harmony-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Tim Ellison (JIRA)" <j...@apache.org>
Subject [jira] Updated: (HARMONY-4679) [startup] java.net.URL implementation without regex classes loading
Date Tue, 28 Aug 2007 10:25:30 GMT

     [ https://issues.apache.org/jira/browse/HARMONY-4679?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Tim Ellison updated HARMONY-4679:
---------------------------------

    Attachment: harmony-4679.patch

Thanks for the patch Naumova, though I admit I found it a bit hard to follow the logic.  Based
on your suggestion I implemented your suggestion as attached in harmony-4679.patch, please
take a look and let me know if that achieves the same effect.  If so I'll commit it.

Reading a bit further down, if we have checked the protocol adheres to that given pattern,
I wonder why we then check whether "protocol.indexOf('/') >= 0"  surely this can never
be true?

Tim


> [startup] java.net.URL implementation without regex classes loading
> -------------------------------------------------------------------
>
>                 Key: HARMONY-4679
>                 URL: https://issues.apache.org/jira/browse/HARMONY-4679
>             Project: Harmony
>          Issue Type: Improvement
>          Components: Classlib
>            Reporter: Naumova Natalya 
>            Assignee: Tim Ellison
>         Attachments: harmony-4679.patch, java.net.URL.without.regex.patch
>
>
> When any class loads java.net.URL is used. However it loads ~60 classes from regex.jar.
We can avoid this to re-write using of java.lang.String.matches method in net.URL, by using
indexOf and substring instead of. Then no regex classes are loaded during the startup. Speedup
on HelloWorld is ~7%.

-- 
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