Github user yhuai commented on a diff in the pull request:
https://github.com/apache/spark/pull/4610#discussion_r24720131
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/json/JSONRelation.scala ---
@@ -104,21 +116,35 @@ private[sql] case class JSONRelation(
override def buildScan() =
JsonRDD.jsonStringToRow(baseRDD, schema,
sqlContext.conf.columnNameOfCorruptRecord)
+ private def isTemporaryFile(file: Path): Boolean = {
+ file.getName == "_temporary"
+ }
+
override def insert(data: DataFrame, overwrite: Boolean) = {
- val filesystemPath = new Path(path)
- val fs =
filesystemPath.getFileSystem(sqlContext.sparkContext.hadoopConfiguration)
+
+ // If the path exists, it must be a directory, error for not.
+ // Otherwise we create a directory with the path name.
+ if (fs.exists(filesystemPath) &&
!fs.getFileStatus(filesystemPath).isDirectory) {
+ sys.error("a CREATE [TEMPORARY] TABLE AS SELECT statement need the
path must be directory")
+ }
if (overwrite) {
+ val temporaryPath = new Path(path, "_temporary")
+ val dataPath = new Path(path, "data")
+ // Write the data.
+ data.toJSON.saveAsTextFile(temporaryPath.toUri.toString)
+ val pathsToDelete = fs.listStatus(filesystemPath).filter(
+ f => !isTemporaryFile(f.getPath)).map(_.getPath)
+
try {
- fs.delete(filesystemPath, true)
--- End diff --
We need to do the same thing for delete at here as what we will do for
createRelation.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
