[
https://issues.apache.org/jira/browse/SOLR-6452?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14127542#comment-14127542
]
Tomás Fernández Löbbe edited comment on SOLR-6452 at 9/9/14 9:01 PM:
---------------------------------------------------------------------
Thanks for the patch [~simpleBread]. As an optimization, why don't you make the
{{missingStats}} a {{Map<Integer, Integer>}} and use the ords as keys instead
of the terms. That way you don't need to do the lookupOrd for all docs, and you
do it only once per term in the {{accumulateMissing()}} method.
was (Author: tomasflobbe):
Thanks for the patch Xu. As an optimization, why don't you make the
{{missingStats}} a {{Map<Integer, Integer>}} and use the ords as keys instead
of the terms. That way you don't need to do the lookupOrd for all docs, and you
do it only once per term in the {{accumulateMissing()}} method.
> StatsComponent "missing" stat won't work with docValues=true and indexed=false
> ------------------------------------------------------------------------------
>
> Key: SOLR-6452
> URL: https://issues.apache.org/jira/browse/SOLR-6452
> Project: Solr
> Issue Type: Bug
> Affects Versions: 5.0, 4.10
> Reporter: Tomás Fernández Löbbe
> Attachments: SOLR-6452-trunk.patch, SOLR-6452-trunk.patch
>
>
> StatsComponent can work with DocValues, but it still required to use
> indexed=true for the "missing" stat to work. Missing values should be
> obtained from the docValues too.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
