How to enable field validators in SQLModel

[mohit2152sharma]

First Check

• I added a very descriptive title here.
• I used the GitHub search to find a similar question and didn't find it.
• I searched the SQLModel documentation, with the integrated search.
• I already searched in Google "How to X in SQLModel" and didn't find any information.
• I already read and followed all the tutorial in the docs and didn't find an answer.
• I already checked if it is not related to SQLModel but to Pydantic.
• I already checked if it is not related to SQLModel but to SQLAlchemy.

Commit to Help

• I commit to help with one of those options 👆

Example Code

class User(SQLModel, table=True):  # pyright: ignore
    __tablename__ = "users"  # type: ignore
    id: int = Field(default=None, primary_key=True)
    uuid: UUID4 = Field(default_factory=uuid4)
    email: EmailStr = Field(sa_column=Column("email", String, unique=True))
    isVerified: bool = False

    class Config:
        arbitrary_types_allowed = True

u1 = User(email="abc")
print(u1.email)

Description

Pydantic has a type EmailStr which can check for whether the string is a valid email or not. I have a column in my database for emails.

When I use the following code:

class User(SQLModel, table=True):  # pyright: ignore
    __tablename__ = "users"  # type: ignore
    id: int = Field(default=None, primary_key=True)
    uuid: UUID4 = Field(default_factory=uuid4)
    email: EmailStr
    isVerified: bool = False

    class Config:
        arbitrary_types_allowed = True

sqlmodel complains with following error

Traceback (most recent call last):
  File "/Users/mohitsharma/github/saral/saral-backend/api/utils/database.py", line 64, in <module>
    class User(SQLModel, table=True):  # pyright: ignore
  File "/Users/mohitsharma/miniconda3/envs/backend/lib/python3.11/site-packages/sqlmodel/main.py", line 474, in __new__
    col = get_column_from_field(v)
          ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/mohitsharma/miniconda3/envs/backend/lib/python3.11/site-packages/sqlmodel/main.py", line 618, in get_column_from_field
    sa_type = get_sqlalchemy_type(field)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/mohitsharma/miniconda3/envs/backend/lib/python3.11/site-packages/sqlmodel/main.py", line 607, in get_sqlalchemy_type
    raise ValueError(f"{type_} has no matching SQLAlchemy type")
ValueError: <class 'pydantic.networks.EmailStr'> has no matching SQLAlchemy type

I can do away with this error by using sa_column but that also drops the additional checks that pydantic provides on email string. For example:

class User(SQLModel, table=True):  # pyright: ignore
    __tablename__ = "users"  # type: ignore
    id: int = Field(default=None, primary_key=True)
    uuid: UUID4 = Field(default_factory=uuid4)
    email: EmailStr = Field(sa_column=Column("email", String, unique=True))
    isVerified: bool = False

    class Config:
        arbitrary_types_allowed = True

u1 = User(email="abc")
print(u1.email)

This runs, but it also considers "abc" as a valid email field.

How can I have a custom type with validation rules in sql model? For example:

class User:
    someColumn: ArbitraryType = Field(default_factory=some_func)
# ArbitraryType has some validation rules

Operating System

macOS

Operating System Details

No response

SQLModel Version

0.0.14

Python Version

3.11.5

Additional Context

No response

[philipp-bliznuk]

I use sa_type instead of sa_column to mix pydantic types with SQLAlchemy types:

from pydantic import EmailStr
from sqlmodel import Field, SQLModel


class User(SQLModel, table=True):  # pyright: ignore

    ...
    email: EmailStr = Field(
        sa_type=String(),  # type: ignore[call-overload]
        unique=True,
        index=True,
        nullable=False,
        description="The email of the user",
    )

and that gives me email validation from pydantic and in database it's a regular string

or:

from pydantic import IPvAnyNetwork
from sqlmodel import Field, SQLModel
from sqlalchemy import ARRAY, String
from sqlalchemy.sql.expression import false


class Token(SQLModel, table=True):
    """Token Model, represents the token for access to service APIs."""
    ...
    internal: bool = Field(
        sa_column_kwargs={"server_default": false()},
        default=False,
        description="Whether the token is an internal system token or not",
    )
    allowed_ips: list[IPvAnyNetwork] = Field(
        sa_type=ARRAY(String),  # type: ignore[call-overload]
        default=[],
        description="The list of allowed IPv4 or IPv6 addresses/subnets for the token",
    )