Repository: spark Updated Branches: refs/heads/master 86ea64dd1 -> 28112657e
[SPARK-12236][SQL] JDBC filter tests all pass if filters are not really pushed down https://issues.apache.org/jira/browse/SPARK-12236 Currently JDBC filters are not tested properly. All the tests pass even if the filters are not pushed down due to Spark-side filtering. In this PR, Firstly, I corrected the tests to properly check the pushed down filters by removing Spark-side filtering. Also, `!=` was being tested which is actually not pushed down. So I removed them. Lastly, I moved the `stripSparkFilter()` function to `SQLTestUtils` as this functions would be shared for all tests for pushed down filters. This function would be also shared with ORC datasource as the filters for that are also not being tested properly. Author: hyukjinkwon <gurwls...@gmail.com> Closes #10221 from HyukjinKwon/SPARK-12236. Project: http://git-wip-us.apache.org/repos/asf/spark/repo Commit: http://git-wip-us.apache.org/repos/asf/spark/commit/28112657 Tree: http://git-wip-us.apache.org/repos/asf/spark/tree/28112657 Diff: http://git-wip-us.apache.org/repos/asf/spark/diff/28112657 Branch: refs/heads/master Commit: 28112657ea5919451291c21b4b8e1eb3db0ec8d4 Parents: 86ea64d Author: hyukjinkwon <gurwls...@gmail.com> Authored: Tue Dec 15 17:02:14 2015 -0800 Committer: Michael Armbrust <mich...@databricks.com> Committed: Tue Dec 15 17:02:14 2015 -0800 ---------------------------------------------------------------------- .../datasources/parquet/ParquetFilterSuite.scala | 15 --------------- .../scala/org/apache/spark/sql/jdbc/JDBCSuite.scala | 10 ++++------ .../org/apache/spark/sql/test/SQLTestUtils.scala | 15 +++++++++++++++ 3 files changed, 19 insertions(+), 21 deletions(-) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/spark/blob/28112657/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetFilterSuite.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetFilterSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetFilterSuite.scala index daf41bc..6178e37 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetFilterSuite.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetFilterSuite.scala @@ -110,21 +110,6 @@ class ParquetFilterSuite extends QueryTest with ParquetTest with SharedSQLContex checkBinaryFilterPredicate(predicate, filterClass, Seq(Row(expected)))(df) } - /** - * Strip Spark-side filtering in order to check if a datasource filters rows correctly. - */ - protected def stripSparkFilter(df: DataFrame): DataFrame = { - val schema = df.schema - val childRDD = df - .queryExecution - .executedPlan.asInstanceOf[org.apache.spark.sql.execution.Filter] - .child - .execute() - .map(row => Row.fromSeq(row.toSeq(schema))) - - sqlContext.createDataFrame(childRDD, schema) - } - test("filter pushdown - boolean") { withParquetDataFrame((true :: false :: Nil).map(b => Tuple1.apply(Option(b)))) { implicit df => checkFilterPredicate('_1.isNull, classOf[Eq[_]], Seq.empty[Row]) http://git-wip-us.apache.org/repos/asf/spark/blob/28112657/sql/core/src/test/scala/org/apache/spark/sql/jdbc/JDBCSuite.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/test/scala/org/apache/spark/sql/jdbc/JDBCSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/jdbc/JDBCSuite.scala index 8c24aa3..a360947 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/jdbc/JDBCSuite.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/jdbc/JDBCSuite.scala @@ -176,12 +176,10 @@ class JDBCSuite extends SparkFunSuite with BeforeAndAfter with SharedSQLContext } test("SELECT * WHERE (simple predicates)") { - assert(sql("SELECT * FROM foobar WHERE THEID < 1").collect().size === 0) - assert(sql("SELECT * FROM foobar WHERE THEID != 2").collect().size === 2) - assert(sql("SELECT * FROM foobar WHERE THEID = 1").collect().size === 1) - assert(sql("SELECT * FROM foobar WHERE NAME = 'fred'").collect().size === 1) - assert(sql("SELECT * FROM foobar WHERE NAME > 'fred'").collect().size === 2) - assert(sql("SELECT * FROM foobar WHERE NAME != 'fred'").collect().size === 2) + assert(stripSparkFilter(sql("SELECT * FROM foobar WHERE THEID < 1")).collect().size === 0) + assert(stripSparkFilter(sql("SELECT * FROM foobar WHERE THEID = 1")).collect().size === 1) + assert(stripSparkFilter(sql("SELECT * FROM foobar WHERE NAME = 'fred'")).collect().size === 1) + assert(stripSparkFilter(sql("SELECT * FROM foobar WHERE NAME > 'fred'")).collect().size === 2) } test("SELECT * WHERE (quoted strings)") { http://git-wip-us.apache.org/repos/asf/spark/blob/28112657/sql/core/src/test/scala/org/apache/spark/sql/test/SQLTestUtils.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/test/scala/org/apache/spark/sql/test/SQLTestUtils.scala b/sql/core/src/test/scala/org/apache/spark/sql/test/SQLTestUtils.scala index 9214569..e87da15 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/test/SQLTestUtils.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/test/SQLTestUtils.scala @@ -180,6 +180,21 @@ private[sql] trait SQLTestUtils } /** + * Strip Spark-side filtering in order to check if a datasource filters rows correctly. + */ + protected def stripSparkFilter(df: DataFrame): DataFrame = { + val schema = df.schema + val childRDD = df + .queryExecution + .executedPlan.asInstanceOf[org.apache.spark.sql.execution.Filter] + .child + .execute() + .map(row => Row.fromSeq(row.toSeq(schema))) + + sqlContext.createDataFrame(childRDD, schema) + } + + /** * Turn a logical plan into a [[DataFrame]]. This should be removed once we have an easier * way to construct [[DataFrame]] directly out of local data without relying on implicits. */ --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@spark.apache.org For additional commands, e-mail: commits-h...@spark.apache.org
