[
https://issues.apache.org/jira/browse/SOLR-5116?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13729696#comment-13729696
]
David Smiley commented on SOLR-5116:
------------------------------------
By raising that question, I think you're conflating issues that should be kept
separate. Leave caching of such queries to SOLR-5093. To sorta answer that
question, a bitset in the filter cache is almost always going to be faster than
nearly any other query. Of course it's got to be queried the first time. And
Solr caches filter queries in normalized to their "positive" form; a negative
query will re-use it.
> Rewrite field:* as if field:[* TO *] so field type can optimize
> ---------------------------------------------------------------
>
> Key: SOLR-5116
> URL: https://issues.apache.org/jira/browse/SOLR-5116
> Project: Solr
> Issue Type: Improvement
> Components: query parsers
> Reporter: David Smiley
>
> The queries {{field:*}} and {{field:[* TO *]}} are semantically equivalent
> but the latter is potentially must faster since the field type can intercept
> it to return a more optimal implementation than a brute force scan of the
> terms. The trie numeric fields do this. The spatial ones don't but they
> could be rewritten easily to do so.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
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]
