[osg-users] OSX tessellator callback options.

Robert Osfield robert.osfield at gmail.com
Wed Oct 3 14:39:19 PDT 2007

Hi All,

In a attempt to fix build problems under OSX we've really been round
the houses.  First up we have Andy Skinner reporting build problem
with the Cmake build under OSX with the original form of tesselator
callback under Apple - that was building with (...).

Next it looks like going for () is fine, in keeping with other
platforms.  This gets checked in works for Andy, works for others too.

Later with have Adrian reporting an error with this change and the
requirement back from (...).

Then others too report just a warning and no error.

Then others reporting error on CMake XCode build again.

For all the effort on this I don't yet have a clear picture of what is
going on.  Clearly the number of different ways to build the OSG under
OSX is killing us right now, oh well actually is stems from Apple
habit of taking things that are done in a standard way and twisting oh
so a little just to make things pain for no gain.

So to get a clearer picture I need some direct feedback on particular question:

  A) CMake Makefile build - does it require 1. (...) or  2. ()
  B) CMake Xcode build - does it require 1. (...) or 2. ()
  C) Xcode build - does it require 1. (...) or 2. ()

>From the various traffic, often seemingly contridictory, it looks like:

  A requires 2.
  B. requires 1.
  C. requires 1.

I say A. 2 because this is what I believe Andy Skinner is using and
the reason why we embarked on this whole journey.  However, we have
had a report tonight that actually A requires 1. too, suggesting there
are even more variables at play, if so exactly what.

So please can users provide some feedback and telll us exactly what
build combination you are using under OSX.  Could you also copy and
past the contents of the glu.h relating to the tessellator callback.


More information about the osg-users mailing list