andygrove opened a new pull request, #4404:
URL: https://github.com/apache/datafusion-comet/pull/4404

   ## Which issue does this PR close?
   
   <!-- No dedicated issue: CI infrastructure change to cut pull request build 
time. -->
   
   N/A
   
   ## Rationale for this change
   
   The Spark SQL test workflow runs every Spark version (3.4, 3.5, 4.0, 4.1) on 
every pull request, and the Iceberg test workflows run every Iceberg version 
(1.8.1, 1.9.1, 1.10.0). This is valuable coverage on `main` but is expensive to 
run on every pull request iteration. Reducing the matrix on pull requests 
shortens feedback time and frees runner capacity, while pushes to `main` still 
get the full sweep so regressions on the less-common versions are still caught 
before release.
   
   ## What changes are included in this PR?
   
   The varying matrix dimension in each workflow is now selected with a 
conditional `fromJSON` expression keyed on `github.event_name`:
   
   - `spark_sql_test.yml`: pull requests test only Spark 3.5 and 4.1; pushes to 
`main` and manual `workflow_dispatch` runs test the full set (3.4, 3.5, 4.0, 
4.1). The 7-entry `module` matrix is unchanged, so pull request runs drop from 
28 jobs to 14.
   - `iceberg_spark_test.yml`: pull requests test only the latest Iceberg 
version (1.10.0); pushes to `main` and manual `workflow_dispatch` runs test the 
full set (1.8.1, 1.9.1, 1.10.0). This applies to all three Iceberg jobs 
(`iceberg-spark`, `iceberg-spark-extensions`, `iceberg-spark-runtime`).
   
   Manual `workflow_dispatch` runs intentionally use the full matrix so the 
complete sweep can still be triggered on demand.
   
   ## How are these changes tested?
   
   These are CI configuration changes. Both modified workflows pass 
`actionlint` (the same check run by `validate_workflows.yml`), and the inline 
matrix JSON was verified to parse. The conditional matrix behavior will be 
observable on this pull request (reduced matrix) versus subsequent pushes to 
`main` (full matrix).


-- 
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]

Reply via email to