[
https://issues.apache.org/jira/browse/AVRO-814?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Stu Hood resolved AVRO-814.
---------------------------
Resolution: Later
> In the case of a union this might not do the right thing.
Ah, yea... I see. Thanks for the explanation Doug!
I'll resolve this issue to wait for AVRO-248, which seems like it will remove a
lot of uncertainty from .specific.
> SpecificDatumReader.newRecord always looks up the classname
> -----------------------------------------------------------
>
> Key: AVRO-814
> URL: https://issues.apache.org/jira/browse/AVRO-814
> Project: Avro
> Issue Type: Bug
> Reporter: Stu Hood
>
> Even when given an instance of the record class, SpecificDatumReader looks up
> the Class in the SpecificData Class cache. This breaks decoding into
> subclasses of a generated record class without programmatically modifying the
> schema of the subclass, and is probably unnecessary.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
