Repository navigation
fix(website): align events e2e past-card filter with the page's end-time status model (red main repair) - #17613
christian-byrne wants to merge 1 commit into
Conversation
…ime status model The 'cards carry the same upcoming and past CTAs' test picked the past card with a naive startDateTime < Date.now() filter, but the page derives past/upcoming from eventStatus (end-time based: start + 1h default duration). Once a linked event (UCAN, 2026-09-13T14:00-07:00) started, the test treated its still-upcoming card as past, expected a LEARN MORE link that the page correctly does not render, and failed red main (CI: Website E2E push run 34782685000 on 8190935). Select the past card from pastEvents - the same derived, end-time-based model the page renders - so the test and page agree by construction.
🌐 Website E2ETip All tests passed.
🔗 Website PreviewWebsite Preview: https://comfy-website-preview-pr-17613.vercel.app Matches production — what ships if we release right now. This commit: https://website-frontend-pvbdm07du-comfyui.vercel.app Last updated: 2026-09-13T21:40:59Z for |
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: ASSERTIVE Plan: Team Run ID: 📒 Files selected for processing (1)
Included review availability: Your plan provides up to 10 included reviews per hour; 6 remain after this review. 📝 WalkthroughWalkthroughThe past-card E2E test now selects a linked event from ChangesEvents E2E Coverage
Priority: ⬇️ Low Estimated code review effort: 2 (Simple) | ~10 minutes Change: Bug fix Suggested reviewers: Merge Risk: ⚪ Minimal · up to This change removes the timing-sensitive event selection that caused the website E2E failure, with no product-code impact. 🚥 Pre-merge checks | ✅ 7✅ Passed checks (7 passed)
✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Comment |
Codecov Report✅ All modified and coverable lines are covered by tests. @@ Coverage Diff @@
## main #17613 +/- ##
=======================================
Coverage 82.71% 82.71%
=======================================
Files 2730 2730
Lines 137651 137651
Branches 37213 37213
=======================================
Hits 113856 113856
Misses 22786 22786
Partials 1009 1009
Flags with carried forward coverage won't be shown. Click here to find out more. 🚀 New features to boost your workflow:
|
|
Dup. Same repair carried by #17612 + bbc #723. Full context for agent readersConcurrent sweeper lane filed this first. Same root cause (test's start-time past filter vs the page's end-time-based eventStatus during a linked event's live window, UCAN started 21:00Z). Closing to keep one repair PR for review. Root-cause evidence also in that thread. |
Justification: red
mainrepair —CI: Website E2Epush run on 8190935 failed at exact head; test-only root-cause fix, no product code touched.Evidence
main, head8190935cd, jobwebsite-e2efailed 21:12:21Z, failed through retries:[desktop] e2e/events.spec.ts:508— cards carry the same upcoming and past CTAs as the list;LEARN MORElink "element(s) not found" on the UCAN card; 517 passed, 1 flaky (visual screenshot, passed on retry)).Date.parse(event.startDateTime) < Date.now()filter, but the page derives past/upcoming viaeventStatus, which is end-time based (startDateTime+ 1h default duration). The UCAN event starts2026-09-13T14:00:00-07:00= 21:00:00Z, four minutes before the run, so during its live window the page still renders the card as upcoming (calendar menu / Register CTA, no LEARN MORE link) while the test expected a past CTA. The same test passed on the prior main E2E run about 30 min earlier on essentially identical code; the head commit (1.55.8 version bump, #17609) only touchespackage.jsonand locale files — the diff is unrelated.pastEvents(the same derived, end-time-based model the page renders) instead of the start-time filter. Test-only, one hunk; oxfmt check clean. The failure self-heals after about 01:30Z but recurs for every future linked event during its live window, so the filter is fixed rather than skipped (no FLAKY skip — the test still runs and passes in and out of live windows).