}
            }
            return buf.toString();
        }
    }

    public ArtifactVersion matchVersion(List<ArtifactVersion> versions) {
        // TODO could be more efficient by sorting the list and then moving along the restrictions in order?

        ArtifactVersion matched = null;
        for (ArtifactVersion version : versions) {
            if (containsVersion(version)) {

    reverse(array, fromIndex, toIndex);
  }

  /**
   * Reverses the elements of {@code array}. This is equivalent to {@code
   * Collections.reverse(Longs.asList(array))}, but is likely to be more efficient.
   *
   * @since 23.1
   */
  public static void reverse(long[] array) {
    checkNotNull(array);
    reverse(array, 0, array.length);
  }

  /**

    reverse(array, fromIndex, toIndex);
  }

  /**
   * Reverses the elements of {@code array}. This is equivalent to {@code
   * Collections.reverse(Longs.asList(array))}, but is likely to be more efficient.
   *
   * @since 23.1
   */
  public static void reverse(long[] array) {
    checkNotNull(array);
    reverse(array, 0, array.length);
  }

  /**

    return file(new File(first), more);
  }

  /** Returns a {@code File} object for the given path parts. */
  private static File file(File first, String... more) {
    // not very efficient, but should definitely be correct
    File file = first;
    for (String name : more) {
      file = new File(file, name);
    }
    return file;
  }

 *     .url("http://publicobject.com/helloworld.txt")
 *     .build();
 * ```
 *
 * If it is only necessary to force a cached response to be validated by the server, use the more
 * efficient `max-age=0` directive instead:
 *
 * ```java
 * Request request = new Request.Builder()
 *     .cacheControl(new CacheControl.Builder()
 *         .maxAge(0, TimeUnit.SECONDS)
 *         .build())

#### Endpoints

Endpoints have an optimized code path, as they are by far the most frequently updated resource - in a steady cluster, this will often be the *only* change, caused by scale up/down.

   * Reader#skip(long) skip} to the end of the stream, and return the total number of chars that
   * were skipped.
   *
   * <p>Note that for sources that implement {@link #lengthIfKnown} to provide a more efficient
   * implementation, it is <i>possible</i> that this method will return a different number of chars
   * than would be returned by reading all of the chars.
   *

  /**
   * Creates a new {@code ConcurrentHashMultiset} containing the specified elements, using the
   * default initial capacity, load factor, and concurrency settings.
   *
   * <p>This implementation is highly efficient when {@code elements} is itself a {@link Multiset}.
   *
   * @param elements the elements that the multiset should contain
   */
  public static <E> ConcurrentHashMultiset<E> create(Iterable<? extends E> elements) {

    return file(new File(first), more);
  }

  /** Returns a {@code File} object for the given path parts. */
  private static File file(File first, String... more) {
    // not very efficient, but should definitely be correct
    File file = first;
    for (String name : more) {
      file = new File(file, name);
    }
    return file;
  }

    return true;
  }

  /**
   * Each instance of MinMaxPriorityQueue encapsulates two instances of Heap: a min-heap and a
   * max-heap. Conceptually, these might each have their own array for storage, but for efficiency's
   * sake they are stored interleaved on alternate heap levels in the same array (MMPQ.queue).
   */
  @WeakOuter
  class Heap {
    final Ordering<E> ordering;