[
https://issues.apache.org/jira/browse/LUCENE-6566?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15215973#comment-15215973
]
Robert Muir commented on LUCENE-6566:
-------------------------------------
Are you sure we should do this? In many systems, such as GeoJSON these polygons
become ambiguous. I don't think lucene should have heuristics here?
Instead we could spend our time speeding up polygon support, improving testing,
supporting multipolygon and polygon rings, etc, etc.
In https://datatracker.ietf.org/doc/draft-ietf-geojson/?include_text=1 it says:
{noformat}
In representing features that cross the antimeridian,
interoperability is improved by cutting geometries so that no single
part crosses the antimeridian.
...
A rectangle extending from 40 degrees N, 170 degrees E across the
antimeridian to 50 degrees N, 170 degrees W should be cut in two and
represented as a MultiPolygon.
{noformat}
> Handle "crosses dateline" cases in BKPointInPolygonQuery
> --------------------------------------------------------
>
> Key: LUCENE-6566
> URL: https://issues.apache.org/jira/browse/LUCENE-6566
> Project: Lucene - Core
> Issue Type: Bug
> Reporter: Michael McCandless
> Assignee: Michael McCandless
> Fix For: master, 6.1
>
>
> Just like LUCENE-6560, but we should also handle for the polygon case, which
> seems harder ...
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
