pytestify 1.5.0

Last updated:

0 purchases

pytestify 1.5.0 Image
pytestify 1.5.0 Images
Add to Cart

Description:

pytestify 1.5.0

pytestify


A tool to automatically change unittest to pytest. Similar to
unittest2pytest,
but with a few more features and written using AST and tokenize, rather
than lib2to3.
Big thanks to pyupgrade, which
this project has learned from.
Installation
pip install pytestify
Usage
pytestify path/to/file.py
or
pytestify path/to/folder/
Optional arguments

--keep-method-casing
--with-count-equal

Please read over all changes that pytestify makes. It's a new
package, so there are bound to be issues.
Implemented features
Test class names
Remove TestCase parent class, and make sure tests start with Test. We are keeping the test classes themselves, but you can remove them manually.
class TestThing(unittest.TestCase): # class TestThing:
class TestThing(TestCase, ClassB): # class TestThing(ClassB):
class ThingTest(unittest.TestCase): # class TestThing:
class Thing(unittest.TestCase): # class TestThing:

Setup / teardowns
def setUp(self): # def setup_method(self):
def tearDown(self): # def teardown_method(self):
def setUpClass(self): # def setup_class(self):
def tearDownClass(self): # def teardown_class(self):

Asserts
Rewrite unittest assert methods using the assert keyword.
# asserting one thing
self.assertTrue(a) # assert a
self.assertFalse (a) # assert not a
self.assertIsNone(a) # assert a is None
self.assertIsNotNone(a) # assert a is not None

# asserting two things
self.assertEqual(a, b) # assert a == b
self.assertNotEqual(a, b) # assert a != b
self.assertIs(a, b) # assert a is b
self.assertIsNot(a, b) # assert a is not b
self.assertIn(a, b) # assert a in b
self.assertNotIn(a, b) # assert a not in b
self.assertListEqual(a, b) # assert a == b
self.assertDictEqual(a, b) # assert a == b
self.assertSetEqual(a, b) # assert a == b
self.assertGreater(a, b) # assert a > b
self.assertLess(a, b) # assert a < b
self.assertGreaterEqual(a, b) # assert a >= b
self.assertLessEqual(a, b) # assert a <= b
self.assertRegex(a, b) # assert a.search(b)
self.assertNotRegex(a, b) # assert not a.search(b)

self.assertAlmostEqual(a, b)
# assert a == pytest.approx(b)
self.assertAlmostEqual(a, b, places=2)
# assert a == pytest.approx(b, abs=0.01)
self.assertAlmostEquals(a, b, delta=2)
# assert a == pytest.approx(b, abs=2)


# improves the assert if reasonable
self.assertEqual(a, None) # assert a is None
self.assertEqual(a, True) # assert a is True


# error messages
self.assertTrue(a, msg='oh no!') # assert a, 'oh no!'

Multi-line asserts
Since assert (a == b, 'err') is equivalent to asserting a tuple, and thus is always True.
self.assertEqual( # assert a == \
a, # b
b,
)

self.assertEqual( # assert a == \
a, # b, \
b, # 'oh no!'
msg='oh no!'
)

camelCase to snake_case
Disable this behavior with --keep-method-casing
def testThing(self): # def test_thing(self):
def testHTTPThing(self): # def test_httpthing(self):

assertCountEqual
The assertCountEqual rewrite is risky, so opt-in with pytest path/to/file --with-count-equal.
self.assertItemsEqual(a, b) # assert sorted(a) == sorted(b)
self.assertCountEqual(a, b) # assert sorted(a) == sorted(b)

Note that pytest has no version of either of these methods. See
this thread for more
information. You can also use
unittest's implementation.
Exceptions
self.assertRaises(OSError) # pytest.raises(OSError)
self.assertWarns(OSError) # pytest.warns(OSError)
with self.assertRaises(OSError) as e: # with pytest.raises(OSError) as e
with self.assertWarns(OSError) as e: # with pytest.warns(OSError) as e

Skipping / Expecting failure
# decorated
@unittest.skip('some reason') # @pytest.mark.skip('some reason')
@unittest.skipIf(some_bool) # @pytest.mark.skipif(some_bool)
@unittest.skipUnless(some_bool) # @pytest.mark.skipif(not some_bool)
@unittest.expectedFailure # @pytest.mark.xfail

# not decorated
unittest.skip('some reason') # pytest.skip('some reason')
unittest.skipTest('some reason') # pytest.skip('some reason')
unittest.fail('some reason') # pytest.fail('some reason')

License:

For personal and professional use. You cannot resell or redistribute these repositories in their original state.

Customer Reviews

There are no reviews.