public inbox for cygwin-apps@cygwin.com
 help / color / mirror / Atom feed
From: Daisuke Fujimura <booleanlabel@gmail.com>
To: cygwin-apps@cygwin.com
Subject: Re: [ITA] ruby 3.2.2
Date: Wed, 26 Apr 2023 06:10:28 +0900	[thread overview]
Message-ID: <CAA3frXR9dsu18KNi2vCxM3Yofab-JkdTKSF54=9hd6ZFLVuSQQ@mail.gmail.com> (raw)
In-Reply-To: <6e1f2251-0e38-7de1-7a3d-88f90b1eeef0@dronecode.org.uk>

Thank you for your response.

Following the announcement of the previous update, I would like to
note that the binaries need to be recompiled.
- https://www.mail-archive.com/cygwin-announce@cygwin.com/msg08753.html

> I applied my usual workaround (which is adding 'ruby_23' to an internal
> list of things which are allowed to not be provided), and set the job to
> rerun, which seems to have succeeded.

Does this mean adding `ruby_23` to `depend2` in some packages in setup.ini?

I would like to know more about the specifications of setup.ini. Is
there any documentation available somewhere?


On Tue, Apr 25, 2023 at 10:52 PM Jon Turney <jon.turney@dronecode.org.uk> wrote:
>
> On 25/04/2023 10:56, Daisuke Fujimura via Cygwin-apps wrote:
> >> I don't think so. Please, go ahead and deploy.
> >
> > I tried to deploy twice, but it failed.
> >
> > First attempt:
> > - https://github.com/cygwin/scallywag/actions/runs/4791077183
> >
> > ```
> > ERROR: package 'ruby-tcltk' version '3.2.2-1' has empty install tar
> > file, but it's not in ['virtual', '_obsolete'] category
> > ERROR: error while validating merged x86_64 packages for Daisuke Fujimura
> > SUMMARY: 2 ERROR(s)
> > ```
> >
> > I fixed it according to the error message.
> > - https://cygwin.com/cgit/cygwin-packages/ruby/commit/?id=e1bc357d4ca0423b5ec92aaeb3846adf7351efa3
> >
>
> Thanks.
>
> If you subsequently adopt the ruby-tk package, please remember to add
> ruby_tk_OBSOLETES="ruby-tcltk" there (as that's the preferred way to
> record obsoletions nowadays)
>
> > Second attempt:
> > - https://github.com/cygwin/scallywag/actions/runs/4791758304
> >
> > ```
> > ERROR: package 'ruby-caca' version '0.99.beta19-4' depends: 'ruby_23',
> > but nothing satisfies that
> > ERROR: package 'ruby-marisa' version '0.2.4-2' depends: 'ruby_23', but
> > nothing satisfies that
> > ERROR: package 'ruby-marisa' version '0.2.4-3' depends: 'ruby_23', but
> > nothing satisfies that
> > ERROR: package 'ruby-openbabel' version '2.3.2-6' depends: 'ruby_23',
> > but nothing satisfies that
> > ERROR: package 'ruby-openbabel' version '2.3.2-5' depends: 'ruby_23',
> > but nothing satisfies that
> > ERROR: package 'ruby-openwsman' version '2.6.3-3' depends: 'ruby_23',
> > but nothing satisfies that
> > ERROR: package 'ruby-openwsman' version '2.6.3-2' depends: 'ruby_23',
> > but nothing satisfies that
> > ERROR: package 'ruby-xapian' version '1.2.24-1' depends: 'ruby_23',
> > but nothing satisfies that
> > ERROR: package 'ruby-xapian' version '1.4.5-1' depends: 'ruby_23', but
> > nothing satisfies that
> > ERROR: package 'ruby-zinnia' version '0.06-8' depends: 'ruby_23', but
> > nothing satisfies that
> > ERROR: package 'ruby-zinnia' version '0.06-9' depends: 'ruby_23', but
> > nothing satisfies that
> > ERROR: package 'subversion-ruby' version '1.11.1-1' depends:
> > 'ruby_23', but nothing satisfies that
> > ERROR: x86_64 package set has errors after removing stale packages
> > ERROR: error while evaluating stale packages for Daisuke Fujimura
> > SUMMARY: 14 ERROR(s)
> > ```
> >
> > When I deployed ruby-3.2.2, ruby-2.3.6 (which provides ruby_23) was
> > moved to the vault, resulting in some ruby-* subpackages being unable
> > to satisfy the dependency of ruby_23.
> >
> > To resolve this, the following methods are being considered:
> >
> > - Do not move ruby-2.3.6 to the vault (I cannot do this myself).
> > - Rebuild the subpackages.
> > - Any other methods?
>
> Yeah, calm needs to learn how to deal with this scenario better.
>
> Probably what should actually happen here is that these packages get
> expired at the same time that the last thing providing ruby_23 gets
> expired (as they can no longer be installed and thus keeping them is a
> bit pointless)
>
> > Is there anything I can do to deploy ruby-3.2.2?
>
> I applied my usual workaround (which is adding 'ruby_23' to an internal
> list of things which are allowed to not be provided), and set the job to
> rerun, which seems to have succeeded.
>
>
> Apologies for the inconvenience, and thanks again for updating ruby!
>

  reply	other threads:[~2023-04-25 21:11 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-04-19 22:42 Daisuke Fujimura
2023-04-20  3:28 ` Marco Atzeri
2023-04-20 10:50   ` Jon Turney
2023-04-20 16:13     ` Jon Turney
2023-04-21 19:36       ` Daisuke Fujimura
2023-04-21 20:06         ` Jon Turney
2023-04-22 12:04           ` Daisuke Fujimura
2023-04-23 13:35             ` Jon Turney
2023-04-23 23:44               ` Daisuke Fujimura
2023-04-24 20:10                 ` Jon Turney
2023-04-25  9:56                   ` Daisuke Fujimura
2023-04-25 13:52                     ` Jon Turney
2023-04-25 21:10                       ` Daisuke Fujimura [this message]
2023-04-25 22:13                         ` Jon Turney
2023-04-25 22:37                           ` Daisuke Fujimura
2023-06-08 14:38 ` [GOLDSTAR] " Jon Turney
2023-06-13 15:04   ` Andrew Schulman
2023-06-13 23:24     ` Daisuke Fujimura

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAA3frXR9dsu18KNi2vCxM3Yofab-JkdTKSF54=9hd6ZFLVuSQQ@mail.gmail.com' \
    --to=booleanlabel@gmail.com \
    --cc=cygwin-apps@cygwin.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).