Lori A. Burns

Ok, yes, I just ran a full tests on today's master, and it's clean. (And the Windows issue is fixed, too.)

No, guess=read is specifically disabled https://github.com/psi4/psi4/blob/master/psi4/driver/p4util/python_helpers.py#L451 . As to how to run the calc, there's ways to deposit and collect scratch files in a qcng.compute run, and that could be...

Are these rdkit-specific? I'm thinking they're a little too specific for the general molecule class. Have you reconsidered since first opening the PR, or would they still be useful, @dr-marsmm?

Agreed, I need to add the pydantic

The conda package v0.26.0 as of an hour ago works in the presence of pydantic v1 or v2. It's still using the v1 API, so I'll leave this open for...

@awvwgk, your `Union[int, float]` construction looked right to me, but upon testing, it was casting values like 3.00001 to 3 in an unpredictable way. I switched the multiplicities to being...

> Why would you have an error for fractional multiplicities when you can pass fractional charges with validation error? "... without validation error"? To be clear, mult=3.0 is fine; only...

Sorry for the delay. So long as what you need is unvalidated pass-through of non-int chg and mult, I think all this PR needs is an extra message for charge...

Ok, https://github.com/loriab/QCElemental/commit/342d7cc4adea5367d950dc04d2324946def009db is my current proposal for allowing float multiplicity. It's sitting atop #343 for now. Investigation showed that the validation routine does handle float charges and checking internal consistency...

Thanks for all your work on this so far, @Lnaden. I've updated through psi4 with pydantic v2 API and hit a couple serialization bugs/changes that I hope you can either...