thegetty / crom

Python library to make creation of CIDOC CRM easier by mapping classes/predicates to python objects
Apache License 2.0
49 stars 16 forks source link

New vocab entries, default boundary check, allow participant #60

Closed azaroth42 closed 5 years ago

kasei commented 5 years ago

@azaroth42 AAT 300008389 is now being used for both a city instance, and a CityName class. Is that intentional? Similarly for nation/CountryName (300128207).

(Sorry I didn't catch this before approval and merging!)

azaroth42 commented 5 years ago

It is intentional. If it causes problems we can change it though :)

kasei commented 5 years ago

Would this be an appropriate modeling using both of those vocab elements?

{
  "id": "urn:uuid:b80463e2-b295-4a33-b95f-c1c9835fac70",
  "type": "Place",
  "_label": "Ghent, Belgium",
  "classified_as": [
    {
      "id": "http://vocab.getty.edu/aat/300008389",
      "type": "Type",
      "_label": "City"
    }
  ],
  "identified_by": [
    {
      "id": "urn:uuid:aa10b83f-4856-4a72-bc65-e2f1e611276d",
      "type": "Name",
      "content": "Ghent",
      "classified_as": [
        {
          "id": "http://vocab.getty.edu/aat/300008389",
          "type": "Type",
          "_label": "City"
        }
      ]
    }
  ]
}
azaroth42 commented 5 years ago

I don't think you would need to classify the name like that, as it's clearly the name of a city ... but I don't think it would be wrong per se, just unnecessary.

The intent is for distinguishing city names in addresses, when the different parts of the address are broken up. e.g. street number, street name, city, state, postal code, country.

kasei commented 5 years ago

So keep the classification on the Place, and then just use model.Name?

azaroth42 commented 5 years ago

Yep :)