[
https://issues.apache.org/jira/browse/LUCENE-8140?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16344971#comment-16344971
]
Karl Wright commented on LUCENE-8140:
-------------------------------------
Hi [~ivera], first of all if this passes all random tests I'm fine with
committing the code as-is. Please also commit to branch_6x and branch_7x when
you do.
Second, about recursion when there are a lot of edges -- beyond a few hundred
edges, the performance of GeoComplexPolygon far exceeds the performance of the
tiled polygon approach. That's why it was written.
> Checks for coplanarity when creating polygons shows numerical issues
> --------------------------------------------------------------------
>
> Key: LUCENE-8140
> URL: https://issues.apache.org/jira/browse/LUCENE-8140
> Project: Lucene - Core
> Issue Type: Bug
> Components: modules/spatial3d
> Reporter: Ignacio Vera
> Assignee: Karl Wright
> Priority: Major
> Attachments: LUCENE-8140-test.patch, LUCENE-8140.patch
>
>
> Coplanarity checks in GeoPolygonFactory shows numerical errors when the
> distance between two points is very small compared to the distance of the
> other two points. The situation is as follows:
> Having three points A, B, & C and the distance between A & B is very small
> compared to the distance between A & C, then the plane AC contains all points
> (co-planar) but the plane defined by AB does not contain C because of
> numerical issues. This situation makes some polygons fail to build.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
