I tried facetting also but not worked smoothly for me. Case i had mentioned in email is dummy one and my actual index is with 12 lakh docs and 2 GB size on single machine. Each of tenant_pool field value has 20-30 tokens. Getting all terms in tenant_pool is fast in seconds but when i go with facet path after filter criteria then that is very slow. Because it is reading whole field from disk and i am only interested in terms.
On Fri, Jan 16, 2015 at 1:48 PM, Ahmet Arslan <iori...@yahoo.com.invalid> wrote: > Hi Naresh, > > Yup terms component does not respect q or fq parameter. > Luckily, thats easy with facet component. Example : > facet=true&facet.field=tenant_pool&q=type:1 > > Please see more here : > https://cwiki.apache.org/confluence/display/solr/Faceting > > happy faceting, > ahmet > > > > On Friday, January 16, 2015 10:13 AM, Naresh Yadav <nyadav....@gmail.com> > wrote: > Hi ahmet, > > Thanks, now i understand better, i will not try my usecase with grouping. > Actually i am interested in unique terms in a field i.e tenant_pool. That i > get perfectly with http://www.imagesup.net/?di=614212438580 > > But i am not able to get terms after applying some filter say "type":"1". > That is I need unique terms in "tenant_pool" field for "type":"1" query and > answer will be P1, L1. > Please suggest me if i can get this with out reading each doc from disk. > > > On Fri, Jan 16, 2015 at 1:28 PM, Ahmet Arslan <iori...@yahoo.com.invalid> > wrote: > > > Hi Naresh, > > > > I have never grouped on a tokenised field and I am not sure it makes > sense > > to do so. > > > > Reading back ref-guide it says this about group.field parameter > > > > "The name of the field by which to group results. The field must be > > single-valued, and either be indexed or a field type that has a value > > source and works in a function query, such as ExternalFileField. It must > > also be a string-based field, such as StrField or TextField" > > > > > > https://cwiki.apache.org/confluence/display/solr/Result+Grouping > > > > Therefore, it should be single valued. P.S. Don't get confused with > > TextField type, for example it could create single token when used with > > keyword tokenizer. > > > > Ahmet > > > > On Friday, January 16, 2015 4:43 AM, Naresh Yadav <nyadav....@gmail.com> > > wrote: > > Hi ahmet, > > > > If you observe output ngroups is 1 and returning only one group P1. > > But my expectation is it should return three groups P1, L1, L2 as my > > field is tokenized with space. > > > > Please correct me if wrong? > > > > > > On 1/15/15, Ahmet Arslan <iori...@yahoo.com.invalid> wrote: > > > > > > > > > Hi Naresh, > > > > > > Everything looks correct, what is the problem here? > > > > > > If you want to see more than one document per group, there is a > parameter > > > for that which defaults to 1. > > > > > > Ahmet > > > > > > > > > > > > On Thursday, January 15, 2015 9:02 AM, Naresh Yadav < > > nyadav....@gmail.com> > > > wrote: > > > Hi all, > > > > > > I had done following configuration to test Solr grouping concept. > > > > > > solr version : 4.6.1 (tried in latest version 4.10.3 also) > > > Schema : http://www.imagesup.net/?di=10142124357616 > > > Solrj code to insert docs : > > http://www.imagesup.net/?di=10142124381116 > > > Response Group's : http://www.imagesup.net/?di=1114212438351 > > > Response Terms' : http://www.imagesup.net/?di=614212438580 > > > > > > Please let me know if am i doing something wrong her > > >
