Mailing-List: contact dev-help@lucene.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@lucene.apache.org
Date: Mon, 5 Mar 2012 14:35:57 +0000 (UTC)
From: "Martijn van Groningen (Commented) (JIRA)" <jira@apache.org>
To: dev@lucene.apache.org
Message-ID: 
 <212897094.22482.1330958157220.JavaMail.tomcat@hel.zones.apache.org>
In-Reply-To: 
 <637453396.35862.1329212639489.JavaMail.tomcat@hel.zones.apache.org>
Subject: [jira] [Commented] (LUCENE-3778) Create a grouping convenience
 class
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/LUCENE-3778?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13222368#comment-13222368 ] 

Martijn van Groningen commented on LUCENE-3778:
-----------------------------------------------

{quote}
One can actually use GroupingSearch in a shard'd env, on each shard,
right? It's just that then you merge them like normal on the front
end (ie, TopGroups.merge). Is that the only reason for the "... in a
non distributed environment" javadoc warning?
{quote}
In a sharded env one needs to first execute all the first pass collector on all shards, merge the returned search groups, execute the second pass search (with the top N merged search groups as argument) on most of the shards and finally merge the TopGroups from all shards into a topN TopGroups. Also grouping by docblock and grouping features like allGroups and groupHead don't work in a normal sharded environment (unless you partition the groups properly). The docs caching also only makes sense when performing grouping on a local index. This usage is very different then non distributed grouping, that is why I think it is better to have a separate grouping convenience class for distributed grouping (DistributedGroupSearch?). 

bq. Maybe the ctor should take docValuesType / diskResidentValues
Makes sense!

bq. Maybe you should pass the groupSort, groupsOffset, groupsLimit to the search method (instead of setters)?
Maybe we just should have defaults for these options? Sort.RELEVANCE, 0 and 10?
                
> Create a grouping convenience class
> -----------------------------------
>
>                 Key: LUCENE-3778
>                 URL: https://issues.apache.org/jira/browse/LUCENE-3778
>             Project: Lucene - Java
>          Issue Type: Improvement
>          Components: modules/grouping
>            Reporter: Martijn van Groningen
>         Attachments: LUCENE-3778.patch
>
>
> Currently the grouping module has many collector classes with a lot of different options per class. I think it would be a good idea to have a GroupUtil (Or another name?) convenience class. I think this could be a builder, because of the many options (sort,sortWithinGroup,groupOffset,groupCount and more) and implementations (term/dv/function) grouping has.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org