Also, the best approach was to use already existing standards.

So, before even starting to code **FastAPI**, I spent several months studying the specs for OpenAPI, JSON Schema, OAuth2, etc. Understanding their relationship, overlap, and differences.

## Design { #design }

Then I spent some time designing the developer "API" I wanted to have as a user (as a developer using FastAPI).

To update an item you can use the [HTTP `PUT`](https://developer.mozilla.org/en-US/docs/Web/HTTP/Methods/PUT) operation.

You can use the `jsonable_encoder` to convert the input data to data that can be stored as JSON (e.g. with a NoSQL database). For example, converting `datetime` to `str`.

{* ../../docs_src/body_updates/tutorial001_py310.py hl[28:33] *}

`PUT` is used to receive data that should replace the existing data.

     -config-ep http://127.0.0.1:5556/dex/.well-known/openid-configuration \
     -cscopes groups,openid,email,profile
```

```
~ mc admin policy create admin allaccess.json
```

Contents of `allaccess.json`

```json
{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": [
        "s3:*"
      ],
      "Resource": [
        "arn:aws:s3:::*"
      ]

    }

    @Test
    void testCreateDir_ok() {
        // Create a directory
        Response response = given().contentType("application/json")
                .header("Authorization", getTestToken())
                .body("{\"name\":\"" + NAME_PREFIX + "testdir\"}")
                .when()
                .post(getApiPath() + "/createDir/");

* docs: documentação
* FastAPI app: aplicação FastAPI
* framework: framework (do not translate)
* feature: funcionalidade
* guides: tutoriais
* I/O (as in "input and output"): I/O (do not translate to "E/S")
* JSON Schema: JSON Schema
* library: biblioteca
* lifespan: lifespan (do not translate to "vida útil")
* list (as in Python list): list
* Machine Learning: Aprendizado de Máquina

	}
	return n.Time, nil
}

func (n DeletedAt) MarshalJSON() ([]byte, error) {
	if n.Valid {
		return json.Marshal(n.Time)
	}
	return json.Marshal(nil)
}

func (n *DeletedAt) UnmarshalJSON(b []byte) error {
	if string(b) == "null" {
		n.Valid = false
		return nil
	}
	err := json.Unmarshal(b, &n.Time)
	if err == nil {
		n.Valid = true
	}
	return err
}

        ("inf", 422),
        ("-inf", 422),
        ("nan", 422),
        ("0", 200),
        ("342", 200),
    ],
)
def test_allow_inf_nan_body(value: str, code: int):
    response = client.post("/", json=value)

        <build-type id="bt270451" />
      </build-types>
    </webhook>
    <webhook url="https://gradle-bot.grdev.net/teamcity" enabled="true" template="legacy-tailored-json">
      <states>
        <state type="buildAddedToQueue" enabled="false" />
        <state type="buildRemovedFromQueue" enabled="false" />
        <state type="buildStarted" enabled="false" />

FastAPI 會使用這個回傳型別來：

* 驗證回傳的資料。
    * 如果資料無效（例如缺少欄位），代表你的應用程式程式碼有問題，沒有回傳應該回傳的內容，FastAPI 會回傳伺服器錯誤，而不是回傳不正確的資料。如此你和你的用戶端都能確定會收到預期的資料與資料結構。
* 在 OpenAPI 的「路徑操作」中為回應新增 JSON Schema。
    * 這會被自動文件使用。
    * 也會被自動用戶端程式碼產生工具使用。
* 使用 Pydantic 將回傳資料**序列化**為 JSON，而 Pydantic 是用 **Rust** 撰寫的，因此會 **更快很多**。

但更重要的是：

* 它會將輸出資料限制並過濾為回傳型別中定義的內容。
    * 這對安全性特別重要，下面會再看到更多細節。

## `response_model` 參數 { #response-model-parameter }

 - [Git](https://github.com/codelibs/fess-ds-git)
 - [Gitbucket](https://github.com/codelibs/fess-ds-gitbucket)
 - [G Suite](https://github.com/codelibs/fess-ds-gsuite)
 - [JSON](https://github.com/codelibs/fess-ds-json)
 - [Office 365](https://github.com/codelibs/fess-ds-office365)
 - [S3](https://github.com/codelibs/fess-ds-s3)
 - [Salesforce](https://github.com/codelibs/fess-ds-salesforce)