# Dependency Injection

Use dependencies when:

* They can't be declared in Pydantic validation and require additional logic
* The logic depends on external resources or could block in any other way
* Other dependencies need their results (it's a sub-dependency)
* The logic can be shared by multiple endpoints to do things like error early, handle authentication, etc.
* They need to handle cleanup (e.g., DB sessions, file handles), using dependencies with `yield`
* Their logic needs input data from the request, like headers, query parameters, etc.

## Dependencies with `yield` and `scope`

When using dependencies with `yield`, they can have a `scope` that defines when the exit code is run.

Use the default scope `"request"` to run the exit code after the response is sent back.

```python
from typing import Annotated

from fastapi import Depends, FastAPI

app = FastAPI()


def get_db():
    db = DBSession()
    try:
        yield db
    finally:
        db.close()


DBDep = Annotated[DBSession, Depends(get_db)]


@app.get("/items/")
async def read_items(db: DBDep):
    return db.query(Item).all()
```

Use the scope `"function"` when they should run the exit code after the response data is generated but before the response is sent back to the client.

```python
from typing import Annotated

from fastapi import Depends, FastAPI

app = FastAPI()


def get_username():
    try:
        yield "Rick"
    finally:
        print("Clean up before response is sent")

UserNameDep = Annotated[str, Depends(get_username, scope="function")]

@app.get("/users/me")
def get_user_me(username: UserNameDep):
    return username
```

## Class Dependencies

Avoid creating class dependencies when possible.

If a class is needed, instead create a regular function dependency that returns a class instance.

Do this:

```python
from dataclasses import dataclass
from typing import Annotated

from fastapi import Depends, FastAPI

app = FastAPI()


@dataclass
class DatabasePaginator:
    offset: int = 0
    limit: int = 100
    q: str | None = None

    def get_page(self) -> dict:
        # Simulate a page of data
        return {
            "offset": self.offset,
            "limit": self.limit,
            "q": self.q,
            "items": [],
        }


def get_db_paginator(
    offset: int = 0, limit: int = 100, q: str | None = None
) -> DatabasePaginator:
    return DatabasePaginator(offset=offset, limit=limit, q=q)


PaginatorDep = Annotated[DatabasePaginator, Depends(get_db_paginator)]


@app.get("/items/")
async def read_items(paginator: PaginatorDep):
    return paginator.get_page()
```

instead of this:

```python
# DO NOT DO THIS
from typing import Annotated

from fastapi import Depends, FastAPI

app = FastAPI()


class DatabasePaginator:
    def __init__(self, offset: int = 0, limit: int = 100, q: str | None = None):
        self.offset = offset
        self.limit = limit
        self.q = q

    def get_page(self) -> dict:
        # Simulate a page of data
        return {
            "offset": self.offset,
            "limit": self.limit,
            "q": self.q,
            "items": [],
        }


@app.get("/items/")
async def read_items(paginator: Annotated[DatabasePaginator, Depends()]):
    return paginator.get_page()
```

# Other Tools

## uv

If uv is available, use it to manage dependencies.

## Ruff

If Ruff is available, use it to lint and format the code. Consider enabling the FastAPI rules.

## ty

If ty is available, use it to check types.

## Asyncer

When needing to run blocking code inside of async functions, or async code inside of blocking functions, suggest using Asyncer.

Prefer it over AnyIO or asyncio.

Install:

```bash
uv add asyncer
```

Run blocking sync code inside of async with `asyncify()`:

```python
from asyncer import asyncify
from fastapi import FastAPI

app = FastAPI()


def do_blocking_work(name: str) -> str:
    # Some blocking I/O operation
    return f"Hello {name}"


@app.get("/items/")
async def read_items():
    result = await asyncify(do_blocking_work)(name="World")
    return {"message": result}
```

And run async code inside of blocking sync code with `syncify()`:

```python
from asyncer import syncify
from fastapi import FastAPI

app = FastAPI()


async def do_async_work(name: str) -> str:
    return f"Hello {name}"


@app.get("/items/")
def read_items():
    result = syncify(do_async_work)(name="World")
    return {"message": result}
```

## SQLModel for SQL databases

When working with SQL databases, prefer using SQLModel as it is integrated with Pydantic and will allow declaring data validation with the same models.

Prefer it over SQLAlchemy.

## HTTPX

Use HTTPX for handling HTTP communication (e.g. with other APIs). It supports sync and async usage.

Prefer it over Requests.

# Path Operations and Routing

## Including Routers

When declaring routers, prefer to add router-level parameters like prefix, tags, and shared dependencies to the router itself instead of in `include_router()`.

Do this:

```python
from fastapi import APIRouter, FastAPI

app = FastAPI()

router = APIRouter(prefix="/items", tags=["items"])


@router.get("/")
async def list_items():
    return []


app.include_router(router)
```

Instead of:

```python
# DO NOT DO THIS
from fastapi import APIRouter, FastAPI

app = FastAPI()

router = APIRouter()


@router.get("/")
async def list_items():
    return []


app.include_router(router, prefix="/items", tags=["items"])
```

There could be exceptions, but try to follow this convention.

Apply shared dependencies at the router level via `dependencies=[Depends(...)]`.

## Use one HTTP operation per function

Don't mix HTTP operations in a single function. Having one function per HTTP operation helps separate concerns and organize the code.

Do this:

```python
from fastapi import FastAPI
from pydantic import BaseModel

app = FastAPI()


class Item(BaseModel):
    name: str


@app.get("/items/")
async def list_items():
    return []


@app.post("/items/")
async def create_item(item: Item):
    return item
```

Instead of:

```python
# DO NOT DO THIS
from fastapi import FastAPI, Request
from pydantic import BaseModel

app = FastAPI()


class Item(BaseModel):
    name: str


@app.api_route("/items/", methods=["GET", "POST"])
async def handle_items(request: Request):
    if request.method == "GET":
        return []
```

# Pydantic

## Do not use Ellipsis

Do not use `...` as a default value for required parameters or model fields. It's not needed and not recommended.

Do this, without Ellipsis (`...`):

```python
from typing import Annotated

from fastapi import FastAPI, Query
from pydantic import BaseModel, Field

app = FastAPI()


class Item(BaseModel):
    name: str
    description: str | None = None
    price: float = Field(gt=0)


@app.post("/items/")
async def create_item(item: Item, project_id: Annotated[int, Query()]):
    return item
```

Instead of:

```python
# DO NOT DO THIS
from typing import Annotated

from fastapi import FastAPI, Query
from pydantic import BaseModel, Field

app = FastAPI()


class Item(BaseModel):
    name: str = ...
    description: str | None = None
    price: float = Field(..., gt=0)


@app.post("/items/")
async def create_item(item: Item, project_id: Annotated[int, Query(...)]):
    return item
```

## Do not use Pydantic RootModels

Do not use Pydantic `RootModel`; instead use regular type annotations with `Annotated` and Pydantic validation utilities.

For example, for a list with validations:

```python
from typing import Annotated

from fastapi import Body, FastAPI
from pydantic import Field

app = FastAPI()


@app.post("/items/")
async def create_items(items: Annotated[list[int], Field(min_length=1), Body()]):
    return items
```

Instead of:

```python
# DO NOT DO THIS
from typing import Annotated

from fastapi import FastAPI
from pydantic import Field, RootModel

app = FastAPI()


class ItemList(RootModel[Annotated[list[int], Field(min_length=1)]]):
    pass


@app.post("/items/")
async def create_items(items: ItemList):
    return items
```

FastAPI supports these type annotations and will create a Pydantic `TypeAdapter` for them, so types work normally without custom wrapper models.

# Responses

## Return Type or Response Model

When possible, include a return type. It will be used to validate, filter, document, and serialize the response.

```python
from fastapi import FastAPI
from pydantic import BaseModel

app = FastAPI()


class Item(BaseModel):
    name: str
    description: str | None = None


@app.get("/items/me")
async def get_item() -> Item:
    return Item(name="Plumbus", description="All-purpose home device")
```

Return types or response models filter data to avoid exposing sensitive information. They also let Pydantic serialize data on the Rust side for performance.

The return type doesn't have to be a Pydantic model. It can be a different type, like a list of integers, a dict, etc.

## When to use `response_model`

If the return type is not the same as the type that you want to use to validate, filter, or serialize, use the `response_model` parameter on the decorator.

```python
from typing import Any

from fastapi import FastAPI
from pydantic import BaseModel

app = FastAPI()


class Item(BaseModel):
    name: str
    description: str | None = None


@app.get("/items/me", response_model=Item)
async def get_item() -> Any:
    return {"name": "Foo", "description": "A very nice Item"}
```

This is particularly useful when filtering data to expose only the public fields and avoid exposing sensitive information.

```python
from typing import Any

from fastapi import FastAPI
from pydantic import BaseModel

app = FastAPI()


class InternalItem(BaseModel):
    name: str
    description: str | None = None
    secret_key: str


class Item(BaseModel):
    name: str
    description: str | None = None


@app.get("/items/me", response_model=Item)
async def get_item() -> Any:
    item = InternalItem(
        name="Foo", description="A very nice Item", secret_key="supersecret"
    )
    return item
```

# Streaming

## Stream JSON Lines

To stream JSON Lines, declare the return type and use `yield` to return the data.

```python
@app.get("/items/stream")
async def stream_items() -> AsyncIterable[Item]:
    for item in items:
        yield item
```

## Server-Sent Events (SSE)

To stream Server-Sent Events, use `response_class=EventSourceResponse` and `yield` items from the endpoint.

Plain objects are automatically JSON-serialized as `data:` fields, declare the return type so the serialization is done by Pydantic:

```python
from collections.abc import AsyncIterable

from fastapi import FastAPI
from fastapi.sse import EventSourceResponse
from pydantic import BaseModel

app = FastAPI()


class Item(BaseModel):
    name: str
    price: float


@app.get("/items/stream", response_class=EventSourceResponse)
async def stream_items() -> AsyncIterable[Item]:
    yield Item(name="Plumbus", price=32.99)
    yield Item(name="Portal Gun", price=999.99)
```

For full control over SSE fields (`event`, `id`, `retry`, `comment`), yield `ServerSentEvent` instances:

```python
from collections.abc import AsyncIterable

from fastapi import FastAPI
from fastapi.sse import EventSourceResponse, ServerSentEvent

app = FastAPI()


@app.get("/events", response_class=EventSourceResponse)
async def stream_events() -> AsyncIterable[ServerSentEvent]:
    yield ServerSentEvent(data={"status": "started"}, event="status", id="1")
    yield ServerSentEvent(data={"progress": 50}, event="progress", id="2")
```

Use `raw_data` instead of `data` to send pre-formatted strings without JSON encoding:

```python
yield ServerSentEvent(raw_data="plain text line", event="log")
```

## Stream bytes

To stream bytes, declare a `response_class=` of `StreamingResponse` or a sub-class, and use `yield` to return the data.

```python
from fastapi import FastAPI
from fastapi.responses import StreamingResponse
from app.utils import read_image

app = FastAPI()


class PNGStreamingResponse(StreamingResponse):
    media_type = "image/png"

@app.get("/image", response_class=PNGStreamingResponse)
def stream_image_no_async_no_annotation():
    with read_image() as image_file:
        yield from image_file
```

prefer this over returning a `StreamingResponse` directly:

```python
# DO NOT DO THIS

import anyio
from fastapi import FastAPI
from fastapi.responses import StreamingResponse
from app.utils import read_image

app = FastAPI()


class PNGStreamingResponse(StreamingResponse):
    media_type = "image/png"


@app.get("/")
async def main():
    return PNGStreamingResponse(read_image())
```