I have tests that I want to parameterize, but there are certain tests that should only be applied to one value of the parameters. To give a specific example, below, I would like to apply parameters one and two to test_A, but only supply parameter one to test_B. 
Current Code
@pytest.fixture(params=['one', 'two'])
def data(request):
    if request.param == 'one'
         data = 5
    return data
def test_A(data):
    assert True
def test_B(data):
    assert True
Desired Results
I basically want something that looks like this, but I can't figure out how to code this properly in pytest:
@pytest.fixture(params=['one', 'two'])
def data(request):
    data = 5
    return data
def test_A(data):
    assert True
@pytest.skipif(param=='two')
def test_B(data):
    assert True
                Building on your answer, you can check the input and call pytest.skip() if you don't want the test to run.
You can do the check in the test:
def test_A(data):
    assert True
def test_B(data):
    if data.param == 'two':
        pytest.skip()
    assert 'foo' == 'foo'
Or you could redefine the test fixture in a subclass:
class TestA:
    def test_A(self, data):
        assert True
class TestB:
    @pytest.fixture
    def data(self, data):
        if data.param == 'two':
            pytest.skip()
        return data
    def test_B(self, data):
        assert 'foo' == 'foo'
One other minor suggestion: your Data class can be replaced with a namedtuple, i.e.
import collections
Data = collections.namedtuple('Data', 'data, param')
                        I found a working solution, but I also welcome more solutions as this feels a tad "hacky":
class Data:
    def__init__(self, data, param):
        self.data = data
        self.param = param
@pytest.fixture(params=['one', 'two'])
def data(request):
    data = 5
    return Data(data, request.param)
def test_A(data):
    assert True
def test_B(data):
    if data.param == 'two':
        assert True
    else:
        assert 'foo' == 'foo'
                        If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With