[Mailman-Users] Reject *to* non-subscriber

[glen martin]

I take John's points well, and indeed had thought of those myself.  In
the case of this particular list, I think the small noncompliance is 
being done by accident rather than connivance, so bouncing the posts
would seem to be a useful way to reinforce subscriber training. :)

Custom handlers look interesting, thanks for the pointer.  Having read
the page it seems that to use my custom handler for just one list, in
all methods I have to hard-code a new handler chain in entirety.  Do I
correctly understand that one can't add the new handler as a delta for a
single list (eg somehow akin to GLOBAL_PIPELINE.insert described for
global list changes)?  Goal being to permit further global changes to be
reflected in the single list as well without having to remember to
update the chain on the single list.

BTW, I didn't say it before, but I've been very impressed with Mailman. 
Very easy to set up, has run flawlessly for my subscribers. All the
efforts of the contributers are much appreciated.

glen


Mark Sapiro wrote:
> John W. Baxter wrote:
>
>   
>> On 10/11/07 2:33 PM, "glen martin" <lists at locutory.org> wrote:
>>
>>     
>>> I run a private list where the list-members have decided they don't want
>>> non-subscribers on the threads at all.  But occasionally one will forget
>>> and cross-post. I'd like to bounce subscriber messages where a
>>> non-subscriber is anywhere on the receiver list.  Is there anything like
>>> this currently available or contemplated?
>>>       
>> It can't be fully done because of Bcc.  (Except in a very specific scenario
>> which likely isn't achievable.)
>>
>> It could be done with respect to To: and Cc:, but that might well just give
>> a false sense of privateness.
>>     
>
>
> John makes good points.
>
> To answer your original question, this would have to be done with a
> custom handler
> <http://www.python.org/cgi-bin/faqw-mm.py?req=show&file=faq04.067.htp>.
> There are no plans for anything more direct in the future.
>
>