Re: ANNOUNCEMENT: meta-python2 has been created

[Adrian Bunk <bunk@stusta.de>]

On Mon, Dec 02, 2019 at 08:10:17PM -0800, Tim Orling wrote:
> On Mon, Dec 2, 2019 at 8:07 AM Martin Jansa <martin.jansa@gmail.com> wrote:
> 
> > Thanks
> >
> > It would be nice to delete all branches except master. Only master branch
> > makes sense now for meta-python2, right?
> >
> > " The stable branches were also kept, to make it easier to refer back to
> prior history
> and allow developers to use meta-python2 only in their workflow."
> 
> I can see use cases where a developer may want to replace meta-python with
> meta-python2 and still be on a "stable" release. It also makes reviewing
> what existed in which release much easier with cgit or git log --oneline,
> which I do VERY frequently. If this is contentious then the OE-TSC will
> make the decision.
>...

Martin beat me since I was about to make the same comment.

Adding a layer with contents that in the best case duplicates what is 
already in another layer (and in the worst case is broken) only creates 
confusion.

Replacing meta-python with meta-python2 sounds like a weird usecase,
and LAYERDEPENDS from layers like meta-multimedia or meta-networking
usually do not even make it an option to not include meta-python.

In practice the most common case where meta-python2 would be included in 
a zeus or earlier distribution would be when someone includes both the 
meta-python and meta-python2 layers.

cu
Adrian