Hi Noel,

Ahhh now I am getting some attention... I mean feedback :-)

> I'm not at all in favor of the particular direction being taken for this
> bounce handler, but I still have more reading to do.  It seems to me that
it
> is attempting to build a pipeline within in a mailet, rather than take
> advantage of matchers, mailets and the processor pipeline.  My current
> belief, which still begs more reading, is that there will be good code,
but
> that the code related to his complex configuration is going to be wasted.

OK, So I did point out (and so did Danny) in my previous posts that what I
was doing was very similar in concept to the mailet/matcher/processor stuff
we already have in James, and begged for feedback.

However there are several things I just can 't do with the current Mailet
API as it stands today (I know that you guys have been asking for feedback,
unfortunately I needed to start coding a Mailet to provide this <g>).
I think Danny pointed out several additions to v3 Mailets which would help
solve many of the requirements I need for this Mailet.

Now I have the option to stop working on my bounce handler and maybe try and
help out with Mailet v3 coding, or try and finnish a basic prototype and
submit the code so that it could be used as a partial "requirements list".
Once coding of v3 Mailets is underway I can start migrating the Mailet code
to the new API.

I also agree that what I am doing now config wise will be wasted, but thats
not a big deal. Its not that much code.

Thanks,
Sergei





--
To unsubscribe, e-mail:   <mailto:[EMAIL PROTECTED]>
For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>

Reply via email to