hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Phabricator (Commented) (JIRA)" <>
Subject [jira] [Commented] (HIVE-2478) Support dry run option in hive
Date Thu, 22 Dec 2011 19:53:32 GMT


Phabricator commented on HIVE-2478:

ashutoshc has requested changes to the revision "HIVE-2478 [jira] Support dry run option in

  common/src/java/org/apache/hadoop/hive/conf/ Name the property as hive.exec.dryrun.
Also, add it in hive-default.xml.template
  ql/src/java/org/apache/hadoop/hive/ql/ There seems to be a confusion
about these different mode means. You have not used PLAN anywhere. Also, 'off' is confusing.
I don't think there is any need of OFF. Allow three values: PARSE, ANALYZE & PLAN.  and
then match these against the user provided value. In hiveConf, default value returned is empty
string "".
  ql/src/java/org/apache/hadoop/hive/ql/ Use, conf.getVar() instead.
  ql/src/java/org/apache/hadoop/hive/ql/ Without thus does it result in NPE
later? Because returned schema is empty object. It will be good to add that in comment, if
that is so.
  ql/src/java/org/apache/hadoop/hive/ql/ I think you want to end this phase
after sem.validate().
  ql/src/java/org/apache/hadoop/hive/ql/ This should be DryMode.PLAN
  ql/src/test/queries/clientnegative/dryrun_bad_fetch_serde.q:1 Please add some comments at
the top of script, explaining which phase should succeed and which should fail in these tests.
You can also come up with better filenaemes for these tests which explains the intent.

  This applies for all tests


> Support dry run option in hive
> ------------------------------
>                 Key: HIVE-2478
>                 URL:
>             Project: Hive
>          Issue Type: Improvement
>          Components: Configuration
>    Affects Versions: 0.9.0
>            Reporter: kalyan ram
>            Assignee: Sushanth Sowmyan
>            Priority: Minor
>         Attachments: HIVE-2478-1.patch, HIVE-2478-2.patch, HIVE-2478-3.patch, HIVE-2478.D927.1.patch
> Hive currently doesn't support a dry run option. For some complex queries we just want
to verify the query syntax initally before running it. A dry run option where just the parsing
is done without actual execution is a good option.

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:!default.jspa
For more information on JIRA, see:


View raw message