tortoise-pagination

Usage

Supposing in myapp.schema you have a pydantic BaseModel to represent your model

from fastapi import Depends
from tortoise_pagination import Pagination, Page

from myapp.main import app
from myapp.models import MyModel
from myapp.schema import MySchema


@app.get('/mymodel')
async def my_view(pagination: Depends(Pagination.from_query)) -> Page[MySchema]:
    return await pagination.paginated_response(MyModel.all(), MySchema)

now you can request with:

curl http://localhost:8000/mymodel?offset=0&limit=20

returned structure:

• items list[MySchema]
• count: NonNegativeInt -> the number of entries for this queryset (MyModel.all().count()) wich the frontend will need to be able to display a pagination

Computed field

Sometime you may want to add computed fields, however pydantic is not async capable so sometime it may become a challenge to achieve that we do:

from fastapi import Depends
from fastapi.routing import APIRouter
from myapp.models import Product
from tortoise.contrib.pydantic import pydantic_model_creator

from tortoise_pagination import Page, Pagination

ProductBaseSchema = pydantic_model_creator(Product, include=("id", "price", "weight"))


class ProductSchema(ProductBaseSchema):
    name: str
    price_per_kilogram: float | None


router = APIRouter(prefix="products")


async def _compute_price_per_kilogram(product: Product) -> float | None:
    try:
        return product.price / product.weight
    except ZeroDivisionError:
        return None


@router.get("")
async def list_products(
    pagination: Depends(Pagination.from_query),
) -> Page[ProductSchema]:
    products = Products.all()
    return await pagination.get_custom_paginated_response(
        queryset=products,
        schema=ProductSchema,
        extra_fields={
            "name": lambda product: product.name.title(),
            "price_per_kilogram": _compute_price_per_kilogram,
        },
    )

Lazy schema

It's possible to build a custom schema from the extra fields definition

from tortoise_pagination import build_pydantic_model_with_extra_fields


def _get_name(product: Product) -> str:
    return product.name.title()


async def _compute_price_per_kilogram(product: Product) -> float | None:
    try:
        return product.price / product.weight
    except ZeroDivisionError:
        return None


EXTRA_FIELDS = {
    "name": _get_name,
    "price_per_kilogram": _compute_price_per_kilogram
}

# This model will have `name` and `price_per_kilogram` with the other fields
# note that you must have return annotation types in given function for extra
# fields, otherwise Any will be used (ex for lambda functions).
ProductWithExtraFieldsSchema = build_pydantic_model_with_extra_fields(
    ProductSchema,
    "ProductWithExtraFields",
    EXTRA_FIELDS,
)