Makes sense to me... thanks for NOT staying out.

I don't like special case code any more than the next guy, but when I think
the only alternative is running everything through cat!?!?

Well that's annoying - and not that it would really be that special case
nessecarily - just generally speaking if you are writing to a socket stream
or character device don't lock cause you can't.

that is pretty general, and allows delivery to daemons (like spamd) without
having to call spamc first.

I think that would be "a good thing".

But not a special hack for "/dev/null" - you are right there.

Thoughts on this other idea?

m/

> -----Original Message-----
> From: Jon Nelson [mailto:[EMAIL PROTECTED]
> Sent: Wednesday, January 14, 2004 5:38 AM
> To: Mitch \(WebCob\)
> Cc: Jeff Jansen; Courier Users List
> Subject: RE: [courier-users] deleting messages reviewed...
>
>
> On Wed, 14 Jan 2004, Mitch \(WebCob\) wrote:
>
> > Right - I think I remember you mentioning that before...
> question for Sam I
> > guess is which way is the most efficient?
> >
> > Or is there another way we haven't thought of...
> >
> > I wonder if the flock can't be skipped in the case of a
> /dev/null delivery -
> > that would make this simple and "to /dev/null" would work...
>
> OK, I've been staying out of this discussion, but adding MORE
> special-case code to ANY program for something like this is just plain
> stupid in my opinion.  It *needlessly* complicates the code, and
> "special cases" are just that -- cases that differ from the norm.
>
> Please, if you want to throw away messages (and you are using maildrop),
> simply set EXITCODE to 0 and exit.  Done.  *No* more programs are
> invoked, *no* more I/O need occur, *no* more system resources consumed,
> *no* more stupid special-case in the code.
>
> Note: (exit)codes other than 0 may also be appropriate here.
> 0 works, I use it.
>
> --
> Ensign Walnut approaches Dr. Crusher with caution...
>
> Jon Nelson <[EMAIL PROTECTED]>
> C and Python Code Gardener
>



-------------------------------------------------------
This SF.net email is sponsored by: Perforce Software.
Perforce is the Fast Software Configuration Management System offering
advanced branching capabilities and atomic changes on 50+ platforms.
Free Eval! http://www.perforce.com/perforce/loadprog.html
_______________________________________________
courier-users mailing list
[EMAIL PROTECTED]
Unsubscribe: https://lists.sourceforge.net/lists/listinfo/courier-users

Reply via email to