znazz75.de

Writing Your First REST API in Python with FastAPI

Claude

FastAPI has become the default choice for new Python REST APIs, largely because it gets three things right at once: it’s genuinely fast (built on Starlette and Pydantic), it generates interactive API documentation automatically, and it catches type errors before they become runtime bugs.

A Minimal API

pip install fastapi uvicorn

from fastapi import FastAPI, HTTPException
from pydantic import BaseModel

app = FastAPI()

class Item(BaseModel):
    name: str
    price: float
    in_stock: bool = True

items: dict[int, Item] = {}

@app.get("/items/{item_id}")
def get_item(item_id: int):
    if item_id not in items:
        raise HTTPException(status_code=404, detail="Item not found")
    return items[item_id]

@app.post("/items/{item_id}")
def create_item(item_id: int, item: Item):
    items[item_id] = item
    return {"id": item_id, "item": item}

Running It

uvicorn main:app --reload

# Interactive docs are generated automatically at:
# http://127.0.0.1:8000/docs

Why the Pydantic Models Matter

Declaring Item as a Pydantic model isn’t just documentation — FastAPI validates every incoming request against it automatically. Send a request missing the price field or with a string where a float is expected, and you get a clear 422 error with details, before your own code ever runs.

Adding Async for I/O-Bound Endpoints

import httpx

@app.get("/weather/{city}")
async def get_weather(city: str):
    async with httpx.AsyncClient() as client:
        resp = await client.get(f"https://api.example.com/weather/{city}")
        return resp.json()

Making an endpoint async def lets FastAPI serve other requests while this one is waiting on a slow network call — genuinely useful once your API starts calling other services rather than just an in-memory dict.

, ,

Recommended

Categories