superfluous usage of "INC_PR" in oe-core?

superfluous usage of "INC_PR" in oe-core?

[Robert P. J. Day]

  given my understanding of the rationale for INC_PR, it would seem as
if at least *some* of these assignments are superfluous:

  recipes-connectivity/connman/connman.inc:INC_PR = "r20"
  recipes-graphics/xorg-driver/xorg-driver-common.inc:INC_PR = "r21"
  recipes-graphics/xorg-xserver/xserver-xorg.inc:INC_PR = "r8"
  recipes-graphics/xorg-font/font-alias_1.0.3.bb:PR = "${INC_PR}.3"
  recipes-graphics/xorg-font/xorg-font-common.inc:INC_PR = "r2"
  recipes-graphics/xorg-font/encodings_1.0.5.bb:PR = "${INC_PR}.1"
  recipes-graphics/xorg-app/xorg-app-common.inc:INC_PR = "r8"
  recipes-kernel/linux/linux-yocto.inc:INC_PR = "r4"

rday

Re: [OE-core] superfluous usage of "INC_PR" in oe-core?

[Richard Purdie]

On Sun, 2020-05-03 at 13:07 -0400, Robert P. J. Day wrote:
>   given my understanding of the rationale for INC_PR, it would seem
> as
> if at least *some* of these assignments are superfluous:
> 
>   recipes-connectivity/connman/connman.inc:INC_PR = "r20"
>   recipes-graphics/xorg-driver/xorg-driver-common.inc:INC_PR = "r21"
>   recipes-graphics/xorg-xserver/xserver-xorg.inc:INC_PR = "r8"
>   recipes-graphics/xorg-font/font-alias_1.0.3.bb:PR = "${INC_PR}.3"
>   recipes-graphics/xorg-font/xorg-font-common.inc:INC_PR = "r2"
>   recipes-graphics/xorg-font/encodings_1.0.5.bb:PR = "${INC_PR}.1"
>   recipes-graphics/xorg-app/xorg-app-common.inc:INC_PR = "r8"
>   recipes-kernel/linux/linux-yocto.inc:INC_PR = "r4"

This is a bit of a historical relic, I've sent a patch to clean it up.

Cheers,

Richard

Re: [OE-core] superfluous usage of "INC_PR" in oe-core?

[Robert P. J. Day]

On Tue, 5 May 2020, Richard Purdie wrote:

> On Sun, 2020-05-03 at 13:07 -0400, Robert P. J. Day wrote:
> >   given my understanding of the rationale for INC_PR, it would seem
> > as
> > if at least *some* of these assignments are superfluous:
> >
> >   recipes-connectivity/connman/connman.inc:INC_PR = "r20"
> >   recipes-graphics/xorg-driver/xorg-driver-common.inc:INC_PR = "r21"
> >   recipes-graphics/xorg-xserver/xserver-xorg.inc:INC_PR = "r8"
> >   recipes-graphics/xorg-font/font-alias_1.0.3.bb:PR = "${INC_PR}.3"
> >   recipes-graphics/xorg-font/xorg-font-common.inc:INC_PR = "r2"
> >   recipes-graphics/xorg-font/encodings_1.0.5.bb:PR = "${INC_PR}.1"
> >   recipes-graphics/xorg-app/xorg-app-common.inc:INC_PR = "r8"
> >   recipes-kernel/linux/linux-yocto.inc:INC_PR = "r4"
>
> This is a bit of a historical relic, I've sent a patch to clean it up.

  just to be clear, while there's not much value to INC_PR anymore,
it's still officially supported, yes? (there are a small number of
occurrences in meta-openembedded as well.)

rday

Re: [OE-core] superfluous usage of "INC_PR" in oe-core?

[Richard Purdie]

On Tue, 2020-05-05 at 08:19 -0400, Robert P. J. Day wrote:
> On Tue, 5 May 2020, Richard Purdie wrote:
> 
> > On Sun, 2020-05-03 at 13:07 -0400, Robert P. J. Day wrote:
> > >   given my understanding of the rationale for INC_PR, it would
> > > seem
> > > as
> > > if at least *some* of these assignments are superfluous:
> > > 
> > >   recipes-connectivity/connman/connman.inc:INC_PR = "r20"
> > >   recipes-graphics/xorg-driver/xorg-driver-common.inc:INC_PR =
> > > "r21"
> > >   recipes-graphics/xorg-xserver/xserver-xorg.inc:INC_PR = "r8"
> > >   recipes-graphics/xorg-font/font-alias_1.0.3.bb:PR =
> > > "${INC_PR}.3"
> > >   recipes-graphics/xorg-font/xorg-font-common.inc:INC_PR = "r2"
> > >   recipes-graphics/xorg-font/encodings_1.0.5.bb:PR =
> > > "${INC_PR}.1"
> > >   recipes-graphics/xorg-app/xorg-app-common.inc:INC_PR = "r8"
> > >   recipes-kernel/linux/linux-yocto.inc:INC_PR = "r4"
> > 
> > This is a bit of a historical relic, I've sent a patch to clean it
> > up.
> 
>   just to be clear, while there's not much value to INC_PR anymore,
> it's still officially supported, yes? (there are a small number of
> occurrences in meta-openembedded as well.)

It doesn't really have "support", its a convention. It can be used just
like any other intermediate variable. That said its a rather dated one
which isn't really appropriate now so we can probably remove the
documentation.conf entry. I've not looked at what the manuals say...

Cheers,

Richard