Hi Paul,
On 3/13/26 01:33, Charles Plessy wrote:
I asked for advice on debian-devel about bulk removal bugs.
Le Sat, Mar 14, 2026 at 07:23:11AM +0100, Paul Gevers a écrit :
Not on d-devel, but: Message-ID:
<[email protected]>
I asked for advice on debian-devel about bulk removal bugs.
https://lists.debian.org/debian-devel/2025/12/msg00106.html
I think that point-to-point replies deviate us from the core of the
discussion.
In any case, I have finished the bulk updates that were automatable,
wich was more than 500 packages. They will be all out of the delayed
queue by the end of the week-end. How I reached that point is now a
thing of the past; we had misunderstandings but I think that we can
overcome and eventually forget them. I understand that people are
frustrated on how long it took but I have a full-time job and a family.
I will follow up with requests for removal and targeted uploads.
Soon, most of the r-cran and all of the r-bioc packages will be
available only on 64-bit little-endian architectures, which is still
broader than what upstream supports. People aiming seriously at
supporting scientific computation with R and Debian packages on other
architectures should engage with the R community first. On my side, I
will not upload anymore packages where we turn off autopkgtests on
specific architectures in order to hide the fact that they perform
incorrect computations.
Dirk, I think that we can give you the ownership of the packages that
you need if you prefer not drop architectures in your packages that
depend on ours. I am also fine for making case-by-case exceptions in
the architecture restricions when there is a serious need.
Lastly, porters who would like to assess the feasibility of supporting
r-cran/bioc packages on their architectures can make local builds and
autopkgtests in environments that provides architecture-is-64-bit and
architecture-is-little-endian. I am fine with extending the list of
architecures we support if the porters show that they can provide
patches that the R community is willing to accept.
Cheers,
Charles
--
Charles Plessy Nagahama, Yomitan, Okinawa, Japan
Debian Med packaging team http://www.debian.org/devel/debian-med
Tooting from work, https://fediscience.org/@charles_plessy
Tooting from home, https://framapiaf.org/@charles_plessy
