Errors¶

Error classes and utilities for validation error handling.

Classes¶

BridgeValidationError ¶

BridgeValidationError(
    message: str | list[Any] | dict[str, Any],
    field_name: str = "_schema",
    data: Any | None = None,
    valid_data: dict[str, Any] | None = None,
    **kwargs: Any,
)

Bases: ValidationError

Validation error that bridges Marshmallow and Pydantic error formats.

Extends Marshmallow's ValidationError to track partially valid data, enabling graceful handling of multi-field validation failures.

ATTRIBUTE	DESCRIPTION
`messages`	Dict of field -> error messages (Marshmallow format)
`valid_data`	Dict of successfully validated fields
`data`	Original input data

Usage¶

Catching Validation Errors¶

from pydantic_marshmallow import schema_for, BridgeValidationError

UserSchema = schema_for(User)
schema = UserSchema()

try:
    user = schema.load({
        "name": "",
        "email": "invalid",
        "age": -5
    })
except BridgeValidationError as e:
    # Error messages by field
    print(e.messages)
    # {'name': ['String should have at least 1 character'],
    #  'email': ['value is not a valid email address'],
    #  'age': ['Input should be greater than or equal to 0']}

    # Original input data
    print(e.data)
    # {'name': '', 'email': 'invalid', 'age': -5}

    # Fields that passed validation
    print(e.valid_data)
    # {}

Marshmallow Compatibility¶

BridgeValidationError extends Marshmallow's ValidationError, so standard exception handling works:

from marshmallow import ValidationError

try:
    user = schema.load(invalid_data)
except ValidationError as e:
    # Works with both Marshmallow and Bridge errors
    print(e.messages)

Partial Validation¶

When using partial loading, valid_data contains fields that passed:

try:
    user = schema.load(
        {"name": "Alice", "email": "invalid"},
        partial=True
    )
except BridgeValidationError as e:
    print(e.valid_data)
    # {'name': 'Alice'}  # email failed, but name was valid