[
https://issues.apache.org/jira/browse/DRILL-3963?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14977438#comment-14977438
]
ASF GitHub Bot commented on DRILL-3963:
---------------------------------------
Github user sudheeshkatkam commented on the pull request:
https://github.com/apache/drill/pull/214#issuecomment-151680840
@amithadke I added a unit test to this
[branch](https://github.com/sudheeshkatkam/drill/commits/local-sequence-files)
that fails. Note that user0:group0 creates a view with permission 750 on top of
a file with permission 700, so user1:group0 should be able to query the view.
The change in ScanBatch is to make sure all readers are initialized during
fragment execution, and not setup time. Ideally the test case should ensure
that at least two readers are created within one fragment.
> Read raw key value bytes from sequence files
> --------------------------------------------
>
> Key: DRILL-3963
> URL: https://issues.apache.org/jira/browse/DRILL-3963
> Project: Apache Drill
> Issue Type: New Feature
> Reporter: amit hadke
> Assignee: amit hadke
>
> Sequence files store list of key-value pairs. Keys/values are of type hadoop
> writable.
> Provide a format plugin that reads raw bytes out of sequence files which can
> be further deserialized by a udf(from hadoop writable -> drill type)
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
