this.totalCount = elemCount + totalCount(left) + totalCount(right);
    }

    private void recomputeHeight() {
      this.height = 1 + max(height(left), height(right));
    }

    private void recompute() {
      recomputeMultiset();
      recomputeHeight();
    }

    private AvlNode<E> rebalance() {
      switch (balanceFactor()) {
        case -2:

    reallyTestNoMatchThenMatch(matcher.precomputed().negate(), s);
  }

  private void doTestNoMatchThenMatch(CharMatcher matcher, String s) {
    reallyTestNoMatchThenMatch(matcher, s);
    reallyTestMatchThenNoMatch(matcher.negate(), s);
    reallyTestNoMatchThenMatch(matcher.precomputed(), s);
    reallyTestMatchThenNoMatch(matcher.negate().precomputed(), s);

   * - median with compute taking a double-collection and with computeInPlace;
   * - quartiles with index and with indexes taking int-varargs, and with compute taking a
   *   double-collection and with computeInPlace;
   * - scale with index and with indexes taking int-varargs, and with all overloads of compute
   *   taking a double-collection and with computeInPlace;

   * - median with compute taking a double-collection and with computeInPlace;
   * - quartiles with index and with indexes taking int-varargs, and with compute taking a
   *   double-collection and with computeInPlace;
   * - scale with index and with indexes taking int-varargs, and with all overloads of compute
   *   taking a double-collection and with computeInPlace;

@pytest.mark.parametrize(
    "path", ["/a/compute", "/a/compute/", "/b/compute", "/b/compute/"]
)
def test_post(path):
    data = {"a": 2, "b": "foo"}
    response = client.post(path, json=data)
    assert response.status_code == 200, response.text
    assert data == response.json()


@pytest.mark.parametrize(
    "path", ["/a/compute", "/a/compute/", "/b/compute", "/b/compute/"]
)
def test_post_invalid(path):

   * memory, so it is only likely to be worthwhile if the precomputed matcher is queried very often.
   *
   * <p>This method has no effect (returns {@code this}) when called in GWT: it's unclear whether a
   * precomputed matcher is faster, but it certainly would consume more memory (which doesn't seem
   * like a worthwhile tradeoff in a browser).
   */
  public CharMatcher precomputed() {
    return Platform.precomputeCharMatcher(this);
  }

    // b/80241237

    Cache<String, String> c = CacheBuilder.newBuilder().maximumSize(1).build();

    assertThat(c.asMap().compute("hash-1", (k, v) -> "a")).isEqualTo("a");
    assertThat(c.asMap().compute("hash-1", (k, v) -> "b")).isEqualTo("b");
    assertThat(c.asMap().compute("hash-1", (k, v) -> "c")).isEqualTo("c");
    assertThat(c.size()).isEqualTo(1);
    assertThat(c.asMap().computeIfAbsent("hash-2", k -> "")).isEqualTo("");

   * memory, so it is only likely to be worthwhile if the precomputed matcher is queried very often.
   *
   * <p>This method has no effect (returns {@code this}) when called in GWT: it's unclear whether a
   * precomputed matcher is faster, but it certainly would consume more memory (which doesn't seem
   * like a worthwhile tradeoff in a browser).
   */
  public CharMatcher precomputed() {
    return Platform.precomputeCharMatcher(this);
  }

 *
 * <h3>Examples</h3>
 *
 * <p>To compute the median:
 *
 * {@snippet :
 * double myMedian = median().compute(myDataset);
 * }
 *
 * where {@link #median()} has been statically imported.
 *
 * <p>To compute the 99th percentile:
 *
 * {@snippet :
 * double myPercentile99 = percentiles().index(99).compute(myDataset);
 * }
 *

            print(
                'ERROR: TensorFlow only supports small CUDA compute'
                ' capabilities of sm_30 and higher. Please re-specify the list'
                ' of compute capabilities excluding version %s.' % ver)
            all_valid = False
          if ver < 35:
            print('WARNING: XLA does not support CUDA compute capabilities '