Skip to content

Validation control

Constraints without models

@wire validates every annotated parameter through pydantic, so one constrained parameter needs neither a BaseModel nor a TypeAdapter: put the constraint in the signature with Annotated, preferably behind a type alias so it is reusable and strictly typed:

from pydantic import AfterValidator, Field

type Limit = Annotated[int, Field(gt=0, le=100)]
type Username = Annotated[str, Field(min_length=2, max_length=32)]


def normalize(value: str) -> str:
    return value.strip().lower()


type NormalizedUsername = Annotated[Username, AfterValidator(normalize)]


@wire
def search_users(query: Username, limit: Limit = 20) -> str:
    ...

Invalid input raises ValidationError; valid input is coerced to the annotated type. Every pydantic annotation works: Field constraints, pydantic types such as PositiveInt, and custom validators. Factory parameters are validated the same way, so constraints travel with the dependency graph.

Turning validation off

@wire validates arguments and the return value with pydantic. Turn either off per function when a hot path does not need it:

@wire(cast=False, cast_result=False)
def operation(value: int) -> int:
    ...

wired() accepts the same cast and cast_result flags per declaration.

Project-wide defaults

The configured form of wire returns a reusable decorator. To apply your preferred options everywhere, bind them once in a project module and import that binding instead of wireme.wire:

# myapp/di.py
import wireme

wire = wireme.wire(cast=False, cast_result=False)

# everywhere else
from myapp.di import wire

The binding works on functions, methods, and classes. There is no process-global configuration by design: a bound decorator is explicit, import-order safe, and cannot change the behavior of libraries that use Wireme themselves. At call sites that need different options, use the full wireme.wire(...) form.

Errors

Wireme exposes:

from wireme import ValidationError, WiremeError
  • WiremeError is the base error exposed by Wireme.
  • ValidationError represents dependency input or result validation failures.

Project-specific DI errors may inherit from WiremeError.

Runnable examples

examples/field_constraints.py, examples/validation.py

Next: Protocol dependencies