Field Constraints and Validators

Pydantic v2 (rewritten in Rust, much faster than v1) validates with type hints plus Field constraints (min_length, ge, pattern). For custom rules use @field_validator (one field) and @model_validator(mode="after") (cross-field checks). Validators should raise ValueError with clear messages; FastAPI turns them into 422 responses with field locations.

from datetime import date
from typing import Self
from pydantic import BaseModel, Field, field_validator, model_validator


class BookingIn(BaseModel):
    guest_name: str = Field(min_length=2, max_length=100)
    check_in: date
    check_out: date
    guests: int = Field(ge=1, le=8)
    promo_code: str | None = None

    @field_validator("promo_code")
    @classmethod
    def normalize_code(cls, v: str | None) -> str | None:
        return v.strip().upper() if v else None

    @model_validator(mode="after")
    def check_dates(self) -> Self:
        if self.check_out <= self.check_in:
            raise ValueError("check_out must be after check_in")
        return self

Common pitfalls

• Using v1 decorators (@validator, @root_validator) — deprecated in v2.
• Forgetting @classmethod on field validators.
• Doing I/O (database lookups) inside validators.

Key points

• Field constraints for simple rules
• field_validator / model_validator for custom rules
• Keep validators pure

Further reading: Pydantic: Validators