CLI to validate a scenario data file

[danielhuppmann]

To make it easier for non-expert users to check their data against a DataStructureDefinition, we need a simple CLI that takes an input file (IAMC scenario data) and validates it against a the definitions.

Basically, this should be a reduced version of cli_run_workflow, but only take an input file (required) and a definitions folder (default to 'definitions'). No output, only raising errors if validation fails.

I suggest to call the CLI nomenclature validate-scenarios (and the function cli_validate_scenarios).

[phackstock]

I find the name validate-scenarios a little bit misleading. In my mind I thought the function validates the values in the scenario column and not an entire data frame. I'd suggest calling it validate-data.
What might be useful in addition is an optional dimension argument that defaults to region, variable but can be extended to scenario, subannual, etc...

[danielhuppmann]

Sorry for not being sufficiently clear in my proposal - in my understanding, validate-scenarios should do the following:

1. check that the scenario data conforms to the codelists (either specified as keyword argument or via nomenclature.yaml or for all available sub-folders}
2. check that the data-values satisfy any given DataValidator instances
3. check that the meta-values satisfy any given MetaValidator instances

#419 implements step 1, with 2 and 3 to be added as we move forward...

[phackstock]

Ah ok thanks for the clarification. To avoid further misunderstandings, by "added as we move forward" you mean that #419, will be merged and then follow-up RPs will implement 2 and 3, or should this all be in #419?
I'm not fully happy with the name validate-scenarios but if you think it's good, I won't object.
Finally, even though rarely used, we could also include applying RequiredDataValidators.

[danielhuppmann]

Keep the PRs clean and simple, so I suggest to merge #419. Not sure how soon we can move forward with the next steps.

[phackstock]

Agreed on the strategy of nice, small PRs.
@dc-almeida, in this case you should edit the description of #419 and remove the "closes" as otherwise this issue will be closed upon merging which is not what we want.