[jira] [Commented] (SOLR-12983) JavabinLoader should avoid creating String Objects and create UTF8CharSequence fields from byte[]

["Yonik Seeley (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SOLR-12983?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16989972#comment-16989972 ] 

Yonik Seeley commented on SOLR-12983:
-------------------------------------

Folks, in the future we really need more description about what is trying to be achieved and why.  The only description in this whole issue was

"Javabin stings already contain Strings in UTF8 byte[] format. String fields can be created directly from those"

The why, and what it is supposed to achieve is totally missing.
Perhaps it is to receive UTF-8 and be able to directly index it for string values?  If so, it doesn't look like this happens.
Or is it really just to attempt to save re-encoding cost when forwarding documents?  It's made everything else slower (and buggier, and harder to understand), so if this was the sole motivation, it would have been nice to know up front.

 

 

> JavabinLoader should avoid creating String Objects and create UTF8CharSequence  fields from byte[]
> --------------------------------------------------------------------------------------------------
>
>                 Key: SOLR-12983
>                 URL: https://issues.apache.org/jira/browse/SOLR-12983
>             Project: Solr
>          Issue Type: Improvement
>            Reporter: Noble Paul
>            Assignee: Noble Paul
>            Priority: Blocker
>             Fix For: 7.7.1, 8.0
>
>         Attachments: SOLR-12983.patch
>
>
> Javabin stings already contain Strings in UTF8 byte[] format. String fields can be created directly from those



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org