Finish converting AppWrappers to v1beta2

Author: dgrove-oss

src/codeflare_sdk/cluster/awload.py

@@ -62,7 +62,7 @@ def submit(self) -> None:
             api_instance = client.CustomObjectsApi(api_config_handler())
             api_instance.create_namespaced_custom_object(
                 group="workload.codeflare.dev",
-                version="v1beta1",
+                version="v1beta2",
                 namespace=self.namespace,
                 plural="appwrappers",
                 body=self.awyaml,
@@ -87,7 +87,7 @@ def remove(self) -> None:
             api_instance = client.CustomObjectsApi(api_config_handler())
             api_instance.delete_namespaced_custom_object(
                 group="workload.codeflare.dev",
-                version="v1beta1",
+                version="v1beta2",
                 namespace=self.namespace,
                 plural="appwrappers",
                 name=self.name,

src/codeflare_sdk/cluster/cluster.py

@@ -199,7 +199,7 @@ def up(self):
                         aw = yaml.load(f, Loader=yaml.FullLoader)
                         api_instance.create_namespaced_custom_object(
                             group="workload.codeflare.dev",
-                            version="v1beta1",
+                            version="v1beta2",
                             namespace=namespace,
                             plural="appwrappers",
                             body=aw,
@@ -208,7 +208,7 @@ def up(self):
                     aw = yaml.safe_load(self.app_wrapper_yaml)
                     api_instance.create_namespaced_custom_object(
                         group="workload.codeflare.dev",
-                        version="v1beta1",
+                        version="v1beta2",
                         namespace=namespace,
                         plural="appwrappers",
                         body=aw,
@@ -250,7 +250,7 @@ def down(self):
             if self.config.appwrapper:
                 api_instance.delete_namespaced_custom_object(
                     group="workload.codeflare.dev",
-                    version="v1beta1",
+                    version="v1beta2",
                     namespace=namespace,
                     plural="appwrappers",
                     name=self.app_wrapper_name,
@@ -681,7 +681,7 @@ def _check_aw_exists(name: str, namespace: str) -> bool:
         api_instance = client.CustomObjectsApi(api_config_handler())
         aws = api_instance.list_namespaced_custom_object(
             group="workload.codeflare.dev",
-            version="v1beta1",
+            version="v1beta2",
             namespace=namespace,
             plural="appwrappers",
         )
@@ -741,7 +741,7 @@ def _app_wrapper_status(name, namespace="default") -> Optional[AppWrapper]:
         api_instance = client.CustomObjectsApi(api_config_handler())
         aws = api_instance.list_namespaced_custom_object(
             group="workload.codeflare.dev",
-            version="v1beta1",
+            version="v1beta2",
             namespace=namespace,
             plural="appwrappers",
         )
@@ -811,7 +811,7 @@ def _get_app_wrappers(
         api_instance = client.CustomObjectsApi(api_config_handler())
         aws = api_instance.list_namespaced_custom_object(
             group="workload.codeflare.dev",
-            version="v1beta1",
+            version="v1beta2",
             namespace=namespace,
             plural="appwrappers",
         )
@@ -908,7 +908,7 @@ def _map_to_app_wrapper(aw) -> AppWrapper:
     if "status" in aw:
         return AppWrapper(
             name=aw["metadata"]["name"],
-            status=AppWrapperStatus(aw["status"]["state"].lower()),
+            status=AppWrapperStatus(aw["status"]["phase"].lower()),
         )
     return AppWrapper(
         name=aw["metadata"]["name"],

src/codeflare_sdk/templates/base-template.yaml

@@ -81,7 +81,7 @@ def update_names(yaml, item, appwrapper_name, cluster_name, namespace):
     metadata = yaml.get("metadata")
     metadata["name"] = appwrapper_name
     metadata["namespace"] = namespace
-    lower_meta = item.get("generictemplate", {}).get("metadata")
+    lower_meta = item.get("template", {}).get("metadata")
     lower_meta["name"] = cluster_name
     lower_meta["namespace"] = namespace
 
@@ -140,11 +140,11 @@ def update_nodes(
     head_memory,
     head_gpus,
 ):
-    if "generictemplate" in item.keys():
-        head = item.get("generictemplate").get("spec").get("headGroupSpec")
+    if "template" in item.keys():
+        head = item.get("template").get("spec").get("headGroupSpec")
         head["rayStartParams"]["num-gpus"] = str(int(head_gpus))
 
-        worker = item.get("generictemplate").get("spec").get("workerGroupSpecs")[0]
+        worker = item.get("template").get("spec").get("workerGroupSpecs")[0]
         # Head counts as first worker
         worker["replicas"] = workers
         worker["minReplicas"] = workers
@@ -240,7 +240,7 @@ def write_components(
     if not os.path.exists(directory_path):
         os.makedirs(directory_path)
 
-    components = user_yaml.get("spec", "resources")["resources"].get("GenericItems")
+    components = user_yaml.get("spec", "resources").get("components")
     open(output_file_name, "w").close()
     lq_name = local_queue or get_default_kueue_name(namespace)
     cluster_labels = labels
@@ -250,14 +250,12 @@ def write_components(
         )
     with open(output_file_name, "a") as outfile:
         for component in components:
-            if "generictemplate" in component:
+            if "template" in component:
                 labels = component["generictemplate"]["metadata"]["labels"]
                 labels.update({"kueue.x-k8s.io/queue-name": lq_name})
                 labels.update(cluster_labels)
                 outfile.write("---\n")
-                yaml.dump(
-                    component["generictemplate"], outfile, default_flow_style=False
-                )
+                yaml.dump(component["template"], outfile, default_flow_style=False)
     print(f"Written to: {output_file_name}")
 
 
@@ -269,19 +267,19 @@ def load_components(
     labels: dict,
 ):
     component_list = []
-    components = user_yaml.get("spec", "resources")["resources"].get("GenericItems")
+    components = user_yaml.get("spec", "resources").get("components")
     lq_name = local_queue or get_default_kueue_name(namespace)
     cluster_labels = labels
     if not local_queue_exists(namespace, lq_name):
         raise ValueError(
             "local_queue provided does not exist or is not in this namespace. Please provide the correct local_queue name in Cluster Configuration"
         )
     for component in components:
-        if "generictemplate" in component:
-            labels = component["generictemplate"]["metadata"]["labels"]
+        if "template" in component:
+            labels = component["template"]["metadata"]["labels"]
             labels.update({"kueue.x-k8s.io/queue-name": lq_name})
             labels.update(cluster_labels)
-            component_list.append(component["generictemplate"])
+            component_list.append(component["template"])
 
     resources = "---\n" + "---\n".join(
         [yaml.dump(component) for component in component_list]
@@ -323,7 +321,7 @@ def generate_appwrapper(
     user_yaml = read_template(template)
     appwrapper_name, cluster_name = gen_names(name)
     resources = user_yaml.get("spec", "resources")
-    item = resources["resources"].get("GenericItems")[0]
+    item = resources.get("components")[0]
     update_names(
         user_yaml,
         item,

src/codeflare_sdk/utils/generate_yaml.py

@@ -1,112 +1,110 @@
-apiVersion: workload.codeflare.dev/v1beta1
+apiVersion: workload.codeflare.dev/v1beta2
 kind: AppsWrapper
 metadata:
   labels:
     orderedinstance: cpu.small_gpu.large
   nam: unit-test-cluster
   namspace: ns
 spec:
-  resources:
-    GenericItems:
-    - generictemplate:
-        apiVersion: ray.io/v1
-        kind: RayCluster
-        metadata:
-          labels:
-            controller-tools.k8s.io: '1.0'
-          name: unit-test-cluster
-          namespace: ns
-        spec:
-          autoscalerOptions:
-            idleTimeoutSeconds: 60
-            imagePullPolicy: Always
-            resources:
-              limits:
-                cpu: 500m
-                memory: 512Mi
-              requests:
-                cpu: 500m
-                memory: 512Mi
-            upscalingMode: Default
-          enableInTreeAutoscaling: false
-          headGroupSpec:
-            rayStartParams:
-              block: 'true'
-              dashboard-host: 0.0.0.0
-              num-gpus: '0'
-            serviceType: ClusterIP
-            template:
-              spec:
-                containers:
-                - env:
-                  - name: MY_POD_IP
-                    valueFrom:
-                      fieldRef:
-                        fieldPath: status.podIP
-                  image: quay.io/project-codeflare/ray:latest-py39-cu118
-                  imagePullPolicy: Always
-                  lifecycle:
-                    preStop:
-                      exec:
-                        command:
-                        - /bin/sh
-                        - -c
-                        - ray stop
-                  name: ray-head
-                  ports:
-                  - containerPort: 6379
-                    name: gcs
-                  - containerPort: 8265
-                    name: dashboard
-                  - containerPort: 10001
-                    name: client
-                  resources:
-                    limits:
-                      cpu: 2
-                      memory: 8G
-                      nvidia.com/gpu: 0
-                    requests:
-                      cpu: 2
-                      memory: 8G
-                      nvidia.com/gpu: 0
-          rayVersion: 1.12.0
-          workerGroupSpecs:
-          - groupName: small-group-unit-test-cluster
-            maxReplicas: 2
-            minReplicas: 2
-            rayStartParams:
-              block: 'true'
-              num-gpus: '7'
-            replicas: 2
-            template:
-              metadata:
-                annotations:
-                  key: value
-                labels:
-                  key: value
-              spec:
-                containers:
-                - env:
-                  - name: MY_POD_IP
-                    valueFrom:
-                      fieldRef:
-                        fieldPath: status.podIP
-                  image: quay.io/project-codeflare/ray:latest-py39-cu118
-                  lifecycle:
-                    preStop:
-                      exec:
-                        command:
-                        - /bin/sh
-                        - -c
-                        - ray stop
-                  name: machine-learning
-                  resources:
-                    limits:
-                      cpu: 4
-                      memory: 6G
-                      nvidia.com/gpu: 7
-                    requests:
-                      cpu: 3
-                      memory: 5G
-                      nvidia.com/gpu: 7
-      replicas: 1
+  components:
+  - template:
+      apiVersion: ray.io/v1
+      kind: RayCluster
+      metadata:
+        labels:
+          controller-tools.k8s.io: '1.0'
+        name: unit-test-cluster
+        namespace: ns
+      spec:
+        autoscalerOptions:
+          idleTimeoutSeconds: 60
+          imagePullPolicy: Always
+          resources:
+            limits:
+              cpu: 500m
+              memory: 512Mi
+            requests:
+              cpu: 500m
+              memory: 512Mi
+          upscalingMode: Default
+        enableInTreeAutoscaling: false
+        headGroupSpec:
+          rayStartParams:
+            block: 'true'
+            dashboard-host: 0.0.0.0
+            num-gpus: '0'
+          serviceType: ClusterIP
+          template:
+            spec:
+              containers:
+              - env:
+                - name: MY_POD_IP
+                  valueFrom:
+                    fieldRef:
+                      fieldPath: status.podIP
+                image: quay.io/project-codeflare/ray:latest-py39-cu118
+                imagePullPolicy: Always
+                lifecycle:
+                  preStop:
+                    exec:
+                      command:
+                      - /bin/sh
+                      - -c
+                      - ray stop
+                name: ray-head
+                ports:
+                - containerPort: 6379
+                  name: gcs
+                - containerPort: 8265
+                  name: dashboard
+                - containerPort: 10001
+                  name: client
+                resources:
+                  limits:
+                    cpu: 2
+                    memory: 8G
+                    nvidia.com/gpu: 0
+                  requests:
+                    cpu: 2
+                    memory: 8G
+                    nvidia.com/gpu: 0
+        rayVersion: 1.12.0
+        workerGroupSpecs:
+        - groupName: small-group-unit-test-cluster
+          maxReplicas: 2
+          minReplicas: 2
+          rayStartParams:
+            block: 'true'
+            num-gpus: '7'
+          replicas: 2
+          template:
+            metadata:
+              annotations:
+                key: value
+              labels:
+                key: value
+            spec:
+              containers:
+              - env:
+                - name: MY_POD_IP
+                  valueFrom:
+                    fieldRef:
+                      fieldPath: status.podIP
+                image: quay.io/project-codeflare/ray:latest-py39-cu118
+                lifecycle:
+                  preStop:
+                    exec:
+                      command:
+                      - /bin/sh
+                      - -c
+                      - ray stop
+                name: machine-learning
+                resources:
+                  limits:
+                    cpu: 4
+                    memory: 6G
+                    nvidia.com/gpu: 7
+                  requests:
+                    cpu: 3
+                    memory: 5G
+                    nvidia.com/gpu: 7