Github user HyukjinKwon commented on a diff in the pull request:
https://github.com/apache/spark/pull/19487#discussion_r144580099
--- Diff:
core/src/main/scala/org/apache/spark/internal/io/HadoopMapReduceCommitProtocol.scala
---
@@ -48,6 +49,16 @@ class HadoopMapReduceCommitProtocol(jobId: String, path:
String)
@transient private var committer: OutputCommitter = _
/**
+ * Checks whether there are files to be committed to a valid output
location.
+ *
+ * As committing and aborting a job occurs on driver, where
`addedAbsPathFiles` is always null,
+ * it is necessary to check whether a valid output path is specified.
+ * [[HadoopMapReduceCommitProtocol#path]] need not be a valid
[[org.apache.hadoop.fs.Path]] for
+ * committers not writing to distributed file systems.
+ */
+ private val hasValidPath = Try { new Path(path) }.isSuccess
--- End diff --
I think we should explicitly catch the exception BTW.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
