Async endpoints in Python

An async def handler runs on the event loop, so awaits do not block it.

import asyncio

from fastapi import FastAPI
from fastapi.testclient import TestClient

app = FastAPI()


async def fetch(name: str) -> dict:
    await asyncio.sleep(0.01)
    return {"name": name, "steps": len(name) * 10}


@app.get("/summary")
async def summary():
    tours = await asyncio.gather(fetch("basics"), fetch("pythonic"))
    return {"tours": tours, "total": sum(t["steps"] for t in tours)}


client = TestClient(app)
print(client.get("/summary").json())

How it works

1. Use async when the work awaits; plain def runs in a threadpool.
2. asyncio.gather fans out inside one request.
3. The TestClient drives async handlers the same way.

Keywords and builtins used here

• async
• await
• def
• dict
• fetch
• for
• len
• print
• return
• str
• sum
• summary

The run, in numbers

Lines

21

Characters to type

451

Tokens

137

Three-star pace

110 tpm

At the three-star pace of 110 tokens a minute, this run takes about 75 seconds.

Type this snippet

Step 2 of 3 in Dependencies & async, step 11 of 19 in Web services & data access.

← Previous Next →