[
https://issues.apache.org/jira/browse/HIVE-15795?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15851798#comment-15851798
]
Josh Elser commented on HIVE-15795:
-----------------------------------
(caveat: it appears that the qtests aren't actually running after fallout from
HIVE-14460... will be trying to get that fixed)
Another suggestion, can you please add a qtest file for the index support? In
general, a qtest is some input SQL and then some expected output.
A quick example:
https://github.com/apache/hive/blob/master/accumulo-handler/src/test/queries/positive/accumulo_queries.q
https://github.com/apache/hive/blob/master/accumulo-handler/src/test/results/positive/accumulo_queries.q.out
You can add a new ".q" file to the positives and then just make an example
".out" file (for now).
Then, do the following to build Hive (with your patch) and then run the test
(which is in a different module):
{noformat}
$ mvn clean install -DskipTests
$ cd itest
$ mvn clean package -Dtest=TestAccumuloCliDriver
{noformat}
You should see the tests being run via:
{noformat}
[INFO] --- maven-surefire-plugin:2.19.1:test (default-test) @
hive-it-qfile-accumulo ---
-------------------------------------------------------
T E S T S
-------------------------------------------------------
Java HotSpot(TM) 64-Bit Server VM warning: ignoring option MaxPermSize=256M;
support was removed in 8.0
Running org.apache.hadoop.hive.cli.TestAccumuloCliDriver
Tests run: 6, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 155.283 sec -
in org.apache.hadoop.hive.cli.TestAccumuloCliDriver
Results :
Tests run: 6, Failures: 0, Errors: 0, Skipped: 0
{noformat}
At this point, you can observe the generated output, verify that it's actually
correct for .q file you wrote, and then add that as an .q.out file (replacing
the stub you created initially). LMK if this isn't clear -- I can provide
something on top of your patch to show-by-example :)
> Support Accumulo Index Tables in Hive Accumulo Connector
> --------------------------------------------------------
>
> Key: HIVE-15795
> URL: https://issues.apache.org/jira/browse/HIVE-15795
> Project: Hive
> Issue Type: Improvement
> Components: Accumulo Storage Handler
> Reporter: Mike Fagan
> Assignee: Mike Fagan
> Priority: Minor
> Attachments: HIVE-15795.1.patch
>
>
> Ability to specify an accumulo index table for an accumulo-hive table.
> This would greatly improve performance for non-rowid query predicates
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
