You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@impala.apache.org by "Zoltan Ivanfi (JIRA)" <ji...@apache.org> on 2017/09/26 11:27:00 UTC

[jira] [Created] (IMPALA-5982) String columns saved to Parquet files should be annotated with the UTF8 logical type

Zoltan Ivanfi created IMPALA-5982:
-------------------------------------

             Summary: String columns saved to Parquet files should be annotated with the UTF8 logical type
                 Key: IMPALA-5982
                 URL: https://issues.apache.org/jira/browse/IMPALA-5982
             Project: IMPALA
          Issue Type: Bug
            Reporter: Zoltan Ivanfi


When creating Parquet files, Impala doesn't add the proper logical type corresponding to the string SQL type. String columns should be annotated with the [{{UTF8}}|https://github.com/apache/parquet-format/blob/master/LogicalTypes.md#utf8-strings] logical type.

The lack of the {{UTF8}} logical type annotation makes it harder to consume the data using other tools, or even by Impala itself if the files are manually moved around in the filesystem and a new table has to be created based on their metadata.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)