------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.
http://www.exim.org/bugzilla/show_bug.cgi?id=245
------- Comment #6 from ph10@??? 2006-11-14 09:50 -------
OK, I've thought about this a bit. If I were implementing ACLs now, I would not
invent "endpass". It seemed like a good idea at the time (when ACLs were
simpler) but it has always proved very hard to explain to people. It is not
necessary for completeness, and it is no longer used in the default
configuration. I wouldn't say it is deprecated, but perhaps "de-emphasised"
might be reasonable. Therefore, I think the idea of extending it is perhaps not
a good one.
Taking that away from your patch then gives a reasonably clean spec for
"message" for all the verbs if we ignore "endpass" for the moment: "message" is
used when any verb causes the ACL to send a return code (and thereby stop
processing the ACL). For an "accept" verb (without "endpass") this is when it
accepts. If "endpass" is used (which can be documented as a non-ideal fudge),
"message" is used for rejection, for backwards compatibility. I will
investigate implementing something along those lines.
[Not, incidentally, that I think anybody is going to read messages sent with a
250 OK response. I'm far too cynical to think that will happen. :-]
--
Configure bugmail:
http://www.exim.org/bugzilla/userprefs.cgi?tab=email
