options.addOption(Option.builder(Character.toString(VERBOSE))
                .longOpt("verbose")
                .desc("Produce execution verbose output")
                .build());
        options.addOption(Option.builder(Character.toString(ERRORS))
                .longOpt("errors")
                .desc("Produce execution error messages")
                .build());
        options.addOption(Option.builder(Character.toString(NON_RECURSIVE))

### 4. Excluded Types

- Already compressed objects are not fit for compression since they do not have compressible patterns.
Such objects do not produce efficient [`LZ compression`](https://en.wikipedia.org/wiki/LZ77_and_LZ78)
which is a fitness factor for a lossless data compression.

Pre-compressed input typically compresses in excess of 2GiB/s per core,

- Follow through the wizard as above for Veeam Backup and Replication as the steps are the same between both products

- Under Backup Infrastructure -> Backup Repositories, right click and "Add Backup Repository"

- Follow the wizard.  Under the "Object Storage Backup Repository" section, choose the MinIO object storage you created above

        String result3 = crawlingInfoHelper.generateId(input3);
        assertNotNull(result3);
        assertEquals(128, result3.length());

        // Test that different inputs produce different results
        assertFalse(result1.equals(result2));
        assertFalse(result2.equals(result3));
        assertFalse(result1.equals(result3));
    }

    @Test

Il suffit de traiter la requête dans un bloc `try`/`except` :

{* ../../docs_src/custom_request_and_route/tutorial002_an_py310.py hl[14,16] *}

Si une exception se produit, l'instance de `Request` sera toujours dans la portée, ce qui nous permet de lire et d'utiliser le corps de la requête lors du traitement de l'erreur :

{* ../../docs_src/custom_request_and_route/tutorial002_an_py310.py hl[17:19] *}

Importez `EventSourceResponse` depuis `fastapi.sse` :

{* ../../docs_src/server_sent_events/tutorial001_py310.py ln[1:25] hl[4,22] *}

Chaque élément produit avec `yield` est encodé en JSON et envoyé dans le champ `data:` d’un événement SSE.

  }

  /**
   * This test creates a long random sequence of inputs, then a lot of differently configured sinks
   * process it; all should produce the same answer, the only difference should be the number of
   * process()/processRemaining() invocations, due to alignment.
   */
  @AndroidIncompatible // slow. TODO(cpovirk): Maybe just reduce iterations under Android.

{* ../../docs_src/stream_data/tutorial001_py310.py ln[32:35] hl[33] *}

This also means that with `StreamingResponse` you have the **freedom** and **responsibility** to produce and encode the data bytes exactly as you need them to be sent, independent of the type annotations. 🤓

### Stream Bytes { #stream-bytes }

	VPGATHERDQ X2, 664(X7*8), X2    // ERROR "mask, index, and destination registers should be distinct"
	VPGATHERDQ X7, 664(X2*8), X2    // ERROR "mask, index, and destination registers should be distinct"
	// Non-X0 for Yxr0 should produce an error
	BLENDVPD X1, (BX), X2           // ERROR "invalid instruction"
	// Check offset overflow. Must fit in int32.
	MOVQ 2147483647+1(AX), AX       // ERROR "offset too large"

     */
    LlmChatResponse generateAnswer(String userMessage, List<Map<String, Object>> documents, List<LlmMessage> history);

    /**
     * Regenerates a search query when the previous query failed to produce relevant results.
     *
     * @param userMessage the user's original message
     * @param failedQuery the query that failed
     * @param failureReason the reason for failure ("no_results" or "no_relevant_results")