Improve new CLI testing ensuring complete coverage of arguments cases

user_tools/src/spark_rapids_tools/cmdli/argprocessor.py

@@ -173,6 +173,12 @@ def validate_onprem_with_cluster_name(self):
                 'invalid_argument',
                 f'Cannot run cluster by name with platform [{CspEnv.ONPREM}]\n  Error:')
 
+    def validate_onprem_with_cluster_props(self):
+        if self.platform == CspEnv.ONPREM:
+            raise PydanticCustomError(
+                'invalid_argument',
+                f'Cannot run cluster by properties with platform [{CspEnv.ONPREM}]\n  Error:')
+
     def init_extra_arg_cases(self) -> list:
         return []
 
@@ -229,12 +235,19 @@ def get_or_set_platform(self) -> CspEnv:
 
     def post_platform_assignment_validation(self, assigned_platform):
         # do some validation after we decide the cluster type
-        if self.argv_cases[1] == ArgValueCase.VALUE_A:
-            if assigned_platform == CspEnv.ONPREM:
-                # it is not allowed to run cluster_by_name on an OnPrem platform
+        if assigned_platform == CspEnv.ONPREM:
+            cluster_case = self.argv_cases[1]
+            eventlogs_case = self.argv_cases[2]
+            if cluster_case == ArgValueCase.VALUE_A:
+                # it is not allowed to run cluster by name on an OnPrem platform
                 raise PydanticCustomError(
                     'invalid_argument',
                     f'Cannot run cluster by name with platform [{CspEnv.ONPREM}]\n  Error:')
+            if cluster_case == ArgValueCase.VALUE_B and eventlogs_case == ArgValueCase.UNDEFINED:
+                # it is not allowed to run cluster by props on an OnPrem platform without eventlogs
+                raise PydanticCustomError(
+                    'invalid_argument',
+                    f'Cannot run cluster by properties with platform [{CspEnv.ONPREM}] without eventlogs\n  Error:')
 
 
 @dataclass
@@ -278,6 +291,13 @@ def define_invalid_arg_cases(self):
                 [ArgValueCase.VALUE_A, ArgValueCase.VALUE_A, ArgValueCase.IGNORE]
             ]
         }
+        self.rejected['Cluster By Properties Cannot go with OnPrem'] = {
+            'valid': False,
+            'callable': partial(self.validate_onprem_with_cluster_props),
+            'cases': [
+                [ArgValueCase.VALUE_A, ArgValueCase.VALUE_C, ArgValueCase.UNDEFINED]
+            ]
+        }
 
     def define_detection_cases(self):
         self.detected['Define Platform from Cluster Properties file'] = {

user_tools/tests/spark_rapids_tools_ut/conftest.py

@@ -16,7 +16,7 @@
 
 import sys
 
-import pytest   # pylint: disable=import-error
+import pytest  # pylint: disable=import-error
 
 
 def get_test_resources_path():
@@ -46,9 +46,10 @@ def gen_cpu_cluster_props():
 # all csps except onprem
 csps = ['dataproc', 'dataproc_gke', 'emr', 'databricks_aws', 'databricks_azure']
 all_csps = csps + ['onprem']
+autotuner_prop_path = 'worker_info.yaml'
 
 
-class SparkRapidsToolsUT:   # pylint: disable=too-few-public-methods
+class SparkRapidsToolsUT:  # pylint: disable=too-few-public-methods
 
     @pytest.fixture(autouse=True)
     def get_ut_data_dir(self):

user_tools/tests/spark_rapids_tools_ut/resources/worker_info.yaml

@@ -0,0 +1,19 @@
+system:
+  numCores: 32
+  memory: 212992MiB
+  numWorkers: 5
+gpu:
+  memory: 15109MiB
+  count: 4
+  name: T4
+softwareProperties:
+  spark.driver.maxResultSize: 7680m
+  spark.driver.memory: 15360m
+  spark.executor.cores: '8'
+  spark.executor.instances: '2'
+  spark.executor.memory: 47222m
+  spark.executorEnv.OPENBLAS_NUM_THREADS: '1'
+  spark.scheduler.mode: FAIR
+  spark.sql.cbo.enabled: 'true'
+  spark.ui.port: '0'
+  spark.yarn.am.memory: 640m