# Lifespan Events { #lifespan-events }

You can define logic (code) that should be executed before the application **starts up**. This means that this code will be executed **once**, **before** the application **starts receiving requests**.

The same way, you can define logic (code) that should be executed when the application is **shutting down**. In this case, this code will be executed **once**, **after** having handled possibly **many requests**.

@app.get("/items/{item_id}")
async def read_items(item_id: str):
    return items[item_id]


def test_read_items():
    # Before the lifespan starts, "items" is still empty
    assert items == {}

    with TestClient(app) as client:
        # Inside the "with TestClient" block, the lifespan starts and items added
        assert items == {"foo": {"name": "Fighters"}, "bar": {"name": "Tenders"}}

        response = client.get("/items/foo")

    /**
     * Called when a processing phase starts.
     *
     * @param phase the phase name (e.g., "intent", "search", "evaluate", "fetch", "answer")
     * @param message a human-readable message describing what's happening
     */
    void onPhaseStart(String phase, String message);

    /**
     * Called when a processing phase starts with additional context data.
     *

  * It currently requires a JVM and will download the Gradle distribution if it is not already present.
* A Gradle version installed on the system
  * This starts the [Gradle Launcher](https://blog.gradle.org/how-gradle-works-1#local-gradle-distribution-in-cli), which itself requires a JVM
* The [Gradle Tooling API client](https://docs.gradle.org/8.8/userguide/third_party_integration.html#embedding)

    }
    assertThat(taskFaker.executeCallCount).isEqualTo(1)

    // Running the red task starts another thread, so the two can run in parallel.
    taskFaker.runNextTask()
    assertThat(log).containsExactly("red:starting@0")
    assertThat(taskFaker.executeCallCount).isEqualTo(2)

    // Next the blue task starts.
    taskFaker.runNextTask()
    assertThat(log).containsExactly(
      "red:starting@0",

`Depends()` erhält einen `scope`-Parameter, der sein kann:

* `"function"`: startet die Abhängigkeit vor der *Pfadoperation-Funktion*, die den Request bearbeitet, beendet die Abhängigkeit nach dem Ende der *Pfadoperation-Funktion*, aber **bevor** die Response an den Client zurückgesendet wird. Die Abhängigkeitsfunktion wird also **um** die *Pfadoperation-**Funktion*** **herum** ausgeführt.

## `fastapi dev` { #fastapi-dev }

Das Ausführen von `fastapi dev` startet den Entwicklermodus.

        session.addUserMessage("User 2");
        session.addAssistantMessage("Assistant 2");
        session.addUserMessage("User 3");
        session.addAssistantMessage("Assistant 3");

        // Trimming to 4 starts at index 2 which is user - no skip needed
        session.trimHistory(4);

        final List<ChatMessage> messages = session.getMessages();
        assertEquals(4, messages.size());

	"github.com/minio/minio-go/v7/pkg/set"
	xhttp "github.com/minio/minio/internal/http"
	"github.com/minio/minio/internal/s3select/jstream"
)

// startWithConds - map which indicates if a given condition supports starts-with policy operator
var startsWithConds = map[string]bool{
	"$acl":                     true,
	"$bucket":                  false,
	"$cache-control":           true,
	"$content-type":            true,

 * is by declaring that step to be the last step of the pipeline. Nevertheless, we refer to the
 * "value" of a successful step or the "result" (value or exception) of any step.
 *
 * <ol>
 *   <li>A pipeline starts at its leaf step (or steps), which is created from either a callable
 *       block or a {@link ListenableFuture}.
 *   <li>Each other step is derived from one or more input steps. At each step, zero or more objects