Am 06.11.23 um 18:32 schrieb Fabio Valentini:
On Mon, Nov 6, 2023 at 6:28 PM Julian Sikorski <belegdol@xxxxxxxxx> wrote:
Am 06.11.23 um 17:17 schrieb Michael J Gruber:
Am Mo., 6. Nov. 2023 um 17:12 Uhr schrieb Fabio Valentini
<decathorpe@xxxxxxxxx <mailto:decathorpe@xxxxxxxxx>>:
On Mon, Nov 6, 2023 at 5:09 PM Julian Sikorski <belegdol@xxxxxxxxx
<mailto:belegdol@xxxxxxxxx>> wrote:
>
> Hello,
>
> I have accidentally built f39/rawhide branch of gnumeric and
> gnome-chemistry-utils for f38 and f37 side tags (f39 too but
rawhide and
> f39 are the same commit). Can this be fixed? Or is the effort not
worth
> it? Thanks.
Are the contents of the package the same except for the changelog
entries / Release field?
Then I would not bother trying to fix it.
It is kind of unfortunate that koji does not have guardrails in place
to prevent building branch X against target Y!=X, but that's probably
something that actually needs to work for Modularity or other
configurations ...
Fabio
The spec files differ in content (as far as flatpaks are concerned). But
since this package uses separate branches without forwards/merges but
cherry-picks only, i.e. clearly separates branches in git, it would be
surprising not to care about the branch now ... I guess git branches are
still somewhat foreign to how we drive our dist.
Michael
Hello,
for gnome-chemistry-utils, the differences are only with release field
and changelog. For gnumeric, f37 has a slightly different
desktop-file-install call in addition.
I would prefer to fix it properly. Due to fewer mass rebuilds older
gnome-chemistry-utils branches should have lower release than the newer
ones. If the problem remains unfixed, I would have to bump the release
to make sure the numbers don't go backwards. For gnumeric this is a bit
less of a problem since there are new versions still being released so
release number would reset soon-ish anyway.
I tried untagging the erroneous builds with
$ koji untag-build
While it worked, I was not able to rebuild from correct branches because
koji was complaining that a build already exists:
https://koji.fedoraproject.org/koji/taskinfo?taskID=108660999
https://koji.fedoraproject.org/koji/taskinfo?taskID=108661023
Can I request to have the builds removed so that they can be rebuilt
from correct branches?
No, NVRs for successful builds are unique in koji.
The only way to resubmit something that already succeeded is to bump release.
Fabio
Thanks for confirming. I have bumped the releases accordingly.
Best regards,
Julian
_______________________________________________
devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx
Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx
Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
