* requested permits, by an invocation acquire(permits), are served from:
   *
   * - stored permits (if available)
   *
   * - fresh permits (for any remaining permits)
   *
   * How this works is best explained with an example:
   *
   * For a RateLimiter that produces 1 token per second, every second that goes by with the
   * RateLimiter being unused, we increase storedPermits by 1. Say we leave the RateLimiter unused

You can also use the *path operation decorator* parameters `response_model_include` and `response_model_exclude`.

They take a `set` of `str` with the name of the attributes to include (omitting the rest) or to exclude (including the rest).

This can be used as a quick shortcut if you have only one Pydantic model and want to remove some data from the output.

/// tip

   * structure (to make it possible to throw ConcurrentModificationException in the iterator). Note
   * that we choose not to make this volatile, so we do less of a "best effort" to track such
   * errors, for better performance.
   *
   * <p>For a new instance, where the arrays above have not yet been allocated, the value of {@code

    requests with the same POST data get the same cache entry.

 *  New: `HttpLoggingInterceptor.redactQueryParams()` configures the query parameters to redact
    in logs. For best security, don't put sensitive information in query parameters.

 *  New: `ConnectionPool.setPolicy()` configures a minimum connection pool size for a target
    address. Use this to proactively open HTTP connections.

    For example, to run all tests under tensorflow/python, do:

    ```bash
    bazel test ${flags} //tensorflow/python/...
    ```

    For a single component e.g. softmax op:

    ```bash
    bazel test ${flags} tensorflow/python/kernel_tests/nn_ops:softmax_op_test
    ```

    For a single/parameterized test e.g. `test_capture_variables` in
    `tensorflow/python/saved_model/load_test.py`:

   * building caches whose key or value types are incompatible with the types accepted by the
   * weigher already provided; the {@code CacheBuilder} type cannot do this. For best results,
   * simply use the standard method-chaining idiom, as illustrated in the documentation at top,
   * configuring a {@code CacheBuilder} and building your {@link Cache} all in a single statement.
   *

   * previously at {@code src} has already been removed from the data structures.
   */
  private void moveEntryToIndex(int src, int dest) {
    if (src == dest) {
      return;
    }
    int predecessor = prevInInsertionOrder[src];
    int successor = nextInInsertionOrder[src];
    setSucceeds(predecessor, dest);
    setSucceeds(dest, successor);

   * example, thread A may be executing [responseBodyStart] while thread B executes [canceled].
   * Implementations must support such concurrent calls.
   *
   * Note that cancellation is best-effort and that a call may proceed normally after it has been
   * canceled. For example, happy-path events like [requestHeadersStart] and [requestHeadersEnd] may

        for (NetworkInterfaceInfo info : response.getInterfaces()) {
            if (info.isUsableForChannel()) {
                remoteInterfaces.add(info);
            }
        }

        // Sort by score (best interfaces first)
        remoteInterfaces.sort((a, b) -> Integer.compare(b.getScore(), a.getScore()));

        log.debug("Discovered {} remote network interfaces", remoteInterfaces.size());
    }

   * recognize certain cases where distinct {@code Equivalence} instances would in fact behave
   * identically. However, as code migrates to {@code java.util.function}, that behavior will
   * disappear. It is best not to depend on it.
   */
  @Override
  public boolean equals(@Nullable Object object) {
    return super.equals(object);
  }

  /**
   * Returns a hash code for {@code t}.
   *