test(web): fix the ConnectedApps flake that failed two deploys - #699
Merged
Merged
Conversation
The mock returned a new t per render; the component's load effect depends on t, so every render reloaded the list and a late reload restored a disconnected row. Assert one load. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Why
ConnectedApps › Disconnect revokes and removes only that rowfailed the #691 deploy (2026-10-04, re-run passed) and the #698 deploy, and a deploy on 2026-10-01 (then "fixed" with a 3s timeout).Root cause
The test's
useTranslationmock returned a newton every render.ConnectedAppsloads grants inuseEffect(..., [isAuthenticated, t]), so in the test every render re-ran the load; after Disconnect filtered "Claude" out, a late reload resolved the full list again and put it back. Timing-dependent → worse on a loaded runner. The real hook memoizest(useCallback), so production is unaffected.Fix
Hoisted one stable
tfor the mock; dropped the 3s timeout; addedexpect(listOAuthGrants).toHaveBeenCalledTimes(1), which fails with the old mock (verified) — so the cause can't come back silently.Verified
5/5 runs green locally; reverting the mock makes the new assertion fail.
🤖 Generated with Claude Code