pytest-rerunfailures是用于当我们的测试用例由于某些原因，进行请求的再次尝试
-
安装
pip install pytest-rerunfailures -
案例
通过@pytest.mark.flaky 进行失败重试
- reruns 请求尝试的次数
- reruns_delay 间隔的时间
import pytest @pytest.mark.parametrize('a,b,result', [ (1, 1, 3), (2, 2, 4), (100, 100, 200), (0.1, 0.1, 0.2), (-1, -1, -2) ], ids=['int', 'int', 'bignum', 'float', 'fushu']) # 参数化 @pytest.mark.flaky(reruns=6, reruns_delay=2) def test_add(a, b, result): # cal = Calculator() assert result == a + b if __name__ == '__main__': # "--reruns 5 --reruns-delay 1", pytest.main(['-vs', "rerunfailures_demo.py"])
@pytest.mark.flaky(reruns=6, reruns_delay=2)当失败后我们尝试6次，每次间隔2秒，结果如下collecting ... collected 5 items rerunfailures_demo.py::test_add[int0] RERUN rerunfailures_demo.py::test_add[int0] RERUN rerunfailures_demo.py::test_add[int0] RERUN rerunfailures_demo.py::test_add[int0] RERUN rerunfailures_demo.py::test_add[int0] RERUN rerunfailures_demo.py::test_add[int0] RERUN rerunfailures_demo.py::test_add[int0] FAILED rerunfailures_demo.py::test_add[int1] PASSED rerunfailures_demo.py::test_add[bignum] PASSED rerunfailures_demo.py::test_add[float] PASSED rerunfailures_demo.py::test_add[fushu] PASSED ================================== FAILURES =================================== _______________________________ test_add[int0] ________________________________ a = 1, b = 1, result = 3 @pytest.mark.parametrize('a,b,result', [ (1, 1, 3), (2, 2, 4), (100, 100, 200), (0.1, 0.1, 0.2), (-1, -1, -2) ], ids=['int', 'int', 'bignum', 'float', 'fushu']) # 参数化 @pytest.mark.flaky(reruns=6, reruns_delay=2) def test_add(a, b, result): # cal = Calculator() > assert result == a + b E assert 3 == 2 E +3 E -2 rerunfailures_demo.py:26: AssertionError =========================== short test summary info =========================== FAILED rerunfailures_demo.py::test_add[int0] - assert 3 == 2 ==================== 1 failed, 4 passed, 6 rerun in 12.20s ====================
正常情况下一条用例如果有多条断言，一条断言失败了，其他断言就不会执行了，而使用pytest-assume可以继续执行下面的断言
-
安装
pip install pytest-assume -
案例
import pytest def test_assume(): print('登录操作') pytest.assume(1 == 2) print('搜索操作') pytest.assume(2 == 2) print('加购操作') pytest.assume(3 == 2) if __name__ == '__main__': pytest.main(["-vs", "assume_demo.py"])
-
运行结果
collecting ... collected 1 item assume_demo.py::test_assume 登录操作 搜索操作 加购操作 FAILED ================================== FAILURES =================================== _________________________________ test_assume _________________________________ tp = <class 'pytest_assume.plugin.FailedAssumption'>, value = None, tb = None def reraise(tp, value, tb=None): try: if value is None: value = tp() if value.__traceback__ is not tb: > raise value.with_traceback(tb) E pytest_assume.plugin.FailedAssumption: E 2 Failed Assumptions: E E assume_demo.py:17: AssumptionFailure E >> pytest.assume(1 == 2) E AssertionError: assert False E E assume_demo.py:21: AssumptionFailure E >> pytest.assume(3 == 2) E AssertionError: assert False C:\python_tools\python3\lib\site-packages\six.py:718: FailedAssumption =========================== short test summary info =========================== FAILED assume_demo.py::test_assume - pytest_assume.plugin.FailedAssumption: ============================== 1 failed in 0.10s ==============================
正常情况下，用例默认执行顺序是自上而下的，对于一些有上下文依赖关系的用例，可是通过 pytest-ordering 来设置执行顺序，当然，通过setup、teardown和fixture来解决也是可以的
-
安装
pip install pytest-ordering -
案例
import pytest @pytest.mark.run(order=2) def test_foo(): assert True @pytest.mark.run(order=1) def test_bar(): assert True
使用该插件可以标记一个testcase作为其他testcase的依赖，当依赖项执行失败时，那些依赖它的test将会被跳过。
-
安装
pip install pytest-dependency -
案例
使用方法： 用 @pytest.mark.dependency()对所依赖的方法进行标记，使用@pytest.mark.dependency(depends=["test_name"])引用依赖,test_name可以是多个。
import pytest @pytest.mark.dependency() def test_01(): assert False @pytest.mark.dependency(depends=["test_01"]) def test_02(): print("执行测试2")
-
运行结果
collecting ... collected 2 items dependency_demo.py::test_01 FAILED dependency_demo.py::test_02 SKIPPED (test_02 depends on test_01) ================================== FAILURES =================================== ___________________________________ test_01 ___________________________________ @pytest.mark.dependency() def test_01(): > assert False E assert False dependency_demo.py:17: AssertionError =========================== short test summary info =========================== FAILED dependency_demo.py::test_01 - assert False ======================== 1 failed, 1 skipped in 0.11s =========================
- 平常我们功能测试用例非常多时，比如有1千条用例，假设每个用例执行需要1分钟，如果单个测试人员执行需要1000分钟才能跑完
- 当项目非常紧急时，会需要协调多个测试资源来把任务分成两部分，于是执行时间缩短一半，如果有10个小伙伴，那么执行时间就会变成十分之一，大大节省了测试时间
- 为了节省项目测试时间，10个测试同时并行测试，这就是一种分布式场景
分布式执行用例的原则：
-
用例之间是独立的，没有依赖关系，完全可以独立运行用例执行没有顺序要求，随机顺序都能正常执行每个用例都能重复运行，运行结果不会影响其他用例
-
安装
pip install pytest-xdis -
案例
pytest -n 2 (2代表2个CPU)，pytest -n auto
- nauto：可以自动检测到系统的CPU核数；从测试结果来看，检测到的是逻辑处理器的数量，即假12核使用auto等于利用了所有CPU来跑用例，此时CPU占用率会特别高
import pytest @pytest.mark.parametrize('a,b,result', [ (1, 1, 3), (2, 2, 4), (100, 100, 200), (0.1, 0.1, 0.2), (-1, -1, -2) ], ids=['int', 'int', 'bignum', 'float', 'fushu']) # 参数化 @pytest.mark.flaky(reruns=6, reruns_delay=2) def test_add(a, b, result): # cal = Calculator() assert result == a + b if __name__ == '__main__': # "--reruns 5 --reruns-delay 1", pytest.main(['-vs', "rerunfailures_demo.py", "-n 2"])
-
结果
plugins: allure-pytest-2.9.45, assume-2.4.3, dependency-0.5.1, forked-1.4.0, html-3.1.1, metadata-1.11.0, ordering-0.6, rerunfailures-10.2, xdist-2.5.0 gw0 I / gw1 I [gw0] win32 Python 3.8.9 cwd: C:\source_code\python_work\python-tutorial\pytests\plug_in_unit [gw1] win32 Python 3.8.9 cwd: C:\source_code\python_work\python-tutorial\pytests\plug_in_unit [gw0] Python 3.8.9 (tags/v3.8.9:a743f81, Apr 6 2021, 14:02:34) [MSC v.1928 64 bit (AMD64)] [gw1] Python 3.8.9 (tags/v3.8.9:a743f81, Apr 6 2021, 14:02:34) [MSC v.1928 64 bit (AMD64)] gw0 [5] / gw1 [5] scheduling tests via LoadScheduling rerunfailures_demo.py::test_add[int1] rerunfailures_demo.py::test_add[int0] [gw1] PASSED rerunfailures_demo.py::test_add[int1] rerunfailures_demo.py::test_add[float] [gw1] PASSED rerunfailures_demo.py::test_add[float] rerunfailures_demo.py::test_add[fushu] [gw1] PASSED rerunfailures_demo.py::test_add[fushu] [gw0] RERUN rerunfailures_demo.py::test_add[int0] rerunfailures_demo.py::test_add[int0] [gw0] RERUN rerunfailures_demo.py::test_add[int0] rerunfailures_demo.py::test_add[int0] [gw0] RERUN rerunfailures_demo.py::test_add[int0] rerunfailures_demo.py::test_add[int0] [gw0] RERUN rerunfailures_demo.py::test_add[int0] rerunfailures_demo.py::test_add[int0] [gw0] RERUN rerunfailures_demo.py::test_add[int0] rerunfailures_demo.py::test_add[int0] [gw0] RERUN rerunfailures_demo.py::test_add[int0] rerunfailures_demo.py::test_add[int0] [gw0] FAILED rerunfailures_demo.py::test_add[int0] rerunfailures_demo.py::test_add[bignum] [gw0] PASSED rerunfailures_demo.py::test_add[bignum] ================================== FAILURES =================================== _______________________________ test_add[int0] ________________________________ [gw0] win32 -- Python 3.8.9 C:\python_tools\python3\python.exe a = 1, b = 1, result = 3 @pytest.mark.parametrize('a,b,result', [ (1, 1, 3), (2, 2, 4), (100, 100, 200), (0.1, 0.1, 0.2), (-1, -1, -2) ], ids=['int', 'int', 'bignum', 'float', 'fushu']) # 参数化 @pytest.mark.flaky(reruns=6, reruns_delay=2) def test_add(a, b, result): # cal = Calculator() > assert result == a + b E assert 3 == 2 E +3 E -2 rerunfailures_demo.py:26: AssertionError =========================== short test summary info =========================== FAILED rerunfailures_demo.py::test_add[int0] - assert 3 == 2 ==================== 1 failed, 4 passed, 6 rerun in 13.41s ==================== Process finished with exit code 0