[
https://issues.apache.org/jira/browse/LUCENE-1777?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12738249#action_12738249
]
Michael McCandless commented on LUCENE-1777:
--------------------------------------------
Indeed, this is the exception I see:
{code}
[junit] java.lang.NumberFormatException: Infinite or NaN
[junit] at java.math.BigDecimal.<init>(BigDecimal.java:716)
[junit] at
org.apache.lucene.spatial.tier.CartesianPolyFilterBuilder.getBoxShape(CartesianPolyFilterBuilder.java:81)
[junit] at
org.apache.lucene.spatial.tier.CartesianPolyFilterBuilder.getBoundingArea(CartesianPolyFilterBuilder.java:109)
[junit] at
org.apache.lucene.spatial.tier.DistanceQueryBuilder.<init>(DistanceQueryBuilder.java:58)
[junit] at
org.apache.lucene.spatial.tier.TestCartesian.testRange(TestCartesian.java:152)
{code}
It's happening because the distance (miles) param to CartesianPolyFilterBuilder
is an int, and 0.5 rounds down to 0. Putting a floor at 1 prevents the
exception (resulting in a floor on the bounding area that can be filtered0,
though, I'm not sure why we can't in general have finer resolution than 1 mile
(ie why mile can't be switched to a double instead).
> Error on distance query where miles < 1.0
> -----------------------------------------
>
> Key: LUCENE-1777
> URL: https://issues.apache.org/jira/browse/LUCENE-1777
> Project: Lucene - Java
> Issue Type: Bug
> Components: contrib/spatial
> Affects Versions: 2.9
> Reporter: Glen Stampoultzis
> Attachments: LUCENE-1777.patch
>
>
> If miles is under 1.0 distance query will break.
> To reproduce modify the file
> http://svn.apache.org/viewvc/lucene/java/trunk/contrib/spatial/src/test/org/apache/lucene/spatial/tier/TestCartesian.java?revision=794721
> And set the line:
> final double miles = 6.0;
> to
> final double miles = 0.5;
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
---------------------------------------------------------------------
To unsubscribe, e-mail: java-dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: java-dev-h...@lucene.apache.org
