[
https://issues.apache.org/jira/browse/SOLR-12795?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Amrit Sarkar updated SOLR-12795:
--------------------------------
Description:
Today facetStream takes a "bucketSizeLimit" parameter. Here is what the doc
says about this parameter - The number of buckets to include. This value is
applied to each dimension.
Now let's say we create a facet stream with 3 nested facets. For example
"year_i,month_i,day_i" and provide 10 as the bucketSizeLimit.
FacetStream would return 10 results to us for this facet expression while the
total number of unqiue values are 1000 (10*10*10 )
The API should have a separate parameter "limit" which limits the number of
tuples while bucketSizeLimit should be used to specify the size of each bucket
in the JSON Facet API.
was:
Let's look at an observation regarding "bucketSizeLimit" in facetStream; and
how we interpret it as a "limit". Suppose for 3 nested facets, bucketSizeLimit
= 10, we receive total 1000 rows. since bucketSizeLimit = limit; ONLY the first
top-level facet value's count will be returned; out of 10*10*10, 1*1*10th rows
will be fetched. And the behavior will be consistent for any bucketSizeLimit we
set,
How about we have a separate parameter "limit" other than "bucketSizeLimit"
which can be set to any arbitrary number (though should be <
bucketSizeLimit^no_of_nested_facets), and that limit can be said "500". In this
way, we will have the true SQL limit feature in place in FacetStream.
> Introduce 'limit' parameter in FacetStream.
> -------------------------------------------
>
> Key: SOLR-12795
> URL: https://issues.apache.org/jira/browse/SOLR-12795
> Project: Solr
> Issue Type: Sub-task
> Security Level: Public(Default Security Level. Issues are Public)
> Components: streaming expressions
> Reporter: Amrit Sarkar
> Priority: Major
> Attachments: SOLR-12795.patch
>
>
> Today facetStream takes a "bucketSizeLimit" parameter. Here is what the doc
> says about this parameter - The number of buckets to include. This value is
> applied to each dimension.
> Now let's say we create a facet stream with 3 nested facets. For example
> "year_i,month_i,day_i" and provide 10 as the bucketSizeLimit.
> FacetStream would return 10 results to us for this facet expression while the
> total number of unqiue values are 1000 (10*10*10 )
> The API should have a separate parameter "limit" which limits the number of
> tuples while bucketSizeLimit should be used to specify the size of each
> bucket in the JSON Facet API.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
