Re: lists vs. subscribe (was Re: mutt to follow discussions.)

2003-05-30 Thread Thomas Roessler
On 2003-05-25 13:23:32 -0400, Gregory Seidman wrote:

 I'm trying to figure this out myself. In particular, it looks
 like there might be a bug in either the documentation or (I hope)
 mutt 1.4. The documentation says that the ~l pattern matches
 messages to known lists (anything matched in a subscribe or lists
 config command), but it actually only matches subscribed lists
 (ignoring anything in a lists config command).

It's, in fact, a bug in the documentation: ~l is supposed to match
messages which have been received through a mailing list.  It does
not match messages which have been sent to you directly, and which
may have been CCed to some mailing list you happen to know about.

 Someone please correct me if I'm wrong, but I believe the
 distinction between known (lists command) and subscribed
 (subscribe command) has to do with what the Mail-Followup-To
 header contains when followup_to is true. If you are subscribed,
 it contains the list address; if not, it contains your address.
 Is this correct, or is there more to it?

For what follows, it's important to note that the set of known lists
is always a superset of the set of subscribed lists.

The notion of a known list is mainly important for the list-reply
command; that command will send a reply to all *known* lists to
which a message is directed (unless there's a mail-followup-to
header).

The logic for the mail-followup-to header is a little more
complicated, and involves both subscribed and known lists.  

First of all, mutt will only generate a mail-followup-to header if
(1) the $followup_to option is set, and (2) the message goes to a
*known* list.  The header will then point to *all* recipients of the
message (i.e., all the lists and anyone who may be CCed).  

Your own address will be added to the mail-followup-to header if you
are not subscribed to any of the lists to which you direct
responses.

In pseudo-code:

if ((message goes to known list)  (we haven't obtained \
mail_followup_to from user or from message we reply to))) {

copy all recipients to mail-followup-to;

remove self from mail-followup-to;

if (!(mail-followup-to points to subscribed list)) {
add self to mail-followup-to;
}
}

Hope that helps,
-- 
Thomas Roessler [EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: lists vs. subscribe (was Re: mutt to follow discussions.)

2003-05-30 Thread Thomas Roessler
On 2003-05-29 14:24:38 -0700, Bill Moseley wrote:

 And to follow up to my own post.  Mutt 1.5.4i (2003-03-19) does
 not set $use_domain yes by default.  The docs say:

 use_domain

 Type: boolean
 Default: yes

 When set, Mutt will qualify all local addresses (ones without the @host 
 portion) with the value of `` $hostname''. If unset, no addresses will 
 be qualified. 

 But I needed to set it in my .muttrc to get mutt to qualify my address 
 in the Mail-Followup-To header.  So a doc bug.

Not a doc bug -- that part of the documentation is immediately
created from the relevant program code.  But, of course, the Debian
package may unset use_domain in the systemwide configuration file
/etc/Muttrc.

Regards,
-- 
Thomas Roessler [EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Upgrading slink - potato - any problems?

2000-06-20 Thread Thomas Roessler
On http://debian.schlittermann.de/, I read that there is
no easy upgrade path from slink to potato.

Is this still true?