reorder function arguments to put variables first and parameters second

[wholmgren]

I think we should reorder a handful of function arguments to put the most variable-like arguments first and most parameter-like arguments last. I'm guessing that a lot people will think that I'm just screwing with the API for no reason, but I do have a few reasons!

1. consistency is important (and underrated)
2. we can make better use of default arguments.

Here are the functions that I know about so far:

ashraeiam(b, aoi) ---> ashraeiam(aoi, b=0.05)

# updated to reflect Cliff's comment below
physicaliam(K, L, n, aoi) ---> physicaliam(aoi, n=1.526, K=4., L=0.002)

# see #200
singlediode(module, photocurrent, saturation_current, resistance_series, resistance_shunt, nNsVth) --->
singlediode(photocurrent, saturation_current, resistance_series, resistance_shunt, nNsVth)

# I just noticed that the matlab sapm api is very different. # see #198
sapm(module, poa_direct, poa_diffuse, temp_cell, airmass_absolute, aoi) ---> 
sapm(poa_direct, poa_diffuse, temp_cell, airmass_absolute, aoi, module)

snlinverter(inverter, v_dc, p_dc) ---> snlinverter(v_dc, p_dc, inverter)

[jforbess]

While this is a pain, messing with the api, I do think it makes sense, and
is worthwhile going forward.

On Thu, Jun 23, 2016 at 10:04 AM, Will Holmgren [email protected]
wrote:

I think we should reorder a handful of function arguments to put the most
variable-like arguments first and most parameter-like arguments last. I'm
guessing that a lot people will think that I'm just screwing with the API
for no reason, but I do have a few reasons!

1. consistency is important (and underrated)
2. we can make better use of default arguments.

Here are the functions that I know about so far:

ashraeiam(b, aoi) --> ashraeiam(aoi, b=0.05)

physicaliam(K, L, n, aoi) --> physicaliam(aoi, K=4., L=0.002, n=1.526)

singlediode(module, photocurrent, saturation_current, resistance_series, resistance_shunt, nNsVth) -->
singlediode(photocurrent, saturation_current, resistance_series, resistance_shunt, nNsVth, module)

I just noticed that the matlab sapm api is very different, so that's another issue

sapm(module, poa_direct, poa_diffuse, temp_cell, airmass_absolute, aoi) -->
sapm(poa_direct, poa_diffuse, temp_cell, airmass_absolute, aoi, module)

snlinverter(inverter, v_dc, p_dc) --> snlinverter(v_dc, p_dc, inverter)

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
#197, or mute the thread
https://github.com/notifications/unsubscribe/AH66AYymLA_VB4jlhl2Rk5Ft6943zpFAks5qOryPgaJpZM4I9Bua
.

[cwhanse]

I agree, this effort makes sense in general. A few specific tweaks:
For physicaliam, I'd order as aoi, n, K, L. K and L can be safely set to the defaults as the function's values are almost entirely determined by n.
See my comment on #198 for the sapm.

[wholmgren]

Good point, Cliff. I updated that line.

ref #200.