Due to back porting another change the fix for 2.2 is now also needed in 2.4.

FWIW: I have not done any tests but the comment below is about the case I suspected to be the cuase for your problem:

See rG0988e49c45 which implements time and group but not yet the split thing because we are not shure that is good idea to have this w/o any implementation support.

The fix actually does the same as my suggested workaround.

There is an easy workaround: Append an exclamation mark to the adsk key. This way gpg will only search for this subkey.
An example with my test keys:

secring.gpg is only used by unsupported legacy versions of GnuPG. Since 2.1 it is not anymore used.

Should be fixed now; see commit above.

FWIW: WriteFile and write are more different than in using a HANDLE vs. a libc file descriptor. Despite that a HANDLE might be a 64 bit pointer, it is guaranteed that the value fits into a 32 bit variable. But they still index different objects. The return code and error values are also different.

Much simpler: write is only used in the callbacks and over there gpgme_io_writen[n] shall be used anyway.

Hmm, for the latter this:

Kleopatra test case (similar to gpg):

Seems it gets a record but is not able to parse it (gnupg/dirmngr/dns-stuff.c:getsrv-standard) in your setup. Not sure why it loops - need to debug it.

Fixed in 2.4

I see the problem: The Qt Pinentry does not implement the BUTTON_INFO status and thus we don't get a fully canceled error back (gpg-agent maps the cancel error to fully-cancel if the close button was used). Should be easy to fix in pinentry (set pinentry->close_button in the close eventhandler).

For me it works if I fully cancel (i.e. close the Window at the first prompt):

Just let me note that we used to have such an API : the former gcry_ac_ functions. However, it turned out that they were more complicated to use.

FWIW, we should anyway move on Widnows to the gpgrt provided setenv and getenv which are directly based on the W32API. The problem here is only that we have a lot of getenv in out code and need a wrapper. That wrapper would then also need to provide a static string as getenv does. A first step would be to wrap all getenv into gnupg-getenv calls.

GnuPG is and can't be FIPS-140-3 compliant due to the way it is implemented. We may eventually employ the new hash-and-sign API of Libgcrypt to move into this direction but that has not yet been done. However, this also requires the use of the new indicator API and the, well, a RedHat kernel.

--openpgp means the current OpenPGP standard as implemented by GnuPG. This was important in the first few years of OpenPGP but not anymore today. The option --rfc4880 might be what you want. Please keep also in mind that the preference list declares what a concrete implementation supports and not necessary what's in an RFC.

This is a standard C pattern to declare that one is not interested in the return value. In this case a return value won't help us because we can't log that anyway because we are in a signal handler.

You are right, it is a bad habit not to check this. Thanks for your patch.

We need the 64 bit version for the GpgOL because there are 32 and 64 bit versions of outlook. Thus we also need a 64 bit gpgme and in turn a 64 bit libassuan and libgpg-error. I can't remember why we don't append the 6 to the gpgme dll, though.

Guessing that it works now.

Meanwhile fixed.

We have new box meanwhile.

We do manual approvals.

backported to 2.2