Re: [Exim] Using the textual part of email address

Page principale
Supprimer ce message
Répondre à ce message
Auteur: Ross Boylan
Date:  
À: Jim Knoble, exim-users
CC: ross Boylan
Sujet: Re: [Exim] Using the textual part of email address
At 01:09 AM 8/15/02 -0400, Jim Knoble wrote:

>How are you receiving your mail now? Via POP/IMAP, or via some other
>method?


POP3


>If you have any measure of control over the mail server that parses
>local parts for your domain, then you should use local-part prefixes or
>suffixes for this.
>
>For example, mail to <rossboylan+ross@???> would go to you, and
>mail to <rossboylan+karen@???> would go to your wife. The way
>that would work is that the upstream mail server interprets the '+' and
>everything following it in the local-part to be a "suffix" which it can
>ignore when routing the message. When your machine gets it, it
>interprets everything before (and including) the '+' as a "prefix"
>which can be ignored when delivering.


Thanks, I didn't know about that. Unfortunately, the + trick does seem to
work with my ISP, but it does not work with my "permanent" address. To add
insult to injury, the failure message comes from exim on their end! Here
it is:

This message was created automatically by mail delivery software (Exim).
A message that you sent could not be delivered to one or more of its
recipients. This is a permanent error. The following address(es) failed:
RossBoylan+ross@???
SMTP error from remote mailer after RCPT
TO:<RossBoylan+ross@???>:
host mxpool.postoffice.net [165.212.65.113]: 550
<RossBoylan+ross@???>... User not known




>Exim 3.x's directors have options which handle these prefixes and
>suffixes (don't confuse them with the prefix and suffix options for the
>appendfile transport, which are completely different).
>
>Alternatively, if you don't have that much control over the upstream
>server (for example, your ISP is unable or unwilling to implement
>local-part suffixes or prefixes), you may wish to subscribe to an
>aliasing service such as pobox.com. You would set up several aliases
>in the pobox.com domain---for example, <ross@???> and
><karen@???>---both of which you would configure to forward to
>your mailbox at your ISP. When you receive the mail, you parse the
>recipient headers (To:, CC:, Resent-To:, Resent-CC:) in the same way
>you're talking about now, but this time you've got actual addresses
>to key on, rather than display text. This way other folks besides
>yourself can use the addresses without having to remember to set the
>display text appropriately.


Does this give enough information to sort things out? I really need the
original address the message was sent to, and that still seems to be
obscured. Suppose someone sends mail To: karen@???,
ross@???. How do I know, when I'm handling a particularly delivery,
which recipient it is for?


>You may find that procmail or similar delivery agents are better suited
>for distinguishing addresses in headers and acting on them.


Why do you say that?

>--
>jim knoble | jmknoble@??? | http://www.pobox.com/~jmknoble/
>(GnuPG fingerprint: 31C4:8AAC:F24E:A70C:4000::BBF4:289F:EAA8:1381:1491)