Physical constants

[MilesCranmer]

We could have a submodule that delivers common physical constants. Already I am finding myself having to type out the speed of light by hand; it would be useful if it were just built-in.

This could work as follows:

using DynamicQuantities
import DynamicQuantities: Units as U, Constants as C

Then, C would let you access constants, and U the units. This is similar to how astropy does it, to avoid mixing namespaces.

e.g., the energy of a 500nm photon

wavelength = 500*U.nm
energy = C.c*C.h/wavelength

This Constants namespace could also be exposed within the @u_str macro, so you could also write this as

energy = (1/500) * u"C.c*C.h/nm"

---

I'd be interested to know if @odow or @ChrisRackauckas have any thoughts or ideas in this direction.

[odow]

That seems very reasonable to me.

[j-fu]

Hi, how does this square with the physical constants defined in Unitful
or those in
https://github.com/JuliaPhysics/PhysicalConstants.jl?

Codes using these constants from different sources should be guaranteed to work together.
I think it might make sense to check this in CI...

[MilesCranmer]

Could you be more specific, I don't understand your question. PhysicalConstants.jl uses Unitful.jl unfortunately, so we cannot use it directly. The constants in PhysicalConstants.jl are taken from the same source as in #27 - from CODATA 2018. By CI do you mean to check the constants are equal?

[j-fu]

By CI do you mean to check the constants are equal?

Right. If you think it makes sense the PR would be on me...

[MilesCranmer]

I understand, that sounds good to me! Thanks.

[MilesCranmer]

Closed with #32