防止仅仅因为参数不同而“重复”测试代码。

Question

在我的测试套件中，我有不同的集成测试和稳定性测试。

例如,

@pytest.mark.integration
def test_integration_total_devices(settings, total_devices):
    assert total_devices == settings['integration']['nodes']['total']

@pytest.mark.stability
def test_stability_total_devices(settings, total_devices):
    assert total_devices == settings['stability']['nodes']['total']

正如您所注意到的，它是完全相同的代码，只是从配置中读取一个不同的参数。

如何防止这种重复代码的情况？设置的值是不同的，所以我不能只是：

@pytest.mark.integration
@pytest.mark.stability
def test_integration_total_devices(settings, total_devices):
    assert total_devices == settings['nodes']['total']

我忘了提一下(为了提醒我，谢谢@dzejdzej )，看来热测试参数化不起作用。当我想运行两个“标记”时，它可以工作，但是标记的目的是能够独立地运行其中一个标记的测试，例如pytest -m integration。然而，就我测试而言，每当我设置参数化时，它都会同时运行。

@pytest.mark.parametrize('type', (
    pytest.param('stability', marks=pytest.mark.stability),
    pytest.param('integration', marks=pytest.mark.integration),
))
@pytest.mark.integration
@pytest.mark.stability
def test_total_devices(settings, total_devices, type):
    assert total_devices == settings[type]['nodes']['total']

Answer 1

请看一看pytest参数化https://docs.pytest.org/en/latest/parametrize.html

应该可以这样做：

@pytest.mark.parametrize('area,total_devices', (
    pytest.param('stability', 10, marks=pytest.mark.stability),
    pytest.param('integration', 15, marks=pytest.mark.integration),
))
def test_integration_total_devices(area, total_devices):
    assert total_devices == settings.get(area)['nodes']['total']