kube-controller manager now use SO_REUSEPORT socket option when listening on address defined by --bind-address and --secure-port flags, when running on Unix systems (Windows is NOT supported). This allows to run multiple instances of those processes on a single host with the same configuration, which allows to update/restart them in a graceful way, without causing downtime. ([#88893](https://github.com/kubernetes/kubernetes/pull/88893), [@invidian](https://github.com/invidian)) [SIG API Machinery, Scheduling...

	writeResponse(w, err.HTTPStatusCode, encodedErrorResponse, mimeJSON)
}

// writeCustomErrorResponseJSON - similar to writeErrorResponseJSON,
// but accepts the error message directly (this allows messages to be
// dynamically generated.)
func writeCustomErrorResponseJSON(ctx context.Context, w http.ResponseWriter, err APIError,
	errBody string, reqURL *url.URL,
) {
	reqInfo := logger.GetReqInfo(ctx)

- Graduated relaxed DNS search string validation to GA. For the Pod API, `.spec.dnsConfig.searches`
  now allows an underscore (`_`) where a dash (`-`) would be allowed, and it allows search strings be a single dot `.`. ([#132036](https://github.com/kubernetes/kubernetes/pull/132036), [@adrianmoisey](https://github.com/adrianmoisey)) [SIG Network and Testing]

	// When versionID is empty, we read directly from the `null` versionID for healing.
	if versionID == "" {
		versionID = nullVersionID
	}

	// Perform quick read without lock.
	// This allows to quickly check if all is ok or all are missing.
	_, errs := readAllFileInfo(healCtx, storageDisks, "", bucket, object, versionID, false, false)
	if isAllNotFound(errs) {
		err := errFileNotFound
		if versionID != "" {

    visibility = ["//visibility:public"],
)

# By default, XLA GPU is compiled into tensorflow when building with
# --config=cuda even when `with_xla_support` is false. The config setting
# here allows us to override the behavior if needed.
config_setting(
    name = "no_xla_deps_in_cuda",
    define_values = {"no_xla_deps_in_cuda": "true"},
    visibility = ["//visibility:public"],
)

github.com/juju/ratelimit
https://github.com/juju/ratelimit
----------------------------------------------------------------
All files in this repository are licensed as follows. If you contribute
to this repository, it is assumed that you license your contribution
under the same license unless you state otherwise.

 * equals</i>. That is, {@code a.compareTo(b)} or {@code comparator.compare(a, b)} should equal zero
 * <i>if and only if</i> {@code a.equals(b)}. If this advice is not followed, the resulting
 * collection will not correctly obey its specification.
 *
 * <p>See the Guava User Guide article on <a href=
 * "https://github.com/google/guava/wiki/ImmutableCollectionsExplained">immutable collections</a>.

* ✨ Add support for `dependencies` in WebSocket routes. PR [#4534](https://github.com/tiangolo/fastapi/pull/4534) by [@paulo-raca](https://github.com/paulo-raca).
* ✨ Add exception handler for `WebSocketRequestValidationError` (which also allows to override it). PR [#6030](https://github.com/tiangolo/fastapi/pull/6030) by [@kristjanvalur](https://github.com/kristjanvalur).

### Refactors

      val slashCount = input.slashCount(pos, limit)
      if (slashCount >= 2 || base == null || base.scheme != this.scheme) {
        // Read an authority if either:
        //  * The input starts with 2 or more slashes. These follow the scheme if it exists.
        //  * The input scheme exists and is different from the base URL's scheme.
        //
        // The structure of an authority is:
        //   username:password@host:port
        //

proprietary. To prevent this, we have made it clear that any patent must
be licensed for everyone's free use or not licensed at all.

The precise terms and conditions for copying, distribution and
modification follow.

TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION

0. This License applies to any program or other work which contains a
notice placed by the copyright holder saying it may be distributed under