On 18/11/13 16:46 +0000, Mark McLoughlin wrote:
On Mon, 2013-11-18 at 17:37 +0100, Julien Danjou wrote:
On Mon, Nov 18 2013, Mark McLoughlin wrote:
> I'm struggling to care about this until I have some insight into why
> it's important. And it's a bit frustrating to have to guess the
> rationale for this. Like commit messages, blueprints should be as much
> about the why as the what.
Sure. I ought to think that having an application that wants to leverage
oslo.messaging but is not using oslo.config and is retrieving its
parameter from another way is a good enough argument.
It's a theoretical benefit versus the very practical "design an API for
the use cases that are actually important to OpenStack projects".
I agree with Mark here. I don't think the discussion here should be
around whether not depending on oslo.config is the right thing. As far
as I can see, we all agree on the fact that libraries should have few
dependencies and they shouldn't force anything on the application
using them.
That being said, I think the discussion here should go around whether
this is the right time to make the change or not. The benefit of doing
it now is that we haven't released oslo.messaging yet, which means
there are no *strong* API requirements. However, I'd like us to
consider how that will help with the migration of existing projects
from oslo-rpc to oslo.messaging.
I think this is - or at least should be - our priority right now as
far as oslo.messaging is concerned.
Cheers,
FF
--
@flaper87
Flavio Percoco
_______________________________________________
OpenStack-dev mailing list
OpenStack-dev@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
