[
https://issues.apache.org/jira/browse/HIVE-12339?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15049636#comment-15049636
]
Sushanth Sowmyan commented on HIVE-12339:
-----------------------------------------
Hi, I believe this limitation was by design. WebHCat was designed to be a
job-submitter that could sit on a gateway that users can access, but not expose
any direct underlying data directly, and instead defer to WebHDFS to manage
that. This decision also gets around the question of how to handle returns of
things like selects when the returns are truly large. For most part, hive query
returns are expected to be large.
Whether or not we want to re-evaluate that is up to community decision.
For most of the usecases associated with remote-running of queries and fetching
results, HS2 via jdbc is the recommended way of using hive.
> WebHCat should have an option to return a result instead of jobId
> -----------------------------------------------------------------
>
> Key: HIVE-12339
> URL: https://issues.apache.org/jira/browse/HIVE-12339
> Project: Hive
> Issue Type: New Feature
> Components: WebHCat
> Affects Versions: 1.2.1
> Reporter: Artem Ervits
> Priority: Minor
>
> the following command should have an option to get result of the hive query
> rather than saving output to hdfs
> curl -s -d execute="select+*+from+sample_08;" \
> -d statusdir="outputofthejob" \
> 'http://localhost:50111/templeton/v1/hive?user.name=root'
> this currently returns jobid in JSON format, it would be nice to get the
> actual result as part of JSON.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
