[
https://issues.apache.org/jira/browse/LUCENE-3467?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Simon Willnauer updated LUCENE-3467:
------------------------------------
Attachment: LUCENE-3467.patch
here is a patch that factors out multiple base classes shared across variant
specific docvalues writers / readers. Numerics subclasses
FixedStraightBytesImpl.Writer and reuses and only encodes incoming values as a
fixed sized bytes. Other byte variants share common base classes and override
necessary parts for their specific variant.
This removes a large amount of code from docvalues which I should have done
earlier. However, this is a first step towards fixing the merge problem since
we can now almost exclusively operate on bytes rather than ints, floats,
doubles etc. on the implementation level.
> Cut over numeric docvalues to fixed straight bytes
> --------------------------------------------------
>
> Key: LUCENE-3467
> URL: https://issues.apache.org/jira/browse/LUCENE-3467
> Project: Lucene - Java
> Issue Type: Improvement
> Components: core/index
> Affects Versions: 4.0
> Reporter: Simon Willnauer
> Fix For: 4.0
>
> Attachments: LUCENE-3467.patch
>
>
> Currently numeric docvalues types are encoded and stored individually which
> creates massive duplication of writing / indexing code. Yet, almost all of
> them (except packed ints) are essentially a fixed straight bytes variant.
--
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]
