rich7420 commented on code in PR #6845:
URL: https://github.com/apache/datafusion-comet/pull/6845#discussion_r4301795339
##########
native/spark-expr/src/map_funcs/map_extract.rs:
##########
@@ -206,6 +208,28 @@ fn spark_map_extract(
// representation the format permits rather than one known to arrive here.
Guard it anyway:
// Spark returns NULL for a NULL map under both ANSI modes, for
`element_at` and `GetMapValue`
// alike, and only `element_at` has a nullable-input guard upstream of
this kernel.
+ if matches!(value_type, DataType::List(_) | DataType::LargeList(_)) {
Review Comment:
Addressed at `0de9779bf`: nested string/binary children now use typed range
copies, with short/skewed nested-string benchmarks covering empty and retained
entry ranges under null maps, plus sliced-input regressions.
The paired release probe for the reported shape measured 9.8% faster than
main with 75% null maps and 19.5% faster without null maps. Integer controls
were sensitive to relinking, so this is not a blanket no-regression claim. All
27 focused Rust tests, 10 Spark 4.1.3 native SQL queries and [upstream
CI](https://github.com/apache/datafusion-comet/actions/runs/38160028540) passed.
Could you apply `run-spark-4.1-tests` for broader coverage? I tried the fork
CI path, but GitHub rejects dispatch with “Actions has been disabled for this
repository.”
--
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]
