Constraint Data Types

Data types play a pivotal role in validating JSON data for compliance with the schema. Essentially, data types determine the kind of data that a JSON element or value can contain. This mechanism serves as a fundamental process in maintaining the accuracy, consistency, and integrity of JSON document and its structure throughout the system, where data quality and reliability are vital.

In the schema document, data types are denoted by the # prefix. Here is an outline of all data types, including their subtypes, used in the schema document to validate a JSON document. When using multiple data types for validation, it indicates that the JSON value is considered valid if it complies with any of the specified alternative data types. All of these data types and their subtypes offer the flexibility of selecting the most appropriate type based on requirements.

#any
 ┬
 ├ #primitive
 │  ┬
 │  ├ #string
 │  │  ┬
 │  │  └ #datetime
 │  │     ┬
 │  │     ├ #date
 │  │     └ #time
 │  ├ #number
 │  │  ┬
 │  │  ├ #integer
 │  │  ├ #float
 │  │  └ #double  
 │  ├ #boolean
 │  └ #null  
 └ #composite
    ┬
    ├ #array
    └ #object

The Any Data Type

This data type accepts any valid JSON value that conforms to the JSON standard. It is the least restrictive data type and serves as the parent type for all other data types defined in this schema, each of which imposes more specific constraints. Here is the specification of JSON document containing rules of all valid JSON values. Following is the syntax for specifying this data type:

#any

The Primitive Data Type

This serves as the foundational data type for all non-composite or primitive JSON values. While it is not intended for direct use in a schema, it organizes common functionalities across other sub-data types. Nevertheless, it remains valid for use, as illustrated by the following syntax:

#primitive

The Composite Data Type

As the parent data type for two kinds of composite JSON values, namely arrays and objects, this composite type consists of zero or more JSON values. It is not designed to be used in a schema but rather the organization of shared functionalities among other sub-data types. Despite this, it is still valid for use, demonstrated by the following syntax:

#composite

The Array Data Type

This data type represents the JSON array type and accepts any JSON array specified by the JSON standard. The specification document for JSON provides details about the various syntax and forms of JSON arrays. Below is the syntax for specifying this data type:

#array

The Object Data Type

This data type represents the JSON object type and accepts any JSON object specified by the JSON standard. The specification document for JSON provides details about the different syntax and forms of JSON objects. Following is the syntax for specifying this data type:

#object

The String Data Type

This is one of the most commonly used data types in a JSON document, designed to accept any JSON string as specified by the JSON standard. The syntax for specifying this data type is as follows:

#string

The Date Time Data Type

The date-time data type serves as the parent data type for both date and time types. It is a subtype of JSON string type and thus formatted as per the JSON string specification. It is not intended to be used directly in a schema, but it provides common functionality for both date and time types.

#datetime

The Date Data Type

The date data type accepts a string representation of a date, conforming to the ISO 8601 standard (date part only). This is the default configuration, which can be modified using the directive described here. It is a subtype of date-time type and thus also formatted as per the JSON string specification. Detailed explanations of the ISO 8601 standard can be found in this document. Furthermore, you can refer to this document for a detailed description of the date pattern associated with this data type. To define this data type in schema, use the following syntax:

#date

The Time Data Type

The time data type accepts a string representation of a time (including both date and time parts), in accordance with the ISO 8601 standard. This default configuration can be modified using the directive described here. Similar to the date data type, it is a subtype of date-time data type and thus also formatted as per the JSON string specification. Here is the ISO 8601 standard document, which contains detailed explanations. Furthermore, you can refer to this document for a detailed description of the date-time pattern associated with this data type. To define this data type in schema, use the following syntax:

#time

The Number Data Type

The number data type serves as the parent data type for all numeric types accommodated by the schema, including integer, floating-point, and exponent-based numbers. It accepts any JSON numerical value in accordance with the JSON standard. The syntax for specifying the number type is as follows:

#number

The Integer Data Type

The integer data type is a subtype of the number data type that only allows integral numbers or whole numbers without any fraction and exponent and thus provides constraints for a wide range of real-world scenarios where numbers cannot involve decimal points or exponents. To specify the integer type in schema, use the following syntax:

#integer

The Float Data Type

The float data type is also a subtype of the number data type that only accepts floating point numbers and does not allow exponent in numbers or integral numbers. This constraint is useful for various real-world applications that require numbers to be exclusively in floating-point format. To specify the float type in schema, use the following syntax:

#float

The Double Data Type

The double data type, as a subtype of the number data type, exclusively accepts numbers with exponents. It can either be an integral number with an exponent or a floating-point number with an exponent. This constraint distinguishes it from other number formats and makes it particularly useful for handling large numbers with exponents. To specify the double type in a schema, use the following syntax:

#double

The Boolean Data Type

The boolean data type is a binary or switch-based data type that only accepts two values, namely true and false. It is particularly useful in situations where toggling and switching are necessary. To specify the boolean type in the schema, use the following syntax:

#boolean

The Null Data Type

The null data type serves as a special constraint within JSchema documents, facilitating the controlled use of null in place of other JSON elements or values. Typically, it is combined with other data types to permit the use of null for specific JSON elements or values. This can set constraints for scenarios in which an array without any elements and an object without any properties can either have null or only be allowed to be empty.

Additionally, the @nonempty constraint functions can be employed to further control the use of empty values within a JSON document. To specify the null type in the schema, use the following syntax:

#null