[
https://issues.apache.org/jira/browse/LUCENE-6780?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14740949#comment-14740949
]
Michael McCandless commented on LUCENE-6780:
--------------------------------------------
OK I committed a new randomized test {{TestGeoUtils.testRandomRectsAndCircles}}
...
It's a simple test: make a random rect and circle, use the GeoUtils API to see
if they are disjoint or if the circle fully contains the rect (else, skip it),
and then randomly pick points inside the rect and confirm they are either
within or not within the circle.
It sometimes fails, e.g.:
{noformat}
ant test -Dtestcase=TestGeoUtils -Dtests.method=testRandomRectsAndCircles
-Dtests.seed=D4C2F5423D90BDD8 -Dtests.slow=true
-Dtests.linedocsfile=/lucenedata/hudson.enwiki.random.lines.txt.fixed
-Dtests.locale=mt_MT -Dtests.timezone=Australia/Yancowinna -Dtests.asserts=true
-Dtests.file.encoding=UTF-8 -Dtests.verbose=true
{noformat}
But it's entirely possible we need to relax the test for boundary conditions,
however when I look at the GeoUtils impls, we seem to consistently use
haversine for all the checks ... but maybe even so the math can produce false
errors?
> GeoPointDistanceQuery doesn't work with a large radius?
> -------------------------------------------------------
>
> Key: LUCENE-6780
> URL: https://issues.apache.org/jira/browse/LUCENE-6780
> Project: Lucene - Core
> Issue Type: Bug
> Reporter: Michael McCandless
> Attachments: LUCENE-6780.patch, LUCENE-6780.patch, LUCENE-6780.patch,
> LUCENE-6780.patch
>
>
> I'm working on LUCENE-6698 but struggling with test failures ...
> Then I noticed that TestGeoPointQuery's test never tests on large distances,
> so I modified the test to sometimes do so (like TestBKDTree) and hit test
> failures.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
