[
https://issues.apache.org/jira/browse/PARQUET-1015?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16035070#comment-16035070
]
Wes McKinney commented on PARQUET-1015:
---------------------------------------
I think we can use the smallest possible type (e.g. INT32) since this will get
encoded to nothing anyway. We will want to store the original Arrow schema as
JSON in the Parquet file metadata
What are we doing with DictionaryArray on the write path now?
> Object categoricals are not serialized when only None is present
> ----------------------------------------------------------------
>
> Key: PARQUET-1015
> URL: https://issues.apache.org/jira/browse/PARQUET-1015
> Project: Parquet
> Issue Type: Bug
> Components: parquet-cpp
> Affects Versions: cpp-1.1.0
> Reporter: Marco Neumann
> Priority: Minor
> Fix For: cpp-1.2.0
>
>
> The following code sample fails with {{pyarrow.lib.ArrowNotImplementedError:
> NotImplemented: unhandled type}} but should not:
> {noformat}
> import pandas as pd
> import pyarrow as pa
> import pyarrow.parquet as pq
> df = pd.DataFrame({'x': [None]})
> df['x'] = df['x'].astype('category')
> table = pa.Table.from_pandas(df)
> buf = pa.InMemoryOutputStream()
> pq.write_table(table, buf)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
