[
https://issues.apache.org/jira/browse/SPARK-15245?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15277979#comment-15277979
]
Hyukjin Kwon commented on SPARK-15245:
--------------------------------------
Sorry for leaving comments again and again but I think this JIRA might not have
to be closed (my PR was closed though) because there is a kind of hidden option
{{basePath}} for reading partitioned table for datasources,
[here|https://github.com/apache/spark/blob/f7b7ef41662d7d02fc4f834f3c6c4ee8802e949c/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/PartitioningAwareFileCatalog.scala#L170-L173]
and {{stream()}} API is overwriting this,
[here|https://github.com/apache/spark/blob/f7b7ef41662d7d02fc4f834f3c6c4ee8802e949c/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/DataSource.scala#L183].
So, I feel like the message has to be corrected anyway..
> stream API throws an exception with an incorrect message when the path is not
> a direcotry
> -----------------------------------------------------------------------------------------
>
> Key: SPARK-15245
> URL: https://issues.apache.org/jira/browse/SPARK-15245
> Project: Spark
> Issue Type: Improvement
> Components: SQL
> Reporter: Hyukjin Kwon
> Priority: Trivial
>
> {code}
> val path = "tmp.csv" // This is not a directory
> val cars = spark.read
> .format("csv")
> .stream(path)
> .write
> .option("checkpointLocation", "streaming.metadata")
> .startStream("tmp")
> {code}
> This throws an exception as below.
> {code}
> java.lang.IllegalArgumentException: Option 'basePath' must be a directory
> at
> org.apache.spark.sql.execution.datasources.PartitioningAwareFileCatalog.basePaths(PartitioningAwareFileCatalog.scala:180)
> at
> org.apache.spark.sql.execution.datasources.PartitioningAwareFileCatalog.inferPartitioning(PartitioningAwareFileCatalog.scala:117)
> at
> org.apache.spark.sql.execution.datasources.ListingFileCatalog.partitionSpec(ListingFileCatalog.scala:54)
> at
> org.apache.spark.sql.execution.datasources.PartitioningAwareFileCatalog.allFiles(PartitioningAwareFileCatalog.scala:65)
> at
> org.apache.spark.sql.execution.datasources.DataSource.resolveRelation(DataSource.scala:350)
> at
> org.apache.spark.sql.execution.datasources.DataSource.org$apache$spark$sql$execution$datasources$DataSource$$dataFrameBuilder$1(DataSource.scala:197)
> at
> org.apache.spark.sql.execution.datasources.DataSource$$anonfun$createSource$1.apply(DataSource.scala:201)
> at
> org.apache.spark.sql.execution.datasources.DataSource$$anonfun$createSource$1.apply(DataSource.scala:201)
> at
> org.apache.spark.sql.execution.streaming.FileStreamSource.getBatch(FileStreamSource.scala:101)
> at
> org.apache.spark.sql.execution.streaming.StreamExecution$$anonfun$5.apply(StreamExecution.scala:313)
> at
> org.apache.spark.sql.execution.streaming.StreamExecution$$anonfun$5.apply(StreamExecution.scala:310)
> at
> scala.collection.TraversableLike$$anonfun$flatMap$1.apply(TraversableLike.scala:241)
> {code}
> It seems {{path}} is set to {{basePath}} in {{DataSource}}. This might be
> great if it has a better message.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
