On 9/25/2018 2:14 PM, Hanjan, Harinder wrote:
Hello!
When starting a new topic on the mailing list, do not reply to an existing message. Your thread is buried within a thread originally titled "Extracting top level URL when indexing document".
https://home.apache.org/~hossman/#threadhijack
Notice that the Communities facet has 2 non zero results. I understand this is because I'm using fq to get only documents which contain BANFF TRAIL but those documents also contain PARKDALE.
Facets return information for what the document that match the query contain. ALL of the information. The query that returned those matches is not examined at all when calculating facets, only the *results* of the query are examined. I don't think there's any way you can exclude the information that you want to exclude, other than removing it from the documents entirely. I would imagine that the PARKDALE information is required in those documents for other purposes and probably can't be removed.
Thanks, Shawn
