[
https://issues.apache.org/jira/browse/LANG-510?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12752237#action_12752237 ]
Henri Yandell commented on LANG-510:
------------------------------------
Said 'merely' walking method needs to match JVM speed though before we'd want the rest of the codebase to be dependent on it. Which I doubt we'd want to do.
Here is the Harmony String.java implementation:
http://svn.apache.org/repos/asf/harmony/enhanced/classlib/trunk/modules/luni/src/main/java/java/lang/String.javaIt uses both direct private variable access on the subString and it works at the array level without having to incur charAt method calls.
Various naive implementations of unoptimized that I just hacked together only reach the 3x speed (and no idea if they're actually accurate). So I think that's it for the idea of moving the parts of StringUtils that rely on indexOf over to CharSequence.
> Convert StringUtils API to take CharSequence
> --------------------------------------------
>
> Key: LANG-510
> URL:
https://issues.apache.org/jira/browse/LANG-510> Project: Commons Lang
> Issue Type: Improvement
> Reporter: Henri Yandell
> Fix For: 3.0
>
>
> Wherever possible, use CharSequence and not String in the StringUtils API.
> substring can go to subSequence inside code. The general substring method can be replaced with a subSequence method and substring deprecated(?). One question is whether to implement a CharSequence.indexOf type method. Given that it's merely a walking method(?!?), this might be quite handy and would allow a bunch of StringUtils methods to move over.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
