[
https://issues.apache.org/jira/browse/LUCENE-7168?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15223348#comment-15223348
]
Karl Wright commented on LUCENE-7168:
-------------------------------------
While XYZSolids are used by Lucene for BKD trees, I cannot say whether or not
they would never be used by anyone else. The functionality is general, after
all. In fact, the functionality was introduced because we had issues without
it from the BKD tree implementation. Perhaps those problems are fixed but
whether degeneracy support *for that use case* is needed still depends strongly
on the relative values of the "known" minimum 32-bit resolution of the tree and
the actual value of Vector.MINIMUM_RESOLUTION. So at the moment I'm not even
convinced that the BKD implementation doesn't need degeneracy support.
I already made all the variants of XYZSolids package-private and
lucene.internal. Is this enough to hide this complication from users?
> Remove geo3d test leniency
> --------------------------
>
> Key: LUCENE-7168
> URL: https://issues.apache.org/jira/browse/LUCENE-7168
> Project: Lucene - Core
> Issue Type: Bug
> Reporter: Michael McCandless
>
> Today the test hides possible failures by leniently handling quantization
> issues.
> We should fix it to do what geo2d tests now do: pre-quantized indexed points,
> but don't quantize query shapes.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
