Hi Ulrich,

I feel the appropriate thing to do with this code is to remove the
assert and remove the target parameter completely.  The function in
question is only ever called from the gluBuild3DMipmaps function which
really should only be called from 3D textures.

Personally I'm not in favour of assert, much preferring checks for
validity to be done at compile where possible, of invalid values might
occur at runtime that one always checks for them so am happy to recode
to avoid use of assert, or simply removing them or replace them with
something that isn't just done on a debug build.

Robert.

On 30 June 2013 13:39, Ulrich Hertlein <[email protected]> wrote:
> Hi Robert,
>
> One of the latest compiler warning clean-ups broke the debug build on OS X in
> src/osg/glu/libutil/mipmap.c because of an 'assert(target == ...'.
>
> Index: src/osg/glu/libutil/mipmap.cpp
> ===================================================================
> --- src/osg/glu/libutil/mipmap.cpp      (revision 13596)
> +++ src/osg/glu/libutil/mipmap.cpp      (working copy)
> @@ -7427,10 +7427,12 @@
>
>
>  static void closestFit3D(GLTexImage3DProc gluTexImage3D,
> -                         GLenum /*target*/, GLint width, GLint height, GLint 
> depth,
> +                         GLenum target, GLint width, GLint height, GLint 
> depth,
>                           GLint internalFormat, GLenum format, GLenum type,
>                           GLint *newWidth, GLint *newHeight, GLint *newDepth)
>  {
> +    (void)target; // To avoid 'unused' warning in release
> +
>  #if !defined(OSG_GLES1_AVAILABLE) && !defined(OSG_GLES2_AVAILABLE)
>     GLint widthPowerOf2= nearestPower(width);
>     GLint heightPowerOf2= nearestPower(height);
>
> Cheers,
> /ulrich
>
> _______________________________________________
> osg-submissions mailing list
> [email protected]
> http://lists.openscenegraph.org/listinfo.cgi/osg-submissions-openscenegraph.org
>
_______________________________________________
osg-submissions mailing list
[email protected]
http://lists.openscenegraph.org/listinfo.cgi/osg-submissions-openscenegraph.org

Reply via email to