mergiraf 0.19.0

A syntax-aware merge driver for Git
Documentation
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
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
use std::{cell::LazyCell, collections::HashMap, ops::Range};

use crate::{
    ast::AstNode,
    line_based::LINE_BASED_METHOD,
    matching::Matching,
    merge_result::MergeResult,
    pcs::Revision,
    settings::{ConflictRegexes, DisplaySettings},
};

pub(crate) const PARSED_MERGE_DIFF2_DETECTED: &str =
    "Mergiraf cannot solve conflicts displayed in the diff2 style";

/// A file which potentially contains merge conflicts, parsed as such.
#[derive(Debug, Clone, Eq, PartialEq, Hash)]
pub struct ParsedMerge<'a> {
    /// The actual contents of the parsed merge
    pub chunks: Vec<MergedChunk<'a>>,
    /// List of correspondences between sections of the reconstructed left revision and the merge output
    left: Vec<OffsetMap>,
    /// List of correspondences between sections of the reconstructed right revision and the merge output
    right: Vec<OffsetMap>,
    /// List of correspondences between sections of the reconstructed base revision and the merge output
    base: Vec<OffsetMap>,
}

/// A chunk in a file with merge conflicts: either a readily merged chunk or a conflict.
#[derive(Debug, Clone, Eq, PartialEq, Hash)]
pub enum MergedChunk<'a> {
    /// A readily-merged chunk
    Resolved {
        /// The byte offset at which this merged chunk can be found
        offset: usize,
        /// Its textual contents (including the last newline before any conflict)
        contents: &'a str,
    },
    /// A diff3-style conflict
    ///
    /// The diff3 format allows representing conflicts where some (or all) sides may have no final
    /// newline. We recognize this property, and preserve whatever newline was present in the original sides.
    Conflict {
        /// The left part of the conflict, with the final newline preserved (if present)
        left: Option<&'a str>,
        /// The base (or ancestor) part of the conflict, with the final newline preserved (if present)
        base: Option<&'a str>,
        /// The right part of the conflict, with the final newline preserved (if present)
        right: Option<&'a str>,
        /// The name of the left revision (potentially empty)
        left_name: Option<&'a str>,
        /// The name of the base revision (potentially empty)
        base_name: Option<&'a str>,
        /// The name of the right revision (potentially empty)
        right_name: Option<&'a str>,
    },
}

/// A correspondence between a section of a reconstructed revision and the merge output
#[derive(Debug, Clone, Eq, PartialEq, Hash)]
struct OffsetMap {
    /// The start of the section in the reconstructed revision
    rev_start: usize,
    /// The start of the section in the original merge output
    merged_start: usize,
    /// The common length of the section on both sides
    length: usize,
}

impl<'a> ParsedMerge<'a> {
    /// Parse a file into a series of chunks.
    /// Fails if the conflict markers do not appear in a consistent order.
    pub(crate) fn parse(source: &'a str, settings: &DisplaySettings) -> Result<Self, String> {
        let mut chunks = Vec::new();

        let ConflictRegexes {
            diff2: diff2conflict,
            diff3: diff3conflict,
            diff3_no_newline: diff3conflict_no_newline,
            ..
        } = settings.conflict_regexes();

        let mut remaining_source = source;
        while !remaining_source.is_empty() {
            let diff3_captures = LazyCell::new(|| diff3conflict.captures(remaining_source));
            let diff3_no_newline_captures =
                LazyCell::new(|| diff3conflict_no_newline.captures(remaining_source));

            // the 3 regexes each match more things than the last in this order:
            // 1) diff2            -- by ignoring the base marker and base rev text
            // 2) diff3_no_newline -- by, well, ignoring the final newline
            // 3) diff3            -- only matches a diff3 conflict ending with a newline
            //
            // so we run them in the opposite order:
            // 1) if diff3 matches, take that
            // 2) if diff3_no_newline matches, take that
            // 3) if diff2 matches, then we know that this isn't a misrecognized diff3, and bail out
            let resolved_end = if let Some(occurrence) =
                (diff3_captures.as_ref()).or_else(|| diff3_no_newline_captures.as_ref())
            {
                occurrence
                    .get(0)
                    .expect("whole match is guaranteed to exist")
                    .start()
            } else if diff2conflict.is_match(remaining_source) {
                return Err(PARSED_MERGE_DIFF2_DETECTED.to_owned());
            } else {
                remaining_source.len()
            };

            if resolved_end > 0 {
                // SAFETY: `remaining_source` is a suffix of `source`
                let offset =
                    unsafe { (remaining_source.as_ptr()).offset_from_unsigned(source.as_ptr()) };
                chunks.push(MergedChunk::Resolved {
                    offset,
                    contents: &remaining_source[..resolved_end],
                });
            }

            if let Some(captures) =
                (diff3_captures.as_ref()).or_else(|| diff3_no_newline_captures.as_ref())
            {
                chunks.push(MergedChunk::Conflict {
                    left_name: captures.get(1).map(|m| m.as_str()),
                    left: captures.get(2).map(|m| m.as_str()),
                    base_name: captures.get(3).map(|m| m.as_str()),
                    base: captures.get(4).map(|m| m.as_str()),
                    right: captures.get(5).map(|m| m.as_str()),
                    right_name: captures.get(6).map(|m| m.as_str()),
                });

                remaining_source = &remaining_source[captures
                    .get(0)
                    .expect("whole match is guaranteed to exist")
                    .end()..];
            } else {
                remaining_source = &remaining_source[resolved_end..];
            }
        }
        Ok(ParsedMerge::new(chunks))
    }

    /// Construct a parsed merge by indexing the provided chunks
    fn new(chunks: Vec<MergedChunk<'a>>) -> Self {
        let mut left_offset = 0;
        let mut base_offset = 0;
        let mut right_offset = 0;
        let mut left = Vec::new();
        let mut base = Vec::new();
        let mut right = Vec::new();
        for chunk in &chunks {
            match chunk {
                MergedChunk::Resolved { offset, contents } => {
                    let length = contents.len();
                    left.push(OffsetMap {
                        rev_start: left_offset,
                        merged_start: *offset,
                        length,
                    });
                    base.push(OffsetMap {
                        rev_start: base_offset,
                        merged_start: *offset,
                        length,
                    });
                    right.push(OffsetMap {
                        rev_start: right_offset,
                        merged_start: *offset,
                        length,
                    });
                    left_offset += length;
                    base_offset += length;
                    right_offset += length;
                }
                MergedChunk::Conflict {
                    left, base, right, ..
                } => {
                    left_offset += left.map_or(0, str::len);
                    base_offset += base.map_or(0, str::len);
                    right_offset += right.map_or(0, str::len);
                }
            }
        }
        ParsedMerge {
            chunks,
            left,
            right,
            base,
        }
    }

    /// Reconstruct the source of a revision based on the merged output.
    ///
    /// Because some changes from both revisions have likely already been
    /// merged in the non-conflicting sections, this is not the original revision,
    /// but rather a half-merged version of it.
    pub(crate) fn reconstruct_revision(&self, revision: Revision) -> String {
        self.chunks
            .iter()
            .map(|chunk| match *chunk {
                MergedChunk::Resolved { contents, .. } => contents,
                MergedChunk::Conflict {
                    left, base, right, ..
                } => match revision {
                    Revision::Base => base.unwrap_or_default(),
                    Revision::Left => left.unwrap_or_default(),
                    Revision::Right => right.unwrap_or_default(),
                },
            })
            .collect()
    }

    /// Find out at which index of the merged file a byte range in the reconstructed revision can be found.
    ///
    /// The returned index will only be returned if the entire range of the reconstructed
    /// revision lies in a fully merged part of the merged file (without overlapping any conflict).
    pub(crate) fn rev_range_to_merged_range(
        &self,
        range: Range<usize>,
        revision: Revision,
    ) -> Option<Range<usize>> {
        let length = range.end - range.start;
        let start = range.start;
        let offset_maps = match revision {
            Revision::Base => &self.base,
            Revision::Left => &self.left,
            Revision::Right => &self.right,
        };
        let matched_start = Self::binary_search(offset_maps, start, length)?;
        Some(matched_start..matched_start + length)
    }

    /// Generate a matching between the ASTs of two revisions generated by this parsed merge,
    /// by matching elements whenever they correspond to the same merged range.
    pub(crate) fn generate_matching<'b>(
        &self,
        first_revision: Revision,
        second_revision: Revision,
        first_tree: &'b AstNode<'b>,
        second_tree: &'b AstNode<'b>,
    ) -> Matching<'b> {
        let first_index = self.index_tree_by_merged_ranges(first_revision, first_tree);
        let second_index = self.index_tree_by_merged_ranges(second_revision, second_tree);
        let mut matching = Matching::new();
        let nodes = first_index.iter().filter_map(|(range, first_node)| {
            second_index
                .get(range)
                .map(|second_node| (*first_node, *second_node))
        });
        matching.extend(nodes);
        matching
    }

    fn index_tree_by_merged_ranges<'b>(
        &self,
        revision: Revision,
        tree: &'b AstNode<'b>,
    ) -> HashMap<(&'static str, Range<usize>), &'b AstNode<'b>> {
        let mut map = HashMap::new();
        self.recursively_index_node(revision, tree, &mut map);
        map
    }

    fn recursively_index_node<'b>(
        &self,
        revision: Revision,
        node: &'b AstNode<'b>,
        map: &mut HashMap<(&'static str, Range<usize>), &'b AstNode<'b>>,
    ) {
        match self.rev_range_to_merged_range(node.byte_range.clone(), revision) {
            Some(range) => {
                map.insert((node.kind, range), node);
            }
            None => {
                node.children
                    .iter()
                    .for_each(|child| self.recursively_index_node(revision, child, map));
            }
        };
    }

    /// Render the parsed merge back to a string representation
    pub(crate) fn render(&self, settings: &DisplaySettings) -> String {
        self.chunks.iter().fold(String::new(), |mut result, chunk| {
            match chunk {
                MergedChunk::Resolved { contents, .. } => result.push_str(contents),
                MergedChunk::Conflict {
                    left, base, right, ..
                } => {
                    // we check whether all 3 sides of the conflict[^1] used ot end with a newline.
                    // If any of them didn't, then the conflict should be rendered in a special way:
                    // - a newline is added to all three sides (even if the particular side used to
                    //   have a newline already)
                    // - *no* newline is added after the right marker, i.e. at the end of conflict
                    //
                    // [^1]: the ones that weren't empty, anyway
                    let add_after_right_marker = if let (None, None, None) = (base, left, right) {
                        unreachable!("wouldn't have been a conflict in the first place")
                    } else {
                        left.is_none_or(|l| l.ends_with('\n'))
                            && base.is_none_or(|b| b.ends_with('\n'))
                            && right.is_none_or(|r| r.ends_with('\n'))
                    };
                    let add_after_lines = !add_after_right_marker;

                    result.push_str(&settings.left_marker_or_default());
                    result.push('\n');
                    result.push_str(left.unwrap_or_default());
                    if add_after_lines {
                        result.push('\n');
                    }

                    if settings.diff3 {
                        result.push_str(&settings.base_marker_or_default());
                        result.push('\n');
                        result.push_str(base.unwrap_or_default());
                        if add_after_lines {
                            result.push('\n');
                        }
                    }

                    result.push_str(&settings.middle_marker_or_default());
                    result.push('\n');

                    result.push_str(right.unwrap_or_default());
                    if add_after_lines {
                        result.push('\n');
                    }
                    result.push_str(&settings.right_marker_or_default());
                    if add_after_right_marker {
                        result.push('\n');
                    }
                }
            }
            result
        })
    }

    /// If the parsed merge contains no conflicts, "render" it by concatenating all the chunks.
    /// Otherwise, return `None`.
    ///
    /// This is helpful when we want to compare the contents of a merge with some string, and we
    /// know that the latter doesn't contain any conflicts as well. An additional benefit is not
    /// requiring [`DisplaySettings`] to render, unlike [`Self::render`]
    pub(crate) fn render_conflictless(&self) -> Option<String> {
        self.chunks
            .iter()
            .map(|c| match c {
                MergedChunk::Resolved { contents, .. } => Some(*contents),
                MergedChunk::Conflict { .. } => None,
            })
            .collect()
    }

    fn binary_search(slice: &[OffsetMap], start: usize, length: usize) -> Option<usize> {
        let mut left = 0;
        let mut right = slice.len();
        while left < right {
            let guess = left.midpoint(right);
            let offset = slice
                .get(guess)
                .expect("Programming error in binary search, oops!");
            if offset.rev_start <= start && start + length <= offset.rev_start + offset.length {
                return Some(offset.merged_start + start - offset.rev_start);
            } else if left + 1 == right {
                break;
            }
            if offset.rev_start <= start {
                left = guess;
            }
            if offset.rev_start >= start {
                right = guess;
            }
        }
        None
    }

    /// Number of conflicts in this merge
    pub fn conflict_count(&self) -> usize {
        self.chunks
            .iter()
            .filter(|chunk| matches!(chunk, MergedChunk::Conflict { .. }))
            .count()
    }

    /// Number of bytes of conflicting content, which is an attempt
    /// at quantifying the effort it takes to resolve the conflicts.
    pub fn conflict_mass(&self) -> usize {
        self.chunks
            .iter()
            .map(|chunk| match chunk {
                MergedChunk::Resolved { .. } => 0,
                MergedChunk::Conflict {
                    base, left, right, ..
                } => {
                    base.map_or(0, str::len) + left.map_or(0, str::len) + right.map_or(0, str::len)
                }
            })
            .sum()
    }

    /// Whether the merge is empty when rendered
    pub(crate) fn is_empty(&self) -> bool {
        // NOTE: `.iter.all()` is trivially true for an empty `self.chunks`
        self.chunks.iter().all(|c| {
            // 1. if any chunk is a conflict, we'll need conflict markers => not empty
            // 2. if any resolved chunk is not empty, its render will be.. not empty as well
            matches!(c, MergedChunk::Resolved { contents: "", .. })
        })
    }

    /// Render into a merge result with the provided settings
    #[allow(clippy::wrong_self_convention)]
    pub(crate) fn into_merge_result(&self, settings: &DisplaySettings<'_>) -> MergeResult {
        MergeResult {
            contents: self.render(settings),
            conflict_count: self.conflict_count(),
            conflict_mass: self.conflict_mass(),
            method: LINE_BASED_METHOD,
            // the line-based merge might have come from a non-syntax-aware tool,
            // and we cautiously assume that it does have issues
            has_additional_issues: true,
        }
    }

    /// Attempt to extract OIDs from the first conflict's marker names (left, base, right).
    /// Returns (left_oid, base_oid, right_oid) if all are present and look like OIDs.
    pub(crate) fn extract_conflict_oids(&self) -> Option<(&str, &str, &str)> {
        fn is_oid(s: &&str) -> bool {
            s.len() == 40 && s.chars().all(|c| c.is_ascii_hexdigit())
        }
        self.chunks.iter().find_map(|chunk| {
            if let MergedChunk::Conflict {
                base_name,
                left_name,
                right_name,
                ..
            } = chunk
            {
                itertools::izip!(
                    base_name.filter(is_oid),
                    left_name.filter(is_oid),
                    right_name.filter(is_oid),
                )
                .next()
            } else {
                None
            }
        })
    }
}

#[cfg(test)]
mod tests {
    use std::borrow::Cow;

    use crate::test_utils::ctx;

    use super::*;

    #[track_caller]
    fn parse(source: &str) -> ParsedMerge<'_> {
        ParsedMerge::parse(source, &DisplaySettings::default()).expect("unexpected parse error")
    }

    #[test]
    fn it_works() {
        let source = "
we reached a junction.
<<<<<<< left
let's go to the left!
||||||| base
where should we go?
=======
turn right please!
>>>>>>>
rest of file
";
        let parsed = parse(source);

        let expected_parse = ParsedMerge::new(vec![
            MergedChunk::Resolved {
                offset: 0,
                contents: "\nwe reached a junction.\n",
            },
            MergedChunk::Conflict {
                left: Some("let's go to the left!\n"),
                base: Some("where should we go?\n"),
                right: Some("turn right please!\n"),
                left_name: Some("left"),
                base_name: Some("base"),
                right_name: None,
            },
            MergedChunk::Resolved {
                offset: 127,
                contents: "rest of file\n",
            },
        ]);

        assert_eq!(parsed, expected_parse);
        assert_eq!(
            parsed.reconstruct_revision(Revision::Base),
            "\nwe reached a junction.\nwhere should we go?\nrest of file\n"
        );
        assert_eq!(
            parsed.reconstruct_revision(Revision::Left),
            "\nwe reached a junction.\nlet's go to the left!\nrest of file\n"
        );
        assert_eq!(
            parsed.reconstruct_revision(Revision::Right),
            "\nwe reached a junction.\nturn right please!\nrest of file\n"
        );

        assert_eq!(
            parsed.rev_range_to_merged_range(1..11, Revision::Base),
            Some(1..11)
        );
        assert_eq!(
            parsed.rev_range_to_merged_range(1..11, Revision::Left),
            Some(1..11)
        );
        assert_eq!(
            parsed.rev_range_to_merged_range(1..11, Revision::Right),
            Some(1..11)
        );
        assert_eq!(
            parsed.rev_range_to_merged_range(11..41, Revision::Base),
            None
        );
        assert_eq!(
            parsed.rev_range_to_merged_range(11..41, Revision::Left),
            None
        );
        assert_eq!(
            parsed.rev_range_to_merged_range(11..41, Revision::Right),
            None
        );
        assert_eq!(
            parsed.rev_range_to_merged_range(25..28, Revision::Base),
            None
        );
        assert_eq!(
            parsed.rev_range_to_merged_range(25..28, Revision::Left),
            None
        );
        assert_eq!(
            parsed.rev_range_to_merged_range(25..28, Revision::Right),
            None
        );
        assert_eq!(
            parsed.rev_range_to_merged_range(45..49, Revision::Base),
            Some(128..132)
        );
        assert_eq!(
            parsed.rev_range_to_merged_range(47..49, Revision::Left),
            Some(128..130)
        );
        assert_eq!(
            parsed.rev_range_to_merged_range(45..48, Revision::Right),
            Some(129..132)
        );
        assert_eq!(
            parsed.rev_range_to_merged_range(180..183, Revision::Base),
            None
        );
        assert_eq!(
            parsed.rev_range_to_merged_range(190..193, Revision::Left),
            None
        );
        assert_eq!(
            parsed.rev_range_to_merged_range(200..203, Revision::Right),
            None
        );
    }

    mod parse {
        use super::*;

        #[test]
        fn start_with_conflict() {
            let source = "\
<<<<<<< left
let's go to the left!
||||||| base
where should we go?
=======
turn right please!
>>>>>>>
rest of file
";
            let parsed = parse(source);

            let expected_parse = ParsedMerge::new(vec![
                MergedChunk::Conflict {
                    left: Some("let's go to the left!\n"),
                    base: Some("where should we go?\n"),
                    right: Some("turn right please!\n"),
                    left_name: Some("left"),
                    base_name: Some("base"),
                    right_name: None,
                },
                MergedChunk::Resolved {
                    offset: 103,
                    contents: "rest of file\n",
                },
            ]);

            assert_eq!(parsed, expected_parse);
            assert_eq!(
                parsed.reconstruct_revision(Revision::Base),
                "where should we go?\nrest of file\n"
            );
            assert_eq!(
                parsed.reconstruct_revision(Revision::Left),
                "let's go to the left!\nrest of file\n"
            );
            assert_eq!(
                parsed.reconstruct_revision(Revision::Right),
                "turn right please!\nrest of file\n"
            );
        }

        #[test]
        fn end_with_conflict() {
            let source = "
we reached a junction.
<<<<<<< left
let's go to the left!
||||||| base
where should we go?
=======
turn right please!
>>>>>>>
";
            let parsed = parse(source);

            let expected_parse = ParsedMerge::new(vec![
                MergedChunk::Resolved {
                    offset: 0,
                    contents: "\nwe reached a junction.\n",
                },
                MergedChunk::Conflict {
                    left: Some("let's go to the left!\n"),
                    base: Some("where should we go?\n"),
                    right: Some("turn right please!\n"),
                    left_name: Some("left"),
                    base_name: Some("base"),
                    right_name: None,
                },
            ]);

            assert_eq!(parsed, expected_parse);
            assert_eq!(
                parsed.reconstruct_revision(Revision::Base),
                "\nwe reached a junction.\nwhere should we go?\n"
            );
            assert_eq!(
                parsed.reconstruct_revision(Revision::Left),
                "\nwe reached a junction.\nlet's go to the left!\n"
            );
            assert_eq!(
                parsed.reconstruct_revision(Revision::Right),
                "\nwe reached a junction.\nturn right please!\n"
            );
        }

        #[test]
        fn diff2() {
            let source = "\
my_struct_t instance = {
<<<<<<< LEFT
    .foo = 3,
    .bar = 2,
=======
>>>>>>> RIGHT
};
";

            let parse_err = ParsedMerge::parse(source, &DisplaySettings::default())
                .expect_err("expected a parse failure for diff2 conflicts");

            assert_eq!(parse_err, PARSED_MERGE_DIFF2_DETECTED);
        }

        #[test]
        fn non_standard_conflict_marker_size() {
            let parsed_expected = ParsedMerge::new(vec![
                MergedChunk::Resolved {
                    offset: 0,
                    contents: "resolved line\n",
                },
                MergedChunk::Conflict {
                    left: Some("left line\n"),
                    base: Some("base line\n"),
                    right: Some("right line\n"),
                    left_name: Some("LEFT"),
                    base_name: Some("BASE"),
                    right_name: Some("RIGHT"),
                },
            ]);

            let conflict_with_4 = "\
resolved line
<<<< LEFT
left line
|||| BASE
base line
====
right line
>>>> RIGHT
";
            let parsed_with_4 = ParsedMerge::parse(
                conflict_with_4,
                &DisplaySettings::from_conflict_marker_size(4),
            )
            .expect("could not parse a conflict with `conflict_marker_size=4`");
            assert_eq!(parsed_with_4, parsed_expected);

            let conflict_with_9 = "\
resolved line
<<<<<<<<< LEFT
left line
||||||||| BASE
base line
=========
right line
>>>>>>>>> RIGHT
";
            let parsed_with_9 = ParsedMerge::parse(
                conflict_with_9,
                &DisplaySettings::from_conflict_marker_size(9),
            )
            .expect("could not parse a conflict with `conflict_marker_size=9`");
            assert_eq!(parsed_with_9, parsed_expected);
        }

        #[test]
        fn left_marker_not_at_line_start() {
            let source = "\
my_struct_t instance = {
 <<<<<<< LIAR LEFT
    .foo = 3,
    .bar = 2,
||||||| BASE
    .foo = 3,
=======
>>>>>>> RIGHT
};
";
            let parsed = ParsedMerge::parse(source, &DisplaySettings::default())
                .expect("should just not see this conflict at all");

            let expected_parse = ParsedMerge::new(vec![MergedChunk::Resolved {
                offset: 0,
                contents: source,
            }]);

            assert_eq!(parsed, expected_parse);
        }

        #[test]
        fn base_marker_not_at_line_start() {
            let source = "\
my_struct_t instance = {
<<<<<<< LEFT
    .foo = 3,
    .bar = 2,
 ||||||| LIAR BASE
    .foo = 3,
=======
>>>>>>> RIGHT
};
";
            let parse_err = ParsedMerge::parse(source, &DisplaySettings::default()).expect_err(
                "because of the missing base marker, this should like a diff2-style conflict",
            );

            assert_eq!(parse_err, PARSED_MERGE_DIFF2_DETECTED);
        }

        #[test]
        fn middle_marker_not_at_line_start() {
            let source = "\
my_struct_t instance = {
<<<<<<< LEFT
    .foo = 3,
    .bar = 2,
||||||| BASE
    .foo = 3,
 =======
>>>>>>> RIGHT
};
";
            let parsed = ParsedMerge::parse(source, &DisplaySettings::default())
                .expect("should ignore the malformed conflict");

            let expected = ParsedMerge::new(vec![MergedChunk::Resolved {
                offset: 0,
                contents: source,
            }]);

            assert_eq!(parsed, expected);
        }

        #[test]
        fn right_marker_not_at_line_start() {
            let source = "\
my_struct_t instance = {
<<<<<<< LEFT
    .foo = 3,
    .bar = 2,
||||||| BASE
    .foo = 3,
=======
 >>>>>>> LIAR RIGHT
};
";
            let parsed = ParsedMerge::parse(source, &DisplaySettings::default())
                .expect("should ignore the malformed conflict");

            let expected = ParsedMerge::new(vec![MergedChunk::Resolved {
                offset: 0,
                contents: source,
            }]);

            assert_eq!(parsed, expected);
        }

        #[test]
        fn diff3_then_diff3_is_lazy() {
            let source = "\
<<<<<<< LEFT
// a comment
||||||| BASE
=======
// hi
>>>>>>> RIGHT
<<<<<<< LEFT
use bytes;
||||||| BASE
use io;
=======
use os;
>>>>>>> RIGHT
";

            let parsed = parse(source);

            let unwanted_non_lazy = ParsedMerge::new(vec![MergedChunk::Conflict {
                left_name: Some("LEFT"),
                left: Some("// a comment\n"),
                base_name: Some("BASE"),
                base: Some(
                    "=======\n// hi\n>>>>>>> RIGHT\n<<<<<<< LEFT\nuse bytes;\n||||||| BASE\nuse io;\n",
                ),
                right: Some("use os;\n"),
                right_name: Some("RIGHT"),
            }]);

            assert_ne!(
                parsed, unwanted_non_lazy,
                "the regex is greedy -- it should've stopped after the first 'RIGHT'!"
            );

            let expected = ParsedMerge::new(vec![
                MergedChunk::Conflict {
                    left_name: Some("LEFT"),
                    left: Some("// a comment\n"),
                    base_name: Some("BASE"),
                    base: None,
                    right: Some("// hi\n"),
                    right_name: Some("RIGHT"),
                },
                MergedChunk::Conflict {
                    left_name: Some("LEFT"),
                    left: Some("use bytes;\n"),
                    base_name: Some("BASE"),
                    base: Some("use io;\n"),
                    right: Some("use os;\n"),
                    right_name: Some("RIGHT"),
                },
            ]);

            assert_eq!(parsed, expected);
        }

        #[test]
        fn diff3_then_diff3_wo_newline() {
            let source = "\
<<<<<<< LEFT
// a comment
||||||| BASE
=======
// hi
>>>>>>> RIGHT
<<<<<<< LEFT
use bytes;
||||||| BASE
use io;
=======
use os;
>>>>>>> RIGHT";

            let parsed = parse(source);

            let expected = ParsedMerge::new(vec![
                MergedChunk::Conflict {
                    left_name: Some("LEFT"),
                    left: Some("// a comment\n"),
                    base_name: Some("BASE"),
                    base: None,
                    right: Some("// hi\n"),
                    right_name: Some("RIGHT"),
                },
                MergedChunk::Conflict {
                    left_name: Some("LEFT"),
                    left: Some("use bytes;"),
                    base_name: Some("BASE"),
                    base: Some("use io;"),
                    right: Some("use os;"),
                    right_name: Some("RIGHT"),
                },
            ]);

            assert_eq!(parsed, expected);
        }

        #[test]
        fn diff3_is_with_final_newline_when_possible() {
            let source = "\
<<<<<<< left
let's go to the left!
||||||| base
where should we go?
=======
turn right please!
>>>>>>>
";

            let parsed = parse(source);

            let unwanted_wo_final_newline = ParsedMerge::new(vec![
                MergedChunk::Conflict {
                    left_name: Some("left"),
                    left: Some("let's go to the left!"),
                    base_name: Some("base"),
                    base: Some("where should we go?"),
                    right: Some("turn right please!"),
                    right_name: None,
                },
                MergedChunk::Resolved {
                    offset: 102,
                    contents: "\n",
                },
            ]);

            assert_ne!(parsed, unwanted_wo_final_newline);
        }
    }

    mod render {
        use super::*;
        #[test]
        fn non_standard_conflict_marker_size() {
            let merge = ParsedMerge::new(vec![
                MergedChunk::Resolved {
                    offset: 0,
                    contents: "resolved line\n",
                },
                MergedChunk::Conflict {
                    left_name: None,
                    left: Some("left line\n"),
                    base: Some("base line\n"),
                    right: Some("right line\n"),
                    right_name: None,
                    base_name: None,
                },
            ]);

            let rendered_with_4 = merge.render(&DisplaySettings::from_conflict_marker_size(4));
            let expected_with_4 = "\
resolved line
<<<< LEFT
left line
|||| BASE
base line
====
right line
>>>> RIGHT
";
            assert_eq!(rendered_with_4, expected_with_4);

            let rendered_with_9 = merge.render(&DisplaySettings::from_conflict_marker_size(9));
            let expected_with_9 = "\
resolved line
<<<<<<<<< LEFT
left line
||||||||| BASE
base line
=========
right line
>>>>>>>>> RIGHT
";
            assert_eq!(rendered_with_9, expected_with_9);
        }

        #[test]
        fn no_final_newline() {
            // meanings of the used shortenings:
            // - wo              - without final newline
            // - w               - with final newline
            // - expected_w_wo_w - expected from base_w, left_wo, right_w

            let base_wo = "base";
            let base_w = "base\n";

            let left_wo = "left";
            let left_w = "left\n";

            let right_wo = "right";
            let right_w = "right\n";

            fn chunk(base: &str, left: &str, right: &str) -> String {
                ParsedMerge::new(vec![MergedChunk::Conflict {
                    left: Some(left),
                    base: Some(base),
                    right: Some(right),
                    left_name: None,
                    base_name: None,
                    right_name: None,
                }])
                .render(&DisplaySettings::default())
            }

            let expected_wo_wo_wo = "\
<<<<<<< LEFT
left
||||||| BASE
base
=======
right
>>>>>>> RIGHT";
            let rendered = chunk(base_wo, left_wo, right_wo);
            assert_eq!(rendered, expected_wo_wo_wo);

            let expected_wo_w_wo = "\
<<<<<<< LEFT
left

||||||| BASE
base
=======
right
>>>>>>> RIGHT";
            let rendered = chunk(base_wo, left_w, right_wo);
            assert_eq!(rendered, expected_wo_w_wo);

            // wo_wo_w case should be symmetrical to wo_w_wo

            let expected_wo_w_w = "\
<<<<<<< LEFT
left

||||||| BASE
base
=======
right

>>>>>>> RIGHT";
            let rendered = chunk(base_wo, left_w, right_w);
            assert_eq!(rendered, expected_wo_w_w);

            let expected_w_wo_wo = "\
<<<<<<< LEFT
left
||||||| BASE
base

=======
right
>>>>>>> RIGHT";
            let rendered = chunk(base_w, left_wo, right_wo);
            assert_eq!(rendered, expected_w_wo_wo);

            let expected_w_w_wo = "\
<<<<<<< LEFT
left

||||||| BASE
base

=======
right
>>>>>>> RIGHT";
            let rendered_w_w_wo = chunk(base_w, left_w, right_wo);
            assert_eq!(rendered_w_w_wo, expected_w_w_wo);

            // w_wo_w should be symmetrical to w_w_wo
        }
    }

    #[test]
    fn parse_then_render_is_identity() {
        let source = "\
my_struct_t instance = {
<<<<<<< LEFT
    .foo = 3,
    .bar = 2,
||||||| BASE
    .foo = 3,
=======
>>>>>>> RIGHT
};
";

        let parsed = parse(source);

        let expected_parse = ParsedMerge::new(vec![
            MergedChunk::Resolved {
                offset: 0,
                contents: "my_struct_t instance = {\n",
            },
            MergedChunk::Conflict {
                left: Some("    .foo = 3,\n    .bar = 2,\n"),
                base: Some("    .foo = 3,\n"),
                right: None,
                left_name: Some("LEFT"),
                base_name: Some("BASE"),
                right_name: Some("RIGHT"),
            },
            MergedChunk::Resolved {
                offset: 115,
                contents: "};\n",
            },
        ]);

        assert_eq!(parsed, expected_parse);
        assert_eq!(parsed.conflict_count(), 1);
        assert_eq!(parsed.conflict_mass(), 42);

        // render the parsed conflict and check it's equal to the source
        let rendered = parsed.render(&DisplaySettings::default());

        assert_eq!(rendered, source);
    }

    mod matching {
        use super::*;

        #[test]
        fn it_works() {
            let ctx = ctx();
            let source = "\
struct MyType {
    field: bool,
<<<<<<< LEFT
    foo: int,
    bar: String,
||||||| BASE
    foo: String,
=======
>>>>>>> RIGHT
};
";

            let parsed = parse(source);

            let left_rev = parsed.reconstruct_revision(Revision::Left);
            let right_rev = parsed.reconstruct_revision(Revision::Right);

            let parsed_left = ctx.parse("a.rs", &left_rev);
            let parsed_right = ctx.parse("a.rs", &right_rev);

            let matching = parsed.generate_matching(
                Revision::Left,
                Revision::Right,
                parsed_left,
                parsed_right,
            );

            let mytype_left = parsed_left[0][1];
            let mytype_right = parsed_right[0][1];
            let closing_bracket_left = parsed_left[0][2][7];
            let closing_bracket_right = parsed_right[0][2][3];

            assert!(matching.are_matched(mytype_left, mytype_right));
            assert!(matching.are_matched(closing_bracket_left, closing_bracket_right));

            assert_eq!(matching.len(), 7);
        }

        #[test]
        fn identical_ranges_but_different_kinds() {
            let ctx = ctx();
            let source = "\
{
}:

{
  foo.bar = \"Hello World\";
<<<<<<< LEFT
  foo.baz = \"Mergiraf is fun :)\";
||||||| BASE
=======
  foo.foo = \"Test\";
>>>>>>> RIGHT
}
";

            let parsed = parse(source);

            let base_rev = parsed.reconstruct_revision(Revision::Base);
            let left_rev = parsed.reconstruct_revision(Revision::Left);

            let parsed_base = ctx.parse("a.nix", &base_rev);
            let parsed_left = ctx.parse("a.nix", &left_rev);

            let binding_set_base = parsed_base[0][2][1];
            assert_eq!(binding_set_base.kind, "binding_set");
            let binding_left = parsed_left[0][2][1][0];
            assert_eq!(binding_left.kind, "binding");
            // two nodes of different types have the same range
            assert_eq!(binding_set_base.byte_range, binding_left.byte_range);

            let matching =
                parsed.generate_matching(Revision::Base, Revision::Left, parsed_base, parsed_left);

            // the two nodes are not matched despite having the same range
            assert!(matching.get_from_left(binding_set_base).is_none());
            assert!(matching.get_from_right(binding_left).is_none());
        }
    }

    mod add_revision_names {
        use super::*;

        #[test]
        fn it_works() {
            let source = "\
<<<<<<< my_left
let's go to the left!
||||||| my_base
where should we go?
=======
turn right please!
>>>>>>> my_right
rest of file
";
            let parsed = parse(source);

            let initial_settings = DisplaySettings::default();

            let mut enriched_settings = initial_settings.clone();
            enriched_settings.add_revision_names(&parsed);

            let manually_enriched_settings = {
                let mut settings = initial_settings;
                settings.left_revision_name = Some(Cow::Borrowed("my_left"));
                settings.base_revision_name = Some(Cow::Borrowed("my_base"));
                settings.right_revision_name = Some(Cow::Borrowed("my_right"));
                settings
            };

            assert_eq!(enriched_settings, manually_enriched_settings);
        }

        #[test]
        fn no_names() {
            let source = "\
<<<<<<<
let's go to the left!
|||||||
where should we go?
=======
turn right please!
>>>>>>>
rest of file
";
            let parsed = parse(source);

            let initial_settings = DisplaySettings::default();

            let mut enriched_settings = initial_settings.clone();
            enriched_settings.add_revision_names(&parsed);

            assert_eq!(enriched_settings, initial_settings);
        }

        #[test]
        fn no_conflict() {
            let source = "\
start of file
rest of file
";
            let parsed = parse(source);

            let initial_settings = DisplaySettings::default();

            let mut enriched_settings = initial_settings.clone();
            enriched_settings.add_revision_names(&parsed);

            assert_eq!(enriched_settings, initial_settings);
        }
    }

    #[test]
    fn is_empty() {
        const fn resolved(contents: &str) -> MergedChunk<'_> {
            MergedChunk::Resolved {
                contents,
                offset: 0,
            }
        }

        const fn conflict<'a>(
            base: Option<&'a str>,
            left: Option<&'a str>,
            right: Option<&'a str>,
        ) -> MergedChunk<'a> {
            MergedChunk::Conflict {
                left,
                base,
                right,
                left_name: None,
                base_name: None,
                right_name: None,
            }
        }

        #[track_caller]
        fn is<const N: usize>(chunks: [MergedChunk<'_>; N]) {
            assert!(ParsedMerge::new(chunks.into()).is_empty())
        }

        #[track_caller]
        fn is_not<const N: usize>(chunks: [MergedChunk<'_>; N]) {
            assert!(!ParsedMerge::new(chunks.into()).is_empty())
        }

        is([]);
        is([resolved("")]);
        is([resolved(""), resolved("")]);
        is_not([resolved("hello")]);

        is_not([conflict(None, None, None)]);
        is_not([conflict(Some(""), None, None)]);
        is_not([conflict(Some("hello"), None, None)]);

        is_not([conflict(None, None, None)]);
        is_not([conflict(None, Some(""), None)]);
        is_not([conflict(None, Some("hello"), None)]);

        is_not([conflict(None, None, None)]);
        is_not([conflict(None, None, Some(""))]);
        is_not([conflict(None, None, Some("hello"))]);
    }
}