FastAPI依靠注入：参数共享与逻辑复用

扫描二维码关注或者微信搜一搜：编程智域 前端至全栈交换与成长
第一章：依靠注入核心原理

1.1 依靠树构建机制

1. from fastapi import Depends
4. def auth_service():
5.     return OAuth2Scheme()
8. def db_conn(auth: dict = Depends(auth_service)):
9.     return Database(creds=auth)
12. @app.get("/data")
13. async def get_data(conn=Depends(db_conn)):
14.     return conn.query()

复制代码

依靠树可视化：
graph TDget_data --> db_conn --> auth_service1.2 作用域控制

1. from fastapi import Depends, FastAPI
2. from sqlalchemy.orm import sessionmaker
4. SessionLocal = sessionmaker(autocommit=False)
7. # 请求级作用域
8. def get_db():
9.     db = SessionLocal()
10.     try:
11.         yield db
12.     finally:
13.         db.close()
16. # 应用级单例
17. cache = LRUCache(size=100)
20. def get_cache():
21.     return cache

复制代码

第二章：Pydantic深度集成

2.1 动态模子注入

1. from pydantic import create_model
4. def dynamic_model(fields: dict):
5.     return create_model('DynamicModel', **fields)
8. class FilterFactory:
9.     @classmethod
10.     def create(cls, model: BaseModel):
11.         class QueryParams(model):
12.             limit: int = 100
13.             offset: int = 0
15.         return QueryParams
18. @app.get("/search")
19. async def search(params=Depends(FilterFactory.create(User))):
20.     return params.dict()

复制代码

2.2 校验逻辑复用

[code]from pydantic import validator, root_validatorclass GeoValidator:    @classmethod    def lat_validator(cls, v):        if not -90