Open demeringo opened 8 months ago
Trying to represents impacts in a json schema leads to question: should we make all fields of an impact object mandatory ?
My understanding in the draft response.json and https://github.com/Boavizta/PCR-SI_dataformat/blob/main/ImpactObject.md is that we tend to define impacts objects as an fixed list of all (4) licecycle steps X all (18) indicators.
{ "BLD" : { "GWP": 123, "ADPe": 13, "ADPf": 23, "WU": 12, "...": ... }, "DIS" : { "GWP": 123, "ADPe": 13, "ADPf": 23, "WU": 12, "...": ... }, "USE" : { "GWP": 123, "ADPe": 13, "ADPf": 23, "WU": 12, "...": ... }, "EOL" : { "GWP": 123, "ADPe": 13, "ADPf": 23, "WU": 12, "...": ... } }
Should we use arrays instead (maybe with explictly named properties) that would allow unknwown values to not appear ?
We can let each of the LCA steps and indicators as optional fields, depending on the background engine. The PCR indicators must be field for standard compliance.
Trying to represents impacts in a json schema leads to question: should we make all fields of an impact object mandatory ?
My understanding in the draft response.json and https://github.com/Boavizta/PCR-SI_dataformat/blob/main/ImpactObject.md is that we tend to define impacts objects as an fixed list of all (4) licecycle steps X all (18) indicators.
Should we use arrays instead (maybe with explictly named properties) that would allow unknwown values to not appear ?