[jira] [Commented] (ARROW-2510) [Python] Segmentation fault when converting empty column as categorical

["Uwe L. Korn (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/ARROW-2510?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16460079#comment-16460079 ] 

Uwe L. Korn commented on ARROW-2510:
------------------------------------

[~fjetter] I cannot reproduce the behaviour I expect that you are describing. The following code succeeds on master:

{code}
import pandas as pd
import pyarrow as pa
import pyarrow.parquet as pq

df = pd.DataFrame({'a': ['str']})
df['a'] = df['a'].astype('category')
pq.write_metadata(pa.Table.from_pandas(df).schema, '_common_metadata')
t = pq.read_table('_common_metadata')
t.to_pandas(categories=['a'])
{code}

> [Python] Segmentation fault when converting empty column as categorical
> -----------------------------------------------------------------------
>
>                 Key: ARROW-2510
>                 URL: https://issues.apache.org/jira/browse/ARROW-2510
>             Project: Apache Arrow
>          Issue Type: Bug
>          Components: C++, Python
>            Reporter: Florian Jetter
>            Assignee: Uwe L. Korn
>            Priority: Minor
>             Fix For: 0.10.0
>
>
> When converting an empty column to categorical in pandas I get a segmentation fault



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)