Calls are executed in one of two ways:

 * **Synchronous:** your thread blocks until the response is readable.
 * **Asynchronous:** you enqueue the request on any thread, and get [called back](https://square.github.io/okhttp/5.x/okhttp/okhttp3/-callback/) on another thread when the response is readable.

     */
    RelevanceEvaluationResult evaluateResults(String userMessage, String query, List<Map<String, Object>> searchResults);

    /**
     * Generates an answer using document content (synchronous version for non-enhanced flow).
     *
     * @param userMessage the user's message
     * @param documents the documents with content
     * @param history the conversation history
     * @return the chat response

                threadName.get().contains("ForkJoinPool") || !threadName.get().equals(Thread.currentThread().getName()));
        pool.shutdown();
    }

    // --- Synchronous execute() test ---

    @Test
    public void test_SynchronousExecute_ReturnsResponse() throws Exception {
        // ## Arrange ##
        String body = "sync response body";

        }
        return getAvailableClient().evaluateResults(userMessage, query, searchResults);
    }

    /**
     * Generates an answer using document content (synchronous).
     *
     * @param userMessage the user's message
     * @param documents the documents with content
     * @param history the conversation history
     * @return the chat response

之所以稱為「非同步」，是因為電腦/程式不需要與那些耗時的任務「同步」，等待任務完成的精確時間，然後才能取得結果並繼續工作。

相反地，非同步系統在任務完成後，可以讓任務稍微等一下（幾微秒），等待電腦/程式完成手頭上的其他工作，然後再回來取得結果繼續進行。

相對於「非同步」（asynchronous），「同步」（synchronous）也常被稱作「順序性」（sequential），因為電腦/程式會依序執行所有步驟，即便這些步驟涉及等待，才會切換到其他任務。

### 並行與漢堡 { #concurrency-and-burgers }

上述非同步程式碼的概念有時也被稱為「並行」，它不同於「平行」。

並行和平行都與 "不同的事情或多或少同時發生" 有關。

但並行和平行之間的細節是完全不同的。

=== ":material-language-kotlin: Kotlin"
    ```kotlin
      private val executor = Executors.newScheduledThreadPool(1)
      private val client = OkHttpClient()

can be controlled with the `x509rsacrt` setting. Using `x509rsacrt=0` restores
the Go 1.23 behavior.

### Go 1.23

Go 1.23 changed the channels created by package time to be unbuffered
(synchronous), which makes correct use of the [`Timer.Stop`](/pkg/time/#Timer.Stop)
and [`Timer.Reset`](/pkg/time/#Timer.Reset) method results much easier.
The [`asynctimerchan` setting](/pkg/time/#NewTimer) disables this change.

    }

    /**
     * Returns the final value of the associated {@link ClosingFuture}, or throws an exception as
     * {@link Future#get()} would.
     *
     * <p>Because the asynchronous operation has already completed, this method is synchronous and
     * returns immediately.
     *
     * @throws CancellationException if the computation was cancelled
     * @throws ExecutionException if the computation threw an exception

    return app


_T = TypeVar("_T")


# Vendored from starlette.routing to avoid importing private symbols
class _AsyncLiftContextManager(AbstractAsyncContextManager[_T]):
    """
    Wraps a synchronous context manager to make it async.

    This is vendored from Starlette to avoid importing private symbols.
    """

    def __init__(self, cm: AbstractContextManager[_T]) -> None:
        self._cm = cm

    *   SignatureRunner is now supported for models with no signatures.

### Bug Fixes and Other Changes

* `tf.data`
    * Add optional `synchronous` argument to `map`, to specify that the `map` should run synchronously, as opposed to be parallelizable when `options.experimental_optimization.map_parallelization=True`. This saves memory compared to setting `num_parallel_calls=1`.