myfaces-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Mike Kienenberger (JIRA)" <...@myfaces.apache.org>
Subject [jira] Commented: (MYFACES-2889) [PERF] Remove String.intern() calls in FlashELResolver and ImplicitObjectResolver
Date Wed, 25 Aug 2010 19:04:15 GMT

    [ https://issues.apache.org/jira/browse/MYFACES-2889?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12902582#action_12902582
] 

Mike Kienenberger commented on MYFACES-2889:
--------------------------------------------

I am not commenting on whether intern should be used in this situation, as I don't know.

However, for maps, if there are several items in the map which have the same hashcode, then
those individual items will be compared using equality operations.

No idea how often that would happen in this particular case.


> [PERF] Remove String.intern() calls in FlashELResolver and ImplicitObjectResolver
> ---------------------------------------------------------------------------------
>
>                 Key: MYFACES-2889
>                 URL: https://issues.apache.org/jira/browse/MYFACES-2889
>             Project: MyFaces Core
>          Issue Type: Improvement
>          Components: JSR-314
>    Affects Versions: 2.0.2-SNAPSHOT
>         Environment: JBoss AS 6 M4, Parleys.com JSF2 app
>            Reporter: Jan-Kees van Andel
>
> I've been doing some profiling and I see pretty much activity in FlashELResolver.castAndIntern()
and ImplicitObjectResolver.castAndIntern().
> When I replace the "return s.intern()" lines by  "return s", both methods have (of course)
much better performance.
> But I'm pretty sure someone put them there with a reason, like memory footprint.
> However, I don't see any difference in memory footprint.
> Any ideas? Do we want to keep the intern() calls?

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message