[
https://issues.apache.org/jira/browse/KUDU-2300?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16366013#comment-16366013
]
Grant Henke commented on KUDU-2300:
-----------------------------------
I think that Kudu internally can optimize it's ranges into slightly different
ranges but still correct ranges. For example kudu might increment by the
smallest unit and convert < to <=.
I agree this could be confusing because it doesn't match exactly the same
syntax as when created, but it is in fact the same logic.
I need to look deeper to validate that is in-fact what is happening.
In your example what values/command did you use to create the range partition?
> Partition schema doesn't show correct type of bounds for range partitions
> -------------------------------------------------------------------------
>
> Key: KUDU-2300
> URL: https://issues.apache.org/jira/browse/KUDU-2300
> Project: Kudu
> Issue Type: Bug
> Components: master
> Affects Versions: 1.5.0
> Reporter: Andre Araujo
> Priority: Major
>
> The Partition Schema section of the master Web UI always show the range
> partition with an {{EXCLUSIVE}} upper bound and an {{INCLUSIVE}} lower
> bounce, regardless of what the actual bounds' types are.
> For example, the partition below was created with two {{INCLUSIVE}} bounds,
> but the upper bound is shown incorrectly:
> {code:java}
> HASH (CALLING_NUMBER_INT, CALLED_NUMBER_INT) PARTITIONS 2,
> RANGE (PERIOD_START_TIME) (
> PARTITION 2018-02-15T00:00:00.000001Z <= VALUES <
> 2018-02-16T00:00:00.000000Z
> ){code}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
