[
https://issues.apache.org/jira/browse/SPARK-17012?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16196190#comment-16196190
]
Hyukjin Kwon commented on SPARK-17012:
--------------------------------------
Couldn't we just replace values after loading it into Spark?
> Reading data frames via CSV - Allow to specify default value for integers
> -------------------------------------------------------------------------
>
> Key: SPARK-17012
> URL: https://issues.apache.org/jira/browse/SPARK-17012
> Project: Spark
> Issue Type: Improvement
> Affects Versions: 2.0.0
> Reporter: Aseem Bansal
>
> Currently the option that we have in DataFrameReader is nullValue which
> allows us one default. But say in our data frame we have string and integers
> and we want to specify the default for strings and integers differently that
> is currently not possible.
> If it is done for different data types then it should be possible to allow to
> specify the schema to be nullable false when inferring schema (as a new
> option).
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
