Github user viirya commented on a diff in the pull request:
https://github.com/apache/spark/pull/22417#discussion_r217619116
--- Diff:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/CodeGeneratorWithInterpretedFallback.scala
---
@@ -59,6 +59,6 @@ abstract class CodeGeneratorWithInterpretedFallback[IN,
OUT] extends Logging {
}
}
- protected def createCodeGeneratedObject(in: IN): OUT
+ protected def createCodeGeneratedObject(in: IN,
subexpressionEliminationEnabled: Boolean): OUT
--- End diff --
If we are going to get config directly from `SQLConf`, why do we need to
pass it as a parameter here?
Can we keep this API untouched and let `createCodeGeneratedObject`
implementation to take care of it?
Like:
```scala
object UnsafeProjection
extends CodeGeneratorWithInterpretedFallback[Seq[Expression],
UnsafeProjection] {
override protected def createCodeGeneratedObject(in: Seq[Expression]):
UnsafeProjection = {
GenerateUnsafeProjection.generate(in,
SQLConf.get.subexpressionEliminationEnabled)
}
...
}
```
I think it is much simpler.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
