duplicate suppression, sieve, loops, redirect and lost email

[Kevin P. Fleming]

Rob Siemborski wrote:
> On Fri, 30 Aug 2002, Ken Murchison wrote:
> 
> 
>>It already does keep track of msgid/rcpt pairs for redirect, but I still
>>don't see how this helps us.  How does the script know that it
>>redirected back to itself?  Perhaps adding the server's hostname to the
>>X-Sieve header would help us determine that this message is coming back
>>to us via a loop.
> 
> 
> I don't think this is good enough, since I could have:
> 
> redirect "leg+cyrus at andrew.cmu.edu"
> 
> in my sieve script, which (if only the server name was included) would
> then get dropped (or bounced or whatever).
> 
> However, putting the expected rcpt-to value in X-Sieve might be
> interesting, since then we have a user and a host name that has already
> seen the message.
> 
> -Rob
> 
> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
> Rob Siemborski * Andrew Systems Group * Cyert Hall 207 * 412-268-7456
> Research Systems Programmer * /usr/contributed Gatekeeper
> 
> 

I'm not sure I see the problem here...

What I'm suggesting is a lot simpler: when a message is redirected by 
the sieve script, it's message id is stored.

If, during a subsequent invocation of the script, the same message id is 
requested to be redirected again, the redirection is ignored (or turned 
into a keep).

For this purpose, it doesn't matter where the message was redirected 
_to_ the first time, because it's obvious that somehow it came back to 
the same mailbox since otherwise this script wouldn't be processing the 
message.

This would even handle the case where the message gets redirected by 
multiple scripts only to end up back at the original mailbox again.