Ok sounds good. Grouping is an interesting distrib case. On Fri, Jun 10, 2011 at 9:27 AM, Michael McCandless <[email protected]> wrote: > I'm actually working on something like this, basically a utility > method to merge N TopDocs into 1. I want to do this for grouping as > well to make it easy to do grouping across shards. > > Mike McCandless > > http://blog.mikemccandless.com > > On Fri, Jun 10, 2011 at 12:25 PM, Jason Rutherglen > <[email protected]> wrote: >>> its fine to have some higher-level class to support this crap, but it >>> shouldnt be some transparent searcher. >> >> I'll create a patch and post to a Jira. >> >> On a side note, for multi threaded calls I noticed there's a lock on >> the PQ in IndexSearcher, is the performance of that OK? >> >> On Fri, Jun 10, 2011 at 9:21 AM, Robert Muir <[email protected]> wrote: >>> On Fri, Jun 10, 2011 at 12:18 PM, Jason Rutherglen >>> <[email protected]> wrote: >>>> I think we only need to resurrect the merge score/field-docs code, in >>>> it's own class. Eg, each sub-node is expected to create it's own >>>> score/field-docs, then the merge code is centralized. >>>> >>>>> Maybe we should bring back the 'merge results' part of multi searcher >>>>> without the query rewrite. >>>> >>>> This is how Solr works today. >>>> >>> >>> no its not, the problem with multisearcher was it was too low of a level. >>> >>> its fine to have some higher-level class to support this crap, but it >>> shouldnt be some transparent searcher. >>> >>> --------------------------------------------------------------------- >>> To unsubscribe, e-mail: [email protected] >>> For additional commands, e-mail: [email protected] >>> >>> >> >> --------------------------------------------------------------------- >> To unsubscribe, e-mail: [email protected] >> For additional commands, e-mail: [email protected] >> >> > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > >
--------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
