Repository navigation
test(apps): replace the fixed wait in app_art_route with await_until - #20274
Conversation
The declared-FIFO test bounded its request with asyncio.wait_for(..., timeout=10). The ceiling exists only to turn a hang into a failure, but a 10 s number is a bet on the runner's speed, which is the commonest source of a flaky test in this repository. Await the request with await_until instead: it fails by name after a lost-run bound of half the test's own timeout, and the status assertion is unchanged. The coroutine is wrapped in asyncio.ensure_future because await_until takes an event or a future. The file's line in wait_discipline_baseline.txt is removed, since its count is now zero.
UX Review (Opus 5.5) — ⏭️ skippedRevision |
First Principles Review (Opus 5.5) — ⏭️ skippedRevision |
Design Review (Opus 5.5) — ✅ PASSDesign-level review of Design-Verdict: PASS Nothing to check. [DESIGN-REVIEWED] bf911b4 Description read[DESCRIPTION-READ] 020a294fbdcdbd7427253ddf784946b01276d8e7c3ce7daccca75a2db5da1596 That sha256 names everything this verdict read: the PR description, plus the 1 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with |
Opus 5.5 Review — ✅ no blocking findingsReviewed Verdict parsed from the review's SHA-scoped output markers for commit False positive or not applicable? A repository writer can comment: |
GPT 6.1 Review — ✅ no blocking findingsGPT 6.1 completed its review of This comment is updated in place on each push. Review detailsNo findings. False positive or not applicable? A repository writer can comment: |
|
Closing and reopening so CI runs on hosted runners: same-repo runs from this account queue on the fleet label without a runner (details on #19683). |
KiroCrewCaptain
left a comment
There was a problem hiding this comment.
Tier 1 auto-approve: test (2 files). Criteria: no conflict, no requested changes, security path denylist clean, design-doc gate clean, security checklist all-NO, no reviewer-lane findings, readiness signal fresh for this head SHA, declared category matches the diff. Category: test-only; replaces a fixed asyncio.wait_for timeout in test_app_art_route.py with await_until and drops the file from the wait-discipline baseline.
Problem / Motivation
Goal: Short fixed waits in 1 test file are replaced by waits on the signal they were polling for, so a slow CI runner can no longer turn them red.
The wait-discipline ratchet counts, per file, the sites the repo still tolerates. This PR retires:
1 test_app_art_route.py.Why it matters
A fixed
wait_for(..., timeout=N)is a bet on the runner's speed: on a loaded shard the event arrives after N and the test fails for a reason unrelated to the code under test. The repo tracks these sites per file intest/wait_discipline_baseline.txtand the ratchet lets each count only go down; retiring them is how the suite gets quieter.Not a goal
What changed (motivation → approach → change)
Each site below replaced
asyncio.wait_for(<signal>, timeout=N)withawait_untilfromkiro_crew.testing.wait, which waits on the same signal with the suite's lost-run ceiling instead of a per-site bet.test/test_app_art_route.py:status, _ = await asyncio.wait_for(_get(f"/apps/{APP}/art/assets/piped.webp"), timeout=10)test/wait_discipline_baseline.txt: removed1 test_app_art_route.py.Backwards compatibility
Compatible: test-only change
Removes nothing: a baseline entry is a tolerance the gate stops needing, not a capability.
Tests
test/test_wait_discipline_ratchet.pyfails (1 failed, 38 passed), proving the gate still counts the old site.-W error::RuntimeWarningso every replaced wait is proven awaited; the file's own pass count is unchanged.Manual verification
N/A — unit coverage sufficient: the gate and the file's own tests exercise the changed lines directly.
Screenshots / video
Why no screenshot: nothing renders; this change touches test code and a CI baseline only.
No issue exists; burn-down of the wait-discipline ratchet baseline (precedents #19928, #20014, #19527).