David Hewitt
David Hewitt
Some kind of option on this line seems very reasonable to me, thanks! FWIW we currently infer the seconds / microseconds split for timestamps - see https://github.com/pydantic/pydantic/issues/7940 and https://github.com/pydantic/speedate/issues/14 I...
I expect there will be some speedate changes, yes. There was previously https://github.com/pydantic/speedate/pull/54 but the work seems to have stalled.
Because `rust-numpy` is a separate crate to PyO3, we can't directly implement for `ndarray` types due to the orphan rule. You could probably use a newtype in your struct e.g....
> The solution for that is probably to remove the `impl From for PyClassInitializer` blanket and instead emit customized code via `#[pyclass]`. That could be done as a follow up....
For module traverse / clear - probably the discussion in #5663 is relevant (could be derived on the module state type).
👍 sounds good to me, thanks!
Sorry for the slow reply here. This doesn't really seem to be a specific PyO3 issue, can you please reduce this to more directly give a minimal repro using just...
Yes, that sounds correct. We might want to rework the script to be part of `noxfile.py`, I think at the time I had issues with the netlify build environment which...
Good question, #5551 produces a `__NEW__` slotdef which you should be able to use, I'm going to split that diff up. Feel free to either base off that PR or...
With #5551 merged, this should be able to rebase on top now.