op_name, tape_tensors, input_ids, input_dtypes,
      [gradient_function]() -> GradientFunction* { return gradient_function; },
      [](GradientFunction* ptr) {
        if (ptr) {
          delete ptr;
        }
      });
}
bool Tape::ShouldRecord(
    absl::Span<const AbstractTensorHandle* const> tensors) const {
  std::vector<int64_t> tensor_ids(tensors.size());

    val response =
      getResponse(
        Request.Builder()
          .url(server.url("/"))
          .delete("BODY".toRequestBody(null))
          .build(),
      )
    assertThat(response.code).isEqualTo(200)
    val request = server.takeRequest()
    assertThat(request.method).isEqualTo("DELETE")
    assertThat(request.body.readUtf8()).isEqualTo("BODY")
  }

  @Test

* If you can't understand the question, ask for more **details**.

### Reproduce the problem

For most of the cases and most of the questions there's something related to the person's **original code**.

In many cases they will only copy a fragment of the code, but that's not enough to **reproduce the problem**.

workloads rely on HDFS atomic rename functionality to complete writing data to the datastore. Object storage operations are atomic by nature and they do not require/implement rename API. The default S3A committer emulates renames through copy and delete APIs. This interaction pattern causes significant loss of performance because of the write amplification. _Netflix_, for example, developed two new staging committers - the Directory staging committer and the Partitioned staging committer - to take...

    listener.forbidLock(client.dispatcher)
  }

  @AfterEach
  fun tearDown() {
    if (socksProxy != null) {
      socksProxy!!.shutdown()
    }
    if (cache != null) {
      cache!!.delete()
    }
  }

  @Test
  fun successfulCallEventSequence() {
    server.enqueue(
      MockResponse.Builder()
        .body("abc")
        .build(),
    )
    val call =

 * a specific value. These modification operations are optional, but implementations which support
 * the standard collection operations {@link #add(Object)} or {@link #remove(Object)} are encouraged
 * to implement the related methods as well. Finally, two collection views are provided: {@link
 * #elementSet} contains the distinct elements of the multiset "with duplicates collapsed", and

ction(t,e,n){if(e){o.default(e).removeClass(ue);var i=o.default(e.parentNode).find("> .dropdown-menu .active")[0];i&&o.default(i).removeClass(ue),"tab"===e.getAttribute("role")&&e.setAttribute("aria-selected",!1)}o.default(t).addClass(ue),"tab"===t.getAttribute("role")&&t.setAttribute("aria-selected",!0),d.reflow(t),t.classList.contains(fe)&&t.classList.add(de);var a=t.parentNode;if(a&&"LI"===a.nodeName&&(a=a.parentNode),a&&o.default(a).hasClass("dropdown-menu")){var s=o.default(t).closest(".dro...

          assertNull(it.networkResponse)
        }

      assertEquals(1, cache.hitCount())
      assertEquals(1, cache.networkCount())
      assertEquals(2, cache.requestCount())
    } finally {
      cache.delete()
    }
  }

  private fun OkHttpClient.get(url: String) {
    val request = Request.Builder().url(url).build()
    val response = this.newCall(request).execute()

    response.use {

Sie können `response_model` in jeder möglichen *Pfadoperation* verwenden:

* `@app.get()`
* `@app.post()`
* `@app.put()`
* `@app.delete()`
* usw.

//// tab | Python 3.10+

```Python hl_lines="17  22  24-27"
{!> ../../docs_src/response_model/tutorial001_py310.py!}
```

////

//// tab | Python 3.9+

import java.util.stream.Stream;
import java.util.stream.StreamSupport;
import javax.annotation.CheckForNull;
import org.checkerframework.checker.nullness.qual.Nullable;

/**
 * Static utility methods related to {@code Stream} instances.
 *
 * @since NEXT (but since 21.0 in the JRE flavor)
 */
@GwtCompatible
@ElementTypesAreNonnullByDefault
@SuppressWarnings("Java7ApiChecker")
/*