[
https://issues.apache.org/jira/browse/LUCENE-6351?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14352301#comment-14352301
]
Michael McCandless commented on LUCENE-6351:
--------------------------------------------
Sorry, I agree this got a lot worse with LUCENE-6161.
But I already opened LUCENE-6315 to simplify PrefixCodedTerms by cutting over
to BytesRefIterator. I think that's a good step?
With Iterator<Term> I didn't like that we had to do a field.equals() every
time. The vast majority of delete-by-term in Lucene is against a single
"unique identifier" field.
> change PrefixCodedTerms.iterator to java.util.Iterator
> ------------------------------------------------------
>
> Key: LUCENE-6351
> URL: https://issues.apache.org/jira/browse/LUCENE-6351
> Project: Lucene - Core
> Issue Type: Task
> Reporter: Robert Muir
>
> The current api is horrible... no performance advantage can justify this.
> We should use java.util.iterator like before again.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
