fastapi-uuidbase62

fastapi-uuidbase62 is intended to provide Pydantic and FastAPI functionality that exposes UUID values as nicely-formatted Stripe-like string values, with validated prefixing. For example, a User UUID identifier of "f8711c37-c1d1-4961-ba3c-98cdc5b4fda8" with a "user" prefix becomes "user_7yNMTpVy8ddRxYKGJqtk7e".

Benefits

Why take this approach?

• users: easy text selection of Stripe-like formatted values (fast double-click or keyboard-based) compared to UUIDs ( try double clicking the samples above!)
• database: using UUIDs in your database is likely much more efficient storage-wise as well as more performant
• prefixing:
  • allows for programmatic routing within a microservice architecture
  • easier to reason about debug output, logging messages (lower cognitive load/overhead)

Installation

fastapi-uuidbase62 installation is much the same as any other Python package.

pip install fastapi-uuidbase62

Python Support

Python 3.7, 3.8, 3.9, 3.10, 3.11 are supported and covered by the tox test configuration described below.

Usage

This package provides the ability to define a field on a Pydantic model that auto-serializes a UUID value to base62 and auto-prefixes a defined label. This serializes a UUID to a prefixed string when rendering a FastAPI response, and does the reverse when processing an incoming FastAPI request.

In the following example, take note of the following:

• UUIDBase62ModelMixin adds a to_uuidbase62 method to Model to easily convert a UUID or valid base62 prefixed value to a UUIDBase62 value
• the con_uuidbase62 function, which defines the autoprefixing and serializing UUID <-> str field
• the get_validated_uuidbase62_by_model dependency injection function providing validation/serialization on incoming base62-encoded parameters (path, header, query)
  • there is a similar get_validated_uuidbase62 function that does not rely on a Model class/field
• UUIDBase62 instance properties
  • uuidbase62_value.uuid: UUID matching the base62 encoded str
  • uuidbase62_value.base62_str: non-prefixed base62 string value
  • uuidbase62_value.value: prefixed base62 string value, same as str(uuidbase62_value)
  • uuidbase62_value.prefix: the prefix used for this UUIDBase62 instance

import uuid

from fastapi import FastAPI, Depends
from pydantic import BaseModel

from uuidbase62 import con_uuidbase62, UUIDBase62, UUIDBase62ModelMixin, get_validated_uuidbase62_by_model

app = FastAPI()


class Book(UUIDBase62ModelMixin, BaseModel):
    id: con_uuidbase62(prefix="book")
    title: str


@app.get("/", response_model=list[Book])
async def get_item_list():
    # fake fetching a list of books from the DB, yielding book IDs and titles
    return [{
        "id": uuid.uuid4(),
        "title": "Red Mars",
    }]


@app.get("/{item_id}", response_model=Book)
async def get_item(item_id: UUIDBase62 = Depends(get_validated_uuidbase62_by_model(Book, 'id', 'item_id'))):
    # fake fetching from the DB based on `item_id`, yielding a book ID and title
    return {
        "id": uuid.uuid4(),
        "title": "Green Mars",
    }


@app.post("/", response_model=Book)
async def create_item(item: Book):
    book_id = item.id  # UUIDBase62 value
    book_id.uuid  # uuid value
    book_id.base62_str  # non-prefixed base62 string value
    book_id.value  # prefixed base62 string value, same as str(book_id)
    book_id.prefix  # 'book'

    return item.dict()

Development

To set up a development environment, it is recommended to create a Python virtual environment, and then install development requirements. You should probably be using pyenv to manage your local Python versions:

# do this for each supported Python version, all are needed to run complete tests via tox
pyenv install 3.x.x 

# in the project directory, make supported Python versions available; first one listed is the default Python
pyenv local 3.10.x 3.7.x 3.8.x 3.9.x 3.11.x 

# create Python virtual environment
python -m venv venv

# install development dependencies
./venv/bin/pip install -r requirements.txt

Testing

fastapi-uuidbase62 is easily tested via the configuration set up with tox, which configures the tox command line tool:

# run tox, parallel mode
./venv/bin/tox -p

Contributing

Leverage Github issues, and do consider submitting fixes/improvements via pull requests on Github.