[jira] [Commented] (LUCENE-7045) remove all encode/decode hooks from PointRangeQuery

["Michael McCandless (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/LUCENE-7045?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15159452#comment-15159452 ] 

Michael McCandless commented on LUCENE-7045:
--------------------------------------------

+1, this is really nice.  I like having all encode/decode for a given data type in once place, succinctly exposed, and having that same place create the queries.

> remove all encode/decode hooks from PointRangeQuery
> ---------------------------------------------------
>
>                 Key: LUCENE-7045
>                 URL: https://issues.apache.org/jira/browse/LUCENE-7045
>             Project: Lucene - Core
>          Issue Type: Improvement
>            Reporter: Robert Muir
>         Attachments: LUCENE-7045.patch
>
>
> In LUCENE-7043 we added several new point types and just gave them static methods to generate exact/range/prefix/whatever queries.
> I think we should do the same in general, e.g. for IntPoint:
> {code}
> This field defines static factory methods for creating common queries:
> <ul>
>   <li>{@link #newExactQuery newExactQuery()} for matching an exact 1D point.
>   <li>{@link #newRangeQuery newRangeQuery()} for matching a 1D range.
>   <li>{@link #newMultiRangeQuery newMultiRangeQuery()} for matching points/ranges in n-dimensional space.
> </ul>
> {code}
> Then each Point can have types that make sense for it: e.g. multi-dimensional range queries don't make sense at all for IP addresses, but prefix query does, and polygon/distance queries make sense for LatLonPoint and so on (that one is not refactored yet here, followup!)



--
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