class TestMatrix2dEdit(_Base.TestValueEdit):
Widget = UsdQt.valueWidgets.Matrix2dEdit
SuccessValues = [Gf.Matrix2f(1.0, 2.0, 3.0, 4.0),
Gf.Matrix2d(1.0, 2.0, 3.0, 4.0)]
SuccessCastedValues = {(('1.0', '2.0'), ('3.0', '4.0')): Gf.Matrix2f(1.0, 2.0, 3.0, 4.0),
None: Gf.Matrix2d(0.0)}
KeySequences = {('1', '.', '0', QtCore.Qt.Key_Tab, '2', '.', '0', QtCore.Qt.Key_Tab,
'3', '.', '0', QtCore.Qt.Key_Tab, '4', '.', '0'): Gf.Matrix2f(1.0, 2.0, 3.0, 4.0), }
ValueErrorValues = [((1.0, None), (None, 1.0)), Gf.Matrix3f(), "(1.0, 2.0)"]
TypeErrorValues = []
def test_Matrices(self):
self.TestAttrOfType(Sdf.ValueTypeNames.Matrix2d,
Sdf.ValueTypeNames.Matrix2dArray, Gf.Matrix2d(0.0), 1e-12)
self.TestAttrOfType(Sdf.ValueTypeNames.Matrix3d,
Sdf.ValueTypeNames.Matrix3dArray, Gf.Matrix3d(0.0), 1e-12)
self.TestAttrOfType(Sdf.ValueTypeNames.Matrix4d,
Sdf.ValueTypeNames.Matrix4dArray, Gf.Matrix4d(0.0), 1e-12)
def test_Numpy(self):
'''Converting VtArrays to numpy arrays and back.'''
try:
import numpy
except ImportError:
# If we don't have numpy, just skip this test.
return
cases = [
dict(length=33,
fill=(1, 2, 3),
arrayType=Vt.Vec3hArray,
expLen=11,
expVal=Gf.Vec3h(1, 2, 3)),
dict(length=33,
fill=(1, 2, 3),
arrayType=Vt.Vec3fArray,
expLen=11,
expVal=Gf.Vec3f(1, 2, 3)),
dict(length=33,
fill=(1, 2, 3),
arrayType=Vt.Vec3dArray,
expLen=11,
expVal=Gf.Vec3d(1, 2, 3)),
dict(length=12,
fill=(1, 2, 3, 4),
arrayType=Vt.Vec4dArray,
expLen=3,
expVal=Gf.Vec4d(1, 2, 3, 4)),
dict(length=12,
fill=(1, 2, 3, 4),
arrayType=Vt.QuatdArray,
expLen=3,
expVal=Gf.Quatd(4, (1, 2, 3))),
dict(length=12,
fill=(1, 2, 3, 4),
arrayType=Vt.QuatfArray,
expLen=3,
expVal=Gf.Quatf(4, (1, 2, 3))),
dict(length=12,
fill=(1, 2, 3, 4),
arrayType=Vt.QuathArray,
expLen=3,
expVal=Gf.Quath(4, (1, 2, 3))),
dict(length=40,
fill=(1, 0, 0, 1),
arrayType=Vt.Matrix2fArray,
expLen=10,
expVal=Gf.Matrix2f()),
dict(length=40,
fill=(1, 0, 0, 1),
arrayType=Vt.Matrix2dArray,
expLen=10,
expVal=Gf.Matrix2d()),
dict(length=90,
fill=(1, 0, 0, 0, 1, 0, 0, 0, 1),
arrayType=Vt.Matrix3fArray,
expLen=10,
expVal=Gf.Matrix3f()),
dict(length=90,
fill=(1, 0, 0, 0, 1, 0, 0, 0, 1),
arrayType=Vt.Matrix3dArray,
expLen=10,
expVal=Gf.Matrix3d()),
dict(length=160,
fill=(1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1),
arrayType=Vt.Matrix4fArray,
expLen=10,
expVal=Gf.Matrix4f()),
dict(length=160,
fill=(1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1),
arrayType=Vt.Matrix4dArray,
expLen=10,
expVal=Gf.Matrix4d()),
dict(length=24,
fill=(4.25, 8.5),
arrayType=Vt.Range1dArray,
expLen=12,
expVal=Gf.Range1d(4.25, 8.5)),
dict(length=24,
fill=(4.25, 8.5),
arrayType=Vt.Range1dArray,
expLen=12,
expVal=Gf.Range1d(4.25, 8.5)),
dict(
length=24,
fill=(-1, -1, 1, 1),
arrayType=Vt.Range2dArray,
expLen=6,
expVal=Gf.Range2d((-1, -1), (1, 1)),
)
]
for case in cases:
Array, length, fill, expLen, expVal = (case['arrayType'],
case['length'],
case['fill'],
case['expLen'],
case['expVal'])
src = Vt.DoubleArray(length, fill)
result = Array.FromNumpy(numpy.array(src, copy=False))
self.assertEqual(len(list(result)), expLen)
self.assertTrue(all([x == expVal for x in result]), \
'%s != %s' % (list(result), [expVal]*len(list(result))))
# Formerly failed, now produces a 1-d length-1 array.
self.assertTrue(
Vt.Vec3dArray.FromNumpy(numpy.array([1, 2, 3])) == Vt.Vec3dArray([(
1, 2, 3)]))
with self.assertRaises((ValueError, TypeError)):
# Two dimensional, but second dimension only 2 (needs 3).
a = numpy.array([[1, 2], [3, 4]])
v = Vt.Vec3dArray(a)
# Some special-case empty array cases.
self.assertEqual(numpy.array(Vt.DoubleArray()).shape, (0, ))
self.assertEqual(numpy.array(Vt.Vec4fArray()).shape, (0, 4))
self.assertEqual(numpy.array(Vt.Matrix2dArray()).shape, (0, 2, 2))
self.assertEqual(numpy.array(Vt.Matrix2fArray()).shape, (0, 2, 2))
self.assertEqual(numpy.array(Vt.Matrix3dArray()).shape, (0, 3, 3))
self.assertEqual(numpy.array(Vt.Matrix3fArray()).shape, (0, 3, 3))
self.assertEqual(numpy.array(Vt.Matrix4dArray()).shape, (0, 4, 4))
self.assertEqual(numpy.array(Vt.Matrix4fArray()).shape, (0, 4, 4))
for C in (Vt.DoubleArray, Vt.Vec4fArray, Vt.Matrix3dArray,
Vt.Matrix4dArray):
self.assertEqual(C(numpy.array(C())), C())
# Support non-contiguous numpy arrays -- slicing numpy arrays is a
# convenient way to get these.
self.assertEqual(
Vt.FloatArray.FromNumpy(
numpy.array(Vt.FloatArray(range(33)))[::4]),
Vt.FloatArray(9, (0, 4, 8, 12, 16, 20, 24, 28, 32)))
self.assertEqual(
Vt.Vec3dArray.FromNumpy(
numpy.array(Vt.Vec3fArray([(1, 2, 3), (4, 5, 6),
(7, 8, 9)]))[::-2]),
Vt.Vec3dArray([(7, 8, 9), (1, 2, 3)]))
# Check that python sequences will convert to arrays as well.
self.assertEqual(Vt.DoubleArray([1, 2, 3, 4]),
Vt.DoubleArray((1, 2, 3, 4)))
# Formerly failed, now works, producing a linear Vec4dArray.
self.assertEqual(Vt.Vec4dArray.FromNumpy(numpy.array([1, 2, 3, 4])),
Vt.Vec4dArray([(1, 2, 3, 4)]))
# Simple 1-d double array.
da = Vt.DoubleArray(10, range(10))
self.assertEqual(Vt.DoubleArray(numpy.array(da)), da)
# Vec3dArray.
va = Vt.Vec3dArray(10, [(1, 2, 3), (2, 3, 4)])
self.assertEqual(Vt.Vec3dArray.FromNumpy(numpy.array(va)), va)
Gf.Quath(0, 0, 1, 0)]
],
"quatf": [
Gf.Quatf(0, 0, 0, 1),
[Gf.Quatf(1, 0, 0, 0),
Gf.Quatf(0, 1, 0, 0),
Gf.Quatf(0, 0, 1, 0)]
],
"quatd": [
Gf.Quatd(0, 0, 0, 1),
[Gf.Quatd(1, 0, 0, 0),
Gf.Quatd(0, 1, 0, 0),
Gf.Quatd(0, 0, 1, 0)]
],
"matrix2d": [
Gf.Matrix2d(1, 0, 0, 0),
[
Gf.Matrix2d(0, 1, 0, 0),
Gf.Matrix2d(0, 0, 2, 0),
Gf.Matrix2d(0, 0, 3, 0)
]
],
"matrix3d": [
Gf.Matrix3d(0, 0, 0, 0, 0, 0, 0, 0, 0),
[
Gf.Matrix3d(0, 0, 0, 0, 1, 0, 0, 0, 0),
Gf.Matrix3d(0, 0, 0, 0, 0, 0, 0, 0, 2),
Gf.Matrix3d(0, 0, 4, 0, 0, 0, 0, 0, 0)
]
],
"matrix4d": [