Creating your first schema

JSON Schema is a vocabulary that you can use to annotate and validate JSON documents. This tutorial guides you through the process of creating a JSON Schema.

After creating your JSON Schema, you can then validate example data against your schema by using a validator in a language of your choice. Please, visit Tools and select the validator that better suit your needs.

If you already know how to create JSON Schemas and you are looking for different JSON Schema use cases like schema generation, code generation, documentation, UI generation or JSON Schema processing or conversion, please visit Tools and explore the amazing tooling available in the JSON Schema Ecosystem.

Table of Contents

Overview

The example we use in this guide is a product catalog that stores its data using JSON objects, like the following:

data

{ "productId": 1, "productName": "A green door", "price": 12.50, "tags": [ "home", "green" ]}

Each product in the catalog has:

• productId: an identifier for the product
• productName: the product name
• price: the cost to the consumer
• tags: an optional array of identifying tags

The JSON object is human-readable, but it doesn’t include any context or metadata. There’s no way to tell from looking at the object what the keys mean or what the possible inputs are. JSON Schema is a standard for providing answers to these questions. In this guide, you will create a JSON Schema document that describes the structure, constraints, and data types for a set of JSON data.

Introduction to JSON Schema

The instance is the JSON document that is being validated or described, and the schema is the document that contains the description.

The most basic schema is a blank JSON object, which constrains nothing, allows anything, and describes nothing:

data

{}

By adding validation keywords to the schema, you can apply constraints to an instance. For example, you can use the type keyword to constrain an instance to an object, array, string, number, boolean, or null:

data

{ "type": "string" }

JSON Schema is hypermedia-ready and ideal for annotating your existing JSON-based HTTP API. JSON Schema documents are identified by URIs, which can be used in HTTP link headers and within JSON Schema documents to allow for recursive definitions.

Create a schema definition

To create a basic schema definition, define the following keywords:

• $schema: specifies which draft of the JSON Schema standard the schema adheres to.
• $id: sets a URI for the schema. You can use this unique URI to refer to elements of the schema from inside the same document or from external JSON documents.
• title and description: state the intent of the schema. These keywords don’t add any constraints to the data being validated.
• type: defines the first constraint on the JSON data. In the product catalog example below, this keyword specifies that the data must be a JSON object.

For example:

schema

{ "$schema": "https://json-schema.org/draft/2020-12/schema", "$id": "https://example.com/product.schema.json", "title": "Product", "description": "A product in the catalog", "type": "object"}

The keywords are defined using JSON keys. Typically, the data being validated is contained in a JSON data document, but JSON Schema can also validate JSON data contained in other content types, such as text or XML files.

In JSON Schema terminology, $schema and $id are schema keywords, title and description are schema annotations, and type is a validation keyword.

Define properties

This section adds the properties keyword. In JSON Schema terms, properties is a validation keyword. When you define properties, you create an object where each property represents a key in the JSON data that’s being validated. You can also specify which properties defined in the object are required.

Add the properties object

Using the product catalog example, productId is a numeric value that uniquely identifies a product. Since this is the canonical identifier for the product, it’s required.

To add the properties object to the schema:

1. Add the properties validation keyword to the end of the schema:

   1 ...
   2   "title": "Product",
   3   "description": "A product from Acme's catalog",
   4   "type": "object",
   5   "properties": {
   6     "productId": {}
   7   }

2. Add the productId keyword, along with the following schema annotations:

   • description: describes what productId is. In this case, it’s the product’s unique identifier.

   • type: defines what kind of data is expected. For this example, since the product identifier is a numeric value, use integer.

     1...
     2  "properties": {
     3    "productId": {
     4      "description": "The unique identifier for a product",
     5      "type": "integer"
     6    }
     7  }

With the new properties validation keyword, the overall schema looks like this:

schema

{ "$schema": "https://json-schema.org/draft/2020-12/schema", "$id": "https://example.com/product.schema.json", "title": "Product", "description": "A product from Acme's catalog", "type": "object", "properties": { "productId": { "description": "The unique identifier for a product", "type": "integer" } }}

The following example adds another required key, productName. This value is a string:

schema

{ "$schema": "https://json-schema.org/draft/2020-12/schema", "$id": "https://example.com/product.schema.json", "title": "Product", "description": "A product from Acme's catalog", "type": "object", "properties": { "productId": { "description": "The unique identifier for a product", "type": "integer" }, "productName": { "description": "Name of the product", "type": "string" } }}

The properties object now includes two keys, productId and productName. When JSON data is validated against this schema, validation fails for any documents that contain invalid data in either of these fields.

Define required properties

This section describes how to specify that certain properties are required. This example makes the two existing keys required and adds another required key named price. The price key has a description and type just like the other keys, but it also specifies a minimum value. Because nothing in the store is free, each product requires a price value that’s above zero. Define this using the exclusiveMinimum validation keyword.

To define a required property:

1. Inside the properties object, add the price key. Include the usual schema annotations description and type, where type is a number:

   1   "properties": {
   2     ...
   3     "price": {
   4       "description": "The price of the product",
   5       "type": "number"
   6     }
   7   }

2. Add the exclusiveMinimum validation keyword and set the value to zero:

   1   "price": {
   2     "description": "The price of the product",
   3     "type": "number",
   4     "exclusiveMinimum": 0
   5   }

3. Add the required validation keyword to the end of the schema, after the properties object. Add productID, productName, and the new price key to the array:

   1 ...
   2   "properties": {
   3     ...
   4     "price": {
   5       "description": "The price of the product",
   6       "type": "number",
   7       "exclusiveMinimum": 0
   8     },
   9   },
   10   "required": [ "productId", "productName", "price" ]

With the new required keyword and price key, the overall schema looks like this:

schema

{ "$schema": "https://json-schema.org/draft/2020-12/schema", "$id": "https://example.com/product.schema.json", "title": "Product", "description": "A product from Acme's catalog", "type": "object", "properties": { "productId": { "description": "The unique identifier for a product", "type": "integer" }, "productName": { "description": "Name of the product", "type": "string" }, "price": { "description": "The price of the product", "type": "number", "exclusiveMinimum": 0 } }, "required": [ "productId", "productName", "price" ]}

The exclusiveMinimum validation keyword is set to zero, which means that only values above zero are considered valid. To include zero as a valid option, you could use the minimum validation keyword instead.

Define optional properties

This section describes how to define an optional property. For this example, define a keyword named tags using the following criteria:

• The tags keyword is optional.
• If tags is included, it must contain at least one item.
• All tags must be unique.
• All tags must be text.

To define an optional property:

1. Inside the properties object, add the tags keyword. Include the usual schema annotations description and type, and define type as an array:

   1 ...
   2   "properties": {
   3     ...
   4     "tags": {
   5       "description": "Tags for the product",
   6       "type": "array"
   7     }
   8   }

2. Add a new validation keyword for items to define what appears in the array. For example, string:

   1 ...
   2     "tags": {
   3       "description": "Tags for the product",
   4       "type": "array",
   5       "items": {
   6         "type": "string"
   7       }
   8     }

3. To make sure there is at least one item in the array, use the minItems validation keyword:

   1 ...
   2     "tags": {
   3       "description": "Tags for the product",
   4       "type": "array",
   5       "items": {
   6         "type": "string"
   7       },
   8       "minItems": 1
   9     }

4. To make sure that every item in the array is unique, use the uniqueItems validation keyword and set it to true:

   1 ...
   2     "tags": {
   3       "description": "Tags for the product",
   4       "type": "array",
   5       "items": {
   6         "type": "string"
   7       },
   8       "minItems": 1,
   9       "uniqueItems": true
   10     }

With the new tags keyword, the overall schema looks like this:

schema

{ "$schema": "https://json-schema.org/draft/2020-12/schema", "$id": "https://example.com/product.schema.json", "title": "Product", "description": "A product from Acme's catalog", "type": "object", "properties": { "productId": { "description": "The unique identifier for a product", "type": "integer" }, "productName": { "description": "Name of the product", "type": "string" }, "price": { "description": "The price of the product", "type": "number", "exclusiveMinimum": 0 }, "tags": { "description": "Tags for the product", "type": "array", "items": { "type": "string" }, "minItems": 1, "uniqueItems": true } }, "required": [ "productId", "productName", "price" ]}

Because the new keyword is not required, there are no changes to the required section.

Create a nested data structure

The earlier examples describe a flat schema with only one level. This section describes how to use nested data structures in JSON Schema.

To create a nested data structure:

1. Inside the properties object, create a new key called dimensions:

   1 ...
   2   "properties": {
   3   ...
   4     "dimensions": {}
   5   }

2. Define the type validation keyword as object:

   1 ...
   2   "dimensions": {
   3     "type": "object"
   4   }

3. Add the properties validation keyword to contain the nested data structure. Inside the new properties keyword, add keywords for length, width, and height that all use the number type:

   1 ...
   2   "dimensions": {
   3     "type": "object",
   4     "properties": {
   5       "length": {
   6         "type": "number"
   7       },
   8       "width": {
   9         "type": "number"
   10       },
   11       "height": {
   12         "type": "number"
   13       }
   14     }
   15   }

4. To make each of these properties required, add a required validation keyword inside the dimensions object:

   1 ...
   2   "dimensions": {
   3     "type": "object",
   4     "properties": {
   5       "length": {
   6         "type": "number"
   7       },
   8       "width": {
   9         "type": "number"
   10       },
   11       "height": {
   12         "type": "number"
   13       }
   14     },
   15     "required": [ "length", "width", "height" ]
   16 }

Using the new nested data structures, the overall schema looks like this:

schema

{ "$schema": "https://json-schema.org/draft/2020-12/schema", "$id": "https://example.com/product.schema.json", "title": "Product", "description": "A product from Acme's catalog", "type": "object", "properties": { "productId": { "description": "The unique identifier for a product", "type": "integer" }, "productName": { "description": "Name of the product", "type": "string" }, "price": { "description": "The price of the product", "type": "number", "exclusiveMinimum": 0 }, "tags": { "description": "Tags for the product", "type": "array", "items": { "type": "string" }, "minItems": 1, "uniqueItems": true }, "dimensions": { "type": "object", "properties": { "length": { "type": "number" }, "width": { "type": "number" }, "height": { "type": "number" } }, "required": [ "length", "width", "height" ] } }, "required": [ "productId", "productName", "price" ]}

The new required validation keyword only applies within the scope of the dimensions key.

Add an external reference

This section describes how to reference resources outside of the schema. Sharing schemas across many data structures is a common way to make them easier to use, read, and keep up-to-date. So far, the product catalog schema is self-contained. This section creates a new schema and then references it in the product catalog schema.

The following schema validates a geographical location:

schema

{ "$id": "https://example.com/geographical-location.schema.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "title": "Longitude and Latitude", "description": "A geographical coordinate on a planet (most commonly Earth).", "required": [ "latitude", "longitude" ], "type": "object", "properties": { "latitude": { "type": "number", "minimum": -90, "maximum": 90 }, "longitude": { "type": "number", "minimum": -180, "maximum": 180 } }}

To reference this schema in the product catalog schema:

1. Inside the properties object, add a key named warehouseLocation:

   1 ...
   2   "properties": {
   3   ...
   4     "warehouseLocation": {}
   5   }

2. To link to the external geographical location schema, add the $ref schema keyword and the schema URL:

   1 ...
   2   "warehouseLocation": {
   3     "description": "Coordinates of the warehouse where the product is located.",
   4     "$ref": "https://example.com/geographical-location.schema.json"
   5   }

With the external schema reference, the overall schema looks like this:

schema

{ "$schema": "https://json-schema.org/draft/2020-12/schema", "$id": "https://example.com/product.schema.json", "title": "Product", "description": "A product from Acme's catalog", "type": "object", "properties": { "productId": { "description": "The unique identifier for a product", "type": "integer" }, "productName": { "description": "Name of the product", "type": "string" }, "price": { "description": "The price of the product", "type": "number", "exclusiveMinimum": 0 }, "tags": { "description": "Tags for the product", "type": "array", "items": { "type": "string" }, "minItems": 1, "uniqueItems": true }, "dimensions": { "type": "object", "properties": { "length": { "type": "number" }, "width": { "type": "number" }, "height": { "type": "number" } }, "required": [ "length", "width", "height" ] }, "warehouseLocation": { "description": "Coordinates of the warehouse where the product is located.", "$ref": "https://example.com/geographical-location.schema.json" } }, "required": [ "productId", "productName", "price" ]}

Validate JSON data against the schema

Now that you have your JSON Schema is time to validate JSON data against it using a JSON Schema Validator.

A Validator is a tool that implements the JSON Schema specification. All validators works in a similar way: they take a JSON Schema and a JSON Instance as input and they returns the validation result as output.

To try it yourself, please visit Tools and select the validator that better suit your needs, our use the editors available below to explore the different Schemas and Instances and see the different validation results.

What Next?

Now that you know how to create a JSON Schema and use it to validate JSON data, we'd invite you to continue your JSON Schema journey:

• Learn more about JSON Schema by visiting the reference documentation.
• Explore the details of the current version of the Spec 2020-12.

If you already know how to create JSON Schemas and you are looking for different JSON Schema use cases like schema generation, code generation, documentation, UI generation or JSON Schema processing or conversion, please visit Tools and explore the amazing tooling available in the JSON Schema Ecosystem.