mirror of
https://github.com/storytold/effectcraft.git
synced 2026-10-28 20:16:52 -06:00
ci: run the full source quality gate on pull requests
This commit is contained in:
@@ -0,0 +1,23 @@
|
||||
# Source PR CI evidence
|
||||
|
||||
`ci.yml` runs the existing `cargo xtask ci` gate on opened, synchronized,
|
||||
reopened and ready-for-review PRs, without source/path or draft filters, plus main
|
||||
pushes and manual dispatch. The existing platform packaging triggers are unchanged.
|
||||
The read-only hosted Linux job tests the PR merge result and records both head
|
||||
and checkout identities. No PR-target execution or signing/deployment credentials
|
||||
are used. The 150-minute cold-build limit is provisional for this new job layout.
|
||||
|
||||
The runner installs Mesa Lavapipe and probes its Vulkan adapter. A failed probe
|
||||
does not prevent CPU gates from running, but fails the final coverage report.
|
||||
`RUST_TEST_NOCAPTURE=1` exposes the existing GPU tests' adapter/parity/skip messages.
|
||||
The reporter requires a successful gate, executed Rust tests and positive GPU
|
||||
evidence; runtime no-adapter skips fail acceptance. Intentionally ignored platform
|
||||
cases are counted/reported separately. Complete logs and source identity are saved
|
||||
as artifacts. Software Vulkan coverage does not qualify physical GPU drivers,
|
||||
other operating systems or tests deliberately ignored on Linux.
|
||||
|
||||
Run the report regression tests with
|
||||
`python3 -m unittest discover -s .github/scripts -p 'test_*.py' -v`.
|
||||
Trigger simulation and fixture tests do not prove GitHub events or runner images
|
||||
work: after approved publication/landing, inspect actual PR run URLs and results,
|
||||
including a ready-for-review event, before requiring the new check for merges.
|
||||
Reference in New Issue
Block a user