ietf-mta-filters
[Top] [All Lists]

Re: Updated Sieve notification draft

2005-10-03 03:17:40

Michael Haardt wrote:

On Tue, Sep 27, 2005 at 12:10:22PM +0100, Nigel Swinson wrote:
   This document does not dictate the notification method used.
   Examples of possible notification methods are Zephyr and Jabber. The
-    method shall be site-defined.
+    available methods shall be site-defined.

Which makes me wonder if each desired "method" should really be available via a 
require command?

I prefer a require commend for each method

I don't want to require a "require" statement for each URI schema. An URI can be fetched using a test from an external server (e.g. it can be an LDAP attribute value ;-)), in which case the script might not know all URI schemas in advance.

and syntactically checking the URL, generating an error on bad syntax.

I am fine with this part. I've added to 3.1:

   If the :method tag contains a supported URI schema, then the URI MUST
   be checked for validity. An invalid URI syntax or an unsupported URI
   extension MUST cause an error.

Is this what you want?

 And I would go as far as
allowing semantic errors to cause errors on behalf of the implementation.
I ran various SMS gateways over the years and if there is one common
problem, it is people embedding letters and all kinds of special
characters in phone numbers, like @ and 8-bit characters, or doubling
the number or the area prefix which yields way too long numbers.
Simply dropping those numbers results in complaints why important SMS
were not sent.  At one time I had almost 10% bad numbers and I am sick
of complaints.


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