!("find_width_of_character_at_span: early return malformed span");
return 1;
}
let local_begin = self.lookup_byte_offset(sp.lo());
let local_end = self.lookup_byte_offset(sp.hi());
debug!(
"find_width_of_character_at_span: local_begin=`{:?}`, local_end=`{:?}`",
local_begin, local_end
);
let start_index = local_begin.pos.to_usize();
let end_index = local_end.pos.to_usize();
debug!(
"find_width_of_character_at_span: start_index=`{:?}`, end_index=`{:?}`",
start_index, end_index
);
// Disregard indexes that are at the start or end of their spans, they can't fit
// bigger characters.
if (!forwards && end_index == usize::min_value())
|| (forwards && start_index == usize::max_value())
{
debug!("find_width_of_character_at_span: start or end of span, cannot be multibyte");
return 1;
}
let source_len = (local_begin.sf.end_pos - local_begin.sf.start_pos).to_usize();
debug!(
"find_width_of_character_at_span: source_len=`{:?}`",
source_len
);
// Ensure indexes are also not malformed.
if start_index > end_index || end_index > source_len {
debug!("find_width_of_character_at_span: source indexes are malformed");
return 1;
}
// We need to extend the snippet to the end of the src rather than to end_index
// so when searching forwards for boundaries we've got somewhere to
// search.
let src = &local_begin.sf.src;
let snippet = {
let len = src.len();
&src[start_index..len]
};
debug!("find_width_of_character_at_span: snippet=`{:?}`", snippet);
let mut target = if forwards {
end_index + 1
} else {
end_index - 1
};
debug!(
"find_width_of_character_at_span: initial target=`{:?}`",
target
);
while !snippet.is_char_boundary(target - start_index) && target < source_len {
target = if forwards {
target + 1
} else {
match target.checked_sub(1) {
Some(target) => target,
None => {
break;
}
}
};
debug!("find_width_of_character_at_span: target=`{:?}`", target);
}
debug!(
"find_width_of_character_at_span: final target=`{:?}`",
target
);
if forwards {
(target - end_index) as u32
} else {
(end_index - target) as u32
}
}
pub fn get_source_file(&self, filename: &FileName) -> Option<Lrc<SourceFile>> {
for sf in self.files.borrow().source_files.iter() {
if *filename == sf.name {
return Some(sf.clone());
}
}
None
}
/// For a global BytePos compute the local offset within the containing
/// SourceFile
pub fn lookup_byte_offset(&self, bpos: BytePos) -> SourceFileAndBytePos {
let sf = self.lookup_source_file(bpos);
let offset = bpos - sf.start_pos;
SourceFileAndBytePos { sf, pos: offset }
}
/// Converts an absolute BytePos to a CharPos relative to the source_file.
fn bytepos_to_file_charpos(&self, bpos: BytePos) -> CharPos {
let map = self.lookup_source_file(bpos);
self.bytepos_to_file_charpos_with(&map, bpos)
}
fn bytepos_to_file_charpos_with(&self, map: &SourceFile, bpos: BytePos) -> CharPos {
let total_extra_bytes = self.calc_utf16_offset(map, bpos, &mut Default::default());
assert!(
map.start_pos.to_u32() + total_extra_bytes <= bpos.to_u32(),
"map.start_pos = {:?}; total_extra_bytes = {}; bpos = {:?}",
map.start_pos,
total_extra_bytes,
bpos,
);
CharPos(bpos.to_usize() - map.start_pos.to_usize() - total_extra_bytes as usize)
}
/// Converts a span of absolute BytePos to a CharPos relative to the
/// source_file.
pub fn span_to_char_offset(&self, file: &SourceFile, span: Span) -> (u32, u32) {
// We rename this to feel more comfortable while doing math.
let start_offset = file.start_pos;
let mut state = ByteToCharPosState::default();
let start = span.lo.to_u32()
- start_offset.to_u32()
- self.calc_utf16_offset(file, span.lo, &mut state);
let end = span.hi.to_u32()
- start_offset.to_u32()
- self.calc_utf16_offset(file, span.hi, &mut state);
(start, end)
}
/// Calculates the number of excess chars seen in the UTF-8 encoding of a
/// file compared with the UTF-16 encoding.
fn calc_utf16_offset(
&self,
file: &SourceFile,
bpos: BytePos,
state: &mut ByteToCharPosState,
) -> u32 {