3. Ensure you have added or ran the appropriate tests for your PR: https://git.k8s.io/community/contributors/devel/sig-testing/testing.md
4. If you want *faster* PR reviews, read how: https://git.k8s.io/community/contributors/guide/pull-requests.md#best-practices-for-faster-reviews
5. If the PR is unfinished, see how to mark it: https://git.k8s.io/community/contributors/guide/pull-requests.md#marking-unfinished-pull-requests
-->

   - Reject untested PRs unless testing is infeasible, then assist with test setup.

## Tips for Success

- **Small PRs**: Easier to review, faster to merge. Split large changes logically.
- **Clear Commits**: Use `git rebase -i` to refine history before submitting.
- **Engage Early**: Discuss complex changes in issues or Slack (https://slack.min.io) before coding.

# For wheel verification, and uploading
auditwheel ~= 6.1.0
twine ~= 6.1.0
id
urllib3
requests

# For XLA
pyyaml

# For JAX
build ~= 1.2.2
# uv is faster than pip for installing Python packages.
uv ~= 0.5.30

# For running wheel verification script

* **FastAPI**:
    * The same way that Starlette uses Uvicorn and cannot be faster than it, **FastAPI** uses Starlette, so it cannot be faster than it.

   *
   * <p>This differs from {@code Long.bitCount(x) == 1}, because {@code
   * Long.bitCount(Long.MIN_VALUE) == 1}, but {@link Long#MIN_VALUE} is not a power of two.
   */
  // Whenever both tests are cheap and functional, it's faster to use &, | instead of &&, ||
  @SuppressWarnings("ShortCircuitBoolean")
  public static boolean isPowerOfTwo(long x) {
    return x > 0 & (x & (x - 1)) == 0;
  }

  /**

/**
 * @author Jesse Wilson
 */
final class Platform {
  static CharMatcher precomputeCharMatcher(CharMatcher matcher) {
    // CharMatcher.precomputed() produces CharMatchers that are maybe a little
    // faster (and that's debatable), but definitely more memory-hungry. We're
    // choosing to turn .precomputed() into a no-op in GWT, because it doesn't
    // seem to be a worthwhile tradeoff in a browser.
    return matcher;
  }

    return delegate.toArray(other);
  }

  @Override
  public String toString() {
    return delegate.toString();
  }

  // TODO(cpovirk): equals(), as well, in case it's any faster than Sets.equalsImpl?

  @Override
  public int hashCode() {
    return delegate.hashCode();
  }

  abstract ImmutableCollection<E> delegateCollection();

  @Override
  public boolean contains(@Nullable Object target) {
    // The collection's contains() is at least as fast as ImmutableList's
    // and is often faster.
    return delegateCollection().contains(target);
  }

  @Override
  public int size() {
    return delegateCollection().size();
  }

  @Override
  public boolean isEmpty() {

## Code Examples

[Using minio-go library](https://github.com/minio/minio/blob/master/docs/extensions/s3zip/examples/minio-go/main.go)
[Using AWS JS SDK v2](https://github.com/minio/minio/blob/master/docs/extensions/s3zip/examples/aws-js/main.js)
[Using boto3](https://github.com/minio/minio/blob/master/docs/extensions/s3zip/examples/boto3/main.py)

## Requirements and limits

#   request - Cache only for current request + its child requests
#   none    - Disable caching entirely
#
# REFERENCE OPTIONS:
#   hard - Strong reference (stays in memory, faster access)
#   soft - Weak reference (can be garbage collected under memory pressure)
#
# CONFIGURATION BREAKDOWN:
#   SourceCacheKey                   - File and RAW model requests (session/hard for consistent file access)