Hello. Does anyone have any idea why sqwebmail is UTF-8/I18N encoding the
subject line, just because I put a single = sign in it?

See message below for why this is causing me problems.

Thanks...

-- 
Stephen Warren, Software Engineer, Parama Networks, San Jose, CA
[EMAIL PROTECTED]                  http://www.wwwdotorg.org/


----- Original Message ----- 
From: "Jason R. Mastaler" <[EMAIL PROTECTED]>
To: "Stephen Warren" <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTECTED]>
Sent: Monday, August 18, 2003 10:26 PM
Subject: Re: tmda-ofmipd that'll run under tcpserver


> On Mon, 18 Aug 2003, Stephen Warren wrote:
> 
> > OK. I've tracked down the source of the failures using X-TMDA in the
> > subject line via sqwebmail.
> > 
> > If I type "X-TMDA dated=5d" into the subject line, this gets sent to
> > sendit.sh (and hence tmda-sendmail/tmda-inject) as "Subject:
> > =?utf-8?Q?dated=3D10d?=" - I guess sqwebmail is seeing the = in the
> > message and escaping things out so it won't clash with the = for
> > marking it UTF-8. So, TMDA is thinking I want to use the explicit
> > address "=?utf-8?Q?dated=3D10d?="
> > 
> > So, the question is - is this an issue in TMDA because it isn't
> > parsing/decoding the UTF-8 encoded data, or shouldn't sqwebmail be
> > doing this encoding in the first place? Votes? :-)
> 
> I don't think either package is technically in the wrong here.  I
> didn't intend for the X_TMDA_IN_SUBJECT feature to support I18N, but
> that doesn't mean we can't change this.
> 
> I'll look into this as a future enhancement, as we should have the
> tools in TMDA to do it.  Basically, we'd need to grok the encoding
> used, then decode the entire Subject string, then perform the X-TMDA
> stripping, and then re-encode the Subject string sans 'X-TMDA foo'.
> 
> One thing to consider in the meantime however, is why sqwebmail is
> encoding the Subject string with UTF-8 when it's not necessary.  Most
> MUAs do this only on demand, which is probably why this issue has
> never arisen before.

Reply via email to