[
https://issues.apache.org/jira/browse/CRUNCH-277?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Tom White updated CRUNCH-277:
-----------------------------
Attachment: CRUNCH-277.patch
Thanks for reviewing the patch!
I found that the Parquet source is not compatible with CombineFileInputFormat,
since the ParquetRecordReader expects a ParquetInputSplit (which encodes
Parquet block information), rather than a regular FileSplit. To fix this I've
disabled the use of combine files for the Parquet source, and added a new test
to verify it works.
> Support Parquet
> ---------------
>
> Key: CRUNCH-277
> URL: https://issues.apache.org/jira/browse/CRUNCH-277
> Project: Crunch
> Issue Type: New Feature
> Components: IO
> Reporter: Tom White
> Assignee: Tom White
> Attachments: CRUNCH-277.patch, CRUNCH-277.patch
>
>
> Add a source and target for Parquet files.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
