def test_combine_points(self):
op = categories.CountCategories()
(width, height, depth) = (1, 1, 3)
screen = (width, height)
glyphset = _GlyphsetShim(depth)
out = op.allocate(glyphset, screen)
glyph = [0, 0, 1, 1]
existing = op.allocate(glyphset, screen)
op.combine(existing, glyph, ShapeCodes.POINT, 1)
expected = np.zeros((height, width, depth))
expected[0, 0, 1] = 1
self.assertTrue(np.array_equal(existing, expected))
(width, height, depth) = (3, 4, 3)
screen = (width, height)
glyphset = _GlyphsetShim(depth)
glyph = [2, 1, 3, 2]
existing = op.allocate(glyphset, screen)
op.combine(existing, glyph, ShapeCodes.POINT, 2)
expected = np.zeros((height, width, depth), dtype=np.int)
expected[1, 2, 2] = 1
self.assertTrue(np.array_equal(existing, expected))
def test(self):
"""Can the output of categories.CountCategories can be
transformed to vanilla counts?"""
op = categories.ToCounts()
(width, height, depth) = (4, 5, 3)
screen = (width, height)
glyphset = _GlyphsetShim(depth)
aggregator = categories.CountCategories()
aggs = aggregator.allocate(glyphset, screen)
aggs.fill(1)
out = op.shade(aggs)
expected = np.empty((height, width), dtype=np.int32)
expected.fill(depth)
self.assertTrue(np.array_equal(out, expected))
self.assertEquals((height, width), out.shape, "Unexpected out shape")
shape = aggs.shape
aggs = np.arange(0, reduce(operator.mul, shape))
aggs = aggs.reshape((depth, height, width)).transpose((1, 2, 0))
out = op.shade(aggs)
self.assertEquals((height, width), out.shape, "Unexpected out shape")
self.assertEquals(out[0, 0], 0+20+40)
self.assertEquals(out[4, 0], 16+36+56)
self.assertEquals(out[0, 3], 3+23+43)
self.assertEquals(out[4, 3], 19+39+59)
def test_allocate(self):
op = categories.CountCategories()
(width, height, depth) = (6, 3, 7)
screen = (width, height)
glyphset = _GlyphsetShim(depth)
out = op.allocate(glyphset, screen)
expected = np.zeros((height, width, depth))
self.assertTrue(np.array_equal(out, expected))
def test_rollup(self):
op = categories.CountCategories()
(width, height, depth) = (8, 5, 3)
ones = np.ones((depth, height, width))
zeros = np.zeros((depth, height, width))
twos = np.empty((depth, height, width))
twos.fill(2)
threes = np.empty((depth, height, width))
threes.fill(3)
out = op.rollup(ones, zeros)
self.assertTrue(np.array_equal(out, ones))
out = op.rollup(ones, ones)
self.assertTrue(np.array_equal(out, twos))
out = op.rollup(twos, ones)
self.assertTrue(np.array_equal(out, threes))
def test_combine_rect(self):
op = categories.CountCategories()
(width, height, depth) = (3, 4, 2)
screen = (width, height)
glyphset = _GlyphsetShim(depth)
glyph = [0, 0, 2, 2]
existing = op.allocate(glyphset, screen)
op.combine(existing, glyph, ShapeCodes.RECT, 0)
expected = np.array([[[1, 1, 0],
[1, 1, 0],
[0, 0, 0],
[0, 0, 0]],
[[0, 0, 0],
[0, 0, 0],
[0, 0, 0],
[0, 0, 0]]], dtype=np.int).transpose((1, 2, 0))
self.assertEqual(existing.shape, expected.shape)
self.assertTrue(np.array_equal(existing, expected))
def test_uniform(self):
op5 = categories.MinPercent(.5)
op1 = categories.MinPercent(.1)
(width, height, depth) = (4, 6, 4)
screen = (width, height)
glyphset = _GlyphsetShim(depth)
aggregator = categories.CountCategories()
aggs = aggregator.allocate(glyphset, screen)
aggs.fill(2)
above = np.empty((height, width, 4), dtype=np.uint8)
above[:] = op5.above
out = op1.shade(aggs)
self.assertTrue(np.array_equal(out, above))
below = np.empty((height, width, 4), dtype=np.uint8)
below[:] = op5.below
out = op5.shade(aggs)
self.assertTrue(np.array_equal(out, below))
def test(self):
op0 = categories.Select(0)
op1 = categories.Select(1)
op2 = categories.Select(2)
(width, height, depth) = (5, 4, 3)
screen = (width, height)
glyphset = _GlyphsetShim(depth)
aggregator = categories.CountCategories()
aggs = aggregator.allocate(glyphset, screen)
aggs[:, :, 0] = 1
aggs[:, :, 1] = 2
aggs[:, :, 2] = 3
expected = np.empty((height, width, depth), dtype=np.int)
expected[:, :, 0] = 1
expected[:, :, 1] = 2
expected[:, :, 2] = 3
self.assertTrue(np.array_equal(op0.shade(aggs), expected[:, :, 0]))
self.assertTrue(np.array_equal(op1.shade(aggs), expected[:, :, 1]))
self.assertTrue(np.array_equal(op2.shade(aggs), expected[:, :, 2]))