Github user rednaxelafx commented on the issue:
https://github.com/apache/spark/pull/20224
One comment as to using `ThreadLocal[Integer]` for keeping track of the
IDs: I did have an alternative implementation of this PR that declares
`WholeStageCodegenExec` as:
```scala
case class WholeStageCodegenExec(child: SparkPlan)(private val
codegenStageId: Int)
extends UnaryExecNode with CodegenSupport
```
and then explicitly thread the `codegenStageId` recursively in
`CollapseCodegenStages.insertWholeStageCodegen()`, so that the relationship
between the auto-increment of IDs and the insertion order of
`WholeStageCodegenExec`s are explicit.
However that turned out to be much more complicated than just using a
`ThreadLocal[Integer]` and implicitly threading the IDs. So in the end I opted
for the thread-local counter version instead.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
