Miscellaneous Examples

In this page, you will find miscellaneous examples illustrating different uses cases to help you get the most out of your JSON Schemas. Each example comes with accompanying JSON data and explanation.

• A typical minimum schema
• Arrays of things
• Enumerated values
• Regular expression pattern
• Complex object with nested properties
• Conditional validation with dependentRequired
• Conditional validation with dependentSchemas
• Conditional validation with if-else

Basic

This example provides a typical minimum you are likely to see in JSON Schema. It contains:

• $id keyword
• $schema keyword
• title annotation keyword
• type instance data model
• properties validation keyword
• Three keys: firstName, lastName and age each with their own:
  • description annotation keyword.
  • type instance data model (see above).
• minimum validation keyword on the age key.

Data

In the data example, we provide values for the firstName, lastName, and age properties. The values match the defined schema, where firstName is a string, lastName is a string, and age is an integer greater than or equal to zero.

Arrays of things

Arrays are fundamental structures in JSON -- here we demonstrate a couple of ways they can be described:

• An array of string values.
• An array of objects.

We also introduce the following with this example:

• $defs keyword
• $ref keyword

For the fruits property:

• type is set to "array" to indicate it's an array.
• items describes the items within the array. In this case, they should be of type "string".

For the vegetables property:

• type is also set to "array" to indicate it's an array.
• items references the $defs/veggie definition, indicating that the items in the array should conform to the "veggie" schema defined in the $defs section.

Data

The data example shows the usage of arrays. The fruits property contains an array of strings, while the vegetables property contains an array of objects, each adhering to the "veggie" schema definition.

Enumerated values

This example introduces the enum validation keyword which is used with an array of values that includes an integer (42), a boolean (true), a string ("hello"), null, and an array ([1, 2, 3]). This demonstrates how enum can be used to specify a set of allowed values of different types.

Data

The provided data adheres to the schema by using the exact values specified in the enum array: [1, 2, 3].

Regular expression pattern

This example introduces the pattern keyword and defines an object with a property called code that must match a specific regular expression pattern: ^[A-Z]{3}-\d{3}$. The pattern here requires three uppercase letters followed by a hyphen and three digits.

Data

The provided data, "ABC-123", satisfies this pattern defined in the schema.

Complex object with nested properties

The schema below represents a complex object with various properties including name, age, address, and hobbies. The address property is an object with nested properties, and the hobbies property is an array of strings. The name and age properties are required.

Data

The provided data conforms to the schema by including values for the required properties and ensuring the age is an integer greater than or equal to zero. The address object contains all the necessary properties, and the hobbies property is an array of strings.

Conditional validation with dependentRequired

In this example, the dependentRequired keyword is used to specify that the property bar is required when the property foo is present. The schema enforces the condition that if foo exists, then bar must also be present.

Data

As per the schema, when the foo property is present (true), the bar property becomes required. The bar property is provided with the value "Hello World", satisfying the requirement of being a string and ensuring compliance with the dependentRequired condition.

Since both foo and bar are missing, the instance is still valid and in compliance with the dependentRequired condition as well.

The above schema is invalid, since the foo property is present, but bar is not, which invalidates the condition of the dependentRequired keyword.

Conditional validation with dependentSchemas

The given schema showcases the use of the dependentSchemas keyword. It allows defining a subschema that must be satisfied if a certain property is present.

• In this example, the schema defines an object with two properties: foo and propertiesCount. The foo property is of boolean type, while the propertiesCount property is of integer type with a minimum value of 0.
• According to the subschema, when the foo property is present, the propertiesCount property becomes required, and must be an integer with a minimum value of 7.

Data

Here, the foo property is set to true, indicating its presence. As per the schema, when foo is present, the propertiesCount property becomes required. In this case, the propertiesCount property is provided with a value of 10, which satisfies the requirement of being an integer and having a minimum value of 7.

In the above data, propertiesCount is 5 but since foo is missing, propertiesCount does not need to be 7 or more than 7, it only needs to be greater than or equal to 0. Hence, this instance is valid.

In this, we have foo as true, but propertiesCount is 5, and in the schema, propertiesCount is set to have minimum 7 as the input according to the dependentSchemas. Hence, this is an invalid instance.

Conditional validation with if-else

In this schema, we have two properties: isMember and membershipNumber. The conditional validation is based on the value of the isMember property. The validation keywords if, then, and else.

Here's how the validation works in this example:

If the value of isMember is true:

• The then block is applied, which specifies that the membershipNumber property should be a string with a minimum length of 10 and a maximum length of 10.

If the value of isMember is anything other than true:

• The else block is applied, which specifies that the membershipNumber property can be any string.

Data

In this case, the isMember property is set to true, so the then block is applied. The membershipNumber property is a string with a length of 10 characters, satisfying the validation.

In this case, the isMember property is false, so the else block is applied. The membershipNumber property can be any string with minimum length greater than or equal to 15, so it satisfies the validation.