Assume the following JSON from Google GeoCoding API
{
"lon": 1.2345678,
"lat": 1.2345678,
"formatted_adress": "Sunny Ave, 01234 Dev City, New Devland",
"address_components": [
{
"long_name": "Sunny Ave",
"short_name": "Sunny Ave",
"types": [
"route"
]
},
{
"long_name": "Downtown",
"short_name": "Downtown",
"types": [
"sublocality_level_1",
"sublocality",
"political"
]
},
{
"long_name": "Dev City",
"short_name": "Dev City",
"types": [
"locality",
"political"
]
},
{
"long_name": "Dev City",
"short_name": "Dev City",
"types": [
"administrative_area_level_3",
"political"
]
},
{
"long_name": "Somwherechester",
"short_name": "SC",
"types": [
"administrative_area_level_1",
"political"
]
},
{
"long_name": "New Devland",
"short_name": "DE",
"types": [
"country",
"political"
]
},
{
"long_name": "01234",
"short_name": "01234",
"types": [
"postal_code"
]
}
]
}
I need to validate the data as shown (types, properties, etc.).
Except:
address_components MUST contain all objects where types contains at least one element of [locality, country, postal_code]. Additional objects are allowed.
Based on this solution https://github.com/orgs/json-schema-org/discussions/191 I managed to get close enough if types would be a constant string. But I didn't manage to validate against an array of strings.
{
"$schema": "https://json-schema.org/draft/2020-12/schema",
"$id": "",
"title": "Google APIs Geo Coding",
"description": "Necessary address components.",
"type": "object",
"required": [
"formatted_adress",
"lon",
"lat",
"address_components"
],
"properties": {
"formatted_adress": {
"type": "string"
},
"lon": {
"type": "number"
},
"lat": {
"type": "number"
},
"address_components": {
"type": "array",
"items": {
"type": "object",
"required": [
"long_name",
"short_name",
"types"
],
"properties": {
"long_name": {
"type": "string"
},
"short_name": {
"type": "string"
},
"types": {
"type": "string"
}
}
},
"allOf": [
{
"contains": {
"properties": {
"types": {
"const": "locality"
}
}
}
},
{
"contains": {
"properties": {
"types": {
"const": "country"
}
}
}
},
{
"contains": {
"properties": {
"types": {
"const": "postal_code"
}
}
}
}
]
}
}
}
Examples
invalid data => there's no object with postal_code.
{
"lon": 1.2345678,
"lat": 1.2345678,
"formatted_adress": "Sunny Ave, 01234 Dev City, New Devland",
"address_components": [
{
"long_name": "Sunny Ave",
"short_name": "Sunny Ave",
"types": [
"route"
]
},
{
"long_name": "Downtown",
"short_name": "Downtown",
"types": [
"sublocality_level_1",
"sublocality",
"political"
]
},
{
"long_name": "Dev City",
"short_name": "Dev City",
"types": [
"locality",
"political"
]
},
{
"long_name": "Dev City",
"short_name": "Dev City",
"types": [
"administrative_area_level_3",
"political"
]
},
{
"long_name": "Somwherechester",
"short_name": "SC",
"types": [
"administrative_area_level_1",
"political"
]
},
{
"long_name": "New Devland",
"short_name": "DE",
"types": [
"country",
"political"
]
}
]
}
valid data => there're all 3 objects with at least one of the mentioned types - no additional (optional) objects
{
"lon": 1.2345678,
"lat": 1.2345678,
"formatted_adress": "Sunny Ave, 01234 Dev City, New Devland",
"address_components": [
{
"long_name": "Dev City",
"short_name": "Dev City",
"types": [
"locality",
"political"
]
},
{
"long_name": "New Devland",
"short_name": "DE",
"types": [
"country",
"political"
]
},
{
"long_name": "01234",
"short_name": "01234",
"types": [
"postal_code"
]
}
]
}
To handle validation of an array of strings in the types property and ensure the presence of at least one of the required types (locality, country, postal_code), you can use a JSON schema that checks whether the types array contains at least one of these values. Here's a refined version of your schema:
{
"$schema": "https://json-schema.org/draft/2020-12/schema",
"$id": "",
"title": "Google APIs Geo Coding",
"description": "Necessary address components.",
"type": "object",
"required": [
"formatted_adress",
"lon",
"lat",
"address_components"
],
"properties": {
"formatted_adress": {
"type": "string"
},
"lon": {
"type": "number"
},
"lat": {
"type": "number"
},
"address_components": {
"type": "array",
"items": {
"type": "object",
"required": [
"long_name",
"short_name",
"types"
],
"properties": {
"long_name": {
"type": "string"
},
"short_name": {
"type": "string"
},
"types": {
"type": "array",
"items": {
"type": "string"
}
}
}
},
"allOf": [
{
"contains": {
"properties": {
"types": {
"type": "array",
"contains": {
"enum": [
"locality"
]
}
}
}
}
},
{
"contains": {
"properties": {
"types": {
"type": "array",
"contains": {
"enum": [
"country"
]
}
}
}
}
},
{
"contains": {
"properties": {
"types": {
"type": "array",
"contains": {
"enum": [
"postal_code"
]
}
}
}
}
}
]
}
}
}
Key points:
types as an array
The types property is now correctly defined as an array of strings.
contains keyword
Used to ensure that the array contains at least one element from the specified set (locality, country, postal_code).
allOf for multiple checks
The schema ensures that the address_components array contains at least one object with locality, one with country, and one with postal_code as part of the types array.
This schema will correctly validate that address_components includes the required address types.