Repository: spark Updated Branches: refs/heads/master 79cdb9b64 -> 84e2c8bfe
[SQL] Add test case with workaround for reading partitioned Avro files In order to read from partitioned Avro files we need to also set the `SERDEPROPERTIES` since `TBLPROPERTIES` are not passed to the initialization. This PR simply adds a test to make sure we don't break this workaround. Author: Michael Armbrust <mich...@databricks.com> Closes #2340 from marmbrus/avroPartitioned and squashes the following commits: 6b969d6 [Michael Armbrust] fix style fea2124 [Michael Armbrust] Add test case with workaround for reading partitioned avro files. Project: http://git-wip-us.apache.org/repos/asf/spark/repo Commit: http://git-wip-us.apache.org/repos/asf/spark/commit/84e2c8bf Tree: http://git-wip-us.apache.org/repos/asf/spark/tree/84e2c8bf Diff: http://git-wip-us.apache.org/repos/asf/spark/diff/84e2c8bf Branch: refs/heads/master Commit: 84e2c8bfe41837baf2aeffa9741e4dbd14351981 Parents: 79cdb9b Author: Michael Armbrust <mich...@databricks.com> Authored: Wed Sep 10 20:57:38 2014 -0700 Committer: Michael Armbrust <mich...@databricks.com> Committed: Wed Sep 10 20:57:38 2014 -0700 ---------------------------------------------------------------------- .../org/apache/spark/sql/hive/TestHive.scala | 69 +++++++++++++++++++- ...AvroSerDe-0-e4501461c855cc9071a872a64186c3de | 8 +++ .../sql/hive/execution/HiveSerDeSuite.scala | 2 + 3 files changed, 78 insertions(+), 1 deletion(-) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/spark/blob/84e2c8bf/sql/hive/src/main/scala/org/apache/spark/sql/hive/TestHive.scala ---------------------------------------------------------------------- diff --git a/sql/hive/src/main/scala/org/apache/spark/sql/hive/TestHive.scala b/sql/hive/src/main/scala/org/apache/spark/sql/hive/TestHive.scala index a013f3f..6974f3e 100644 --- a/sql/hive/src/main/scala/org/apache/spark/sql/hive/TestHive.scala +++ b/sql/hive/src/main/scala/org/apache/spark/sql/hive/TestHive.scala @@ -269,7 +269,74 @@ class TestHiveContext(sc: SparkContext) extends HiveContext(sc) { |) """.stripMargin.cmd, s"LOAD DATA LOCAL INPATH '${getHiveFile("data/files/episodes.avro")}' INTO TABLE episodes".cmd - ) + ), + // THIS TABLE IS NOT THE SAME AS THE HIVE TEST TABLE episodes_partitioned AS DYNAMIC PARITIONING + // IS NOT YET SUPPORTED + TestTable("episodes_part", + s"""CREATE TABLE episodes_part (title STRING, air_date STRING, doctor INT) + |PARTITIONED BY (doctor_pt INT) + |ROW FORMAT SERDE '${classOf[AvroSerDe].getCanonicalName}' + |STORED AS + |INPUTFORMAT '${classOf[AvroContainerInputFormat].getCanonicalName}' + |OUTPUTFORMAT '${classOf[AvroContainerOutputFormat].getCanonicalName}' + |TBLPROPERTIES ( + | 'avro.schema.literal'='{ + | "type": "record", + | "name": "episodes", + | "namespace": "testing.hive.avro.serde", + | "fields": [ + | { + | "name": "title", + | "type": "string", + | "doc": "episode title" + | }, + | { + | "name": "air_date", + | "type": "string", + | "doc": "initial date" + | }, + | { + | "name": "doctor", + | "type": "int", + | "doc": "main actor playing the Doctor in episode" + | } + | ] + | }' + |) + """.stripMargin.cmd, + // WORKAROUND: Required to pass schema to SerDe for partitioned tables. + // TODO: Pass this automatically from the table to partitions. + s""" + |ALTER TABLE episodes_part SET SERDEPROPERTIES ( + | 'avro.schema.literal'='{ + | "type": "record", + | "name": "episodes", + | "namespace": "testing.hive.avro.serde", + | "fields": [ + | { + | "name": "title", + | "type": "string", + | "doc": "episode title" + | }, + | { + | "name": "air_date", + | "type": "string", + | "doc": "initial date" + | }, + | { + | "name": "doctor", + | "type": "int", + | "doc": "main actor playing the Doctor in episode" + | } + | ] + | }' + |) + """.stripMargin.cmd, + s""" + INSERT OVERWRITE TABLE episodes_part PARTITION (doctor_pt=1) + SELECT title, air_date, doctor FROM episodes + """.cmd + ) ) hiveQTestUtilTables.foreach(registerTestTable) http://git-wip-us.apache.org/repos/asf/spark/blob/84e2c8bf/sql/hive/src/test/resources/golden/Read Partitioned with AvroSerDe-0-e4501461c855cc9071a872a64186c3de ---------------------------------------------------------------------- diff --git a/sql/hive/src/test/resources/golden/Read Partitioned with AvroSerDe-0-e4501461c855cc9071a872a64186c3de b/sql/hive/src/test/resources/golden/Read Partitioned with AvroSerDe-0-e4501461c855cc9071a872a64186c3de new file mode 100644 index 0000000..49c8434 --- /dev/null +++ b/sql/hive/src/test/resources/golden/Read Partitioned with AvroSerDe-0-e4501461c855cc9071a872a64186c3de @@ -0,0 +1,8 @@ +The Eleventh Hour 3 April 2010 11 1 +The Doctor's Wife 14 May 2011 11 1 +Horror of Fang Rock 3 September 1977 4 1 +An Unearthly Child 23 November 1963 1 1 +The Mysterious Planet 6 September 1986 6 1 +Rose 26 March 2005 9 1 +The Power of the Daleks 5 November 1966 2 1 +Castrolava 4 January 1982 5 1 http://git-wip-us.apache.org/repos/asf/spark/blob/84e2c8bf/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveSerDeSuite.scala ---------------------------------------------------------------------- diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveSerDeSuite.scala b/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveSerDeSuite.scala index 8bc7238..7486bfa 100644 --- a/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveSerDeSuite.scala +++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveSerDeSuite.scala @@ -37,4 +37,6 @@ class HiveSerDeSuite extends HiveComparisonTest with BeforeAndAfterAll { createQueryTest("Read with RegexSerDe", "SELECT * FROM sales") createQueryTest("Read with AvroSerDe", "SELECT * FROM episodes") + + createQueryTest("Read Partitioned with AvroSerDe", "SELECT * FROM episodes_part") } --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@spark.apache.org For additional commands, e-mail: commits-h...@spark.apache.org
