fixture-as-parameter.md

import pytest

def test_data_1_is_string(data_1):
    """Test that data is string"""
    assert isinstance(data_1, str)

def test_data_2_is_string(data_2):
    """Test that data is string"""
    assert isinstance(data_2, str)

@pytest.mark.parametrize(
    'data',
    [data_1, data_2]
)
def test_data_is_string_parametrized(data):
    """Parametrized test that data is string"""
    assert isinstance(data, str)

@pytest.fixture()
def data_1():
    """Data fixture for data 1."""
    return 'DATA_1'

@pytest.fixture()
def data_2():
    """Data fixture for data 2."""
    return 'DATA_2'

Zachs-MBP:test zhammer$ pytest test.py 
============================= test session starts ==============================
platform darwin -- Python 3.6.0, pytest-3.3.1, py-1.5.2, pluggy-0.6.0
rootdir: /Users/zhammer/code/test, inifile:
collected 0 items / 1 errors                                                   

==================================== ERRORS ====================================
___________________________ ERROR collecting test.py ___________________________
test.py:14: in <module>
    [data_1, data_2]
E   NameError: name 'data_1' is not defined
!!!!!!!!!!!!!!!!!!! Interrupted: 1 errors during collection !!!!!!!!!!!!!!!!!!!!
=========================== 1 error in 0.12 seconds ============================

I've looked into the 'parametrizing fixtures', but it seems like this syntax:

@pytest.fixture(scope=​'session'​, params=[​'tiny'​, ​'mongo'​])

doesn't support params that are fixtures.

One option is the move the fixture functions above the tests, and simple execute each function in the parametrization to get its data, but this seems ugly / anti-pytest.

import pytest

@pytest.fixture()
def data_1():
    """Data fixture for data 1."""
    return 'DATA_1'

@pytest.fixture()
def data_2():
    """Data fixture for data 2."""
    return 'DATA_2'

def test_data_1_is_string(data_1):
    """Test that data is string"""
    assert isinstance(data_1, str)

def test_data_2_is_string(data_2):
    """Test that data is string"""
    assert isinstance(data_2, str)

@pytest.mark.parametrize(
    'data',
    [data_1(), data_2()]
)
def test_data_is_string_parametrized(data):
    """Parametrized test that data is string"""
    assert isinstance(data, str)

Is there any good way to do this?