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
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
#[cfg(not(feature = "parking_lot"))]
use std::sync::Mutex;
use std::{
    borrow::Cow,
    cmp, fmt,
    hash::{Hash, Hasher},
    ops::{Add, Sub},
    path::PathBuf,
    sync::atomic::AtomicU32,
};

#[cfg(feature = "parking_lot")]
use parking_lot::Mutex;
use serde::{Deserialize, Serialize};
use url::Url;

pub use self::hygiene::{Mark, SyntaxContext};
use crate::{rustc_data_structures::stable_hasher::StableHasher, sync::Lrc};

mod analyze_source_file;
pub mod hygiene;

/// Spans represent a region of code, used for error reporting. Positions in
/// spans are *absolute* positions from the beginning of the `source_map`, not
/// positions relative to `SourceFile`s. Methods on the `SourceMap` 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 `source_map` and you cannot
/// assume that the length of the `span = hi - lo`; there may be space in the
/// `BytePos` range between files.
#[derive(Debug, Clone, Copy, Hash, PartialEq, Eq, Ord, PartialOrd, Serialize, Deserialize)]
#[cfg_attr(
    feature = "rkyv",
    derive(rkyv::Archive, rkyv::Serialize, rkyv::Deserialize)
)]
pub struct Span {
    #[serde(rename = "start")]
    #[cfg_attr(feature = "rkyv", omit_bounds)]
    pub lo: BytePos,
    #[serde(rename = "end")]
    #[cfg_attr(feature = "rkyv", omit_bounds)]
    pub hi: BytePos,
    /// Information about where the macro came from, if this piece of
    /// code was created by a macro expansion.
    #[cfg_attr(feature = "rkyv", omit_bounds)]
    pub ctxt: SyntaxContext,
}

#[cfg(feature = "arbitrary")]
#[cfg_attr(docsrs, doc(cfg(feature = "arbitrary")))]
impl<'a> arbitrary::Arbitrary<'a> for Span {
    fn arbitrary(u: &mut arbitrary::Unstructured<'_>) -> arbitrary::Result<Self> {
        let lo = u.arbitrary::<BytePos>()?;
        let hi = u.arbitrary::<BytePos>()?;

        Ok(Self::new(lo, hi, Default::default()))
    }
}

/// Dummy span, both position and length are zero, syntax context is zero as
/// well.
pub const DUMMY_SP: Span = Span {
    lo: BytePos::DUMMY,
    hi: BytePos::DUMMY,
    ctxt: SyntaxContext::empty(),
};

#[derive(Default)]
pub struct Globals {
    hygiene_data: Mutex<hygiene::HygieneData>,
    #[allow(unused)]
    dummy_cnt: AtomicU32,
}

const DUMMY_RESERVE: u32 = u32::MAX - 2_u32.pow(16);

impl Globals {
    pub fn new() -> Globals {
        Globals {
            hygiene_data: Mutex::new(hygiene::HygieneData::new()),
            dummy_cnt: AtomicU32::new(DUMMY_RESERVE),
        }
    }
}

better_scoped_tls::scoped_tls!(

    /// Storage for span hygiene data.
    ///
    /// This variable is used to manage identifiers or to identify nodes.
    /// Note that it's stored as a thread-local storage, but actually it's shared
    /// between threads.
    ///
    /// # Usages
    ///
    /// ## Span hygiene
    ///
    /// [Mark]s are stored in this variable.
    ///
    /// You can see the document how swc uses the span hygiene info at
    /// https://rustdoc.swc.rs/swc_ecma_transforms_base/resolver/fn.resolver_with_mark.html
    pub static GLOBALS: Globals
);

#[cfg_attr(
    feature = "rkyv",
    derive(rkyv::Archive, rkyv::Serialize, rkyv::Deserialize)
)]
#[derive(Debug, Eq, PartialEq, Clone, Ord, PartialOrd, Hash)]
pub enum FileName {
    Real(#[cfg_attr(feature = "rkyv", with(crate::source_map::EncodePathBuf))] PathBuf),
    /// A macro. This includes the full name of the macro, so that there are no
    /// clashes.
    Macros(String),
    /// call to `quote!`
    QuoteExpansion,
    /// Command line
    Anon,
    /// Hack in src/libsyntax/parse.rs
    MacroExpansion,
    ProcMacroSourceCode,
    Url(#[cfg_attr(feature = "rkyv", with(crate::source_map::EncodeUrl))] Url),
    Internal(String),
    /// Custom sources for explicit parser calls from plugins and drivers
    Custom(String),
}

/// A wrapper that attempts to convert a type to and from UTF-8.
///
/// Types like `OsString` and `PathBuf` aren't guaranteed to be encoded as
/// UTF-8, but they usually are anyway. Using this wrapper will archive them as
/// if they were regular `String`s.
///
/// There is built-in `AsString` supports PathBuf but it requires custom
/// serializer wrapper to handle conversion errors. This wrapper is simplified
/// version accepts errors
#[cfg(feature = "rkyv")]
#[derive(Debug, Clone, Copy)]
pub struct EncodePathBuf;

#[cfg(feature = "rkyv")]
impl rkyv::with::ArchiveWith<PathBuf> for EncodePathBuf {
    type Archived = rkyv::string::ArchivedString;
    type Resolver = rkyv::string::StringResolver;

    #[inline]
    unsafe fn resolve_with(
        field: &PathBuf,
        pos: usize,
        resolver: Self::Resolver,
        out: *mut Self::Archived,
    ) {
        // It's safe to unwrap here because if the OsString wasn't valid UTF-8 it would
        // have failed to serialize
        rkyv::string::ArchivedString::resolve_from_str(field.to_str().unwrap(), pos, resolver, out);
    }
}

#[cfg(feature = "rkyv")]
impl<S> rkyv::with::SerializeWith<PathBuf, S> for EncodePathBuf
where
    S: ?Sized + rkyv::ser::Serializer,
{
    #[inline]
    fn serialize_with(field: &PathBuf, serializer: &mut S) -> Result<Self::Resolver, S::Error> {
        let s = field.to_str().unwrap_or_default();
        rkyv::string::ArchivedString::serialize_from_str(s, serializer)
    }
}

#[cfg(feature = "rkyv")]
impl<D> rkyv::with::DeserializeWith<rkyv::string::ArchivedString, PathBuf, D> for EncodePathBuf
where
    D: ?Sized + rkyv::Fallible,
{
    #[inline]
    fn deserialize_with(
        field: &rkyv::string::ArchivedString,
        _: &mut D,
    ) -> Result<PathBuf, D::Error> {
        Ok(<PathBuf as std::str::FromStr>::from_str(field.as_str()).unwrap())
    }
}

/// A wrapper that attempts to convert a Url to and from String.
#[cfg(feature = "rkyv")]
#[derive(Debug, Clone, Copy)]
pub struct EncodeUrl;

#[cfg(feature = "rkyv")]
impl rkyv::with::ArchiveWith<Url> for EncodeUrl {
    type Archived = rkyv::string::ArchivedString;
    type Resolver = rkyv::string::StringResolver;

    #[inline]
    unsafe fn resolve_with(
        field: &Url,
        pos: usize,
        resolver: Self::Resolver,
        out: *mut Self::Archived,
    ) {
        rkyv::string::ArchivedString::resolve_from_str(field.as_str(), pos, resolver, out);
    }
}

#[cfg(feature = "rkyv")]
impl<S> rkyv::with::SerializeWith<Url, S> for EncodeUrl
where
    S: ?Sized + rkyv::ser::Serializer,
{
    #[inline]
    fn serialize_with(field: &Url, serializer: &mut S) -> Result<Self::Resolver, S::Error> {
        let field = field.as_str();
        rkyv::string::ArchivedString::serialize_from_str(field, serializer)
    }
}

#[cfg(feature = "rkyv")]
impl<D> rkyv::with::DeserializeWith<rkyv::Archived<String>, Url, D> for EncodeUrl
where
    D: ?Sized + rkyv::Fallible,
{
    #[inline]
    fn deserialize_with(field: &rkyv::string::ArchivedString, _: &mut D) -> Result<Url, D::Error> {
        Ok(Url::parse(field.as_str()).unwrap())
    }
}

impl std::fmt::Display for FileName {
    fn fmt(&self, fmt: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
        match *self {
            FileName::Real(ref path) => write!(fmt, "{}", path.display()),
            FileName::Macros(ref name) => write!(fmt, "<{} macros>", name),
            FileName::QuoteExpansion => write!(fmt, "<quote expansion>"),
            FileName::MacroExpansion => write!(fmt, "<macro expansion>"),
            FileName::Anon => write!(fmt, "<anon>"),
            FileName::ProcMacroSourceCode => write!(fmt, "<proc-macro source code>"),
            FileName::Url(ref u) => write!(fmt, "{}", u),
            FileName::Custom(ref s) | FileName::Internal(ref s) => write!(fmt, "<{}>", s),
        }
    }
}

impl From<PathBuf> for FileName {
    fn from(p: PathBuf) -> Self {
        assert!(!p.to_string_lossy().ends_with('>'));
        FileName::Real(p)
    }
}

impl From<Url> for FileName {
    fn from(url: Url) -> Self {
        FileName::Url(url)
    }
}

impl FileName {
    pub fn is_real(&self) -> bool {
        match *self {
            FileName::Real(_) => true,
            FileName::Macros(_)
            | FileName::Anon
            | FileName::MacroExpansion
            | FileName::ProcMacroSourceCode
            | FileName::Custom(_)
            | FileName::QuoteExpansion
            | FileName::Internal(_)
            | FileName::Url(_) => false,
        }
    }

    pub fn is_macros(&self) -> bool {
        match *self {
            FileName::Real(_)
            | FileName::Anon
            | FileName::MacroExpansion
            | FileName::ProcMacroSourceCode
            | FileName::Custom(_)
            | FileName::QuoteExpansion
            | FileName::Internal(_)
            | FileName::Url(_) => false,
            FileName::Macros(_) => true,
        }
    }
}

/// 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, Debug, Default, Hash, PartialEq, Eq)]
#[cfg_attr(
    feature = "diagnostic-serde",
    derive(serde::Serialize, serde::Deserialize)
)]
#[cfg_attr(
    feature = "plugin-base",
    derive(rkyv::Archive, rkyv::Serialize, rkyv::Deserialize)
)]
pub struct MultiSpan {
    primary_spans: Vec<Span>,
    span_labels: Vec<(Span, String)>,
}

extern "C" {
    fn __span_dummy_with_cmt_proxy() -> u32;
}

impl Span {
    #[inline]
    pub fn lo(self) -> BytePos {
        self.lo
    }

    #[inline]
    pub fn new(mut lo: BytePos, mut hi: BytePos, ctxt: SyntaxContext) -> Self {
        if lo > hi {
            std::mem::swap(&mut lo, &mut hi);
        }

        Span { lo, hi, ctxt }
    }

    #[inline]
    pub fn with_lo(&self, lo: BytePos) -> Span {
        Span::new(lo, self.hi, self.ctxt)
    }

    #[inline]
    pub fn hi(self) -> BytePos {
        self.hi
    }

    #[inline]
    pub fn with_hi(&self, hi: BytePos) -> Span {
        Span::new(self.lo, hi, self.ctxt)
    }

    #[inline]
    pub fn ctxt(self) -> SyntaxContext {
        self.ctxt
    }

    #[inline]
    pub fn with_ctxt(&self, ctxt: SyntaxContext) -> Span {
        Span::new(self.lo, self.hi, ctxt)
    }

    /// Returns `true` if this is a dummy span with any hygienic context.
    #[inline]
    pub fn is_dummy(self) -> bool {
        self.lo.0 == 0 && self.hi.0 == 0 || self.lo.0 >= DUMMY_RESERVE
    }

    /// Returns `true` if this is a dummy span with any hygienic context.
    #[inline]
    pub fn is_dummy_ignoring_cmt(self) -> bool {
        self.lo.0 == 0 && self.hi.0 == 0
    }

    /// Returns a new span representing an empty span at the beginning of this
    /// span
    #[inline]
    pub fn shrink_to_lo(self) -> Span {
        self.with_hi(self.lo)
    }

    /// Returns a new span representing an empty span at the end of this span
    #[inline]
    pub fn shrink_to_hi(self) -> Span {
        self.with_lo(self.hi)
    }

    /// Returns `self` if `self` is not the dummy span, and `other` otherwise.
    pub fn substitute_dummy(self, other: Span) -> Span {
        if self.is_dummy() {
            other
        } else {
            self
        }
    }

    /// Return true if `self` fully encloses `other`.
    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)`, 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(self.with_lo(cmp::max(self.lo, other.hi)))
        } else {
            None
        }
    }

    /// Return a `Span` that would enclose both `self` and `end`.
    pub fn to(self, end: Span) -> Span {
        let span_data = self;
        let end_data = end;
        // FIXME(jseyfried): self.ctxt should always equal end.ctxt here (c.f. issue
        // #23480) Return the macro span on its own to avoid weird diagnostic
        // output. It is preferable to have an incomplete span than a completely
        // nonsensical one.
        if span_data.ctxt != end_data.ctxt {
            if span_data.ctxt == SyntaxContext::empty() {
                return end;
            } else if end_data.ctxt == SyntaxContext::empty() {
                return self;
            }
            // both span fall within a macro
            // FIXME(estebank) check if it is the *same* macro
        }
        Span::new(
            cmp::min(span_data.lo, end_data.lo),
            cmp::max(span_data.hi, end_data.hi),
            if span_data.ctxt == SyntaxContext::empty() {
                end_data.ctxt
            } else {
                span_data.ctxt
            },
        )
    }

    /// Return a `Span` between the end of `self` to the beginning of `end`.
    pub fn between(self, end: Span) -> Span {
        let span = self;
        Span::new(
            span.hi,
            end.lo,
            if end.ctxt == SyntaxContext::empty() {
                end.ctxt
            } else {
                span.ctxt
            },
        )
    }

    /// Return a `Span` between the beginning of `self` to the beginning of
    /// `end`.
    pub fn until(self, end: Span) -> Span {
        let span = self;
        Span::new(
            span.lo,
            end.lo,
            if end.ctxt == SyntaxContext::empty() {
                end.ctxt
            } else {
                span.ctxt
            },
        )
    }

    pub fn from_inner_byte_pos(self, start: usize, end: usize) -> Span {
        let span = self;
        Span::new(
            span.lo + BytePos::from_usize(start),
            span.lo + BytePos::from_usize(end),
            span.ctxt,
        )
    }

    #[inline]
    pub fn apply_mark(self, mark: Mark) -> Span {
        let span = self;
        span.with_ctxt(span.ctxt.apply_mark(mark))
    }

    #[inline]
    pub fn remove_mark(&mut self) -> Mark {
        let mut span = *self;
        let mark = span.ctxt.remove_mark();
        *self = Span::new(span.lo, span.hi, span.ctxt);
        mark
    }

    #[inline]
    pub fn adjust(&mut self, expansion: Mark) -> Option<Mark> {
        let mut span = *self;
        let mark = span.ctxt.adjust(expansion);
        *self = Span::new(span.lo, span.hi, span.ctxt);
        mark
    }

    #[inline]
    pub fn glob_adjust(
        &mut self,
        expansion: Mark,
        glob_ctxt: SyntaxContext,
    ) -> Option<Option<Mark>> {
        let mut span = *self;
        let mark = span.ctxt.glob_adjust(expansion, glob_ctxt);
        *self = Span::new(span.lo, span.hi, span.ctxt);
        mark
    }

    #[inline]
    pub fn reverse_glob_adjust(
        &mut self,
        expansion: Mark,
        glob_ctxt: SyntaxContext,
    ) -> Option<Option<Mark>> {
        let mut span = *self;
        let mark = span.ctxt.reverse_glob_adjust(expansion, glob_ctxt);
        *self = Span::new(span.lo, span.hi, span.ctxt);
        mark
    }

    /// Returns `true` if `self` is marked with `mark`.
    ///
    /// Panics if `mark` is not a valid mark.
    #[inline]
    pub fn has_mark(self, mark: Mark) -> bool {
        debug_assert_ne!(
            mark,
            Mark::root(),
            "Cannot check if a span contains a `ROOT` mark"
        );

        let mut ctxt = self.ctxt;

        loop {
            if ctxt == SyntaxContext::empty() {
                return false;
            }

            let m = ctxt.remove_mark();
            if m == mark {
                return true;
            }
            if m == Mark::root() {
                return false;
            }
        }
    }

    /// Dummy span, both position are extremely large numbers so they would be
    /// ignore by sourcemap, but can still have comments
    pub fn dummy_with_cmt() -> Self {
        #[cfg(all(feature = "plugin-mode", target_arch = "wasm32"))]
        {
            let lo = BytePos(unsafe { __span_dummy_with_cmt_proxy() });

            return Span {
                lo,
                hi: lo,
                ctxt: SyntaxContext::empty(),
            };
        }

        #[cfg(not(all(feature = "plugin-mode", target_arch = "wasm32")))]
        return GLOBALS.with(|globals| {
            let lo = BytePos(
                globals
                    .dummy_cnt
                    .fetch_add(1, std::sync::atomic::Ordering::SeqCst),
            );
            Span {
                lo,
                hi: lo,
                ctxt: SyntaxContext::empty(),
            }
        });
    }
}

#[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>,
}

impl Default for Span {
    fn default() -> Self {
        DUMMY_SP
    }
}

impl MultiSpan {
    #[inline]
    pub fn new() -> MultiSpan {
        Self::default()
    }

    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 `true` if this contains only a dummy primary span with any
    /// hygienic context.
    pub fn is_dummy(&self) -> bool {
        let mut is_dummy = true;
        for span in &self.primary_spans {
            if !span.is_dummy() {
                is_dummy = false;
            }
        }
        is_dummy
    }

    /// Replaces all occurrences of one Span with another. Used to move Spans in
    /// areas that don't display well (like std macros). Returns true if
    /// replacements occurred.
    pub fn replace(&mut self, before: Span, after: Span) -> bool {
        let mut replacements_occurred = false;
        for primary_span in &mut self.primary_spans {
            if *primary_span == before {
                *primary_span = after;
                replacements_occurred = true;
            }
        }
        for span_label in &mut self.span_labels {
            if span_label.0 == before {
                span_label.0 = after;
                replacements_occurred = true;
            }
        }
        replacements_occurred
    }

    /// 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 = self
            .span_labels
            .iter()
            .map(|&(span, ref label)| SpanLabel {
                span,
                is_primary: is_primary(span),
                label: Some(label.clone()),
            })
            .collect::<Vec<_>>();

        for &span in &self.primary_spans {
            if !span_labels.iter().any(|sl| sl.span == span) {
                span_labels.push(SpanLabel {
                    span,
                    is_primary: true,
                    label: None,
                });
            }
        }

        span_labels
    }
}

impl From<Span> for MultiSpan {
    fn from(span: Span) -> MultiSpan {
        MultiSpan::from_span(span)
    }
}

impl From<Vec<Span>> for MultiSpan {
    fn from(spans: Vec<Span>) -> MultiSpan {
        MultiSpan::from_spans(spans)
    }
}

pub const NO_EXPANSION: SyntaxContext = SyntaxContext::empty();

/// Identifies an offset of a multi-byte character in a SourceFile
#[cfg_attr(
    feature = "rkyv",
    derive(rkyv::Archive, rkyv::Serialize, rkyv::Deserialize)
)]
#[derive(Copy, Clone, Eq, PartialEq, Debug)]
pub struct MultiByteChar {
    /// The absolute offset of the character in the SourceMap
    pub pos: BytePos,
    /// The number of bytes, >=2
    pub bytes: u8,
}

/// Identifies an offset of a non-narrow character in a SourceFile
#[cfg_attr(
    feature = "rkyv",
    derive(rkyv::Archive, rkyv::Serialize, rkyv::Deserialize)
)]
#[derive(Copy, Clone, Eq, PartialEq, Debug)]
pub enum NonNarrowChar {
    /// Represents a zero-width character
    ZeroWidth(BytePos),
    /// Represents a wide (fullwidth) character
    Wide(BytePos),
    /// Represents a tab character, represented visually with a width of 4
    /// characters
    Tab(BytePos),
}

impl NonNarrowChar {
    fn new(pos: BytePos, width: usize) -> Self {
        match width {
            0 => NonNarrowChar::ZeroWidth(pos),
            2 => NonNarrowChar::Wide(pos),
            4 => NonNarrowChar::Tab(pos),
            _ => panic!("width {} given for non-narrow character", width),
        }
    }

    /// Returns the absolute offset of the character in the SourceMap
    pub fn pos(self) -> BytePos {
        match self {
            NonNarrowChar::ZeroWidth(p) | NonNarrowChar::Wide(p) | NonNarrowChar::Tab(p) => p,
        }
    }

    /// Returns the width of the character, 0 (zero-width) or 2 (wide)
    pub fn width(self) -> usize {
        match self {
            NonNarrowChar::ZeroWidth(_) => 0,
            NonNarrowChar::Wide(_) => 2,
            NonNarrowChar::Tab(_) => 4,
        }
    }
}

impl Add<BytePos> for NonNarrowChar {
    type Output = Self;

    fn add(self, rhs: BytePos) -> Self {
        match self {
            NonNarrowChar::ZeroWidth(pos) => NonNarrowChar::ZeroWidth(pos + rhs),
            NonNarrowChar::Wide(pos) => NonNarrowChar::Wide(pos + rhs),
            NonNarrowChar::Tab(pos) => NonNarrowChar::Tab(pos + rhs),
        }
    }
}

impl Sub<BytePos> for NonNarrowChar {
    type Output = Self;

    fn sub(self, rhs: BytePos) -> Self {
        match self {
            NonNarrowChar::ZeroWidth(pos) => NonNarrowChar::ZeroWidth(pos - rhs),
            NonNarrowChar::Wide(pos) => NonNarrowChar::Wide(pos - rhs),
            NonNarrowChar::Tab(pos) => NonNarrowChar::Tab(pos - rhs),
        }
    }
}

/// A single source in the SourceMap.
#[cfg_attr(
    feature = "rkyv",
    derive(rkyv::Archive, rkyv::Serialize, rkyv::Deserialize)
)]
#[derive(Clone)]
pub struct SourceFile {
    /// 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,
    /// True if the `name` field above has been modified by
    /// `--remap-path-prefix`
    pub name_was_remapped: bool,
    /// The unmapped path of the file that the source came from.
    /// Set to `None` if the `SourceFile` was imported from an external crate.
    pub unmapped_path: Option<FileName>,
    /// Indicates which crate this `SourceFile` was imported from.
    pub crate_of_origin: u32,
    /// The complete source code
    pub src: Lrc<String>,
    /// The source code's hash
    pub src_hash: u128,
    /// The start position of this source in the `SourceMap`
    pub start_pos: BytePos,
    /// The end position of this source in the `SourceMap`
    pub end_pos: BytePos,
    /// Locations of lines beginnings in the source code
    pub lines: Vec<BytePos>,
    /// Locations of multi-byte characters in the source code
    pub multibyte_chars: Vec<MultiByteChar>,
    /// Width of characters that are not narrow in the source code
    pub non_narrow_chars: Vec<NonNarrowChar>,
    /// A hash of the filename, used for speeding up the incr. comp. hashing.
    pub name_hash: u128,
}

impl fmt::Debug for SourceFile {
    fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
        write!(fmt, "SourceFile({})", self.name)
    }
}

impl SourceFile {
    pub fn new(
        name: FileName,
        name_was_remapped: bool,
        unmapped_path: FileName,
        mut src: String,
        start_pos: BytePos,
    ) -> SourceFile {
        debug_assert_ne!(
            start_pos,
            BytePos::DUMMY,
            "BytePos::DUMMY is reserved and `SourceFile` should not use it"
        );

        remove_bom(&mut src);

        let src_hash = {
            let mut hasher: StableHasher = StableHasher::new();
            hasher.write(src.as_bytes());
            hasher.finish()
        };
        let name_hash = {
            let mut hasher: StableHasher = StableHasher::new();
            name.hash(&mut hasher);
            hasher.finish()
        };
        let end_pos = start_pos.to_usize() + src.len();

        let (lines, multibyte_chars, non_narrow_chars) =
            analyze_source_file::analyze_source_file(&src[..], start_pos);

        SourceFile {
            name,
            name_was_remapped,
            unmapped_path: Some(unmapped_path),
            crate_of_origin: 0,
            src: Lrc::new(src),
            src_hash,
            start_pos,
            end_pos: Pos::from_usize(end_pos),
            lines,
            multibyte_chars,
            non_narrow_chars,
            name_hash,
        }
    }

    /// Return the BytePos of the beginning of the current line.
    pub fn line_begin_pos(&self, pos: BytePos) -> BytePos {
        let line_index = self.lookup_line(pos).unwrap();
        self.lines[line_index]
    }

    /// Get a line from the list of pre-computed line-beginnings.
    /// The line number here is 0-based.
    pub fn get_line(&self, line_number: usize) -> Option<Cow<'_, str>> {
        fn get_until_newline(src: &str, begin: usize) -> &str {
            // 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,
            }
        }

        let begin = {
            let line = self.lines.get(line_number)?;
            let begin: BytePos = *line - self.start_pos;
            begin.to_usize()
        };

        Some(Cow::from(get_until_newline(&self.src, begin)))
    }

    pub fn is_real_file(&self) -> bool {
        self.name.is_real()
    }

    pub fn byte_length(&self) -> u32 {
        self.end_pos.0 - self.start_pos.0
    }

    pub fn count_lines(&self) -> usize {
        self.lines.len()
    }

    /// Find the line containing the given position. The return value is the
    /// index into the `lines` array of this SourceFile, not the 1-based line
    /// number. If the `source_file` is empty or the position is located before
    /// the first line, `None` is returned.
    pub fn lookup_line(&self, pos: BytePos) -> Option<usize> {
        if self.lines.is_empty() {
            return None;
        }

        let line_index = lookup_line(&self.lines[..], pos);
        assert!(line_index < self.lines.len() as isize);
        if line_index >= 0 {
            Some(line_index as usize)
        } else {
            None
        }
    }

    pub fn line_bounds(&self, line_index: usize) -> (BytePos, BytePos) {
        if self.start_pos == self.end_pos {
            return (self.start_pos, self.end_pos);
        }

        assert!(line_index < self.lines.len());
        if line_index == (self.lines.len() - 1) {
            (self.lines[line_index], self.end_pos)
        } else {
            (self.lines[line_index], self.lines[line_index + 1])
        }
    }

    #[inline]
    pub fn contains(&self, byte_pos: BytePos) -> bool {
        byte_pos >= self.start_pos && byte_pos <= self.end_pos
    }
}

/// Remove utf-8 BOM if any.
fn remove_bom(src: &mut String) {
    if src.starts_with('\u{feff}') {
        src.drain(..3);
    }
}

// _____________________________________________________________________________
// Pos, BytePos, CharPos
//

pub trait Pos {
    fn from_usize(n: usize) -> Self;
    fn to_usize(&self) -> usize;
    fn from_u32(n: u32) -> Self;
    fn to_u32(&self) -> u32;
}

/// A byte offset. Keep this small (currently 32-bits), as AST contains
/// a lot of them.
///
///
/// # Reserved
///
///  - 0 is reserved for dummy spans. It means `BytePos(0)` means the `BytePos`
///    is synthesized by the compiler.
///
///  - Values larger than `u32::MAX - 2^16` are reserved for the comments.
///
/// `u32::MAX` is special value used to generate source map entries.
#[derive(Clone, Copy, PartialEq, Eq, Hash, PartialOrd, Ord, Debug, Serialize, Deserialize)]
#[serde(transparent)]
#[cfg_attr(feature = "arbitrary", derive(arbitrary::Arbitrary))]
#[cfg_attr(
    feature = "rkyv",
    derive(rkyv::Archive, rkyv::Serialize, rkyv::Deserialize)
)]
pub struct BytePos(#[cfg_attr(feature = "rkyv", omit_bounds)] pub u32);

impl BytePos {
    /// Dummy position. This is reserved for synthesized spans.
    pub const DUMMY: Self = BytePos(0);
    const MIN_RESERVED: Self = BytePos(DUMMY_RESERVE);

    pub const fn is_reserved_for_comments(self) -> bool {
        self.0 >= Self::MIN_RESERVED.0 && self.0 != u32::MAX
    }

    /// Returns true if this is synthesized and has no relevant input source
    /// code.
    pub const fn is_dummy(self) -> bool {
        self.0 == 0
    }
}

/// A character offset. Because of multibyte utf8 characters, a byte offset
/// is not equivalent to a character offset. The SourceMap will convert BytePos
/// values to CharPos values as necessary.
#[cfg_attr(
    feature = "rkyv",
    derive(rkyv::Archive, rkyv::Serialize, rkyv::Deserialize)
)]
#[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 {
    #[inline(always)]
    fn from_usize(n: usize) -> BytePos {
        BytePos(n as u32)
    }

    #[inline(always)]
    fn to_usize(&self) -> usize {
        self.0 as usize
    }

    #[inline(always)]
    fn from_u32(n: u32) -> BytePos {
        BytePos(n)
    }

    #[inline(always)]
    fn to_u32(&self) -> u32 {
        self.0
    }
}

impl Add for BytePos {
    type Output = BytePos;

    #[inline(always)]
    fn add(self, rhs: BytePos) -> BytePos {
        BytePos((self.to_usize() + rhs.to_usize()) as u32)
    }
}

impl Sub for BytePos {
    type Output = BytePos;

    #[inline(always)]
    fn sub(self, rhs: BytePos) -> BytePos {
        BytePos((self.to_usize() - rhs.to_usize()) as u32)
    }
}

impl Pos for CharPos {
    #[inline(always)]
    fn from_usize(n: usize) -> CharPos {
        CharPos(n)
    }

    #[inline(always)]
    fn to_usize(&self) -> usize {
        self.0
    }

    #[inline(always)]
    fn from_u32(n: u32) -> CharPos {
        CharPos(n as usize)
    }

    #[inline(always)]
    fn to_u32(&self) -> u32 {
        self.0 as u32
    }
}

impl Add for CharPos {
    type Output = CharPos;

    #[inline(always)]
    fn add(self, rhs: CharPos) -> CharPos {
        CharPos(self.to_usize() + rhs.to_usize())
    }
}

impl Sub for CharPos {
    type Output = CharPos;

    #[inline(always)]
    fn sub(self, rhs: CharPos) -> CharPos {
        CharPos(self.to_usize() - rhs.to_usize())
    }
}

// _____________________________________________________________________________
// Loc, LocWithOpt, SourceFileAndLine, SourceFileAndBytePos
//

/// A source code location used for error reporting
#[cfg_attr(
    feature = "rkyv",
    derive(rkyv::Archive, rkyv::Serialize, rkyv::Deserialize)
)]
#[derive(Debug, Clone)]
pub struct Loc {
    /// Information about the original source
    pub file: Lrc<SourceFile>,
    /// The (1-based) line number
    pub line: usize,
    /// The (0-based) column offset
    pub col: CharPos,
    /// The (0-based) column offset when displayed
    pub col_display: usize,
}

/// 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<Lrc<SourceFile>>,
}

// used to be structural records. Better names, anyone?
#[derive(Debug)]
pub struct SourceFileAndLine {
    pub sf: Lrc<SourceFile>,
    pub line: usize,
}
#[derive(Debug)]
pub struct SourceFileAndBytePos {
    pub sf: Lrc<SourceFile>,
    pub pos: BytePos,
}

#[derive(Copy, Clone, Debug, PartialEq, Eq)]
#[cfg_attr(
    feature = "rkyv",
    derive(rkyv::Archive, rkyv::Serialize, rkyv::Deserialize)
)]
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,
}

/// Used to create a `.map` file.
#[derive(Copy, Clone, Debug, PartialEq, Eq)]
pub struct LineCol {
    pub line: u32,
    pub col: u32,
}

#[cfg_attr(
    feature = "rkyv",
    derive(rkyv::Archive, rkyv::Serialize, rkyv::Deserialize)
)]
pub struct FileLines {
    pub file: Lrc<SourceFile>,
    pub lines: Vec<LineInfo>,
}

// _____________________________________________________________________________
// SpanLinesError, SpanSnippetError, DistinctSources,
// MalformedSourceMapPositions
//

pub type FileLinesResult = Result<FileLines, SpanLinesError>;

#[derive(Clone, PartialEq, Eq, Debug)]
#[cfg_attr(
    feature = "rkyv",
    derive(rkyv::Archive, rkyv::Serialize, rkyv::Deserialize)
)]
pub enum SpanLinesError {
    IllFormedSpan(Span),
    DistinctSources(DistinctSources),
}

#[derive(Clone, PartialEq, Eq, Debug)]
pub enum SpanSnippetError {
    DummyBytePos,
    IllFormedSpan(Span),
    DistinctSources(DistinctSources),
    MalformedForSourcemap(MalformedSourceMapPositions),
    SourceNotAvailable { filename: FileName },
}

#[derive(Clone, PartialEq, Eq, Debug)]
#[cfg_attr(
    feature = "rkyv",
    derive(rkyv::Archive, rkyv::Serialize, rkyv::Deserialize)
)]
pub struct DistinctSources {
    pub begin: (FileName, BytePos),
    pub end: (FileName, BytePos),
}

#[derive(Clone, PartialEq, Eq, Debug)]
pub struct MalformedSourceMapPositions {
    pub name: FileName,
    pub source_len: usize,
    pub begin_pos: BytePos,
    pub end_pos: BytePos,
}

// Given a slice of line start positions and a position, returns the index of
// the line the position is on. Returns -1 if the position is located before
// the first line.
fn lookup_line(lines: &[BytePos], pos: BytePos) -> isize {
    match lines.binary_search(&pos) {
        Ok(line) => line as isize,
        Err(line) => line as isize - 1,
    }
}

#[cfg(test)]
mod tests {
    use super::{lookup_line, BytePos, Span};

    #[test]
    fn test_lookup_line() {
        let lines = &[BytePos(3), BytePos(17), BytePos(28)];

        assert_eq!(lookup_line(lines, BytePos(0)), -1);
        assert_eq!(lookup_line(lines, BytePos(3)), 0);
        assert_eq!(lookup_line(lines, BytePos(4)), 0);

        assert_eq!(lookup_line(lines, BytePos(16)), 0);
        assert_eq!(lookup_line(lines, BytePos(17)), 1);
        assert_eq!(lookup_line(lines, BytePos(18)), 1);

        assert_eq!(lookup_line(lines, BytePos(28)), 2);
        assert_eq!(lookup_line(lines, BytePos(29)), 2);
    }

    #[test]
    fn size_of_span() {
        assert_eq!(std::mem::size_of::<Span>(), 12);
    }
}