Kubernetes Resource Model (KRM) KCL Specification

The KRM KCL function SDK contains a KRM KCL spec and an interpreter to run KCL codes to generate, mutate or validate Kubernetes resources.

The KCL programming language can be used to:

• Add labels or annotations based on a condition.
• Inject a sidecar container in all KRM resources that contain a PodTemplate.
• Validate all KRM resources using KCL schema.
• Use an abstract model to generate KRM resources.

Test the KRM Function

Unit Tests

You need to put your KCL script source in the functionConfig of kind KCLRun and then the function will run the KCL script that you provide.

This function can be used both declaratively and imperatively.

make test

Integration Tests

You need to put your KCL source code or url in the functionConfig of kind KCLRun and then the function will run the KCL code that you provide.

# Verify that the annotation is added to the `Deployment` resource and the other resource `Service` 
# does not have this annotation.
export TEST_FILE=./pkg/options/testdata/yaml_stream/kcl-run-code.yaml
diff \
  <(cat ${TEST_FILE}) \
  <(cat ${TEST_FILE} | go run main.go)

FunctionConfig

To use a KCLRun as the functionConfig, the KCL source must be specified in the source field. Additional parameters can be specified in the params field. The params field supports any complex data structure as long as it can be represented in YAML.

apiVersion: krm.kcl.dev/v1alpha1
kind: KCLRun
metadata:
  name: conditionally-add-annotations
spec:
  params:
    toMatch:
      config.kubernetes.io/local-config: "true"
    toAdd:
      configmanagement.gke.io/managed: disabled
  source: |
    params = option("params")
    toMatch = params.toMatch
    toAdd = params.toAdd
    items = [item | {
       # If all annotations are matched, patch more annotations
       if all key, value in toMatch {
          item.metadata.annotations[key] == value
       }:
           metadata.annotations: toAdd
    } for item in option("items")]

In the example above, the script accesses the toMatch parameters using option("params").toMatch.

Besides, the source field supports different KCL sources, which can come from a local file, VCS such as github, OCI registry, http, etc. You can see the specific usage here. Take an OCI source as the example.

apiVersion: krm.kcl.dev/v1alpha1
kind: KCLRun
spec:
  params:
    annotations:
      config.kubernetes.io/local-config: "true"
  source: oci://ghcr.io/kcl-lang/set-annotation

Guides for Developing KCL

Here's what you can do in the KCL script:

• Read resources from option("resource_list"). The option("resource_list") complies with the KRM Functions Specification. You can read the input resources from option("items") and the functionConfig from option("functionConfig").
• Return a KRM list for output resources.
• Return an error using assert {condition}, {error_message}.
• Read the environment variables. e.g. option("PATH") (Not yet implemented).
• Read the OpenAPI schema. e.g. option("open_api")["definitions"]["io.k8s.api.apps.v1.Deployment"] (Not yet implemented).

Library

You can directly use KCL standard libraries without importing them, such as regex.match, math.log.

Tutorial

• See here to study more features of KCL.

Examples

• See here for more examples.

Tools and Integrations

• Kubectl KCL Plugin
• Helm KCL Plugin
• Helmfile KCL Plugin
• KPT KCL Plugin
• KCL Operator
• Crossplane KCL Function