And as you would be building the container image again and again during development to check that your code changes are working, there's a lot of accumulated time this would save.

Then, near the end of the `Dockerfile`, we copy all the code. As this is what **changes most frequently**, we put it near the end, because almost always, anything after this step will not be able to use the cache.

```Dockerfile

    return delegate().descendingMultiset();
  }

  /**
   * A skeleton implementation of a descending multiset view. Normally, {@link
   * #descendingMultiset()} will not reflect any changes you make to the behavior of methods such as
   * {@link #add(Object)} or {@link #pollFirstEntry}. This skeleton implementation correctly
   * delegates each of its operations to the appropriate methods of this {@code

			// reset log entries
			atomic.StoreInt64(&dt.entries, 0)
			dt.mutex.Unlock()

			dt.adjust(logCopy)
			return
		}
		dt.mutex.Unlock()
	}
}

// adjust changes the value of the dynamic timeout based on the
// previous results
func (dt *dynamicTimeout) adjust(entries [dynamicTimeoutLogSize]time.Duration) {
	failures, max := 0, time.Duration(0)
	for _, dur := range entries[:] {

        }
        sb.append(comp);
        append = true;
      }
      dummy ^= sb.toString().length();
    }
    return dummy;
  }

  /**
   * Starts with an empty delimiter and changes to the desired value at the end of the iteration.
   */
  @Benchmark
  int assignDelimiter(int reps) {
    int dummy = 0;
    for (int i = 0; i < reps; i++) {
      StringBuilder sb = new StringBuilder();

Doing this, our `GzipRequest` will take care of decompressing the data (if necessary) before passing it to our *path operations*.

After that, all of the processing logic is the same.

But because of our changes in `GzipRequest.body`, the request body will be automatically decompressed when it is loaded by **FastAPI** when needed.

## Accessing the request body in an exception handler

/// tip

        }
        sb.append(comp);
        append = true;
      }
      dummy ^= sb.toString().length();
    }
    return dummy;
  }

  /**
   * Starts with an empty delimiter and changes to the desired value at the end of the iteration.
   */
  @Benchmark
  int assignDelimiter(int reps) {
    int dummy = 0;
    for (int i = 0; i < reps; i++) {
      StringBuilder sb = new StringBuilder();

		delete(parsedLog, "scope")
		// check time is there
		if _, f := parsedLog["time"]; !f {
			t.Fatalf("log %v did not have time", i)
		}
		// but remove time since it changes on each test
		delete(parsedLog, "time")
		want := map[string]any{
			"level": cases[i].level,
			"msg":   cases[i].msg,
		}
		if k := cases[i].key; k != nil {
			want["key"] = *k
		}

 <span style="background-color:#C4A000"><font color="#2E3436">╰─────────────────────────────────────────────────────╯</font></span>

<font color="#4E9A06">INFO</font>:     Will watch for changes in these directories: [&apos;/home/user/code/awesomeapp&apos;]
<font color="#4E9A06">INFO</font>:     Uvicorn running on <b>http://127.0.0.1:8000</b> (Press CTRL+C to quit)

}

@test "API compatibility test passes, ensuring no unexpected changes to the TF API" {
    bazel test $TFCI_BAZEL_COMMON_ARGS //tensorflow/tools/api/tests:api_compatibility_test
    echo "You have to re-generate the TF API goldens and have the API changes reviewed."
    echo "Look at the instructions for ':api_compatibility_test -- --update_goldens=True'"
}

      queue[size] = null;
      return null;
    }
    E toTrickle = elementData(size);
    queue[size] = null;
    MoveDesc<E> changes = fillHole(index, toTrickle);
    if (lastElementAt < index) {
      // Last element is moved to before index, swapped with trickled element.
      if (changes == null) {
        // The trickled element is still after index.
        return new MoveDesc<>(actualLastElement, toTrickle);
      } else {