[
https://issues.apache.org/jira/browse/SOLR-8919?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15553671#comment-15553671
]
Joel Bernstein edited comment on SOLR-8919 at 10/7/16 12:27 AM:
----------------------------------------------------------------
The approach I was going to use is to identify if it's a count distinct query
and then just add a count stream to the select distinct logic. This would be in
MapReduce mode. I believe count distinct is directly supported by the JSON
facet API, so in facet mode, this might involve a change to the FacetStream. Or
for initial FacetMode implementation we could just use the select distinct
facet code and wrap a count stream. This would be easier then changing the
FacetStream.
was (Author: joel.bernstein):
The approach I was going to use is to identify if it's a count distinct query
and then just add a count stream to the select distinct logic. This would be in
MapReduce mode. I believe count distinct is directly supported by the JSON
facet API, so in facet mode, this might involve a change to the FacetStream.
> Add SELECT COUNT(DISTINCT COL) queries to the SQL Interface
> -----------------------------------------------------------
>
> Key: SOLR-8919
> URL: https://issues.apache.org/jira/browse/SOLR-8919
> Project: Solr
> Issue Type: Bug
> Reporter: Joel Bernstein
> Fix For: 6.2, master (7.0)
>
>
> While analyzing the Enron emails for SOLR-8888, I was wishing that
> COUNT(DISTINCT) was implemented. This ticket is to implement it.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
