Take care: gpg is also used on platforms with proprietary compilers which don't support -f options. Thus you need to limit this to gcc.

After some more checking: LLVM-11 introduced the same behaviour in that regard, but appearently not a pragma/attribute to override this: https://releases.llvm.org/11.0.0/tools/clang/docs/ReleaseNotes.html

This reminds me that I should check if kconfig_compiler nowadays supports cross compiling or add that. Back when I started cross compiling kleopatra in 2015 I was lazy and patched in the generated kconfig files. I never really saw the advantage of them but yeah it's more KDEish to use them.

This works now with 0c1bd9076958e584820fadf997ca7d8a248b6888 but needs more testing before this can be relased. It will probably be part of a Gpg4win-4 beta.

Taking since I ran into this problem while working on T5174. In Kleopatra, if one opens the certificate details of one's own keys (i.e. secret key is available), then the tags vanish from the key list.

The C++, CL, Javascript and QT Bindings are all written by hand.

Hi Werner,

can you elaborate?

Ok. I've just added with rev 69539cea316f2d2998eefbd14539f8def3fc07ab more platform specific code for userinformation so this makes even more sense.

Have you tried on the command line: "gpg --gen-key" ? Maybe this gives more helpful debug output. But in the case of "permission denied" this really sounds like a local setup issue as there are also no other reports related to this. Still I'll add the saveguard to make the check in kleo more explicit.

In T5189#140853, @gniibe wrote:

Please check following translations:

"do not detach from the console"
"do not use the internal CCID driver"
"do not use a reader's pinpad"

Those are explanation for the options to instruct gpg-agent or scdaemon, not do something.
It's not a text to users.

It seems you have a pretty good understanding and also test cases at hand. May I ask you to apply the suggested pacthes to master?

Given all the resources we had put on this Python bindings I'd suggest to bite the bullet and replace Swig by handcrafted bindings. More work but we do it for the other bindings as well.

I'd suggest to first try the current version to see whether the bug has been solved.

I think we can close this one, right?

For the context of all subscribed parties I think Werner refers to what Hockeypuck is doing: https://lists.gnupg.org/pipermail/gnupg-users/2020-December/064441.html

Meanwhile there are simpler ideas and code on how to do only authenticated uploads. Thus lowering the prio.

I think the option you are looking for is "--homedir" with that option on the command line you can redirect where GnuPG looks for options and keys.

@glr thanks for the offer. Right now the number of support requests is low enough that we can handle them during the normal triage.

Please try using the current version (3.1.14) and if the problem persists re-open this bug. In this case we will also need a more detailed report.

(Reporter has problems running his own keyserver and accessing it.)

Please try gpg4win 3.1.14 and check whether your problem has gone.

Flagged as high becuase this is RC for Libgcrypt 1.9

rG6850f21d08b2: po: Fix Simplified Chinese Translation. is a fix for adjusting columns; The number means the number of columns.
rGf4a8be0950ea: po: Fix Simplified Chinese Translation. is a fix for:

• SETERROR is a command name of pinentry which should not be translated.
• The message is expected to be displayed in four lines.

Please check following translations:

"do not detach from the console"
"do not use the internal CCID driver"
"do not use a reader's pinpad"

Those are explanation for the options to instruct gpg-agent or scdaemon, not do something.
It's not a text to users.

Sorry, I didn't read your message above, and it's applied and pushed to master, due to exactly same reason (it's so big).
It's easier (at least for me), when it's in git repo.

I've attempted the fix a couple of times now, and it doesn't seem like Kleopatra is running with elevated privileges, but I'm still encountering the permission denial.

Thus better add a

&& !defined(__clang__)

Sure that the FreeBSD compiler does not define it? I am pretty sure it does.

According to list of attributes in the clang 12 documention, there is no such attribute in clang. However, the clang-11 compiler (as seen in Debian) does not define __GNUC__, so the proposed patch does not affect the status when building with clang.