maropu commented on a change in pull request #29092:
URL: https://github.com/apache/spark/pull/29092#discussion_r500694652
##########
File path:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/Optimizer.scala
##########
@@ -1854,3 +1855,37 @@ object OptimizeLimitZero extends Rule[LogicalPlan] {
empty(ll)
}
}
+
+/**
+ * Infers filters from [[Generate]], such that rows that would have been
removed
+ * by this [[Generate]] can be removed earlier - before joins and in data
sources.
+ */
+object InferFiltersFromGenerate extends Rule[LogicalPlan] {
+ def apply(plan: LogicalPlan): LogicalPlan = plan transformUp {
+ case generate @ Generate(e: ExplodeBase, _, false, _, _, _)
+ if e.deterministic && !e.child.foldable =>
+
+ inferFiltersForExplodeLike(generate, e)
+
+ case generate @ Generate(e: Inline, _, false, _, _, _)
+ if e.deterministic && !e.child.foldable =>
+
+ inferFiltersForExplodeLike(generate, e)
+ }
+
+ private def inferFiltersForExplodeLike(generate: Generate, expr:
UnaryExpression): Generate = {
Review comment:
Just a nit suggestion; how about reformatting it like this?
```
def apply(plan: LogicalPlan): LogicalPlan = plan transformUp {
case generate @ Generate(e, _, _, _, _, _)
if !e.deterministic || e.children.forall(_.foldable) => generate
case generate @ Generate(g, _, false, _, _, _) if canInferFilters(g) =>
// Exclude child's constraints to guarantee idempotency
val inferredFilters = ExpressionSet(
Seq(
GreaterThan(Size(g.children.head), Literal(0)),
IsNotNull(g.children.head)
)
) -- generate.child.constraints
if (inferredFilters.nonEmpty) {
generate.copy(child = Filter(inferredFilters.reduce(And),
generate.child))
} else {
generate
}
}
private def canInferFilters(g: Generator): Boolean = g match {
case _: ExplodeBase => true
case _: Inline => true
case _ => false
}
```
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
