[
https://issues.apache.org/jira/browse/SPARK-33844?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
angerszhu updated SPARK-33844:
------------------------------
Description:
After hive-2.3 we will set COLUMN_NAME_DELIMITER to special char when col name
cntains ','since column list and column types in serde is splited by
COLUMN_NAME_DELIMITER.
In spark-2.4.0 + hive-1.2.1 we will failed when INSERT OVERWRITE DIR when
query result schema columan name contains ','as
{code:java}
org.apache.hadoop.hive.serde2.SerDeException:
org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe: columns has 14 elements
while columns.types has 11 elements! at
org.apache.hadoop.hive.serde2.lazy.LazySerDeParameters.extractColumnInfo(LazySerDeParameters.java:146)
at
org.apache.hadoop.hive.serde2.lazy.LazySerDeParameters.<init>(LazySerDeParameters.java:85)
at
org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe.initialize(LazySimpleSerDe.java:125)
at
org.apache.spark.sql.hive.execution.HiveOutputWriter.<init>(HiveFileFormat.scala:119)
at
org.apache.spark.sql.hive.execution.HiveFileFormat$$anon$1.newInstance(HiveFileFormat.scala:103)
at
org.apache.spark.sql.execution.datasources.SingleDirectoryDataWriter.newOutputWriter(FileFormatDataWriter.scala:120)
at
org.apache.spark.sql.execution.datasources.SingleDirectoryDataWriter.<init>(FileFormatDataWriter.scala:108)
at
org.apache.spark.sql.execution.datasources.FileFormatWriter$.org$apache$spark$sql$execution$datasources$FileFormatWriter$$executeTask(FileFormatWriter.scala:287)
at
org.apache.spark.sql.execution.datasources.FileFormatWriter$$anonfun$write$1.apply(FileFormatWriter.scala:219)
at
org.apache.spark.sql.execution.datasources.FileFormatWriter$$anonfun$write$1.apply(FileFormatWriter.scala:218)
at org.apache.spark.scheduler.ResultTask.runTask(ResultTask.scala:90) at
org.apache.spark.scheduler.Task.run(Task.scala:121) at
org.apache.spark.executor.Executor$TaskRunner$$anonfun$12.apply(Executor.scala:461)
at org.apache.spark.util.Utils$.tryWithSafeFinally(Utils.scala:1360) at
org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:467) at
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748){code}
Since this problem has been solved by
[https://github.com/apache/hive/blob/6f4c35c9e904d226451c465effdc5bfd31d395a0/metastore/src/java/org/apache/hadoop/hive/metastore/MetaStoreUtils.java#L1044-L1075]
But I think we can do this in Spark side to make all version work well.
was:
After hive-2.3 we will set COLUMN_NAME_DELIMITER to special char when col name
cntains ','since column list and column types in serde is splited by
COLUMN_NAME_DELIMITER.
In spark-2.4.0 + hive-1.2.1 we will failed when INSERT OVERWRITE DIR when query
result schema columan name contains
','as```org.apache.hadoop.hive.serde2.SerDeException:
org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe: columns has 14 elements
while columns.types has 11 elements! at
org.apache.hadoop.hive.serde2.lazy.LazySerDeParameters.extractColumnInfo(LazySerDeParameters.java:146)
at
org.apache.hadoop.hive.serde2.lazy.LazySerDeParameters.<init>(LazySerDeParameters.java:85)
at
org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe.initialize(LazySimpleSerDe.java:125)
at
org.apache.spark.sql.hive.execution.HiveOutputWriter.<init>(HiveFileFormat.scala:119)
at
org.apache.spark.sql.hive.execution.HiveFileFormat$$anon$1.newInstance(HiveFileFormat.scala:103)
at
org.apache.spark.sql.execution.datasources.SingleDirectoryDataWriter.newOutputWriter(FileFormatDataWriter.scala:120)
at
org.apache.spark.sql.execution.datasources.SingleDirectoryDataWriter.<init>(FileFormatDataWriter.scala:108)
at
org.apache.spark.sql.execution.datasources.FileFormatWriter$.org$apache$spark$sql$execution$datasources$FileFormatWriter$$executeTask(FileFormatWriter.scala:287)
at
org.apache.spark.sql.execution.datasources.FileFormatWriter$$anonfun$write$1.apply(FileFormatWriter.scala:219)
at
org.apache.spark.sql.execution.datasources.FileFormatWriter$$anonfun$write$1.apply(FileFormatWriter.scala:218)
at org.apache.spark.scheduler.ResultTask.runTask(ResultTask.scala:90) at
org.apache.spark.scheduler.Task.run(Task.scala:121) at
org.apache.spark.executor.Executor$TaskRunner$$anonfun$12.apply(Executor.scala:461)
at org.apache.spark.util.Utils$.tryWithSafeFinally(Utils.scala:1360) at
org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:467) at
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)```
Since this problem has been solved by
https://github.com/apache/hive/blob/6f4c35c9e904d226451c465effdc5bfd31d395a0/metastore/src/java/org/apache/hadoop/hive/metastore/MetaStoreUtils.java#L1044-L1075
But I think we cna do this in Spark side to make all version work well.
val hiveColumns = table.schema.map(toHiveColumn) // Note: In Hive the
schema and partition columns must be disjoint sets val (partCols, schema) =
table.schema.map(toHiveColumn).partition \{ c => val (partCols, schema) =
hiveColumns.partition { c =>
table.partitionColumnNames.contains(c.getName) }
hiveTable.setFields(schema.asJava) hiveTable.setPartCols(partCols.asJava)
if (hiveColumns.exists(_.getName.contains(","))) \{
hiveTable.setSerdeParam(serdeConstants.COLUMN_NAME_DELIMITER,
String.valueOf(SerDeUtils.COLUMN_COMMENTS_DELIMITER)) }
test("insert overwrite directory with comma col name") { withTempDir {
dir => val path = dir.toURI.getPath
val v1 = s""" | INSERT OVERWRITE DIRECTORY '${path}'
| STORED AS TEXTFILE | SELECT 1 as a, 'c' as b, if(1 = 1,
"true", "false") """.stripMargin
sql(v1).explain(true)
sql(v1).show()
// checkAnswer( //
spark.read.json(dir.getCanonicalPath), // sql("SELECT 1 as a, 'c'
as b")) } }
> InsertIntoDir failed since query column name contains ',' cause column type
> and column names size not equal
> -----------------------------------------------------------------------------------------------------------
>
> Key: SPARK-33844
> URL: https://issues.apache.org/jira/browse/SPARK-33844
> Project: Spark
> Issue Type: Improvement
> Components: SQL
> Affects Versions: 2.4.0
> Reporter: angerszhu
> Priority: Major
>
>
> After hive-2.3 we will set COLUMN_NAME_DELIMITER to special char when col
> name cntains ','since column list and column types in serde is splited by
> COLUMN_NAME_DELIMITER.
> In spark-2.4.0 + hive-1.2.1 we will failed when INSERT OVERWRITE DIR when
> query result schema columan name contains ','as
> {code:java}
> org.apache.hadoop.hive.serde2.SerDeException:
> org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe: columns has 14 elements
> while columns.types has 11 elements! at
> org.apache.hadoop.hive.serde2.lazy.LazySerDeParameters.extractColumnInfo(LazySerDeParameters.java:146)
> at
> org.apache.hadoop.hive.serde2.lazy.LazySerDeParameters.<init>(LazySerDeParameters.java:85)
> at
> org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe.initialize(LazySimpleSerDe.java:125)
> at
> org.apache.spark.sql.hive.execution.HiveOutputWriter.<init>(HiveFileFormat.scala:119)
> at
> org.apache.spark.sql.hive.execution.HiveFileFormat$$anon$1.newInstance(HiveFileFormat.scala:103)
> at
> org.apache.spark.sql.execution.datasources.SingleDirectoryDataWriter.newOutputWriter(FileFormatDataWriter.scala:120)
> at
> org.apache.spark.sql.execution.datasources.SingleDirectoryDataWriter.<init>(FileFormatDataWriter.scala:108)
> at
> org.apache.spark.sql.execution.datasources.FileFormatWriter$.org$apache$spark$sql$execution$datasources$FileFormatWriter$$executeTask(FileFormatWriter.scala:287)
> at
> org.apache.spark.sql.execution.datasources.FileFormatWriter$$anonfun$write$1.apply(FileFormatWriter.scala:219)
> at
> org.apache.spark.sql.execution.datasources.FileFormatWriter$$anonfun$write$1.apply(FileFormatWriter.scala:218)
> at org.apache.spark.scheduler.ResultTask.runTask(ResultTask.scala:90) at
> org.apache.spark.scheduler.Task.run(Task.scala:121) at
> org.apache.spark.executor.Executor$TaskRunner$$anonfun$12.apply(Executor.scala:461)
> at org.apache.spark.util.Utils$.tryWithSafeFinally(Utils.scala:1360) at
> org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:467) at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> at java.lang.Thread.run(Thread.java:748){code}
> Since this problem has been solved by
> [https://github.com/apache/hive/blob/6f4c35c9e904d226451c465effdc5bfd31d395a0/metastore/src/java/org/apache/hadoop/hive/metastore/MetaStoreUtils.java#L1044-L1075]
> But I think we can do this in Spark side to make all version work well.
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
