查看 `/docs` 时，你会看到字段 `data` 期望接收 base64 编码的字节：

<div class="screenshot">
<img src="/img/tutorial/json-base64-bytes/image01.png">
</div>

你可以发送如下请求：

```json
{
    "description": "Some data",
    "data": "aGVsbG8="
}
```

/// tip | 提示

`aGVsbG8=` 是 `hello` 的 base64 编码。

///

随后 Pydantic 会解码该 base64 字符串，并在模型的 `data` 字段中提供原始字节。

你将会收到类似的响应：

```json
{

    }

    /**
     * Processes a document through the ingest pipeline.
     * Applies all available ingesters to transform the document data.
     *
     * @param paramMap the data store parameters
     * @param dataMap the document data to process
     * @return the processed document data
     */
    protected Map<String, Object> ingest(final DataStoreParams paramMap, final Map<String, Object> dataMap) {

    name: str
    price: float


class Shop(BaseModel):
    name: str


@app.post("/products")
async def create_product(data: Product = Body(media_type=media_type, embed=True)):
    pass  # pragma: no cover


@app.post("/shops")
async def create_shop(
    data: Shop = Body(media_type=media_type),
    included: list[Product] = Body(default=[], media_type=media_type),
):
    pass  # pragma: no cover

            var data = JSON.parse(e.data);
            if (data.sources && data.sources.length > 0 && messageElement) {
                addSourcesToMessage(messageElement, data.sources);
            }
        });

        eventSource.addEventListener('done', function(e) {
            var data = JSON.parse(e.data);
            state.sessionId = data.sessionId;

fi

minio server -S /tmp/no-certs --address ":9001" "http://localhost:9001/tmp/multisitea/data/disterasure/xl{1...4}" \
	"http://localhost:9002/tmp/multisitea/data/disterasure/xl{5...8}" >/tmp/sitea_1.log 2>&1 &

minio server -S /tmp/no-certs --address ":9002" "http://localhost:9001/tmp/multisitea/data/disterasure/xl{1...4}" \
	"http://localhost:9002/tmp/multisitea/data/disterasure/xl{5...8}" >/tmp/sitea_2.log 2>&1 &

[Pydantic](https://docs.pydantic.dev/) is a Python library to perform data validation.

You declare the "shape" of the data as classes with attributes.

And each attribute has a type.

Then you create an instance of that class with some values and it will validate the values, convert them to the appropriate type (if that's the case) and give you an object with all the data.

And you get all the editor support with that resulting object.

            response.readBytesWireFormat(buffer, 0);
        });

        // The validation correctly detects when buffer extends beyond available data
        assertTrue(exception.getMessage().contains("Security buffer extends beyond available data"),
                "Should detect security buffer overflow: " + exception.getMessage());
    }

    /**
     * Test security buffer validation against buffer overflow.

                final char[] data = line.toCharArray();
                int ip, i, j;
                Name name;
                NbtAddress addr;
                char c = '.';

                ip = i = 0;
                for (; i < data.length && c == '.'; i++) {
                    int b = 0x00;

                    for (; i < data.length && (c = data[i]) >= 48 && c <= 57; i++) {

	object := "object"

	data := make([]byte, 1*humanize.MiByte)
	length := int64(len(data))
	_, err = rand.Read(data)
	if err != nil {
		t.Fatal(err)
	}

	oi1, err := obj.PutObject(ctx, bucket, object, mustGetPutObjReader(t, bytes.NewReader(data), length, "", ""), ObjectOptions{})
	if err != nil {
		t.Fatal(err)
	}
	// 2nd version.
	_, _ = rand.Read(data)

        for (Object[] data : testData) {
            String endpoint = (String) data[0];
            String pipeName = (String) data[1];
            String expectedUuid = (String) data[2];
            int expectedMajor = (int) data[3];
            int expectedMinor = (int) data[4];

            // The interfaces are added in setUp()