before the response starts. With duplex, request and response bodies are transmitted
    simultaneously. This can be used to implement interactive conversations within a single HTTP
    call.

    Create duplex calls by overriding the new `RequestBody.isDuplex()` method to return true.
    This simple option dramatically changes the behavior of the request body and of the entire
    call.

      }

    /**
     * Sets the socket factory used to create connections. OkHttp only uses the parameterless
     * [SocketFactory.createSocket] method to create unconnected sockets. Overriding this method,
     * e. g., allows the socket to be bound to a specific local address.
     *
     * If unset, the [system-wide default][SocketFactory.getDefault] socket factory will be used.
     */

   */
  final class WriteThroughEntry extends SimpleEntry<K, V> {
    WriteThroughEntry(K key, V value) {
      super(key, value);
    }

    /*
     * We inherit equals() and hashCode() instead of overriding them to use keyEquivalence and
     * valueEquivalence.
     */

    @Override
    public V setValue(V newValue) {
      put(getKey(), newValue);

         * an array and when a value is potentionally overriden it is added to the array. The single
         * arg option value is retrieved and instead of returning values[values.length-1] it returns
         * values[0] which means that the original value instead of the overridden one is returned
         * (first wins). With properties values are truely overriden since at the end a map is used
         * to merge which means last wins.

   * broadly guaranteed, but it is observed to be true for OpenJDK 1.7. It also can't be guaranteed
   * that the method isn't inadvertently <i>oversizing</i> the returned set.
   *
   * @param expectedSize the number of elements you expect to add to the returned set
   * @return a new, empty {@code LinkedHashSet} with enough capacity to hold {@code expectedSize}

	fi.ModTime = modTime     // set modTime for the new versionID
	if !dstOpts.MTime.IsZero() {
		modTime = dstOpts.MTime
		fi.ModTime = dstOpts.MTime
	}
	// check inline before overwriting metadata.
	inlineData := fi.InlineData()

	fi.Metadata = srcInfo.UserDefined
	srcInfo.UserDefined["etag"] = srcInfo.ETag

	freeVersionID := fi.TierFreeVersionID()
	freeVersionMarker := fi.TierFreeVersion()

		// call will report top level prefixes in deleted state, whereas spark/hadoop interpret this as non-empty
		// and throw a 404 exception. This is especially a problem for spark jobs overwriting the same partition
		// repeatedly. This workaround recursively lists the top 3 entries including delete markers to reflect the
		// correct state of the directory in the list results.

	if err = checkPathLength(dstFilePath); err != nil {
		return err
	}
	if srcIsDir {
		// If source is a directory, we expect the destination to be non-existent but we
		// we still need to allow overwriting an empty directory since it represents
		// an object empty directory.
		dirInfo, err := Lstat(dstFilePath)
		if isSysErrIO(err) {
			return errFaultyDisk
		}
		if err != nil {