fix(signals): group contributor open PRs by repo case-insensitively

[galuis116]

Summary

Fixes #1526.

groupByRepo in buildContributorOpenPrMonitor bucketed a contributor's open PRs by the raw pr.repoFullName, while every other repo-name operation in the same function is case-insensitive (the registered-repo set, the open-PR filter, the repo lookup). GitHub repo names are case-insensitive, so case-variant repoFullName values for one repo split into separate groups — under-counting per-repo open PRs, running duplicate detection only within each partial group, and querying listPullRequests once per casing.

What changed

• groupByRepo now keys buckets on pr.repoFullName.toLowerCase(), matching the function's own repo-name handling.
• The consumer loop derives the canonical repoFullName from a bucket's first PR (its original casing) for the case-sensitive per-repo DB queries — the established key-by-lower / keep-original pattern used in contributor-evidence-graph.ts.

Scope

• Conventional Commit title.
• Focused; no unrelated changes.
• Follows CONTRIBUTING.md; no site//CNAME/VitePress.
• Linked issue ([Bug]: contributor open-PR monitor groups PRs by case-sensitive repoFullName, splitting one repo into two #1526).

Validation

• git diff --check
• npm run typecheck
• npm run test:coverage locally (4619 passed); the new grouping lines are covered by a case-variant test, and the existing same-casing tests are unchanged.
• npm run test:workers
• npm audit --audit-level=moderate
• New unit test: two open PRs for one repo under different casings group together and the case-sensitive query never runs against the case-variant.

If any required check was skipped, explain why:

• OpenAPI/types/migrations unchanged — internal grouping only; the monitor's output shape is identical.

Safety

• No secrets, wallets, hotkeys, PATs, trust scores, or private evidence exposed.
• Public text sanitized and low-noise (monitor output sanitizer test still passes).
• No auth/CORS/session change.
• API/OpenAPI/MCP behavior unchanged.
• No UI changes.
• No changelog edit.

Notes

• Confined to src/signals/contributor-open-pr-monitor.ts; mirrors the case-insensitive repo handling already present in the same function and in contributor-evidence-graph.ts.

[superagent-security]

Superagent didn't find any vulnerabilities or security issues in this PR.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 95.54%. Comparing base (758e524) to head (d20b607).
⚠️ Report is 42 commits behind head on main.
✅ All tests successful. No failed tests found.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1527   +/-   ##
=======================================
  Coverage   95.53%   95.54%           
=======================================
  Files         204      204           
  Lines       22084    22086    +2     
  Branches     7972     7972           
=======================================
+ Hits        21097    21101    +4     
  Misses        412      412           
+ Partials      575      573    -2     

Files with missing lines	Coverage Δ
src/signals/contributor-open-pr-monitor.ts	98.11% <100.00%> (+1.95%)	⬆️

🚀 New features to boost your workflow:

• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[loopover-orb]

Tip

🟩🟩🟩🟩🟩🟩🟩🟩🟩🟩🟩🟩

✅ Gittensory review — safe to merge

2 files · 1 AI reviewers · no blockers · readiness 55/100 · CI green · clean

✅ Approved — safe to merge

Review summary
The `groupByRepo` case-fold (`pr.repoFullName.toLowerCase()`) is exactly the right fix: it collapses case-variant entries for one repo into a single bucket, matching every other case-insensitive repo-name operation in `buildContributorOpenPrMonitor`. The consumer loop recovers original casing from `repoOpen[0]!.repoFullName`, mirroring the key-by-lower/keep-original pattern from `contributor-evidence-graph.ts`. The new integration test drives the real `buildContributorOpenPrMonitor` path through mocked repositories and correctly asserts both grouping and query-casing behaviour. No blockers.

Signal	Result	Evidence
Code review	✅ No blockers	1 reviewers, synthesized
Linked issue	✅ Linked	#1526
Related work	⚠️ 2 scoped overlaps	Top overlaps are listed below; lower-confidence bulk is hidden.
Review load	❌ 8/20	Readiness component derived from cached public PR metadata and labels; size label size:S.
Validation evidence	❌ 5/25	Cached preflight status is hold.
Open PR queue	❌ 3/10	28 open PR(s), 14 likely reviewable, 14 unlinked.
Contributor context	✅ Confirmed Gittensor contributor	galuis116; Gittensor profile; 1266 PR(s), 156 issue(s).
Gate result	✅ Passing	No configured blocker found.

Nits — 5 non-blocking

• src/signals/contributor-open-pr-monitor.ts:61 — `repoOpen[0]!.repoFullName` as the canonical name for `listPullRequests` and `loadContributorRepoOpenPrSignals` is order-dependent: if the first PR in the bucket carries the atypical casing (`Entrius/Allways-UI`), case-sensitive DB queries use that casing and can return a partial result set for `detectPendingPrScenario`; prefer `repo?.fullName ?? repoOpen[0]!.repoFullName` (restructure so `repo` is derived first from the lowercase key, then `repoFullName` falls back to the PR only when no registered repo matches).
• test/unit/contributor-open-pr-monitor.test.ts:203 — the assertion `expect(listPrSpy).toHaveBeenCalledWith(env, "entrius/allways-ui")` is coupled to mock insertion order: the lowercase PR is listed first so `repoOpen[0]` is always `entrius/allways-ui`; reversing the two mock entries would flip the casing and falsify the assertion without any bug in the fix — add a complementary reversed-order case, or address the root by deriving `repoFullName` from `repo.fullName` to make the assertion order-independent.
• src/signals/contributor-open-pr-monitor.ts:63 and :220 — the two added comment blocks are significantly more verbose than the existing style in this file; both re-explain what the code already expresses — trim each to a single short clause or remove entirely.
• src/signals/contributor-open-pr-monitor.ts:61 — derive `repoFullName` from the registered repo's authoritative `fullName` to make the casing deterministic regardless of which PR arrives first: `const repo = repositories.find(e => e.fullName.toLowerCase() === repoOpen[0]!.repoFullName.toLowerCase()) ?? null; const repoFullName = repo?.fullName ?? repoOpen[0]!.repoFullName;` — this also lets you drop the `!` assertion on the find.
• test/unit/contributor-open-pr-monitor.test.ts — add a second `it` block (or a second `mockResolvedValue` variant) where PR 31 (`Entrius/Allways-UI`) is listed first; it should still assert `listPullRequests` is called with the registered canonical casing, proving the fix is robust to DB-delivery order rather than accidentally passing because of the current fixture ordering.

Review context

• Author: galuis116
• Role context: outside_contributor
• Public audience mode: oss maintainer
• Lane context: Repository registration is not available in the local Gittensory cache.
• Public profile languages: JavaScript, Dart, TypeScript, Python, C++, HTML, MDX, Rust
• Official Gittensor activity: 1266 PR(s), 156 issue(s).
• Related work: Items reference the same linked issue feat(anti-abuse): configurable contributor blacklist — auto-label + auto-close banned logins #1425. (issue #1425, PR #1531)
• Related work: Open PR work references issue feat(anti-abuse): configurable contributor blacklist — auto-label + auto-close banned logins #1425. (issue #1425, PR #1531)

Contributor next steps

• Review top overlaps.
• Add scope summary.
• Fix blocker.
• Expect slower review.
• Refresh registry data or choose a registered active repo.
• Check active issues and PRs before submitting.

Signal definitions

• Related work = same linked issue, overlapping active PRs, or title/path similarity.
• Review load = cached public PR metadata such as size labels, changed paths, and preflight status.
• Open PR queue = repo-wide review pressure; it is not a PR quality failure.
• Contributor context = public GitHub/Gittensor identity context; non-Gittensor status is not a blocker.

Review details

Generated from public PR metadata and the diff. Advisory only; deterministic signals remain authoritative.

The `groupByRepo` case-fold (`pr.repoFullName.toLowerCase()`) is exactly the right fix: it collapses case-variant entries for one repo into a single bucket, matching every other case-insensitive repo-name operation in `buildContributorOpenPrMonitor`. The consumer loop recovers original casing from `repoOpen[0]!.repoFullName`, mirroring the key-by-lower/keep-original pattern from `contributor-evidence-graph.ts`. The new integration test drives the real `buildContributorOpenPrMonitor` path through mocked repositories and correctly asserts both grouping and query-casing behaviour. No blockers.

Nits (5)

• src/signals/contributor-open-pr-monitor.ts:61 — `repoOpen[0]!.repoFullName` as the canonical name for `listPullRequests` and `loadContributorRepoOpenPrSignals` is order-dependent: if the first PR in the bucket carries the atypical casing (`Entrius/Allways-UI`), case-sensitive DB queries use that casing and can return a partial result set for `detectPendingPrScenario`; prefer `repo?.fullName ?? repoOpen[0]!.repoFullName` (restructure so `repo` is derived first from the lowercase key, then `repoFullName` falls back to the PR only when no registered repo matches).
• test/unit/contributor-open-pr-monitor.test.ts:203 — the assertion `expect(listPrSpy).toHaveBeenCalledWith(env, "entrius/allways-ui")` is coupled to mock insertion order: the lowercase PR is listed first so `repoOpen[0]` is always `entrius/allways-ui`; reversing the two mock entries would flip the casing and falsify the assertion without any bug in the fix — add a complementary reversed-order case, or address the root by deriving `repoFullName` from `repo.fullName` to make the assertion order-independent.
• src/signals/contributor-open-pr-monitor.ts:63 and :220 — the two added comment blocks are significantly more verbose than the existing style in this file; both re-explain what the code already expresses — trim each to a single short clause or remove entirely.
• src/signals/contributor-open-pr-monitor.ts:61 — derive `repoFullName` from the registered repo's authoritative `fullName` to make the casing deterministic regardless of which PR arrives first: `const repo = repositories.find(e => e.fullName.toLowerCase() === repoOpen[0]!.repoFullName.toLowerCase()) ?? null; const repoFullName = repo?.fullName ?? repoOpen[0]!.repoFullName;` — this also lets you drop the `!` assertion on the find.
• test/unit/contributor-open-pr-monitor.test.ts — add a second `it` block (or a second `mockResolvedValue` variant) where PR 31 (`Entrius/Allways-UI`) is listed first; it should still assert `listPullRequests` is called with the registered canonical casing, proving the fix is robust to DB-delivery order rather than accidentally passing because of the current fixture ordering.

_{🟩 Safe / merged · 🟦 Advisory · 🟨 Held for review · 🟥 Blocked / closed}

---

💰 Earn for open-source contributions like this. Gittensor lets GitHub contributors earn for the work they already do — register to start earning →.

Checked by Gittensory, a quiet PR intelligence layer for OSS maintainers.

• Re-run Gittensory review