[MPlayer-dev-eng] [PATCH] cross-compile: directfb version detection

[Diego Biurrun]

On Wed, Oct 12, 2005 at 07:59:25PM +0200, Aurelien Jacobs wrote:
> On Wed, 12 Oct 2005 11:39:43 +0200
> Diego Biurrun <diego at biurrun.de> wrote:
> 
> > On Mon, Oct 10, 2005 at 12:01:53AM +0200, Aurelien Jacobs wrote:
> > 
> > > @@ -3459,7 +3459,14 @@
> > >  }
> > >  EOF
> > >    if cc_check $_inc_directfb -ldirectfb && "$TMPO" >> "$TMPLOG" ;
> > >    then
> > > -    _directfb_version=`"$TMPO"`
> > > +   cat > $TMPC <<EOF
> > > +#include <directfb_version.h>
> > > +int
> > > +dfb_ver = DIRECTFB_MAJOR_VERSION DIRECTFB_MINOR_VERSION
> > > DIRECTFB_MICRO_VERSION +;
> > > +EOF
> > > +   if $_cc -E $TMPC $_inc_directfb > "$TMPO"; then
> > > +    _directfb_version=`sed -n 's/^dfb_ver[^1-9]*\(.*\)/\1/p'
> > > "$TMPO" | tr -d '() '`
> > >      _def_directfb_version="#define DIRECTFBVERSION
> > >      $_directfb_version" if test "$_directfb_version" -ge 913; then
> > >  	_res_comment="$_directfb_version"
> > 
> > There is a test for this very thing above, what about it?
> 
> The test above allows to verify if we can compile and link against
> directfb, but we can't rely on it to detect the version, since we
> can't assume we will be able to run the binary (remember we are
> talking about cross-compiling ;-)
> That's why I added a second test which use pre-processor only to
> get version.
> In fact, the first test could be modified as it don't need to
> report the version anymore. But I think it is still a good test
> to verify if directfb is available, so we can keep it as is.

Yes, you should modify that test.  Can't you just merge both tests into
one?

> > > @@ -3468,9 +3475,13 @@
> > >  	_directfb=no
> > >  	_res_comment="version >=0.9.13 required"
> > >      fi
> > > -  else
> > > +   else
> > 
> > cosmetics
> 
> It sounds like. But it's not really...
> In fact it's just a diff misbehavior.

No.  Either you fiddled with the whitespace or your editor did.  Maybe
some tabs to spaces (or vice versa) conversion?

Diego