>>>>> "BG" == Ben Gertzfield <[EMAIL PROTECTED]> writes:
>>>>> "John" == John W Baxter <John> writes:
John> You're posting as
John> Content-Type: TEXT/PLAIN; CHARSET=US-ASCII
John> Question for the gurus: could the upper case content type
John> description be causing problems here?
BG> It very well could be. The problem is that *some* headers
BG> are case-sensitive (like Subject) and others are not
BG> (like Content-Type). I haven't been putting in random
BG> .lower() calls on every string, so we'll probably run
BG> into some of these issues.
BG> Barry, do you think the email module should just deal with
BG> this somehow?
There are a number of RFC-ish things I think email should eventually
do (probably not for 1.1). E.g. it should enforce the RFC specified
max and min on numbers of headers (e.g. RFC 2822 says From: must
appear once and only once). We could do some more enforcement of case
sensitivity, but I also think Mailman ought to be more robust.
I suspect that we're not doing case folding on parameters retrieved
with email.Message.Message.get_param(). Note that get_type() already
folds to lowercase, since it's obvious that MIME types are case
insensitive. get_param() /matches/ case insensitively, but returns
the case preserved version of the parameter.
I'm going to try to dig up Ben's patch and I'll double check that the
parameters are being compared case insensitively.
-Barry
_______________________________________________
Mailman-Developers mailing list
[EMAIL PROTECTED]
http://mail.python.org/mailman/listinfo/mailman-developers
