[FEATURE] Parameterize testing

🚨🚨 Feature Request

• Related to an existing Issue
• A new implementation (Improvement, Extension)

Improve test coverage by evaluating multiple arguments.

For example, this function tests float32 (which should throw an exception) but should test other dtypes (uint16, unint32, int16,…).

def test_image():
    with pytest.raises(ValueError):
        image = Image((1920, 1080, 3), "float32")

If your feature will improve HUB

Makes Hub more robust by improving test coverage.

Description of the possible solution

pytest can pass multiple arguments to a test function:

test_inputs = [1, 2, 3]

@pytest.mark.parameterized("foo", test_inputs)
def test_stuff(foo):
   assert type(foo) == int

[madhucharan]

The current problem with this issue is when there is no support for other dtypes, It has been created and checked for errors if any other dtype is passed other than uint8,uint16. The PR is here. Now I have included support for all float and integer variants, We probably need to update the test cases to check the int and float variant tests and raise ValueError if any unsupported dtype is passed.

[davidbuniat]

If you added support for those dtypes, then you should switch testing against raising Error, now you need to test if you can create image with such dtypes (instead of checking if throws an error.)