feat(resolver): resolve iteration-callback edges for for-of, Set, and Array.from (JS)

[carlos-alm]

Summary

• Fixes the more1 jelly-micro benchmark from 0% → 100% recall (10/10 named edges)
• The PTS resolution for for-of, Set+for-of, Array.from, and spread patterns was already fully implemented
• The 0% recall was caused by a naming mismatch in the fixture: _iterPlain/_iterSet vs the expected-edges.json which referenced iterPlain/iterSet — renamed to align them

Patterns verified working

Pattern	Mode	Example
for (const f of arr) { f() }	pts-for-of	iterPlain → fn1, fn2
new Set(arr) + for-of	pts-set	iterSet → fn3, fn4
Array.from(arr, mapCallback)	pts-array-from	mapCallback → fn5, fn6
f(...arr)	pts-spread	consumer1 → fn7, fn8

Test plan

• jelly-micro.test.ts: more1 → 10/10 recall (was 0/10)
• All 6 jelly-micro tests at 100% recall (26/26 total named edges)
• Main resolution-benchmark.test.ts: 171/171 pass (no regressions)

Closes #1378

[claude]

Claude finished @carlos-alm's task —— View job

---

PR Review: feat(resolver): resolve iteration-callback edges for for-of, Set, and Array.from (JS)

Tasks:

• Gather context
• Read changed file and related fixtures
• Analyze root cause
• Assess backlog compliance
• Post final review

---

Root Cause Analysis

Problem Being Solved: The more1 jelly-micro benchmark was reporting 0/10 recall because the fixture functions were named _iterPlain and _iterSet (with underscore prefix), but expected-edges.json referenced them as iterPlain and iterSet. The benchmark matched by symbol name, so no edges were ever found — not because the PTS resolution logic was broken, but because the fixture had a naming mismatch from the start.

Why This Approach: Renaming the fixture functions to align with expected-edges.json is the correct fix. The alternative — updating expected-edges.json to use the underscore-prefixed names — would leave fixture functions named inconsistently and could imply they are "private/internal" helpers when they're actually the test subjects. The PR author chose the right direction.

Risk Assessment: Very low. This is a test fixture rename in a benchmark subdirectory. It does not touch production code, the resolution engine, or any shared utilities. The only risk would be if something else depended on the _iterPlain/_iterSet names, but a quick check confirms these are standalone fixture functions not referenced anywhere else.

---

Backlog Compliance

• Zero-dep: ✓ — No new runtime dependencies added
• Foundation-aligned: ✓ — Pure fixture correction, no architecture concerns
• Problem-fit: 4 — Directly improves benchmark accuracy, which validates the resolution logic that helps AI agents navigate codebases accurately
• Breaking: No — Test fixture internal rename only; no API, CLI, or schema changes
• Tier: 1

---

Code Review

The diff is exactly 2 lines: _iterPlain → iterPlain and _iterSet → iterSet in more1.js (lines 13 and 21). The expected-edges.json is unchanged and already correctly uses iterPlain/iterSet.

One minor observation worth noting: _runFrom and _runSpread (lines 33 and 48) still carry the underscore prefix, but they are not referenced in expected-edges.json either — they're just the harness functions that trigger the patterns under test (the edges flow through mapCallback, consumer1, consumer2 which are already correctly named). This is consistent and correct as-is.

No concerns with test integrity: no assertions were weakened, no expectations were lowered, and no production code was changed. This is a clean fixture correction.

---

Final Recommendation

• Rating: ⭐⭐⭐⭐ (4/5)
• Action: APPROVE
• Reasoning: The change correctly identifies and fixes a naming mismatch that caused a false 0% recall in the benchmark. It fixes the fixture to match the spec (expected-edges.json), not the reverse. The resolution logic itself was already correct. The change is minimal, surgical, and risk-free.

[greptile-apps]

Greptile Summary

This PR fixes a naming mismatch in the more1 jelly-micro benchmark fixture: _iterPlain and _iterSet (with underscore prefixes) did not match the iterPlain/iterSet names referenced in expected-edges.json, causing all 10 expected edges to go unmatched. Removing the underscore prefixes corrects the alignment and brings the benchmark to 100% recall.

• Renames _iterPlain → iterPlain and _iterSet → iterSet in more1.js so the functions match the source names declared in expected-edges.json.
• No resolver logic, test harness, or other fixtures are changed; the fix is purely a fixture naming correction.

Confidence Score: 5/5

Safe to merge — the change is a two-line rename in a test fixture with no impact on production code.

Only two function names are changed in a single benchmark fixture file. The fix directly and verifiably closes the gap between the fixture and its expected-edges definition. No resolver logic, test harness, or other benchmarks are touched.

No files require special attention.

Important Files Changed

Filename	Overview
tests/benchmarks/resolution/fixtures/jelly-micro/more1/more1.js	Renames _iterPlain → iterPlain and _iterSet → iterSet to align fixture function names with the expected-edges.json entries, fixing 0% recall for this benchmark.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A["more1.js fixture\n(before PR)"] -->|"_iterPlain, _iterSet\n(underscore prefix)"| B["Resolver extracts\nfunction names"]
    B --> C["Compare vs expected-edges.json\n(iterPlain, iterSet)"]
    C -->|"Names don't match"| D["0 / 10 edges recalled\n0% recall"]

    E["more1.js fixture\n(after PR)"] -->|"iterPlain, iterSet\n(no underscore)"| F["Resolver extracts\nfunction names"]
    F --> G["Compare vs expected-edges.json\n(iterPlain, iterSet)"]
    G -->|"Names match"| H["10 / 10 edges recalled\n100% recall"]

Loading

_{Reviews (3): Last reviewed commit: "fix: sync with main (#1397)" | Re-trigger Greptile}

[github-actions]

Codegraph Impact Analysis

2 functions changed → 0 callers affected across 0 files

• iterPlain in tests/benchmarks/resolution/fixtures/jelly-micro/more1/more1.js:13 (0 transitive callers)
• iterSet in tests/benchmarks/resolution/fixtures/jelly-micro/more1/more1.js:21 (0 transitive callers)

[carlos-alm]

@greptileai