In JSON Schemas, you can either put a schema per file and then access them using their URL (where you stored them), or a big schema with id
tags.
Here is for one big file:
{
"id": "#root",
"properties": {
"author": {
"id": "#author",
"properties": {
"first_name": {
"type": "string"
},
"last_name": {
"type": "string"
}
},
"type": "object"
},
// author
"author_api": {
"id": "#author_api",
"items": {
"$ref": "author"
},
"type": "array"
},
// authors API
"book": {
"id": "#book",
"properties": {
"author": {
"type": "string"
},
"title": {
"type": "string"
}
},
"type": "object"
},
// books API: list of books written by same author
"books_api": {
"id": "#books_api",
"properties": {
"author": {
"$ref": "author"
},
"books": {
"items": {
"$ref": "book"
},
"type": "array"
}
},
"type": "object"
}
}
}
You can then reference your validator to one of those sub schemas (which are defined with an id
).
From outside of your schema, this:
{ "$ref": "url://to/your/schema#root/properties/book" }
is equivalent to this:
{ "$ref": "url://to/your/schema#book" }
… which is equivalent, from inside, to this:
{ "$ref": "#root/properties/book" }
or this (still from inside):
{ "$ref": "#book" }
See my answer here for more information.
与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…