new fields for preprocess

[raprasad]

• data.dataset.description - default to empty string

• data.dataset.unitOfAnalysis - default to empty string

• data.dataset.structure - wide or long

• variable-level:

  • interpretation - differs depending on whether categorical or continuous,
  • identifier - cross-section or time

{
   "variables": {
    "distance" : {
        "description" : "Distance traveled in km.",
        "interpretation": {
            "standardizedMeasure": "As defined by ...",
            "unit": "kilometers",
        }
    },
   "carColor" : {
         "description" : "Color of the car.",
        "interpretation": {
           "standardizedMeasure": "likert scale",
           "unit": "",
           "categoricalDefinition": { 
              "Label":  {"1": "grey",  "2": "blue",  "3": "orange"},
              "valueDescription": "Car color"
            }
        }
    }
}

(post Tues)

• maybe/for some formats: data.dataset.unit_of_analysis_variables - default to empty list

[raprasad]

These are in:

• [x] data.dataset.description - default to empty string
• [x] data.dataset.unitOfAnalysis - default to empty string
• [x] data.dataset.structure - wide or long