[
https://issues.apache.org/jira/browse/GORA-212?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13646389#comment-13646389
]
Renato Javier Marroquín Mogrovejo commented on GORA-212:
--------------------------------------------------------
Hi Lewis, thanks for getting those changes into a cleaner patch!
What there is to discuss is that how we want to store the schema used while
inserting the new object. Because if we were talking about a single-type union,
we would only have one type (i.e. null + a single type) so it is easy to know
which schema we are using.
However, when using multi-type union data types we should keep some record of
the schema used in order to chose the correct serializer whenever we decide to
obtain the data back out. In other words, we'd need to add a new column for
storing the index of the selected type within the union data type. I am not
sure if we/users want to store an extra column for the ability to use
multi-types unions. What do you think man?
> Verify storage and retrieval of Avro null-multi-type Union data type within
> Gora-Cassandra
> ------------------------------------------------------------------------------------------
>
> Key: GORA-212
> URL: https://issues.apache.org/jira/browse/GORA-212
> Project: Apache Gora
> Issue Type: Improvement
> Components: gora-cassandra
> Affects Versions: 0.3
> Reporter: Renato Javier Marroquín Mogrovejo
> Assignee: Renato Javier Marroquín Mogrovejo
> Priority: Minor
> Labels: gora-cassandra, gora-core
> Fix For: 0.3
>
> Attachments: GORA-212.v1.patch, GORA-212.v2.patch
>
>
> Verify storage and retrieval of Avro null-multi-type Union data type within
> Gora-Cassandra
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
