[
https://issues.apache.org/jira/browse/HBASE-17320?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15778337#comment-15778337
]
Duo Zhang commented on HBASE-17320:
-----------------------------------
There is a semantic change for the Scan object. In the old time, we assume the
scan is a get if we found that the start row and stop row of it are equal.
After we introduce the inclusive of the stopRow, it is more natural for us to
require that the includeStopRow to also be true when we decide whether a scan
is a get.
This lead to the failure of some UTs and I've add a dirty hack in the
constructor Scan(byte[], byte[]) to fix them. If the startRow and stopRow are
equal, then I will set the includeStopRow to true. I'm not sure if I should
keep all the old behaviors as the old behavior seems wrong after we introduce
the inclusive of stop row.
Suggestions are welcomed. Thanks.
> Add inclusive/exclusive support for startRow and endRow of scan
> ---------------------------------------------------------------
>
> Key: HBASE-17320
> URL: https://issues.apache.org/jira/browse/HBASE-17320
> Project: HBase
> Issue Type: Sub-task
> Components: Client, scan
> Affects Versions: 2.0.0
> Reporter: Duo Zhang
> Assignee: Duo Zhang
> Fix For: 2.0.0
>
> Attachments: HBASE-17320-v1.patch, HBASE-17320-v2.patch,
> HBASE-17320.patch
>
>
> This is especially useful when doing reverse scan. HBASE-17168 maybe a more
> powerful solution but we need to be careful about the atomicity, and I do not
> think we will provide the feature to end user. But I think it is OK to
> provide inclusive/exclusive option to end user.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
