Return-Path: <jakarta-ant@ehatchersolutions.com>
Mailing-List: contact dev-help@ant.apache.org; run by ezmlm
Delivered-To: mailing list dev@ant.apache.org
Received: (qmail 12965 invoked from network); 22 Apr 2003 00:49:54 -0000
Received: from h002.c000.snv.cp.net (HELO c000.snv.cp.net) (209.228.32.66)
  by daedalus.apache.org with SMTP; 22 Apr 2003 00:49:54 -0000
Received: (cpmta 679 invoked from network); 21 Apr 2003 17:50:02 -0700
Received: from 68.169.37.64 (HELO ehatchersolutions.com)
  by smtp.hatcher.net (209.228.32.66) with SMTP; 21 Apr 2003 17:50:02 -0700
X-Sent: 22 Apr 2003 00:50:02 GMT
Date: Mon, 21 Apr 2003 20:50:00 -0400
Subject: Re: antlib
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v552)
Cc: Stuart Halloway <stu@develop.com>
To: "Ant Developers List" <dev@ant.apache.org>
From: Erik Hatcher <jakarta-ant@ehatchersolutions.com>
In-Reply-To: <006d01c3084c$50993360$030200c0@DJ0X820J>
Message-Id: <58D3B1C0-745C-11D7-8A73-000393A564E6@ehatchersolutions.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.552)
X-Spam-Rating: daedalus.apache.org 1.6.2 0/1000/N

On Monday, April 21, 2003, at 05:23  PM, Antoine Levy-Lambert wrote:
> I am trying to update the antlib proposal (proposal/sandbox/antlib).

Cool stuff... glad you are on board!  :)

> There are the following features in the antlib proposal:
>
> 1) antlib & antjar
> ------------------------
>
> deployment descriptor called antlib.xml which would go in the META-INF 
> subdirectory of the antlib
>
> looks like that
> <antlib version="1.5" >
>   <task name="mkdir" class="org.apache.tools.ant.taskdefs.Mkdir"/>
>   <task name="javac" class="org.apache.tools.ant.taskdefs.Javac"/>
>
>   <data-type name="description" 
> class="org.apache.tools.ant.types.Description"/>
>   <data-type name="dirset" class="org.apache.tools.ant.types.DirSet"/>
> ....
>
> Note that this mechanism would also be used for the core of ant and 
> for optional.jar.
> So defaults.properties would die. I believe that xml is superior to 
> properties files, and the fact of using the same technique for ant 
> core and for the antlibs [ to find out which tasks and which datatypes 
> are defined ] is good.

Also keep in mind that we can use XDoclet to generate these 
descriptors, if that seems like the right thing to do.  I'd be happy to 
help out with that generation.

> 4) A framework for managing classloaders where you can specify which 
> classloader to use when loading an antlib.

We ought to get Stuart Halloway involved in this effort.  He's got a 
lot of experience and expertise with classloaders and has some specific 
thoughts on what we can do with Ant to get some more classloader 
sanity, I think.  I'm CC'ing him on this mail in case he's not 
monitoring the dev list.

	Erik