kpt VS kustomize

For the kpt tool and kpt kcl SDK, all mutations performed by KCL in place will be checked into git. In reality, KCL is not limited to mutating Kubernetes manifests on the client side. It is also suitable for use in the admission controller (mainly with certain advantages in language features and performance). We also have plans to support the use of KCL by operators in clusters to edit or validate Kubernetes manifests just like Kubevela and CUE language.

Why are you using a conditional in a declaration? There is https://github.com/GoogleContainerTools/kpt

Here is link number 1 - Previous text "kpt"

kustomize: brew install kustomize

It's a built in feature of kustomize https://github.com/kubernetes-sigs/kustomize/blob/master/api/types/helmchartargs.go

I think the combination of Kustomize and helm works in my experience. For advanced use cases, you can also see KRM functions in Kustomize.

See for instance a related issue: https://github.com/kubernetes-sigs/kustomize/issues/3866

Using a strategic merge is the safest way so you avoid the index fragility.

Rendering out the manifests is something I have been pushing for. Not having to understand how every templating tool works and what actually is being changed is key. Though, it gets complicated when you use helm (or any templating/patching tool) that produces many variants. You also lose any release/deployment time hooks that are provided (helm hooks or recently "patched" kustomize env variables).

https://github.com/kubernetes-sigs/kustomize/blob/master/examples/chart.md (edit: oh I see the other commenter also included this link, oops)

Kustomize is an open-source project that “lets you customize raw, template-free YAML files for multiple purposes, leaving the original YAML untouched and usable as is.” It’s now the most popular tool for customizing Kubernetes manifests reasonably, and it’s even built directly into the Kubernetes CLI since K8s v1.14.