How FastAPI dependency injection works with Depends
Depends function, dependency declaration, shared logic, dependency chaining, dependency caching, use_cache parameter, function dependencies
Dependency Injection with Depends
FastAPI's Depends system lets you declare reusable logic — database sessions, auth checks, pagination — once and inject it into any route. Dependencies run before the route handler and can raise HTTPException to abort the request.
Basic dependency
from fastapi import Depends, FastAPI
app = FastAPI()
def get_pagination(skip: int = 0, limit: int = 10):
return {"skip": skip, "limit": limit}
@app.get("/items/")
def list_items(pagination: dict = Depends(get_pagination)):
return pagination
FastAPI calls get_pagination, resolves its own parameters from the request, and injects the return value. The route handler never deals with skip or limit directly.
Chaining dependencies
def verify_token(token: str):
if token != "secret":
raise HTTPException(status_code=401, detail="Invalid token")
return token
def get_current_user(token: str = Depends(verify_token)):
return {"user": "alice", "token": token}
@app.get("/profile")
def profile(user: dict = Depends(get_current_user)):
return user
FastAPI resolves the dependency tree automatically. get_current_user depends on verify_token, and the framework calls them in the right order.
Caching
By default, FastAPI calls a dependency once per request and caches the result. Pass use_cache=False to Depends if you need a fresh call every time it appears.