Should marion_diffuse use diffuse versions of martin_ruiz and schlick?

[markcampanelli]

Describe the bug

While working on #2050, it occurred to me that pvlib.iam.marion_diffuse refers to the direct component IAM functions for martin_ruiz and schlick, i.e., it would seem that these two lines should instead refer to pvlib.iam.martin_ruiz_diffuse and pvlib.iam.schlick_diffuse

To Reproduce
Steps to reproduce the behavior:

1. Inspect the lines at: https://github.com/pvlib/pvlib-python/blob/main/pvlib/iam.py#L630-L631

Expected behavior
I would expect the diffuse IAM models to be used, because marion_diffuse computes diffuse components.

Screenshots
None

Versions:

• pvlib.__version__: v0.11,0
• python: 3.11.7

Additional context
If someone could verify, then I would be happy to fix this as part of #2050.

[echedey-ls]

If I understand correctly, that diffuse IAM uses direct IAM models to compute a diffuse one via integration. So it sounds good to me. It uses this function: https://github.com/pvlib/pvlib-python/blob/13ef8f4e3f07a6b1c5acd7014a2617beeb10867c/pvlib/iam.py#L647C1-L653C34

[markcampanelli]

If I understand correctly, that diffuse IAM uses direct IAM models to compute a diffuse one via integration. So it sounds good to me. It uses this function: https://github.com/pvlib/pvlib-python/blob/13ef8f4e3f07a6b1c5acd7014a2617beeb10867c/pvlib/iam.py#L647C1-L653C34

I think I understand now: pvlib.iam.martin_ruiz_diffuse and pvlib.iam.schlick_diffuse are analogous to pvlib.iam.marion_diffuse, but with a slightly different output interface. I was wondering why one would need a separate IAM function for direct vs. diffuse 🤪. Thanks @echedey-ls!

[markcampanelli]

Closing due to misunderstanding of code. (Github's issue-closing interface is a bit wierd!)

[markcampanelli]

Really, close this.