JSON Schema String Constraints — minLength, maxLength, pattern

Adding String Constraints to Generated Schemas

The basic generator infers "type": "string" from any string value, but real-world validation often requires tighter constraints. After generating the initial schema you can enhance string properties with several keywords.

Available String Keywords

Keyword	Purpose	Example
minLength	Minimum character count	"minLength": 1
maxLength	Maximum character count	"maxLength": 255
pattern	Regex the value must match	"pattern": "^[a-z]+$"
format	Semantic format hint	"format": "email"

Example: Enhancing a Username Field

Starting from a generated property:

{ "username": { "type": "string" } }

You might refine it to:

{
  "username": {
    "type": "string",
    "minLength": 3,
    "maxLength": 30,
    "pattern": "^[a-zA-Z0-9_]+$"
  }
}

This ensures the username is 3-30 characters long and contains only alphanumeric characters or underscores.

The format Keyword

The format keyword is an annotation that validators may enforce depending on configuration. Common values include "email", "uri", "date-time", "uuid", and "ipv4". Unlike pattern, format does not use a custom regex — the validator has built-in logic for each recognized format.

When to Use Each Keyword

• Use minLength/maxLength for database column limits or UI field sizes.
• Use pattern when the value must follow a specific syntax (e.g., slugs, codes, phone numbers).
• Use format for well-known types that validators already understand, reducing the chance of a faulty regex.

Combining these keywords lets you go from a loose generated schema to a production-ready contract that catches invalid data before it reaches your application logic.