lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Robert Muir (JIRA)" <>
Subject [jira] [Commented] (LUCENE-5499) EarlyTerminatingSortingCollector shouldnt require exact Sort match
Date Fri, 07 Mar 2014 13:24:43 GMT


Robert Muir commented on LUCENE-5499:

The basics are: right now we just encode Sort.toString() in the index. But a Sort is just
a collection of SortFields. So if we encode it differently (e.g. each SortField.toString()
separated by INFORMATION_SEPARATOR_ONE, escaping the former in case someone is crazy...) we
can easily have logic like this.

> EarlyTerminatingSortingCollector shouldnt require exact Sort match
> ------------------------------------------------------------------
>                 Key: LUCENE-5499
>                 URL:
>             Project: Lucene - Core
>          Issue Type: Improvement
>            Reporter: Robert Muir
> Today EarlyTerminatingSortingCollector requires that the Sort match exactly at query
and at index time.
> However, now that you can use any Sort (e.g. with multiple sortfields), this should be
> For example, early termination is fine in the following case:
> * index-time: popularity desc, time desc
> * query-time: popularity desc

This message was sent by Atlassian JIRA

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message