[
https://issues.apache.org/jira/browse/PHOENIX-1312?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15061363#comment-15061363
]
James Taylor commented on PHOENIX-1312:
---------------------------------------
Thanks for the updates, [~ram_krish]. Here's some feedback:
- We still want to favor the empty column family if it's referenced in the
where clause. How about creating a Set<byte[]> from the List<byte[],byte[]>
whereConditions and passing this into getGuidePosts() instead of the List? Then
you can first check for the empty column family in the Set and use that if it's
there. Otherwise, you'd get the first element in the Set and use that. No need
to check for existence in the scan.getFamilyMap().
- For the tests, we need to know that the non empty column family stats where
used. We're currently only checking how many guideposts there are which isn't a
definitive check. One easy way would be to put different sized values in the
columns in each column family - that way they'll have a different number of
guideposts so we can at least infer we're using the non empty column family
stats.
> Do not always project the empty column family
> ---------------------------------------------
>
> Key: PHOENIX-1312
> URL: https://issues.apache.org/jira/browse/PHOENIX-1312
> Project: Phoenix
> Issue Type: Bug
> Reporter: James Taylor
> Assignee: ramkrishna.s.vasudevan
> Fix For: 4.7.0
>
> Attachments: PHOENIX-1312_1.patch, PHOENIX-1312_v2.patch,
> PHOENIX-1312_v3.patch, Phoenix-1312.patch
>
>
> Often times, we don't need to, but it seems we always are. See
> MultiCfQueryExecIT.testGuidePostsForMultiCFs() where we run a query like this:
> {code}
> SELECT count(*) FROM multi_cf WHERE e.cpu_utilization IS NOT NULL
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
