* <p><b>Performance note:</b> the returned array is backed by the same array as the builder, so
     * no data is copied as part of this step, but this may occupy more memory than strictly
     * necessary. To copy the data to a right-sized backing array, use {@code .build().trimmed()}.
     */
    public ImmutableIntArray build() {
      return count == 0 ? EMPTY : new ImmutableIntArray(array, 0, count);
    }
  }

  // Instance stuff here

    <exportedPackage>jakarta.inject.*</exportedPackage>
    <exportedPackage>jakarta.annotation.*</exportedPackage>

    <!--
      | We may potentially want to export these, but right now I'm not sure that anything Guice specific needs
      | to be made available to plugin authors. If we find people are getting fancy and want to take advantage of

            // mvn remote-resources:process
            //
            // From the metadata stored on the server which has been created as part of a standard
            // Maven plugin deployment we will find the right PluginDescriptor from the remote
            // repository.

            plugin = findPluginForPrefix(prefix, session);
        }

        int executionIdx = goal.indexOf("@");
        if (executionIdx > 0) {

     * parameters, and as it turns out Firefox and Chrome actually do rather different things, and
     * both say in their comments that they're not really sure what the right approach is. We go
     * with Chrome's behavior (which also experimentally seems to match what IE does), but if you
     * actually want to have a good chance of things working, please avoid double-quotes, newlines,

*A noticeable trait of this expansion is that it chooses unique hosts such the setup provides maximum protection and availability.*

- Choosing an erasure set for the object is decided during `PutObject()`, object names are used to find the right erasure set using the following pseudo code.

```go
// hashes the key returning an integer.
func sipHashMod(key string, cardinality int, id [16]byte) int {
        if cardinality <= 0 {
                return -1

		// array second dimension represents list of disks used per set.
		Sets [][]string `json:"sets"`
		// Distribution algorithm represents the hashing algorithm
		// to pick the right set index for an object.
		DistributionAlgo string `json:"distributionAlgo"`
	} `json:"xl"`
}

// formatErasureV3 struct is same as formatErasureV2 struct except that formatErasureV3.Erasure.Version is "3" indicating

    return list;
  }

  @VisibleForTesting
  static int computeArrayListCapacity(int arraySize) {
    checkNonnegative(arraySize, "arraySize");

    // TODO(kevinb): Figure out the right behavior, and document it
    return Ints.saturatedCast(5L + arraySize + (arraySize / 10));
  }

  /**
   * Creates an {@code ArrayList} instance backed by an array with the specified initial size;

   * Views the array as an immutable list. The array must have only {@code E} elements.
   *
   * <p>The array must be internally created.
   */
  @SuppressWarnings("unchecked") // caller is reponsible for getting this right
  static <E> ImmutableList<E> asImmutableList(Object[] elements) {
    return unsafeDelegateList((List) Arrays.asList(elements));
  }

  public static <E extends Comparable<? super E>> ImmutableList<E> sortedCopyOf(

	if err != nil {
		return time.Time{}, ErrHeader
	}
	if len(sn) == 0 {
		return time.Unix(secs, 0), nil // No sub-second values
	}

	// Parse the nanoseconds.
	// Initialize an array with '0's to handle right padding automatically.
	nanoDigits := [maxNanoSecondDigits]byte{'0', '0', '0', '0', '0', '0', '0', '0', '0'}
	for i := range len(sn) {
		switch c := sn[i]; {
		case c < '0' || c > '9':
			return time.Time{}, ErrHeader

#### Blocking APIs

Blocking APIs are convenient because you get top-to-bottom procedural code without indirection. Network calls work like regular method calls: ask for data and it is returned. If the request fails, you get a stacktrace right where the call was made.

Blocking APIs may be inefficient because you hold a thread idle while waiting on the network. Threads are expensive because they have both a memory overhead and a context-switching overhead.