[aosd-discuss] Is there a standard .Net Aspect software package?
eaddy at cs.columbia.edu
eaddy at cs.columbia.edu
Wed Jun 14 21:16:25 EST 2006
Hi Owen,
<ShamelessPlug>
My group at Columbia U. has been doing a lot of .NET AOP development
recently. We just released Wicca v2, a hybrid AOP system for .NET 2.0
that includes static and dynamic weavers.
The static bytecode weaver is called Phx.Morph and is built on top of the
Microsoft Phoenix compiler framework. Phx.Morph has limited functionality
but is fairly robust, although I wouldn't go so far as to say its suitable
for production.
The dynamic weaver leverages the Microsoft Debugger Edit-and-Continue and
Breakpoint APIs. It's really a proof-of-concept (read: very slow) and not
intended for production.
You can download source, regression tests, and binaries from:
http://www.cs.columbia.edu/~eaddy/wicca
</ShamelessPlug>
Cheers,
Marc Eaddy
PhD Student
Columbia University
> * Castle Project's Dynamic Proxy and nAspect are from what I know
> both under active development
>
>
>
> I went to Castle http://www.castleproject.org/index.php/AspectSharp
>
> Their Dynamic Proxy described itself as a general purpose proxy piece,
> and pointed to Aspect# as a customer of it. I went there - and it had
> become part of Castleproject also!
>
>
>
> I went to download it - and the latest release is from May 2005.
>
>
>
> Which brings me back to my point: "Seems to me like .Net Aspect
> development fell off a cliff about 2 years ago."
>
> 1 year in this case.
>
>
>
> owen
>
>
>
> ________________________________
>
> From: gregory young [mailto:gregoryyoung1 at gmail.com]
> Sent: Wednesday, June 14, 2006 1:18 PM
> To: Corpening, Owen
> Subject: Re: [aosd-discuss] Is there a standard .Net Aspect software
> package?
>
>
>
> Nearly all will work with 2.0 (the dynamic proxy based ones anyways
> which is about 90% of them). They will just not necesarily support
> generics in a full way.
>
>
>
> They operate on closed types so you will still be able to proxy closed
> types as you used to. They will not however allow you to proxy generic
> types (or to use generic arguments when defining point cuts)
>
>
>
> ex: instead of saying that you would like to attach an interceptor to
>
>
>
> List<int>.Add() where it would only apply to generic instances of List
> where T was an int
>
>
>
> you would have to instead use
>
>
>
> public class Foo : List<int> {
>
> }
>
>
>
> then apply to your closed class.
>
>
> Other than that rather large missing but of functionality most should
> work great.
>
>
>
> Castle Project's Dynamic Proxy and nAspect are from what I know both
> under active development. I know castle is working towards a version
> 2.0. I have been working on one with full generic support but there are
> alot of fringe conditions to think through especially dealing with
> mixins and generic proxy reuse.
>
>
>
> I have never has the courage to use EOS in a production environment so I
> can't really speak on it beyond the trivial examples I have used it for.
>
>
>
> Cheers,
>
>
>
> Greg Young
>
>
> On 6/14/06, Corpening, Owen <Owen.Corpening at amd.com> wrote:
>
> Seems to me like .Net Aspect development fell off a cliff about 2 years
> ago.
>
>
>
> Are any of the .Net tools still being updated, do any of them work well,
> is there a consensus similar to AspectJ being a consensus of sorts in
> java?
>
>
>
> I successfully used EOS over a year ago but I question whether any of
> the .Net packages work of each new release of visual studio, and are in
> any way supported.
>
>
>
> Open source? EOS was not ...
>
>
>
> thanks,
>
> Owen Corpening
>
> owen.corpening at amd.com
>
> 602-1000 x59613
>
>
>
>
> _______________________________________________
> discuss mailing list - discuss at aosd.net <mailto:discuss at aosd.net>
>
>
> To unsubscribe and change options, go to:
> http://aosd.net/mailman/listinfo/discuss_aosd.net
>
> Check out the AOSD.net Wiki: http://aosd.net/wiki
>
>
>
>
>
>
> --
> If knowledge can create problems, it is not through ignorance that we
> can solve them.
>
> Isaac Asimov
>
>
More information about the discuss
mailing list