Minor unit confusion, dipole moments

[andersx]

Just a minor inconsistency: In the data.rst file, moments are listed as given in atomic units. while in the data_notes.rst the unit is listed as Debye. I think it should be [Debye/Å^L] in both places.

https://cclib.github.io/data.html vs. https://cclib.github.io/data_notes.html#moments

[langner]

@berquist if this will take more than a few days, could we move the issue to the main repo? I'd like to close down the issues here, since it's an automatically updated repo.

[berquist]

Yes, I'll need more time to look into it.

[bhavaygg]

So, is the correct unit [Debye/Å^L] ?

[langner]

I think the short is answer is yes, but it's not as simple as that.

The long answer is that "L" refers to the exponent over 1/Å, which grows with the rank of the moment. So, there is not one unit for moments. For dipole L=0 and it will be Debye, for quadrupoles it will be L=1 which is the Buckingham, and so on. Which means that maybe it should actually be L-1 depending on where we start counting (the first element is just the reference point).

The even longer answer is that we don't really have good tests for moments at this point... I think the SP unit tests has a zero dipole. So it'd be nice to improve there while settling this matter as well.

And, #89 is relevant to this discussion as well.