[
https://issues.apache.org/jira/browse/LUCENE-5794?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14048083#comment-14048083
]
Ryan Ernst commented on LUCENE-5794:
------------------------------------
The code seems fine, given your description of desired behavior. If I
understand correctly, you would use this in existing code today, which would
speed it up if backed by a random access codec? ie, no worse than today?
> Add a slow random-access ords wrapper
> -------------------------------------
>
> Key: LUCENE-5794
> URL: https://issues.apache.org/jira/browse/LUCENE-5794
> Project: Lucene - Core
> Issue Type: Improvement
> Reporter: Adrien Grand
> Assignee: Adrien Grand
> Priority: Minor
> Fix For: 5.0, 4.10
>
> Attachments: LUCENE-5794.patch
>
>
> Even if you are using an algorithm that requires random access (eg. sorting
> based on the maximum value), it might still be ok to allow for it
> occasionally on a codec that doesn't support random access like
> MemoryDocValuesFormat by having a slow random-access wrapper. This slow
> wrapper would need to be enabled explicitely. This would allow to have
> algorithms that are optimized on random-access codecs but still work in the
> general case.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
