assertThat(stopwatch.elapsed(MILLISECONDS)).isIn(Range.closed(NOT_ENOUGH_MS, DELAY_MS * 2));
    // Is it still computing away anyway?
    assertThat(target.finished).isFalse();
    MILLISECONDS.sleep(ENOUGH_MS);
    assertThat(target.finished).isFalse();
  }

  public void testNewProxy_badMethodWithEnoughTime() throws Exception {
    SampleImpl target = new SampleImpl(DELAY_MS);

                if (scheduledJob.isLoggingEnabled() && logger.isInfoEnabled()) {
                    logger.info("Finished job: id={}", id);
                }
            } else {
                if (scheduledJob.isLoggingEnabled() && logger.isInfoEnabled()) {
                    logger.info("Finished job: id={}, returnValue:\n{}", id, ret);
                }
                jobLog.setScriptResult(ret.toString());
            }

    assertThat(stopwatch.elapsed(MILLISECONDS)).isIn(Range.closed(NOT_ENOUGH_MS, DELAY_MS * 2));
    // Is it still computing away anyway?
    assertThat(target.finished).isFalse();
    MILLISECONDS.sleep(ENOUGH_MS);
    assertThat(target.finished).isFalse();
  }

  public void testNewProxy_badMethodWithEnoughTime() throws Exception {
    SampleImpl target = new SampleImpl(DELAY_MS);

In version 0.121.0, FastAPI added support for `Depends(scope="function")` for dependencies with `yield`.

Using `Depends(scope="function")`, the exit code after `yield` is executed right after the *path operation function* is finished, before the response is sent back to the client.

And when using `Depends(scope="request")` (the default), the exit code after `yield` is executed after the response is sent.

So, during that time, the computer can go and do some other work, while "slow-file" 📝 finishes.

Then the computer / program 🤖 will come back every time it has a chance because it's waiting again, or whenever it 🤖 finished all the work it had at that point. And it 🤖 will see if any of the tasks it was waiting for have already finished, doing whatever it had to do.

This policy protects our limited review budget, allowing us to invest our attention wisely.

A pull request is not a finished product—it is the start of a conversation, and an implied contract: we promise to provide thoughtful feedback, and we expect the contributor to engage with us in return.

## What we expect from contributors

 * to be closed asynchronously <b>after</b> the returned {@code Future} is done: the future
 * completes before closing starts, rather than once it has finished.
 *
 * {@snippet :
 * FluentFuture<UserName> userName =
 *     ClosingFuture.submit(
 *             closer -> closer.eventuallyClose(database.newTransaction(), closingExecutor),
 *             executor)

    assertFalse(singleCallListener.wasCalled());
    assertTrue(listenableFuture.isDone()); // We call AbstractFuture#set above.

    // #addListener() will run the listener immediately because the Future is
    // already finished (we explicitly set the result of it above).
    listenableFuture.addListener(singleCallListener, directExecutor());
    assertThat(listenableFuture.get()).isEqualTo(DATA1);

            }
            if (logNotificationTarget != null) {
                logNotificationTarget.flush();
            }
            destroyContainer();
        }

        logger.info("Finished SuggestCreator.");
        System.exit(exitCode);
    }

    private static void destroyContainer() {
        TimeoutManager.getInstance().stop();
        synchronized (SingletonLaContainerFactory.class) {

The first part of the function, before the `yield`, will be executed **before** the application starts.

And the part after the `yield` will be executed **after** the application has finished.

### Async Context Manager { #async-context-manager }

If you check, the function is decorated with an `@asynccontextmanager`.

That converts the function into something called an "**async context manager**".