[ https://issues.apache.org/jira/browse/SPARK-33025?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17205328#comment-17205328 ]
Takeshi Yamamuro commented on SPARK-33025: ------------------------------------------ This is not a bug but an expected behaviour and please see [https://github.com/apache/spark/pull/18654#issuecomment-315928986] for more details. Yea, we might be able to remove it for csv/json/text, but that is a minor fix, so I personally think we don't need to do so. > Empty file for the first partition > ---------------------------------- > > Key: SPARK-33025 > URL: https://issues.apache.org/jira/browse/SPARK-33025 > Project: Spark > Issue Type: Bug > Components: Spark Core, SQL > Affects Versions: 3.0.1 > Reporter: Evgenii Samusenko > Priority: Minor > > If I create Dataframe with 1 row, Spark will create empty file for the first > partition. > > Example: > val df = Seq(1).toDF("col1").repartition(8) > df1.write.csv("/csv") > > I got 2 files. The first contains the first partition and the second contains > single row from another partition. It is valid also for parquet, text and etc. -- This message was sent by Atlassian Jira (v8.3.4#803005) --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org