Automatic merge from submit-queue (batch tested with PRs 63251, 59166, 63250, 63180, 63169). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
divide statically known typer from dynamically derived restmapper
The CLI factory conflated a statically known typer and a dynamically derived restmapper. This produced confusion and unnecessary indirection. This pull separates dynamically known from statically known.
A future pull will continue the separation of in the builder to indicate the points of variability for clients, which should logically be limited to a negotiated serializer for decoding and an object typer.
@kubernetes/sig-cli-maintainers
/assign @soltysh
/assign @juanvallejo
```release-note
NONE
```
Automatic merge from submit-queue (batch tested with PRs 62982, 63075, 63067, 62877, 63141). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
make describers more generic from the CLI
I've made this change very small so the intent and explanation make sense to people.
Clients are not generic. Client**Configs** are generic. We faced this distinction in the apiserver and it took us a little to hurdle it. When you try to provide a generic example or function, you need to provide Client**Config**, not a kube clientset. The reason is that the code you're calling may have generated their own clientset, may want to use a dynamic one, or may want to a simple restclient. As we seek to make `kubectl` primitives more generally applicable, this is an example we'll want to follow. I suspect we'll be making more changes along these veins as we tease out the generic pieces of `kubectl ` to make a friendly CLI library.
@kubernetes/sig-cli-maintainers
/hold
Holding for a few days to make sure that people have time to read and digest.
```release-note
NONE
```
Automatic merge from submit-queue (batch tested with PRs 60302, 57921, 59042, 60126, 59561). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
Remove pkg/client/unversioned
**What this PR does / why we need it**:
This is removing unused package, and moves the used bits into appropriate placeholders.
**Which issue(s) this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close the issue(s) when PR gets merged)*:
Fixes#25442
**Special notes for your reviewer**:
**Release note**:
```release-note
NONE
```
/assign @deads2k
/assign @sttts
Automatic merge from submit-queue (batch tested with PRs 58955, 58968, 58971, 58963, 58298). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
removes the remainder from ScalerFor method
**What this PR does / why we need it**:
this PR removes existing scalers from `ScalerFor` method
**Release note**:
```release-note
NONE
```
all remaining scalers were replaced by GenericScaler exept JobScaler.
It is not clear whether JobScaler could use generic scaler or not.
For more details see the pull request.
note that we don't change the behaviour of kubectl.
For example it won't scale new resources. That's the end goal.
The first step is to retrofit existing code to use the generic scaler.
Callers must take a dependency on one or the other set of conversions
and default client behavior. Future changes may add a Versioned() type,
but this is an accurate reflection of current code state.
The unified RESTMapper and Typer follow the new rules, but on error will
fallback to the legacy path (while still supporting Unstructured
objects). This allows callers to handle the appropriate distinction
themselves if necessary.
Add a LocalParam() method to the resource.Builder that DRYs up a large
chunk of complicated code in set commands.
Also remove error messages that depended on ObjectKinds() - future
changes will potentially remove this interface and the replacements here
are equivalent.
Delays the error until the first call and then preserves it for others.
More closely matches the intent of the Object() calls. Loaders are now
lazy and don't need to return errors directly.
Sets the stage for collapsing unstructured and structured builders
together.
Automatic merge from submit-queue (batch tested with PRs 53606, 49361). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
add ApproximatePodTemplateForObject factory method
Makes it possible to get at a pod spec template even if an object is scaled to zero, for use with commands that care about pod templates.
**Release note**:
```release-note
NONE
```
Related downstream patch and use-case: https://github.com/openshift/origin/pull/16379
cc @smarterclayton
Automatic merge from submit-queue (batch tested with PRs 51771, 52971). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
pass labelSelector to server side opaquely
**What this PR does / why we need it**:
From @smarterclayton
> The server is responsible for handling label selection for the most part. There is some level of client side processing possible, but for the most part `label selector` should be able to be passed opaquely.
xref #50140
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes #
**Special notes for your reviewer**:
/assign @smarterclayton @liggitt
**Release note**:
```release-note
None
```
