Rules

Definition

Validation rules can be defined in multiple ways.

String Notation

You can write rules as single pipe delimited strings:

rules = {
    "id": "required|integer",
    "email": "required|email",
    "first_name": "required|string",
    "last_name": "required|string",
    "password": "required|min:8|max:255",
}

List Notation

Alternatively to the string notation, rules can be specified as lists of rules:

rules = {
    "id": ["required", "integer"],
    "email": ["required", "email"],
    "first_name": ["required", "string"],
    "last_name": ["required", "string"],
    "password": ["required", "min:8", "max:255"],
}

Function as a Rule

If you require some more custom or complex validation logic than what is provided by the available rules, but you don't want to create a custom rule, you can use a lambda expression or a function as a rule.

The expression or function is expected to return a string if the validation fails, or None if it passes. The returned string is your custom error message.

Note

You can only use a function as a rule when using the list notation.

Tip

Python functions return None by default, so in most cases you probably don't want to explicitly return None.

Examples

An example using a lambda expression:

rules = {
    "some_field": [
        "required",
        "integer",
        lambda field, value, validator: f"The {field} field has to be greater than 2." if value <= 2 else None,
    ],
}

An example using a function:

def custom_validate(field, value, validator):
    if value <= 2:
        return  f"The {field} field has to be greater than 2."


rules = {
    "some_field": ["required", "integer", custom_validate],
}

Tip

Just like custom rules, if needed, you can add implicit and stop flags to your rule function:

def custom_validate(value, **kwargs):
    if not value:
        return  "Value is required!"

custom_validate.implicit = True
custom_validate.stop = True

Provided Arguments

Both lambda expressions and functions will have access to the following keyword arguments:

• field -- name of the field under validation
• value -- value of the field under validation
• validator -- instance of the validator

Tip

If you only need value in your function or expression, you can use **kwargs to omit the rest:

def custom_validate(value, **kwargs):
    if value <= 2:
        return  "Value has to be greater than 2."