Alex Gaynor
Alex Gaynor
IMO it should be a classmethod on `Hash`, to make it clear what it's doing. An instance-method `digsest` just invites questions about how it interacts with the state.
(extremely untested, generated with zed's agent stuff + some basic cleanup)
@woodruffw a) what am I supposed to do about this datetime lint thing? b) are you up for giving this a review to see if it looks like a vaguely...
I feel like it'd be helpful to be able to get that table in CI. I'm developing on a mac where the docker based harnesses are not going to run.
grumble, so this doesn't even fail correctly in openssl
I think the comment has all the info I want, I think it'd be more readable if the table was inverted though: one table per test case, with the table...
oh my god, apparently pyopenssl still uses internals ðŸ˜. Looks like we have `ASN1_STRING_get0_data` and `ASN1_STRING_length` exposed, I'll go fix there.
https://github.com/pyca/pyopenssl/pull/1462 has the fix -- will mark this for the next milestone and expect to rebase/merge after our next pyopenssl release.
I'm pretty ambivalent about this. We have a few choices: 1) Simply implement `__eq__` for our own hashes, don't change the contract (all objects always have `__eq__`). 2) Add `__eq__`...
That can break existing applications, to the extent they're relying on the default `object.__eq__` behavior!