}

        return new PacUnicodeString(length, maxLength, pointer);
    }

    /**
     * Reads a string with length prefix from the stream.
     * @return the decoded string
     * @throws IOException if an I/O error occurs
     * @throws PACDecodingException if the string structure is malformed
     */
    public String readString() throws IOException, PACDecodingException {

```

### Using xl-meta

Executing `xl-meta` will look for an `xl.meta` in the current folder and decode it to JSON. It is also possible to specify multiple files or wildcards, for example `xl-meta ./**/xl.meta` will output decoded metadata recursively. It is possible to view what inline data is stored inline in the metadata using `--data` parameter `xl-meta -data xl.json` will display an id -> data size. To export inline data to a file use the `--export` option.

            byte[] ident = new byte[] { 0, 0, 0, 0, 0, 5 };
            rpc.sid_t st = buildSidT((byte) 1, ident, 10, 20);

            byte[] bytes = SID.toByteArray(st);

            // Act: construct from bytes and re-encode
            SID sid = new SID(bytes, 0);
            byte[] roundTrip = sid.toByteArray();

            // Assert
            assertArrayEquals(bytes, roundTrip);

Если вы не объявите модель ответа, FastAPI использует `jsonable_encoder`, описанный в разделе [JSON-совместимый энкодер](../tutorial/encoder.md), и поместит результат в `JSONResponse`.

Якщо ви не оголосите модель відповіді, FastAPI використає `jsonable_encoder`, пояснений у [Сумісний кодувальник JSON](../tutorial/encoder.md), і помістить результат у `JSONResponse`.

///

## JSON 响应 { #json-responses }

默认情况下 FastAPI 返回 JSON 响应。

如果你声明了一个[响应模型](../tutorial/response-model.md)，FastAPI 会使用 Pydantic 将数据序列化为 JSON。

如果你没有声明响应模型，FastAPI 会使用 [JSON 兼容编码器](../tutorial/encoder.md) 中解释的 `jsonable_encoder`，并把结果放进一个 `JSONResponse`。

* le service cloud

* le développement
* les étapes de développement

* le dict
* le dictionnaire
* l’énumération
* l’enum
* le membre d’enum

* l’encodeur
* le décodeur
* encoder
* décoder

* l’exception
* lever

* l’expression
* l’instruction

* le frontend
* le backend

* la discussion GitHub
* le ticket GitHub

* la performance

	AmzSignedHeaders        = "X-Amz-SignedHeaders"
	AmzSignature            = "X-Amz-Signature"
	AmzCredential           = "X-Amz-Credential"
	AmzSecurityToken        = "X-Amz-Security-Token"
	AmzDecodedContentLength = "X-Amz-Decoded-Content-Length"
	AmzTrailer              = "X-Amz-Trailer"
	AmzMaxParts             = "X-Amz-Max-Parts"
	AmzPartNumberMarker     = "X-Amz-Part-Number-Marker"

            // Then: Should not throw exception
            assertDoesNotThrow(() -> handle.doSendFragment(new byte[10], 0, 10));
        }

        @Test
        @DisplayName("Should handle encode message operation")
        void testEncodeMessage() throws NdrException, DcerpcException {
            // Given: Mock message
            DcerpcMessage mockMessage = mock(DcerpcMessage.class);

* Seriously impressive performance. It is [one of the fastest Python frameworks available, on par with **NodeJS** and **Go**](https://github.com/encode/starlette#performance).
* **WebSocket** support.
* In-process background tasks.
* Startup and shutdown events.
* Test client built on HTTPX.
* **CORS**, GZip, Static Files, Streaming responses.