scottprahl / pypolar Goto Github PK

I think it's a mistake, but my energy to dig into it is flagging.
My concern was aroused when I was getting values of ts that were >1, and noticing that the figure in the docs "Similar to Hecht Figure 4.41" doesn't really look like that figure. (It is 4.41 in my edition, but in a newer edition it's 4.49) I wondered if the difference is immaterial; the amplitudes obey "conservation laws" that are not as simple as r + t = 1, so I tried to make it all work. I ran out of energy. I believe there is a mistake, but I can't unambiguously convince myself that it is so.

fresnel.t_per_amplitude has
ts = 2 * d / (m/n_i)/ (c + d)
which doesn't look correct to me.

while sym_fresnel.t_per_amplitude has
ts = 2 * c / (c + d)
which looks correct to me.

Thanks for this package. It has saved me a lot of time.
-gapster

Am I right that pypolar depends on serial? I had strange issue installing pypolar using conda skeleton and serial from conda-forge. I don't know exactly what or why, but after installing pycolor and then serial my conda was rendered useless., throwing lots of errors. Guessing that the problem lies with the installation of serial, I will ask ...

How does pypolar use serial, and more to the point, can pypolar run without serial? If so, which modules use serial?

As I said, I'm guessing what caused the problem. and since it's a bit hard to see what serial does for pypolar, I'm starting there as a work-around.

I think it would be nice if one could draw the Poincaré sphere. As an example the library py_pol:
https://bitbucket.org/optbrea/py_pol/src/master/py_pol/drawings.py

Which is unfortunately GPL.