On Thu, 2005-04-14 at 11:55 -0400, Serge Knystautas wrote:
> Anyway, I have no objection on principle to the pattern.  Just became 
> something of an antipattern in the SMTP handler.

I don't think it's necessarily a bad way to implement things.  Some of
the James SMTP stream wrapping is actually quite clever.  I think stream
wrapping is fine when working with socket streams (or any other stream
for that matter.)  The only problem is that the logic became difficult
for me to follow when porting the protocol handler to a non-stream based
system like Mina.

Although, as I go back to improve the Mina SMTP handler, I can see where
a lot of the logic in the streams could be moved to the decoder to make
the decoding process more robust.

-Mike


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

Reply via email to