import io import os import sys import pickle import subprocess import unittest from unittest.case import _Outcome from .support import LoggingResult, ResultWithNoStartTestRunStopTestRun class TestCleanUp(unittest.TestCase): def testCleanUp(self): class TestableTest(unittest.TestCase): def testNothing(self): pass test = TestableTest('testNothing') self.assertEqual(test._cleanups, []) cleanups = [] def cleanup1(*args, **kwargs): cleanups.append((1, args, kwargs)) def cleanup2(*args, **kwargs): cleanups.append((2, args, kwargs)) test.addCleanup(cleanup1, 1, 2, 3, four='hello', five='goodbye') test.addCleanup(cleanup2) self.assertEqual(test._cleanups, [(cleanup1, (1, 2, 3), dict(four='hello', five='goodbye')), (cleanup2, (), {})]) self.assertTrue(test.doCleanups()) self.assertEqual(cleanups, [(2, (), {}), (1, (1, 2, 3), dict(four='hello', five='goodbye'))]) def testCleanUpWithErrors(self): class TestableTest(unittest.TestCase): def testNothing(self): pass test = TestableTest('testNothing') outcome = test._outcome = _Outcome() exc1 = Exception('foo') exc2 = Exception('bar') def cleanup1(): raise exc1 def cleanup2(): raise exc2 test.addCleanup(cleanup1) test.addCleanup(cleanup2) self.assertFalse(test.doCleanups()) self.assertFalse(outcome.success) ((_, (Type1, instance1, _)), (_, (Type2, instance2, _))) = reversed(outcome.errors) self.assertEqual((Type1, instance1), (Exception, exc1)) self.assertEqual((Type2, instance2), (Exception, exc2)) def testCleanupInRun(self): blowUp = False ordering = [] class TestableTest(unittest.TestCase): def setUp(self): ordering.append('setUp') if blowUp: raise Exception('foo') def testNothing(self): ordering.append('test') def tearDown(self): ordering.append('tearDown') test = TestableTest('testNothing') def cleanup1(): ordering.append('cleanup1') def cleanup2(): ordering.append('cleanup2') test.addCleanup(cleanup1) test.addCleanup(cleanup2) def success(some_test): self.assertEqual(some_test, test) ordering.append('success') result = unittest.TestResult() result.addSuccess = success test.run(result) self.assertEqual(ordering, ['setUp', 'test', 'tearDown', 'cleanup2', 'cleanup1', 'success']) blowUp = True ordering = [] test = TestableTest('testNothing') test.addCleanup(cleanup1) test.run(result) self.assertEqual(ordering, ['setUp', 'cleanup1']) def testTestCaseDebugExecutesCleanups(self): ordering = [] class TestableTest(unittest.TestCase): def setUp(self): ordering.append('setUp') self.addCleanup(cleanup1) def testNothing(self): ordering.append('test') def tearDown(self): ordering.append('tearDown') test = TestableTest('testNothing') def cleanup1(): ordering.append('cleanup1') test.addCleanup(cleanup2) def cleanup2(): ordering.append('cleanup2') test.debug() self.assertEqual(ordering, ['setUp', 'test', 'tearDown', 'cleanup1', 'cleanup2']) class Test_TextTestRunner(unittest.TestCase): """Tests for TextTestRunner.""" def setUp(self): # clean the environment from pre-existing PYTHONWARNINGS to make # test_warnings results consistent self.pythonwarnings = os.environ.get('PYTHONWARNINGS') if self.pythonwarnings: del os.environ['PYTHONWARNINGS'] def tearDown(self): # bring back pre-existing PYTHONWARNINGS if present if self.pythonwarnings: os.environ['PYTHONWARNINGS'] = self.pythonwarnings def test_init(self): runner = unittest.TextTestRunner() self.assertFalse(runner.failfast) self.assertFalse(runner.buffer) self.assertEqual(runner.verbosity, 1) self.assertEqual(runner.warnings, None) self.assertTrue(runner.descriptions) self.assertEqual(runner.resultclass, unittest.TextTestResult) def test_multiple_inheritance(self): class AResult(unittest.TestResult): def __init__(self, stream, descriptions, verbosity): super(AResult, self).__init__(stream, descriptions, verbosity) class ATextResult(unittest.TextTestResult, AResult): pass # This used to raise an exception due to TextTestResult not passing # on arguments in its __init__ super call ATextResult(None, None, 1) def testBufferAndFailfast(self): class Test(unittest.TestCase): def testFoo(self): pass result = unittest.TestResult() runner = unittest.TextTestRunner(stream=io.StringIO(), failfast=True, buffer=True) # Use our result object runner._makeResult = lambda: result runner.run(Test('testFoo')) self.assertTrue(result.failfast) self.assertTrue(result.buffer) def testRunnerRegistersResult(self): class Test(unittest.TestCase): def testFoo(self): pass originalRegisterResult = unittest.runner.registerResult def cleanup(): unittest.runner.registerResult = originalRegisterResult self.addCleanup(cleanup) result = unittest.TestResult() runner = unittest.TextTestRunner(stream=io.StringIO()) # Use our result object runner._makeResult = lambda: result self.wasRegistered = 0 def fakeRegisterResult(thisResult): self.wasRegistered += 1 self.assertEqual(thisResult, result) unittest.runner.registerResult = fakeRegisterResult runner.run(unittest.TestSuite()) self.assertEqual(self.wasRegistered, 1) def test_works_with_result_without_startTestRun_stopTestRun(self): class OldTextResult(ResultWithNoStartTestRunStopTestRun): separator2 = '' def printErrors(self): pass class Runner(unittest.TextTestRunner): def __init__(self): super(Runner, self).__init__(io.StringIO()) def _makeResult(self): return OldTextResult() runner = Runner() runner.run(unittest.TestSuite()) def test_startTestRun_stopTestRun_called(self): class LoggingTextResult(LoggingResult): separator2 = '' def printErrors(self): pass class LoggingRunner(unittest.TextTestRunner): def __init__(self, events): super(LoggingRunner, self).__init__(io.StringIO()) self._events = events def _makeResult(self): return LoggingTextResult(self._events) events = [] runner = LoggingRunner(events) runner.run(unittest.TestSuite()) expected = ['startTestRun', 'stopTestRun'] self.assertEqual(events, expected) def test_pickle_unpickle(self): # Issue #7197: a TextTestRunner should be (un)pickleable. This is # required by test_multiprocessing under Windows (in verbose mode). stream = io.StringIO("foo") runner = unittest.TextTestRunner(stream) for protocol in range(2, pickle.HIGHEST_PROTOCOL + 1): s = pickle.dumps(runner, protocol) obj = pickle.loads(s) # StringIO objects never compare equal, a cheap test instead. self.assertEqual(obj.stream.getvalue(), stream.getvalue()) def test_resultclass(self): def MockResultClass(*args): return args STREAM = object() DESCRIPTIONS = object() VERBOSITY = object() runner = unittest.TextTestRunner(STREAM, DESCRIPTIONS, VERBOSITY, resultclass=MockResultClass) self.assertEqual(runner.resultclass, MockResultClass) expectedresult = (runner.stream, DESCRIPTIONS, VERBOSITY) self.assertEqual(runner._makeResult(), expectedresult) def test_warnings(self): """ Check that warnings argument of TextTestRunner correctly affects the behavior of the warnings. """ # see #10535 and the _test_warnings file for more information def get_parse_out_err(p): return [b.splitlines() for b in p.communicate()] opts = dict(stdout=subprocess.PIPE, stderr=subprocess.PIPE, cwd=os.path.dirname(__file__)) ae_msg = b'Please use assertEqual instead.' at_msg = b'Please use assertTrue instead.' # no args -> all the warnings are printed, unittest warnings only once p = subprocess.Popen([sys.executable, '_test_warnings.py'], **opts) out, err = get_parse_out_err(p) self.assertIn(b'OK', err) # check that the total number of warnings in the output is correct self.assertEqual(len(out), 12) # check that the numbers of the different kind of warnings is correct for msg in [b'dw', b'iw', b'uw']: self.assertEqual(out.count(msg), 3) for msg in [ae_msg, at_msg, b'rw']: self.assertEqual(out.count(msg), 1) args_list = ( # passing 'ignore' as warnings arg -> no warnings [sys.executable, '_test_warnings.py', 'ignore'], # -W doesn't affect the result if the arg is passed [sys.executable, '-Wa', '_test_warnings.py', 'ignore'], # -W affects the result if the arg is not passed [sys.executable, '-Wi', '_test_warnings.py'] ) # in all these cases no warnings are printed for args in args_list: p = subprocess.Popen(args, **opts) out, err = get_parse_out_err(p) self.assertIn(b'OK', err) self.assertEqual(len(out), 0) # passing 'always' as warnings arg -> all the warnings printed, # unittest warnings only once p = subprocess.Popen([sys.executable, '_test_warnings.py', 'always'], **opts) out, err = get_parse_out_err(p) self.assertIn(b'OK', err) self.assertEqual(len(out), 14) for msg in [b'dw', b'iw', b'uw', b'rw']: self.assertEqual(out.count(msg), 3) for msg in [ae_msg, at_msg]: self.assertEqual(out.count(msg), 1) def testStdErrLookedUpAtInstantiationTime(self): # see issue 10786 old_stderr = sys.stderr f = io.StringIO() sys.stderr = f try: runner = unittest.TextTestRunner() self.assertTrue(runner.stream.stream is f) finally: sys.stderr = old_stderr def testSpecifiedStreamUsed(self): # see issue 10786 f = io.StringIO() runner = unittest.TextTestRunner(f) self.assertTrue(runner.stream.stream is f)