FYI, I'm still planning on working on the Java packaging issues. But, still working on figuring out how to do the work. I've had a lot of other problems come up lately so I have not made too much progress. But, I still expect to be working on this soon.

On Wednesday, February 5, 2020, 9:11:25 AM EST, Fabio Valentini <decathorpe@gmail.com> wrote:


On Wed, Feb 5, 2020 at 2:55 PM Mario Torre <neugens@redhat.com> wrote:
>
> On Wed, Feb 5, 2020 at 2:41 PM Fabio Valentini <decathorpe@gmail.com> wrote:
> > > > Currently impossible due to completely broken Java stack in Fedora.
> > >
> > > When I read this I'm always puzzled by the negativity.
> > >
> > > What is "completely broken" exactly that you need? Things seem to work
> > > quite well.
> >
> > The problem is that the current "work[s] quite well" situation relies
> > entirely on a small number of volunteers (including me), who don't
> > even want to maintain Java packages.
> > If I alone decide to finally give up on Java packages in fedora
> > (because why not?), you can say goodbye to almost 200 Java packages
> > (not including dependencies).

<>

> First of all, thank you for your work! I know it may not mean much but
> it is appreciated.

Thank you.

> On a general side, I would say, however, that this is the nature of
> every collaborative effort, isn't it?

Sure. Some members of the Stewardship SIG also work on DogTag PKI, so
they actually need a working Java stack.
On the other hand, I don't use any Java packages at all (other than
OpenJDK itself), and just took ownership of most packages to prevent
them from being removed and 25% of fedora imploding.

> Also, I can only speak for JMC (since my team owns this) and for
> OpenJDK (since I'm part of the OpenJDK team an Jiri own this), and I
> know the base is sound. And we aren't required to do any of this work,
> but we do it because we want to have Fedora a great platform for Java
> development, so we put efforts in ensuring that OpenJDK (and JMC) are
> packaged correctly.

Your work is appreciated :) OpenJDK works quite well.

> Also, there is always the risk that someone leaves to do better
> things, but this doesn't mean "completely broken" (otherwise why
> haven't the packages been retired?), so I think my first question is
> still valid.

"Completely broken" is a rather pessimistic point of view.
For example, because of our continued work, of the ~240 packages the
Stewardship SIG maintains, *none* failed to build in the f32 mass
rebuild.

However, due to a lack of manpower, we need to focus our effort on
basically 1) keep packages building, 2) fix CVEs filed against our
packages.
Some packages we can keep up-to-date without much effort, but we can't
deal with the few gnarly bits since the people who knew how to deal
with those packages have either left or abandoned non-modular Java
packages.

So to answer your question, the only reasons why not more packages
have been retired is 1) we took them when they were orphaned, 2) we
fix FTBFS issues, and 3) we fix CVEs. But that's really the bare
minimum to prevent packages from being removed entirely. We can reuse
some work from the modular branches, but the modular Java maintainers
have made it clear that they are not interested in helping us. So ...
collaborative effort? Not so much

Fabio

> Cheers,
> Mario
>
>
> --
> Mario Torre
> Associate Manager, Software Engineering
> Red Hat GmbH <https://www.redhat.com>
> 9704 A60C B4BE A8B8 0F30  9205 5D7E 4952 3F65 7898
> _______________________________________________
> devel mailing list -- devel@lists.fedoraproject.org
> To unsubscribe send an email to devel-leave@lists.fedoraproject.org
> 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@lists.fedoraproject.org