fix: run-tests timeout can cancel in-flight Test Runner jobs

[hatayama]

Summary

• After a run-tests timeout or cancel, Unity can now request Test Runner to stop the in-flight job (EditMode included), not only exit Play Mode.
• On Test Framework versions where cancel helpers are unavailable or changed, behavior automatically falls back to the existing bounded Play Mode exit path.

User Impact

• Before: Cancel/timeout relied on Play Mode exit only; EditMode jobs and some PlayMode cases could keep running until they finished on their own.
• After: When Unity Test Framework exposes CancelTestRun, uloop uses it with the run id from Execute. If IsRunActive() is missing (e.g. TF 2.0 guid overload), cancel still proceeds via the public Play Mode exit + capped wait without retrying reflection.

Changes

• Keep and pass the Execute run GUID into cancel stop/restore.
• Resolve CancelTestRun(string) with Public | NonPublic once; cache success and failure.
• Resolve parameterless IsRunActive() only; miss → Option B fallback.
• Pure unit tests for TF 1.3.9 / 1.4 / 2.0 shaped lookup.

Verification

• dotnet test tests/RunTestsExecutionTimeout.UnitTests/RunTestsExecutionTimeout.UnitTests.csproj
• dist/darwin-arm64/uloop compile --project-path "$(git rev-parse --show-toplevel)" (0 errors / 0 warnings)
• CI on this PR

Made with Cursor

[coderabbitai]

Warning

Review limit reached

@hatayama, you've reached your PR review limit, so we couldn't start this review.

Next review available in: 5 minutes

Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available.
You're only billed for reviews past your plan's rate limits ($0.25/file).

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 39b41ecb-1a46-4d0a-ae2f-bd83182c72d5

📥 Commits

Reviewing files that changed from the base of the PR and between 67427ea and 26869df.

⛔ Files ignored due to path filters (1)

• Assets/Tests/Editor/TestRunnerApiCancelMethodLookupEditModeTests.cs.meta is excluded by none and included by none

📒 Files selected for processing (2)

• Assets/Tests/Editor/TestRunnerApiCancelMethodLookupEditModeTests.cs
• Packages/src/Editor/FirstPartyTools/RunTests/TestFramework/TestRunnerApiCancelBridge.cs

📝 Walkthrough

Walkthrough

Changes

Test cancellation flow

Layer / File(s)	Summary
Resolve cancellation API methods Packages/src/Editor/FirstPartyTools/RunTests/TestFramework/TestRunnerApiCancelMethodLookup.cs, tests/RunTestsExecutionTimeout.UnitTests/TestRunnerApiCancelMethodLookupTests.cs	Reflection resolves CancelTestRun(string) and parameterless IsRunActive(), with tests covering supported, public, mismatched, and missing method shapes.
Cache and invoke cancellation methods Packages/src/Editor/FirstPartyTools/RunTests/TestFramework/TestRunnerApiCancelBridge.cs	The bridge caches resolved methods, reports fallback logs, exposes capability checks, and safely invokes cancellation and active-run polling.
Propagate run identifiers during cancellation Packages/src/Editor/FirstPartyTools/RunTests/TestFramework/PlayModeTestExecuter.cs, Packages/src/Editor/FirstPartyTools/RunTests/TestFramework/RunTestsCancelStopRestoreUnityHooks.cs	Test execution returns and captures runGuid, passes it to stop/restore cancellation, and conditionally wires the reflected Unity hooks.

Estimated code review effort: 3 (Moderate) | ~20 minutes

Possibly related PRs

• hatayama/unity-cli-loop#1254: Related to Play Mode test execution recovery and DomainReloadDisableScope handling.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 21.74% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly matches the main change: improving run-tests timeout/cancel behavior for in-flight Test Runner jobs.
Description check	✅ Passed	The description accurately describes the cancellation, fallback behavior, and added tests in this changeset.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch feat/run-tests-cancel-testrun-reflection

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[hatayama]

LGTM (reviewer: Fable). Follow-up verified at 26869df: the EditMode test asserts the lookup resolves CancelTestRun(string) and parameterless IsRunActive() against the real TF 1.3.9 TestRunnerApi (so a future TF upgrade in this repo fails loudly instead of silently degrading), and _resolved is now volatile. The merge-gate device verification also passed: an intentional EditMode infinite-yield hang with --timeout-seconds 8 timed out with a message showing the CancelTestRun path succeeded, the immediate follow-up EditMode run passed without an Editor restart, and IsRunActive reported false afterward. Option A wiring is complete and version-compatible from TF 1.3.9 through the 2.0 previews.