That is not currently configurable, no. However, you can implement your own extractor to do so. I think the original reasoning for '_' over '.' was that Lucene query syntax allows '.' to specify the index name. Having multiple '.' in a field name might break the query parser.
On Sat, Jul 14, 2012 at 11:35 AM, Andrew Zeneski <[email protected]>wrote: > Sean, > > One thing I have been meaning to ask is if we can change the character > used when flattening the fields. I'd like to use a character besides _ so > that I can identify the parent from a field that has _ in its name. I would > probably change it to '.' personally. > > Is this configurable? > > Andrew > > On Jul 14, 2012, at 11:31 AM, Sean Cribbs wrote: > > Hi (what's your name?), > > What a search query returns to you is the "index document", or the result > of transforming the input before adding it to the search indexes; this also > means that deeply nested fields will be flattened. If you want the > original, you can use MapReduce to retrieve the documents resulting from a > search query. > > On Fri, Jul 13, 2012 at 8:54 PM, 7zark7 <[email protected]> wrote: > >> I would like to store JSON documents in a Riak 1.1.x bucket (using >> bitcask), and would like to search for them via REST using JSON property >> values. >> >> E.g. JSON docs with typed properties like this: >> >> { >> "type": "whatever", >> "owner": 123456, >> "text": "Blah blah blah", >> "enabled": true >> } >> >> However, the query response >> (/solr/mybucket/select?q=enabled:true&wt=json) returns each doc's >> properties as JSON strings, and not the original JSON type: >> >> {"responseHeader":{ >> "status":0, >> "QTime":1, >> "params":{ >> "q":"enabled:true", >> "q.op":"or", >> "filter":"", >> "wt":"json" >> } >> },"response":{ >> "numFound":1, >> "start":0, >> "maxScore":"0.353553", >> "docs":[ >> { >> "id":"Obdoy3XbJe3kdh1T2bfxJgISCCL", >> "index":"rest", >> * "fields":{* >> * "enabled": "true",* >> * "owner": "123456",* >> * "text": "Blah blah blah",* >> * "type": "whatever"* >> * },* >> "props":{} >> } >> ] >> }} >> >> This makes the results unusable for me, and I'm concerned I'd have to >> fall back to N calls back to retrieve each original doc, which is not >> feasible. >> >> >> I'm used to CouchDB+Lucene which behaves "correctly" to me. >> >> Is there some way to do what I'd like? E.g. another query method or >> option? >> >> Thanks >> >> >> >> _______________________________________________ >> riak-users mailing list >> [email protected] >> http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com >> >> > > > -- > Sean Cribbs <[email protected]> > Software Engineer > Basho Technologies, Inc. > http://basho.com/ > > _______________________________________________ > riak-users mailing list > [email protected] > http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com > > > -- Sean Cribbs <[email protected]> Software Engineer Basho Technologies, Inc. http://basho.com/
_______________________________________________ riak-users mailing list [email protected] http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com
