How to use SQLAlchemy's most resent PEP 484 Annotated Declarative Table style?

[mhubig]

Hi there I'm wondering how one can utilize the PEP 484 Annotated Declarative Table Style of SQLAlchemy with pgvector

See: https://docs.sqlalchemy.org/en/20/orm/declarative_tables.html#using-annotated-declarative-table-type-annotated-forms-for-mapped-column).

This is a Table definition from a small flask app I'm working on, but since I'm fairly new to SQLAlchemy, I'm not sure if this is the correct way of doing it.

from typing import Iterable

from pgvector.sqlalchemy import Vector
from sqlalchemy.dialects.postgresql import JSONB
from sqlalchemy.orm import Mapped, mapped_column

from flask_app import db

class Product(db.Model):
    __tablename__ = "product"

    id: Mapped[str] = mapped_column(primary_key=True)
    data: Mapped[dict|list] = mapped_column(type_=JSONB)
    tokensize: Mapped[int]= mapped_column()
    checksum: Mapped[str] = mapped_column(db.String(40), index=True)
    embedding: Mapped[Iterable[float]] = db.mapped_column(Vector(1536))

[ankane]

Hi @mhubig, that should work. From my understanding, the type (Iterable[float]) will be discarded.

If the mapped_column() construct indicates an explicit type as passed to the mapped_column.__type argument, then the given Python type is disregarded.