@brtietz I am getting a series of "Save inputs as" dialogs saving files with invalid names like ssc-/ssc-pvsamv1.lk.lk followed by a UI callback error:

Could not evaluate callback function:call_reopt->on_change
[73]: error in call to 'reopt_size_battery()': SSC requires input 'subarray1_rear_soiling_loss', but was not found in the SAM UI or from previous simulations[5]: eval error in statement list

Also, the initial msgbox() that opens asking about changing inputs on the battery dispatch page should be made more general now that the script changes more of those check boxes:

Old:

REopt requires that battery can charge from grid.

Change Battery can Charge from Grid input on Battery Dispatch page automatically? Click No to exit optimization.

Suggested New:

REopt requires specific battery dispatch options.

Change these options automatically on the Battery Dispatch page? Click No to exit optimization.

In my tests (in sam-private so I could run the REopt API call), I am getting errors like the following:

Could not evaluate callback function:call_reopt->on_change
[67]: error in call to 'reopt_size_battery()': Invalid inputs. See 'input_errors'.
[ losses value (1.14904) in Scenario>Site>PV exceeds allowable max 0.99 ][5]: eval error in statement list

The following change to Line 5773 of invoke.cpp dividing the loss by 100 to convert from percent to decimal seems to fix the problem:

(annual_total_loss_percent is defined here in pvsamv1: https://github.com/NREL/ssc/blob/0ec9c8300ddc4b9287002f8a7bd5947a9f20a36a/ssc/cmod_pvsamv1.cpp#L919, losses is the DC loss input for PVWatts: https://github.com/NREL/ssc/blob/0ec9c8300ddc4b9287002f8a7bd5947a9f20a36a/ssc/cmod_pvwattsv8.cpp#L134)

I'm not sure exactly what is going on here because the REopt API inputs documentation does not list losses as an input.

Oh, sorry, @Matthew-Boyd and I have a plan to fix this that he hasn't had a chance to execute yet, sorry to waste your time testing it again now.

I think that might be the link to the REopt V2 documentation - I can't find the link to the older, more comprehensive documentation, but "losses" is a variable in the code: https://github.com/NREL/REopt.jl/blob/master/src/core/pv.jl

Relatedly, I wasn't sure I should make an issue about upgrading to the V2 endpoint, since this issue exists: #1135

No need for a different issue, I added a note to #1135 to considering updates.

Additional commits contain required improvements for NatLabRockies/ssc#925 to solve NatLabRockies/ssc#923. I still need to make the requested messagebox changes.

Additionally, while debugging some issues for ssc 923, I encountered the following wxWidgets error message:

Does this require any changes to the dialog window?

@cpaulgilman Ok, this is finally ready for a second review. Please check out NatLabRockies/ssc#925 if it hasn't been merged prior to your review of this PR.

@cpaulgilman, @brtietz, the sam_1092_update_reopt_script pythonhandler.cpp fails to load on MacOS (found while trying to test lk pull request 36)

@cpaulgilman, @brtietz, the sam_1092_update_reopt_script pythonhandler.cpp fails to load on MacOS (found while trying to test lk pull request 36)

This issue is not specific to develop but is caused by a bogus python_config.json that was checked in in commit c3af95

This means that the current release candidate will fail if someone has not run landbosse before! We need to prevent the python_config.json to be checked in - I am going to put on the .gitignore after fixing.