[
https://issues.apache.org/jira/browse/ARROW-5335?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16856293#comment-16856293
]
Wes McKinney commented on ARROW-5335:
-------------------------------------
I think it would be irresponsible to release without at least adding a check
for all dictionaries being the same
> [Python] Support for converting variable dictionaries to pandas
> ---------------------------------------------------------------
>
> Key: ARROW-5335
> URL: https://issues.apache.org/jira/browse/ARROW-5335
> Project: Apache Arrow
> Issue Type: Improvement
> Components: Python
> Reporter: Wes McKinney
> Priority: Blocker
> Fix For: 0.14.0
>
>
> Address after ARROW-3144. The current code presumes the dictionary is the
> same for all chunks. We should check if the dictionary is the same for all
> chunks, and if not, perform a {{DictionaryType::Unify}} operation and then
> write out into the resulting {{CategoricalBlock}}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
