vranes opened a new pull request, #56815:
URL: https://github.com/apache/spark/pull/56815

   ### What changes were proposed in this pull request?
   
   The two BIN BY tests that assert the operator is rejected now pin 
`spark.sql.binByRelationOperator.enabled` explicitly (`= false`) via 
`withSQLConf`, instead of relying on the config's default:
   
   - `BinBySuite` and `ResolveBinBySuite`: the "BIN BY is gated off by default" 
cases are renamed to "BIN BY is rejected when the operator is disabled" and 
wrap the query in `withSQLConf(BIN_BY_ENABLED.key -> "false")`.
   - `ResolveBinBySuite` also drops the 
`assert(!SQLConf.get.getConf(BIN_BY_ENABLED))` check, which asserted the 
default value directly.
   
   No production code changes; the config default stays `false`.
   
   ### Why are the changes needed?
   
   Tests should not depend on a config's default value. A case that runs 
without pinning the flag fails in any environment that flips the config on (for 
example a config-flip CI lane), even though the behavior under test is 
unchanged. Pinning the flag explicitly keeps the suites self-contained and 
stable across those lanes.
   
   ### Does this PR introduce _any_ user-facing change?
   
   No. Test-only change; the `spark.sql.binByRelationOperator.enabled` default 
is unchanged.
   
   ### How was this patch tested?
   
   Ran the updated suites locally:
   
   - `build/sbt 'catalyst/testOnly *ResolveBinBySuite'` (17 tests, all pass)
   - `build/sbt 'sql/testOnly *BinBySuite'` (3 tests, all pass)
   
   scalastyle is clean for both modules.
   
   ### Was this patch authored or co-authored using generative AI tooling?
   
   Generated-by: Claude Code (Claude Opus 4.8)
   


-- 
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