[Gpg4win-devel] Broken claws mail from gpg4win package on the site

Dr. Peter Voigt pvoigt at uos.de
Thu Jan 9 23:55:19 CET 2014


Am Thu, 9 Jan 2014 17:18:49 +0100
schrieb Bernhard Reiter <bernhard at intevation.de>:

> On Wednesday 08 January 2014 at 23:13:05, Cristian Baboi wrote:
> > >It would probably be helpful to verify/decrypt the message on the
> > >command line using gpgsm. Thus we could exclude e.g. GPGSM
> > >problem.s.
> >
> > Both the command line and this version of claws mail are working.
> >
> > > I've just tried to do so but I failed. I have not enough
> > >knowledge about internal processes, e.g. I do not know, if headers
> > >are to be included and if I have to BASE64 decode before and how
> > >to process MIME boundary strings. Maybe someone can give help on
> > >this.
> 
> Overall it is a good idea in this situation, to try to send an
> attachment which has the same cryptographic properties, e.g.
> encrypted to X,Y, signed by Z and then try gpgsm on the saved
> attachment.
> 
> You can make it possible to decode the MIME part for a signed
> messages, but it is not that easy.
> 

Thanks, Bernhard for clarifying this. I just couldn't believe that
decrypting or verifying an S/MIME mail should be that simple. I quick
view into the "source code" of an S/MIME mail did not even reveal to me,
which encoding out of Quoted-printable, Base64 or Radix64 is used for
which part of the mail.

Nevertheless, if you should have some minutes left, I would like to get
a rough insight into this.

Regards,
Peter

PS.: I cannot verify the signature of your last email for unknown
reason. Did you recently change your X.509 certificate? I checked it
with Claws Mail, Thunderbird and GNU Emacs/Gnus. GNU Emacs/Gnus even
hangs forever and had to be interrupted leaving me with a plain base64
message which is very hard to read :-).


More information about the Gpg4win-devel mailing list