From 1b5ee2003c368d18a5f8c17c2a869ef5770c60a1 Mon Sep 17 00:00:00 2001 From: windpiger Date: Fri, 27 Jan 2017 17:17:17 -0800 Subject: [SPARK-19359][SQL] clear useless path after rename a partition with upper-case by HiveExternalCatalog ## What changes were proposed in this pull request? Hive metastore is not case preserving and keep partition columns with lower case names. If SparkSQL create a table with upper-case partion name use HiveExternalCatalog, when we rename partition, it first call the HiveClient to renamePartition, which will create a new lower case partition path, then SparkSql rename the lower case path to the upper-case. while if the renamed partition contains more than one depth partition ,e.g. A=1/B=2, hive renamePartition change to a=1/b=2, then SparkSql rename it to A=1/B=2, but the a=1 still exists in the filesystem, we should also delete it. ## How was this patch tested? unit test added Author: windpiger Closes #16700 from windpiger/clearUselessPathAfterRenamPartition. --- .../hive/PartitionProviderCompatibilitySuite.scala | 36 ++++++++++++++++++++++ 1 file changed, 36 insertions(+) (limited to 'sql/hive/src/test/scala') diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/hive/PartitionProviderCompatibilitySuite.scala b/sql/hive/src/test/scala/org/apache/spark/sql/hive/PartitionProviderCompatibilitySuite.scala index dca207a72d..1214a92b76 100644 --- a/sql/hive/src/test/scala/org/apache/spark/sql/hive/PartitionProviderCompatibilitySuite.scala +++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/PartitionProviderCompatibilitySuite.scala @@ -19,8 +19,11 @@ package org.apache.spark.sql.hive import java.io.File +import org.apache.hadoop.fs.Path + import org.apache.spark.metrics.source.HiveCatalogMetrics import org.apache.spark.sql.{AnalysisException, QueryTest} +import org.apache.spark.sql.catalyst.TableIdentifier import org.apache.spark.sql.hive.test.TestHiveSingleton import org.apache.spark.sql.internal.SQLConf import org.apache.spark.sql.test.SQLTestUtils @@ -481,4 +484,37 @@ class PartitionProviderCompatibilitySuite assert(spark.sql("show partitions test").count() == 5) } } + + test("partition path created by Hive should be deleted after renamePartitions with upper-case") { + withTable("t", "t1", "t2") { + Seq((1, 2, 3)).toDF("id", "A", "B").write.partitionBy("A", "B").saveAsTable("t") + spark.sql("alter table t partition(A=2, B=3) rename to partition(A=4, B=5)") + + var table = spark.sessionState.catalog.getTableMetadata(TableIdentifier("t")) + var extraHivePath = new Path(table.location + "/a=4") + assert(!extraHivePath.getFileSystem(spark.sessionState.newHadoopConf()) + .exists(extraHivePath), "partition path created by Hive should be deleted " + + "after renamePartitions with upper-case") + + Seq((1, 2, 3, 4)).toDF("id", "A", "B", "C").write.partitionBy("A", "B", "C").saveAsTable("t1") + spark.sql("alter table t1 partition(A=2, B=3, C=4) rename to partition(A=5, B=6, C=7)") + table = spark.sessionState.catalog.getTableMetadata(TableIdentifier("t1")) + extraHivePath = new Path(table.location + "/a=5") + assert(!extraHivePath.getFileSystem(spark.sessionState.newHadoopConf()) + .exists(extraHivePath), "partition path created by Hive should be deleted " + + "after renamePartitions with upper-case") + + Seq((1, 2, 3, 4)).toDF("id", "a", "B", "C").write.partitionBy("a", "B", "C").saveAsTable("t2") + spark.sql("alter table t2 partition(a=2, B=3, C=4) rename to partition(a=4, B=5, C=6)") + table = spark.sessionState.catalog.getTableMetadata(TableIdentifier("t2")) + val partPath = new Path(table.location + "/a=4") + assert(partPath.getFileSystem(spark.sessionState.newHadoopConf()) + .exists(partPath), "partition path of lower-case partition name should not be deleted") + + extraHivePath = new Path(table.location + "/a=4/b=5") + assert(!extraHivePath.getFileSystem(spark.sessionState.newHadoopConf()) + .exists(extraHivePath), "partition path created by Hive should be deleted " + + "after renamePartitions with upper-case") + } + } } -- cgit v1.2.3