1use std::path::{Path, PathBuf};
4
5use fallow_engine::duplicates::{
6 CloneFingerprintSet, clone_fingerprint, dominant_identifier, fingerprint_for_fragment,
7};
8use fallow_output::{
9 CloneFamilyAction, CloneGroupAction, CodeClimateIssue, CodeClimateIssueInput,
10 CodeClimateSeverity, clone_family_actions, clone_group_actions, codeclimate_fingerprint_hash,
11 normalize_uri,
12};
13use fallow_types::duplicates::{
14 CloneFamily, CloneGroup, CloneInstance, DuplicationReport, DuplicationStats, MirroredDirectory,
15 RefactoringSuggestion, clone_location_spread,
16};
17use fallow_types::envelope::AuditIntroduced;
18use fallow_types::serde_path;
19use serde::Serialize;
20
21#[derive(Debug, Clone, Serialize)]
29#[cfg_attr(feature = "schema", derive(schemars::JsonSchema))]
30pub struct AttributedInstance {
31 #[serde(flatten)]
33 pub instance: CloneInstance,
34 pub owner: String,
37}
38
39#[derive(Debug, Clone, Serialize)]
42#[cfg_attr(feature = "schema", derive(schemars::JsonSchema))]
43pub struct AttributedCloneGroup {
44 pub primary_owner: String,
47 pub token_count: usize,
49 pub line_count: usize,
51 #[serde(default, skip_serializing_if = "Option::is_none")]
53 #[cfg_attr(feature = "schema", schemars(with = "f64"))]
54 pub similarity: Option<f64>,
55 pub instances: Vec<AttributedInstance>,
58}
59
60impl AttributedCloneGroup {
61 #[must_use]
63 pub fn fingerprint(&self, fingerprints: &CloneFingerprintSet) -> String {
64 let instances: Vec<_> = self
65 .instances
66 .iter()
67 .map(|instance| instance.instance.clone())
68 .collect();
69 fingerprints.fingerprint_for_parts(&instances, self.token_count, self.line_count)
70 }
71}
72
73#[derive(Debug, Clone, Serialize)]
79#[cfg_attr(feature = "schema", derive(schemars::JsonSchema))]
80pub struct AttributedCloneGroupFinding {
81 #[serde(flatten)]
83 pub group: AttributedCloneGroup,
84 pub fingerprint: String,
89 pub spread: usize,
91 pub actions: Vec<CloneGroupAction>,
93}
94
95impl AttributedCloneGroupFinding {
96 #[allow(
98 dead_code,
99 reason = "kept for focused wrapper tests and non-report construction paths"
100 )]
101 #[must_use]
102 pub fn with_actions(group: AttributedCloneGroup) -> Self {
103 let fingerprint = group.instances.first().map_or_else(
104 || fingerprint_for_fragment(""),
105 |ai| fingerprint_for_fragment(&ai.instance.fragment),
106 );
107 Self::with_fingerprint(group, fingerprint)
108 }
109
110 #[must_use]
112 pub fn with_fingerprint(group: AttributedCloneGroup, fingerprint: String) -> Self {
113 let spread = clone_location_spread(group.instances.iter().map(|instance| {
114 (
115 instance.instance.file.as_path(),
116 instance.instance.start_line,
117 instance.instance.end_line,
118 )
119 }));
120 let actions = clone_group_actions(group.line_count, group.instances.len());
121 Self {
122 group,
123 fingerprint,
124 spread,
125 actions,
126 }
127 }
128}
129
130#[derive(Debug, Clone, Serialize)]
133#[cfg_attr(feature = "schema", derive(schemars::JsonSchema))]
134pub struct DuplicationGroup {
135 pub key: String,
139 pub stats: DuplicationStats,
141 pub clone_groups: Vec<AttributedCloneGroupFinding>,
146 pub clone_families: Vec<CloneFamilyFinding>,
149}
150
151#[derive(Debug, Clone, Serialize)]
153pub struct DuplicationGrouping {
154 pub mode: &'static str,
156 pub groups: Vec<DuplicationGroup>,
158}
159
160#[derive(Debug, Clone, Copy, PartialEq, Eq, Serialize)]
167#[cfg_attr(feature = "schema", derive(schemars::JsonSchema))]
168#[serde(rename_all = "kebab-case")]
169pub enum CloneDemotionReason {
170 NoAddedLines,
174}
175
176impl CloneDemotionReason {
177 #[must_use]
180 pub fn wire_name(self) -> String {
181 match serde_json::to_value(self) {
182 Ok(serde_json::Value::String(name)) => name,
183 _ => String::new(),
184 }
185 }
186}
187
188#[derive(Debug, Clone, Serialize)]
194#[cfg_attr(feature = "schema", derive(schemars::JsonSchema))]
195pub struct CloneGroupFinding {
196 #[serde(flatten)]
198 pub group: CloneGroup,
199 pub fingerprint: String,
204 pub spread: usize,
206 #[serde(default, skip_serializing_if = "Option::is_none")]
213 pub suggested_name: Option<String>,
214 pub actions: Vec<CloneGroupAction>,
218 #[serde(default, skip_serializing_if = "Option::is_none")]
221 pub introduced: Option<AuditIntroduced>,
222 #[serde(default, skip_serializing_if = "Option::is_none")]
227 pub demotion_reason: Option<CloneDemotionReason>,
228}
229
230impl CloneGroupFinding {
231 #[allow(
233 dead_code,
234 reason = "kept for focused wrapper tests and non-report construction paths"
235 )]
236 #[must_use]
237 pub fn with_actions(group: CloneGroup) -> Self {
238 let fingerprint = clone_fingerprint(&group.instances);
239 Self::with_fingerprint(group, fingerprint)
240 }
241
242 #[must_use]
244 pub fn with_fingerprint(group: CloneGroup, fingerprint: String) -> Self {
245 let spread = group.spread();
246 let suggested_name = dominant_identifier(&group);
247 let actions = clone_group_actions(group.line_count, group.instances.len());
248 Self {
249 fingerprint,
250 spread,
251 suggested_name,
252 group,
253 actions,
254 introduced: None,
255 demotion_reason: None,
256 }
257 }
258}
259
260#[derive(Debug, Clone, Serialize)]
271#[cfg_attr(feature = "schema", derive(schemars::JsonSchema))]
272pub struct CloneFamilyFinding {
273 #[serde(serialize_with = "serde_path::serialize_vec")]
275 pub files: Vec<PathBuf>,
276 pub groups: Vec<CloneGroupFinding>,
280 pub total_duplicated_lines: usize,
282 pub total_duplicated_tokens: usize,
284 pub suggestions: Vec<RefactoringSuggestion>,
286 pub actions: Vec<CloneFamilyAction>,
291}
292
293impl CloneFamilyFinding {
294 #[allow(
296 dead_code,
297 reason = "kept for focused wrapper tests and non-report construction paths"
298 )]
299 #[must_use]
300 pub fn with_actions(family: CloneFamily) -> Self {
301 let fingerprints = CloneFingerprintSet::from_groups(&family.groups);
302 Self::with_fingerprints(family, &fingerprints)
303 }
304
305 #[must_use]
308 pub fn with_fingerprints(family: CloneFamily, fingerprints: &CloneFingerprintSet) -> Self {
309 let actions = build_clone_family_actions(
310 &family.groups,
311 family.total_duplicated_lines,
312 &family.suggestions,
313 );
314 Self {
315 files: family.files,
316 groups: family
317 .groups
318 .into_iter()
319 .map(|group| {
320 let fingerprint = fingerprints.fingerprint_for_group(&group);
321 CloneGroupFinding::with_fingerprint(group, fingerprint)
322 })
323 .collect(),
324 total_duplicated_lines: family.total_duplicated_lines,
325 total_duplicated_tokens: family.total_duplicated_tokens,
326 suggestions: family.suggestions,
327 actions,
328 }
329 }
330}
331
332fn build_clone_family_actions(
333 groups: &[CloneGroup],
334 total_duplicated_lines: usize,
335 suggestions: &[RefactoringSuggestion],
336) -> Vec<CloneFamilyAction> {
337 clone_family_actions(
338 groups.len(),
339 total_duplicated_lines,
340 suggestions
341 .iter()
342 .map(|suggestion| suggestion.description.as_str()),
343 )
344}
345
346#[derive(Debug, Clone, Serialize)]
356#[cfg_attr(feature = "schema", derive(schemars::JsonSchema))]
357pub struct DupesReportPayload {
358 pub clone_groups: Vec<CloneGroupFinding>,
360 pub clone_families: Vec<CloneFamilyFinding>,
368 #[serde(default, skip_serializing_if = "Vec::is_empty")]
370 pub mirrored_directories: Vec<MirroredDirectory>,
371 pub stats: DuplicationStats,
373}
374
375impl DupesReportPayload {
376 #[must_use]
378 pub fn from_report(report: &DuplicationReport) -> Self {
379 let fingerprints = CloneFingerprintSet::from_groups(&report.clone_groups);
380 Self {
381 clone_groups: report
382 .clone_groups
383 .iter()
384 .map(|group| {
385 CloneGroupFinding::with_fingerprint(
386 group.clone(),
387 fingerprints.fingerprint_for_group(group),
388 )
389 })
390 .collect(),
391 clone_families: report
392 .clone_families
393 .iter()
394 .map(|family| CloneFamilyFinding::with_fingerprints(family.clone(), &fingerprints))
395 .collect(),
396 mirrored_directories: report.mirrored_directories.clone(),
397 stats: report.stats.clone(),
398 }
399 }
400}
401
402#[must_use]
408#[expect(
409 clippy::cast_possible_truncation,
410 reason = "line numbers are bounded by source size"
411)]
412pub fn build_duplication_codeclimate(
413 report: &DuplicationReport,
414 root: &Path,
415) -> Vec<CodeClimateIssue> {
416 let mut issues = Vec::new();
417
418 for (i, group) in report.clone_groups.iter().enumerate() {
419 let token_str = group.token_count.to_string();
420 let line_count_str = group.line_count.to_string();
421 let fragment_prefix: String = group
422 .instances
423 .first()
424 .map(|inst| inst.fragment.chars().take(64).collect())
425 .unwrap_or_default();
426
427 for instance in &group.instances {
428 let path = codeclimate_path(&instance.file, root);
429 let start_str = instance.start_line.to_string();
430 let fp = codeclimate_fingerprint_hash(&[
431 "fallow/code-duplication",
432 &path,
433 &start_str,
434 &token_str,
435 &line_count_str,
436 &fragment_prefix,
437 ]);
438 issues.push(fallow_output::build_codeclimate_issue(
439 CodeClimateIssueInput {
440 check_name: "fallow/code-duplication",
441 description: &format!(
442 "Code clone group {} ({} lines, {} instances)",
443 i + 1,
444 group.line_count,
445 group.instances.len()
446 ),
447 severity: CodeClimateSeverity::Minor,
448 category: "Duplication",
449 path: &path,
450 begin_line: Some(instance.start_line as u32),
451 fingerprint: &fp,
452 },
453 ));
454 }
455 }
456
457 issues
458}
459
460fn codeclimate_path(path: &Path, root: &Path) -> String {
461 normalize_uri(
462 &path
463 .strip_prefix(root)
464 .unwrap_or(path)
465 .display()
466 .to_string(),
467 )
468}
469
470#[cfg(test)]
471mod tests {
472 use std::path::Path;
473
474 use fallow_output::{CloneFamilyActionType, CloneGroupActionType};
475 use fallow_types::duplicates::{
476 CloneInstance, DuplicationStats, RefactoringKind, RefactoringSuggestion,
477 };
478
479 use super::*;
480
481 fn instance(path: &str) -> CloneInstance {
482 CloneInstance {
483 file: PathBuf::from(path),
484 start_line: 1,
485 end_line: 10,
486 start_col: 0,
487 end_col: 0,
488 fragment: String::new(),
489 }
490 }
491
492 fn group(instances: usize) -> CloneGroup {
493 CloneGroup {
494 instances: (0..instances)
495 .map(|i| instance(&format!("/root/file_{i}.ts")))
496 .collect(),
497 token_count: 100,
498 line_count: 20,
499 similarity: None,
500 }
501 }
502
503 #[test]
504 fn clone_group_finding_position_0_is_extract_shared() {
505 let finding = CloneGroupFinding::with_actions(group(2));
506 assert_eq!(finding.actions.len(), 2);
507 assert_eq!(finding.actions[0].kind, CloneGroupActionType::ExtractShared);
508 assert_eq!(finding.actions[1].kind, CloneGroupActionType::SuppressLine);
509 assert!(finding.introduced.is_none());
510 assert!(finding.demotion_reason.is_none());
511 }
512
513 #[test]
514 fn clone_group_finding_omits_audit_only_fields_outside_audit() {
515 let finding = CloneGroupFinding::with_actions(group(2));
518 let value = serde_json::to_value(&finding).expect("finding serializes");
519 assert!(value.get("introduced").is_none());
520 assert!(value.get("demotion_reason").is_none());
521 }
522
523 #[test]
524 fn clone_demotion_reason_wire_name_matches_serde_representation() {
525 let reason = CloneDemotionReason::NoAddedLines;
526 assert_eq!(
527 serde_json::to_value(reason).expect("reason serializes"),
528 serde_json::Value::String(reason.wire_name())
529 );
530 assert_eq!(reason.wire_name(), "no-added-lines");
531 }
532
533 #[test]
534 fn attributed_clone_group_finding_actions_match_clone_group_shape() {
535 let attributed = AttributedCloneGroup {
536 primary_owner: "src".to_string(),
537 token_count: 100,
538 line_count: 20,
539 similarity: None,
540 instances: vec![
541 AttributedInstance {
542 instance: instance("/root/src/a.ts"),
543 owner: "src".to_string(),
544 },
545 AttributedInstance {
546 instance: instance("/root/src/b.ts"),
547 owner: "src".to_string(),
548 },
549 ],
550 };
551 let finding = AttributedCloneGroupFinding::with_actions(attributed);
552 assert_eq!(finding.actions.len(), 2);
553 assert_eq!(finding.actions[0].kind, CloneGroupActionType::ExtractShared);
554 assert_eq!(finding.actions[1].kind, CloneGroupActionType::SuppressLine);
555 }
556
557 #[test]
558 fn clone_group_finding_surfaces_dominant_identifier() {
559 let fragment = "function parseCsv() { parseCsv(); parseCsv(); return parseCsv; }";
560 let g = CloneGroup {
561 instances: vec![
562 CloneInstance {
563 file: PathBuf::from("/root/a.ts"),
564 start_line: 1,
565 end_line: 3,
566 start_col: 0,
567 end_col: 0,
568 fragment: fragment.to_string(),
569 },
570 CloneInstance {
571 file: PathBuf::from("/root/b.ts"),
572 start_line: 1,
573 end_line: 3,
574 start_col: 0,
575 end_col: 0,
576 fragment: fragment.to_string(),
577 },
578 ],
579 token_count: 100,
580 line_count: 3,
581 similarity: None,
582 };
583 let finding = CloneGroupFinding::with_actions(g);
584 assert_eq!(finding.suggested_name.as_deref(), Some("parseCsv"));
585 }
586
587 #[test]
588 fn clone_group_finding_suggested_name_none_for_unnamed_fragment() {
589 let finding = CloneGroupFinding::with_actions(group(2));
590 assert!(finding.suggested_name.is_none());
591 }
592
593 #[test]
594 fn clone_group_finding_description_pluralises_instance_count() {
595 let single = CloneGroupFinding::with_actions(group(1));
596 assert!(single.actions[0].description.contains("1 instance"));
597 assert!(!single.actions[0].description.contains("1 instances"));
598 let multi = CloneGroupFinding::with_actions(group(3));
599 assert!(multi.actions[0].description.contains("3 instances"));
600 }
601
602 #[test]
603 fn clone_family_finding_position_0_is_extract_shared_then_suggestions_then_suppress() {
604 let family = CloneFamily {
605 files: vec![PathBuf::from("/root/a.ts"), PathBuf::from("/root/b.ts")],
606 groups: vec![group(2), group(2)],
607 total_duplicated_lines: 40,
608 total_duplicated_tokens: 200,
609 suggestions: vec![
610 RefactoringSuggestion {
611 kind: RefactoringKind::ExtractFunction,
612 description: "Extract helper".to_string(),
613 estimated_savings: 10,
614 },
615 RefactoringSuggestion {
616 kind: RefactoringKind::ExtractModule,
617 description: "Extract module".to_string(),
618 estimated_savings: 30,
619 },
620 ],
621 };
622 let finding = CloneFamilyFinding::with_actions(family);
623 assert_eq!(finding.actions.len(), 4);
624 assert_eq!(
625 finding.actions[0].kind,
626 CloneFamilyActionType::ExtractShared
627 );
628 assert_eq!(
629 finding.actions[1].kind,
630 CloneFamilyActionType::ApplySuggestion
631 );
632 assert_eq!(finding.actions[1].description, "Extract helper");
633 assert_eq!(
634 finding.actions[2].kind,
635 CloneFamilyActionType::ApplySuggestion
636 );
637 assert_eq!(finding.actions[2].description, "Extract module");
638 assert_eq!(finding.actions[3].kind, CloneFamilyActionType::SuppressLine);
639 assert_eq!(finding.groups.len(), 2);
640 for inner in &finding.groups {
641 assert_eq!(inner.actions.len(), 2);
642 assert_eq!(inner.actions[0].kind, CloneGroupActionType::ExtractShared);
643 assert_eq!(inner.actions[1].kind, CloneGroupActionType::SuppressLine);
644 }
645 }
646
647 #[test]
648 fn clone_family_finding_with_no_suggestions_emits_two_actions() {
649 let family = CloneFamily {
650 files: vec![PathBuf::from("/root/a.ts")],
651 groups: vec![group(2)],
652 total_duplicated_lines: 20,
653 total_duplicated_tokens: 100,
654 suggestions: Vec::new(),
655 };
656 let finding = CloneFamilyFinding::with_actions(family);
657 assert_eq!(finding.actions.len(), 2);
658 assert_eq!(
659 finding.actions[0].kind,
660 CloneFamilyActionType::ExtractShared
661 );
662 assert_eq!(finding.actions[1].kind, CloneFamilyActionType::SuppressLine);
663 }
664
665 #[test]
666 fn payload_from_report_wraps_all_findings() {
667 let report = DuplicationReport {
668 clone_groups: vec![group(2), group(3)],
669 clone_families: vec![CloneFamily {
670 files: vec![PathBuf::from("/root/a.ts")],
671 groups: vec![group(2)],
672 total_duplicated_lines: 20,
673 total_duplicated_tokens: 100,
674 suggestions: Vec::new(),
675 }],
676 mirrored_directories: Vec::new(),
677 stats: DuplicationStats::default(),
678 };
679 let payload = DupesReportPayload::from_report(&report);
680 assert_eq!(payload.clone_groups.len(), 2);
681 assert_eq!(payload.clone_families.len(), 1);
682 for finding in &payload.clone_groups {
683 assert_eq!(finding.actions.len(), 2);
684 }
685 assert_eq!(payload.clone_families[0].actions.len(), 2);
686 }
687
688 #[test]
689 fn duplication_codeclimate_uses_relative_normalized_paths() {
690 let report = DuplicationReport {
691 clone_groups: vec![CloneGroup {
692 instances: vec![CloneInstance {
693 file: PathBuf::from("/root/app/[id]/page.tsx"),
694 start_line: 4,
695 end_line: 8,
696 start_col: 0,
697 end_col: 0,
698 fragment: "const duplicate = 1;".to_string(),
699 }],
700 token_count: 42,
701 line_count: 5,
702 similarity: None,
703 }],
704 clone_families: Vec::new(),
705 mirrored_directories: Vec::new(),
706 stats: DuplicationStats::default(),
707 };
708
709 let issues = build_duplication_codeclimate(&report, Path::new("/root"));
710
711 assert_eq!(issues.len(), 1);
712 let issue = &issues[0];
713 assert_eq!(issue.check_name, "fallow/code-duplication");
714 assert_eq!(issue.location.path, "app/%5Bid%5D/page.tsx");
715 assert_eq!(issue.location.lines.begin, 4);
716 assert_eq!(issue.categories, vec!["Duplication"]);
717 assert!(issue.description.contains("Code clone group 1"));
718 }
719}