return
    }

    writer.write(name, tagClass, tag) {
      codec.encode(writer, value)
    }
  }

  /**
   * Returns a copy with a context tag. This should be used when the type is ambiguous on its own.
   * For example, the tags in this schema are 0 and 1:
   *
   * ```
   * Point ::= SEQUENCE {
   *   x [0] INTEGER OPTIONAL,
   *   y [1] INTEGER OPTIONAL
   * }
   * ```
   *

 * change the behavior of {@link #equals}, which can lead to unexpected behavior. In this case, you
 * should override {@code equals} as well, either providing your own implementation, or delegating
 * to the provided {@code standardEquals} method.
 *
 * <p>Each of the {@code standard} methods, where appropriate, use {@link Objects#equals} to test

 * side of permissiveness and thus avoid spurious rejection of valid sites. Of course, to actually
 * determine addressability of any host, clients of this class will need to perform their own DNS
 * lookups.
 *
 * <p>During construction, names are normalized in two ways:
 *
 * <ol>
 *   <li>ASCII uppercase characters are converted to lowercase.

 * behavior of {@link #offer} which can lead to unexpected behavior. In this case, you should
 * override {@code offer} as well, either providing your own implementation, or delegating to the
 * provided {@code standardOffer} method.
 *
 * <p><b>{@code default} method warning:</b> This class does <i>not</i> forward calls to {@code

    subjected to URI’s overly-strict validation.
 *  Fix: Don't re-encode `+` as `%20` in encoded URL query strings. OkHttp
    prefers `%20` when doing its own encoding, but will retain `+` when that is
    provided.
 *  Fix: Enforce that callers call `WebSocket.close()` on IO errors. Error
    handling in WebSockets is significantly improved.

   * AggregateFuture} attaches listeners that hold references to one or more inputs. And in the case
   * of {@link CombinedFuture}, the user-supplied callback usually has its own references to inputs.
   */
  /*
   * In certain circumstances, this field might theoretically not be visible to an afterDone() call
   * triggered by cancel(). For details, see the comments on the fields of TimeoutFuture.

 * #add(Object, int)} alone <b>will not</b> change the behavior of {@link #add(Object)}, which can
 * lead to unexpected behavior. In this case, you should override {@code add(Object)} as well,
 * either providing your own implementation, or delegating to the provided {@code standardAdd}
 * method.
 *
 * <p><b>{@code default} method warning:</b> This class does <i>not</i> forward calls to {@code

 *   <li>It is easy for the user to ensure that listeners are never invoked while holding locks.
 * </ul>
 *
 * The last point is subtle. Often the observable object will be managing its own internal state
 * using a lock, however it is dangerous to dispatch listeners while holding a lock because they
 * might run on the {@code directExecutor()} or be otherwise re-entrant (call back into your

    assertNull(getMap().get(k1()));
    assertEquals(k1(), getMap().inverse().get(null));
    assertEquals(1, getMap().size());
  }

  // nb: inverse is run through its own entire suite

  @MapFeature.Require(SUPPORTS_PUT)
  @CollectionSize.Require(ZERO)
  public void testInversePut() {
    getMap().put(k0(), v0());
    getMap().inverse().put(v1(), k1());

 * the behaviour of {@link #offer} which can lead to unexpected behaviour. In this case, you should
 * override {@code offer} as well, either providing your own implementation, or delegating to the
 * provided {@code standardOffer} method.
 *
 * <p><b>{@code default} method warning:</b> This class does <i>not</i> forward calls to {@code