There is some problem with spectral data in the following example

[vidanovic]

This is not necessary problem with WinCalc itself, but need further pondering on how to process this error from WINDOW.

[RDmitchell]

@vidanovic -- is this WINDOW 8.1? or also 7.8?

[vidanovic]

This is in WinCalc/WinCalc-Engine. This comment is under WC repo.

[StephenCzarnecki]

Looking at the spectral data in the database for NFRC 888 the first wavelength is .305. However according to the NFRC standard the minimum wavelength for the SOLAR method is .300. Since the product does not contain measured data over the entire range WinCalc throws an exception accordingly.

However we will may need to look into how exceptions are being handled in either WINDOW or WinCalc (or maybe both).

[StephenCzarnecki]

Looking into this a bit more and I believe the issue is with WINDOW not WinCalc.

I can wrap the call to WinCalc with a try-catch and that caches an exception which means that WinCalc is correctly throwing an exception which is being propagated up to WINDOW.

However simply catching the exception at boundary between WINDOW and WinCalc does not seem to be enough to resolve this issue so we need to investigate how to handle the exception in WINDOW.

i.e. this is not sufficient in CGlzSysRS::SaveWCEOpticalResults

try
{
    Solar = glazing.optical_method_results("SOLAR");
}
catch (std::runtime_error& e)
{
    std::basic_stringstream<TCHAR> msg;
    msg << _T("Error calculating solar spectral data properties: ") << e.what();
    DisplayMessage(msg.str().c_str());
    return;
}

[jcjonsson]

@StephenCzarnecki @vidanovic

Simon explained it to me that the handling of the exception could be improved, but to me this shouldn't happen for IGDB data. However, as it can happen for Optics imported data it is a decision to make how important it is to fix. I do not feel like we are responsible for users who put garbage into the software and then break it, but that is just my opinion.

That record (888) has been deleted since 2022 and people with up to date IGDB libraries might not have access to it when testing. I had to dig pretty deep to find the original file in IGDB v10 which still somehow imports into Optics (despite the sparse header from back then: )

NFRC f52c
{units} IP Microns
{thickness} .156
{conductivity} .520
.305 .001 .050 .050
...

[RDmitchell]

@vidanovic / @jcjonsson -- should we close this issue then?

[RDmitchell]

This record is in the default w7.mdb file that comes with the installer for WINDOW 7.8.74/77.

It shows that 888 is a deleted record, so the user can't actually do the calculation.

But I will put it on my list to update the w7.mdb file so that there is a reasonable example there.

[vidanovic]

@RDmitchell No we should not close this issue. Still needs to be fixed in case when the data are missing.

[RDmitchell]

OK, thx