mail to list+subscribe if already subscribed

mail to list+subscribe if already subscribed

[Jakob Hirsch]

Hi,

mlmmj seems to ignore subscribe-mails from already subscribed addresses. 
Shouldn't it notify the user in this case?

Another thing I'd appreciate: Use a custom header instead of To/CC for 
getting the recipient address. This may not sound useful, but I had 
list-management mails sent to the list, because they had no To/CC 
header, but the envelope recipient was right. I'd tell my MTA to put the 
envelope-rcpt into the header (as it happens already with the 
envelope-sender in Return-Path) and would be on the safe side then.

What do you think?
I'd try to put that into the code (not many changes, I'd say) and send 
patches.

Re: mail to list+subscribe if already subscribed

[Mads Martin Joergensen]

* Jakob Hirsch <jh@plonk.de> [May 05. 2005 22:36]:
> Hi,
> 
> mlmmj seems to ignore subscribe-mails from already subscribed
> addresses.  Shouldn't it notify the user in this case?

You might argue that it should, and the reason it doesn't is that noone
made it so yet.

> Another thing I'd appreciate: Use a custom header instead of To/CC for
> getting the recipient address. This may not sound useful, but I had
> list-management mails sent to the list, because they had no To/CC
> header, but the envelope recipient was right. I'd tell my MTA to put
> the envelope-rcpt into the header (as it happens already with the
> envelope-sender in Return-Path) and would be on the safe side then.

Why do you allow mail to the list where the listaddress is not in To:
or Cc: in the first place?

You should use Delivered-To: for the envelope-rcpt, postfix (and qmail)
already does that.

> What do you think?  I'd try to put that into the code (not many
> changes, I'd say) and send patches.

Uuuuh, patches always sound extremely nice :-)

-- 
Mads Martin Joergensen, http://mmj.dk
"Why make things difficult, when it is possible to make them cryptic
 and totally illogical, with just a little bit more effort?"
                                -- A. P. J.

Re: mail to list+subscribe if already subscribed

[Jakob Hirsch]

Mads Martin Joergensen wrote:

>>mlmmj seems to ignore subscribe-mails from already subscribed
>>addresses.  Shouldn't it notify the user in this case?
> You might argue that it should, and the reason it doesn't is that noone
> made it so yet.

so it's time to change that. :)

> Why do you allow mail to the list where the listaddress is not in To:
> or Cc: in the first place?

Because I thought there's no reason for doing so, but dumb me forgot 
about control messages...

> You should use Delivered-To: for the envelope-rcpt, postfix (and qmail)
> already does that.

Ok, found that in the mlmmj-process manpage.
Exim has an option to add a header called Envelope-to, so it might be a 
good idea to check for that, too. OTOH, it's easy to tell Exim to put 
this into a Delivered-To header.

>>What do you think?  I'd try to put that into the code (not many
>>changes, I'd say) and send patches.
> Uuuuh, patches always sound extremely nice :-)

on my way!

Re: mail to list+subscribe if already subscribed

[Mads Martin Joergensen]

* Jakob Hirsch <jh@plonk.de> [May 06. 2005 14:32]:
> >>mlmmj seems to ignore subscribe-mails from already subscribed
> >>addresses.  Shouldn't it notify the user in this case?
> >You might argue that it should, and the reason it doesn't is that noone
> >made it so yet.
> 
> so it's time to change that. :)

Indeed.

> >Why do you allow mail to the list where the listaddress is not in To:
> >or Cc: in the first place?
> 
> Because I thought there's no reason for doing so, but dumb me forgot 
> about control messages...

And it also prevents spam.

> >You should use Delivered-To: for the envelope-rcpt, postfix (and qmail)
> >already does that.
> 
> Ok, found that in the mlmmj-process manpage.
> Exim has an option to add a header called Envelope-to, so it might be a 
> good idea to check for that, too. OTOH, it's easy to tell Exim to put 
> this into a Delivered-To header.

I thought about using Envelope-To if there is none, but since it's not
added pr. default and one have to add one or the other anyway, I prefer
following the "standard" and add Delivered-To:.

> >>What do you think?  I'd try to put that into the code (not many
> >>changes, I'd say) and send patches.
> >Uuuuh, patches always sound extremely nice :-)
> 
> on my way!

:)

-- 
Mads Martin Joergensen, http://mmj.dk
"Why make things difficult, when it is possible to make them cryptic
 and totally illogical, with just a little bit more effort?"
                                -- A. P. J.