def test_apply_to_mat_column_transform_expand(mat):
# transform to list of tuples
mat = [tuple(row) for row in mat]
n_rows = len(mat)
unique_n_cols = set(map(len, mat))
assert len(unique_n_cols) == 1
n_cols = unique_n_cols.pop()
col_idx = random.randrange(0, n_cols)
mat_t = apply_to_mat_column(mat,
col_idx,
lambda x: (x, x.lower(), x.upper()),
expand=True)
assert n_rows == len(mat_t)
for orig, trans in zip(mat, mat_t):
assert len(orig) == len(trans) - 2
before, x, after = orig[:col_idx + 1], orig[col_idx], orig[col_idx +
1:]
before_t, x_t, after_t = trans[:col_idx + 1], trans[col_idx:col_idx +
3], trans[col_idx +
3:]
assert before == before_t
assert after == after_t
assert len(x_t) == 3
assert x == x_t[0]
assert x.lower() == x_t[1]
assert x.upper() == x_t[2]
def test_apply_to_mat_column_identity(mat, col_idx):
identity_fn = lambda x: x
# transform to list of tuples
mat = [tuple(row) for row in mat]
n_rows = len(mat)
if n_rows > 0: # make sure the supplied matrix is not ragged
unique_n_cols = set(map(len, mat))
assert len(unique_n_cols) == 1
n_cols = unique_n_cols.pop()
else:
n_cols = 0
if n_rows == 0 or (n_rows > 0 and n_cols == 0) or col_idx < 0 or col_idx >= n_cols:
with pytest.raises(ValueError):
apply_to_mat_column(mat, col_idx, identity_fn)
else:
assert _mat_equality(mat, apply_to_mat_column(mat, col_idx, identity_fn))
def test_apply_to_mat_column_transform(mat):
# transform to list of tuples
mat = [tuple(row) for row in mat]
n_rows = len(mat)
unique_n_cols = set(map(len, mat))
assert len(unique_n_cols) == 1
n_cols = unique_n_cols.pop()
col_idx = random.randrange(0, n_cols)
mat_t = apply_to_mat_column(mat, col_idx, lambda x: x.upper())
assert n_rows == len(mat_t)
for orig, trans in zip(mat, mat_t):
assert len(orig) == len(trans)
for x, x_t in zip(orig, trans):
assert x.upper() == x_t.upper()