[oe] [PATCH 02/12] qt4-embedded: added directfb plugin

[Koen Kooi]

On 03-06-09 19:36, Tom Rini wrote:
> On Wed, Jun 03, 2009 at 06:53:01PM +0200, Koen Kooi wrote:
>> On 03-06-09 18:00, Tom Rini wrote:
>>> On Wed, Jun 03, 2009 at 06:37:12PM +0300, Ihar Hrachyshka wrote:
>>>> Signed-off-by: Ihar Hrachyshka<ihar.hrachyshka at gmail.com>
>>>> ---
>>>>    recipes/qt4/qt4-embedded.inc |    4 ++--
>>>>    1 files changed, 2 insertions(+), 2 deletions(-)
>>>>
>>>> diff --git a/recipes/qt4/qt4-embedded.inc b/recipes/qt4/qt4-embedded.inc
>>>> index bae52ab..3d986b2 100644
>>>> --- a/recipes/qt4/qt4-embedded.inc
>>>> +++ b/recipes/qt4/qt4-embedded.inc
>>>> @@ -3,7 +3,7 @@ SECTION = "libs"
>>>>    LICENSE = "GPL QPL"
>>>>    PRIORITY = "optional"
>>>>    HOMEPAGE = "http://www.trolltech.com"
>>>> -DEPENDS += "tslib"
>>>> +DEPENDS += "directfb tslib"
>>>>    INC_PR = "r11"
>>>
>>> So, here's a problem.  We both don't allow use flags and we force
>>> directfb stuff to be done in foo-directfb (libxine / synaesthesia seems
>>> to be the exception to this rule) rather than the main foo package.
>>
>> Isn't this just creating a qt-gfxdriver-directfb.so and not impose
>> runtime deps on directfb in ${PN}?
>
> But how much stuff is now going to pick up "Oh, we have directfb" ?  Or
> are you sure everything is already passing --disable-directfb that might
> default to auto?

That would be bugs in other recipes, and fixing those is out of scope 
for this patchset.

regards,

Koen