hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Anoop Sam John (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-10964) Delete mutation is not consistent with Put wrt timestamp
Date Sat, 12 Apr 2014 03:22:15 GMT

    [ https://issues.apache.org/jira/browse/HBASE-10964?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13967373#comment-13967373

Anoop Sam John commented on HBASE-10964:

Committed to 96+ versions.
[~lhofhansl] that is a +1 for 94?
bq.Decided then not to change it to keep API consistent.
It looks like a compatibility change. So I was also worried initially. Later changed my mind.

If a user creates a Delete object passing a TS and add delete columns to that, it is sure
that user want a TS based version(s) delete. So this is like a bug only

> Delete mutation is not consistent with Put wrt timestamp
> --------------------------------------------------------
>                 Key: HBASE-10964
>                 URL: https://issues.apache.org/jira/browse/HBASE-10964
>             Project: HBase
>          Issue Type: Bug
>          Components: Client
>            Reporter: Anoop Sam John
>            Assignee: Anoop Sam John
>         Attachments: HBASE-10964.patch
> We have	Put constructors which take ts param
> eg: Put(byte[] row, long ts)
> When one creates a Put this way and add columns to it, without giving a specific TS,
these individual cells will honour this Put object's TS. One can use add API which takes a
TS and so can override TS for this Cell.
> For delete also we have similar constructors with and without TS params and delete***()
APIs same way as add(). But delete***() APIs (without taking a specific TS) is not honouring
the Delete object's TS.

This message was sent by Atlassian JIRA

View raw message