[
https://issues.apache.org/jira/browse/ARROW-13007?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17359329#comment-17359329
]
Torstein Sørnes commented on ARROW-13007:
-----------------------------------------
The first would obviously be better, but the second one would do!
If there is support for maps/dictionaries, why not sets? Are they not a simpler
case of the map?
> Supporting Sets (HashSets)
> --------------------------
>
> Key: ARROW-13007
> URL: https://issues.apache.org/jira/browse/ARROW-13007
> Project: Apache Arrow
> Issue Type: Improvement
> Components: Python
> Affects Versions: 4.0.1
> Reporter: Torstein Sørnes
> Priority: Major
>
> Having support for Sets would be great.
>
> It is possible to do:
> {code:java}
> pd.DataFrame([[{'1': None}]], columns =["s"]).to_feather('/tmp/x'){code}
> But using a Set instead of a dictionary is not possible
> {code:java}
> pd.DataFrame([[{'1'}]], columns =["s"]).to_feather('/tmp/x'){code}
>
> {quote}ArrowInvalid: ("Could not convert \{'1'} with type set: did not
> recognize Python value type when inferring an Arrow data type", 'Conversion
> failed for column s with type object')
> {quote}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
