-
Notifications
You must be signed in to change notification settings - Fork 13k
/
Copy pathcodemap.rs
1735 lines (1481 loc) · 59.6 KB
/
codemap.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 CodeMap 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
//! CodeMap and are represented by FileMaps. Byte positions are stored in
//! `spans` and used pervasively in the compiler. They are absolute positions
//! within the CodeMap, which upon request can be converted to line and column
//! information, source code snippets, etc.
pub use self::ExpnFormat::*;
use std::cell::{Cell, RefCell};
use std::ops::{Add, Sub};
use std::path::Path;
use std::rc::Rc;
use std::cmp;
use std::{fmt, fs};
use std::io::{self, Read};
use serialize::{Encodable, Decodable, Encoder, Decoder};
use ast::Name;
// _____________________________________________________________________________
// Pos, BytePos, CharPos
//
pub trait Pos {
fn from_usize(n: usize) -> Self;
fn to_usize(&self) -> usize;
}
/// A byte offset. Keep this small (currently 32-bits), as AST contains
/// a lot of them.
#[derive(Clone, Copy, PartialEq, Eq, Hash, PartialOrd, Ord, Debug)]
pub struct BytePos(pub u32);
/// A character offset. Because of multibyte utf8 characters, a byte offset
/// is not equivalent to a character offset. The CodeMap will convert BytePos
/// values to CharPos values as necessary.
#[derive(Copy, Clone, PartialEq, Eq, Hash, PartialOrd, Ord, Debug)]
pub struct CharPos(pub usize);
// FIXME: Lots of boilerplate in these impls, but so far my attempts to fix
// have been unsuccessful
impl Pos for BytePos {
fn from_usize(n: usize) -> BytePos { BytePos(n as u32) }
fn to_usize(&self) -> usize { let BytePos(n) = *self; n as usize }
}
impl Add for BytePos {
type Output = BytePos;
fn add(self, rhs: BytePos) -> BytePos {
BytePos((self.to_usize() + rhs.to_usize()) as u32)
}
}
impl Sub for BytePos {
type Output = BytePos;
fn sub(self, rhs: BytePos) -> BytePos {
BytePos((self.to_usize() - rhs.to_usize()) as u32)
}
}
impl Encodable for BytePos {
fn encode<S: Encoder>(&self, s: &mut S) -> Result<(), S::Error> {
s.emit_u32(self.0)
}
}
impl Decodable for BytePos {
fn decode<D: Decoder>(d: &mut D) -> Result<BytePos, D::Error> {
Ok(BytePos(d.read_u32()?))
}
}
impl Pos for CharPos {
fn from_usize(n: usize) -> CharPos { CharPos(n) }
fn to_usize(&self) -> usize { let CharPos(n) = *self; n }
}
impl Add for CharPos {
type Output = CharPos;
fn add(self, rhs: CharPos) -> CharPos {
CharPos(self.to_usize() + rhs.to_usize())
}
}
impl Sub for CharPos {
type Output = CharPos;
fn sub(self, rhs: CharPos) -> CharPos {
CharPos(self.to_usize() - rhs.to_usize())
}
}
// _____________________________________________________________________________
// Span, MultiSpan, Spanned
//
/// Spans represent a region of code, used for error reporting. Positions in spans
/// are *absolute* positions from the beginning of the codemap, not positions
/// relative to FileMaps. Methods on the CodeMap can be used to relate spans back
/// to the original source.
/// You must be careful if the span crosses more than one file - you will not be
/// able to use many of the functions on spans in codemap and you cannot assume
/// that the length of the span = hi - lo; there may be space in the BytePos
/// range between files.
#[derive(Clone, Copy, Hash, PartialEq, Eq)]
pub struct Span {
pub lo: BytePos,
pub hi: BytePos,
/// Information about where the macro came from, if this piece of
/// code was created by a macro expansion.
pub expn_id: ExpnId
}
/// A collection of spans. Spans have two orthogonal attributes:
///
/// - they can be *primary spans*. In this case they are the locus of
/// the error, and would be rendered with `^^^`.
/// - they can have a *label*. In this case, the label is written next
/// to the mark in the snippet when we render.
#[derive(Clone)]
pub struct MultiSpan {
primary_spans: Vec<Span>,
span_labels: Vec<(Span, String)>,
}
#[derive(Clone, Debug)]
pub struct SpanLabel {
/// The span we are going to include in the final snippet.
pub span: Span,
/// Is this a primary span? This is the "locus" of the message,
/// and is indicated with a `^^^^` underline, versus `----`.
pub is_primary: bool,
/// What label should we attach to this span (if any)?
pub label: Option<String>,
}
pub const DUMMY_SP: Span = Span { lo: BytePos(0), hi: BytePos(0), expn_id: NO_EXPANSION };
// Generic span to be used for code originating from the command line
pub const COMMAND_LINE_SP: Span = Span { lo: BytePos(0),
hi: BytePos(0),
expn_id: COMMAND_LINE_EXPN };
impl Span {
/// Returns a new span representing just the end-point of this span
pub fn end_point(self) -> Span {
let lo = cmp::max(self.hi.0 - 1, self.lo.0);
Span { lo: BytePos(lo), hi: self.hi, expn_id: self.expn_id}
}
/// Returns `self` if `self` is not the dummy span, and `other` otherwise.
pub fn substitute_dummy(self, other: Span) -> Span {
if self.source_equal(&DUMMY_SP) { other } else { self }
}
pub fn contains(self, other: Span) -> bool {
self.lo <= other.lo && other.hi <= self.hi
}
/// Return true if the spans are equal with regards to the source text.
///
/// Use this instead of `==` when either span could be generated code,
/// and you only care that they point to the same bytes of source text.
pub fn source_equal(&self, other: &Span) -> bool {
self.lo == other.lo && self.hi == other.hi
}
/// Returns `Some(span)`, a union of `self` and `other`, on overlap.
pub fn merge(self, other: Span) -> Option<Span> {
if self.expn_id != other.expn_id {
return None;
}
if (self.lo <= other.lo && self.hi > other.lo) ||
(self.lo >= other.lo && self.lo < other.hi) {
Some(Span {
lo: cmp::min(self.lo, other.lo),
hi: cmp::max(self.hi, other.hi),
expn_id: self.expn_id,
})
} else {
None
}
}
/// Returns `Some(span)`, where the start is trimmed by the end of `other`
pub fn trim_start(self, other: Span) -> Option<Span> {
if self.hi > other.hi {
Some(Span { lo: cmp::max(self.lo, other.hi), .. self })
} else {
None
}
}
}
#[derive(Clone, PartialEq, Eq, RustcEncodable, RustcDecodable, Hash, Debug, Copy)]
pub struct Spanned<T> {
pub node: T,
pub span: Span,
}
impl Encodable for Span {
fn encode<S: Encoder>(&self, s: &mut S) -> Result<(), S::Error> {
s.emit_struct("Span", 2, |s| {
s.emit_struct_field("lo", 0, |s| {
self.lo.encode(s)
})?;
s.emit_struct_field("hi", 1, |s| {
self.hi.encode(s)
})
})
}
}
impl Decodable for Span {
fn decode<D: Decoder>(d: &mut D) -> Result<Span, D::Error> {
d.read_struct("Span", 2, |d| {
let lo = d.read_struct_field("lo", 0, |d| {
BytePos::decode(d)
})?;
let hi = d.read_struct_field("hi", 1, |d| {
BytePos::decode(d)
})?;
Ok(mk_sp(lo, hi))
})
}
}
fn default_span_debug(span: Span, f: &mut fmt::Formatter) -> fmt::Result {
write!(f, "Span {{ lo: {:?}, hi: {:?}, expn_id: {:?} }}",
span.lo, span.hi, span.expn_id)
}
thread_local!(pub static SPAN_DEBUG: Cell<fn(Span, &mut fmt::Formatter) -> fmt::Result> =
Cell::new(default_span_debug));
impl fmt::Debug for Span {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
SPAN_DEBUG.with(|span_debug| span_debug.get()(*self, f))
}
}
pub fn spanned<T>(lo: BytePos, hi: BytePos, t: T) -> Spanned<T> {
respan(mk_sp(lo, hi), t)
}
pub fn respan<T>(sp: Span, t: T) -> Spanned<T> {
Spanned {node: t, span: sp}
}
pub fn dummy_spanned<T>(t: T) -> Spanned<T> {
respan(DUMMY_SP, t)
}
/* assuming that we're not in macro expansion */
pub fn mk_sp(lo: BytePos, hi: BytePos) -> Span {
Span {lo: lo, hi: hi, expn_id: NO_EXPANSION}
}
/// Return the span itself if it doesn't come from a macro expansion,
/// otherwise return the call site span up to the `enclosing_sp` by
/// following the `expn_info` chain.
pub fn original_sp(cm: &CodeMap, sp: Span, enclosing_sp: Span) -> Span {
let call_site1 = cm.with_expn_info(sp.expn_id, |ei| ei.map(|ei| ei.call_site));
let call_site2 = cm.with_expn_info(enclosing_sp.expn_id, |ei| ei.map(|ei| ei.call_site));
match (call_site1, call_site2) {
(None, _) => sp,
(Some(call_site1), Some(call_site2)) if call_site1 == call_site2 => sp,
(Some(call_site1), _) => original_sp(cm, call_site1, enclosing_sp),
}
}
impl MultiSpan {
pub fn new() -> MultiSpan {
MultiSpan {
primary_spans: vec![],
span_labels: vec![]
}
}
pub fn from_span(primary_span: Span) -> MultiSpan {
MultiSpan {
primary_spans: vec![primary_span],
span_labels: vec![]
}
}
pub fn from_spans(vec: Vec<Span>) -> MultiSpan {
MultiSpan {
primary_spans: vec,
span_labels: vec![]
}
}
pub fn push_span_label(&mut self, span: Span, label: String) {
self.span_labels.push((span, label));
}
/// Selects the first primary span (if any)
pub fn primary_span(&self) -> Option<Span> {
self.primary_spans.first().cloned()
}
/// Returns all primary spans.
pub fn primary_spans(&self) -> &[Span] {
&self.primary_spans
}
/// Returns the strings to highlight. We always ensure that there
/// is an entry for each of the primary spans -- for each primary
/// span P, if there is at least one label with span P, we return
/// those labels (marked as primary). But otherwise we return
/// `SpanLabel` instances with empty labels.
pub fn span_labels(&self) -> Vec<SpanLabel> {
let is_primary = |span| self.primary_spans.contains(&span);
let mut span_labels = vec![];
for &(span, ref label) in &self.span_labels {
span_labels.push(SpanLabel {
span: span,
is_primary: is_primary(span),
label: Some(label.clone())
});
}
for &span in &self.primary_spans {
if !span_labels.iter().any(|sl| sl.span == span) {
span_labels.push(SpanLabel {
span: span,
is_primary: true,
label: None
});
}
}
span_labels
}
}
impl From<Span> for MultiSpan {
fn from(span: Span) -> MultiSpan {
MultiSpan::from_span(span)
}
}
// _____________________________________________________________________________
// Loc, LocWithOpt, FileMapAndLine, FileMapAndBytePos
//
/// A source code location used for error reporting
#[derive(Debug)]
pub struct Loc {
/// Information about the original source
pub file: Rc<FileMap>,
/// The (1-based) line number
pub line: usize,
/// The (0-based) column offset
pub col: CharPos
}
/// A source code location used as the result of lookup_char_pos_adj
// Actually, *none* of the clients use the filename *or* file field;
// perhaps they should just be removed.
#[derive(Debug)]
pub struct LocWithOpt {
pub filename: FileName,
pub line: usize,
pub col: CharPos,
pub file: Option<Rc<FileMap>>,
}
// used to be structural records. Better names, anyone?
#[derive(Debug)]
pub struct FileMapAndLine { pub fm: Rc<FileMap>, pub line: usize }
#[derive(Debug)]
pub struct FileMapAndBytePos { pub fm: Rc<FileMap>, pub pos: BytePos }
// _____________________________________________________________________________
// ExpnFormat, NameAndSpan, ExpnInfo, ExpnId
//
/// The source of expansion.
#[derive(Clone, Hash, Debug, PartialEq, Eq)]
pub enum ExpnFormat {
/// e.g. #[derive(...)] <item>
MacroAttribute(Name),
/// e.g. `format!()`
MacroBang(Name),
}
#[derive(Clone, Hash, Debug)]
pub struct NameAndSpan {
/// The format with which the macro was invoked.
pub format: ExpnFormat,
/// Whether the macro is allowed to use #[unstable]/feature-gated
/// features internally without forcing the whole crate to opt-in
/// to them.
pub allow_internal_unstable: bool,
/// The span of the macro definition itself. The macro may not
/// have a sensible definition span (e.g. something defined
/// completely inside libsyntax) in which case this is None.
pub span: Option<Span>
}
impl NameAndSpan {
pub fn name(&self) -> Name {
match self.format {
ExpnFormat::MacroAttribute(s) => s,
ExpnFormat::MacroBang(s) => s,
}
}
}
/// Extra information for tracking spans of macro and syntax sugar expansion
#[derive(Hash, Debug)]
pub struct ExpnInfo {
/// The location of the actual macro invocation or syntax sugar , e.g.
/// `let x = foo!();` or `if let Some(y) = x {}`
///
/// This may recursively refer to other macro invocations, e.g. if
/// `foo!()` invoked `bar!()` internally, and there was an
/// expression inside `bar!`; the call_site of the expression in
/// the expansion would point to the `bar!` invocation; that
/// call_site span would have its own ExpnInfo, with the call_site
/// pointing to the `foo!` invocation.
pub call_site: Span,
/// Information about the expansion.
pub callee: NameAndSpan
}
#[derive(PartialEq, Eq, Clone, Debug, Hash, RustcEncodable, RustcDecodable, Copy)]
pub struct ExpnId(u32);
pub const NO_EXPANSION: ExpnId = ExpnId(!0);
// For code appearing from the command line
pub const COMMAND_LINE_EXPN: ExpnId = ExpnId(!1);
impl ExpnId {
pub fn from_u32(id: u32) -> ExpnId {
ExpnId(id)
}
pub fn into_u32(self) -> u32 {
self.0
}
}
// _____________________________________________________________________________
// FileMap, MultiByteChar, FileName, FileLines
//
pub type FileName = String;
#[derive(Copy, Clone, Debug, PartialEq, Eq)]
pub struct LineInfo {
/// Index of line, starting from 0.
pub line_index: usize,
/// Column in line where span begins, starting from 0.
pub start_col: CharPos,
/// Column in line where span ends, starting from 0, exclusive.
pub end_col: CharPos,
}
pub struct FileLines {
pub file: Rc<FileMap>,
pub lines: Vec<LineInfo>
}
/// Identifies an offset of a multi-byte character in a FileMap
#[derive(Copy, Clone, RustcEncodable, RustcDecodable, Eq, PartialEq)]
pub struct MultiByteChar {
/// The absolute offset of the character in the CodeMap
pub pos: BytePos,
/// The number of bytes, >=2
pub bytes: usize,
}
/// A single source in the CodeMap.
pub struct FileMap {
/// The name of the file that the source came from, source that doesn't
/// originate from files has names between angle brackets by convention,
/// e.g. `<anon>`
pub name: FileName,
/// The complete source code
pub src: Option<Rc<String>>,
/// The start position of this source in the CodeMap
pub start_pos: BytePos,
/// The end position of this source in the CodeMap
pub end_pos: BytePos,
/// Locations of lines beginnings in the source code
pub lines: RefCell<Vec<BytePos>>,
/// Locations of multi-byte characters in the source code
pub multibyte_chars: RefCell<Vec<MultiByteChar>>,
}
impl Encodable for FileMap {
fn encode<S: Encoder>(&self, s: &mut S) -> Result<(), S::Error> {
s.emit_struct("FileMap", 5, |s| {
s.emit_struct_field("name", 0, |s| self.name.encode(s))?;
s.emit_struct_field("start_pos", 1, |s| self.start_pos.encode(s))?;
s.emit_struct_field("end_pos", 2, |s| self.end_pos.encode(s))?;
s.emit_struct_field("lines", 3, |s| {
let lines = self.lines.borrow();
// store the length
s.emit_u32(lines.len() as u32)?;
if !lines.is_empty() {
// In order to preserve some space, we exploit the fact that
// the lines list is sorted and individual lines are
// probably not that long. Because of that we can store lines
// as a difference list, using as little space as possible
// for the differences.
let max_line_length = if lines.len() == 1 {
0
} else {
lines.windows(2)
.map(|w| w[1] - w[0])
.map(|bp| bp.to_usize())
.max()
.unwrap()
};
let bytes_per_diff: u8 = match max_line_length {
0 ... 0xFF => 1,
0x100 ... 0xFFFF => 2,
_ => 4
};
// Encode the number of bytes used per diff.
bytes_per_diff.encode(s)?;
// Encode the first element.
lines[0].encode(s)?;
let diff_iter = (&lines[..]).windows(2)
.map(|w| (w[1] - w[0]));
match bytes_per_diff {
1 => for diff in diff_iter { (diff.0 as u8).encode(s)? },
2 => for diff in diff_iter { (diff.0 as u16).encode(s)? },
4 => for diff in diff_iter { diff.0.encode(s)? },
_ => unreachable!()
}
}
Ok(())
})?;
s.emit_struct_field("multibyte_chars", 4, |s| {
(*self.multibyte_chars.borrow()).encode(s)
})
})
}
}
impl Decodable for FileMap {
fn decode<D: Decoder>(d: &mut D) -> Result<FileMap, D::Error> {
d.read_struct("FileMap", 5, |d| {
let name: String = d.read_struct_field("name", 0, |d| Decodable::decode(d))?;
let start_pos: BytePos = d.read_struct_field("start_pos", 1, |d| Decodable::decode(d))?;
let end_pos: BytePos = d.read_struct_field("end_pos", 2, |d| Decodable::decode(d))?;
let lines: Vec<BytePos> = d.read_struct_field("lines", 3, |d| {
let num_lines: u32 = Decodable::decode(d)?;
let mut lines = Vec::with_capacity(num_lines as usize);
if num_lines > 0 {
// Read the number of bytes used per diff.
let bytes_per_diff: u8 = Decodable::decode(d)?;
// Read the first element.
let mut line_start: BytePos = Decodable::decode(d)?;
lines.push(line_start);
for _ in 1..num_lines {
let diff = match bytes_per_diff {
1 => d.read_u8()? as u32,
2 => d.read_u16()? as u32,
4 => d.read_u32()?,
_ => unreachable!()
};
line_start = line_start + BytePos(diff);
lines.push(line_start);
}
}
Ok(lines)
})?;
let multibyte_chars: Vec<MultiByteChar> =
d.read_struct_field("multibyte_chars", 4, |d| Decodable::decode(d))?;
Ok(FileMap {
name: name,
start_pos: start_pos,
end_pos: end_pos,
src: None,
lines: RefCell::new(lines),
multibyte_chars: RefCell::new(multibyte_chars)
})
})
}
}
impl fmt::Debug for FileMap {
fn fmt(&self, fmt: &mut fmt::Formatter) -> fmt::Result {
write!(fmt, "FileMap({})", self.name)
}
}
impl FileMap {
/// EFFECT: register a start-of-line offset in the
/// table of line-beginnings.
/// UNCHECKED INVARIANT: these offsets must be added in the right
/// order and must be in the right places; there is shared knowledge
/// about what ends a line between this file and parse.rs
/// WARNING: pos param here is the offset relative to start of CodeMap,
/// and CodeMap will append a newline when adding a filemap without a newline at the end,
/// so the safe way to call this is with value calculated as
/// filemap.start_pos + newline_offset_relative_to_the_start_of_filemap.
pub fn next_line(&self, pos: BytePos) {
// the new charpos must be > the last one (or it's the first one).
let mut lines = self.lines.borrow_mut();
let line_len = lines.len();
assert!(line_len == 0 || ((*lines)[line_len - 1] < pos));
lines.push(pos);
}
/// get a line from the list of pre-computed line-beginnings.
/// line-number here is 0-based.
pub fn get_line(&self, line_number: usize) -> Option<&str> {
match self.src {
Some(ref src) => {
let lines = self.lines.borrow();
lines.get(line_number).map(|&line| {
let begin: BytePos = line - self.start_pos;
let begin = begin.to_usize();
// We can't use `lines.get(line_number+1)` because we might
// be parsing when we call this function and thus the current
// line is the last one we have line info for.
let slice = &src[begin..];
match slice.find('\n') {
Some(e) => &slice[..e],
None => slice
}
})
}
None => None
}
}
pub fn record_multibyte_char(&self, pos: BytePos, bytes: usize) {
assert!(bytes >=2 && bytes <= 4);
let mbc = MultiByteChar {
pos: pos,
bytes: bytes,
};
self.multibyte_chars.borrow_mut().push(mbc);
}
pub fn is_real_file(&self) -> bool {
!(self.name.starts_with("<") &&
self.name.ends_with(">"))
}
pub fn is_imported(&self) -> bool {
self.src.is_none()
}
fn count_lines(&self) -> usize {
self.lines.borrow().len()
}
}
/// 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;
/// 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 read_file(&self, path: &Path) -> io::Result<String> {
let mut src = String::new();
fs::File::open(path)?.read_to_string(&mut src)?;
Ok(src)
}
}
// _____________________________________________________________________________
// CodeMap
//
pub struct CodeMap {
pub files: RefCell<Vec<Rc<FileMap>>>,
expansions: RefCell<Vec<ExpnInfo>>,
file_loader: Box<FileLoader>
}
impl CodeMap {
pub fn new() -> CodeMap {
CodeMap {
files: RefCell::new(Vec::new()),
expansions: RefCell::new(Vec::new()),
file_loader: Box::new(RealFileLoader)
}
}
pub fn with_file_loader(file_loader: Box<FileLoader>) -> CodeMap {
CodeMap {
files: RefCell::new(Vec::new()),
expansions: RefCell::new(Vec::new()),
file_loader: file_loader
}
}
pub fn file_exists(&self, path: &Path) -> bool {
self.file_loader.file_exists(path)
}
pub fn load_file(&self, path: &Path) -> io::Result<Rc<FileMap>> {
let src = self.file_loader.read_file(path)?;
Ok(self.new_filemap(path.to_str().unwrap().to_string(), src))
}
fn next_start_pos(&self) -> usize {
let files = self.files.borrow();
match files.last() {
None => 0,
// Add one so there is some space between files. This lets us distinguish
// positions in the codemap, even in the presence of zero-length files.
Some(last) => last.end_pos.to_usize() + 1,
}
}
/// Creates a new filemap without setting its line information. If you don't
/// intend to set the line information yourself, you should use new_filemap_and_lines.
pub fn new_filemap(&self, filename: FileName, mut src: String) -> Rc<FileMap> {
let start_pos = self.next_start_pos();
let mut files = self.files.borrow_mut();
// Remove utf-8 BOM if any.
if src.starts_with("\u{feff}") {
src.drain(..3);
}
let end_pos = start_pos + src.len();
let filemap = Rc::new(FileMap {
name: filename,
src: Some(Rc::new(src)),
start_pos: Pos::from_usize(start_pos),
end_pos: Pos::from_usize(end_pos),
lines: RefCell::new(Vec::new()),
multibyte_chars: RefCell::new(Vec::new()),
});
files.push(filemap.clone());
filemap
}
/// Creates a new filemap and sets its line information.
pub fn new_filemap_and_lines(&self, filename: &str, src: &str) -> Rc<FileMap> {
let fm = self.new_filemap(filename.to_string(), src.to_owned());
let mut byte_pos: u32 = fm.start_pos.0;
for line in src.lines() {
// register the start of this line
fm.next_line(BytePos(byte_pos));
// update byte_pos to include this line and the \n at the end
byte_pos += line.len() as u32 + 1;
}
fm
}
/// Allocates a new FileMap representing a source file from an external
/// crate. The source code of such an "imported filemap" is not available,
/// but we still know enough to generate accurate debuginfo location
/// information for things inlined from other crates.
pub fn new_imported_filemap(&self,
filename: FileName,
source_len: usize,
mut file_local_lines: Vec<BytePos>,
mut file_local_multibyte_chars: Vec<MultiByteChar>)
-> Rc<FileMap> {
let start_pos = self.next_start_pos();
let mut files = self.files.borrow_mut();
let end_pos = Pos::from_usize(start_pos + source_len);
let start_pos = Pos::from_usize(start_pos);
for pos in &mut file_local_lines {
*pos = *pos + start_pos;
}
for mbc in &mut file_local_multibyte_chars {
mbc.pos = mbc.pos + start_pos;
}
let filemap = Rc::new(FileMap {
name: filename,
src: None,
start_pos: start_pos,
end_pos: end_pos,
lines: RefCell::new(file_local_lines),
multibyte_chars: RefCell::new(file_local_multibyte_chars),
});
files.push(filemap.clone());
filemap
}
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)).to_string()
}
/// Lookup source information about a BytePos
pub fn lookup_char_pos(&self, pos: BytePos) -> Loc {
let chpos = self.bytepos_to_file_charpos(pos);
match self.lookup_line(pos) {
Ok(FileMapAndLine { fm: f, line: a }) => {
let line = a + 1; // Line numbers start at 1
let linebpos = (*f.lines.borrow())[a];
let linechpos = self.bytepos_to_file_charpos(linebpos);
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: line,
col: chpos - linechpos,
}
}
Err(f) => {
Loc {
file: f,
line: 0,
col: chpos,
}
}
}
}
// If the relevant filemap is empty, we don't return a line number.
fn lookup_line(&self, pos: BytePos) -> Result<FileMapAndLine, Rc<FileMap>> {
let idx = self.lookup_filemap_idx(pos);
let files = self.files.borrow();
let f = (*files)[idx].clone();
let len = f.lines.borrow().len();
if len == 0 {
return Err(f);
}
let mut a = 0;
{
let lines = f.lines.borrow();
let mut b = lines.len();
while b - a > 1 {
let m = (a + b) / 2;
if (*lines)[m] > pos {
b = m;
} else {
a = m;
}
}
assert!(a <= lines.len());
}
Ok(FileMapAndLine { fm: f, line: a })
}
pub fn lookup_char_pos_adj(&self, pos: BytePos) -> LocWithOpt {
let loc = self.lookup_char_pos(pos);
LocWithOpt {
filename: loc.file.name.to_string(),
line: loc.line,
col: loc.col,
file: Some(loc.file)
}
}
pub fn span_to_string(&self, sp: Span) -> String {
if sp == COMMAND_LINE_SP {
return "<command line option>".to_string();
}
if self.files.borrow().is_empty() && sp.source_equal(&DUMMY_SP) {
return "no-location".to_string();
}
let lo = self.lookup_char_pos_adj(sp.lo);
let hi = self.lookup_char_pos_adj(sp.hi);
return (format!("{}:{}:{}: {}:{}",
lo.filename,
lo.line,
lo.col.to_usize() + 1,
hi.line,
hi.col.to_usize() + 1)).to_string()
}
// Returns true if two spans have the same callee
// (Assumes the same ExpnFormat implies same callee)
fn match_callees(&self, sp_a: &Span, sp_b: &Span) -> bool {
let fmt_a = self
.with_expn_info(sp_a.expn_id,
|ei| ei.map(|ei| ei.callee.format.clone()));
let fmt_b = self
.with_expn_info(sp_b.expn_id,
|ei| ei.map(|ei| ei.callee.format.clone()));
fmt_a == fmt_b
}
/// Returns a formatted string showing the expansion chain of a span
///
/// Spans are printed in the following format:
///
/// filename:start_line:col: end_line:col
/// snippet
/// Callee:
/// Callee span
/// Callsite:
/// Callsite span
///
/// Callees and callsites are printed recursively (if available, otherwise header
/// and span is omitted), expanding into their own callee/callsite spans.
/// Each layer of recursion has an increased indent, and snippets are truncated
/// to at most 50 characters. Finally, recursive calls to the same macro are squashed,
/// with '...' used to represent any number of recursive calls.
pub fn span_to_expanded_string(&self, sp: Span) -> String {
self.span_to_expanded_string_internal(sp, "")
}
fn span_to_expanded_string_internal(&self, sp:Span, indent: &str) -> String {
let mut indent = indent.to_owned();
let mut output = "".to_owned();
let span_str = self.span_to_string(sp);
let mut span_snip = self.span_to_snippet(sp)
.unwrap_or("Snippet unavailable".to_owned());
// Truncate by code points - in worst case this will be more than 50 characters,
// but ensures at least 50 characters and respects byte boundaries.
let char_vec: Vec<(usize, char)> = span_snip.char_indices().collect();
if char_vec.len() > 50 {
span_snip.truncate(char_vec[49].0);
span_snip.push_str("...");
}
output.push_str(&format!("{}{}\n{}`{}`\n", indent, span_str, indent, span_snip));
if sp.expn_id == NO_EXPANSION || sp.expn_id == COMMAND_LINE_EXPN {
return output;
}