Yes, this is ok for me.  I'm more than willing to submit the code to the list 
and let someone else work
out the commit details.  I'll have to write test cases as well, as my current 
test cases for this are executed
manually.  That being said, i make regular use of the task, so that is some 
measure of comfort with it's
stability :)

I can very easily draw up documentation, though I'm unfamiliar with what format 
will be easiest to integrate
with the current ivy build (no sense i me writing document in a format where 
someone will have to convert
it to xdocs or something else).  Any help on that would be appreciated.

Thanks.
 
----
[EMAIL PROTECTED]
"Once you start down the dark path, forever will it
dominate your destiny.  Consume you it will " - Yoda

----- Original Message ----
From: Xavier Hanin <[EMAIL PROTECTED]>
To: [email protected]
Sent: Monday, October 15, 2007 1:20:43 PM
Subject: Re: Ivy task to import common build tasks


On 10/15/07, Matthew Inger <[EMAIL PROTECTED]> wrote:
>
> As part of the ant-contrib project, I had written a task which,
 underneath
> the covers,
> used the ivy library to resolve a jar file, extract it locally in the
 ivy
> cache, and import an
> ant build script.
>
> <ac:importurl org="org" name="name" rev="latest.release" />
>
> The following actions would be performed:
>    1.  Resolve the requested module using ivy's inline resolution
> capabilities
>    2.  Expand the resolved .jar/.zip file to the ivy cache
>    3.  Import the "build.xml" contained in the jar file (the imported
> filename is
>           overridable via the "resource" attribute on the task).
>
> In theory, it's a wrapper for 3 operations:
>
>    <ivy:resolve />
>    <unjar />
>    <import />
>
> This gave me the capability to build re-usable build components which
> could
> be shared amongst different projects.  The components live in the ivy
> repository.
> As a result, it became very easy for me to have one common set of
 build
> tasks,
> which could be updated across the board for all projects simply by
> publishing a
> version of the dependency.
>
> I realize that this is typically accomplished using an ant <import>
> task.  However, that
> has several drawback, which i won't go into here, unless someone
 really
> wants me to.
>
> With the pending release of Ivy 2.0, this task will no longer work
 without
> code changes
> to account for the changes to the API from 1.4.1 to 2.0.
>
> I'm wondering if this task might be better suited to being owned by
 the
> Ivy project itself.
> I am more than willing to donate the code to the Ivy project.  As the
> ant-contrib project
> is published under the Apache 2.0 license, i don't see any issues
 with
> this.
>
> Any thoughts from the dev team?


I like the idea behind this task, and think it could fit in the picture
 of
Ivy. So I think moving this to the Ivy project can be nice. We have to
 think
about several things though:
- since you are not a committer on Ivy (yet), you will have to submit
patches if you later want to improve the task. Is this ok for you?
- do you have documentation and unit test for this task? Documentation
 is
mandatory, and unit test are highly appreciated :-)

Thoughts?

Xavier

----
> [EMAIL PROTECTED]
> "Once you start down the dark path, forever will it
> dominate your destiny.  Consume you it will " - Yoda
>
>
>
>
>
>
>
 
____________________________________________________________________________________
> Need a vacation? Get great deals
> to amazing places on Yahoo! Travel.
> http://travel.yahoo.com/
>



-- 
Xavier Hanin - Independent Java Consultant
http://xhab.blogspot.com/
http://incubator.apache.org/ivy/
http://www.xoocode.org/





      
____________________________________________________________________________________
Check out the hottest 2008 models today at Yahoo! Autos.
http://autos.yahoo.com/new_cars.html

Reply via email to