The NF does not recognize record of parameters as a parameter itself

[casella]

Please check ​Buildings.Fluid.HeatExchangers.DXCoils.BaseClasses.Examples.WetCoil. The NF reports:

[Buildings 7.0.0/Fluid/HeatExchangers/DXCoils/BaseClasses/Examples/WetCoil.mo:10:5-10:18:writable]
Error: Component datCoi of variability parameter has binding 'datCoi'
of higher variability continuous.
[Buildings 7.0.0/Fluid/HeatExchangers/DXCoils/AirCooled/Data/Generic/DXCoil.mo:14:3-19:53:writable]
Error: Could not evaluate structural parameter (or constant): 
wetCoi.datCoi.nSta which gives dimensions of array: sta.
Array dimensions must be known at compile time.

If you look up the ​definition datCoi, its ​type AirCooled.Data.Generic.DXCoil only contains parameters, so I guess the binding should be allowed.

I suspect this may be the reason of other issues, such as #5991 and, possibily #5988.

[perost]

Whether record instances that only contains parameters should be considered parameters themselves was discussed in ​this ticket, and it seems the conclusion was that they shouldn't be.

I opened ​PR 2328 to fix this issue in the Buildings library, since it looks like a simple mistake to me. The other examples in the package that are similar all declare datCoi to be a parameter.

I also don't think this is what's causing the issues you point to since an error is issued if the variability is wrong, which does not happen in those cases.