On Thu, Jun 07, 2018 at 09:01:41AM -0500, Dirk Eddelbuettel wrote:
>
> Consider the following thought experiment:
>
> - the change was made in 2014
> - nobody noticed
There is no need for a thought experiment. I was working around this
bug several times (last time here[1]). The problem simply became
obvious since dh-update-R tries to create a sensible set of dependencies
- formerly the set was manually craftet and it did not became obvious
that its just a bug in r-cran-hmisc.
If you need another proof that somebody noticed and also worked around
feel free to inspect Graham's commit[2].
> - the bug is filed 'grave'
grave
makes the package in question unusable or mostly so
If you install r-cran-hmisc on a fresh installation it is unusable.
> So either nobody uses the package, or it doesn't matter in practical
> use. Both ot these cannot be true at the same.
There is no definition if "practical use". A package should work out of
the box on a fresh installation. R-cran-hmisc does not fulfill this
criterion. Moreover it took me only a couple of minutes to prove that
the issue caused work for at least two fellow developers.
> I am not interested in theoretical consideration quote reference manual or
> policy. I have been here long enough to understand that you are correct in
> the narrow (and here, irrelevant) sense. I simply still have not given up
> hope that you may have some common sense left. I could be wrong.
I would love if we could settle with technical argumentation and leave
out speculations about common sense. Please be so kind and upload a
fixed package instead of defending your inappropriate decrease of the
bug severity.
Thank you
Andreas.
[1]
https://salsa.debian.org/r-pkg-team/r-bioc-biovizbase/commit/c2aab7de43f385fc20ec966861ca3aaa374c3fcc
[2]
https://salsa.debian.org/r-pkg-team/r-bioc-cummerbund/commit/367f89148e51f37188480d17f3f12f3dfa7a77e1
--
http://fam-tille.de
