Question

"""confest.py file."""
import pytest


def pytest_addoption(parser):
    parser.addoption(
        "--cmdopt", action="store", default=None
    )
@pytest.fixture(scope="session", autouse=True)
def cmdopt(request):
    """Reading data from command prompt."""
    return request.config.getoption("--cmdopt")

@pytest.fixture(scope="session", autouse=True)
def do_something(cmdopt)
    """Would need to preprocess the needed one to make it parameterised."""
    return list_of_tuples

测试文件的结构是这样的

""" Test Suite File."""
import pytest

#do_something is in conftest file
@pytest.mark.parametrize("val1,val2", variables = do_something)
def test_case1(val1, val2):
    assert val1 == val2

当前代码实际上会引发错误，因为do_something中使用的conftest.py方法无法正常工作。我该如何纠正代码？

Answer 1

您不能使用固定装置来做到这一点；但是，您可以编写一个pytest_generate_tests挂钩，以参数化val1，val2测试参数。示例：

# conftest.py
import pytest


def pytest_addoption(parser):
    parser.addoption('--range', action='store', type=int, default=None)


def pytest_generate_tests(metafunc):
    if all(arg in metafunc.fixturenames for arg in ('val1', 'val2')):
        hi = metafunc.config.getoption('--range')
        if hi is not None:
            variables = [(i, i) for i in range(hi)]
        metafunc.parametrize("val1,val2", variables)

测试模块：

def test_case1(val1, val2):
    assert val1 == val2

以不同的--range号运行测试将产生不同数量的测试：

$ pytest -v --range 1
...
test_spam.py::test_case1[0-0] PASSED
...
$ pytest -v --range 5
...
test_spam.py::test_case1[0-0] PASSED
test_spam.py::test_case1[1-1] PASSED
test_spam.py::test_case1[2-2] PASSED
test_spam.py::test_case1[3-3] PASSED
test_spam.py::test_case1[4-4] PASSED
...

如何在pytest中为参数化方法传递预处理变量？

1 个答案: