code-analyze-core 0.4.1

Multi-language AST analysis library using tree-sitter
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
// SPDX-FileCopyrightText: 2026 code-analyze-mcp contributors
// SPDX-License-Identifier: Apache-2.0
//! Main analysis engine for extracting code structure from files and directories.
//!
//! Implements the four MCP tools: `analyze_directory` (Overview), `analyze_file` (`FileDetails`),
//! `analyze_symbol` (call graph), and `analyze_module` (lightweight index). Handles parallel processing and cancellation.

use crate::formatter::{
    format_file_details, format_focused_internal, format_focused_summary_internal, format_structure,
};
use crate::graph::{CallGraph, InternalCallChain};
use crate::lang::{language_for_extension, supported_languages};
use crate::parser::{ElementExtractor, SemanticExtractor};
use crate::test_detection::is_test_file;
use crate::traversal::{WalkEntry, walk_directory};
use crate::types::{
    AnalysisMode, FileInfo, ImplTraitInfo, ImportInfo, SemanticAnalysis, SymbolMatchMode,
};
use rayon::prelude::*;
#[cfg(feature = "schemars")]
use schemars::JsonSchema;
use serde::{Deserialize, Serialize};
use std::path::{Path, PathBuf};
use std::sync::Arc;
use std::sync::atomic::{AtomicUsize, Ordering};
use std::time::Instant;
use thiserror::Error;
use tokio_util::sync::CancellationToken;
use tracing::instrument;

#[derive(Debug, Error)]
#[non_exhaustive]
pub enum AnalyzeError {
    #[error("Traversal error: {0}")]
    Traversal(#[from] crate::traversal::TraversalError),
    #[error("Parser error: {0}")]
    Parser(#[from] crate::parser::ParserError),
    #[error("Graph error: {0}")]
    Graph(#[from] crate::graph::GraphError),
    #[error("Formatter error: {0}")]
    Formatter(#[from] crate::formatter::FormatterError),
    #[error("Analysis cancelled")]
    Cancelled,
    #[error("unsupported language: {0}")]
    UnsupportedLanguage(String),
}

/// Result of directory analysis containing both formatted output and file data.
#[derive(Debug, Clone, Serialize)]
#[cfg_attr(feature = "schemars", derive(JsonSchema))]
#[non_exhaustive]
pub struct AnalysisOutput {
    #[cfg_attr(
        feature = "schemars",
        schemars(description = "Formatted text representation of the analysis")
    )]
    pub formatted: String,
    #[cfg_attr(
        feature = "schemars",
        schemars(description = "List of files analyzed in the directory")
    )]
    pub files: Vec<FileInfo>,
    /// Walk entries used internally for summary generation; not serialized.
    #[serde(skip)]
    #[cfg_attr(feature = "schemars", schemars(skip))]
    pub entries: Vec<WalkEntry>,
    /// Subtree file counts computed from an unbounded walk; used by `format_summary`; not serialized.
    #[serde(skip)]
    #[cfg_attr(feature = "schemars", schemars(skip))]
    pub subtree_counts: Option<Vec<(std::path::PathBuf, usize)>>,
    #[serde(skip_serializing_if = "Option::is_none")]
    #[cfg_attr(
        feature = "schemars",
        schemars(
            description = "Opaque cursor token for the next page of results (absent when no more results)"
        )
    )]
    pub next_cursor: Option<String>,
}

/// Result of file-level semantic analysis.
#[derive(Debug, Clone, Serialize)]
#[cfg_attr(feature = "schemars", derive(JsonSchema))]
#[non_exhaustive]
pub struct FileAnalysisOutput {
    #[cfg_attr(
        feature = "schemars",
        schemars(description = "Formatted text representation of the analysis")
    )]
    pub formatted: String,
    #[cfg_attr(
        feature = "schemars",
        schemars(description = "Semantic analysis data including functions, classes, and imports")
    )]
    pub semantic: SemanticAnalysis,
    #[cfg_attr(
        feature = "schemars",
        schemars(description = "Total line count of the analyzed file")
    )]
    #[cfg_attr(
        feature = "schemars",
        schemars(schema_with = "crate::schema_helpers::integer_schema")
    )]
    pub line_count: usize,
    #[serde(skip_serializing_if = "Option::is_none")]
    #[cfg_attr(
        feature = "schemars",
        schemars(
            description = "Opaque cursor token for the next page of results (absent when no more results)"
        )
    )]
    pub next_cursor: Option<String>,
}

impl FileAnalysisOutput {
    /// Create a new `FileAnalysisOutput`.
    #[must_use]
    pub fn new(
        formatted: String,
        semantic: SemanticAnalysis,
        line_count: usize,
        next_cursor: Option<String>,
    ) -> Self {
        Self {
            formatted,
            semantic,
            line_count,
            next_cursor,
        }
    }
}
#[instrument(skip_all, fields(path = %root.display()))]
// public API; callers expect owned semantics
#[allow(clippy::needless_pass_by_value)]
pub fn analyze_directory_with_progress(
    root: &Path,
    entries: Vec<WalkEntry>,
    progress: Arc<AtomicUsize>,
    ct: CancellationToken,
) -> Result<AnalysisOutput, AnalyzeError> {
    // Check if already cancelled
    if ct.is_cancelled() {
        return Err(AnalyzeError::Cancelled);
    }

    // Detect language from file extension
    let file_entries: Vec<&WalkEntry> = entries.iter().filter(|e| !e.is_dir).collect();

    let start = Instant::now();
    tracing::debug!(file_count = file_entries.len(), root = %root.display(), "analysis start");

    // Parallel analysis of files
    let analysis_results: Vec<FileInfo> = file_entries
        .par_iter()
        .filter_map(|entry| {
            // Check cancellation per file
            if ct.is_cancelled() {
                return None;
            }

            let path_str = entry.path.display().to_string();

            // Detect language from extension
            let ext = entry.path.extension().and_then(|e| e.to_str());

            // Try to read file content; skip binary or unreadable files
            let Ok(source) = std::fs::read_to_string(&entry.path) else {
                progress.fetch_add(1, Ordering::Relaxed);
                return None;
            };

            // Count lines
            let line_count = source.lines().count();

            // Detect language and extract counts
            let (language, function_count, class_count) = if let Some(ext_str) = ext {
                if let Some(lang) = language_for_extension(ext_str) {
                    let lang_str = lang.to_string();
                    match ElementExtractor::extract_with_depth(&source, &lang_str) {
                        Ok((func_count, class_count)) => (lang_str, func_count, class_count),
                        Err(_) => (lang_str, 0, 0),
                    }
                } else {
                    ("unknown".to_string(), 0, 0)
                }
            } else {
                ("unknown".to_string(), 0, 0)
            };

            progress.fetch_add(1, Ordering::Relaxed);

            let is_test = is_test_file(&entry.path);

            Some(FileInfo {
                path: path_str,
                line_count,
                function_count,
                class_count,
                language,
                is_test,
            })
        })
        .collect();

    // Check if cancelled after parallel processing
    if ct.is_cancelled() {
        return Err(AnalyzeError::Cancelled);
    }

    tracing::debug!(
        file_count = file_entries.len(),
        duration_ms = u64::try_from(start.elapsed().as_millis()).unwrap_or(u64::MAX),
        "analysis complete"
    );

    // Format output
    let formatted = format_structure(&entries, &analysis_results, None);

    Ok(AnalysisOutput {
        formatted,
        files: analysis_results,
        entries,
        next_cursor: None,
        subtree_counts: None,
    })
}

/// Analyze a directory structure and return formatted output and file data.
#[instrument(skip_all, fields(path = %root.display()))]
pub fn analyze_directory(
    root: &Path,
    max_depth: Option<u32>,
) -> Result<AnalysisOutput, AnalyzeError> {
    let entries = walk_directory(root, max_depth)?;
    let counter = Arc::new(AtomicUsize::new(0));
    let ct = CancellationToken::new();
    analyze_directory_with_progress(root, entries, counter, ct)
}

/// Determine analysis mode based on parameters and path.
#[must_use]
pub fn determine_mode(path: &str, focus: Option<&str>) -> AnalysisMode {
    if focus.is_some() {
        return AnalysisMode::SymbolFocus;
    }

    let path_obj = Path::new(path);
    if path_obj.is_dir() {
        AnalysisMode::Overview
    } else {
        AnalysisMode::FileDetails
    }
}

/// Analyze a single file and return semantic analysis with formatted output.
#[instrument(skip_all, fields(path))]
pub fn analyze_file(
    path: &str,
    ast_recursion_limit: Option<usize>,
) -> Result<FileAnalysisOutput, AnalyzeError> {
    let start = Instant::now();
    let source = std::fs::read_to_string(path)
        .map_err(|e| AnalyzeError::Parser(crate::parser::ParserError::ParseError(e.to_string())))?;

    let line_count = source.lines().count();

    // Detect language from extension
    let ext = Path::new(path)
        .extension()
        .and_then(|e| e.to_str())
        .and_then(language_for_extension)
        .map_or_else(|| "unknown".to_string(), std::string::ToString::to_string);

    // Extract semantic information
    let mut semantic = SemanticExtractor::extract(&source, &ext, ast_recursion_limit)?;

    // Populate the file path on references now that the path is known
    for r in &mut semantic.references {
        r.location = path.to_string();
    }

    // Resolve Python wildcard imports
    if ext == "python" {
        resolve_wildcard_imports(Path::new(path), &mut semantic.imports);
    }

    // Detect if this is a test file
    let is_test = is_test_file(Path::new(path));

    // Extract parent directory for relative path display
    let parent_dir = Path::new(path).parent();

    // Format output
    let formatted = format_file_details(path, &semantic, line_count, is_test, parent_dir);

    tracing::debug!(path = %path, language = %ext, functions = semantic.functions.len(), classes = semantic.classes.len(), imports = semantic.imports.len(), duration_ms = u64::try_from(start.elapsed().as_millis()).unwrap_or(u64::MAX), "file analysis complete");

    Ok(FileAnalysisOutput::new(
        formatted, semantic, line_count, None,
    ))
}

/// Analyze source code from a string buffer without filesystem access.
///
/// This function analyzes in-memory source code by language identifier. The `language`
/// parameter can be either a language name (e.g., `"rust"`, `"python"`, `"go"`) or a file
/// extension (e.g., `"rs"`, `"py"`).
///
/// Accepted language identifiers depend on compiled features. Use [`supported_languages()`] to
/// discover the available language names at runtime, and [`language_for_extension()`] to resolve
/// a file extension to its supported language identifier.
///
/// # Arguments
///
/// * `source` - The source code to analyze
/// * `language` - The language identifier (language name or extension)
/// * `ast_recursion_limit` - Optional limit for AST traversal depth
///
/// # Returns
///
/// - `Ok(FileAnalysisOutput)` on success
/// - `Err(AnalyzeError::UnsupportedLanguage)` if the language is not recognized
/// - `Err(AnalyzeError::Parser)` if parsing fails
///
/// # Notes
///
/// - Python wildcard import resolution is skipped for in-memory analysis (no filesystem path available)
/// - The formatted output uses the standard file-details formatter, so it includes a `FILE:` header with an empty path
#[inline]
pub fn analyze_str(
    source: &str,
    language: &str,
    ast_recursion_limit: Option<usize>,
) -> Result<FileAnalysisOutput, AnalyzeError> {
    // Resolve language: first try as a file extension, then as a language name
    // (case-insensitive match against supported_languages()).
    let lang = language_for_extension(language).or_else(|| {
        let lower = language.to_ascii_lowercase();
        supported_languages()
            .iter()
            .find(|&&name| name == lower)
            .copied()
    });
    let lang = lang.ok_or_else(|| AnalyzeError::UnsupportedLanguage(language.to_string()))?;

    // Extract semantic information
    let mut semantic = SemanticExtractor::extract(source, lang, ast_recursion_limit)?;

    // Populate a stable in-memory sentinel on all reference locations
    for r in &mut semantic.references {
        r.location = "<memory>".to_string();
    }

    // Count lines in the source
    let line_count = source.lines().count();

    // Format output with empty path (no filesystem access)
    let formatted = format_file_details("", &semantic, line_count, false, None);

    Ok(FileAnalysisOutput::new(
        formatted, semantic, line_count, None,
    ))
}

/// Single entry in a call chain (depth-1 direct caller or callee).
#[derive(Debug, Clone, Serialize, Deserialize)]
#[cfg_attr(feature = "schemars", derive(JsonSchema))]
pub struct CallChainEntry {
    #[cfg_attr(
        feature = "schemars",
        schemars(description = "Symbol name of the caller or callee")
    )]
    pub symbol: String,
    #[cfg_attr(
        feature = "schemars",
        schemars(description = "File path relative to the repository root")
    )]
    pub file: String,
    #[cfg_attr(
        feature = "schemars",
        schemars(
            description = "Line number of the definition or call site (1-indexed)",
            schema_with = "crate::schema_helpers::integer_schema"
        )
    )]
    pub line: usize,
}

/// Result of focused symbol analysis.
#[derive(Debug, Serialize)]
#[cfg_attr(feature = "schemars", derive(JsonSchema))]
#[non_exhaustive]
pub struct FocusedAnalysisOutput {
    #[cfg_attr(
        feature = "schemars",
        schemars(description = "Formatted text representation of the call graph analysis")
    )]
    pub formatted: String,
    #[serde(skip_serializing_if = "Option::is_none")]
    #[cfg_attr(
        feature = "schemars",
        schemars(
            description = "Opaque cursor token for the next page of results (absent when no more results)"
        )
    )]
    pub next_cursor: Option<String>,
    /// Production caller chains (partitioned from incoming chains, excluding test callers).
    /// Not serialized; used for pagination in lib.rs.
    #[serde(skip)]
    #[cfg_attr(feature = "schemars", schemars(skip))]
    pub prod_chains: Vec<InternalCallChain>,
    /// Test caller chains. Not serialized; used for pagination summary in lib.rs.
    #[serde(skip)]
    #[cfg_attr(feature = "schemars", schemars(skip))]
    pub test_chains: Vec<InternalCallChain>,
    /// Outgoing (callee) chains. Not serialized; used for pagination in lib.rs.
    #[serde(skip)]
    #[cfg_attr(feature = "schemars", schemars(skip))]
    pub outgoing_chains: Vec<InternalCallChain>,
    /// Number of definitions for the symbol. Not serialized; used for pagination headers.
    #[serde(skip)]
    #[cfg_attr(feature = "schemars", schemars(skip))]
    pub def_count: usize,
    /// Total unique callers before `impl_only` filter. Not serialized; used for FILTER header.
    #[serde(skip)]
    #[cfg_attr(feature = "schemars", schemars(skip))]
    pub unfiltered_caller_count: usize,
    /// Unique callers after `impl_only` filter. Not serialized; used for FILTER header.
    #[serde(skip)]
    #[cfg_attr(feature = "schemars", schemars(skip))]
    pub impl_trait_caller_count: usize,
    /// Direct (depth-1) production callers. `follow_depth` does not affect this field.
    #[serde(skip_serializing_if = "Option::is_none")]
    pub callers: Option<Vec<CallChainEntry>>,
    /// Direct (depth-1) test callers. `follow_depth` does not affect this field.
    #[serde(skip_serializing_if = "Option::is_none")]
    pub test_callers: Option<Vec<CallChainEntry>>,
    /// Direct (depth-1) callees. `follow_depth` does not affect this field.
    #[serde(skip_serializing_if = "Option::is_none")]
    pub callees: Option<Vec<CallChainEntry>>,
}

/// Parameters for focused symbol analysis. Groups high-arity parameters to keep
/// function signatures under clippy's default 7-argument threshold.
#[derive(Clone)]
pub struct FocusedAnalysisConfig {
    pub focus: String,
    pub match_mode: SymbolMatchMode,
    pub follow_depth: u32,
    pub max_depth: Option<u32>,
    pub ast_recursion_limit: Option<usize>,
    pub use_summary: bool,
    pub impl_only: Option<bool>,
}

/// Internal parameters for focused analysis phases.
#[derive(Clone)]
struct FocusedAnalysisParams {
    focus: String,
    match_mode: SymbolMatchMode,
    follow_depth: u32,
    ast_recursion_limit: Option<usize>,
    use_summary: bool,
    impl_only: Option<bool>,
}

/// Type alias for analysis results: (`file_path`, `semantic_analysis`) pairs and impl-trait info.
type AnalysisResults = (Vec<(PathBuf, SemanticAnalysis)>, Vec<ImplTraitInfo>);

/// Phase 1: Collect semantic analysis for all files in parallel.
fn collect_file_analysis(
    entries: &[WalkEntry],
    progress: &Arc<AtomicUsize>,
    ct: &CancellationToken,
    ast_recursion_limit: Option<usize>,
) -> Result<AnalysisResults, AnalyzeError> {
    // Check if already cancelled
    if ct.is_cancelled() {
        return Err(AnalyzeError::Cancelled);
    }

    // Use pre-walked entries (passed by caller)
    // Collect semantic analysis for all files in parallel
    let file_entries: Vec<&WalkEntry> = entries.iter().filter(|e| !e.is_dir).collect();

    let analysis_results: Vec<(PathBuf, SemanticAnalysis)> = file_entries
        .par_iter()
        .filter_map(|entry| {
            // Check cancellation per file
            if ct.is_cancelled() {
                return None;
            }

            let ext = entry.path.extension().and_then(|e| e.to_str());

            // Try to read file content
            let Ok(source) = std::fs::read_to_string(&entry.path) else {
                progress.fetch_add(1, Ordering::Relaxed);
                return None;
            };

            // Detect language and extract semantic information
            let language = if let Some(ext_str) = ext {
                language_for_extension(ext_str)
                    .map_or_else(|| "unknown".to_string(), std::string::ToString::to_string)
            } else {
                "unknown".to_string()
            };

            if let Ok(mut semantic) =
                SemanticExtractor::extract(&source, &language, ast_recursion_limit)
            {
                // Populate file path on references
                for r in &mut semantic.references {
                    r.location = entry.path.display().to_string();
                }
                // Populate file path on impl_traits (already extracted during SemanticExtractor::extract)
                for trait_info in &mut semantic.impl_traits {
                    trait_info.path.clone_from(&entry.path);
                }
                progress.fetch_add(1, Ordering::Relaxed);
                Some((entry.path.clone(), semantic))
            } else {
                progress.fetch_add(1, Ordering::Relaxed);
                None
            }
        })
        .collect();

    // Check if cancelled after parallel processing
    if ct.is_cancelled() {
        return Err(AnalyzeError::Cancelled);
    }

    // Collect all impl-trait info from analysis results
    let all_impl_traits: Vec<ImplTraitInfo> = analysis_results
        .iter()
        .flat_map(|(_, sem)| sem.impl_traits.iter().cloned())
        .collect();

    Ok((analysis_results, all_impl_traits))
}

/// Phase 2: Build call graph from analysis results.
fn build_call_graph(
    analysis_results: Vec<(PathBuf, SemanticAnalysis)>,
    all_impl_traits: &[ImplTraitInfo],
) -> Result<CallGraph, AnalyzeError> {
    // Build call graph. Always build without impl_only filter first so we can
    // record the unfiltered caller count before discarding those edges.
    CallGraph::build_from_results(
        analysis_results,
        all_impl_traits,
        false, // filter applied below after counting
    )
    .map_err(std::convert::Into::into)
}

/// Phase 3: Resolve symbol and apply `impl_only` filter.
/// Returns (`resolved_focus`, `unfiltered_caller_count`, `impl_trait_caller_count`).
/// CRITICAL: Must capture `unfiltered_caller_count` BEFORE `retain()`, then apply `retain()`,
/// then compute `impl_trait_caller_count`.
fn resolve_symbol(
    graph: &mut CallGraph,
    params: &FocusedAnalysisParams,
) -> Result<(String, usize, usize), AnalyzeError> {
    // Resolve symbol name using the requested match mode.
    let resolved_focus = if params.match_mode == SymbolMatchMode::Exact {
        let exists = graph.definitions.contains_key(&params.focus)
            || graph.callers.contains_key(&params.focus)
            || graph.callees.contains_key(&params.focus);
        if exists {
            params.focus.clone()
        } else {
            return Err(crate::graph::GraphError::SymbolNotFound {
                symbol: params.focus.clone(),
                hint: "Try match_mode=insensitive for a case-insensitive search, or match_mode=prefix to list symbols starting with this name.".to_string(),
            }
            .into());
        }
    } else {
        graph.resolve_symbol_indexed(&params.focus, &params.match_mode)?
    };

    // Count unique callers for the focus symbol before applying impl_only filter.
    let unfiltered_caller_count = graph.callers.get(&resolved_focus).map_or(0, |edges| {
        edges
            .iter()
            .map(|e| &e.neighbor_name)
            .collect::<std::collections::HashSet<_>>()
            .len()
    });

    // Apply impl_only filter now if requested, then count filtered callers.
    // Filter all caller adjacency lists so traversal and formatting are consistently
    // restricted to impl-trait edges regardless of follow_depth.
    let impl_trait_caller_count = if params.impl_only.unwrap_or(false) {
        for edges in graph.callers.values_mut() {
            edges.retain(|e| e.is_impl_trait);
        }
        graph.callers.get(&resolved_focus).map_or(0, |edges| {
            edges
                .iter()
                .map(|e| &e.neighbor_name)
                .collect::<std::collections::HashSet<_>>()
                .len()
        })
    } else {
        unfiltered_caller_count
    };

    Ok((
        resolved_focus,
        unfiltered_caller_count,
        impl_trait_caller_count,
    ))
}

/// Type alias for `compute_chains` return type: (`formatted_output`, `prod_chains`, `test_chains`, `outgoing_chains`, `def_count`).
type ChainComputeResult = (
    String,
    Vec<InternalCallChain>,
    Vec<InternalCallChain>,
    Vec<InternalCallChain>,
    usize,
);

/// Helper function to convert InternalCallChain data to CallChainEntry vec.
/// Takes the first (depth-1) element of each chain and converts it to a CallChainEntry.
/// Returns None if chains is empty, otherwise returns a vec of up to 10 entries.
fn chains_to_entries(
    chains: &[InternalCallChain],
    root: Option<&std::path::Path>,
) -> Option<Vec<CallChainEntry>> {
    if chains.is_empty() {
        return None;
    }
    let entries: Vec<CallChainEntry> = chains
        .iter()
        .take(10)
        .filter_map(|chain| {
            let (symbol, path, line) = chain.chain.first()?;
            let file = match root {
                Some(root) => path
                    .strip_prefix(root)
                    .unwrap_or(path.as_path())
                    .to_string_lossy()
                    .into_owned(),
                None => path.to_string_lossy().into_owned(),
            };
            Some(CallChainEntry {
                symbol: symbol.clone(),
                file,
                line: *line,
            })
        })
        .collect();
    if entries.is_empty() {
        None
    } else {
        Some(entries)
    }
}

/// Phase 4: Compute chains and format output.
fn compute_chains(
    graph: &CallGraph,
    resolved_focus: &str,
    root: &Path,
    params: &FocusedAnalysisParams,
    unfiltered_caller_count: usize,
    impl_trait_caller_count: usize,
) -> Result<ChainComputeResult, AnalyzeError> {
    // Compute chain data for pagination (always, regardless of summary mode)
    let def_count = graph.definitions.get(resolved_focus).map_or(0, Vec::len);
    let incoming_chains = graph.find_incoming_chains(resolved_focus, params.follow_depth)?;
    let outgoing_chains = graph.find_outgoing_chains(resolved_focus, params.follow_depth)?;

    let (prod_chains, test_chains): (Vec<_>, Vec<_>) =
        incoming_chains.iter().cloned().partition(|chain| {
            chain
                .chain
                .first()
                .is_none_or(|(name, path, _)| !is_test_file(path) && !name.starts_with("test_"))
        });

    // Format output with pre-computed chains
    let mut formatted = if params.use_summary {
        format_focused_summary_internal(
            graph,
            resolved_focus,
            params.follow_depth,
            Some(root),
            Some(&incoming_chains),
            Some(&outgoing_chains),
        )?
    } else {
        format_focused_internal(
            graph,
            resolved_focus,
            params.follow_depth,
            Some(root),
            Some(&incoming_chains),
            Some(&outgoing_chains),
        )?
    };

    // Add FILTER header if impl_only filter was applied
    if params.impl_only.unwrap_or(false) {
        let filter_header = format!(
            "FILTER: impl_only=true ({impl_trait_caller_count} of {unfiltered_caller_count} callers shown)\n",
        );
        formatted = format!("{filter_header}{formatted}");
    }

    Ok((
        formatted,
        prod_chains,
        test_chains,
        outgoing_chains,
        def_count,
    ))
}

/// Analyze a symbol's call graph across a directory with progress tracking.
// public API; callers expect owned semantics
#[allow(clippy::needless_pass_by_value)]
pub fn analyze_focused_with_progress(
    root: &Path,
    params: &FocusedAnalysisConfig,
    progress: Arc<AtomicUsize>,
    ct: CancellationToken,
) -> Result<FocusedAnalysisOutput, AnalyzeError> {
    let entries = walk_directory(root, params.max_depth)?;
    let internal_params = FocusedAnalysisParams {
        focus: params.focus.clone(),
        match_mode: params.match_mode.clone(),
        follow_depth: params.follow_depth,
        ast_recursion_limit: params.ast_recursion_limit,
        use_summary: params.use_summary,
        impl_only: params.impl_only,
    };
    analyze_focused_with_progress_with_entries_internal(
        root,
        params.max_depth,
        &progress,
        &ct,
        &internal_params,
        &entries,
    )
}

/// Internal implementation of focused analysis using pre-walked entries and params struct.
#[instrument(skip_all, fields(path = %root.display(), symbol = %params.focus))]
fn analyze_focused_with_progress_with_entries_internal(
    root: &Path,
    _max_depth: Option<u32>,
    progress: &Arc<AtomicUsize>,
    ct: &CancellationToken,
    params: &FocusedAnalysisParams,
    entries: &[WalkEntry],
) -> Result<FocusedAnalysisOutput, AnalyzeError> {
    // Check if already cancelled
    if ct.is_cancelled() {
        return Err(AnalyzeError::Cancelled);
    }

    // Check if path is a file (hint to use directory)
    if root.is_file() {
        let formatted =
            "Single-file focus not supported. Please provide a directory path for cross-file call graph analysis.\n"
                .to_string();
        return Ok(FocusedAnalysisOutput {
            formatted,
            next_cursor: None,
            prod_chains: vec![],
            test_chains: vec![],
            outgoing_chains: vec![],
            def_count: 0,
            unfiltered_caller_count: 0,
            impl_trait_caller_count: 0,
            callers: None,
            test_callers: None,
            callees: None,
        });
    }

    // Phase 1: Collect file analysis
    let (analysis_results, all_impl_traits) =
        collect_file_analysis(entries, progress, ct, params.ast_recursion_limit)?;

    // Check for cancellation before building the call graph (phase 2)
    if ct.is_cancelled() {
        return Err(AnalyzeError::Cancelled);
    }

    // Phase 2: Build call graph
    let mut graph = build_call_graph(analysis_results, &all_impl_traits)?;

    // Check for cancellation before resolving the symbol (phase 3)
    if ct.is_cancelled() {
        return Err(AnalyzeError::Cancelled);
    }

    // Phase 3: Resolve symbol and apply impl_only filter
    let (resolved_focus, unfiltered_caller_count, impl_trait_caller_count) =
        resolve_symbol(&mut graph, params)?;

    // Check for cancellation before computing chains (phase 4)
    if ct.is_cancelled() {
        return Err(AnalyzeError::Cancelled);
    }

    // Phase 4: Compute chains and format output
    let (formatted, prod_chains, test_chains, outgoing_chains, def_count) = compute_chains(
        &graph,
        &resolved_focus,
        root,
        params,
        unfiltered_caller_count,
        impl_trait_caller_count,
    )?;

    // Compute depth-1 chains for structured output fields (always direct relationships only,
    // regardless of `follow_depth` used for the text-formatted output).
    let (depth1_callers, depth1_test_callers, depth1_callees) = if params.follow_depth <= 1 {
        // Chains already at depth 1; reuse the partitioned vecs.
        let callers = chains_to_entries(&prod_chains, Some(root));
        let test_callers = chains_to_entries(&test_chains, Some(root));
        let callees = chains_to_entries(&outgoing_chains, Some(root));
        (callers, test_callers, callees)
    } else {
        // follow_depth > 1: re-query at depth 1 to get only direct edges.
        let incoming1 = graph
            .find_incoming_chains(&resolved_focus, 1)
            .unwrap_or_default();
        let outgoing1 = graph
            .find_outgoing_chains(&resolved_focus, 1)
            .unwrap_or_default();
        let (prod1, test1): (Vec<_>, Vec<_>) = incoming1.into_iter().partition(|chain| {
            chain
                .chain
                .first()
                .is_none_or(|(name, path, _)| !is_test_file(path) && !name.starts_with("test_"))
        });
        let callers = chains_to_entries(&prod1, Some(root));
        let test_callers = chains_to_entries(&test1, Some(root));
        let callees = chains_to_entries(&outgoing1, Some(root));
        (callers, test_callers, callees)
    };

    Ok(FocusedAnalysisOutput {
        formatted,
        next_cursor: None,
        callers: depth1_callers,
        test_callers: depth1_test_callers,
        callees: depth1_callees,
        prod_chains,
        test_chains,
        outgoing_chains,
        def_count,
        unfiltered_caller_count,
        impl_trait_caller_count,
    })
}

/// Analyze a symbol's call graph using pre-walked directory entries.
pub fn analyze_focused_with_progress_with_entries(
    root: &Path,
    params: &FocusedAnalysisConfig,
    progress: &Arc<AtomicUsize>,
    ct: &CancellationToken,
    entries: &[WalkEntry],
) -> Result<FocusedAnalysisOutput, AnalyzeError> {
    let internal_params = FocusedAnalysisParams {
        focus: params.focus.clone(),
        match_mode: params.match_mode.clone(),
        follow_depth: params.follow_depth,
        ast_recursion_limit: params.ast_recursion_limit,
        use_summary: params.use_summary,
        impl_only: params.impl_only,
    };
    analyze_focused_with_progress_with_entries_internal(
        root,
        params.max_depth,
        progress,
        ct,
        &internal_params,
        entries,
    )
}

#[instrument(skip_all, fields(path = %root.display(), symbol = %focus))]
pub fn analyze_focused(
    root: &Path,
    focus: &str,
    follow_depth: u32,
    max_depth: Option<u32>,
    ast_recursion_limit: Option<usize>,
) -> Result<FocusedAnalysisOutput, AnalyzeError> {
    let entries = walk_directory(root, max_depth)?;
    let counter = Arc::new(AtomicUsize::new(0));
    let ct = CancellationToken::new();
    let params = FocusedAnalysisConfig {
        focus: focus.to_string(),
        match_mode: SymbolMatchMode::Exact,
        follow_depth,
        max_depth,
        ast_recursion_limit,
        use_summary: false,
        impl_only: None,
    };
    analyze_focused_with_progress_with_entries(root, &params, &counter, &ct, &entries)
}

/// Analyze a single file and return a minimal fixed schema (name, line count, language,
/// functions, imports) for lightweight code understanding.
#[instrument(skip_all, fields(path))]
pub fn analyze_module_file(path: &str) -> Result<crate::types::ModuleInfo, AnalyzeError> {
    let source = std::fs::read_to_string(path)
        .map_err(|e| AnalyzeError::Parser(crate::parser::ParserError::ParseError(e.to_string())))?;

    let file_path = Path::new(path);
    let name = file_path
        .file_name()
        .and_then(|s| s.to_str())
        .unwrap_or("unknown")
        .to_string();

    let line_count = source.lines().count();

    let language = file_path
        .extension()
        .and_then(|e| e.to_str())
        .and_then(language_for_extension)
        .ok_or_else(|| {
            AnalyzeError::Parser(crate::parser::ParserError::ParseError(
                "unsupported or missing file extension".to_string(),
            ))
        })?;

    let semantic = SemanticExtractor::extract(&source, language, None)?;

    let functions = semantic
        .functions
        .into_iter()
        .map(|f| crate::types::ModuleFunctionInfo {
            name: f.name,
            line: f.line,
        })
        .collect();

    let imports = semantic
        .imports
        .into_iter()
        .map(|i| crate::types::ModuleImportInfo {
            module: i.module,
            items: i.items,
        })
        .collect();

    Ok(crate::types::ModuleInfo {
        name,
        line_count,
        language: language.to_string(),
        functions,
        imports,
    })
}

/// Resolve Python wildcard imports to actual symbol names.
///
/// For each import with items=`["*"]`, this function:
/// 1. Parses the relative dots (if any) and climbs the directory tree
/// 2. Finds the target .py file or __init__.py
/// 3. Extracts symbols (functions and classes) from the target
/// 4. Honors __all__ if defined, otherwise uses function+class names
///
/// All resolution failures are non-fatal: debug-logged and the wildcard is preserved.
fn resolve_wildcard_imports(file_path: &Path, imports: &mut [ImportInfo]) {
    use std::collections::HashMap;

    let mut resolved_cache: HashMap<PathBuf, Vec<String>> = HashMap::new();
    let Ok(file_path_canonical) = file_path.canonicalize() else {
        tracing::debug!(file = ?file_path, "unable to canonicalize current file path");
        return;
    };

    for import in imports.iter_mut() {
        if import.items != ["*"] {
            continue;
        }
        resolve_single_wildcard(import, file_path, &file_path_canonical, &mut resolved_cache);
    }
}

/// Resolve one wildcard import in place. On any failure the import is left unchanged.
fn resolve_single_wildcard(
    import: &mut ImportInfo,
    file_path: &Path,
    file_path_canonical: &Path,
    resolved_cache: &mut std::collections::HashMap<PathBuf, Vec<String>>,
) {
    let module = import.module.clone();
    let dot_count = module.chars().take_while(|c| *c == '.').count();
    if dot_count == 0 {
        return;
    }
    let module_path = module.trim_start_matches('.');

    let Some(target_to_read) = locate_target_file(file_path, dot_count, module_path, &module)
    else {
        return;
    };

    let Ok(canonical) = target_to_read.canonicalize() else {
        tracing::debug!(target = ?target_to_read, import = %module, "unable to canonicalize path");
        return;
    };

    if canonical == file_path_canonical {
        tracing::debug!(target = ?canonical, import = %module, "cannot import from self");
        return;
    }

    if let Some(cached) = resolved_cache.get(&canonical) {
        tracing::debug!(import = %module, symbols_count = cached.len(), "using cached symbols");
        import.items.clone_from(cached);
        return;
    }

    if let Some(symbols) = parse_target_symbols(&target_to_read, &module) {
        tracing::debug!(import = %module, resolved_count = symbols.len(), "wildcard import resolved");
        import.items.clone_from(&symbols);
        resolved_cache.insert(canonical, symbols);
    }
}

/// Locate the .py file that a wildcard import refers to. Returns None if not found.
fn locate_target_file(
    file_path: &Path,
    dot_count: usize,
    module_path: &str,
    module: &str,
) -> Option<PathBuf> {
    let mut target_dir = file_path.parent()?.to_path_buf();

    for _ in 1..dot_count {
        if !target_dir.pop() {
            tracing::debug!(import = %module, "unable to climb {} levels", dot_count.saturating_sub(1));
            return None;
        }
    }

    let target_file = if module_path.is_empty() {
        target_dir.join("__init__.py")
    } else {
        let rel_path = module_path.replace('.', "/");
        target_dir.join(format!("{rel_path}.py"))
    };

    if target_file.exists() {
        Some(target_file)
    } else if target_file.with_extension("").is_dir() {
        let init = target_file.with_extension("").join("__init__.py");
        if init.exists() { Some(init) } else { None }
    } else {
        tracing::debug!(target = ?target_file, import = %module, "target file not found");
        None
    }
}

/// Read and parse a target .py file, returning its exported symbols.
fn parse_target_symbols(target_path: &Path, module: &str) -> Option<Vec<String>> {
    use tree_sitter::Parser;

    let source = match std::fs::read_to_string(target_path) {
        Ok(s) => s,
        Err(e) => {
            tracing::debug!(target = ?target_path, import = %module, error = %e, "unable to read target file");
            return None;
        }
    };

    // Parse once with tree-sitter
    let lang_info = crate::languages::get_language_info("python")?;
    let mut parser = Parser::new();
    if parser.set_language(&lang_info.language).is_err() {
        return None;
    }
    let tree = parser.parse(&source, None)?;

    // First, try to extract __all__ from the same tree
    let mut symbols = Vec::new();
    extract_all_from_tree(&tree, &source, &mut symbols);
    if !symbols.is_empty() {
        tracing::debug!(import = %module, symbols = ?symbols, "using __all__ symbols");
        return Some(symbols);
    }

    // Fallback: extract functions/classes from the tree
    let root = tree.root_node();
    let mut cursor = root.walk();
    for child in root.children(&mut cursor) {
        if matches!(child.kind(), "function_definition" | "class_definition")
            && let Some(name_node) = child.child_by_field_name("name")
        {
            let name = source[name_node.start_byte()..name_node.end_byte()].to_string();
            if !name.starts_with('_') {
                symbols.push(name);
            }
        }
    }
    tracing::debug!(import = %module, fallback_symbols = ?symbols, "using fallback function/class names");
    Some(symbols)
}

/// Extract __all__ from a tree-sitter tree.
fn extract_all_from_tree(tree: &tree_sitter::Tree, source: &str, result: &mut Vec<String>) {
    let root = tree.root_node();
    let mut cursor = root.walk();
    for child in root.children(&mut cursor) {
        if child.kind() == "simple_statement" {
            // simple_statement contains assignment and other statement types
            let mut simple_cursor = child.walk();
            for simple_child in child.children(&mut simple_cursor) {
                if simple_child.kind() == "assignment"
                    && let Some(left) = simple_child.child_by_field_name("left")
                {
                    let target_text = source[left.start_byte()..left.end_byte()].trim();
                    if target_text == "__all__"
                        && let Some(right) = simple_child.child_by_field_name("right")
                    {
                        extract_string_list_from_list_node(&right, source, result);
                    }
                }
            }
        } else if child.kind() == "expression_statement" {
            // Fallback for older Python AST structures
            let mut stmt_cursor = child.walk();
            for stmt_child in child.children(&mut stmt_cursor) {
                if stmt_child.kind() == "assignment"
                    && let Some(left) = stmt_child.child_by_field_name("left")
                {
                    let target_text = source[left.start_byte()..left.end_byte()].trim();
                    if target_text == "__all__"
                        && let Some(right) = stmt_child.child_by_field_name("right")
                    {
                        extract_string_list_from_list_node(&right, source, result);
                    }
                }
            }
        }
    }
}

/// Extract string literals from a Python list node.
fn extract_string_list_from_list_node(
    list_node: &tree_sitter::Node,
    source: &str,
    result: &mut Vec<String>,
) {
    let mut cursor = list_node.walk();
    for child in list_node.named_children(&mut cursor) {
        if child.kind() == "string" {
            let raw = source[child.start_byte()..child.end_byte()].trim();
            // Strip quotes: "name" -> name
            let unquoted = raw.trim_matches('"').trim_matches('\'').to_string();
            if !unquoted.is_empty() {
                result.push(unquoted);
            }
        }
    }
}

#[cfg(test)]
mod tests {
    use super::*;
    use crate::formatter::format_focused_paginated;
    use crate::graph::InternalCallChain;
    use crate::pagination::{PaginationMode, decode_cursor, paginate_slice};
    use std::fs;
    use std::path::PathBuf;
    use tempfile::TempDir;

    #[cfg(feature = "lang-rust")]
    #[test]
    fn analyze_str_rust_happy_path() {
        let source = "fn hello() -> i32 { 42 }";
        let result = analyze_str(source, "rs", None);
        assert!(result.is_ok());
    }

    #[cfg(feature = "lang-python")]
    #[test]
    fn analyze_str_python_happy_path() {
        let source = "def greet(name):\n    return f'Hello {name}'";
        let result = analyze_str(source, "py", None);
        assert!(result.is_ok());
    }

    #[cfg(feature = "lang-rust")]
    #[test]
    fn analyze_str_rust_by_language_name() {
        let source = "fn hello() -> i32 { 42 }";
        let result = analyze_str(source, "rust", None);
        assert!(result.is_ok());
    }

    #[cfg(feature = "lang-python")]
    #[test]
    fn analyze_str_python_by_language_name() {
        let source = "def greet(name):\n    return f'Hello {name}'";
        let result = analyze_str(source, "python", None);
        assert!(result.is_ok());
    }

    #[cfg(feature = "lang-rust")]
    #[test]
    fn analyze_str_rust_mixed_case() {
        let source = "fn hello() -> i32 { 42 }";
        let result = analyze_str(source, "RuSt", None);
        assert!(result.is_ok());
    }

    #[cfg(feature = "lang-python")]
    #[test]
    fn analyze_str_python_mixed_case() {
        let source = "def greet(name):\n    return f'Hello {name}'";
        let result = analyze_str(source, "PyThOn", None);
        assert!(result.is_ok());
    }

    #[test]
    fn analyze_str_unsupported_language() {
        let result = analyze_str("code", "brainfuck", None);
        assert!(
            matches!(result, Err(AnalyzeError::UnsupportedLanguage(lang)) if lang == "brainfuck")
        );
    }

    #[cfg(feature = "lang-rust")]
    #[test]
    fn test_symbol_focus_callers_pagination_first_page() {
        let temp_dir = TempDir::new().unwrap();

        // Create a file with many callers of `target`
        let mut code = String::from("fn target() {}\n");
        for i in 0..15 {
            code.push_str(&format!("fn caller_{:02}() {{ target(); }}\n", i));
        }
        fs::write(temp_dir.path().join("lib.rs"), &code).unwrap();

        // Act
        let output = analyze_focused(temp_dir.path(), "target", 1, None, None).unwrap();

        // Paginate prod callers with page_size=5
        let paginated = paginate_slice(&output.prod_chains, 0, 5, PaginationMode::Callers)
            .expect("paginate failed");
        assert!(
            paginated.total >= 5,
            "should have enough callers to paginate"
        );
        assert!(
            paginated.next_cursor.is_some(),
            "should have next_cursor for page 1"
        );

        // Verify cursor encodes callers mode
        assert_eq!(paginated.items.len(), 5);
    }

    #[test]
    fn test_symbol_focus_callers_pagination_second_page() {
        let temp_dir = TempDir::new().unwrap();

        let mut code = String::from("fn target() {}\n");
        for i in 0..12 {
            code.push_str(&format!("fn caller_{:02}() {{ target(); }}\n", i));
        }
        fs::write(temp_dir.path().join("lib.rs"), &code).unwrap();

        let output = analyze_focused(temp_dir.path(), "target", 1, None, None).unwrap();
        let total_prod = output.prod_chains.len();

        if total_prod > 5 {
            // Get page 1 cursor
            let p1 = paginate_slice(&output.prod_chains, 0, 5, PaginationMode::Callers)
                .expect("paginate failed");
            assert!(p1.next_cursor.is_some());

            let cursor_str = p1.next_cursor.unwrap();
            let cursor_data = decode_cursor(&cursor_str).expect("decode failed");

            // Get page 2
            let p2 = paginate_slice(
                &output.prod_chains,
                cursor_data.offset,
                5,
                PaginationMode::Callers,
            )
            .expect("paginate failed");

            // Format paginated output
            let formatted = format_focused_paginated(
                &p2.items,
                total_prod,
                PaginationMode::Callers,
                "target",
                &output.prod_chains,
                &output.test_chains,
                &output.outgoing_chains,
                output.def_count,
                cursor_data.offset,
                Some(temp_dir.path()),
                true,
            );

            // Assert: header shows correct range for page 2
            let expected_start = cursor_data.offset + 1;
            assert!(
                formatted.contains(&format!("CALLERS ({}", expected_start)),
                "header should show page 2 range, got: {}",
                formatted
            );
        }
    }

    #[test]
    fn test_chains_to_entries_empty_returns_none() {
        // Arrange
        let chains: Vec<InternalCallChain> = vec![];

        // Act
        let result = chains_to_entries(&chains, None);

        // Assert
        assert!(result.is_none());
    }

    #[test]
    fn test_chains_to_entries_with_data_returns_entries() {
        // Arrange
        let chains = vec![
            InternalCallChain {
                chain: vec![("caller1".to_string(), PathBuf::from("/root/lib.rs"), 10)],
            },
            InternalCallChain {
                chain: vec![("caller2".to_string(), PathBuf::from("/root/other.rs"), 20)],
            },
        ];
        let root = PathBuf::from("/root");

        // Act
        let result = chains_to_entries(&chains, Some(root.as_path()));

        // Assert
        assert!(result.is_some());
        let entries = result.unwrap();
        assert_eq!(entries.len(), 2);
        assert_eq!(entries[0].symbol, "caller1");
        assert_eq!(entries[0].file, "lib.rs");
        assert_eq!(entries[0].line, 10);
        assert_eq!(entries[1].symbol, "caller2");
        assert_eq!(entries[1].file, "other.rs");
        assert_eq!(entries[1].line, 20);
    }

    #[test]
    fn test_symbol_focus_callees_pagination() {
        let temp_dir = TempDir::new().unwrap();

        // target calls many functions
        let mut code = String::from("fn target() {\n");
        for i in 0..10 {
            code.push_str(&format!("    callee_{:02}();\n", i));
        }
        code.push_str("}\n");
        for i in 0..10 {
            code.push_str(&format!("fn callee_{:02}() {{}}\n", i));
        }
        fs::write(temp_dir.path().join("lib.rs"), &code).unwrap();

        let output = analyze_focused(temp_dir.path(), "target", 1, None, None).unwrap();
        let total_callees = output.outgoing_chains.len();

        if total_callees > 3 {
            let paginated = paginate_slice(&output.outgoing_chains, 0, 3, PaginationMode::Callees)
                .expect("paginate failed");

            let formatted = format_focused_paginated(
                &paginated.items,
                total_callees,
                PaginationMode::Callees,
                "target",
                &output.prod_chains,
                &output.test_chains,
                &output.outgoing_chains,
                output.def_count,
                0,
                Some(temp_dir.path()),
                true,
            );

            assert!(
                formatted.contains(&format!(
                    "CALLEES (1-{} of {})",
                    paginated.items.len(),
                    total_callees
                )),
                "header should show callees range, got: {}",
                formatted
            );
        }
    }

    #[test]
    fn test_symbol_focus_empty_prod_callers() {
        let temp_dir = TempDir::new().unwrap();

        // target is only called from test functions
        let code = r#"
fn target() {}

#[cfg(test)]
mod tests {
    use super::*;
    #[test]
    fn test_something() { target(); }
}
"#;
        fs::write(temp_dir.path().join("lib.rs"), code).unwrap();

        let output = analyze_focused(temp_dir.path(), "target", 1, None, None).unwrap();

        // prod_chains may be empty; pagination should handle it gracefully
        let paginated = paginate_slice(&output.prod_chains, 0, 100, PaginationMode::Callers)
            .expect("paginate failed");
        assert_eq!(paginated.items.len(), output.prod_chains.len());
        assert!(
            paginated.next_cursor.is_none(),
            "no next_cursor for empty or single-page prod_chains"
        );
    }

    #[test]
    fn test_impl_only_filter_header_correct_counts() {
        let temp_dir = TempDir::new().unwrap();

        // Create a Rust fixture with:
        // - A trait definition
        // - An impl Trait for SomeType block that calls the focus symbol
        // - A regular (non-trait-impl) function that also calls the focus symbol
        let code = r#"
trait MyTrait {
    fn focus_symbol();
}

struct SomeType;

impl MyTrait for SomeType {
    fn focus_symbol() {}
}

fn impl_caller() {
    SomeType::focus_symbol();
}

fn regular_caller() {
    SomeType::focus_symbol();
}
"#;
        fs::write(temp_dir.path().join("lib.rs"), code).unwrap();

        // Call analyze_focused with impl_only=Some(true)
        let params = FocusedAnalysisConfig {
            focus: "focus_symbol".to_string(),
            match_mode: SymbolMatchMode::Insensitive,
            follow_depth: 1,
            max_depth: None,
            ast_recursion_limit: None,
            use_summary: false,
            impl_only: Some(true),
        };
        let output = analyze_focused_with_progress(
            temp_dir.path(),
            &params,
            Arc::new(AtomicUsize::new(0)),
            CancellationToken::new(),
        )
        .unwrap();

        // Assert the result contains "FILTER: impl_only=true"
        assert!(
            output.formatted.contains("FILTER: impl_only=true"),
            "formatted output should contain FILTER header for impl_only=true, got: {}",
            output.formatted
        );

        // Assert the retained count N < total count M
        assert!(
            output.impl_trait_caller_count < output.unfiltered_caller_count,
            "impl_trait_caller_count ({}) should be less than unfiltered_caller_count ({})",
            output.impl_trait_caller_count,
            output.unfiltered_caller_count
        );

        // Assert format is "FILTER: impl_only=true (N of M callers shown)"
        let filter_line = output
            .formatted
            .lines()
            .find(|line| line.contains("FILTER: impl_only=true"))
            .expect("should find FILTER line");
        assert!(
            filter_line.contains(&format!(
                "({} of {} callers shown)",
                output.impl_trait_caller_count, output.unfiltered_caller_count
            )),
            "FILTER line should show correct N of M counts, got: {}",
            filter_line
        );
    }

    #[test]
    fn test_callers_count_matches_formatted_output() {
        let temp_dir = TempDir::new().unwrap();

        // Create a file with multiple callers of `target`
        let code = r#"
fn target() {}
fn caller_a() { target(); }
fn caller_b() { target(); }
fn caller_c() { target(); }
"#;
        fs::write(temp_dir.path().join("lib.rs"), code).unwrap();

        // Analyze the symbol
        let output = analyze_focused(temp_dir.path(), "target", 1, None, None).unwrap();

        // Extract CALLERS count from formatted output
        let formatted = &output.formatted;
        let callers_count_from_output = formatted
            .lines()
            .find(|line| line.contains("FOCUS:"))
            .and_then(|line| {
                line.split(',')
                    .find(|part| part.contains("callers"))
                    .and_then(|part| {
                        part.trim()
                            .split_whitespace()
                            .next()
                            .and_then(|s| s.parse::<usize>().ok())
                    })
            })
            .expect("should find CALLERS count in formatted output");

        // Compute expected count from prod_chains (unique first-caller names)
        let expected_callers_count = output
            .prod_chains
            .iter()
            .filter_map(|chain| chain.chain.first().map(|(name, _, _)| name))
            .collect::<std::collections::HashSet<_>>()
            .len();

        assert_eq!(
            callers_count_from_output, expected_callers_count,
            "CALLERS count in formatted output should match unique-first-caller count in prod_chains"
        );
    }
}