[
https://issues.apache.org/jira/browse/SPARK-18943?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15764103#comment-15764103
]
Apache Spark commented on SPARK-18943:
--------------------------------------
User 'HyukjinKwon' has created a pull request for this issue:
https://github.com/apache/spark/pull/16351
> Avoid per-record type dispatch in CSV when reading
> --------------------------------------------------
>
> Key: SPARK-18943
> URL: https://issues.apache.org/jira/browse/SPARK-18943
> Project: Spark
> Issue Type: Improvement
> Components: SQL
> Reporter: Hyukjin Kwon
>
> This is similar with https://issues.apache.org/jira/browse/SPARK-16434
> {{CSVRelation.csvParser}} does type dispatch for each value in each row.
> We can prevent this because the schema is already kept in {{CSVRelation}}.
> So, converters can be created first according to the schema, and then apply
> them to each.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
