Python quick use of REST API

                                                                                                                                                                                                                        

#​

Fastapi is a python-based framework that encourages the use of Pydantic and OpenAPI (formerly Swagger) for documentation, Docker for rapid development and deployment, and simple testing based on the Starlette framework.

It provides many benefits such as automatic OpenAPI validation and documentation without adding unnecessary bloat. I think there's a good balance between not providing any built-in functionality and providing too much built-in functionality.

Getting Started

Install fastapi and ASGI server (e.g. uvicorn):

Make sure you are using Python 3.6.7 If pip and python give you a python 2 version, you may have to use pip3

and

python3

. Also, check out my article on getting started with python.

pip install fastapi uvicorn

and add the old "hello world" in the main.py

file:

from fastapi import FastAPI

app = FastAPI()

@app.get("/")
def home():
    return {"Hello": "World"}

Run development

then To run for development, you can run uvicorn main:app --reload That’s all a simple server does! Now you can check //localhost:8000/ to see the "Home Page". And, as you can see, the JSON response "just works"! You can also get Swagger UI for free at //localhost:8000/docs.

Validation

As mentioned before, it is easy to validate the data (and generate Swagger documentation for the accepted data formats). Just add the

Query

import from fastapi and use it to force validation: <pre class="brush:php;toolbar:false">from fastapi import FastAPI, Query @app.get('/user') async def user(     *,     user_id: int = Query(..., title=&quot;The ID of the user to get&quot;, gt=0) ):   return { 'user_id': user_id }</pre> The first parameter ... is the default value if the user does not provide a value This default value is provided. If set to None, there is no default value and the parameter is optional. In order that there is no default value and the parameter is mandatory, use Ellipsis, or

...

instead. If you run this code, you will automatically see the update on the swagger UI:

The Swagger UI allows you to view the new /user route and use a specific Make the request with your user ID

If you enter any user ID, you will see that it will automatically perform the request for you, for example //localhost:8000/user?user_id=1. In the page you can only see the user ID echoed!

If you want to use the path parameter instead (so that it is

/user/1, then just enter and use Path

instead of

Query. It is also possible to combine both Post routes

If you have a

POST route, you can just define the input

@app.post('/user/update')
async def update_user(
    *,
    user_id: int,
    really_update: int = Query(...)
):
    pass

in this case, you can see that user_id is only defined as an integer without Query or Path; which means it will be in the POST request body. If You accept more complex data structures, such as JSON data, you should look into request models.

Request and Response Models You can use Pydantic models to record and declare detailed Request and response models. Not only does this allow you to have automatic OpenAPI documentation for all your models, but it also validates the request and response models to ensure that any POST data entered is correct and that the data returned conforms to the model.

Just declare the model like this:

from pydantic import BaseModel

class User(BaseModel):
    id:: int
    name: str
    email: str

Then if you want to take the user model as input you can do this:

async def update_user(*, user: User):
    pass

Or if you want to Used as output:

@app.get('/user')
async def user(
    *,
    user_id: int = Query(..., title="The ID of the user to get", gt=0),
    response_model=User
):
  my_user = get_user(user_id)
  return my_user

Routing and decomposing a larger API

You can use ###APIRouter### to decompose the api into routes. For example, I have Found this in my API ###app/routers/v1/__init__.py######

from fastapi import APIRouter
from .user import router as user_router

router = APIRouter()

router.include_router(
    user_router,
    prefix='/user',
    tags=['users'],
)

###Then you can find this in ###app/routers/v1/user.py## # using the user code above - just import ###APIRouter### and use ###@ router.get('/')### instead of ###@ app.get(' /user')###. It will automatically route to ###/user /### because the route is prefix relative. ###

from fastapi import APIRouter

router = APIRouter()

@router.get('/')
async def user(
    *,
    user_id: int = Query(..., title="The ID of the user to get", gt=0),
    response_model=User
):
  my_user = get_user(user_id)
  return my_user

###Finally, use all ## in your application #v1###router, just edit ###main.py### to: ###

from fastapi import FastAPI
from app.routers import v1

app = FastAPI()

app.include_router(
    v1.router,
    prefix="/api/v1"
)

### You can chain routers at will this way, allowing you to split large applications and have versions ized API. #########Dockerizing and Deploying#########One of the authors of Fastapi makes Dockerizing surprisingly easy! The default ###Dockerfile### is 2 OK!###

FROM tiangolo/uvicorn-gunicorn-fastapi:python3.7

COPY ./app /app

是否想通过自动重新加载进行 Dockerize 开发？这是我在撰写文件中使用的秘方：

version: "3"
services:
  test-api:
    build: ..
    entrypoint: '/start-reload.sh'
    ports:
        - 8080:80
    volumes:
        - ./:/app

这会将当前目录挂载为app并将在任何更改时自动重新加载。您可能还想将app / app用于更大的应用程序。

有用的网址

所有这些信息都来自 Fastapi网站，该文档具有出色的文档，我鼓励您阅读。此外，作者在 Gitter 上非常活跃并乐于助人！

结论

就是这样-我希望本指南对您有所帮助，并且您会像我一样喜欢使用 Fastapi。

推荐教程：Python教程

The above is the detailed content of Python quick use of REST API. For more information, please follow other related articles on the PHP Chinese website!