: ImmutableList.<E>construct(elements.clone());
  }

  /**
   * Returns an immutable list containing the given elements, sorted according to their natural
   * order. The sorting algorithm used is stable, so elements that compare as equal will stay in the
   * order in which they appear in the input.
   *
   * <p>If your data has no duplicates, or you wish to deduplicate elements, use {@code

func (e ErasureInfo) GetChecksumInfo(partNumber int) (ckSum ChecksumInfo) {
	for _, sum := range e.Checksums {
		if sum.PartNumber == partNumber {
			// Return the checksum
			return sum
		}
	}
	return ChecksumInfo{Algorithm: DefaultBitrotAlgorithm}
}

// ShardFileSize - returns final erasure size from original size.
func (e ErasureInfo) ShardFileSize(totalLength int64) int64 {
	if totalLength == 0 {
		return 0
	}

	values.Set(storageRESTFilePath, path)
	values.Set(storageRESTOffset, strconv.Itoa(int(offset)))
	values.Set(storageRESTLength, strconv.Itoa(len(buf)))
	if verifier != nil {
		values.Set(storageRESTBitrotAlgo, verifier.algorithm.String())
		values.Set(storageRESTBitrotHash, hex.EncodeToString(verifier.sum))
	} else {
		values.Set(storageRESTBitrotAlgo, "")
		values.Set(storageRESTBitrotHash, "")
	}

	// undefined.
	if !st.Mode().IsRegular() {
		return 0, errIsNotRegular
	}

	if verifier == nil {
		n, err = file.ReadAt(buffer, offset)
		return int64(n), err
	}

	h := verifier.algorithm.New()
	if _, err = io.Copy(h, io.LimitReader(file, offset)); err != nil {
		return 0, err
	}

	if n, err = io.ReadFull(file, buffer); err != nil {
		return int64(n), err
	}

			partPath := pathJoin(object, metaArr[index].DataDir, fmt.Sprintf("part.%d", partNumber))
			readers[index] = newBitrotReader(disk, metaArr[index].Data, bucket, partPath, tillOffset,
				checksumInfo.Algorithm, checksumInfo.Hash, erasure.ShardSize())

			// Prefer local disks
			prefer[index] = disk.Hostname() == ""
		}

		written, err := erasure.Decode(ctx, writer, readers, partOffset, partLength, partSize, prefer)

        behavior, set `inject_prefetch=True` in
        `tf.data.experimental.OptimizationOptions`.
    *   Added a new value to `tf.data.Options.autotune.autotune_algorithm`:
        `STAGE_BASED`. If the autotune algorithm is set to `STAGE_BASED`, then
        it runs a new algorithm that can get the same performance with lower
        CPU/memory usage.
    *   Added