[ https://issues.apache.org/jira/browse/SPARK-19614?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15869593#comment-15869593 ]
Takeshi Yamamuro commented on SPARK-19614: ------------------------------------------ Is it much common for users to explicitly set NULL? Anyway, you can set NULL with types; {code} Seq((1, 0), (2, 0)).toDF("a", "b").selectExpr("a", "b", "CAST(NULL AS INT)") {code} Is this not easy enough for your case? > add type-preserving null function > --------------------------------- > > Key: SPARK-19614 > URL: https://issues.apache.org/jira/browse/SPARK-19614 > Project: Spark > Issue Type: Improvement > Components: SQL > Affects Versions: 2.1.0 > Reporter: Nick Dimiduk > Priority: Trivial > > There's currently no easy way to extend the columns of a DataFrame with null > columns that also preserves the type. {{lit(null)}} evaluates to > {{Literal(null, NullType)}}, despite any subsequent hinting, for instance > with {{Column.as(String, Metadata)}}. This comes up when programmatically > munging data from disparate sources. A function such as {{null(DataType)}} > would be nice. -- This message was sent by Atlassian JIRA (v6.3.15#6346) --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org