[
https://issues.apache.org/jira/browse/HBASE-9508?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13765594#comment-13765594
]
stack commented on HBASE-9508:
------------------------------
bq. The equivalent of "removing" them would be to just not commit the
deprecated methods to trunk right?
Yes.
And then there is the suggestion that since 0.98 is just after 0.96, that we
leave the deprecated in till the release after.
And after removal, we'd then have getTable which returns a TableName and no
getTableName method. Would make users wonder. At least if there is a
getTableName still in place, there'd be no API hole (and a pointer to getTable).
> Restore some API mistakenly removed in client, mapred*, and common
> ------------------------------------------------------------------
>
> Key: HBASE-9508
> URL: https://issues.apache.org/jira/browse/HBASE-9508
> Project: HBase
> Issue Type: Bug
> Components: Usability
> Reporter: stack
> Assignee: stack
> Priority: Critical
> Fix For: 0.98.0, 0.96.0
>
> Attachments: 9508.txt, 9508v2.txt
>
>
> Here is contrib to the API compatibility story. I went over Aleks'
> compatibility report and restored removed or overriden methods and
> constructors, stuff that was in 0.94 non-deprecated and removed in 0.96.
> This patch is not comprehensive because some removals cannot be restored as
> in those that used take Writables (more on this later from Jon).
> The changes included here are mostly restore of methods that took a table
> name as a byte array replaced by a TableName object.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
