Re: Duplicity 3.1.0 Released

--===============6627340576059846248==
Content-Type: multipart/signed; boundary="=-=-=";
	micalg=pgp-sha512; protocol="application/pgp-signature"

--=-=-=
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable

Kenneth Loafman <[email protected]> writes:

> It turns out that gpg has changed the way it handles key usage, so failur=
e or success would end up dependent upon what
> gpg version you were running.  Originally the encrypt key we were using d=
oubled as a sign key and that worked until now.=20
> I had to use 'gpg --edit-key 839E6A2856538CCF` to also allow encryption a=
s well as signing.  It was not a lack of
> password but a lack of usage flag in the key.  I took out the restriction=
 on Docker as well.

Thank you very much Kenneth! It solves the problem for me.

>
> I've attached a gzipped patch to provide the changes.
>
> ...Thanks,
> ...Ken
>
> On Wed, Jul 1, 2026 at 6:03=E2=80=AFAM Kenneth Loafman <[email protected]=
om> wrote:
>
>  Interesting that it would fail now.  It ran on Linux, macOS, and GitLab =
CI before the release.
>
>  I'll look into it.
>
>  ...Ken
>
>  On Mon, Jun 29, 2026 at 5:03=E2=80=AFPM Sam James via Duplicity-talk <du=
[email protected]> wrote:
>
>  Scott Hannahs via Duplicity-talk <[email protected]> writes:
>
>  > Thanks for the update.  Trying to adjust my fink-project build system =
for this.  I get a couple of failures
>  when running
>  > the suite of tests.  I know the pylint and the black tests are not nec=
essary but I dislike failures in my
>  output.  Put it
>  > down to programmer OCD.  I am sure the tests all passed on your system=
 before you released it, but I am curious
>  what is
>  > different on my system that all but these 3 tests would pass.  I am us=
ing python 3.10 and pylint 2.14.5.  Now
>  updating
>  > pylint (and astroid) to current version.
>  >
>  > [...]
>  >
>  > 3. This is further on a specific test for a code regression number 908=
.  I am including the long log here
>  because I am
>  > not sure what is relevant.
>  >
>  >  testing/functional/test_regression.py::RegressionTest::test_issue908
>  >  FAILED [ 72%]
>
>  I'm seeing this as well (https://bugs.gentoo.org/978226) but not figured
>  out why yet. I'll note that it's skipped in CI (the "Won't work on
>  docker") thing but I'm not sure why, as it seems like it should be OK in
>  Docker too. In any case, it's not just you, just I didn't have enough
>  info to report it yet.
>
>  >  =E2=80=94=E2=80=94
>  >  _________________________ RegressionTest.test_issue908 ______________=
___________
>  >
>  >  self =3D <testing.functional.test_regression.RegressionTest testMetho=
d=3Dtest_issue908>
>  >
>  >      @unittest.skipIf(os.path.exists("/.dockerenv"), "Won't work on do=
cker")
>  > [...]
>
>  sam
>  _______________________________________________
>  Duplicity-talk mailing list
>  [email protected]
>  https://lists.nongnu.org/mailman/listinfo/duplicity-talk

--=-=-=
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----

iQEBBAEWCgCpFiEEJaa7iN2bdkxrVUHCc4QJ9SDfkZAFAmpP8BcbFIAAAAAABAAO
bWFudTIsMi41KzEuMTIsMiwyXxSAAAAAAC4AKGlzc3Vlci1mcHJAbm90YXRpb25z
Lm9wZW5wZ3AuZmlmdGhob3JzZW1hbi5uZXQyNUE2QkI4OEREOUI3NjRDNkI1NTQx
QzI3Mzg0MDlGNTIwREY5MTkwDxxzYW1AZ2VudG9vLm9yZwAKCRBzhAn1IN+RkI5k
AQCTunaSvJpPM9Ic9D0OAA1Q/P3sAbEpU5D/pL1kk+bygwEA+V8SSrKDpi2m5uon
66IK6C8nLYyD9VmItE6/qp818w8=
=5cwO
-----END PGP SIGNATURE-----
--=-=-=--


--===============6627340576059846248==
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: base64
Content-Disposition: inline

X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18KRHVwbGljaXR5
LXRhbGsgbWFpbGluZyBsaXN0CkR1cGxpY2l0eS10YWxrQG5vbmdudS5vcmcKaHR0cHM6Ly9saXN0
cy5ub25nbnUub3JnL21haWxtYW4vbGlzdGluZm8vZHVwbGljaXR5LXRhbGsK

--===============6627340576059846248==--