First, import `BackgroundTasks` and define a parameter in your *path operation function* with a type declaration of `BackgroundTasks`:

{* ../../docs_src/background_tasks/tutorial001_py310.py hl[1,13] *}

**FastAPI** will create the object of type `BackgroundTasks` for you and pass it as that parameter.

## Create a task function { #create-a-task-function }

Create a function to be run as the background task.

在某些情況下，你可能仍需要使用 Pydantic 版本的 `dataclasses`。例如，當自動產生的 API 文件出現錯誤時。

這種情況下，你可以把標準的 `dataclasses` 直接換成 `pydantic.dataclasses`，它是可直接替換（drop-in replacement）的：

{* ../../docs_src/dataclasses_/tutorial003_py310.py hl[1,4,7:10,13:16,22:24,27] *}

1. 我們仍然從標準的 `dataclasses` 匯入 `field`。
2. `pydantic.dataclasses` 是 `dataclasses` 的可直接替換版本。
3. `Author` dataclass 內含一個 `Item` dataclass 的清單。
4. `Author` dataclass 被用作 `response_model` 參數。

    where an integer (but not int typed) is used. Encoding this as a float
    results in failed round-tripping between encode and parse.
    Our Id type is a prime example of this.

    >>> decimal_encoder(Decimal("1.0"))
    1.0

    >>> decimal_encoder(Decimal("1"))
    1

    >>> decimal_encoder(Decimal("NaN"))
    nan
    """
    exponent = dec_value.as_tuple().exponent
    if isinstance(exponent, int) and exponent >= 0:

    assertThat(Doubles.contains(EMPTY, 1.0)).isFalse();
    assertThat(Doubles.contains(ARRAY1, 2.0)).isFalse();
    assertThat(Doubles.contains(ARRAY234, 1.0)).isFalse();
    assertThat(Doubles.contains(new double[] {-1.0}, -1.0)).isTrue();
    assertThat(Doubles.contains(ARRAY234, 2.0)).isTrue();
    assertThat(Doubles.contains(ARRAY234, 3.0)).isTrue();
    assertThat(Doubles.contains(ARRAY234, 4.0)).isTrue();

   * (10, ∞) => "baz"} then {@code rangeMap.subRangeMap(Range.open(3, 12))} would return a range map
   * with the entries {@code (3, 5] => "foo", (6, 8) => "bar", (10, 12) => "baz"}.
   *
   * <p>The returned range map supports all optional operations that this range map supports, except
   * for {@code asMapOfRanges().iterator().remove()}.
   *

Если присмотреться, функция декорирована `@asynccontextmanager`.

Это превращает функцию в «асинхронный менеджер контекста».

{* ../../docs_src/events/tutorial003_py310.py hl[1,13] *}

Менеджер контекста в Python — это то, что можно использовать в операторе `with`. Например, `open()` можно использовать как менеджер контекста:

```Python
with open("file.txt") as file:
    file.read()
```

Primeiro, importe `BackgroundTasks` e defina um parâmetro na sua *função de operação de rota* com uma declaração de tipo `BackgroundTasks`:

{* ../../docs_src/background_tasks/tutorial001_py310.py hl[1,13] *}

O **FastAPI** criará o objeto do tipo `BackgroundTasks` para você e o passará como esse parâmetro.

## Crie uma função de tarefa { #create-a-task-function }

If you check, the function is decorated with an `@asynccontextmanager`.

That converts the function into something called an "**async context manager**".

{* ../../docs_src/events/tutorial003_py310.py hl[1,13] *}

A **context manager** in Python is something that you can use in a `with` statement, for example, `open()` can be used as a context manager:

```Python
with open("file.txt") as file:
    file.read()
```

Se você verificar, a função está decorada com um `@asynccontextmanager`.

Isso converte a função em algo chamado "**gerenciador de contexto assíncrono**".

{* ../../docs_src/events/tutorial003_py310.py hl[1,13] *}

Um **gerenciador de contexto** em Python é algo que você pode usar em uma declaração `with`, por exemplo, `open()` pode ser usado como um gerenciador de contexto:

```Python
with open("file.txt") as file:

	hammerRWMutex(t, 1, 1, n)
	hammerRWMutex(t, 1, 3, n)
	hammerRWMutex(t, 1, 10, n)
	hammerRWMutex(t, 4, 1, n)
	hammerRWMutex(t, 4, 3, n)
	hammerRWMutex(t, 4, 10, n)
	hammerRWMutex(t, 10, 1, n)
	hammerRWMutex(t, 10, 3, n)
	hammerRWMutex(t, 10, 10, n)
	hammerRWMutex(t, 10, 5, n)
}

// Borrowed from rwmutex_test.go
func TestUnlockPanic(t *testing.T) {
	defer func() {
		if recover() == nil {