In some cases there might be databook quantities only used for calibration/validation purposes. At the moment if a quantity can appear in the datebook, a value must be provided for it. Consider whether to (a) allow marking quantities in the framework as 'optional' which will result in successful validation of a datebook without data for such quantities, and/or (b) automatically considering a quantity optional if it is not used to inform (directly or indirectly) any transitions. Some considerations are
- For compartments and characteristics, inputs could only be optional if the setup weight is 0
- How likely is it that downstream user code relies on data being available for quantities that don't inform transitions. Would it be useful for a framework author to be able to guarantee that data of some sort exists for a given quantity (that would be an argument in favour of (a) over (b)
In some cases there might be databook quantities only used for calibration/validation purposes. At the moment if a quantity can appear in the datebook, a value must be provided for it. Consider whether to (a) allow marking quantities in the framework as 'optional' which will result in successful validation of a datebook without data for such quantities, and/or (b) automatically considering a quantity optional if it is not used to inform (directly or indirectly) any transitions. Some considerations are