Add unit tests for the opctl backends

Author: mrDzurb

ads/jobs/ads_job.py

@@ -502,7 +502,7 @@ def from_dict(cls, config: dict) -> "Job":
         Raises
         ------
         NotImplementedError
-            If the type of the intrastructure or runtime is not supported.
+            If the type of the infrastructure or runtime is not supported.
         """
         if not isinstance(config, dict):
             raise ValueError("The config data for initializing the job is invalid.")

ads/opctl/backend/ads_ml_pipeline.py

@@ -130,7 +130,10 @@ def init(
 
             # define a pipeline
             pipeline = (
-                Pipeline(**(self.config.get("infrastructure", {}) or {}))
+                Pipeline(
+                    name="Pipeline Name",
+                    spec=(self.config.get("infrastructure", {}) or {}),
+                )
                 .with_step_details([pipeline_step])
                 .with_dag(["pipeline_step_name_1"])
                 .build()

tests/unitary/with_extras/opctl/test_opctl_dataflow_backend.py

@@ -3,28 +3,46 @@
 # Copyright (c) 2022, 2023 Oracle and/or its affiliates.
 # Licensed under the Universal Permissive License v 1.0 as shown at https://oss.oracle.com/licenses/upl/
 
+import json
 import os
 import tempfile
 from pathlib import Path
 from unittest.mock import patch
-import json
 
 import pytest
+import yaml
 
 from ads.opctl.backend.ads_dataflow import DataFlowBackend
 
 
 class TestDataFlowBackend:
-    def test_dataflow_apply(self):
-        config = {
+    @property
+    def curr_dir(self):
+        return os.path.dirname(os.path.abspath(__file__))
+
+    @property
+    def config(self):
+        return {
             "execution": {
                 "backend": "dataflow",
+                "auth": "api_key",
                 "oci_profile": "DEFAULT",
                 "oci_config": "~/.oci/config",
-            }
+            },
+            "infrastructure": {
+                "compartment_id": "ocid1.compartment.oc1..<unique_id>",
+                "driver_shape": "VM.Standard.E2.4",
+                "executor_shape": "VM.Standard.E2.4",
+                "logs_bucket_uri": "oci://bucket@namespace",
+                "script_bucket": "oci://bucket@namespace/prefix",
+                "num_executors": "1",
+                "spark_version": "3.2.1",
+            },
         }
-        with pytest.raises(ValueError):
-            DataFlowBackend(config).apply()
+
+    def test_dataflow_apply(self):
+        with pytest.raises(NotImplementedError):
+            DataFlowBackend(self.config).apply()
 
     @patch("ads.jobs.builders.infrastructure.dataflow.DataFlowApp.create")
     @patch("ads.opctl.backend.ads_dataflow.Job.run")
@@ -68,3 +86,30 @@ def test_dataflow_run(self, file_upload, job_run, job_create):
                 "oci://<bucket_name>@<namespace>/<prefix>",
                 False,
             )
+
+    @pytest.mark.parametrize(
+        "runtime_type",
+        ["dataFlow", "dataFlowNotebook"],
+    )
+    def test_init(self, runtime_type, monkeypatch):
+        """Ensures that starter YAML can be generated for every supported runtime of the Data Flow."""
+        monkeypatch.delenv("NB_SESSION_OCID", raising=False)
+
+        with tempfile.TemporaryDirectory() as td:
+            test_yaml_uri = os.path.join(td, f"dataflow_{runtime_type}.yaml")
+            expected_yaml_uri = os.path.join(
+                self.curr_dir, "test_files", f"dataflow_{runtime_type}.yaml"
+            )
+
+            DataFlowBackend(self.config).init(
+                uri=test_yaml_uri,
+                overwrite=False,
+                runtime_type=runtime_type,
+            )
+
+            with open(test_yaml_uri, "r") as stream:
+                test_yaml_dict = yaml.safe_load(stream)
+            with open(expected_yaml_uri, "r") as stream:
+                expected_yaml_dict = yaml.safe_load(stream)
+
+            assert test_yaml_dict == expected_yaml_dict

tests/unitary/with_extras/opctl/test_opctl_ml_job_backend.py

@@ -6,6 +6,8 @@
 import tempfile
 import os
 from unittest.mock import patch
+import pytest
+import yaml
 
 from ads.opctl.backend.ads_ml_job import MLJobBackend
 from ads.jobs import Job, DataScienceJobRun
@@ -106,3 +108,31 @@ def test_run_with_image(self, rt, job_run, job_create):
         rt.with_cmd.assert_called_with("-n,hello-world,-c,~/.oci/config,-p,DEFAULT")
         job_create.assert_called()
         job_run.assert_called()
+
+    @pytest.mark.parametrize(
+        "runtime_type",
+        ["container", "script", "python", "notebook", "gitPython"],
+    )
+    def test_init(self, runtime_type, monkeypatch):
+        """Ensures that starter YAML can be generated for every supported runtime of the Job."""
+
+        monkeypatch.delenv("NB_SESSION_OCID", raising=False)
+
+        with tempfile.TemporaryDirectory() as td:
+            test_yaml_uri = os.path.join(td, f"job_{runtime_type}.yaml")
+            expected_yaml_uri = os.path.join(
+                self.curr_dir, "test_files", f"job_{runtime_type}.yaml"
+            )
+
+            MLJobBackend(self.config).init(
+                uri=test_yaml_uri,
+                overwrite=False,
+                runtime_type=runtime_type,
+            )
+
+            with open(test_yaml_uri, "r") as stream:
+                test_yaml_dict = yaml.safe_load(stream)
+            with open(expected_yaml_uri, "r") as stream:
+                expected_yaml_dict = yaml.safe_load(stream)
+
+            assert test_yaml_dict == expected_yaml_dict

tests/unitary/with_extras/opctl/test_opctl_ml_pipeline_backend.py

@@ -3,19 +3,23 @@
 # Copyright (c) 2022, 2023 Oracle and/or its affiliates.
 # Licensed under the Universal Permissive License v 1.0 as shown at https://oss.oracle.com/licenses/upl/
 
-import unittest
+
+import os
+import tempfile
 from unittest.mock import patch
 
-try:
-    from ads.opctl.backend.ads_ml_pipeline import PipelineBackend
-    from ads.pipeline import Pipeline, PipelineRun
-except (ImportError, AttributeError) as e:
-    raise unittest.SkipTest(
-        "OCI MLPipeline is not available. Skipping the MLPipeline tests."
-    )
+import pytest
+import yaml
+
+from ads.opctl.backend.ads_ml_pipeline import PipelineBackend
+from ads.pipeline import Pipeline, PipelineRun
 
 
 class TestMLPipelineBackend:
+    @property
+    def curr_dir(self):
+        return os.path.dirname(os.path.abspath(__file__))
+
     @property
     def config(self):
         return {
@@ -27,7 +31,13 @@ def config(self):
                 "oci_profile": "DEFAULT",
                 "ocid": "test",
                 "auth": "api_key",
-            }
+            },
+            "infrastructure": {
+                "compartment_id": "ocid1.compartment.oc1..<unique_id>",
+                "project_id": "ocid1.datascienceproject.oc1.<unique_id>",
+                "log_group_id": "ocid1.loggroup.oc1.iad.<unique_id>",
+                "log_id": "ocid1.log.oc1.iad.<unique_id>",
+            },
         }
 
     @patch(
@@ -123,3 +133,36 @@ def test_watch(self, mock_from_ocid, mock_watch):
         backend.watch()
         mock_from_ocid.assert_called_with("test_pipeline_run_id")
         mock_watch.assert_called_with(log_type="custom_log")
+
+    @pytest.mark.parametrize(
+        "runtime_type",
+        ["container", "script", "python", "notebook", "gitPython"],
+    )
+    def test_init(self, runtime_type, monkeypatch):
+        """Ensures that starter YAML can be generated for every supported runtime of the Data Flow."""
+
+        monkeypatch.delenv("NB_SESSION_OCID", raising=False)
+        monkeypatch.setenv(
+            "NB_SESSION_COMPARTMENT_OCID",
+            self.config["infrastructure"]["compartment_id"],
+        )
+        monkeypatch.setenv("PROJECT_OCID", self.config["infrastructure"]["project_id"])
+
+        with tempfile.TemporaryDirectory() as td:
+            test_yaml_uri = os.path.join(td, f"pipeline_{runtime_type}.yaml")
+            expected_yaml_uri = os.path.join(
+                self.curr_dir, "test_files", f"pipeline_{runtime_type}.yaml"
+            )
+
+            PipelineBackend(self.config).init(
+                uri=test_yaml_uri,
+                overwrite=False,
+                runtime_type=runtime_type,
+            )
+
+            with open(test_yaml_uri, "r") as stream:
+                test_yaml_dict = yaml.safe_load(stream)
+            with open(expected_yaml_uri, "r") as stream:
+                expected_yaml_dict = yaml.safe_load(stream)
+
+            assert test_yaml_dict == expected_yaml_dict

tests/unitary/with_extras/opctl/test_opctl_model_deployment_backend.py

@@ -4,13 +4,22 @@
 # Copyright (c) 2023 Oracle and/or its affiliates.
 # Licensed under the Universal Permissive License v 1.0 as shown at https://oss.oracle.com/licenses/upl/
 
+import os
+import tempfile
 from unittest.mock import patch
 
-from ads.opctl.backend.ads_model_deployment import ModelDeploymentBackend
+import pytest
+import yaml
+
 from ads.model import ModelDeployment
+from ads.opctl.backend.ads_model_deployment import ModelDeploymentBackend
 
 
 class TestModelDeploymentBackend:
+    @property
+    def curr_dir(self):
+        return os.path.dirname(os.path.abspath(__file__))
+
     @property
     def config(self):
         return {
@@ -27,7 +36,17 @@ def config(self):
                 "log_type": "predict",
                 "log_filter": "test_filter",
                 "interval": 3,
-            }
+            },
+            "infrastructure": {
+                "compartment_id": "ocid1.compartment.oc1..<unique_id>",
+                "project_id": "ocid1.datascienceproject.oc1.<unique_id>",
+                "log_group_id": "ocid1.loggroup.oc1.iad.<unique_id>",
+                "log_id": "ocid1.log.oc1.iad.<unique_id>",
+                "shape_name": "VM.Standard.E2.4",
+                "bandwidth_mbps": 10,
+                "replica": 1,
+                "web_concurrency": 10,
+            },
         }
 
     @patch("ads.opctl.backend.ads_model_deployment.ModelDeployment.deploy")
@@ -103,3 +122,33 @@ def test_watch(self, mock_from_id, mock_watch):
         mock_watch.assert_called_with(
             log_type="predict", interval=3, log_filter="test_filter"
         )
+
+    @pytest.mark.parametrize(
+        "runtime_type",
+        ["container", "conda"],
+    )
+    def test_init(self, runtime_type, monkeypatch):
+        """Ensures that starter YAML can be generated for every supported runtime of the Data Flow."""
+
+        # For every supported runtime generate a YAML -> test_files
+        # On second iteration remove a temporary code and compare result YAML.
+        monkeypatch.delenv("NB_SESSION_OCID", raising=False)
+
+        with tempfile.TemporaryDirectory() as td:
+            test_yaml_uri = os.path.join(td, f"modeldeployment_{runtime_type}.yaml")
+            expected_yaml_uri = os.path.join(
+                self.curr_dir, "test_files", f"modeldeployment_{runtime_type}.yaml"
+            )
+
+            ModelDeploymentBackend(self.config).init(
+                uri=test_yaml_uri,
+                overwrite=False,
+                runtime_type=runtime_type,
+            )
+
+            with open(test_yaml_uri, "r") as stream:
+                test_yaml_dict = yaml.safe_load(stream)
+            with open(expected_yaml_uri, "r") as stream:
+                expected_yaml_dict = yaml.safe_load(stream)
+
+            assert test_yaml_dict == expected_yaml_dict