def test_div(self, field_def, div_def):
size = np.random.random(3) + 1.0
cells = np.random.randint(8, 32, (3))
nodes = grid_filters.coordinates0_point(cells, size)
my_locals = locals() # needed for list comprehension
field = np.stack([
np.broadcast_to(eval(f, globals(), my_locals), cells)
for f in field_def
],
axis=-1)
field = field.reshape(
tuple(cells) + ((3, 3) if len(field_def) == 9 else (3, )))
div = np.stack([
np.broadcast_to(eval(c, globals(), my_locals), cells)
for c in div_def
],
axis=-1)
if len(div_def) == 3:
div = div.reshape(tuple(cells) + ((3, )))
else:
div = div.reshape(tuple(cells))
assert np.allclose(div, grid_filters.divergence(size, field))
def test_add_divergence(self, default, shape):
if shape == 'vector':
default.add_calculation('#F#[:,:,0]', 'x', '1', 'just a vector')
if shape == 'tensor':
default.add_calculation('#F#[:,:,:]', 'x', '1', 'just a tensor')
x = default.place('x')
default.add_divergence('x')
in_file = default.place('divergence(x)')
in_memory = grid_filters.divergence(
default.size, x.reshape(tuple(default.cells) +
x.shape[1:])).reshape(in_file.shape)
assert (in_file == in_memory).all()