[
https://issues.apache.org/jira/browse/LUCENE-3589?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13156715#comment-13156715
]
Robert Muir commented on LUCENE-3589:
-------------------------------------
We really cannot just let BytesRef pile up into some mega API.
When we add things to the public API, we have to eventually support this stuff
in future releases.
If we are hell-bent to keep the api, we must fix the bogosities about it,
otherwise its unmaintainable, and should be removed:
* the fact it is copy(short) is wrong, if i delete the copy(short) and
copy(int) methods everything compiles as normal, leading me to believe they are
unused. this is because anything using thse gets promoted to copy(long). So the
methods must be renamed to copyShort, copyInt, etc to prevent these types of
mistakes.
* why do we reset the offset to 0?
In general i'm just wondering why this is needed in bytesref itself... do
people know about ByteArrayDataInput and ByteArrayDataOutput?
Separately, I'm still going to remove all unused bytesref methods at the
current moment... i'll leave this one be though.
> BytesRef copy short missed the length setting
> ---------------------------------------------
>
> Key: LUCENE-3589
> URL: https://issues.apache.org/jira/browse/LUCENE-3589
> Project: Lucene - Java
> Issue Type: Bug
> Components: core/other
> Affects Versions: 4.0
> Environment: linux 64bit jdk 6
> Reporter: peter chang
> Assignee: Robert Muir
> Labels: patch
> Fix For: 4.0
>
> Attachments: BytesRef_bug_lucene3589.patch
>
>
> when storing a short type integer to BytesRef, BytesRef missed the length
> setting. then it will cause the storage size is ZERO if no continuous options
> on this BytesRef
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
