You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Apache Spark (JIRA)" <ji...@apache.org> on 2015/12/18 16:09:46 UTC
[jira] [Commented] (SPARK-12369) DataFrameReader fails on globbing
parquet paths
[ https://issues.apache.org/jira/browse/SPARK-12369?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15064059#comment-15064059 ]
Apache Spark commented on SPARK-12369:
--------------------------------------
User 'yanakad' has created a pull request for this issue:
https://github.com/apache/spark/pull/10379
> DataFrameReader fails on globbing parquet paths
> -----------------------------------------------
>
> Key: SPARK-12369
> URL: https://issues.apache.org/jira/browse/SPARK-12369
> Project: Spark
> Issue Type: Bug
> Components: SQL
> Affects Versions: 1.5.2
> Reporter: Yana Kadiyska
>
> Start with a list of parquet paths where some or all do not exist:
> {noformat}
> val paths=List("/foo/month=05/*.parquet","/foo/month=06/*.parquet")
> sqlContext.read.parquet(paths:_*)
> java.lang.NullPointerException
> at org.apache.hadoop.fs.Globber.glob(Globber.java:218)
> at org.apache.hadoop.fs.FileSystem.globStatus(FileSystem.java:1625)
> at org.apache.spark.deploy.SparkHadoopUtil.globPath(SparkHadoopUtil.scala:251)
> at org.apache.spark.deploy.SparkHadoopUtil.globPathIfNecessary(SparkHadoopUtil.scala:258)
> at org.apache.spark.sql.DataFrameReader$$anonfun$3.apply(DataFrameReader.scala:264)
> at org.apache.spark.sql.DataFrameReader$$anonfun$3.apply(DataFrameReader.scala:260)
> at scala.collection.TraversableLike$$anonfun$flatMap$1.apply(TraversableLike.scala:251)
> at scala.collection.TraversableLike$$anonfun$flatMap$1.apply(TraversableLike.scala:251)
> at scala.collection.immutable.List.foreach(List.scala:318)
> at scala.collection.TraversableLike$class.flatMap(TraversableLike.scala:251)
> at scala.collection.AbstractTraversable.flatMap(Traversable.scala:105)
> at org.apache.spark.sql.DataFrameReader.parquet(DataFrameReader.scala:260)
> {noformat}
> It would be better to produce a dataframe from the paths that do exist and log a warning that a path was missing. Not sure for "all paths are missing case" -- probably return an emptyDF with no schema since that method already does so on empty path list.But I would prefer not to have to pre-validate paths
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org