destructive_command_guard 0.4.3

A Claude Code hook that blocks destructive commands before they execute
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
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
//! Test helper utilities for pack unit testing.
//!
//! This module provides reusable assertion functions and utilities for testing
//! pack patterns. Use these helpers to ensure consistent test structure and
//! informative failure messages across all pack tests.
//!
//! # Usage
//!
//! ```rust,ignore
//! use crate::packs::test_helpers::*;
//!
//! #[test]
//! fn test_my_pack() {
//!     let pack = my_pack::create_pack();
//!
//!     // Test destructive patterns block with expected reasons
//!     assert_blocks(&pack, "dangerous-command", "expected reason substring");
//!
//!     // Test safe patterns allow commands
//!     assert_allows(&pack, "safe-command");
//!
//!     // Test unrelated commands are not matched
//!     assert_no_match(&pack, "unrelated-command");
//! }
//! ```

use crate::packs::{Pack, Severity};
use std::fmt::Write;
use std::time::{Duration, Instant};

/// Maximum time allowed for a single pattern match operation.
///
/// Pattern matching should be sub-millisecond for typical commands.
/// We use 500ms to accommodate coverage instrumentation overhead while
/// still catching catastrophic regex backtracking (which would be 10s+).
pub const PATTERN_MATCH_TIMEOUT: Duration = Duration::from_millis(500);

/// Assert that a pack blocks a command with a reason containing the expected substring.
///
/// # Panics
///
/// Panics if:
/// - The pack does not block the command
/// - The block reason does not contain `expected_reason_substring`
///
/// # Example
///
/// ```rust,ignore
/// assert_blocks(&pack, "git reset --hard", "destroys uncommitted changes");
/// ```
#[track_caller]
pub fn assert_blocks(pack: &Pack, command: &str, expected_reason_substring: &str) {
    let result = pack.check(command);

    match result {
        Some(matched) => {
            assert!(
                matched.reason.contains(expected_reason_substring),
                "Command '{}' was blocked but with unexpected reason.\n\
                 Expected reason to contain: '{}'\n\
                 Actual reason: '{}'",
                command,
                expected_reason_substring,
                matched.reason
            );
        }
        None => {
            panic!(
                "Expected pack '{}' to block command '{}' but it was allowed.\n\
                 Pack has {} safe patterns and {} destructive patterns.\n\
                 Keywords: {:?}",
                pack.id,
                command,
                pack.safe_patterns.len(),
                pack.destructive_patterns.len(),
                pack.keywords
            );
        }
    }
}

/// Assert that a pack blocks a command with the specified pattern name.
///
/// This is useful for testing that a specific pattern matches rather than
/// just any pattern. Pattern names are used for allowlisting.
///
/// # Panics
///
/// Panics if:
/// - The pack does not block the command
/// - The pattern that matched does not have the expected name
#[track_caller]
pub fn assert_blocks_with_pattern(pack: &Pack, command: &str, expected_pattern_name: &str) {
    let result = pack.check(command);

    match result {
        Some(matched) => match matched.name {
            Some(name) => {
                assert_eq!(
                    name, expected_pattern_name,
                    "Command '{command}' was blocked by pattern '{name}' but expected '{expected_pattern_name}'"
                );
            }
            None => {
                panic!(
                    "Command '{}' was blocked but by an unnamed pattern.\n\
                         Expected pattern name: '{}'\n\
                         Reason: '{}'",
                    command, expected_pattern_name, matched.reason
                );
            }
        },
        None => {
            panic!(
                "Expected pack '{}' to block command '{}' with pattern '{}' but it was allowed",
                pack.id, command, expected_pattern_name
            );
        }
    }
}

/// Assert that a pack blocks a command with the specified severity level.
///
/// Use this to verify that Critical, High, Medium, and Low severity patterns
/// are correctly classified.
///
/// # Panics
///
/// Panics if:
/// - The pack does not block the command
/// - The matched pattern does not have the expected severity
#[track_caller]
pub fn assert_blocks_with_severity(pack: &Pack, command: &str, expected_severity: Severity) {
    let result = pack.check(command);

    match result {
        Some(matched) => {
            assert_eq!(
                matched.severity, expected_severity,
                "Command '{}' was blocked with severity {:?} but expected {:?}.\n\
                 Pattern: {:?}\n\
                 Reason: '{}'",
                command, matched.severity, expected_severity, matched.name, matched.reason
            );
        }
        None => {
            panic!(
                "Expected pack '{}' to block command '{}' with severity {:?} but it was allowed",
                pack.id, command, expected_severity
            );
        }
    }
}

/// Assert that a pack allows a command (no destructive pattern matches).
///
/// This can mean either:
/// - A safe pattern explicitly allows the command, OR
/// - No patterns match at all
///
/// # Panics
///
/// Panics if the pack blocks the command.
#[track_caller]
pub fn assert_allows(pack: &Pack, command: &str) {
    let result = pack.check(command);

    if let Some(matched) = result {
        panic!(
            "Expected pack '{}' to allow command '{}' but it was blocked.\n\
             Pattern: {:?}\n\
             Reason: '{}'\n\
             Severity: {:?}",
            pack.id, command, matched.name, matched.reason, matched.severity
        );
    }
}

/// Assert that a safe pattern explicitly matches a command.
///
/// This is stricter than `assert_allows` - it verifies that a safe pattern
/// actually matches, not just that no destructive pattern matched.
///
/// # Panics
///
/// Panics if no safe pattern matches the command.
#[track_caller]
pub fn assert_safe_pattern_matches(pack: &Pack, command: &str) {
    assert!(
        pack.matches_safe(command),
        "Expected a safe pattern in pack '{}' to match command '{}' but none did.\n\
         Safe patterns ({}):\n{}",
        pack.id,
        command,
        pack.safe_patterns.len(),
        pack.safe_patterns
            .iter()
            .map(|p| format!("  - {}", p.name))
            .collect::<Vec<_>>()
            .join("\n")
    );
}

/// Assert that no safe pattern matches the command.
///
/// Use this to verify that a command is NOT allowlisted.
///
/// # Panics
///
/// Panics if a safe pattern matches the command.
#[track_caller]
pub fn assert_no_safe_match(pack: &Pack, command: &str) {
    if pack.matches_safe(command) {
        let matched = pack
            .safe_patterns
            .iter()
            .find(|p| p.regex.is_match(command));

        panic!(
            "Expected no safe patterns in pack '{}' to match command '{}' but one did.\n\
             Matched safe pattern: {:?}",
            pack.id,
            command,
            matched.map(|p| p.name)
        );
    }
}

/// Assert that no pattern in the pack matches the command.
///
/// Use this to verify specificity - that patterns don't accidentally match
/// unrelated commands due to overly broad regexes.
///
/// # Panics
///
/// Panics if any pattern (safe or destructive) matches the command.
#[track_caller]
pub fn assert_no_match(pack: &Pack, command: &str) {
    // Check safe patterns
    if pack.matches_safe(command) {
        let matched_safe = pack
            .safe_patterns
            .iter()
            .find(|p| p.regex.is_match(command));

        panic!(
            "Expected no patterns in pack '{}' to match command '{}' but safe pattern matched.\n\
             Matched safe pattern: {:?}",
            pack.id,
            command,
            matched_safe.map(|p| p.name)
        );
    }

    // Check destructive patterns
    if let Some(matched) = pack.matches_destructive(command) {
        panic!(
            "Expected no patterns in pack '{}' to match command '{}' but destructive pattern matched.\n\
             Pattern: {:?}\n\
             Reason: '{}'",
            pack.id, command, matched.name, matched.reason
        );
    }
}

/// Assert that a pattern matches within the allowed time budget.
///
/// Pattern matching should be fast. This helper ensures regex patterns
/// don't have catastrophic backtracking or performance issues.
///
/// **Note:** This function pre-warms the pack's patterns before measuring,
/// so that lazy regex compilation doesn't affect the timing. The goal is
/// to measure pattern matching performance, not compilation time.
///
/// # Panics
///
/// Panics if pattern matching takes longer than `PATTERN_MATCH_TIMEOUT`.
#[track_caller]
pub fn assert_matches_within_budget(pack: &Pack, command: &str) {
    // Pre-warm: trigger lazy compilation of all patterns by running a check.
    // This ensures we measure actual matching time, not compilation time.
    let _ = pack.check("__warmup__");

    let start = Instant::now();
    let _ = pack.check(command);
    let elapsed = start.elapsed();

    assert!(
        elapsed < PATTERN_MATCH_TIMEOUT,
        "Pattern matching for command '{}' in pack '{}' took {:?}, exceeding budget of {:?}.\n\
         This may indicate catastrophic regex backtracking.",
        command,
        pack.id,
        elapsed,
        PATTERN_MATCH_TIMEOUT
    );
}

/// Test a batch of commands that should all be blocked.
///
/// Returns a summary of results for debugging.
///
/// # Panics
///
/// Panics if any command in the batch is not blocked or has an unexpected reason.
///
/// # Example
///
/// ```rust,ignore
/// let commands = vec![
///     "git reset --hard",
///     "git reset --hard HEAD",
///     "git reset --hard HEAD~1",
/// ];
/// test_batch_blocks(&pack, &commands, "reset");
/// ```
#[track_caller]
pub fn test_batch_blocks(pack: &Pack, commands: &[&str], reason_substring: &str) {
    let mut failures = Vec::new();

    for cmd in commands {
        let result = pack.check(cmd);
        match result {
            Some(matched) => {
                if !matched.reason.contains(reason_substring) {
                    failures.push(format!(
                        "  '{cmd}': blocked but reason '{}' doesn't contain '{reason_substring}'",
                        matched.reason
                    ));
                }
            }
            None => {
                failures.push(format!("  '{cmd}': allowed (should be blocked)"));
            }
        }
    }

    assert!(
        failures.is_empty(),
        "Batch block test failed for pack '{}':\n{}",
        pack.id,
        failures.join("\n")
    );
}

/// Test a batch of commands that should all be allowed.
///
/// # Panics
///
/// Panics if any command in the batch is blocked.
///
/// # Example
///
/// ```rust,ignore
/// let commands = vec![
///     "git status",
///     "git log",
///     "git diff",
/// ];
/// test_batch_allows(&pack, &commands);
/// ```
#[track_caller]
pub fn test_batch_allows(pack: &Pack, commands: &[&str]) {
    let mut failures = Vec::new();

    for cmd in commands {
        if let Some(matched) = pack.check(cmd) {
            failures.push(format!(
                "  '{cmd}': blocked by {:?} - '{}'",
                matched.name, matched.reason
            ));
        }
    }

    assert!(
        failures.is_empty(),
        "Batch allow test failed for pack '{}':\n{}",
        pack.id,
        failures.join("\n")
    );
}

/// Get detailed match information for debugging.
///
/// This is useful when writing tests to understand why a pattern did or
/// didn't match.
#[must_use]
pub fn debug_match_info(pack: &Pack, command: &str) -> String {
    let mut info = format!("Match info for '{command}' in pack '{}':\n", pack.id);

    // Check keyword matching
    let might_match = pack.might_match(command);
    let _ = writeln!(
        info,
        "  Keywords ({:?}): {}",
        pack.keywords,
        if might_match {
            "MAY match"
        } else {
            "quick-rejected"
        }
    );

    if !might_match {
        return info;
    }

    // Check safe patterns
    info.push_str("  Safe patterns:\n");
    for pattern in &pack.safe_patterns {
        let matches = pattern.regex.is_match(command);
        let _ = writeln!(
            info,
            "    - {}: {}",
            pattern.name,
            if matches { "MATCH" } else { "no match" }
        );
    }

    // Check destructive patterns
    info.push_str("  Destructive patterns:\n");
    for pattern in &pack.destructive_patterns {
        let matches = pattern.regex.is_match(command);
        let _ = writeln!(
            info,
            "    - {:?}: {} (severity: {:?})",
            pattern.name,
            if matches { "MATCH" } else { "no match" },
            pattern.severity
        );
    }

    info
}

/// Validate a pack's definition.
///
/// Checks for:
/// - Empty ID/Name/Description
/// - Empty keywords (must have at least one)
/// - Empty pattern lists (must have at least one safe or destructive pattern)
/// - Keywords are valid
///
/// # Panics
///
/// Panics if any validation check fails.
pub fn validate_pack(pack: &Pack) {
    assert!(!pack.id.is_empty(), "Pack ID must not be empty");
    assert!(
        pack.id
            .chars()
            .all(|c| c.is_ascii_lowercase() || c == '.' || c == '_' || c.is_numeric()),
        "Pack ID '{}' should be lowercase with dots, underscores, or digits only",
        pack.id
    );

    // Required fields
    assert!(
        !pack.name.is_empty(),
        "Pack '{}' name must not be empty",
        pack.id
    );
    assert!(
        !pack.description.is_empty(),
        "Pack '{}' description must not be empty",
        pack.id
    );
    assert!(
        !pack.keywords.is_empty(),
        "Pack '{}' must have at least one keyword",
        pack.id
    );

    // Pattern validation
    assert_patterns_compile(pack);
    assert_all_patterns_have_reasons(pack);
    assert_unique_pattern_names(pack);
}

/// Verify that all patterns in a pack compile successfully.
///
/// This is a sanity check to ensure no regex syntax errors exist.
#[track_caller]
pub fn assert_patterns_compile(pack: &Pack) {
    // Safe patterns
    for pattern in &pack.safe_patterns {
        // Just accessing the regex is enough - it's compiled at pack creation
        let _ = pattern.regex.as_str();
    }

    // Destructive patterns
    for pattern in &pack.destructive_patterns {
        let _ = pattern.regex.as_str();
    }
}

/// Verify that all destructive patterns have non-empty reasons.
///
/// # Panics
///
/// Panics if any destructive pattern has an empty reason string.
#[track_caller]
pub fn assert_all_patterns_have_reasons(pack: &Pack) {
    for pattern in &pack.destructive_patterns {
        assert!(
            !pattern.reason.is_empty(),
            "Destructive pattern {:?} in pack '{}' has empty reason",
            pattern.name,
            pack.id
        );
    }
}

/// Verify that all named patterns have unique names within the pack.
///
/// # Panics
///
/// Panics if any two patterns (safe or destructive) share the same name.
#[track_caller]
pub fn assert_unique_pattern_names(pack: &Pack) {
    let mut names = std::collections::HashSet::new();

    // Check safe patterns
    for pattern in &pack.safe_patterns {
        assert!(
            names.insert(pattern.name),
            "Duplicate safe pattern name '{}' in pack '{}'",
            pattern.name,
            pack.id
        );
    }

    // Check destructive patterns
    for pattern in &pack.destructive_patterns {
        if let Some(name) = pattern.name {
            assert!(
                names.insert(name),
                "Duplicate destructive pattern name '{}' in pack '{}'",
                name,
                pack.id
            );
        }
    }
}

// ============================================================================
// Logging Integration
// ============================================================================

use crate::logging::{PackTestLogConfig, PackTestLogger};

/// A test runner that integrates with `PackTestLogger` for structured output.
///
/// This allows running pack tests with detailed logging and generating
/// JSON reports for CI/CD integration.
///
/// # Example
///
/// ```rust,ignore
/// let mut runner = LoggedPackTestRunner::new(&pack, PackTestLogConfig::default());
/// runner.assert_blocks("git reset --hard", "destroys uncommitted");
/// runner.assert_allows("git status");
/// let report = runner.finish();
/// println!("{}", report);
/// ```
pub struct LoggedPackTestRunner<'a> {
    pack: &'a Pack,
    logger: PackTestLogger,
}

impl<'a> LoggedPackTestRunner<'a> {
    /// Create a new test runner for a pack.
    #[must_use]
    pub fn new(pack: &'a Pack, config: PackTestLogConfig) -> Self {
        Self {
            pack,
            logger: PackTestLogger::new(&pack.id, &config),
        }
    }

    /// Create a test runner with debug-mode logging.
    #[must_use]
    pub fn debug(pack: &'a Pack) -> Self {
        Self {
            pack,
            logger: PackTestLogger::debug_mode(&pack.id),
        }
    }

    /// Assert that a command is blocked and log the result.
    ///
    /// # Panics
    ///
    /// Panics if the command is not blocked or reason doesn't match.
    #[track_caller]
    pub fn assert_blocks(&mut self, command: &str, expected_reason_substring: &str) {
        let start = Instant::now();
        let result = self.pack.check(command);
        #[allow(clippy::cast_possible_truncation)]
        let duration_us = start.elapsed().as_micros() as u64; // Safe: test durations won't exceed u64

        if let Some(matched) = &result {
            let passed = matched.reason.contains(expected_reason_substring);
            self.logger.log_pattern_match_detailed(
                matched.name.unwrap_or("unnamed"),
                command,
                true,
                duration_us,
                Some(&format!("{:?}", matched.severity)),
                Some(matched.reason),
            );
            self.logger.log_test_result_detailed(
                "assert_blocks",
                passed,
                if passed { "" } else { "reason mismatch" },
                matched.name,
                Some(command),
            );
            assert!(
                passed,
                "Command '{}' blocked but with unexpected reason.\n\
                 Expected: '{}'\n\
                 Actual: '{}'",
                command, expected_reason_substring, matched.reason
            );
        } else {
            self.logger.log_test_result_detailed(
                "assert_blocks",
                false,
                "command was allowed",
                None,
                Some(command),
            );
            panic!(
                "Expected pack '{}' to block command '{}' but it was allowed",
                self.pack.id, command
            );
        }
    }

    /// Assert that a command is allowed and log the result.
    ///
    /// # Panics
    ///
    /// Panics if the command is blocked.
    #[track_caller]
    pub fn assert_allows(&mut self, command: &str) {
        let start = Instant::now();
        let result = self.pack.check(command);
        #[allow(clippy::cast_possible_truncation)]
        let duration_us = start.elapsed().as_micros() as u64; // Safe: test durations won't exceed u64

        if let Some(matched) = result {
            self.logger.log_pattern_match_detailed(
                matched.name.unwrap_or("unnamed"),
                command,
                true,
                duration_us,
                Some(&format!("{:?}", matched.severity)),
                Some(matched.reason),
            );
            self.logger.log_test_result_detailed(
                "assert_allows",
                false,
                &format!("blocked by {:?}", matched.name),
                matched.name,
                Some(command),
            );
            panic!(
                "Expected pack '{}' to allow command '{}' but it was blocked",
                self.pack.id, command
            );
        } else {
            self.logger
                .log_pattern_match("none", command, false, duration_us);
            self.logger
                .log_test_result_detailed("assert_allows", true, "", None, Some(command));
        }
    }

    /// Run a batch of blocking assertions.
    ///
    /// # Panics
    ///
    /// Panics if any command is not blocked or has unexpected reason.
    #[track_caller]
    pub fn test_batch_blocks(&mut self, commands: &[&str], reason_substring: &str) {
        for cmd in commands {
            self.assert_blocks(cmd, reason_substring);
        }
    }

    /// Run a batch of allowing assertions.
    ///
    /// # Panics
    ///
    /// Panics if any command is blocked.
    #[track_caller]
    pub fn test_batch_allows(&mut self, commands: &[&str]) {
        for cmd in commands {
            self.assert_allows(cmd);
        }
    }

    /// Log a summary and get the JSON report.
    #[must_use]
    pub fn finish(&self) -> String {
        let total = self.logger.test_result_count();
        // All tests passed if we got here without panic
        self.logger.log_summary(total, total, 0);
        self.logger.report_json()
    }

    /// Get the underlying logger for additional customization.
    #[must_use]
    pub const fn logger(&self) -> &PackTestLogger {
        &self.logger
    }
}

/// Create a debug-mode test runner for a pack.
///
/// This is a convenience function for quick debugging.
#[must_use]
pub fn create_debug_runner(pack: &Pack) -> LoggedPackTestRunner<'_> {
    LoggedPackTestRunner::debug(pack)
}

// ============================================================================
// Isomorphism Test Infrastructure
// ============================================================================
//
// This section provides utilities for golden/snapshot testing of the evaluator.
// Use these helpers to verify that refactors preserve exact behavior.

use crate::Config;
use crate::allowlist::AllowlistLayer;
use crate::evaluator::{
    EvaluationDecision, EvaluationResult, MatchSource, evaluate_command_with_pack_order,
};
use crate::packs::{DecisionMode, REGISTRY};
use std::path::Path;

/// A stable, comparable snapshot of an evaluation result for golden testing.
///
/// This struct captures all meaningful aspects of an evaluation that tests
/// should verify remain unchanged across refactors.
///
/// # Example
///
/// ```rust,ignore
/// let snapshot = eval_snapshot("git reset --hard");
/// assert_eq!(snapshot.decision, "deny");
/// assert_eq!(snapshot.rule_id, Some("core.git:reset-hard".into()));
/// ```
#[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
pub struct EvalSnapshot {
    /// The command that was evaluated.
    pub command: String,
    /// Decision: "allow" or "deny".
    pub decision: String,
    /// Effective mode: "deny", "warn", or "log", or None if allowed cleanly.
    pub effective_mode: Option<String>,
    /// Combined pack:pattern ID (e.g., "core.git:reset-hard").
    pub rule_id: Option<String>,
    /// Pack that matched (e.g., "core.git").
    pub pack_id: Option<String>,
    /// Pattern name within the pack (e.g., "reset-hard").
    pub pattern_name: Option<String>,
    /// Match source: "pack", "config", "heredoc", or "legacy".
    pub match_source: Option<String>,
    /// Reason substring (first 100 chars).
    pub reason_preview: Option<String>,
    /// Whether evaluation was truncated due to time budget.
    pub skipped_due_to_budget: bool,
    /// Allowlist layer if overridden: "project", "user", or "system".
    pub allowlist_layer: Option<String>,
    /// Preview of matched text (if available).
    pub matched_text_preview: Option<String>,
}

impl EvalSnapshot {
    /// Create a snapshot from an evaluation result.
    #[must_use]
    pub fn from_result(command: &str, result: &EvaluationResult) -> Self {
        let decision = match result.decision {
            EvaluationDecision::Allow => "allow",
            EvaluationDecision::Deny => "deny",
        };

        let effective_mode = result.effective_mode.map(|m| match m {
            DecisionMode::Deny => "deny".to_string(),
            DecisionMode::Warn => "warn".to_string(),
            DecisionMode::Log => "log".to_string(),
        });

        let (pack_id, pattern_name, rule_id, match_source, reason_preview, matched_text_preview) =
            result
                .pattern_info
                .as_ref()
                .map_or((None, None, None, None, None, None), |info| {
                    let pack = info.pack_id.clone();
                    let pattern = info.pattern_name.clone();
                    let rule = pack
                        .as_ref()
                        .zip(pattern.as_ref())
                        .map(|(p, n)| format!("{p}:{n}"));
                    let source = Some(match info.source {
                        MatchSource::Pack => "pack".to_string(),
                        MatchSource::ConfigOverride => "config".to_string(),
                        MatchSource::HeredocAst => "heredoc".to_string(),
                        MatchSource::LegacyPattern => "legacy".to_string(),
                    });
                    // Truncate reason to ~100 chars, but safely handle UTF-8 boundaries
                    let reason = if info.reason.len() > 100 {
                        // Find a safe truncation point at a char boundary
                        let mut end = 100;
                        while end > 0 && !info.reason.is_char_boundary(end) {
                            end -= 1;
                        }
                        Some(info.reason[..end].to_string())
                    } else {
                        Some(info.reason.clone())
                    };
                    (
                        pack,
                        pattern,
                        rule,
                        source,
                        reason,
                        info.matched_text_preview.clone(),
                    )
                });

        let allowlist_layer = result.allowlist_override.as_ref().map(|ao| match ao.layer {
            AllowlistLayer::Project => "project".to_string(),
            AllowlistLayer::User => "user".to_string(),
            AllowlistLayer::System => "system".to_string(),
        });

        Self {
            command: command.to_string(),
            decision: decision.to_string(),
            effective_mode,
            rule_id,
            pack_id,
            pattern_name,
            match_source,
            reason_preview,
            skipped_due_to_budget: result.skipped_due_to_budget,
            allowlist_layer,
            matched_text_preview,
        }
    }
}

/// Expected log fields from a corpus test case.
///
/// These are the fields specified in `[case.log]` sections of corpus TOML files.
#[derive(Debug, Clone, Default, serde::Serialize, serde::Deserialize)]
pub struct ExpectedLog {
    /// Expected decision: "allow" or "deny".
    pub decision: Option<String>,
    /// Expected effective mode: "deny", "warn", or "log".
    pub mode: Option<String>,
    /// Expected pack ID.
    pub pack_id: Option<String>,
    /// Expected pattern name.
    pub pattern_name: Option<String>,
    /// Expected rule ID (pack:pattern).
    pub rule_id: Option<String>,
    /// Substring that reason must contain.
    pub reason_contains: Option<String>,
    /// Expected allowlist layer.
    pub allowlist_layer: Option<String>,
}

/// A corpus test case loaded from TOML.
#[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
pub struct CorpusTestCase {
    /// Human-readable description.
    pub description: String,
    /// The command to evaluate.
    pub command: String,
    /// Expected outcome: "allow" or "deny".
    pub expected: String,
    /// Optional rule ID that should match (for deny cases).
    #[serde(default)]
    pub rule_id: Option<String>,
    /// Optional expected log fields for detailed verification.
    #[serde(default)]
    pub log: Option<ExpectedLog>,
}

/// A corpus file containing multiple test cases.
#[derive(Debug, serde::Deserialize)]
struct CorpusFile {
    #[serde(rename = "case")]
    cases: Vec<CorpusTestCase>,
}

/// Category of corpus test based on directory name.
#[derive(Debug, Clone, Copy, PartialEq, Eq, Hash)]
pub enum CorpusCategory {
    /// Commands that should be denied (true positives).
    TruePositives,
    /// Commands that should be allowed (false positive prevention).
    FalsePositives,
    /// Bypass attempts that should still be denied.
    BypassAttempts,
    /// Edge cases where any outcome is acceptable (tests parsing/stability).
    EdgeCases,
}

impl CorpusCategory {
    /// Parse category from directory name.
    #[must_use]
    pub fn from_dir_name(name: &str) -> Option<Self> {
        match name {
            "true_positives" => Some(Self::TruePositives),
            "false_positives" => Some(Self::FalsePositives),
            "bypass_attempts" => Some(Self::BypassAttempts),
            "edge_cases" => Some(Self::EdgeCases),
            _ => None,
        }
    }

    /// Get the expected decision for this category.
    #[must_use]
    pub const fn expected_decision(&self) -> Option<&'static str> {
        match self {
            Self::TruePositives | Self::BypassAttempts => Some("deny"),
            Self::FalsePositives => Some("allow"),
            Self::EdgeCases => None, // Any outcome OK
        }
    }
}

/// Load corpus test cases from a TOML file.
///
/// # Errors
///
/// Returns an error if the file cannot be read or parsed.
#[allow(clippy::missing_errors_doc)]
pub fn load_corpus_file(path: &Path) -> Result<Vec<CorpusTestCase>, String> {
    let content = std::fs::read_to_string(path)
        .map_err(|e| format!("Failed to read {}: {e}", path.display()))?;
    let corpus: CorpusFile =
        toml::from_str(&content).map_err(|e| format!("Failed to parse {}: {e}", path.display()))?;
    Ok(corpus.cases)
}

/// Load all corpus test cases from a directory.
///
/// Recursively loads all `.toml` files from the given directory.
///
/// # Errors
///
/// Returns an error if any file cannot be read or parsed.
#[allow(clippy::missing_errors_doc)]
pub fn load_corpus_dir(
    dir: &Path,
) -> Result<Vec<(CorpusCategory, String, CorpusTestCase)>, String> {
    let mut cases = Vec::new();
    let categories = [
        "true_positives",
        "false_positives",
        "bypass_attempts",
        "edge_cases",
    ];

    for category_name in &categories {
        let category_dir = dir.join(category_name);
        if !category_dir.exists() {
            continue;
        }

        let Some(category) = CorpusCategory::from_dir_name(category_name) else {
            continue;
        };

        let entries = std::fs::read_dir(&category_dir)
            .map_err(|e| format!("Failed to read {}: {e}", category_dir.display()))?;

        for entry in entries.flatten() {
            let path = entry.path();
            if path.extension().is_some_and(|ext| ext == "toml") {
                let file_cases = load_corpus_file(&path)?;
                let file_name = path
                    .file_name()
                    .map(|n| n.to_string_lossy().to_string())
                    .unwrap_or_default();
                for case in file_cases {
                    cases.push((category, file_name.clone(), case));
                }
            }
        }
    }

    Ok(cases)
}

/// Evaluate a command and return a snapshot for comparison.
///
/// Uses the default configuration with all core packs enabled.
///
/// # Example
///
/// ```rust,ignore
/// let snapshot = eval_snapshot("git reset --hard");
/// assert_eq!(snapshot.decision, "deny");
/// ```
#[must_use]
pub fn eval_snapshot(command: &str) -> EvalSnapshot {
    eval_snapshot_with_config(command, &Config::default())
}

/// Evaluate a command with a specific configuration and return a snapshot.
#[must_use]
pub fn eval_snapshot_with_config(command: &str, config: &Config) -> EvalSnapshot {
    let enabled_packs = config.enabled_pack_ids();
    let enabled_keywords = REGISTRY.collect_enabled_keywords(&enabled_packs);
    let ordered_packs = REGISTRY.expand_enabled_ordered(&enabled_packs);
    let keyword_index = REGISTRY.build_enabled_keyword_index(&ordered_packs);
    let compiled_overrides = config.overrides.compile();
    let allowlists = crate::LayeredAllowlist::default();
    let heredoc_settings = config.heredoc_settings();

    let result = evaluate_command_with_pack_order(
        command,
        &enabled_keywords,
        &ordered_packs,
        keyword_index.as_ref(),
        &compiled_overrides,
        &allowlists,
        &heredoc_settings,
    );

    EvalSnapshot::from_result(command, &result)
}

/// Diff two evaluation snapshots and return human-readable differences.
///
/// Returns `None` if the snapshots are identical.
#[must_use]
pub fn diff_snapshots(expected: &EvalSnapshot, actual: &EvalSnapshot) -> Option<String> {
    let mut diffs = Vec::new();

    if expected.decision != actual.decision {
        diffs.push(format!(
            "  decision: expected '{}', got '{}'",
            expected.decision, actual.decision
        ));
    }

    if expected.effective_mode != actual.effective_mode {
        diffs.push(format!(
            "  effective_mode: expected {:?}, got {:?}",
            expected.effective_mode, actual.effective_mode
        ));
    }

    if expected.rule_id != actual.rule_id {
        diffs.push(format!(
            "  rule_id: expected {:?}, got {:?}",
            expected.rule_id, actual.rule_id
        ));
    }

    if expected.pack_id != actual.pack_id {
        diffs.push(format!(
            "  pack_id: expected {:?}, got {:?}",
            expected.pack_id, actual.pack_id
        ));
    }

    if expected.pattern_name != actual.pattern_name {
        diffs.push(format!(
            "  pattern_name: expected {:?}, got {:?}",
            expected.pattern_name, actual.pattern_name
        ));
    }

    if expected.match_source != actual.match_source {
        diffs.push(format!(
            "  match_source: expected {:?}, got {:?}",
            expected.match_source, actual.match_source
        ));
    }

    if expected.skipped_due_to_budget != actual.skipped_due_to_budget {
        diffs.push(format!(
            "  skipped_due_to_budget: expected {}, got {}",
            expected.skipped_due_to_budget, actual.skipped_due_to_budget
        ));
    }

    if expected.allowlist_layer != actual.allowlist_layer {
        diffs.push(format!(
            "  allowlist_layer: expected {:?}, got {:?}",
            expected.allowlist_layer, actual.allowlist_layer
        ));
    }

    if diffs.is_empty() {
        None
    } else {
        Some(format!(
            "Snapshot mismatch for command: {}\n{}\n\nReproduce:\n  dcg explain '{}'",
            expected.command,
            diffs.join("\n"),
            expected.command.replace('\'', "'\\''")
        ))
    }
}

/// Verify a corpus test case passes.
///
/// Returns `Ok(())` if the test passes, or `Err(message)` with details on failure.
#[allow(clippy::missing_errors_doc, clippy::too_many_lines)]
pub fn verify_corpus_case(case: &CorpusTestCase, category: CorpusCategory) -> Result<(), String> {
    // Get both snapshot (for most checks) and full result (for reason_contains check)
    let config = Config::default();
    let enabled_packs = config.enabled_pack_ids();
    let enabled_keywords = REGISTRY.collect_enabled_keywords(&enabled_packs);
    let ordered_packs = REGISTRY.expand_enabled_ordered(&enabled_packs);
    let keyword_index = REGISTRY.build_enabled_keyword_index(&ordered_packs);
    let compiled_overrides = config.overrides.compile();
    let allowlists = crate::LayeredAllowlist::default();
    let heredoc_settings = config.heredoc_settings();

    let result = evaluate_command_with_pack_order(
        &case.command,
        &enabled_keywords,
        &ordered_packs,
        keyword_index.as_ref(),
        &compiled_overrides,
        &allowlists,
        &heredoc_settings,
    );
    let snapshot = EvalSnapshot::from_result(&case.command, &result);

    // Get full reason for reason_contains check (not truncated)
    let full_reason = result.pattern_info.as_ref().map(|info| &info.reason);

    // Check basic decision
    let expected_decision = case.expected.as_str();
    if snapshot.decision != expected_decision {
        return Err(format!(
            "Decision mismatch:\n\
             Description: {}\n\
             Command: {}\n\
             Expected: {}\n\
             Actual: {}\n\
             Rule ID: {:?}\n\
             Reproduce: dcg explain '{}'",
            case.description,
            case.command,
            expected_decision,
            snapshot.decision,
            snapshot.rule_id,
            case.command.replace('\'', "'\\''")
        ));
    }

    // Check rule_id if specified
    if let Some(ref expected_rule_id) = case.rule_id {
        if snapshot.rule_id.as_deref() != Some(expected_rule_id.as_str()) {
            return Err(format!(
                "Rule ID mismatch:\n\
                 Description: {}\n\
                 Command: {}\n\
                 Expected rule_id: {}\n\
                 Actual rule_id: {:?}\n\
                 Reproduce: dcg explain '{}'",
                case.description,
                case.command,
                expected_rule_id,
                snapshot.rule_id,
                case.command.replace('\'', "'\\''")
            ));
        }
    }

    // Check detailed log expectations if present
    if let Some(ref log) = case.log {
        if let Some(ref expected_decision) = log.decision {
            if snapshot.decision != *expected_decision {
                return Err(format!(
                    "Log decision mismatch: expected {expected_decision}, got {}",
                    snapshot.decision
                ));
            }
        }

        if let Some(ref expected_mode) = log.mode {
            if snapshot.effective_mode.as_deref() != Some(expected_mode.as_str()) {
                return Err(format!(
                    "Log mode mismatch: expected {expected_mode}, got {:?}",
                    snapshot.effective_mode
                ));
            }
        }

        if let Some(ref expected_pack_id) = log.pack_id {
            if snapshot.pack_id.as_deref() != Some(expected_pack_id.as_str()) {
                return Err(format!(
                    "Log pack_id mismatch: expected {expected_pack_id}, got {:?}",
                    snapshot.pack_id
                ));
            }
        }

        if let Some(ref expected_pattern_name) = log.pattern_name {
            if snapshot.pattern_name.as_deref() != Some(expected_pattern_name.as_str()) {
                return Err(format!(
                    "Log pattern_name mismatch: expected {expected_pattern_name}, got {:?}",
                    snapshot.pattern_name
                ));
            }
        }

        if let Some(ref expected_rule_id) = log.rule_id {
            if snapshot.rule_id.as_deref() != Some(expected_rule_id.as_str()) {
                return Err(format!(
                    "Log rule_id mismatch: expected {expected_rule_id}, got {:?}",
                    snapshot.rule_id
                ));
            }
        }

        if let Some(ref reason_contains) = log.reason_contains {
            // Use full_reason (not truncated reason_preview) for reason_contains check
            let contains = full_reason.is_some_and(|r| r.contains(reason_contains));
            if !contains {
                return Err(format!(
                    "Log reason_contains mismatch: expected to contain '{reason_contains}', got {full_reason:?}"
                ));
            }
        }

        if let Some(ref expected_allowlist_layer) = log.allowlist_layer {
            if snapshot.allowlist_layer.as_deref() != Some(expected_allowlist_layer.as_str()) {
                return Err(format!(
                    "Log allowlist_layer mismatch: expected {expected_allowlist_layer}, got {:?}",
                    snapshot.allowlist_layer
                ));
            }
        }
    }

    // Category-based validation (for cases without explicit log expectations)
    if case.log.is_none() {
        if let Some(cat_decision) = category.expected_decision() {
            if snapshot.decision != cat_decision {
                return Err(format!(
                    "Category-based decision mismatch:\n\
                     Description: {}\n\
                     Command: {}\n\
                     Category: {:?} (expects {})\n\
                     Actual: {}",
                    case.description, case.command, category, cat_decision, snapshot.decision
                ));
            }
        }
    }

    Ok(())
}

/// Assert a command produces the expected snapshot.
///
/// Panics with a detailed diff if the snapshots don't match.
///
/// # Panics
///
/// Panics if the actual evaluation result doesn't match the expected snapshot.
#[track_caller]
pub fn assert_eval_snapshot(command: &str, expected: &EvalSnapshot) {
    let actual = eval_snapshot(command);
    if let Some(diff) = diff_snapshots(expected, &actual) {
        panic!("Evaluation snapshot mismatch:\n{diff}");
    }
}

/// Assert a command produces a specific decision.
///
/// This is a simpler helper for basic decision verification.
///
/// # Panics
///
/// Panics if the decision doesn't match.
#[track_caller]
pub fn assert_decision(command: &str, expected_decision: &str) {
    let snapshot = eval_snapshot(command);
    assert_eq!(
        snapshot.decision,
        expected_decision,
        "Decision mismatch for command: {}\nExpected: {}\nActual: {}\nRule ID: {:?}\n\nReproduce: dcg explain '{}'",
        command,
        expected_decision,
        snapshot.decision,
        snapshot.rule_id,
        command.replace('\'', "'\\''")
    );
}

/// Assert a command is denied with a specific rule.
///
/// # Panics
///
/// Panics if the command is not denied or the rule doesn't match.
#[track_caller]
pub fn assert_denies_with_rule(command: &str, expected_rule_id: &str) {
    let snapshot = eval_snapshot(command);
    assert_eq!(
        snapshot.decision, "deny",
        "Expected deny for command: {}\nActual: {}\nRule ID: {:?}",
        command, snapshot.decision, snapshot.rule_id
    );
    assert_eq!(
        snapshot.rule_id.as_deref(),
        Some(expected_rule_id),
        "Rule mismatch for command: {}\nExpected: {}\nActual: {:?}",
        command,
        expected_rule_id,
        snapshot.rule_id
    );
}

/// Assert a command is allowed (not blocked).
///
/// # Panics
///
/// Panics if the command is denied.
#[track_caller]
pub fn assert_allows_command(command: &str) {
    let snapshot = eval_snapshot(command);
    assert_eq!(
        snapshot.decision, "allow",
        "Expected allow for command: {}\nActual: {}\nBlocked by: {:?}",
        command, snapshot.decision, snapshot.rule_id
    );
}

/// Batch verify corpus test cases.
///
/// Returns a summary of pass/fail counts and details of failures.
#[must_use]
pub fn verify_corpus_batch(
    cases: &[(CorpusCategory, String, CorpusTestCase)],
) -> (usize, usize, Vec<String>) {
    let mut passed = 0;
    let mut failed = 0;
    let mut failures = Vec::new();

    for (category, file, case) in cases {
        match verify_corpus_case(case, *category) {
            Ok(()) => passed += 1,
            Err(msg) => {
                failed += 1;
                failures.push(format!("[{file}] {msg}"));
            }
        }
    }

    (passed, failed, failures)
}

#[cfg(test)]
mod tests {
    use super::*;
    use crate::packs::core;

    #[test]
    fn test_assert_blocks_works() {
        let pack = core::git::create_pack();
        assert_blocks(&pack, "git reset --hard", "destroys uncommitted changes");
    }

    #[test]
    fn test_assert_allows_works() {
        let pack = core::git::create_pack();
        assert_allows(&pack, "git status");
        assert_allows(&pack, "git log");
    }

    #[test]
    fn test_assert_safe_pattern_matches_works() {
        let pack = core::git::create_pack();
        assert_safe_pattern_matches(&pack, "git checkout -b feature");
    }

    #[test]
    fn test_assert_no_match_works() {
        let pack = core::git::create_pack();
        assert_no_match(&pack, "ls -la");
        assert_no_match(&pack, "cargo build");
    }

    #[test]
    fn test_batch_blocks_works() {
        let pack = core::git::create_pack();
        let commands = vec![
            "git reset --hard",
            "git reset --hard HEAD",
            "git reset --hard HEAD~1",
        ];
        test_batch_blocks(&pack, &commands, "reset");
    }

    #[test]
    fn test_batch_allows_works() {
        let pack = core::git::create_pack();
        let commands = vec!["git status", "git log", "git diff"];
        test_batch_allows(&pack, &commands);
    }

    #[test]
    fn test_debug_match_info_provides_useful_output() {
        let pack = core::git::create_pack();
        let info = debug_match_info(&pack, "git reset --hard");
        assert!(info.contains("core.git"));
        assert!(info.contains("reset-hard"));
        assert!(info.contains("MATCH"));
    }

    #[test]
    fn test_patterns_compile_and_validate() {
        let pack = core::git::create_pack();
        assert_patterns_compile(&pack);
        assert_all_patterns_have_reasons(&pack);
        assert_unique_pattern_names(&pack);
    }

    #[test]
    fn test_assert_blocks_with_pattern_works() {
        let pack = core::git::create_pack();
        assert_blocks_with_pattern(&pack, "git reset --hard", "reset-hard");
    }

    #[test]
    fn test_assert_blocks_with_severity_works() {
        let pack = core::git::create_pack();
        assert_blocks_with_severity(&pack, "git reset --hard", Severity::Critical);
    }

    // =========================================================================
    // LoggedPackTestRunner Tests
    // =========================================================================

    #[test]
    fn test_logged_runner_creation() {
        let pack = core::git::create_pack();
        let runner = LoggedPackTestRunner::debug(&pack);
        assert_eq!(runner.logger().test_result_count(), 0);
    }

    #[test]
    fn test_logged_runner_assert_blocks() {
        let pack = core::git::create_pack();
        let mut runner = LoggedPackTestRunner::debug(&pack);
        runner.assert_blocks("git reset --hard", "destroys uncommitted");
        assert_eq!(runner.logger().test_result_count(), 1);
    }

    #[test]
    fn test_logged_runner_assert_allows() {
        let pack = core::git::create_pack();
        let mut runner = LoggedPackTestRunner::debug(&pack);
        runner.assert_allows("git status");
        assert_eq!(runner.logger().test_result_count(), 1);
    }

    #[test]
    fn test_logged_runner_batch_operations() {
        let pack = core::git::create_pack();
        let mut runner = LoggedPackTestRunner::debug(&pack);
        runner.test_batch_allows(&["git status", "git log"]);
        assert_eq!(runner.logger().test_result_count(), 2);
    }

    #[test]
    fn test_logged_runner_finish_produces_json() {
        let pack = core::git::create_pack();
        let mut runner = LoggedPackTestRunner::debug(&pack);
        runner.assert_allows("git status");
        let report = runner.finish();
        assert!(report.contains("\"pack\""));
        assert!(report.contains("core.git"));
    }

    #[test]
    fn test_create_debug_runner_helper() {
        let pack = core::git::create_pack();
        let runner = create_debug_runner(&pack);
        assert_eq!(runner.logger().test_result_count(), 0);
    }

    // =========================================================================
    // Isomorphism Test Infrastructure Tests
    // =========================================================================

    #[test]
    fn test_eval_snapshot_deny() {
        let snapshot = eval_snapshot("git reset --hard");
        assert_eq!(snapshot.decision, "deny");
        assert_eq!(snapshot.rule_id, Some("core.git:reset-hard".to_string()));
        assert_eq!(snapshot.pack_id, Some("core.git".to_string()));
        assert_eq!(snapshot.pattern_name, Some("reset-hard".to_string()));
        assert_eq!(snapshot.match_source, Some("pack".to_string()));
        assert!(snapshot.reason_preview.is_some());
        assert!(!snapshot.skipped_due_to_budget);
    }

    #[test]
    fn test_eval_snapshot_allow() {
        let snapshot = eval_snapshot("git status");
        assert_eq!(snapshot.decision, "allow");
        assert!(snapshot.rule_id.is_none());
        assert!(snapshot.pack_id.is_none());
    }

    #[test]
    fn test_assert_decision_works() {
        assert_decision("git reset --hard", "deny");
        assert_decision("git status", "allow");
    }

    #[test]
    fn test_assert_denies_with_rule_works() {
        assert_denies_with_rule("git reset --hard", "core.git:reset-hard");
        assert_denies_with_rule("git clean -fd", "core.git:clean-force");
    }

    #[test]
    fn test_assert_allows_command_works() {
        assert_allows_command("git status");
        assert_allows_command("ls -la");
    }

    #[test]
    fn test_diff_snapshots_identical() {
        let s1 = eval_snapshot("git reset --hard");
        let s2 = eval_snapshot("git reset --hard");
        assert!(diff_snapshots(&s1, &s2).is_none());
    }

    #[test]
    fn test_diff_snapshots_different() {
        let s1 = eval_snapshot("git reset --hard");
        let s2 = eval_snapshot("git status");
        let diff = diff_snapshots(&s1, &s2);
        assert!(diff.is_some());
        let diff_text = diff.unwrap();
        assert!(diff_text.contains("decision"));
        assert!(diff_text.contains("Reproduce"));
    }

    #[test]
    fn test_corpus_category_from_dir_name() {
        assert_eq!(
            CorpusCategory::from_dir_name("true_positives"),
            Some(CorpusCategory::TruePositives)
        );
        assert_eq!(
            CorpusCategory::from_dir_name("false_positives"),
            Some(CorpusCategory::FalsePositives)
        );
        assert_eq!(
            CorpusCategory::from_dir_name("bypass_attempts"),
            Some(CorpusCategory::BypassAttempts)
        );
        assert_eq!(
            CorpusCategory::from_dir_name("edge_cases"),
            Some(CorpusCategory::EdgeCases)
        );
        assert!(CorpusCategory::from_dir_name("unknown").is_none());
    }

    #[test]
    fn test_corpus_category_expected_decision() {
        assert_eq!(
            CorpusCategory::TruePositives.expected_decision(),
            Some("deny")
        );
        assert_eq!(
            CorpusCategory::BypassAttempts.expected_decision(),
            Some("deny")
        );
        assert_eq!(
            CorpusCategory::FalsePositives.expected_decision(),
            Some("allow")
        );
        assert_eq!(CorpusCategory::EdgeCases.expected_decision(), None);
    }

    #[test]
    fn test_verify_corpus_case_pass() {
        let case = CorpusTestCase {
            description: "git reset --hard should be blocked".to_string(),
            command: "git reset --hard".to_string(),
            expected: "deny".to_string(),
            rule_id: Some("core.git:reset-hard".to_string()),
            log: None,
        };
        let result = verify_corpus_case(&case, CorpusCategory::TruePositives);
        assert!(result.is_ok(), "Expected pass: {result:?}");
    }

    #[test]
    fn test_verify_corpus_case_fail_wrong_decision() {
        let case = CorpusTestCase {
            description: "git status should NOT be blocked".to_string(),
            command: "git status".to_string(),
            expected: "deny".to_string(), // Wrong: git status is allowed
            rule_id: None,
            log: None,
        };
        let result = verify_corpus_case(&case, CorpusCategory::TruePositives);
        assert!(result.is_err());
        let err = result.unwrap_err();
        assert!(err.contains("Decision mismatch"));
    }

    #[test]
    fn test_verify_corpus_case_with_log_expectations() {
        let case = CorpusTestCase {
            description: "git reset --hard with log checks".to_string(),
            command: "git reset --hard".to_string(),
            expected: "deny".to_string(),
            rule_id: Some("core.git:reset-hard".to_string()),
            log: Some(ExpectedLog {
                decision: Some("deny".to_string()),
                mode: Some("deny".to_string()),
                pack_id: Some("core.git".to_string()),
                pattern_name: Some("reset-hard".to_string()),
                rule_id: Some("core.git:reset-hard".to_string()),
                reason_contains: Some("uncommitted".to_string()),
                allowlist_layer: None,
            }),
        };
        let result = verify_corpus_case(&case, CorpusCategory::TruePositives);
        assert!(result.is_ok(), "Expected pass: {result:?}");
    }

    #[test]
    fn test_load_corpus_file() {
        // Test loading a real corpus file
        let path = std::path::Path::new("tests/corpus/true_positives/git_destructive.toml");
        if path.exists() {
            let cases = load_corpus_file(path);
            assert!(cases.is_ok(), "Failed to load: {cases:?}");
            let cases = cases.unwrap();
            assert!(!cases.is_empty());
            // First case should be git reset --hard
            assert!(cases[0].command.contains("git reset"));
        }
    }

    #[test]
    fn test_eval_snapshot_serialization() {
        let snapshot = eval_snapshot("git reset --hard");
        let json = serde_json::to_string(&snapshot).unwrap();
        assert!(json.contains("deny"));
        assert!(json.contains("core.git:reset-hard"));

        // Round-trip
        let deserialized: EvalSnapshot = serde_json::from_str(&json).unwrap();
        assert_eq!(snapshot, deserialized);
    }
}