Bison used to be the de-facto standard yacc ;-)

I think that a simple way is defining a table (string -> token) by ourselves in yylex, not enabling %token-table.
(Then, we don't need to depend on the feature of string with %token, which is not supported by POSIX yacc.)

Note that I'm referring to file based keys, not card based.

I tested this on 2.3, and it doesn't seem to be fixed. When adding an existing ECDSA subkey I don't get the option to choose whether to make it a signing or encrypting subkey. Instead it only allows me to choose an encrypting subkey.

Thanks for your findings. I recall that I read this in the announcement and cursed about this new tendency in GNU to break long standing APIs.

Looks like yytoknum was removed from Bison in version 3.8: http://git.savannah.gnu.org/cgit/bison.git/commit/?id=1efe31185ff6b0bc22ff527098971bedf1ace5f4

Danke -

As long as we can't replicate this, it does not make sense to keep this bug open. Please re-open it if you run into it again in a replicatable way.

Problem/Bug still persists in current version (gpg4win 3.1.16) --> reopen

sorry for a confusion. We do not plan to certify DSA so disregard the second part of the patch.

Sorry, I just discovered that I had to click on "Save All" in order for the file to be actually stored in the disk and then it works.

Here it goes...

Please hit "mostra de registro..." link in the blue box and show us its content (you may want to check that it does not show sensitive data)

Thanks for the log, however, I would suggest to use 3.1.16 and try again.

Do we really need to support DSA in FIPS mode? I mean standard DSA and not ECDSA.

Argh, sorry for bugging. Clearing comment out - I simply missed fact that my tests are run with random messages, so with 5% probability another password will be interpreted as 'good' for the first SKESK.

Works for me:

$ gpg --version
gpg (GnuPG) 2.2.27
libgcrypt 1.9.4-unknown
Copyright (C) 2021 Free Software Foundation, Inc.
License GNU GPL-3.0-or-later <https://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.

The usual procedure for downgrading is

1. Uninstall the currently installed version
2. Install the older version

You should never ever downgrade. What is the problem with the new 2.2.32?

What do you mean by asking on a ML or on IRC for networking help?

Hi, I have installed 2.2.32, but still get the same error.

Thank you for your reply! I have updated version numbers and the used OS. I will try with GnuPG 2.2.32

I can't tell you why you get this error. However, since Oct 1 the keyserver access does in many case not work anymnore. This has been fixed in GnuPG 2.2.32, which I released a few minutes ago. You may install this on top of gpg4win 3.1.16.

Backported to 2.2.32

You mean Gpg4win. The solution for Gpg4win 3.1.x is to install the latest GnUPG LTS installer for Windows on top of the latest Gpg4win version. See
https://lists.gnupg.org/pipermail/gnupg-announce/2021q3/000464.html
Noet that there will very soon be a 2.2.32 to fix a problem with Let's encrypt protected keyservers (T5639).

Just for everbody else who might be waiting for a new release. Workaround is to simply use the previous version: https://www.gpg4win.de/change-history-de.html (3.1.15)

Thanks for the report. However, for 1.4 we will only apply important real world security patches. A brief review did not reveal any setious problems. Theoretical memory leaks will not be fixed. Note that your report also includes patches to parts of the code which are not anymore used.

Currently, I am using --lock-never config to avoid generating lock file as a workaround.

For 2.3, when you use PC/SC, please use the disable-ccid option in your .gnupg/scdaemon.conf.

okidok - understand and thank you.

Quite possibe and thanks for the report. However, this is a dev state of the things and thus not expected to work. I'll keep this open as a reminder for me, but in general I would prefer to get a report at the gnupg-devel ML.

Sorry, a hostname with slash is simply not allowed by IETF standards. Given that the hostname is part of temporary file names, you will run into an error. Yes, we could remap the slash in the mktemp function but there are lot of other plzces where the hostname is used and certain properties are expected.

Well this seems to be a gcc 4.2 bug. But well, forward declarations should go into a separate file so that tehre is only one place which would require changes. In this case it does not matter.

Hello again Werner,

Thank you for reporting.
Fixed in master.

Hi @Lambd0x, with Thunderbird having migrated to a different main OpenPGP implementation and Enigmail not supporting old thunderbirds version anymore (in two days https://www.enigmail.net/index.php/en/home/news/71-2021-08-31-end-of-support-for-thunderbird) and new GnuPG versions over the last three years. Do you still have problems?

@werner I think @Rombobeorn suggests something like

Sorry, I can't read all your comments about this. The percent escaping is correct and required. If you want to use the output in a script you can get it without percent escaping by using for example

Hello again Werner,

OK, effect avoided. Only dued to a big sequence of '-' characters that this bug reporting system interpreted as special control code characters for making font bigger.

As I anticipated here's screenshot I also saved.
P.S. Just in case you might wonder why I tried to execute 'man' not existing command is because I also incorrectly tried to run it 2 days ago when after I tested some GnuPG commands in a CMD window, and again same window self-closed with no error messages or any dump creation and again I'm sure at that time I didn't run 'mode con' before those GnuPG commands (I'm unsure which ones but they probably some were the same I also already reported here). Please also note that making 'Stato del dispositivo CON:' so big was unintended and very likely just an effect of cut & paste I did, so now I'll try to see if I can avoid that effect.

Hello again Werner,

As the bug I located is a simple fix, I think it can be also applied to 2.2.

Ok, but the problem is that by default CMD command windows do not use any Unicode set of characters, and their default font is 'Raster characters' and the typical display of characters inside these windows, at least for Italian people because this was the default since MS-DOS age, is to only consider using MS-DOS ASCII codepage 850, not Unicode.
And this is also why before showing the error message I ran the command mode.
Anyway because of the 2 characters getting displayed instead of single expected 'è' (whose character code is 0x82) I can also tell you that by using Charmap.exe again (just in case you may not know it allows display of any Unicode or Raster font characters and their hexadecimal codes) and so switching its default font from System (Unicode) to 8514oem (raster, corresponding to 8514oem.fon) I've been able to determine which should be the 2 wrong displayed characters hexadecimal codes : 'Ã' (0xC3), and '¨' (0xA8) but here obviously both characters cannot be represented correctly because this site is using 'Segoe UI' font, which is Unicode too.
So the only main way to check that I identified the correct characters will be if you (or someone else you know) with another Windows client run Charmap.exe switch font to 8514oem (raster) and check that displayed characters corresponding to 0xC3 and 0xA8 are the same that got displayed in the error message bitmap I provided (a part from probably being able to find where in Italian messages localization rather then in source code those 2 characters have been used incorrectly).
P.S. Because in a past similar bug about Italian localization I saw a comment from you about this kind of option, please just let me know in case you think I might possibly be able to help you to review involved 'GPG' Italian error message(s) for this bug.

Well, as I've said in the comment above, there doesn't seem to be any correction towarads --passphrase-fd not requiring --pinentry-mode loopback (still works withou)... and --no-default-keyring still gives the impression that it would be needed (while --no-keyring works as well).

Please don't, if you really feel like tha tis not resolved please re-open this ticket.

That pretty much looks like the other errors you have with Unicode characters - which we can't replicate.

This is all build from the same source. We could fix that but I'll give that a lo priority. Thanks for reporting.

P.S. Also note that my Gpg4win 3.1.16 installation status is still last T5593 described one.

Here's the file I anticipated (this time uploading worked... :-D ) (my Gpg4win 3.1.16 installation status is still T5593 described one).

@werner shall I open a new ticket for the remaining stuff?

Thanks. This fixes the invalid packet errors when using --list-packets or when trying to decrypt the file without secret key.

Fixed in rGcc6152b802f2: gpg: Skip the packet when not used for AEAD., but I put wrong bug-id in the commit message.