You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@uima.apache.org by "Marshall Schor (JIRA)" <de...@uima.apache.org> on 2017/08/14 14:31:00 UTC

[jira] [Resolved] (UIMA-5533) improve sorted index insert and faster findEq when lots of equal items

     [ https://issues.apache.org/jira/browse/UIMA-5533?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Marshall Schor resolved UIMA-5533.
----------------------------------
    Resolution: Fixed

> improve sorted index insert and faster findEq when lots of equal items
> ----------------------------------------------------------------------
>
>                 Key: UIMA-5533
>                 URL: https://issues.apache.org/jira/browse/UIMA-5533
>             Project: UIMA
>          Issue Type: Improvement
>          Components: Core Java Framework
>    Affects Versions: 2.10.0SDK
>            Reporter: Marshall Schor
>            Assignee: Marshall Schor
>            Priority: Minor
>             Fix For: 2.10.1SDK
>
>
> Use the same technique used in UIMA v3 for sorted indexes to sort by their comparator, and then if equal, to sort by their FS hashcode (== id).  This fixes a performance problem when there's lots of comparator-equal (but not identical) FSs added to a sorted index.  It also makes the iteration order consistent with UIMA V3, enabling verification/compares of the two versions.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)