Github user marmbrus commented on a diff in the pull request:
https://github.com/apache/spark/pull/482#discussion_r12265745
--- Diff:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/Optimizer.scala
---
@@ -87,6 +88,56 @@ object ColumnPruning extends Rule[LogicalPlan] {
/**
* Replaces [[catalyst.expressions.Expression Expressions]] that can be
statically evaluated with
+ * equivalent [[catalyst.expressions.Literal Literal]] values. This rule
is more specific with
+ * Null value propagation from bottom to top of the expression tree.
+ */
+object NullPropagation extends Rule[LogicalPlan] {
+ def apply(plan: LogicalPlan): LogicalPlan = plan transform {
+ case q: LogicalPlan => q transformExpressionsUp {
+ // Skip redundant folding of literals.
+ case l: Literal => l
--- End diff --
Yes, but in the case of `ConstantFolding` the subsequent pattern _will_
match `Literal`, since a `Literal` is technically `foldable`. Matching the
next pattern causes the rule to invoke the expression evaluator and create an
identical, wasted object.
In `NullPropogation`, a `Literal` will not match any of the later rules.
So in essence you are second guessing the code generated by the pattern
matcher. While there may be extreme cases where that is required for
performance, I don't think this is one of them.
---
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.
---
