[
https://issues.apache.org/jira/browse/SPARK-2010?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14052068#comment-14052068
]
Kan Zhang commented on SPARK-2010:
----------------------------------
Sounds reasonable. Named tuple is a better fit than dictionary for struct type.
Presumably it is due to lack of pickling support for named tuple that we
resorted to dictionary for python schema definition. But for nested
dictionaries, we should treat them as map type.
> Support for nested data in PySpark SQL
> --------------------------------------
>
> Key: SPARK-2010
> URL: https://issues.apache.org/jira/browse/SPARK-2010
> Project: Spark
> Issue Type: Improvement
> Components: SQL
> Reporter: Michael Armbrust
> Assignee: Kan Zhang
>
--
This message was sent by Atlassian JIRA
(v6.2#6252)
