[ 
https://issues.apache.org/jira/browse/SOLR-7253?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

David Donohue updated SOLR-7253:
--------------------------------
    Description: 
Solr 5.0 added stats to its pivot facet component, so that this query 

{code}
facet=true&stats=true&stats.field={!tag=t1}income&facet.pivot={!stats=t1}tags&facet.prefix=university
{code}

returns stats for each facet with a tag that starts with "university".  Very 
fast.
It returns these stats:  min,max,count,missing,sum,sumOfSquares,mean,stddev

The issue is that it returns these stats for ALL facets matching the query 
criteria.  No ability to limit or sort, e.g. to return the top 20 earning 
universities.

If this functionality can be added, then Solr could deliver on a more complete 
ability for analytics, rivaling more complex SQL aggregate queries.

  was:
Solr 5.0 added stats to its pivot facet component, so that this query 

facet=true&stats=true&stats.field={!tag=t1}income&facet.pivot={!stats=t1}tags&facet.prefix=university

returns stats for each facet with a tag that starts with "university".  Very 
fast.
It returns these stats:  min,max,count,missing,sum,sumOfSquares,mean,stddev

The issue is that it returns these stats for ALL facets matching the query 
criteria.  No ability to limit or sort, e.g. to return the top 20 earning 
universities.

If this functionality can be added, then Solr could deliver on a more complete 
ability for analytics, rivaling more complex SQL aggregate queries.


> sort, limit for PivotFacet stats
> --------------------------------
>
>                 Key: SOLR-7253
>                 URL: https://issues.apache.org/jira/browse/SOLR-7253
>             Project: Solr
>          Issue Type: New Feature
>    Affects Versions: 5.0
>            Reporter: David Donohue
>              Labels: Analytics, Facets
>
> Solr 5.0 added stats to its pivot facet component, so that this query 
> {code}
> facet=true&stats=true&stats.field={!tag=t1}income&facet.pivot={!stats=t1}tags&facet.prefix=university
> {code}
> returns stats for each facet with a tag that starts with "university".  Very 
> fast.
> It returns these stats:  min,max,count,missing,sum,sumOfSquares,mean,stddev
> The issue is that it returns these stats for ALL facets matching the query 
> criteria.  No ability to limit or sort, e.g. to return the top 20 earning 
> universities.
> If this functionality can be added, then Solr could deliver on a more 
> complete ability for analytics, rivaling more complex SQL aggregate queries.



--
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

Reply via email to