Yicong-Huang opened a new pull request, #59195:
URL: https://github.com/apache/spark/pull/59195
### What changes were proposed in this pull request?
The Arrow eval type handler tests built handlers with a hand-written
`_RunnerConf` stand-in that exposed only a few hard-coded fields. This replaces
it with the worker's real `RunnerConf`, constructed from a plain conf dict
(`RunnerConf({...})`), matching the pattern already used by the incremental
aggregate/window handler tests added in SPARK-59727. Window-frame tests now
pass the bound types through the real conf key
(`RunnerConf({"window_bound_types": "unbounded"})` / `"bounded"`) instead of
mutating a fake attribute. The identical `_RunnerConf` stand-in in the sibling
`test_base_eval_type_handlers.py` is removed the same way.
### Why are the changes needed?
The fake config duplicated field defaults that can silently drift from the
real `RunnerConf`, and it hid the actual conf keys the handlers read. Building
the real config from a conf dict exercises the same code path as production and
keeps the tests honest. This is a test improvement following the Arrow eval
handler migrations under SPARK-59415.
### Does this PR introduce _any_ user-facing change?
No. Test-only change.
### How was this patch tested?
`pyspark.eval_handlers.tests.test_arrow_eval_type_handlers` and
`pyspark.eval_handlers.tests.test_base_eval_type_handlers` pass (38 tests).
### Was this patch authored or co-authored using generative AI tooling?
Generated-by: Claude Opus 4.8
This pull request and its description were written by Isaac.
--
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]
