This is just a warning.

Use the pdf version.
An info file will be installed with gnupg.

Hello Werner,

Hallo Werner!

For the given use case you should ask the former employee to revoke the uid.
And in case you can't contact him, the signers may revoke their signatures
(--edit-key, "revsig").

Werner, please note: this issue is kkc.

It is unofficial and the FAQ is outdated anyway.

OpenPGP!
Marcus means, we need a similar pinentry message which ask for correctness of
fingerprint of the certificate which the user want to certified (like the
pinentry if you enabled allow-mark-trusted option for S/MIME).

OpenPGP or CMS?

EDIT: The NO_SECKEY comes from a second key for which I do not have the private
key. But all I get after pressing cancel is BAD_PASSPHRASE, which could happen
as well if the user really got it right on the second try.

Hmmm... I am talking about GnuPG 1.4.10 here. I am not getting anything like
this. I am, however, using gpg-agent from gpg2.

Actually there is a queue. However it needs a redesign to check again right
before the passpharse dialog comes up. I think this is a real bug.

What about this:

That's a fair point. However it's just not there in glibc (at least I couldn't
find it). But if that's the only offending part the first two hunks could just
be skipped.

Does not look pretty.
I ask myself why glib does not implement its own printf functions to get rid of
such macros. I recently changed the gnupg code base to use the estream printf
functions which now allows us to use modern printf format specifiers. That is
much nicer than cluttering the code with casts or macros.

Applied. Thanks.

Hi Werner,
the G_* stuff is from GLIB/GTK+ so for a GTK+ object it's what somebody would
expect. If you want to stay close to the rest of the pinentry coding style I'm
or course fine with that.

No, this proprietary software.

Marcus, please decide whether we should apply them. Afaics, the options are not
the standard ones we use. The strange G_GSIZE_FORMAT macros look strange, we
better cast to int or use "%zd". Using the unused attribute to mark unused args
is not portable; better we access them (void)foo;

Here's the link to the GTK+ 3 transition:

D129: 294_0002-Add-gseal-gtk-compat.h-from-gedit.patch

Compat with older GTK+

D130: 293_0001-Make-DSEAL-safe.patch

This still affects 0.8.0.

I understand (and exactly it is why I do not want use some weak RNG but use
something robust in crytsetup).

No, this would violate the design of the RNG. It is already hard enough to come
up with good random and we don't want to weake it anymore.

No the warning will stay. I will add a comment to show that this is an optional
part of the API.

I see, so there is no immediate action needed and I can release a full
features libassuan 2.0.0 even without this feature, and in a future version
this warning will be gone, right?

It is liklely that we remove gpgkey2ssh entirely. It used to be a test program
for the gpg-agent implementation. I never used it.

The libassuan provides this feature and we don't want to remove it. GnuPG
however does not use it antmore. In fact, in the development branch we are
working on replacing all use of stdio by our estream code which features an
fopencookie like fucntion and thus it will be possible to use features like
logging to sockets aso on system without a native funopen/fopencookie.

In card edit it is:

Changed in all branches (svn rev 5256).
The prompts are now:

Taking a look at the code, it would require to adjust

May I ask for the status? Seems, there were no objections.

Agreed.

Implemented since 1.0.2