[
https://issues.apache.org/jira/browse/LUCENE-1045?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12545567
]
Grant Ingersoll commented on LUCENE-1045:
-----------------------------------------
This patch seems a bit strange to me (sorry for getting to it so late). I
think the better way might be to have the FieldSortedHitQueue#comparatorAuto
method use the ExtendedFieldCache, instead of creating the dependency in
FieldCacheImpl on ExtendedFieldCache to have long support.
Then, you could just override getAuto in ExtendedFieldCache.
This way, those who have longs and want them treated as such get them from the
ExtendedFC, otherwise, those just using FieldCacheImpl, get the behavior they
are used to, namely no long support.
> SortField.AUTO doesn't work with long
> -------------------------------------
>
> Key: LUCENE-1045
> URL: https://issues.apache.org/jira/browse/LUCENE-1045
> Project: Lucene - Java
> Issue Type: Bug
> Components: Search
> Affects Versions: 2.2
> Reporter: Daniel Naber
> Priority: Minor
> Fix For: 2.3
>
> Attachments: auto-long-sorting.diff, TestDateSort.java
>
>
> This is actually the same as LUCENE-463 but I cannot find a way to re-open
> that issue. I'm attaching a test case by dragon-fly999 at hotmail com that
> shows the problem and a patch that seems to fix it.
> The problem is that a long (as used for dates) cannot be parsed as an
> integer, and the next step is then to parse it as a float, which works but
> which is not correct. With the patch the following parsers are used in this
> order: int, long, float.
--
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: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
