the system. In Kubernetes, a controller is a control loop that watches the shared
state of the cluster through the apiserver and makes changes attempting to move the
current state towards the desired state. Examples of controllers that ship with
Kubernetes today are the replication controller, endpoints controller, namespace
controller, and serviceaccounts controller.`,
		PersistentPreRunE: func(*cobra.Command, []string) error {
			// silence client-go warnings.

	if err != nil && routeName != model.RDSHttpProxy && !strings.HasPrefix(routeName, model.UnixAddressPrefix) {
		// TODO: This is potentially one place where envoyFilter ADD operation can be helpful if the
		// user wants to ship a custom RDS. But at this point, the match semantics are murky. We have no
		// object to match upon. This needs more thought. For now, we will continue to return nil for
		// unknown routes
		return nil, false
	}

include::sample[dir="snippets/dependencyManagement/declaringDependencies-artifactOnly/groovy",files="build.gradle[tags=artifact-only-dependency-declaration]"]
====

	"shell":                                "\U0001f41a",
	"shield":                               "\U0001f6e1\ufe0f",
	"shinto_shrine":                        "\u26e9\ufe0f",
	"ship":                                 "\U0001f6a2",
	"shirt":                                "\U0001f455",
	"shit":                                 "\U0001f4a9",
	"shoe":                                 "\U0001f45e",

	return m
}

// clangos lists the operating systems where we prefer clang to gcc.
var clangos = []string{
	"darwin", "ios", // macOS 10.9 and later require clang
	"freebsd", // FreeBSD 10 and later do not ship gcc
	"openbsd", // OpenBSD ships with GCC 4.2, which is now quite old.
}

// compilerEnvLookup returns the compiler settings for goos/goarch in map m.
// kind is "CC" or "CXX".

It means that you will be tempted to switch it off.

So why should you bother?

Dependency verification is about **trust** in what you get and what you ship.

Without dependency verification it's easy for an attacker to compromise your supply chain.
There are many real world examples of tools compromised by adding a malicious dependency.

==== Backward compatibility

Our approach is to only do backwards-breaking Kotlin upgrades on a major Gradle release. We will always clearly document which Kotlin version we ship and announce upgrade plans before a major release.

* Make kubectl edit not convert GV on edits ([#23437](https://github.com/kubernetes/kubernetes/pull/23437), [@DirectXMan12](https://github.com/DirectXMan12))
* don't ship kube-registry-proxy and pause images in tars. ([#23605](https://github.com/kubernetes/kubernetes/pull/23605), [@mikedanese](https://github.com/mikedanese))

=== Potential breaking changes

[[kotlin_1_7_10]]
==== Upgrade to Kotlin 1.7.10

The embedded Kotlin has been updated to https://github.com/JetBrains/kotlin/releases/tag/v1.7.10[Kotlin 1.7.10].

Gradle doesn't ship with the `kotlin-gradle-plugin` but the upgrade to 1.7.10 can bring the new version.
For example when you use the `kotlin-dsl` plugin.

���[�_Qw����B4^l����40��d�hX\�1�R~��F��`���ͷ>xP�L_Q'���I�I26��a�
 ��F�8H�O�C���z7��(�oi-� ��Xؚ�oFV���~\�>&���?�WӅ
��#r��F�� |������?��rw��W�������Iw�{!0
2w�&(kV`�Zu9�墈8��~�@��3+6_���6�N�=�J}Lr��4�HU�����_�#+/�ź)Z-b>SS�)�Ζ^����u�M6'��hip����#�Ѐ�*0kz�P�G*��� �9����^�/^�����73i�$� HިW����w�h��6t�:K�VȽ�+
� e�c���������x�ߗ���D��� ���:���_���‹�z����,UMF��H���x���6%Vnj��V����Bӓp����ذ�܇a�3������xц#��ϝ ^�w$���3*A�] .�tL�ѝ~�� ڻU��'�Q�pV t7��--S[������/^���*�1�GK����*"cc��m�Ͻ�,��1����{x...