Mailing List Archive

On Thu, 13 Jun 2019 21:27, look@my.amazin.horse said:
> I integrated one more of dkg's tests. Subkey revocations are now imported even
> in the absence of subkey binding certificates. I believe this patch series now
> covers the low hanging fruit wrt import of partial keys.

As remarked over at the WG list I am not convinced of that whole change.
The patches I have seen are pretty intrusive and change long standing
behaviour of gpg.

You also missed to send a DCO.


Salam-Shalom,

Werner

--
Die Gedanken sind frei. Ausnahmen regelt ein Bundesgesetz.

On Thu 2019-06-13 21:27:40 +0200, Vincent Breitmoser wrote:
> I integrated one more of dkg's tests. Subkey revocations are now imported even
> in the absence of subkey binding certificates. I believe this patch series now
> covers the low hanging fruit wrt import of partial keys.
>
> This makes GnuPG fully compatible with the split identity/non-identity
> information model of keys.openpgp.org, and fixes the issue mentioned in our FAQ:
> https://keys.openpgp.org/about/faq#older-gnupg
>
> Please let me know if there are any issues remaining with this patch, and I will
> try to resolve them.

I have reviewed these changes, and they look good to me. I've pushed a
fix-4393 branch to the upstream git repository containing those changes
with minor commit message cleanup, and my signoff on them.

--dkg

Hi,

thanks for considering this patch series,

> As remarked over at the WG list I am not convinced of that whole change.
> The patches I have seen are pretty intrusive and change long standing
> behaviour of gpg.

Can you be more specific about your concerns?

Surely if GnuPG encounters a valid revocation certificate and is in a position
to import it without too much effort or subverting reasonable user expectations,
it should do so?

I understand how the patches are a little awkward, as they basically just remove
checks for error cases that have existed for a long time. I can't say that
I understand all the implications this might have, so a close review would be
greatly appreciated.

> You also missed to send a DCO.

Sorry, it's been a while since I read the contribution guidelines. I'll send
one.

- V


_______________________________________________
Gnupg-devel mailing list
Gnupg-devel@gnupg.org
http://lists.gnupg.org/mailman/listinfo/gnupg-devel

> As remarked over at the WG list I am not convinced of that whole change.

I interpreted your reply "useful feature request" and triage as normal on T4393
as agreement that this is generally a desirable feature?

If you are only unhappy with the method of the patch, please give some specifics
and I will try to fix them.

> The patches I have seen are pretty intrusive and change long standing
> behaviour of gpg.

Note that this patch changes behavior mostly in a way that was already
achievable before through the use of "import-drop-uids".

- V


_______________________________________________
Gnupg-devel mailing list
Gnupg-devel@gnupg.org
http://lists.gnupg.org/mailman/listinfo/gnupg-devel