procmail
[Top] [All Lists]

Re: More Matching questions - regexp

2000-09-27 14:49:38
At 09:32 2000-09-27 -0500, tomcat(_at_)visi(_dot_)com wrote:
strips off all header lines when replying (or replying and
including the original message) except for (sic)

[snip]

except that it should gleefully *USE* the Reply-To header to override the address in the From:, for determining where the message should be sent. This header should not be retained in the headers within the REPLY, but the sofware should take it into account.

Excepting that there's probably some braindead mailer out there someplace that won't support it right, you should be able to specify a Reply-To like so:

        Reply-To: originatoraddress(_at_)domain(_dot_)com, 
gatekeeper(_at_)bozo(_dot_)com

(the gatekeeper address will be cleartexted in the TO of the reply, which means they'll likely get copies of continued discussions if the originator uses REPLY-ALL -- but NOT if they just do REPLY)

If you did it this way, the webform perl script could set this header on the message going to the officer (possibly also setting From: to the originator address as well, if supported by the SMTP server settings), with a carbon copy going to the gatekeeper address (from the perl script). Gatekeeper then sees the original message - doesn't DO anything with it other than logging it.

Officer receives the message, hits reply, at which time BOTH of the above addresses appear in their TO: line, and they compose and send their reply, copy goes _directly_ to originator, as well as to the gatekeeper, which again, receives a copy and does nothing more than logging it.

The webform perl script could hock a line into the top of the original sent body reminding the admin to not change the subject (or to at least keep the "reference code" (hash value) in the subject).

I can't change any of these fields.  The Destination replies to the
Gatekeeper

Why can't you change any of these fields? Procmail (via formail) will gladly allow you to change them.

To: Gatekeeper(_at_)bozo(_dot_)com
From: clubofficer(_at_)bozo(_dot_)com
Subject: Re: Webmail: The user's subject here *- bozo(_at_)bozo(_dot_)com

Looking at your problem, it REALLY seems like the perl script should just be setting a reply-to (if not a From: for that matter, but perhaps your server's SMTP relaying config won't allow that) and mailing the club officer(s), who, when they reply to the message, will end up sending a message to the reply-to address -- the web user. If you need copies logged by the gatekeeper, do it as I suggested futher above.

An alternate way, if you must relay the message through the gatekeeper is to see if your ISP doesn't support plussed addresses, then mangle the relay address into the gatekeeper's address (you'll have some fun dealing with this). Have you even considered just putting it within parens?

        From: <Gatekeeper(_at_)bozo(_dot_)com> 
(realuseraddress(_at_)wherever(_dot_)com)
        To: clubofficer(_at_)bozo(_dot_)com

Though some cheezeball email client will no doubt mangle the From (just replying with the address, and not the parenthetical text), so this may not be an appropriate solution.

Unfortunatley, some braindead corporate mail clients (Lotus notes comes to mind) mangle plussed addresses as well, because those clients don't obey RFC standards. If your officers ARE NOT on such systems, you can probably ignore this concern.

I want a copy of the original and of the reply in the Gatekeeper account,
because I write the subjects to a log file, and can
tell if the club officers have responded to the Originators
email.  In the logfile I should see pairs:

What happens when the originator then sends a new message, this time directly to the club officer, now entirely bypassing the gateway? Or do you plan to mask the officer's address (note: they may have it in their .sig!) to prevent future direct contacts, for which you'll be entirely out of the loop?

I can say from personal preference that if I have an _email_ address for someone, I'd rather use that than use some kludgy webform to make contact.

BTW - if your "webform" expects the user to have valid email, providing a mailto: instead of using a webform will increase your chances that the email address they use is valid (some people seem not to be able to relate their address properly, but usually their email client has it configured fine - and if not, they probaby won't get your messages anyway).

I absolutely refuse to have folks email the club
and not know if a response from the club got sent.  After
3 days or so, if the club officer did not respond, I respond
or forward the original to someone else to respond to.

Okay, the logic finally comes out. Would asking officers to voluntarily BCC the gatekeeper not work? If they don't, then you can slap them around.


Yea, have the webform send it to your gateway, poss with a certain subject prefix (which confirms it came from the webform), which you strip, then forward the message to the appropriate officer:

Subject: [VISI-WEBFORM] real subject

:0
* ^Subject:[    ]*\[VISI-WEBFORM\][     ]*/\.*
{
        REALSUBJECT=$MATCH
}


Finally, procmail deals with the reply by looking for "Re: Webmail:"
(of course, I will make the first line in the body of the original

As indicated in a previous post of mine, you can't really expect that the reply WILL be prefixed with "Re:" - some mailers do things their own way:
        [Re:] Webmail
        RE2: webmail
        (etc)

The concept of having the webform insert some prefix into the message sent initially will catch the FIRST message, you strip that prefix off, and anything else that comes through you don't have to worry about being a webform or RE:reply, because the presence of the token would have identified the webform (though the From: of the webform system could have served that purpose, unless you change that).

You could also have the webform assign an incrementing number, or a hash (of the date/time/number) into the subject.

Now, Originator and Destination can reply until hell freezes over,
Gatekeeper is out of the loop.

So long as you're clear that this is what happens - AND that future contacts from the originator may go directly to an officer, without involving your system.

---
 Please DO NOT carbon me on list replies.  I'll get my copy from the list.

 Sean B. Straw / Professional Software Engineering
 Post Box 2395 / San Rafael, CA  94912-2395


_______________________________________________
procmail mailing list
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail

<Prev in Thread] Current Thread [Next in Thread>