From: Joshua Lock <josh@linux.intel.com>
To: openembedded-core@lists.openembedded.org
Subject: Re: [RFC] Working toward a GNOME layer
Date: Tue, 26 Apr 2011 13:26:37 -0700 [thread overview]
Message-ID: <1303849597.28281.39.camel@scimitar> (raw)
In-Reply-To: <A64BF647-3F77-4108-BAED-FBF48FE567D7@dominion.thruhere.net>
On Fri, 2011-04-22 at 19:34 +0200, Koen Kooi wrote:
> Op 22 apr 2011, om 18:23 heeft Joshua Lock het volgende geschreven:
>
> > On Thu, 2011-04-21 at 20:12 +0200, Koen Kooi wrote:
> >> Op 21 apr 2011, om 19:41 heeft Joshua Lock het volgende geschreven:
> >>
> >>> On Thu, 2011-04-21 at 19:29 +0200, Koen Kooi wrote:
> >>>> Op 21 apr 2011, om 18:40 heeft Joshua Lock het volgende geschreven:
> >>>>
> >>>>> On Thu, 2011-04-21 at 16:05 +0100, Paul Eggleton wrote:
> >>>>>> On Thursday 21 April 2011 15:02:49 Koen Kooi wrote:
> >>>>>>> and possibly more. I would like to create a meta-gnome layer in the
> >>>>>>> meta-openembedded repository where new recipes get added and things from
> >>>>>>> meta-demoapps can get moved over into. Long term recipes-gnome in oe-core
> >>>>>>> should move there as well.
> >>>>>>>
> >>>>>>> What are your thoughts on this?
> >>>>>
> >>>>> +1
> >>>>>
> >>>>>>
> >>>>>> From my perspective this sounds like a great idea. The only question would be
> >>>>>> how much of the "GNOME" libs would remain in oe-core as some of them are quite
> >>>>>> widely used outside of GNOME proper; however that can easily be worked out as
> >>>>>> these things mature.
> >>>>>
> >>>>> +1
> >>>>>
> >>>>> My personal opinion would be that we start with glib & gtk+ (plus their
> >>>>> dependencies, i.e. pango, atk, etc) in core and move the rest out to a
> >>>>> layer.
> >>>>>
> >>>>> I feel that Gtk+ is used by enough non-gnome software that it belongs in
> >>>>> core but others may disagree?
> >>>>>
> >>>>> Between meta/recipes-gnome and meta-demoapps we have a reasonable start
> >>>>> to a meta-gnome/
> >>>>
> >>>> Where did meta-demoapps go? It's not in OE-core anymore by the looks of it.
> >>>
> >>> Hmm, still exists for me:
> >>>
> >>> joshual@vorpal:~/Projects/Yocto/oe-core/meta-demoapps[master]
> >>> $ pwd
> >>> /home/joshual/Projects/Yocto/oe-core/meta-demoapps
> >>>
> >>>
> >>>>
> >>>>> I'd be happy to help with this layer.
> >>>>
> >>>> Awesome, do you have any objection to put meta-gnome into the meta-openembedded repo for the time being? Once we get better tooling we can move it elsewhere, of course.
> >>>
> >>> None whatsoever. I keep meaning to push some recipes into that layer
> >>> anyway.
> >>
> >> I just sent out 10 patches for review that import recipes-gnome from meta-demoapps into meta-gnome. Please review :)
> >
> > The way I see it there are two approaches, tidy & test the recipes then
> > merge *or* merge then fix.
> >
> > If we're going for the latter approach let's get your patches merged!
>
> While I'd love the first approach, the second is a lot more practical
> for 'yocto' recipes, I would reserve the vetting for the ones imported
> from OE. Does that sound OK?
Agreed.
>
> > This does raise another question, is meta-oe striving for the same
> > standards of metadata as oe-core? i.e. SRC_URI & license checksums,
> > updated patch syntax, etc.
>
> It does, but it still has crud in it from "the early days" which needs
> to get cleaned up. No license checksum is fatal, so that one is
> covered, but the source checksums aren't. I would be nice to make the
> following errors fatal (if they haven't already):
>
> 1) license checksums
> 2) LDFLAGS (gnu-hash)
> 3) source checksums
Bug report in email? :-)
>
> > Also, how much gnome do we want to support? Are we trying to be all new
> > and shiny and drop deprecated libraries (gnome-vfs)?
>
> Personally I would go for importing 2.30 from OE and see if there are
> 2.32 updates available. 2.30 should be relatively gnome-vfs free. So
> for this specific example, let's try to avoid gnome-vfs. In a broader
> sense, let's get gnome 2 in and look at gnome 3 later. I suspect gnome
> 3 will be a world of hurt with its clutter and hence openGL
> requirements.
I agree with this approach. I think the gobject-introspection will be a
bigger pain point than GL though.
>
> > Just trying to work out what patches to work on ;-)
>
> Which reminds me, the gtk+ in meta-oe needs to get compared to the one
> in oe-core to see what's different, I don't want to keep a copy in
> meta-oe.
>
> > Perhaps we can
> > define a policy of what's appropriate for the layer in a README?
>
> Are you volunteering to draft a README? If so, go for it :)
I guess so. I'll try and submit some patches.
Joshua
--
Joshua Lock
Yocto Build System Monkey
Intel Open Source Technology Centre
next prev parent reply other threads:[~2011-04-26 20:29 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-04-21 14:02 [RFC] Working toward a GNOME layer Koen Kooi
2011-04-21 15:05 ` Paul Eggleton
2011-04-21 16:40 ` Joshua Lock
2011-04-21 17:29 ` Koen Kooi
2011-04-21 17:41 ` Joshua Lock
2011-04-21 18:01 ` Koen Kooi
2011-04-21 18:12 ` Koen Kooi
2011-04-22 16:23 ` Joshua Lock
2011-04-22 17:34 ` Koen Kooi
2011-04-26 20:26 ` Joshua Lock [this message]
2011-04-21 17:57 ` Saul Wold
2011-04-21 19:23 ` Richard Purdie
2011-04-22 16:25 ` Joshua Lock
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=1303849597.28281.39.camel@scimitar \
--to=josh@linux.intel.com \
--cc=openembedded-core@lists.openembedded.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.