Re: F29 System Wide Change: Modules for Everyone

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Wed, Jul 04, 2018 at 08:50:32AM -0400, Neal Gompa wrote:
> On Wed, Jul 4, 2018 at 7:38 AM Zbigniew Jędrzejewski-Szmek
> <zbyszek@xxxxxxxxx> wrote:
> >
> > On Wed, Jun 20, 2018 at 02:20:59PM -0400, Stephen Gallagher wrote:
> > > It also carries with it the implication that the supported package managers
> > > must handle module updates correctly. (It does *not* require that they all
> > > be able to handle selecting/switching module streams, but it must honor
> > > whichever stream was set via DNF).
> >
> > I expect that getting support in all the package managers is the hard part.
> > I'm happy to see this explicitly stated in the Change page and included in
> > the Contingency section.
> >
> > What package managements softwares need to be updated apart from packagekit?
> 
> Since we're ignoring YUM, the main issue at this point is that
> modulemd isn't understood by libsolv, so handling of modules is very
> strange.

Yes, I think yum can be ignored. Removal of yum has been proposed and
is generally supported:
https://fedoraproject.org/wiki/Changes/Deprecate_YUM_3.

> > What is the current status of those changes?
> 
> With DNF 3.x, libmodulemd is used through gi for the python frontend.
> With libdnf 0.15.x, libmodulemd is linked in and we're forced to do
> our own depsolving for modules, because the solver doesn't know
> anything about them.
> 
> That said, my understanding is that the repodata format is still kind
> of up in the air, so until the module repodata format is rationalized
> to be sane w.r.t. rpm-md, then I suspect this will not get fixed.
> 
> No one I've talked to is particularly enthused with the idea of having
> YAML parsing for repodata, because of how insane YAML tends to be. And
> libsolv is intended to be mostly self-contained, so it has its own
> internal representation of rpm-md, comps, etc. YAML as input is fine,
> but we're more comfortable with the idea of the output being XML so we
> can keep using the same validated XML parser for everything. Multiple
> parsers make things much harder...
> 
> Incidentally, one of the reasons why PackageKit doesn't understand
> comps anymore is because libdnf hasn't transitioned to using libsolv
> for comps processing yet. That's something that's coming eventually,
> and with it will be a simpler API for dealing with comps groups.

Hmm, all those details… What about a checklist? ;)
- dnf ✓
- dnfdragora ?
- pkcon ?
- Gnome Software ?
- yum - (not needed)
- something else ?

Zbyszek
_______________________________________________
devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx/message/I3ASE33IT24MKYMXDR4ILRCNL65PRM6T/




[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Index of Archives]     [Fedora Announce]     [Fedora Users]     [Fedora Kernel]     [Fedora Testing]     [Fedora Formulas]     [Fedora PHP Devel]     [Kernel Development]     [Fedora Legacy]     [Fedora Maintainers]     [Fedora Desktop]     [PAM]     [Red Hat Development]     [Gimp]     [Yosemite News]

  Powered by Linux