On 28Sep2009 00:28, Bruno Wolff III <bruno@xxxxxxxx> wrote:
| On Mon, Sep 28, 2009 at 07:13:10 +0800,
| Ed Greshko <Ed.Greshko@xxxxxxxxxxx> wrote:
| > As mentioned several times, the string enclosed within " in the To:
| > header is a text comment field and not an address. It matters not that
| > is contains encoded text. The comma is not a field separator. Yet the
| > reply generated by mutt treated it as such.
|
| No it didn't. What it did was escape the invalid rfc 2047 escapes. All of
| the comment part of the address ended up between the quotes because the
| invalid escapes were between the quotes. So while it may have looked odd, it
| (the cc header) contained one address (fedora-list@xxxxxxxxxx).
Hmm. That's roughly true, looking at the header Ed forwarded to me:
Cc: =?us-ascii?Q?=3D=3FISO-8859-1=3FQ=3F_Community=5Fassistance=2C_=5Fenco?=
=?us-ascii?B?dXJhZ2VtZW50LCA9MDlhbmRfYWR2aWNlX2Zvcl8/PSA9P0lTTy04ODU5LTE/?=
=?us-ascii?B?UT91c2luZ19GZWRvcmEu?= ?= <fedora-list@xxxxxxxxxx>
except for the final ?= which seems a bit bogus.
However, I think you'll agree is looks pretty horrible.
I've just saved your message to a new mbox and opened it. Mutt presents it
just fine. Looking at it in the raw, Bruno's message looks fine except
for the CC: line, which is like this:
Cc: =?us-ascii?Q?=3D=3FISO-8859-1=3FQ=3F_Community=5Fassistance=2C_=5Fenco?=
=?us-ascii?B?dXJhZ2VtZW50LCA9MDlhbmRfYWR2aWNlX2Zvcl8/PSA9P0lTTy04ODU5LTE/?=
=?us-ascii?B?UT91c2luZ19GZWRvcmEu?= ?= <fedora-list@xxxxxxxxxx>
The most notable thing about this is that both the quotes and the commas
are 2407 encoded. So an un2407aware mailer will see just one address and
no quotes or commas.
Ed, how sure are you that TBird is handling that line correctly?
If it's breaking that up into multiple addresses, why?
There's two ways to approach this: parse in 822 style first, then
2407decode, which should produce one address; or 2407decode then 822
parse, which should produce one address.
I'm going to hand decode that line and check... Ick.
Post decode, that line then contains more RFC207 content, including
commas. And no quotes:-(
So, it sounds like TBird decodes, then parses. And explodes in Ed's
face.
Bruno, I notice you're running mutt-1.5.18. I'm running 1.5.20 and it
produces much nicer results, which are visibly good to the human eye. In
fact it takes Frank's not properly 2407 encoded header and undoes it
and writes an unencoded address, nice and clear.
Maybe you could upgrade - it would save a lot of bickering:-)
Though a dig through the changelog doesn't show me a relevant looking
change:-(
Hmm. I just tried 1.5.18 on Frank's original message with the odd
rfc2407 stuff; it behaves just fine too.
I'd really like to know what happened to Bruno's setup.
Bruno, could you attach Frank's original message, as it is in _your_
mail folder, to your reply for inspection? I'd like to compare it
against my copy and the copy Ed sent me. I have trouble believing
they're all the same.
Cheers,
--
Cameron Simpson <cs@xxxxxxxxxx> DoD#743
http://www.cskk.ezoshosting.com/cs/
Six trillion RFID tags is four orders of magnitude bigger than any electronic item ever made.
- overhead by WIRED at the Intelligent Printing conference Oct2006
--
fedora-list mailing list
fedora-list@xxxxxxxxxx
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list
Guidelines: http://fedoraproject.org/wiki/Communicate/MailingListGuidelines
