Query time

2023-02-08 Thread Mike
I have a standalone Solr server and an index of millions of documents. Some queries that e.g. more than 1 million times exist takes a long time. I only need the first 100 results, can I make solr stop ranking and sort by the first 100 hits? How can i limit the search time of sometimes more than 10

Re: Query time

2023-02-08 Thread Andy Lester
Please include your schema and some sample queries so we have specifics to go on. > On Feb 8, 2023, at 9:00 AM, Mike wrote: > > I have a standalone Solr server and an index of millions of documents. > Some queries that e.g. more than 1 million times exist takes a long time. > I only need the

Re: Query time

2023-02-08 Thread Alessandro Benedetti
The short answer is no, you can't ask Solr to stop searching after the first 100 hits are found, the retrieval and ranking phase needs to be completed to be able to return the top-K. If you don't find the candidates(retrieval) and rank them, how would you know what is the top-K at all? Then, the l

Re: Indexing documents using ExtractHandler returning metadata

2023-02-08 Thread Sergio García Maroto
Found the solution. We have to captureAttr=true on solrconfix.xml and upfrefix=_ignored DocContentS false true ignored_ -MM-dd After on schema.xml. Add On Fri, 3 Feb 2023 at 16:50, Sergio García Maroto wrote: > Hi, > > I am indexing documents using tika and ExtractRequest handler.

Re: Query time

2023-02-08 Thread Tomás Fernández Löbbe
> Then, the long answer is that Apache Solr implements already approaches for > 'early termination' such as Block Max WAND from Solr 8(thanks Lucene for > this: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7148045/) to optimise > query time and 'skip un-worthy candidates'. > Note that this is not

Re: Query time

2023-02-08 Thread Mikhail Khludnev
I suppose if we sort=_docid_ asc it will break after collecting rows rows. Obviously, it disables all scoring, but perhaps you can pre-sort index. On Wed, Feb 8, 2023 at 7:36 PM Tomás Fernández Löbbe wrote: > > Then, the long answer is that Apache Solr implements already approaches > for > > 'ea

Re: NullPointerException when using facets in More Like This Query

2023-02-08 Thread Grace Sainsbury
Thanks for the quick fix! It seems to have resolved the issue. On Tue, 7 Feb 2023 at 16:29, Mikhail Khludnev wrote: > https://github.com/apache/solr/pull/1339 > Reviews are welcome. > > On Tue, Feb 7, 2023 at 10:54 PM Grace Sainsbury wrote: > > > I've created a ticket for this here: > > > > htt

MoreLikeThis highlighting question

2023-02-08 Thread Grace Sainsbury
Thanks again for fixing the NPE issue so quickly. I've noticed another issue, but I'm not sure if it's a configuration issue on my end. If I enable highlighting by providing the hl and hl.q parameters, I don't get any highlighting in the results. There is no highlighting section in the returned res

Re: MoreLikeThis highlighting question

2023-02-08 Thread Mikhail Khludnev
Hello, Grace. Now, MoreLikeThis handler carries the burden of invoking FacetComponet's internals, but it does nothing for highlighting, and there is no way to plug HighlightingComponent in it via solrconfig.xml. So, there are two options: someone makes hands dirty to plug HighlightingComponent into