[
https://issues.apache.org/jira/browse/SOLR-209?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12488754
]
Rida Benjelloun commented on SOLR-209:
--------------------------------------
Hi Yonik,
The MultiFieldsHitsfacets use lucene hits to create the facet. For each field
on which we want to make a facet, I get the field content from the lucene
document in the search results, and I check if the Map key is equals to the
field content, if no, I add it as a key in the Hashmap, else I get the Map
content wish is an Integer and I increment it. Finally I sort the HashMap.
With this approach you can tokenized your fields, and it will not have any
impact on the faceting mechanism because I use the string content to make the
facets, however the field must be stored.
Regards.
Rida Benjelloun
> Multifields and multivalued facets
> ----------------------------------
>
> Key: SOLR-209
> URL: https://issues.apache.org/jira/browse/SOLR-209
> Project: Solr
> Issue Type: Improvement
> Components: search
> Affects Versions: 1.1.0
> Environment: Java
> Reporter: Rida Benjelloun
> Fix For: 1.1.0
>
> Attachments: MultiFieldsHitsFacets.java, MultiFieldsHitsFacets.patch
>
>
> MultiFieldsHitsFacets, increase the performance of faceting in multiValued
> fields, buy creating facets from Lucene Hits. It also allows the creation of
> facet using multiple fields. The fields must be separated by single space
> Example : facet.field=subject subjectG subjectA .
> Rida Benjelloun
> [EMAIL PROTECTED]
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
