Cyril Brulebois <kibi@debian.org> (2024-12-29):
> I can't believe this just happened.
>
> There was no advance warning for the installer team (that I'm aware of),
> the removal not only removed the package from unstable but also from
> testing right afterwards, which just sent the D-I Trixie Alpha 1 preps
> flying in million pieces.
>
> Spotted while rebuilding an image locally, where the tooling suddenly
> stopped finding usr-is-merged, while #1091649 was getting filed by
> Andrey (thank you).
>
> Any chance that could be reverted?
Oh no. I assumed this was all coordinated. We can't revert although that looks like no longer a factor judging by the next bit here:
And since I don't anticipate a revert to be either possible or
well-known territory, 39+nmu1 on its way.
This looks to have been successful. If anyone needs a hand with anything (NEW or other changes to the archive), consider me on the hook.
diffoscope confirms, comparing 39 binaries and 39+nmu1 ones:
- the usr-is-merged binary is identical except for the extra changelog
entry and different timestamps;
- ditto for the usrmerge binary, with also dh_installdeb/13.11.9 vs.
dh_installdeb/13.23 in comments above the debhelper-added snippets.
AFAICT there should be no functional changes there.
Cheers,
--
Cyril Brulebois (kibi@debian.org) <https://debamax.com/>
D-I release manager -- Release team member -- Freelance Consultant