Ref
A value here may be written two ways, and both conform. A reader has to handle either.
- string — a reference, as @type: @id writes one
- object — Something the record points at rather than describes.
Examples
"https://bibframe.example/works/9937996a-1083-4357-b05f-4be85f7b7354"{
"@id": "http://id.loc.gov/vocabulary/relators/aut",
"rdfs:label": [
"author"
]
}Trimmed from a real record rather than written for the occasion, and checked against this definition by tests/test_examples.py. A wrong example is worse than none, because it is the first thing anyone copies.
Properties
| Property | Takes | What BIBFRAME says it is |
|---|---|---|
rdfs:label | Text | defined by RDF Schema |
source | Ref | Resource from which value or label came or was derived, such as the formal source/scheme from which a classification number is taken or derived, list from which an agent name is taken or derived, source within which an identifier is unique. |
Every property here holds an array, and a property this definition does not name is still checked: it must be an array too. Naming a property buys a check on what its values look like, and nothing else.
The schema
https://blue-core-lod.github.io/bibframe-json/schema/dialect/Ref.json is this definition on its own, and https://blue-core-lod.github.io/bibframe-json/schema/dialect.json is the whole shape with every definition inlined. They are checked against each other over the conformance corpus, so either is safe to depend on.