alessandrobenedetti commented on PR #2489:
URL: https://github.com/apache/solr/pull/2489#issuecomment-2174445957
> I did the work of migrating to a QParser but don't have permissions to
push to your PR branch.
>
> Beware, I'm reverting most/all of the files that you modified of existing
Solr source files. If you would prefer I submit a separate PR, I'll do that.
Instead of `combiner=true`, you'd do `q={!combine}`.
I don't remember if this is the best way to invite you as a collaborator but
I just added you on our repo, hopefully that works!
I wasn't entirely sure a Query Parser is the right place for this, given we
are not really building a "new query" but then I thought : "hey we have the
boolean query parser, that more or less does the same", so I have nothing
against it!
You can go ahead and add your commit, I don't care much if you overwrote
many of my changes, I just want a nice and clean solution, however we come up
to it! Worst case scenario we go back in commits and re-use what's necessary!
>
> Wouldn't the docSet (assuming it's needed, like for faceting) be computed
on each subQuery, and if so how is that rolled up to the final QueryResult? I
see you tested that faceting works but I suspect there may be a very sad
performance bug here.
In regards to the docSet It was rolled up in:
`SortedIntDocSet docSet = new SortedIntDocSet(combinedResultsDocIds,
combinedResultsLength);
combinedRankedList.setDocSet(docSet);`
combinedResultsDocIds is the merged array of doc Ids.
I admit I haven't spent the required time on this part and it was quickly
done as a draft, so feel free to recommend a different way/correct me!
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
