Github user marmbrus commented on a diff in the pull request:
https://github.com/apache/spark/pull/11256#discussion_r53875740
--- Diff:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/Optimizer.scala
---
@@ -313,97 +313,85 @@ object SetOperationPushDown extends Rule[LogicalPlan]
with PredicateHelper {
*/
object ColumnPruning extends Rule[LogicalPlan] {
def apply(plan: LogicalPlan): LogicalPlan = plan transform {
- case a @ Aggregate(_, _, e @ Expand(projects, output, child))
- if (e.outputSet -- a.references).nonEmpty =>
- val newOutput = output.filter(a.references.contains(_))
- val newProjects = projects.map { proj =>
- proj.zip(output).filter { case (e, a) =>
+ // Prunes the unused columns from project list of
Project/Aggregate/Window/Expand
+ case p @ Project(_, p2: Project) if (p2.outputSet --
p.references).nonEmpty =>
+ p.copy(child = p2.copy(projectList =
p2.projectList.filter(p.references.contains)))
+ case p @ Project(_, a: Aggregate) if (a.outputSet --
p.references).nonEmpty =>
+ p.copy(
+ child = a.copy(aggregateExpressions =
a.aggregateExpressions.filter(p.references.contains)))
+ case p @ Project(_, w: Window) if (w.outputSet --
p.references).nonEmpty =>
+ p.copy(child = w.copy(
+ projectList = w.projectList.filter(p.references.contains),
+ windowExpressions =
w.windowExpressions.filter(p.references.contains)))
+ case a @ Project(_, e @ Expand(_, _, grandChild)) if (e.outputSet --
a.references).nonEmpty =>
+ val newOutput = e.output.filter(a.references.contains(_))
+ val newProjects = e.projections.map { proj =>
+ proj.zip(e.output).filter { case (e, a) =>
newOutput.contains(a)
}.unzip._1
}
- a.copy(child = Expand(newProjects, newOutput, child))
+ a.copy(child = Expand(newProjects, newOutput, grandChild))
+ // TODO: support some logical plan for Dataset
--- End diff --
Maybe create a JIRA for this. Should be pretty easy.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
