duplicity-team team mailing list archive

[Kenneth Loafman <kenneth@xxxxxxxxxxx>]

I have no problem with that.  A lot of those distributions are at least a
dozen versions behind and are not keeping up-to-date anyway.  As web hosts
constantly change their API in non-backwards compatible manners, duplicity
has to change and the changes go into the next version.  Since the distros
lock to a particular version, the bug caused by the web host is not fixed
until the user upgrades duplicity outside the bounds of their repo
restrictions.  That's perfectly fine with me.  The distro maintainers are
alway able to patch a fix in if they need, but I've seen few signs of that.

I imagine in a couple of years I'll still get requests for support of 0.6.x
duplicity and my answer will always be to upgrade.  If we can't make
progress during LTS, then I refuse to play their games.

Meanwhile if anyone wants to preserve the old duplicity, they are free to
clone 1.2.3 (the last planned py2/3 release) at any time.  The web hosts
will soon guarantee that it's out of date.

...Ken


On Sun, Jul 9, 2023 at 5:22 PM Jakob Bohm via Duplicity-talk <
duplicity-talk@xxxxxxxxxx> wrote:

> On 06/07/2023 16:35, Kenneth Loafman via Duplicity-talk wrote:
>
> Intro
>
> This will serve to keep track of major changes coming down the pike for
> v2.0 duplicity.  I'll update with any new changes
>
> I have made a new project called duplicity-py3
> <https://gitlab.com/duplicity/duplicity-py3>. It passes all the tests we
> have. I’m working on more comprehensive testing now.
> Timeline
>
> At this point we have released 3 alpha and 3 beta versions with almost
> minimal from testers. I’d like to get the testing done before the release
> if at all possible.
>
> The main branch has been merged into *duplicity-py3*, and *main* has is
> now in a release freeze. When we release 2.0, I’ll just replace main's repo
> with the duplicity-py3 repo and be done with it, down to just one project
> again. Between now and then I have a tentative release schedule:
> date release
> 07/10 RC0
> 07/17 RC1
> 07/24 RC2
> 07/27 2.0.0 Python 3.8+ only from now on
>
>    - 2.7 is End Of Life
>    - 3.0 to 3.7 are End Of Life
>    - remove future (includes builtins)
>    - pure Python 3 code only (1 C extension)
>
>
> Please be aware that LTS Linux distributions will be some releases behind
> upstream Python and will rely on their own Python version for other system
> functions.  Thus users on those systems will be practically unable to use
> anything that requires Python 3.8 for a few more years.  Here are some
> examples found from online lookups, note especially Ubuntu 18.04 LTS using
> Python 3.6 for another 5 years .
>
> Debian 10 Buster - Exp 2024-06-30 - Python 3.7
> Debian 11 Bullseye - Exp 2026-06-30? - Python 3.9
> Debian 12 Bookworm - Exp ??? - Python 3.11
> Ubuntu 14.04 Trusty Tahr - Exp 2024-04-01 - Python 3.4
> Ubuntu 16.04 Xenial Xerus - Exp 2026-04-01 - Python 3.5
> Ubuntu 18.04 Bionic Beaver - Exp 2028-04-01 -Python 3.6
> Ubuntu 20.04 Focal Fossa - Exp 2030-04-01 - Python 3.8
> Ubuntu 22.04 Jammy Jellyfish - Exp 2032-04-01 - Python 3.10
> Ubuntu 23.04 Lunar Lobster - Exp 2024-01-01 - Python 3.11
>
> P.S.
>
> I have no problem using s3-boto3, it's my current choice
>
> Enjoy
>
> --
> Jakob Bohm, CIO, Partner, WiseMo A/S.  https://www.wisemo.com
> Transformervej 29, 2860 Søborg, Denmark.  Direct +45 31 13 16 10
> This public discussion message is non-binding and may contain errors.
> WiseMo - Remote Service Management for PCs, Phones and Embedd
>
> _______________________________________________
> Duplicity-talk mailing list
> Duplicity-talk@xxxxxxxxxx
> https://lists.nongnu.org/mailman/listinfo/duplicity-talk
>