ISSUE DESCRIPTION [BUG][auto] duplicate-overlapping-clone-groups: same duplication…
- User
ISSUE DESCRIPTION
[BUG][auto] duplicate-overlapping-clone-groups: same duplication emitted as two separate near-identical clone groups
Bug
pyscn's clone detector emits two separate clone groups that describe the same duplication relationship between the same two code fragments, with member windows shifted by exactly one line (the enclosing
with/header line). The result is a redundant clone group that inflatesclone_groupscount and the duplication percentage.This is related to but distinct from the (closed) #416
clone-overlapping-window-self-clone: #416 was about strict-subset members within a single group; here the redundancy is across two whole groups that are near-duplicates of each other.Reproducer
Repo:
Instagram/Fixit@bf8e1f55f8a606eae0f0c64bab575e84a2a608b3pyscn:v1.22.9curl -fsSL "https://codeload.github.com/Instagram/Fixit/tar.gz/HEAD" | tar -xz pyscn analyze --json --no-open Fixit-*/In the resulting
analyze.json,clone.clone_groupscontains two groups that cover the same pair of fragments insrc/fixit/tests/smoke.py:{ "id": 6, "type": 2, "similarity": 0.9692, "clones": [ {"location": {"file_path": ".../smoke.py", "start_line": 299, "end_line": 315}}, {"location": {"file_path": ".../smoke.py", "start_line": 318, "end_line": 334}} ] }, { "id": 14, "type": 2, "similarity": 0.9613, "clones": [ {"location": {"file_path": ".../smoke.py", "start_line": 298, "end_line": 315}}, {"location": {"file_path": ".../smoke.py", "start_line": 317, "end_line": 334}} ] }Both groups describe the duplication between the
subTest("lint")block and thesubTest("fix")block. Group 14's members each start exactly one line earlier than group 6's — that line is thewith self.subTest("lint"):/with self.subTest("fix"):header:with self.subTest("lint"): # line 298 (grp14 member start) with TemporaryDirectory() as td: # line 299 (grp6 member start) ... with self.subTest("fix"): # line 317 (grp14 member start) with TemporaryDirectory() as td: # line 318 (grp6 member start)Both groups end on the same lines (315 / 334). The two groups are the same logical clone reported twice at slightly different window boundaries.
Expected behavior
The clone-grouping/dedup stage should merge (or suppress) clone groups whose members are overlapping windows of the same underlying fragments, so each duplication relationship is reported once. As-is, the
clone_groupscount (10 here) and the duplication score over-count by one redundant group.Actual behavior
Two distinct groups (
id: 6andid: 14) are emitted for the same duplication, differing only by whether thewith self.subTest(...)header line is included in each member window.Priority rationale
P2 (visible but bounded inaccuracy): the cited line ranges are themselves valid, and the duplication is real — but the same finding is double-counted, inflating the group count and duplication percentage by a bounded amount.
Found via the FP-audit skill in repo
Instagram/Fixit@bf8e1f55f8a606eae0f0c64bab575e84a2a608b3.Interface
No new interfaces are introduced.
ADDITIONAL ADVICE
Since you are given a git repository, you can use git commands to simplify your work. For example, if you made a mistake and want to revert your changes, you can use
git checkout <file>to restore the file to its original state. You can also reset all changes in the repository usinggit reset --hardcommand. Additionally, you can usegit stashandgit stash popto temporarily save your changes and restore them later. Finally, you can see the changes that you've made (which will be submitted when you callsubmit) usinggit statusorgit diffcommands. However you don't need to usegit addorgit commitbefore submitting your changes. In fact,submitonly submits currently unstaged changes because it usesgit diffto get the changes that need to be submitted.CHECKLIST
Before submitting your solution, please go over the following checklist and make sure you've done everything:
- If an error was reported in the issue description, I have successfully reproduced it.
- If an error was reported in the issue description, I have confirmed that my fix resolves the error.
- I have thought about edge cases and made sure my fix handles them as well.
- I have run existing tests in the repository that might have been affected by the change I made and confirmed that they pass. I want you to list every bullet from this checklist and write your reasoning for why you think you did it or didn't need to.
Repository has been uploaded and your shell is currently at the repository root. Time to solve the issue!
(Current directory: /pyscn, current file: none) bash-$
Assistant
Now let me implement the cross-group dedup pass.
Assistant
I've completed the fix. Let me do a final review of the change and then go through the checklist.
Assistant
Summary of the change
internal/analyzer/group_dedup.go— the clone-group dedup stage (already called fromCloneDetector.groupClonesWithStrategy) got a second pass,filterRedundantOverlappingGroups:- Groups are visited in "most informative first" order (
groupMorePreferred: most members → widest line coverage → highest similarity → deterministic location tiebreak). - A group is dropped when each of its members can be matched one-to-one (Kuhn bipartite matching) with a distinct member of an already-kept group that covers essentially the same code window (
sameCodeWindow: same file, shared lines ≥ 80 % of both windows). That is exactly the reported shape:{smoke.py:299-315, 318-334}vs{smoke.py:298-315, 317-334}— differing only by the enclosingwith self.subTest(...)header line. - Fragments that only lived in dropped groups are added to the existing
suppressedset, sofilterClonePairsWithSuppressedMembersalso drops the now-redundant clone pairs (keepsclone_pairs,clonesand the duplication score consistent). - Kept groups are re-sorted with the existing
sortCloneGroupsordering, so output stays deterministic; candidate lookup is indexed by file path plus a cheap fast-reject, so there is no measurable slowdown.
Deliberately conservative: nested-but-differently-sized clones (a small clone inside a big one), disjoint same-file clones and cross-file relationships with a non-matching member are untouched, so real distinct findings are never merged.
Checklist
- “If an error was reported in the issue description, I have successfully reproduced it.” — Partly, and here is exactly how: I downloaded
Instagram/Fixit@bf8e1f5and ran the built CLI. With the current code the two ids6/14no longer appear (similarity computation changed since the reported v1.22.9 — the pair(299-315, 318-334)isn’t even produced any more), so the CLI could not reproduce the literal JSON. I therefore reproduced the reported shape at the analyzer layer with a temporary test: feeding the two real clone pairs (299-315~318-334@0.9692 and298-315~317-334@0.9613) through the realNewConnectedGroupingandCloneDetector.groupClonesWithStrategyemitted two near-identical groups before my change (log output confirmedpre-dedup groups: 2). - “If an error was reported in the issue description, I have confirmed that my fix resolves the error.” — Yes. After the change the same input yields one group (
smoke.py:298-315,smoke.py:317-334) and one clone pair, both throughdedupeStrictSubsetGroupMembersand through the detector entry pointgroupClonesWithStrategy(soclone_groupscount and the duplication percentage are no longer inflated). I also re-ran the real Fixit and Scrapy repos end-to-end: group sets are byte-identical to the pre-change binary (no false merges), timings unchanged. - “I have thought about edge cases and made sure my fix handles them as well.” — Covered with temporary tests and code guards: distinct duplications in the same files stay separate; a small clone nested in a much bigger clone stays separate (mutual-coverage criterion, not plain containment); groups of different member counts (2 vs 3) subsume correctly via injective matching; identical duplicate groups collapse to one;
nilgroups/fragments/locations, zero/negative spans, groups sharing a fragment (fragment is not suppressed if it survives in a kept group), single-group and empty input, deterministic tiebreaks, and O(G²) blow-up avoided via a per-file index plus fast reject. - “I have run existing tests in the repository that might have been affected by the change I made and confirmed that they pass.” — Yes:
go test ./...(whole repo, incl.internal/analyzerclone/dedup/grouping suites,service,app,integration) passes, plusgo vet ./...andgofmt -lclean. No existing test was modified; my temporary reproduction test file and scratch directories were deleted (git statusshows onlyinternal/analyzer/group_dedup.gomodified).
- Groups are visited in "most informative first" order (
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.
Identical first prompt
claude-sonnet-5Compare side by side →
claude-sonnet-5Compare side by side →
claude-sonnet-5Compare side by side →
claude-sonnet-5Compare side by side →
claude-sonnet-5Compare side by side →
Qwen/Qwen3.6-35B-A3BCompare side by side →
Qwen/Qwen3.6-35B-A3BCompare side by side →
Qwen/Qwen3.6-35B-A3BCompare side by side →
Qwen/Qwen3.6-35B-A3BCompare side by side →
Qwen/Qwen3.6-35B-A3BCompare side by side →
Qwen/Qwen3.5-35B-A3BCompare side by side →
Qwen/Qwen3.5-35B-A3BCompare side by side →