Hello list,
Tarqi Kazan, who has found the bug with `-L', asked for SMTP
GSS-API authentication and had the will to test and a working
Kerberos installation (my hero). So that was collaboration that
wouldn't have been possible without the internet! (Tarqi lives
in Düsseldorf (he said Ddorf).)
Thank you very much, Tarqi!
(Unfortunately Tarqi has vanished his mail account (tarqi AT cfs
DOT or GS), which is why he's not in Cc:.)
Also on [next] there is:
- The possibility to use PgUp / PgDown / Home / End instead of
z[-+0$] on xterm-compatible terminals
- Support for empty lines in macro and account definition blocks
(as required by POSIX for startup files in general)
- ^C in compose mode with *ignore* set acts now POSIX compatible
- If *encoding* is set to base64 then we don't ignore that user
wish and use quoted-printable (when 7bit doesn't suffice)
- We don't (temporarily) set the $LESS environment variable
anymore (even if that was not set)
- `resend' should be truly fixed regardless of what the
configuration actually is, and (fwiw) can also resend to
pipe and file addressees.
Some new commands: `urlenc' and `urldec', `setenv' and `unsetenv'
(like `set' and `unset' but also synchronise the program
environment; those actively fail unless all resource files have been
loaded, which hopefully circumvents possible security implications).
Tarqi also correctly pointed that S-nail is a complete mess
relating to -auth- and password variables.
This will **change** (i.e., without remaining backward compatible)
in v15.0, but i just implemented a generic URL parse and
credential lookup facility which can be used already today by
setting the *v15-compat* variable:
PROTO://[USER[:PASSWORD]@]server[:port][/path]
If no USER is given:
*user-HOST* then *user* (which defaults to *USER* which defaults
to getpwuid(getuid()) a.k.a. the current user).
No more prompt asking for user names.
If no PASSWORD is given:
*password-USER@HOST*, then *password-HOST* then *password*.
Prompt on the terminal if all else fails (and the authentication
type requires a password).
Authentication:
*PROTO-auth-USER@HOST* then *PROTO-auth-HOST* then *PROTO-auth*
(which has a protocol-specific default).
Note I:
For SMTP the rules are a bit more complicated:
regardless of *v15-compat* we will *always* work relative to
*from*, not to the smtp:// account in respect to S/MIME
(*smime-sign* and *smime-sign-include-certs*).
This is so because S/MIME verification uses `From:' (or
`Sender:').
For the SMTP 'MAIL FROM:<>' we also use *from* _unless_ the new
*smtp-hostname* is also set (and setting it to an empty string
automatically deduces *hostname* or, as a last resort, the local
hostname for it), in which case that is used as `HOST' (and the
SMTP account user is `USER').
So, for one you should make yourself familiar with what is
possible now with several different settings of *from*,
*sender*, *from*, *hostname*, *smtp-hostname* etc..
But even with the Yandex provider (which requires that 'MAIL
FROM:<>' uses @yandex.XY) S-nail can now be driven with
define _sdaoden_yandex {
call _sdaoden_shared
set v15-compat
set user=sdaoden password=PASSWORD hostname=yandex.com
set smtp=smtps://smtp.yandex.ru:465 smtp-auth=plain smtp-hostname=
ghost xi "fi %:imaps://imap.yandex.ru
}
account sn_sf {
localopts 1
call _sdaoden_yandex
set from="[email protected] (Steffen Nurpmeso)"
}
so that the next announcement mail will again be able to use
the projects mail address.
Note that the `Message-Id:' header is (1) shortened a bit and
(2) reflects several different settings from above, in
particular the *smtp-hostname* thing.
Note III:
IMAP is not yet done completely for cases where the URL includes
a `:PASSWORD' and/or `/path'; i'll enter the jungle and fix
that for v14.7, however.
All credential and authentication variables not mentioned above
will continue to work until v15.0, and then vanish.
From my testing it all seems to work.
As usual care has been taken to update the manual (ok, i've
forgotten to mention the cursor key change in last release).
I hope you have a nice time in the summer, should there otherwise
be silence around here...
Ciao,
------------------------------------------------------------------------------
Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
• 3 signs your SCM is hindering your productivity
• Requirements for releasing software faster
• Expert tips and advice for migrating your SCM now
http://p.sf.net/sfu/perforce
_______________________________________________
S-nail-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/s-nail-users
