Karthikeyan Singaravelan <[email protected]> added the comment:
A possible test case to trigger this code path based on
testUncheckedHashBasedPyc would be as below. check_hash_based_pycs can be
patched to be "always" so that the hash is validated using
_bootstrap_external._validate_hash_pyc. The difference between "state = old"
for pyc and "state = new" in py would ensure the hashes are different to raise
ImportError expected and the test picks up state = new as the updated source
code. Without fixing the typo trying to make hash validation would throw
NameError.
@unittest.mock.patch('_imp.check_hash_based_pycs', 'always')
def test_checked_hash_based_change_pyc(self):
source = b"state = 'old'"
source_hash = importlib.util.source_hash(source)
bytecode = importlib._bootstrap_external._code_to_hash_pyc(
compile(source, "???", "exec"),
source_hash,
False,
)
files = {TESTMOD + ".py": (NOW, "state = 'new'"),
TESTMOD + ".pyc": (NOW - 20, bytecode)}
def check(mod):
self.assertEqual(mod.state, 'new')
self.doTest(None, files, TESTMOD, call=check)
----------
nosy: +xtreak
_______________________________________
Python tracker <[email protected]>
<https://bugs.python.org/issue39033>
_______________________________________
_______________________________________________
Python-bugs-list mailing list
Unsubscribe:
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
