accumulo-notifications mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Josh Elser (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (ACCUMULO-2758) LocalityGroupUtil.encodeColumnFamily() ignores offset parameter
Date Fri, 16 May 2014 10:56:35 GMT

    [ https://issues.apache.org/jira/browse/ACCUMULO-2758?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13999133#comment-13999133
] 

Josh Elser commented on ACCUMULO-2758:
--------------------------------------

[~gary2086], a couple of things:

* Please follow the [contributor guidelines|http://accumulo.apache.org/git.html#contributors]
for future patches. This helps us accurately give you credit for the work you're submitting.
* It looks like you removed the public method. As callers might be depending on this, we should
preserve that method. I believe Bill's original description better hints towards moving the
implementation from the private method to the public method (which you deleted). The private
method's implementation isn't sufficient for the intent given its arguments, but works perfectly
fine when operating over the entire ByteSequence.

Does that make sense?

> LocalityGroupUtil.encodeColumnFamily() ignores offset parameter
> ---------------------------------------------------------------
>
>                 Key: ACCUMULO-2758
>                 URL: https://issues.apache.org/jira/browse/ACCUMULO-2758
>             Project: Accumulo
>          Issue Type: Bug
>    Affects Versions: 1.4.5, 1.5.1
>            Reporter: Bill Havanki
>            Assignee: Gary Singh
>            Priority: Minor
>              Labels: newbie
>             Fix For: 1.5.2, 1.6.1, 1.7.0
>
>         Attachments: ACCUMULO-2758-2.patch, ACCUMULO-2758.patch
>
>
> The method {{LocalityGroupUtil.encodeColumnFamily(ByteSequence)}} heeds the current offset
and length of the byte sequence, passing them to an private method that does the encoding.
However, the private method ignores its offset parameter, and always starts from the beginning
of the byte array.
> Fortunately, it appears that there are no callers for the public method, and the other
caller of the private method always passes an offset of zero. So, the bug is probably not
expressed. The method with the bug should probably just be eliminated.



--
This message was sent by Atlassian JIRA
(v6.2#6252)

Mime
View raw message