[
https://issues.apache.org/jira/browse/LUCENE-5260?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13792662#comment-13792662
]
Michael McCandless commented on LUCENE-5260:
--------------------------------------------
Thanks Areek, patch looks great! I like the hasPayloads() up-front
introspection.
In UnsortedTermFreqIteratorWrapper.payload(), why do we set currentOrd
as a side effect? Shouldn't next() already do that? Maybe, we should
instead assert currentOrd == ords[curPos]? Also, can we break that
sneaky currentOrd assignment in next into its own line before?
> Make older Suggesters more accepting of TermFreqPayloadIterator
> ---------------------------------------------------------------
>
> Key: LUCENE-5260
> URL: https://issues.apache.org/jira/browse/LUCENE-5260
> Project: Lucene - Core
> Issue Type: Improvement
> Components: core/search
> Reporter: Areek Zillur
> Attachments: LUCENE-5260.patch
>
>
> As discussed in https://issues.apache.org/jira/browse/LUCENE-5251, it would
> be nice to make the older suggesters accepting of TermFreqPayloadIterator and
> throw an exception if payload is found (if it cannot be used).
> This will also allow us to nuke most of the other interfaces for
> BytesRefIterator.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
