```

////

//// tab | 🐍 3️⃣.1️⃣0️⃣ & 🔛

```Python hl_lines="6-7"
{!> ../../docs_src/dependencies/tutorial005_py310.py!}
```

////

⚫️ 📣 📦 🔢 🔢 `q` `str`, & ⤴️ ⚫️ 📨 ⚫️.

👉 🙅 (🚫 📶 ⚠), ✋️ 🔜 ℹ 👥 🎯 🔛 ❔ 🎧-🔗 👷.

## 🥈 🔗, "☑" & "⚓️"

⤴️ 👆 💪 ✍ ➕1️⃣ 🔗 🔢 ("☑") 👈 🎏 🕰 📣 🔗 🚮 👍 (⚫️ "⚓️" 💁‍♂️):

//// tab | 🐍 3️⃣.6️⃣ & 🔛

FastAPI 支持创建含**子依赖项**的依赖项。

并且，可以按需声明任意**深度**的子依赖项嵌套层级。

**FastAPI** 负责处理解析不同深度的子依赖项。

### 第一层依赖项

下列代码创建了第一层依赖项：

```Python hl_lines="8-9"
{!../../docs_src/dependencies/tutorial005.py!}
```

这段代码声明了类型为 `str` 的可选查询参数 `q`，然后返回这个查询参数。

这个函数很简单（不过也没什么用），但却有助于让我们专注于了解子依赖项的工作方式。

### 第二层依赖项

接下来，创建另一个依赖项函数，并同时用该依赖项自身再声明一个依赖项（所以这也是一个「依赖项」）：

```Python hl_lines="13"
{!../../docs_src/dependencies/tutorial005.py!}
```

from fastapi import FastAPI
from fastapi.testclient import TestClient
from pydantic import BaseModel

app = FastAPI()


class Product(BaseModel):
    name: str
    description: str = None  # type: ignore
    price: float


@app.get("/product")
async def create_item(product: Product):
    return product


client = TestClient(app)


def test_get_with_body():

from fastapi import FastAPI, Path, Query, status
from fastapi.testclient import TestClient

app = FastAPI()


@app.get("/{repeated_alias}")
def get_parameters_with_repeated_aliases(
    path: str = Path(..., alias="repeated_alias"),
    query: str = Query(..., alias="repeated_alias"),
):
    return {"path": path, "query": query}


client = TestClient(app)


def test_get_parameters():

@app.get("/float/{param:float}")
def float_convertor(param: float = Path()):
    return {"float": param}


@app.get("/path/{param:path}")
def path_convertor(param: str = Path()):
    return {"path": param}


@app.get("/query/")
def query_convertor(param: str = Query()):
    return {"query": param}


client = TestClient(app)


def test_route_converters_int():
    # Test integer conversion

	// The size field ends at the first space.
	nStr, rest, ok := strings.Cut(s, " ")
	if !ok {
		return "", "", s, ErrHeader
	}

	// Parse the first token as a decimal integer.
	n, perr := strconv.ParseInt(nStr, 10, 0) // Intentionally parse as native int
	if perr != nil || n < 5 || n > int64(len(s)) {
		return "", "", s, ErrHeader
	}
	n -= int64(len(nStr) + 1) // convert from index in s to index in rest
	if n <= 0 {

      checkPositionIndexes(off, off + len, cbuf.length);
    }

    @Override
    public void write(String str) {
      checkNotNull(str);
    }

    @Override
    public void write(String str, int off, int len) {
      checkPositionIndexes(off, off + len, str.length());
    }

    @Override
    public Writer append(@CheckForNull CharSequence csq) {
      return this;
    }

from typing_extensions import Annotated


class Hero(SQLModel, table=True):
    id: Union[int, None] = Field(default=None, primary_key=True)
    name: str = Field(index=True)
    age: Union[int, None] = Field(default=None, index=True)
    secret_name: str


sqlite_file_name = "database.db"
sqlite_url = f"sqlite:///{sqlite_file_name}"

connect_args = {"check_same_thread": False}

from pydantic import BaseModel

app = FastAPI()

oid = OpenIdConnect(openIdConnectUrl="/openid", auto_error=False)


class User(BaseModel):
    username: str


def get_current_user(oauth_header: Optional[str] = Security(oid)):
    if oauth_header is None:
        return None
    user = User(username=oauth_header)
    return user


@app.get("/users/me")

### Создание класса `Enum`

Импортируйте `Enum` и создайте подкласс, который наследуется от `str` и `Enum`.

Мы наследуемся от `str`, чтобы документация API могла понять, что значения должны быть типа `string` и отображалась правильно.

Затем создайте атрибуты класса с фиксированными допустимыми значениями: