[leafnode-list] Re: User authentication
Whiskers
catwheezel at operamail.com
Wed Sep 17 15:24:13 CEST 2008
On Wed, 17 Sep 2008 01:44:12 +0200 "Matthias Andree"
<matthias.andree at gmx.de> wrote:
> Whiskers schrieb am 2008-09-17:
> > On Tue, 16 Sep 2008 21:07:45 +0200 "Robert Grimm"
> > <lists at news.robgri.de> wrote:
[...]
> > Odd!
> >
> > Although your email has a heading
> >
> > Content-Transfer-Encoding: base64
> >
> > it appears to be plain text only - so my Claws Mail couldn't decode it
> > <grin> (but reading the 'source' is perfectly clear).
> >
> > Is that a side-effect of posting to the list via Gmane's usenet
> > gateway?
>
> AFAICT, it's not. I received Robert's message directly from the list;
> it was declared base64, it contained base64 ("as advertised"), and mutt
> could decode it without further ado.
>
> So look for something along your inbound path, or a Claws bug.
I've looked along my inbound path. This may be a combination of a Claws
bug and my non-standard way of getting the messages in this list.
Claws has never failed to decode Base 64 correctly for me, although such
encoding is very rare in the emails I get. But perhaps Claws doesn't know
what to do when told that something is Base 64 when it isn't. And the
email as it exists on my hard disc is very definitely plain text, despite
the Base 64 coding indicated in the header.
So if the email left the sender as Base 64, how did it change to plain
text before getting to me? I think the answer is that this list gets to
me at a webmail 'inbox' (Operamail) which I read locally in Claws Mail via
"FreePops" - a program which scrapes the webmail pages and presents the
emails in a local POP server which Claws Mail gets them from. It looks as
though Operamail's web server decodes the Base 64 but of course the
encoding header is still present, so Claws Mail gets a conflict it can't
handle.
Interesting.
--
-- ^^^^^^^^^^
-- Whiskers
-- ~~~~~~~~~~
More information about the leafnode-list
mailing list