manuzhang opened a new pull request, #16800:
URL: https://github.com/apache/iceberg/pull/16800
## Summary
This updates Spark CI to generate its test matrix from the files changed in
a pull request.
For pull requests that only touch version-scoped Spark directories such as
`spark/v4.1/**`, the workflow now runs only that Spark version's jobs. Changes
to shared code, build files, workflow files, docs mixed with Spark changes, or
any unknown path continue to run the full Spark matrix.
## Why
The existing workflow always expands the full Spark matrix once Spark CI is
triggered, so a Spark 4.1-only change still runs Spark 4.0 jobs. This keeps
coverage for shared changes while avoiding unrelated version-specific test jobs.
The detector intentionally does not duplicate the workflow's `paths-ignore`
list. It only narrows the matrix when every changed path is under a
version-scoped Spark directory; otherwise it falls back to the full matrix.
## Validation
- Parsed `.github/workflows/spark-ci.yml` with Ruby YAML loading.
- Ran `bash -n` against the generated detector script.
- Simulated changed-file detection for:
- `spark/v4.1/**` selecting only Spark 4.1 jobs
- `.github/workflows/spark-ci.yml` selecting the full matrix
- docs plus `spark/v4.1/**` selecting the full matrix
- Ran `git diff --check`.
`actionlint` was not installed locally, so I could not run it before
updating the draft PR.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
