[
https://issues.apache.org/jira/browse/LUCENE-7290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15289940#comment-15289940
]
Karl Wright commented on LUCENE-7290:
-------------------------------------
Failure analysis:
{code}
[junit4] 1> root cell: cell=43116 x: -2 TO 2013125215, y: -2 TO 1, z:
745164843 TO 2140283542, splits: 0
[junit4] 1> cycle: cell=43116 x: -2 TO 2013125215, y: -2 TO 1, z:
745164843 TO 2140283542, splits: 0 queue.size()=0
[junit4] 1> minx=-9.32364588596373E-10 maxx=0.9384833319044121
miny=-9.32364588596373E-10 maxy=9.323645885963728E-10 minz=0.34738265614008795
maxz=0.9977622925243912
[junit4] 1> GeoArea.OVERLAPS: keep splitting
[junit4] 1> now split on x=283727372
[junit4] 1> split cell1: cell=43117 parentCellID=43116 x: -2 TO
283727372, y: -2 TO 1, z: 745164843 TO 2140283542, splits: 1
[junit4] 1> split cell2: cell=43118 parentCellID=43116 x: 283727372 TO
2013125215, y: -2 TO 1, z: 745164843 TO 2140283542, splits: 1
{code}
{code}
[junit4] 1> doc=3174 should match but did not
[junit4] 1> point=[X=-0.01741337087280333, Y=-2.3309114714909325E-10,
Z=0.9976113452615002]
{code}
Looks like doc=3174 is not within the root cell.
X=-0.01741337087280333
but
minx=-9.32364588596373E-10 maxx=0.9384833319044121
So, maybe bounds are wrong? But I can't see how this point should be
considered within the shape at all... The shape:
{code}
[junit4] > Throwable #1: java.lang.AssertionError: invalid hits for
shape=GeoNorthRectangle: {planetmodel=PlanetModel.WGS84,
bottomlat=0.35451471030934045(20.312196676028226),
leftlon=9.908337057950734E-15(5.677058954136481E-13),
rightlon=2.891004593509811E-11(1.6564236176104631E-9)}
{code}
This is an extremely thin slice from the north pole down to latitude 20.312.
Both left longitude and right longitude are positive, though small --
essentially almost 0.0. This would lead to points being in-set only if they
have positive X. How, then, can it be thinking that any point with
X=-0.01741337087280333 can be in-set?
The only answer I have is that this slice is just *barely* non-degenerate. The
point in question must be within the MINIMUM_RESOLUTION distance of both
planes, even though it is clearly in negative territory. Gotta think about the
right solution.
> geo3d test failure
> ------------------
>
> Key: LUCENE-7290
> URL: https://issues.apache.org/jira/browse/LUCENE-7290
> Project: Lucene - Core
> Issue Type: Bug
> Components: modules/spatial3d
> Reporter: Karl Wright
> Assignee: Karl Wright
>
> {code}
> Error Message:
> invalid hits for shape=GeoNorthRectangle: {planetmodel=PlanetModel.WGS84,
> bottomlat=0.35451471030934045(20.312196676028226),
> leftlon=9.908337057950734E-15(5.677058954136481E-13),
> rightlon=2.891004593509811E-11(1.6564236176104631E-9)}
> Stack Trace:
> java.lang.AssertionError: invalid hits for shape=GeoNorthRectangle:
> {planetmodel=PlanetModel.WGS84,
> bottomlat=0.35451471030934045(20.312196676028226),
> leftlon=9.908337057950734E-15(5.677058954136481E-13),
> rightlon=2.891004593509811E-11(1.6564236176104631E-9)}
> at
> __randomizedtesting.SeedInfo.seed([1D5C0D197B0122B4:AD23708DF44C8C28]:0)
> at org.junit.Assert.fail(Assert.java:93)
> at
> org.apache.lucene.spatial3d.TestGeo3DPoint.testGeo3DRelations(TestGeo3DPoint.java:440)
> at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
> at
> sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
> at
> sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
> at java.lang.reflect.Method.invoke(Method.java:498)
> at
> com.carrotsearch.randomizedtesting.RandomizedRunner.invoke(RandomizedRunner.java:1764)
> ...
> {code}
> and
> {code}
> Build Log:
> [...truncated 8902 lines...]
> [junit4] Suite: org.apache.lucene.spatial3d.TestGeo3DPoint
> [junit4] 1> doc=3174 should match but did not
> [junit4] 1> point=[X=-0.01741337087280333, Y=-2.3309114714909325E-10,
> Z=0.9976113452615002]
> [junit4] 2> NOTE: reproduce with: ant test -Dtestcase=TestGeo3DPoint
> -Dtests.method=testGeo3DRelations -Dtests.seed=1D5C0D197B0122B4
> -Dtests.multiplier=3 -Dtests.slow=true -Dtests.locale=lt
> -Dtests.timezone=Europe/Kiev -Dtests.asserts=true -Dtests.file.encoding=UTF-8
> [junit4] FAILURE 7.13s J1 | TestGeo3DPoint.testGeo3DRelations <<<
> [junit4] > Throwable #1: java.lang.AssertionError: invalid hits for
> shape=GeoNorthRectangle: {planetmodel=PlanetModel.WGS84,
> bottomlat=0.35451471030934045(20.312196676028226),
> leftlon=9.908337057950734E-15(5.677058954136481E-13),
> rightlon=2.891004593509811E-11(1.6564236176104631E-9)}
> [junit4] > at
> __randomizedtesting.SeedInfo.seed([1D5C0D197B0122B4:AD23708DF44C8C28]:0)
> [junit4] > at
> org.apache.lucene.spatial3d.TestGeo3DPoint.testGeo3DRelations(TestGeo3DPoint.java:440)
> [junit4] > at java.lang.Thread.run(Thread.java:745)
> [junit4] IGNOR/A 0.01s J1 | TestGeo3DPoint.testRandomBig
> [junit4] > Assumption #1: 'nightly' test group is disabled (@Nightly())
> [junit4] 2> NOTE: test params are: codec=Lucene62,
> sim=RandomSimilarity(queryNorm=false,coord=crazy): {}, locale=lt,
> timezone=Europe/Kiev
> [junit4] 2> NOTE: Linux 4.4.0-21-generic i386/Oracle Corporation
> 1.8.0_92 (32-bit)/cpus=12,threads=1,free=43619256,total=64946176
> [junit4] 2> NOTE: All tests run in this JVM: [TestGeo3DPoint]
> [junit4] Completed [10/11 (1!)] on J1 in 74.75s, 13 tests, 1 failure, 1
> skipped <<< FAILURES!
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
