Accepting request 966568 from home:sbradnick:branches:devel:languages:python

- Adding _service file
- Cleaning up Source0 declaration since PyPI doesn't have a tarball
- Updating hash for 'accelerate/test' Source[1,2]
- Adding code to %prep section to deal w/ how _service pulls down
  files for the gzip'd tarball

- Update to 3.1.6
  * Pulled directly from https://github.com/mcfletch/pyopengl since
    pypi.org (via py2pack fetch) doesn't have this version available.
  * Seems that some packages which use PyOpenGL and accelerate have
    a soft requirement that they're the same version.

OBS-URL: https://build.opensuse.org/request/show/966568
OBS-URL: https://build.opensuse.org/package/show/devel:languages:python/python-opengl-accelerate?expand=0&rev=18

test_arraydatatypeaccel.py

@@ -3,12 +3,18 @@ from OpenGL.arrays import arraydatatype as adt
 from OpenGL.arrays import vbo
 from OpenGL import GL
 from OpenGL._bytes import integer_types
+import pytest
 try:
     import numpy 
 except ImportError:
     numpy = None
+try:
+    import OpenGL_accelerate
+except ImportError:
+    pytest.skip('Accelerate not installed, skipping', allow_module_level=True)
 
 class _BaseTest( object ):
+    array = None
     def setUp( self ):
         self.handler = adt.ArrayDatatype
         assert self.handler.isAccelerated
@@ -38,39 +44,41 @@ class _BaseTest( object ):
         p = self.handler.arrayToGLType( self.array )
         assert p == GL.GL_FLOAT
 
-if numpy:
-    # Skip if modifies the functions, which are *shared* between the 
-    # classes...
-    #@pytest.mark.skipif( not numpy, reason="Numpy not available")
-    class TestNumpy( _BaseTest, unittest.TestCase ):
-        def setUp( self ):
-            self.array = numpy.array( [[1,2,3],[4,5,6]],'f')
-            super(TestNumpy,self).setUp()
-        def test_dataPointer( self ):
-            p = self.handler.dataPointer( self.array )
-            assert isinstance( p, integer_types)
-            assert p == self.array.ctypes.data
-        def test_zeros( self ):
-            p = self.handler.zeros( (2,3,4), 'f' )
-            assert p.shape == (2,3,4)
-            assert p.dtype == numpy.float32
-        def test_asArrayConvert( self ):
-            p = self.handler.asArray( self.array, GL.GL_DOUBLE )
-            assert p is not self.array 
-            assert p.dtype == numpy.float64
-            p = self.handler.asArray( self.array, 'd' )
-            assert p is not self.array 
-            assert p.dtype == numpy.float64
-        def test_zeros_typed( self ):
-            z = self.handler.zeros( (2,3,4), GL.GL_FLOAT)
-            assert z.shape == (2,3,4)
-            assert z.dtype == numpy.float32
-        def test_downconvert( self ):
-            p = self.handler.asArray( numpy.array( [1,2,3],'d'), GL.GL_FLOAT )
-            assert p.dtype == numpy.float32
-        def test_zeros_small( self ):
-            z = self.handler.zeros( 0, GL.GL_BYTE )
-            assert z.dtype == numpy.byte, z
+# Skip if modifies the functions, which are *shared* between the 
+# classes...
+@pytest.mark.skipif( not numpy, reason="Numpy not available")
+class TestNumpy( _BaseTest, unittest.TestCase ):
+    def setUp( self ):
+        super(TestNumpy,self).setUp()
+        self.array = numpy.array( [[1,2,3],[4,5,6]],'f')
+        handler = adt.ArrayDatatype.getHandler( self.array )
+        handler.registerReturn( )
+
+    def test_dataPointer( self ):
+        p = self.handler.dataPointer( self.array )
+        assert isinstance( p, integer_types)
+        assert p == self.array.ctypes.data
+    def test_zeros( self ):
+        p = self.handler.zeros( (2,3,4), 'f' )
+        assert p.shape == (2,3,4)
+        assert p.dtype == numpy.float32
+    def test_asArrayConvert( self ):
+        p = self.handler.asArray( self.array, GL.GL_DOUBLE )
+        assert p is not self.array 
+        assert p.dtype == numpy.float64
+        p = self.handler.asArray( self.array, 'd' )
+        assert p is not self.array 
+        assert p.dtype == numpy.float64
+    def test_zeros_typed( self ):
+        z = self.handler.zeros( (2,3,4), GL.GL_FLOAT)
+        assert z.shape == (2,3,4)
+        assert z.dtype == numpy.float32
+    def test_downconvert( self ):
+        p = self.handler.asArray( numpy.array( [1,2,3],'d'), GL.GL_FLOAT )
+        assert p.dtype == numpy.float32
+    def test_zeros_small( self ):
+        z = self.handler.zeros( (0,), GL.GL_BYTE )
+        assert z.dtype == numpy.byte, z
 
 class TestVBO( _BaseTest, unittest.TestCase ):
     def setUp( self ):