Re: Balsa + libESMTP patch

[Pawel Salek <pawsa TheoChem kth se>]

(quest a quick reply).

On 2001-05-02 17:21 Carlos Morgado wrote:
> > The SMTP client in Balsa is buggy to the extent its unusable except
> > in a small set of circumstances.  Aside from that it cannot easily
> 
> i must be one of the lucky few. 

present SMTP code in balsa works for me most of the time but of course
it should be replaced by something better in the long term. There is no
question that the internet is more hostile that few years ago and
supporting for example AUTH is a must, IMO.

> no. i'm saying balsa shouldn't make such a change overnight. i'm
> saying there 
> should be a 1.1 release with default --with-libesmtp=yes and then on
> the next 
> libmutt review phase out the old code. 
> i'm saying maybe there should be a libesmtp branch. pawel ?

This is rather major change, so I think we should branch for it, release
one version with it and ask for response. this has to be done smoothly
because branches tend to get rotten and unmaintained in the long term.
Releasing it to wider audience is crucial because people have sometimes
totally different setups (different Linux distributions, AIX, Thru64,
Sun etc). Brian, on which boxes have you tested your code so far?

> > | i am yet to see a mta listening on port 587. setting the default
> to 
> > | something that doesn't work in the default case is not very good.
> > 
> > We have to agree to disagree on this one.  The justification in
> > RFC 2476 for the use of port 587 for mail submission is good enough
> > for me.
> > 
> 
> the justification that *everyone* will have to correct the default
> setting is good for me.

This may be a problem because of one more reason: the configuration file
format changes. But I guess we can modify the initialization wizard to
suggest using port 25 and giving some comments on MSA port 587. Also, we
can alter save-restore.c to read Port entry and append read value to the
host string. Frankly, I have seen just one host listening to port 587
(but if we are not on the cutting edge, who will be?)

/Pawel
-- 
Pawel Salek (pawsa@theochem.kth.se) http://www.theochem.kth.se/~pawsa/
Theoretical Chemistry Division, KTH voice: +46 8 790-8202