HyukjinKwon commented on a change in pull request #25398: [SPARK-28659][SQL]
Use data source if convertible in insert overwrite directory
URL: https://github.com/apache/spark/pull/25398#discussion_r329354901
##########
File path:
sql/core/src/main/scala/org/apache/spark/sql/execution/SparkSqlParser.scala
##########
@@ -1396,6 +1396,16 @@ class SparkSqlAstBuilder(conf: SQLConf) extends
AstBuilder(conf) {
compressed = false,
properties = rowStorage.properties ++ fileStorage.properties)
- (ctx.LOCAL != null, storage, Some(DDLUtils.HIVE_PROVIDER))
+ val fileFormat = extractFileFormat(fileStorage.serde)
+ (ctx.LOCAL != null, storage, Some(fileFormat))
+ }
+
+ private def extractFileFormat(serde: Option[String]): String = {
Review comment:
I think @maropu has kind of its point. Why don't you just use `STORED AS
file_format` instead? BTW, I think it should be in `RelationConversions` with
`spark.sql.hive.convertMetastoreXXX`.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
