-
Notifications
You must be signed in to change notification settings - Fork 15
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Update to the latest c-code template #73
Conversation
Add a regression test for CFLAGS not having -Ofast, which is known to break things. See zopefoundation/meta#155 for reference.
Hm, according to coveralls the new test does not run: https://coveralls.io/builds/52359729/source?filename=src/zodbpickle/tests/test_compile_flags.py |
Good catch! That would be because
and, for tox -e coverage,
(And similar things in buildout.cfg and GitHub Actions, most of them by way of custom test commands in I'll see what I can do (probably rename test_pickle_{2,3}.py so their file names don't start with test, after all they're included in the test suite explicitly inside test_pickle.py). |
Rename test_pickle_{2,3}.py because these are not supposed to be collected automatically: they use syntax incompatible with Python 3 and 2 respectively. Get rid of the -m zodbpickle.tests.test_pickle$ hack that was used to avoid the aforementioned syntax errors, and was also preventing the new test_compile_flags.py from being executed. Get rid of unittest.makeSuite() because I hear it's getting deprecated.
Wiped some caches:
Retried the two failing jobs with debug logging. Found cp310 and cp311 wheels in the Linux-pip-pypy-3.7 cache.
Wiped it
Retried again. |
Linux-pip-2.7 cache also has things like
At this point I think every single cache is poisoned with bad cp310 and cp311 wheels, so
I would love to know how that sort of thing happens. |
Another cache existed: Linux-pip-3.11.0-beta.3.
Are there more?
None of them start with "Linux-pip-" |
Theory: when you add a new Python version to the build matrix, there's no Linux-pip-3.10 cache, so it uses a compatible Linux-pip-2.7 or something, builds a bunch of new wheels, and then instead of creating Linux-pip-3.10, updates Linux-pip-2.7. Nah, can't be right, that would mean the new cache for Linux-pip-3.10 wouldn't ever get created. Could it be that the cache action both updates the old compatible cache and creates the new one? Would be strange to me... |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM.
Thank you for making the test setup way easier!
Add a regression test for CFLAGS not having -Ofast, which is known to break things. See zopefoundation/meta#155 for reference.