[oe] Question about bluez4 and PROVIDES_append_angstrom

[Denys Dmytriyenko]

On Thu, Oct 22, 2009 at 09:24:07PM +0200, Marco Cavallini wrote:
> Hello,
> first of all I must say that I am not a OE expert and I'd like to learn
> and understand, so please apologize if my question is too silly for most
> of you :-)
> 
> I am quite confused about a few lines I found into the following files.
> The result of these lines faced me to errors using KaeilOS distro.
> 
> This modification looks (to me) related to bluez4 and Angstrom only and
> I wonder why some package recipes (bluez4*.bb) contains some lines that
> strictly ties them to Angstrom distribution breaking other distributions.

Can you please be more specific as to what is broken for your distribution?

> Files I found are the following:
> recipes/bluez/bluez4_4.35.bb:PROVIDES_append_angstrom =
>  " bluez-utils bluez-libs"
> recipes/bluez/bluez4_4.37.bb:PROVIDES_append_angstrom =
>  " bluez-utils bluez-libs"
> recipes/bluez/bluez4.inc:PROVIDES_append_angstrom =
>  " bluez-utils-dbus bluez-utils bluez-libs"
> 
> Lines are similar to these:
> # For angstrom we want this to replace at least bluez-libs
> PROVIDES_append_angstrom = " bluez-utils-dbus bluez-utils bluez-libs"
> RPROVIDES_bluez4-dev_angstrom = "bluez-libs-dev"
> 
> My question is: why these lines are not into angstrom*.bb distro recipe
> making bluez*.bb distro-independent (agnostic)?
> 
> Thank you in advance for your hints.

PROVIDES_append_angstrom is an override in bluez4 recipe specific to Angstrom. 
For other distributions it won't do anything. For Angstrom, it will _extend_ 
the default PROVIDES variable with extra values.

It's pretty standard to see small alterations for specific machines or distros 
in recipes, when most of the recipe is still common.

-- 
Denys