[
https://issues.apache.org/jira/browse/LUCENE-6487?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14553274#comment-14553274
]
Karl Wright edited comment on LUCENE-6487 at 5/20/15 11:11 PM:
---------------------------------------------------------------
Note that the general ellipsoid treatment will be some percentage slower than
the prior sphere treatment. The critical parts (e.g. shape membership) are
unchanged, but the following have taken a small hit:
- initial setup of objects, e.g. computing points from lat/lon
- computing the bounds of shapes
- computing the arc distance between two points
I have not added a "surface distance" abstraction at this time, in part because
there is no accurate ellipsoidal calculation I can find, only approximations.
So distance measures continue to be computed as arc distances, normal
distances, or direct (linear) distances. Should an accurate surface
calculation become available this can readily be added.
(Further research shows that computing elliptical curve distances are in fact
part of a class of problem under the name of "elliptic integrals", which has
only numeric solutions. So it is unlikely that we will be implementing this in
geo3d soon.)
was (Author: kwri...@metacarta.com):
Note that the general ellipsoid treatment will be some percentage slower than
the prior sphere treatment. The critical parts (e.g. shape membership) are
unchanged, but the following have taken a small hit:
- initial setup of objects, e.g. computing points from lat/lon
- computing the bounds of shapes
- computing the arc distance between two points
I have not added a "surface distance" abstraction at this time, in part because
there is no accurate ellipsoidal calculation I can find, only approximations.
So distance measures continue to be computed as arc distances, normal
distances, or direct (linear) distances. Should an accurate surface
calculation become available this can readily be added.
> Add WGS84 capability to geo3d support
> -------------------------------------
>
> Key: LUCENE-6487
> URL: https://issues.apache.org/jira/browse/LUCENE-6487
> Project: Lucene - Core
> Issue Type: Improvement
> Components: modules/spatial
> Reporter: Karl Wright
> Attachments: LUCENE-6487.patch
>
>
> WGS84 compatibility has been requested for geo3d. This involves working with
> an ellipsoid rather than a unit sphere. The general formula for an ellipsoid
> is:
> x^2/a^2 + y^2/b^2 + z^2/c^2 = 1
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
