You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lucene.apache.org by "Robert Muir (JIRA)" <ji...@apache.org> on 2016/03/22 05:44:25 UTC
[jira] [Updated] (LUCENE-7127) remove epsilon-based testing from
lucene/spatial
[ https://issues.apache.org/jira/browse/LUCENE-7127?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Robert Muir updated LUCENE-7127:
--------------------------------
Attachment: LUCENE-7127.patch
Here's my patch. Now GeoPoint and LatLonPoint are tested the exact same way (they are already generally returning exact same answers on any reasonable dataset, modulo bugs).
Tests generally pass but there are sporatic corner case failures: e.g.
{noformat}
[junit4] Started J0 PID(16923@localhost).
[junit4] Suite: org.apache.lucene.spatial.geopoint.search.TestLegacyGeoPointQuery
[junit4] 1> T1: id=1623 should not match but did
[junit4] 1> small=false query=GeoPointDistanceQuery: field=point: Center: [-78.95940156653523,-61.08293957076967] Distance: 5705904.858922883 meters] docID=1623
[junit4] 1> lat=-67.50789809972048 lon=95.3347154147923
[junit4] 1> deleted?=false
[junit4] 1> docID=1623 centerLon=-78.95940156653523 centerLat=-61.08293957076967 pointLon=95.3347154147923 pointLat=-67.50789809972048 distanceMeters=5715359.332616319 vs radiusMeters=5705904.858922883
[junit4] 2> ??? 22, 2016 7:36:32 ? com.carrotsearch.randomizedtesting.RandomizedRunner$QueueUncaughtExceptionsHandler uncaughtException
[junit4] 2> WARNING: Uncaught exception in thread: Thread[T1,5,TGRP-TestLegacyGeoPointQuery]
[junit4] 2> java.lang.AssertionError: wrong hit (first of possibly more)
[junit4] 2> at __randomizedtesting.SeedInfo.seed([2E99F05640D677C6]:0)
[junit4] 2> at org.junit.Assert.fail(Assert.java:93)
[junit4] 2> at org.apache.lucene.spatial.util.BaseGeoPointTestCase$VerifyHits.test(BaseGeoPointTestCase.java:553)
[junit4] 2> at org.apache.lucene.spatial.util.BaseGeoPointTestCase$2._run(BaseGeoPointTestCase.java:772)
[junit4] 2> at org.apache.lucene.spatial.util.BaseGeoPointTestCase$2.run(BaseGeoPointTestCase.java:637)
[junit4] 2>
[junit4] 2> NOTE: reproduce with: ant test -Dtestcase=TestLegacyGeoPointQuery -Dtests.method=testRandomWithThreads -Dtests.seed=2E99F05640D677C6 -Dtests.locale=ar-EG -Dtests.timezone=Antarctica/Syowa -Dtests.asserts=true -Dtests.file.encoding=US-ASCII
{noformat}
I will start to try to debug through some of these, see if i can get things happy.
> remove epsilon-based testing from lucene/spatial
> ------------------------------------------------
>
> Key: LUCENE-7127
> URL: https://issues.apache.org/jira/browse/LUCENE-7127
> Project: Lucene - Core
> Issue Type: Test
> Reporter: Robert Muir
> Attachments: LUCENE-7127.patch
>
>
> Currently, the random tests here allow a TOLERANCE and will fail if the error exceeds. But this is not fun to debug! It also keeps the door wide open for bugs to creep in.
> Alternatively, we can rework the tests like we did for sandbox/ points. This means the test is aware of the index-time quantization and so it can demand exact answers.
> Its more difficult at first, because even floating point error can cause a failure. It requires us to maybe work through corner cases/rework optimizations. If any epsilons must be added, they can be added to the optimizations themselves (e.g. bounding box) instead of the user's result.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org