search

SunPower / PVMismatch

An explicit Python PV system IV & PV curve trace calculator which can also calculate mismatch.
http://sunpower.github.io/PVMismatch/
BSD 3-Clause "New" or "Revised" License
78 stars 28 forks source link

Files missing from contrib.gen_coeffs? #114

Closed cwhanse closed 4 years ago

cwhanse commented 4 years ago

example.py has references to gen_coeffs.PVMODULES, gen_coeffs.gen_two_diode, and other gen_xxx functions. Am I missing something, or is PVMismatch missing something?

mikofski commented 4 years ago

Sorry, they're in contrib/gen_coeffs/__init__.py

cwhanse commented 4 years ago

Thanks. Is it typical to put code in the init.py file? I didn't think to look there.

mikofski commented 4 years ago

Sorry, I'm not sure what's Pythonic. I think it's okay if they're constants and methods that apply to the entire sub-package, but I may be going rogue. I've learned a lot since PVMismatch was started like 7 years ago, so maybe I wouldn't do that again. It just irks me that the dunder module has to be there, but is just empty doing nothing, seems like a waste, so I usually try to use it somehow, but maybe that's wrong?

I definitely think it's okay to use __init__.py for creating or maintaining an API to sub-modules and packages beneath it the way pvlib does now, or to contain __version__ and __author__ information if it's at the top of the package, for use in setup.py and for users to extract for meta info.

cwhanse commented 4 years ago

I'm not sure what's Pythonic.

Neither am I. "wrong" is a strong word.

Since you brought up python conventions, what is the appetite for migrating towards PEP8? e.g. function names from capital letters to lower case, etc.

mikofski commented 4 years ago

Ugh, so many conflicts. I think this is a @chetan201 question, but IMO we should make this code black, even if it has to go through deprecations or break code. Probably shouldn't be too painful since there are probably so few users.

cwhanse commented 4 years ago

If it's OK to really depart from the current release, my project would support an overhaul of PVMismatch. Just tossing the idea out there. In the end our projects need a computational engine for mismatched IV curve calculations, that interfaces readily with pvlib, and which will persist for a long time (we'll re-use it).

mikofski commented 4 years ago

Hi Cliff, maybe change the title of this issue to, "Move code from init.py, use better name" or perhaps consider closing it?