[
https://issues.apache.org/jira/browse/SPARK-4052?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Yin Huai updated SPARK-4052:
----------------------------
Description:
{code}
val sqlContext = new org.apache.spark.sql.hive.HiveContext(sc)
import sqlContext.createSchemaRDD
val rdd = sc.parallelize(("key", "value") :: Nil)
// Test1: This one fails.
case class Test1(m: scala.collection.Map[String, String])
val rddOfTest1 = rdd.map { case (k, v) => Test1(Map(k->v)) }
rddOfTest1.registerTempTable("t1")
/* Stack trace
scala.MatchError: scala.collection.Map[String,String] (of class
scala.reflect.internal.Types$TypeRef$$anon$5)
at
org.apache.spark.sql.catalyst.ScalaReflection$.schemaFor(ScalaReflection.scala:53)
at
org.apache.spark.sql.catalyst.ScalaReflection$$anonfun$schemaFor$1.apply(ScalaReflection.scala:64)
at
org.apache.spark.sql.catalyst.ScalaReflection$$anonfun$schemaFor$1.apply(ScalaReflection.scala:62)
...
*/
// Test2: This one is fine.
case class Test2(m: scala.collection.immutable.Map[String, String])
val rddOfTest2 = rdd.map { case (k, v) => Test2(Map(k->v)) }
rddOfTest2.registerTempTable("t2")
sqlContext.sql("SELECT m FROM t2").collect
sqlContext.sql("SELECT m['key'] FROM t2").collect
// Test3: This one fails.
import org.apache.spark.sql._
val schema = StructType(StructField("m", MapType(StringType, StringType), true)
:: Nil)
val rowRDD = rdd.map { case (k, v) =>
Row(scala.collection.mutable.HashMap(k->v)) }
val schemaRDD = sqlContext.applySchema(rowRDD, schema)
schemaRDD.registerTempTable("t3")
sqlContext.sql("SELECT m FROM t3").collect
sqlContext.sql("SELECT m['key'] FROM t3").collect
sqlContext.sql("CREATE TABLE testHiveTable(m MAP <STRING, STRING>)")
sqlContext.sql("INSERT OVERWRITE TABLE testHiveTable SELECT m FROM t3")
/* Stack trace
14/10/22 19:30:56 INFO DAGScheduler: Job 4 failed: runJob at
InsertIntoHiveTable.scala:124, took 1.384579 s
org.apache.spark.SparkException: Job aborted due to stage failure: Task 1 in
stage 4.0 failed 4 times, most recent failure: Lost task 1.3 in stage 4.0 (TID
12, yins-mbp): java.lang.ClassCastException: scala.collection.mutable.HashMap
cannot be cast to scala.collection.immutable.Map
org.apache.spark.sql.hive.execution.InsertIntoHiveTable$$anonfun$wrapperFor$5.apply(InsertIntoHiveTable.scala:96)
org.apache.spark.sql.hive.execution.InsertIntoHiveTable$$anonfun$wrapperFor$5.apply(InsertIntoHiveTable.scala:96)
org.apache.spark.sql.hive.execution.InsertIntoHiveTable$$anonfun$org$apache$spark$sql$hive$execution$InsertIntoHiveTable$$writeToFile$1$1.apply(InsertIntoHiveTable.scala:148)
org.apache.spark.sql.hive.execution.InsertIntoHiveTable$$anonfun$org$apache$spark$sql$hive$execution$InsertIntoHiveTable$$writeToFile$1$1.apply(InsertIntoHiveTable.scala:145)
*/
{code}
> Use scala.collection.Map for pattern matching instead of using Predef.Map (it
> is scala.collection.immutable.Map)
> ----------------------------------------------------------------------------------------------------------------
>
> Key: SPARK-4052
> URL: https://issues.apache.org/jira/browse/SPARK-4052
> Project: Spark
> Issue Type: Bug
> Components: SQL
> Affects Versions: 1.1.0
> Reporter: Yin Huai
> Priority: Minor
>
> {code}
> val sqlContext = new org.apache.spark.sql.hive.HiveContext(sc)
> import sqlContext.createSchemaRDD
> val rdd = sc.parallelize(("key", "value") :: Nil)
> // Test1: This one fails.
> case class Test1(m: scala.collection.Map[String, String])
> val rddOfTest1 = rdd.map { case (k, v) => Test1(Map(k->v)) }
> rddOfTest1.registerTempTable("t1")
> /* Stack trace
> scala.MatchError: scala.collection.Map[String,String] (of class
> scala.reflect.internal.Types$TypeRef$$anon$5)
> at
> org.apache.spark.sql.catalyst.ScalaReflection$.schemaFor(ScalaReflection.scala:53)
> at
> org.apache.spark.sql.catalyst.ScalaReflection$$anonfun$schemaFor$1.apply(ScalaReflection.scala:64)
> at
> org.apache.spark.sql.catalyst.ScalaReflection$$anonfun$schemaFor$1.apply(ScalaReflection.scala:62)
> ...
> */
> // Test2: This one is fine.
> case class Test2(m: scala.collection.immutable.Map[String, String])
> val rddOfTest2 = rdd.map { case (k, v) => Test2(Map(k->v)) }
> rddOfTest2.registerTempTable("t2")
> sqlContext.sql("SELECT m FROM t2").collect
> sqlContext.sql("SELECT m['key'] FROM t2").collect
> // Test3: This one fails.
> import org.apache.spark.sql._
> val schema = StructType(StructField("m", MapType(StringType, StringType),
> true) :: Nil)
> val rowRDD = rdd.map { case (k, v) =>
> Row(scala.collection.mutable.HashMap(k->v)) }
> val schemaRDD = sqlContext.applySchema(rowRDD, schema)
> schemaRDD.registerTempTable("t3")
> sqlContext.sql("SELECT m FROM t3").collect
> sqlContext.sql("SELECT m['key'] FROM t3").collect
> sqlContext.sql("CREATE TABLE testHiveTable(m MAP <STRING, STRING>)")
> sqlContext.sql("INSERT OVERWRITE TABLE testHiveTable SELECT m FROM t3")
> /* Stack trace
> 14/10/22 19:30:56 INFO DAGScheduler: Job 4 failed: runJob at
> InsertIntoHiveTable.scala:124, took 1.384579 s
> org.apache.spark.SparkException: Job aborted due to stage failure: Task 1 in
> stage 4.0 failed 4 times, most recent failure: Lost task 1.3 in stage 4.0
> (TID 12, yins-mbp): java.lang.ClassCastException:
> scala.collection.mutable.HashMap cannot be cast to
> scala.collection.immutable.Map
>
> org.apache.spark.sql.hive.execution.InsertIntoHiveTable$$anonfun$wrapperFor$5.apply(InsertIntoHiveTable.scala:96)
>
> org.apache.spark.sql.hive.execution.InsertIntoHiveTable$$anonfun$wrapperFor$5.apply(InsertIntoHiveTable.scala:96)
>
> org.apache.spark.sql.hive.execution.InsertIntoHiveTable$$anonfun$org$apache$spark$sql$hive$execution$InsertIntoHiveTable$$writeToFile$1$1.apply(InsertIntoHiveTable.scala:148)
>
> org.apache.spark.sql.hive.execution.InsertIntoHiveTable$$anonfun$org$apache$spark$sql$hive$execution$InsertIntoHiveTable$$writeToFile$1$1.apply(InsertIntoHiveTable.scala:145)
> */
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
