[
https://issues.apache.org/jira/browse/LUCENE-7073?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15183287#comment-15183287
]
Robert Muir commented on LUCENE-7073:
-------------------------------------
I will defer the tokenized issue, it impacts all fieldtypes. And its easy
enough for code to do indexed && tokenized to work around it.
But the ctor, hashcode, equals, and the test need to be fixed, I will make a
patch.
> FieldType problems with Points
> ------------------------------
>
> Key: LUCENE-7073
> URL: https://issues.apache.org/jira/browse/LUCENE-7073
> Project: Lucene - Core
> Issue Type: Bug
> Reporter: Robert Muir
>
> The "copy constructor" is buggy and ignores the incoming points properties,
> because of a bug:
> {code}
> public FieldType(FieldType ref) {
> ...
> // these two assignments have no effect.
> this.dimensionCount = dimensionCount;
> this.dimensionNumBytes = dimensionNumBytes;
> {code}
> Another issue is that tokenized() returns true for all these point fields,
> which is bogus, and it confuses things like transitioning to points in some
> benchmark code (trying to clean up places still using LegacyNumeric* !!!).
> This code checks if (tokenized()) { try to use tokenstream }
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
