Re: [proaudio] libflashsupport-jack with V4L1 removed

[Dominique Michel <dominique.michel@xxxxxxxxx>]

Le Thu, 13 Oct 2011 21:56:42 +0200,
Goran Mekić <meka@xxxxxxxxxx> a écrit :

> On 10/12/11 20:13, Dominique Michel wrote:
> > Le Wed, 12 Oct 2011 11:36:58 +0200,
> > Goran Mekic<meka@xxxxxxxxxx>  a écrit :
> >
> >> Hello,
> >>       This is my first post to this list, so excuse me if I did
> >> something wrong. Attached are two files,
> >> libflashsupport-jack-9999.ebuild.patch is output of "svn diff", and
> >> the other file is the patch removing V4L1 support. Is this the
> >> right format to submit patches? Regrets!
> >
> > Thanks for the patches.
> >
> > And yes, I think that it is OK with those patches. I will try to
> > apply them later. The good thing with patches is that it is easier
> > and faster to review the changes.
> >
> > You changed the KEYWORD from "" to ~amd64. This is a live ebuild
> > (as indicate revision .9999). We do respect the gentoo policy
> > regarding the KEYWORD. Is is explained somewhere in the doc (ebuild
> > _howto or the devman I guess) than ~arch is for ebuilds that need
> > more testing, and "" (no keyword) is for ebuilds that cannot be
> > tested. Typically, an ebuild that cannot be tested is a live ebuild
> > that will download a source code that can change at any time. So
> > the original KEYWORD was the correct one. You must write ** into
> > your package.keywords.
> >
> > ~arch will be OK for a new ebuild that download an archive that
> > doesn't change with time. And in that case, it is better to specify
> > both ~amd64 and ~x86 even if you can test only one of those arch. We
> > are an overlay with limited man power and we don't open bugs in
> > order to test the different architectures. It is specific cases
> > where only x86 will work, but with most of the ebuilds, it will be
> > no problem to specify the 2 archs, even if you can test only one of
> > them. Well, this is my opinion, but I am not the only one here.
> >
> > Also, with a loopback, you wont need libflashsupport:
> > http://alsa.opensrc.org/Jack_and_Loopback_device_as_Alsa-to-Jack_bridge
> > The flash player wont appear in qjackctl's graph. You will get a
> > cloop and ploop instead (for capture and playback loops), and flash
> > will connect to them trough ALSA. The plus is that you can connect
> > and disconnect the cloop and the ploop without to crash the ALSA
> > program. You can even stop and restart the jack server without
> > problem, the sound will be back when you restart JACK. And you will
> > also get an outstanding stability.
> >
> > Ciao,
> > Dominique
> >
> >>
> >> --
> >> FreeB(eer)S(ex)D(rugs) are the real daemons
> >>
> >
> 	It's a dead project, and an ebuild was "standing" in the
> overlay, so I just patched it and used it for quite a while, so I
> thought it's stable enough to get ~amd64. I do not have ~x86 to test,
> so I've omitted it. As there is no tar ball around, I thought it's ok
> to keyword it. I can change that and send another patch or something?
> 
> 	Thanx a lot for the loopback link.
> 
> 
I added your patch. And also an info with a link to the loopback.

-- 
"We have the heroes we deserve."