Repository navigation
fix(ci): stabilize Python 3.15 quality and recovery checks - #247
Conversation
There was a problem hiding this comment.
Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.
An organization owner can choose the repository organization as the billing source for reviews requested by members with a paid Copilot license. Learn about Copilot code review billing.
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
|
复核 2026-11-05 的当前状态:本 PR head 这里的 PR CI 已覆盖当前 当前 main 自身仍有四组失败:quality artifact 记录 以上是现有 CI、产物和补丁的复核,没有重跑或修改代码。现有自动 review 摘要针对旧 head |
… ci] Preserve PR #248 fingerprint boundary changes. The combined tree passed all 48 recovery, live-agent-space and internet tests against an isolated disk-backed PostgreSQL cluster.
Problem
Python 3.15 exposed three independent issues: Ruff/format drift in
deploy/check_home_seo.py; pytest-asyncio's unset fixture-loop-scope warning becoming an error; and restored applications retaining a certificate validator bound to the pre-restore clock.Changes
CertificateValidatora forwarding clock so replacingApplication.clockafter restore takes effect immediately.Original PR head:
70007351fe1872899b7c623b99169e88a47107a6.Verified current base:
c71a41d8b2b3601c63e734c344c8e58b5a52b676.Integration head:
2c65570a10fe72a52464d2adcea06a4f64e31964.Verified tree:
bf8a2d0fa55529a0ca88191aa1858e58734cd77a.Current development-machine validation — 2026-11-20
The exact combined tree was tested in an independent checkout, with Python 3.15.2, pytest 9.1.1, and pytest-asyncio 1.4.0.
.venv/bin/ruff check .: passed..venv/bin/ruff format --check .: passed..venv/bin/python -W error -m compileall -q src tests conformance scripts: passed..venv/bin/pytest -vv --tb=short tests/test_complete_recovery_proof.py tests/test_live_agent_space.py tests/test_internet.py: 48 passed, 0 failed, in 540.75 seconds.git diff --checkand clean-source readback passed.d69f2111f1566dd825bb70a5b8ddf3c4bec4ccf6; the GitHub integration tree above is byte-for-byte identical.Evidence:
/home/lightjunction/Documents/GITHUB/pr-cleanup-org-20261120/evidence-msg/, including10-full-disk.json,10-full-disk.log, and the PostgreSQL log.Preserved environment failures and their correction
Earlier attempts are retained as failures, not counted as acceptance:
/tmpis an 8 GiB tmpfs; only 508 MiB remained at diagnosis. Its 831.6 MiB scope peak distinguishes that failure from the earlier OOM./tmp. The private cluster and test directories were cleaned by the task's own finalizer.The Internet subset also passed all 15 tests independently before the complete 48-test run. No timeout, quality gate, assertion, or compatibility check in the repository was weakened. Registry-mirror-only lockfile churn was discarded; the original dependency lock was retained.
Scope
All 32 current workflow files were inspected for deployment consequences. They describe validation, including an isolated installed-wheel deployment rehearsal; this change does not invoke production deployment. No workflow was manually dispatched or rerun, and no release, tag, real account authorization, or live payment was performed. This targeted acceptance does not claim the entire repository's optional platform/integration matrix.