feat(ci): add CRD schema generation for kubeconform validation

Add scripts to extract OpenAPI v3 schemas from CRD definitions in all
kustomize overlays and write JSON schema files to ci/crd-schemas/ for
kubeconform validation. This allows kubeconform to validate CRD instances
(Elasticsearch, Kibana, CNPG Cluster, VictoriaMetrics, etc.) instead of
skipping or erroring on them.

- ci/generate-crd-schemas.py: extracts schemas from CRD YAML on stdin
- ci/generate-crd-schemas.sh: iterates overlays, pipes to Python script
- ci/validate-apps.sh, ci/validate-clusters.sh: add local schema-location fallback
- Makefile: add generate-schemas target
- add generate-schemas step to kubeconform woodpecker pipeline so schemas

ci/generate-crd-schemas.py

@@ -0,0 +1,45 @@
+#!/usr/bin/env python3
+"""
+Extract OpenAPI v3 schemas from CRD YAML on stdin and write JSON schema files
+to the output directory for use with kubeconform.
+
+Usage: kustomize build ... | python3 ci/generate-crd-schemas.py <output-dir>
+"""
+import sys
+import json
+import os
+
+import yaml
+
+
+def main() -> int:
+    output_dir = sys.argv[1] if len(sys.argv) > 1 else "ci/crd-schemas"
+    count = 0
+
+    for doc in yaml.safe_load_all(sys.stdin):
+        if not doc or doc.get("kind") != "CustomResourceDefinition":
+            continue
+
+        group = doc["spec"]["group"]
+        kind = doc["spec"]["names"]["kind"]
+        group_dir = os.path.join(output_dir, group)
+        os.makedirs(group_dir, exist_ok=True)
+
+        for ver in doc["spec"].get("versions", []):
+            if not ver.get("served", True):
+                continue
+            schema = ver.get("schema", {}).get("openAPIV3Schema")
+            if not schema:
+                continue
+            fname = os.path.join(group_dir, f"{kind.lower()}_{ver['name']}.json")
+            with open(fname, "w") as f:
+                json.dump({"$schema": "http://json-schema.org/schema#", **schema}, f, indent=2)
+                f.write("\n")
+            print(f"  wrote {fname}", file=sys.stderr)
+            count += 1
+
+    return count
+
+
+if __name__ == "__main__":
+    print(main())

ci/generate-crd-schemas.sh

@@ -0,0 +1,23 @@
+#!/usr/bin/env bash
+# Extract OpenAPI v3 schemas from CRD definitions in all kustomize overlays
+# and write JSON schema files to ci/crd-schemas/ for kubeconform validation.
+#
+# Run this script whenever CRD versions change, then commit the output.
+# Usage: ci/generate-crd-schemas.sh [output-dir]
+
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+OUTPUT_DIR="${1:-${SCRIPT_DIR}/crd-schemas}"
+mkdir -p "$OUTPUT_DIR"
+
+total=0
+
+while IFS= read -r -d "" k; do
+  dir="$(dirname "$k")"
+  n=$(kustomize build --enable-helm "$dir" 2>/dev/null \
+    | python3 "$SCRIPT_DIR/generate-crd-schemas.py" "$OUTPUT_DIR") || continue
+  total=$((total + n))
+done < <(find apps/overlays clusters -name kustomization.yaml -print0 | sort -z)
+
+echo "Generated ${total} schema(s) in ${OUTPUT_DIR}" >&2

ci/validate-apps.sh

@@ -6,6 +6,7 @@ KUBE_VERSION="1.33.7"
 schema_args=(
   -schema-location "https://raw.githubusercontent.com/yannh/kubernetes-json-schema/master/{{.NormalizedKubernetesVersion}}-standalone{{.StrictSuffix}}/{{.ResourceKind}}{{.KindSuffix}}.json"
   -schema-location "https://raw.githubusercontent.com/datreeio/CRDs-catalog/main/{{.Group}}/{{.ResourceKind}}_{{.ResourceAPIVersion}}.json"
+  -schema-location "file://${PWD}/ci/crd-schemas/{{.Group}}/{{.ResourceKind}}_{{.ResourceAPIVersion}}.json"
 )
 
 while IFS= read -r -d "" k; do
@@ -18,6 +19,6 @@ while IFS= read -r -d "" k; do
         -summary \
         -output pretty \
         -verbose \
-        -skip CustomResourceDefinition,GpuDevicePlugin,LBNodeAgent,ServiceGroup \
+        -skip GpuDevicePlugin,LBNodeAgent,ServiceGroup \
         "${schema_args[@]}"
 done < <(find apps/overlays -name kustomization.yaml -print0)

ci/validate-clusters.sh

@@ -6,6 +6,7 @@ KUBE_VERSION="1.33.7"
 schema_args=(
   -schema-location "https://raw.githubusercontent.com/yannh/kubernetes-json-schema/master/{{.NormalizedKubernetesVersion}}-standalone{{.StrictSuffix}}/{{.ResourceKind}}{{.KindSuffix}}.json"
   -schema-location "https://raw.githubusercontent.com/datreeio/CRDs-catalog/main/{{.Group}}/{{.ResourceKind}}_{{.ResourceAPIVersion}}.json"
+  -schema-location "file://${PWD}/ci/crd-schemas/{{.Group}}/{{.ResourceKind}}_{{.ResourceAPIVersion}}.json"
 )
 
 while IFS= read -r -d "" k; do