Repository: spark Updated Branches: refs/heads/branch-1.0 357d16bcd -> a803e009c
[SPARK-2210] cast to boolean on boolean value gets turned into NOT((boolean_condition) = 0) ``` explain select cast(cast(key=0 as boolean) as boolean) aaa from src ``` should be ``` [Physical execution plan:] [Project [(key#10:0 = 0) AS aaa#7]] [ HiveTableScan [key#10], (MetastoreRelation default, src, None), None] ``` However, it is currently ``` [Physical execution plan:] [Project [NOT((key#10=0) = 0) AS aaa#7]] [ HiveTableScan [key#10], (MetastoreRelation default, src, None), None] ``` Author: Reynold Xin <[email protected]> Closes #1144 from rxin/booleancast and squashes the following commits: c4e543d [Reynold Xin] [SPARK-2210] boolean cast on boolean value should be removed. (cherry picked from commit 61756409736a64bd42577782cb7468557fa0b642) Signed-off-by: Reynold Xin <[email protected]> Project: http://git-wip-us.apache.org/repos/asf/spark/repo Commit: http://git-wip-us.apache.org/repos/asf/spark/commit/a803e009 Tree: http://git-wip-us.apache.org/repos/asf/spark/tree/a803e009 Diff: http://git-wip-us.apache.org/repos/asf/spark/diff/a803e009 Branch: refs/heads/branch-1.0 Commit: a803e009c94c9dd4223003e7695fc3f885454c43 Parents: 357d16b Author: Reynold Xin <[email protected]> Authored: Thu Jun 19 23:58:23 2014 -0700 Committer: Reynold Xin <[email protected]> Committed: Thu Jun 19 23:58:30 2014 -0700 ---------------------------------------------------------------------- .../catalyst/analysis/HiveTypeCoercion.scala | 4 +++- .../hive/execution/HiveTypeCoercionSuite.scala | 25 +++++++++++++++++++- 2 files changed, 27 insertions(+), 2 deletions(-) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/spark/blob/a803e009/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/HiveTypeCoercion.scala ---------------------------------------------------------------------- diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/HiveTypeCoercion.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/HiveTypeCoercion.scala index 6d331fb..c0714bc 100644 --- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/HiveTypeCoercion.scala +++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/HiveTypeCoercion.scala @@ -251,7 +251,9 @@ trait HiveTypeCoercion { def apply(plan: LogicalPlan): LogicalPlan = plan transformAllExpressions { // Skip nodes who's children have not been resolved yet. case e if !e.childrenResolved => e - + // Skip if the type is boolean type already. Note that this extra cast should be removed + // by optimizer.SimplifyCasts. + case Cast(e, BooleanType) if e.dataType == BooleanType => e case Cast(e, BooleanType) => Not(Equals(e, Literal(0))) case Cast(e, dataType) if e.dataType == BooleanType => Cast(If(e, Literal(1), Literal(0)), dataType) http://git-wip-us.apache.org/repos/asf/spark/blob/a803e009/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveTypeCoercionSuite.scala ---------------------------------------------------------------------- diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveTypeCoercionSuite.scala b/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveTypeCoercionSuite.scala index e030c8e..cc8744c 100644 --- a/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveTypeCoercionSuite.scala +++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveTypeCoercionSuite.scala @@ -17,8 +17,12 @@ package org.apache.spark.sql.hive.execution +import org.apache.spark.sql.catalyst.expressions.{Cast, Equals} +import org.apache.spark.sql.execution.Project +import org.apache.spark.sql.hive.test.TestHive + /** - * A set of tests that validate type promotion rules. + * A set of tests that validate type promotion and coercion rules. */ class HiveTypeCoercionSuite extends HiveComparisonTest { val baseTypes = Seq("1", "1.0", "1L", "1S", "1Y", "'1'") @@ -28,4 +32,23 @@ class HiveTypeCoercionSuite extends HiveComparisonTest { createQueryTest(s"$i + $j", s"SELECT $i + $j FROM src LIMIT 1") } } + + test("[SPARK-2210] boolean cast on boolean value should be removed") { + val q = "select cast(cast(key=0 as boolean) as boolean) from src" + val project = TestHive.hql(q).queryExecution.executedPlan.collect { case e: Project => e }.head + + // No cast expression introduced + project.transformAllExpressions { case c: Cast => + assert(false, "unexpected cast " + c) + c + } + + // Only one Equals + var numEquals = 0 + project.transformAllExpressions { case e: Equals => + numEquals += 1 + e + } + assert(numEquals === 1) + } }
