Repository navigation
perf(ci): Cut Docker smoke-test time on large runners - #588
Conversation
|
You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard. |
![cubic-dev-ai[bot]](https://avatars.githubusercontent.com/in/1082092?s=60&v=4){kind=small}
There was a problem hiding this comment.
All reported issues were addressed across 2 files
Shadow auto-approve: would not auto-approve because issues were found.
Re-trigger cubic
CyMule
changed the title
|
You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard. |
There was a problem hiding this comment.
No issues found across 2 files
Shadow auto-approve: would not auto-approve. This PR does not meet the repository auto-approval settings.
Re-trigger cubic
There was a problem hiding this comment.
0 issues found across 2 files (changes from recent commits).
Shadow auto-approve: would not auto-approve. This PR does not meet the repository auto-approval settings.
Re-trigger cubic
2 participants
What
Two changes to the Docker smoke-test job:
/and Docker's storage filesystem have at least 30 GiB free. Smaller runners still clean up as before..gzformat checks use the one-page PDF instead of the full paper. The.gzcase compresses that PDF in pytest's temp directory, so the unusedlayout-parser-paper.pdf.gzfixture is removed.Why
These checks only confirm that each file type is accepted and parsed. The full paper is still processed by the table-extraction, strategy, and parallel-mode smoke tests, so the generic format cases don't need it. All 37 cases and their explicit-MIME, inferred-MIME, and CSV assertions remain.
Validation
actionlint and Ruff pass, and all 37 smoke tests collect. Both changed cases were run against a controlled response, and the gzip decompresses byte-for-byte to the source PDF. Both disk branches (enough space and low space) were tested with mocked commands. Full timing comes from this PR's CI (baseline).