Re: [PATCH 5/5] fake-metadata: updated to may_resolve() API

[Iago Toral <itoral igalia com>]

On Thu, 24 Feb 2011 12:55:17 +0100, "Juan A." Suárez Romero 
<jasuarez igalia com> wrote:
> On Tue, 2011-02-22 at 16:32 +0100, Guillaume Emont wrote:
> > ---
> >  src/fake-metadata/grl-fake-metadata.c |   41 
> > ++++++++++++++------------------
> >  1 files changed, 18 insertions(+), 23 deletions(-)
>
>
> This is not a question for Guillaume, but for everybody.
>
> Is it worth to keep fake-metadata in Grilo?
>
> It was some time ago, when we needed a fake metadata source to do 
> some
> tests.
>
> But right now, I think it is not worth to keep it, and we should 
> remove
> it.
>
> For the matter of tests, actually we should have a fake metadata but 
> in
> the unit tests, as a mockup.

I think a fake-metadata provider is interesting for testing things like 
the full-resolution stuff that Guillaume was doing these days but it is 
true that it should not live with the rest of the plugins any more, just 
let's do what you say and move it to the tests.

Iago