Copilot commented on code in PR #12444:
URL: https://github.com/apache/gluten/pull/12444#discussion_r3530658512
##########
backends-velox/src-delta40/main/scala/org/apache/spark/sql/delta/GlutenOptimisticTransaction.scala:
##########
@@ -48,6 +49,14 @@ class GlutenOptimisticTransaction(delegate:
OptimisticTransaction)
writeOptions: Option[DeltaOptions],
isOptimize: Boolean,
additionalConstraints: Seq[Constraint]): Seq[FileAction] = {
+ // Velox has no Arrow representation for VariantType: the native write
path inserts a
+ // RowToVeloxColumnarExec transition whose SparkArrowUtil.toArrowSchema
call throws
+ // `Unsupported data type: variant` at runtime. Delegate writes whose
schema contains a
+ // variant column to the vanilla Delta write path instead of offloading
them.
+ if (inputData.schema.existsRecursively(_.isInstanceOf[VariantType])) {
+ logInfo("Input schema contains VariantType; falling back to the vanilla
Delta write path.")
+ return super.writeFiles(inputData, writeOptions, isOptimize,
additionalConstraints)
+ }
hasWritten = true
Review Comment:
The early `return super.writeFiles(...)` path skips the `hasWritten = true`
assignment that this override uses to track that a write occurred. Even if
`super.writeFiles` currently sets it, keeping `hasWritten` consistent in this
override avoids behavior differences between the native/offloaded and fallback
paths.
--
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]
