diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/Optimizer.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/Optimizer.scala index 0df16b7a56c56310803ea9e66a4dbd0ed4dc3cc7..4952ba3b2b99d56aad310e4a22d7b1090df62213 100644 --- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/Optimizer.scala +++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/Optimizer.scala @@ -710,7 +710,7 @@ object PushDownPredicate extends Rule[LogicalPlan] with PredicateHelper { val (pushDown, rest) = candidates.partition { cond => val replaced = replaceAlias(cond, aliasMap) - replaced.references.subsetOf(aggregate.child.outputSet) + cond.references.nonEmpty && replaced.references.subsetOf(aggregate.child.outputSet) } val stayUp = rest ++ containingNonDeterministic diff --git a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/FilterPushdownSuite.scala b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/FilterPushdownSuite.scala index 55836f96f7e0eba990f260299d81c942ccc06af6..019f132d94cb259f30363f842f8413bdf077bba1 100644 --- a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/FilterPushdownSuite.scala +++ b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/FilterPushdownSuite.scala @@ -687,6 +687,23 @@ class FilterPushdownSuite extends PlanTest { comparePlans(optimized, correctAnswer) } + test("SPARK-17712: aggregate: don't push down filters that are data-independent") { + val originalQuery = LocalRelation.apply(testRelation.output, Seq.empty) + .select('a, 'b) + .groupBy('a)(count('a)) + .where(false) + + val optimized = Optimize.execute(originalQuery.analyze) + + val correctAnswer = testRelation + .select('a, 'b) + .groupBy('a)(count('a)) + .where(false) + .analyze + + comparePlans(optimized, correctAnswer) + } + test("broadcast hint") { val originalQuery = BroadcastHint(testRelation) .where('a === 2L && 'b + Rand(10).as("rnd") === 3)