zhang-arvin opened a new pull request, #18105:
URL: https://github.com/apache/iceberg/pull/18105
## Purpose
Fixes #18103. Resolves flaky test failures in `TestViews` caused by
colliding temporary function names.
## Root cause
`viewName()` generated suffixes with `new Random().nextInt(1000000)`, which
can produce duplicate names across the parameterized catalog runs of the same
SparkSession (e.g. `readFromViewReferencingTempFunction` creates a temporary
function named `viewName("test_avg")` on each catalog and never drops it),
leading to `ROUTINE_ALREADY_EXISTS` errors. The same pattern exists in
`createViewReferencingTempFunction` and
`createViewReferencingQualifiedTempFunction` for Spark 3.5 / 4.0 / 4.1 / 4.2.
## Fix
Replace the random suffix with a process-unique monotonically increasing
`AtomicInteger` counter (`NAME_SUFFIX.incrementAndGet()`), so names can never
collide within a JVM run and the `java.util.Random` import is removed. This
also applies to the shared `viewName()` helper used by all other view tests.
Files changed:
- spark/v3.5/spark-extensions/.../TestViews.java
- spark/v4.0/spark-extensions/.../TestViews.java
- spark/v4.1/spark-extensions/.../TestViews.java
- spark/v4.2/spark-extensions/.../TestViews.java
--
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]
