[oe] [PATCH 1/2] protobuf: add protobuf-c recipe

[Bruce Ashfield]

On Thu, Apr 28, 2016 at 4:02 AM, Felipe Ferreri Tonello <
eu at felipetonello.com> wrote:

> Hi Bruce,
>
> On 20/04/16 15:32, Bruce Ashfield wrote:
> > On Wed, Apr 20, 2016 at 4:32 AM, Felipe Ferreri Tonello <
> > eu at felipetonello.com> wrote:
> >
> >> Hi Bruce,
> >>
> >> On 19/04/16 20:55, Bruce Ashfield wrote:
> >>> On Tue, Apr 19, 2016 at 3:46 PM, Bruce Ashfield <
> >> bruce.ashfield at gmail.com>
> >>> wrote:
> >>>
> >>>>
> >>>>
> >>>> On Tue, Apr 19, 2016 at 11:40 AM, Felipe F. Tonello <
> >> eu at felipetonello.com>
> >>>> wrote:
> >>>>
> >>>>> Initial version of recipe. The main package could be split into two
> to
> >>>>> separate the compiler. This also applies to protobuf recipe.
> >>>>>
> >>>>>
> >>>> More precisely, this is the initial version outside of
> >> meta-virtualization
> >>>> which enea added
> >>>> in 2012 :) It was added as a dependency for criu (hence why it was put
> >> in
> >>>> meta-virt).
> >>>>
> >>>> If we move it to meta-oe, we at least owe that other implementation a
> >>>> reference in the commit.
> >>>>
> >>>>
> >>> s/if/when/.
> >>>
> >>> I'm happy to purge all the protobuf* recipes from meta-virt, since they
> >>> were only there as
> >>> support mechanisms (and I wasn't involved in their original merge).
> >>>
> >>> But if you can take a look at what's in the meta-virt recipe, I'll do
> >> some
> >>> runtime testing with your
> >>> variant here, and drop the meta-virt ones when I can confirm criu
> works.
> >>>
> >>
> >> Regular protobuf recipe is already part of meta-oe. That's why I added
> >> this one there too.
> >>
> >
> > Sure. That's obvious from the layer index, as is the existence of the one
> > in meta-virt.
>
> I didn 't take it from meta-virt. The original protobuf recipe is in
> meta-oe.
>

Yup, I wasn't implying that you did. Sorry if it came across that way, I
was just
pointing out the layer index, since it is a good place to check when
considering
a new recipe.

The ones in meta-virt need cleanup (and I'll drop them once this is in
meta-oe),
but that's something we could have tackled (literally) years ago if the
repetition
had been picked up.


>
> >
> > Credit, where credit is due. It's not my work, so I'm not asking for any
> > credit,
> > but simply duplicating something that already exists without a nod to the
> > older one
> > isn't ideal.
> >
> > All I was asking was that if you could a link to the meta-virt one in the
> > commit header
> > so that someone not familiar with the layer index can see the two options
> > .. and
> > at the same time I was wondering if you'd seen the meta-virt one and did
> > this one
> > differently for technical reasons. That makes it easier for me to drop
> > recipes as they
> > get cloned around to new (and better) locations.
>
> I am fine with that. Can you please provide the git hash so I can link it?
>
>
That's great, just a shout out something like:

----
replaces/obsoletes: commit 0d31f8448412 [Added protobuf-c criu dependency]
in meta-virtualization
----

Would be great, then we see the duplication, and why it was ever in
meta-virt in the first
place.

(the full commit log is below)


commit 0d31f844841249262fe142c4c03853dd1f647e2d

Author: David Nystr?m <david.nystrom at enea.com>

Date:   Thu Dec 6 10:54:55 2012 +0100


    Added protobuf-c criu dependency

    Signed-off-by: David Nystr?m <david.nystrom at enea.com>

Cheers,

Bruce


> Thanks in advance,
> Felipe
>
> --
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel at lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-devel
>
>


-- 
"Thou shalt not follow the NULL pointer, for chaos and madness await thee
at its end"