Skip to content

test: stabilize resume polling boundary under CI load - #20

Merged
AnsonDev42 merged 1 commit into
mainfrom
codex/stabilize-resume-ci-test
Nov 11, 2026
Merged

AnsonDev42 merged 1 commit into
mainfrom
codex/stabilize-resume-ci-test

Conversation

@AnsonDev42

Copy link
Copy Markdown
Owner

The resume-polling test assumed the initial materialization slice always reached exactly 1,000 rows. On a busy Python 3.12 tag runner, the time budget correctly flushed 999 rows first, causing a false assertion failure. A controlled clock reproduced the same failure against both the pre-review base and the released runtime.

Pin this test's work-slice clock so its count-boundary and poll-before-materialize assertions are deterministic; retain separate time-budget tests and the real polling clock. Document the testing rule. Runtime code, version metadata and the published v0.0.8 tag remain unchanged.

Proof

  • uv run pytest -q: 345 passed, 9 skipped; coverage gate retained.
  • Ruff lint/format and strict MkDocs build passed.
  • Independent review and all current-head CI required before merge.
  • No provider requests.

@AnsonDev42
AnsonDev42 marked this pull request as ready for review November 11, 2026 23:06
@AnsonDev42
AnsonDev42 merged commit 6dad50a into main Nov 11, 2026
16 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant