[
https://issues.apache.org/jira/browse/LUCENE-10378?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17488790#comment-17488790
]
Ignacio Vera edited comment on LUCENE-10378 at 2/8/22, 11:48 AM:
-----------------------------------------------------------------
Hi Gautam!
Thanks for the PR but you are right, I think the idea from Adrien is to avoid
visiting nodes when we know that it is inside our query. In that case you can
call directly PointTree#size to compute the number of points/documents to be
added to the final result.
I think you can start by implementing the following function in PointValues:
{code:java}
/**
* Finds the number of points matching the provided conditions. This method
does not enforce live
* documents, therefore it should only be used when there are no deleted
documents.
*/
public final long pointCount(BiFunction<byte[], byte[], Relation>
nodeComparator, Predicate<byte[]> leafComparator) throws IOException;{code}
This is the function you want to call when you know the number of points is
equal to the number of documents.
was (Author: ivera):
Hi Gautam!
Thanks for the PR but you are right, I think the idea from Adrien is to avoid
visiting nodes when we know that it is inside our query. In that case you can
call directly PointTree#size to compute the number of points/documents to be
added to the final result.
I think you can start by implementing the following function in PointValues:
{code:java}
/**
* Finds the number of points matching the provided conditions. This method
does not enforce live
* documents, therefore it should only be used when there are no deleted
documents.
*/
public final long pointCount(BiFunction<byte[], byte[], Relation>
nodeComparator, Predicate<byte[]> leafComparator) throws IOException;{code}
This is the function you want to call when you know the number of points is
equal to the number of values.
> Implement Weight#count on PointRangeQuery
> -----------------------------------------
>
> Key: LUCENE-10378
> URL: https://issues.apache.org/jira/browse/LUCENE-10378
> Project: Lucene - Core
> Issue Type: Task
> Reporter: Adrien Grand
> Priority: Minor
> Time Spent: 10m
> Remaining Estimate: 0h
>
> When there are no deletions and the field is single-valued (docCount == size)
> and has a single-dimension, PointRangeQuery could implement {{Weight#count}}
> by only counting matches on the two leaves that cross with the query.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org
For additional commands, e-mail: issues-h...@lucene.apache.org
