[Feature] add generic config for generic tests

.changes/unreleased/Features-20240531-150816.yaml

@@ -0,0 +1,6 @@
+kind: Features
+body: Update data_test to accept arbitrary config options
+time: 2024-05-31T15:08:16.431966-05:00
+custom:
+    Author: McKnight-42
+    Issue: "10197"

core/dbt/parser/generic_test_builders.py

@@ -125,62 +125,71 @@
         self.name: str = groups["test_name"]
         self.namespace: str = groups["test_namespace"]
         self.config: Dict[str, Any] = {}
+        self.initialize_config(render_ctx, column_name)
 
-        # This code removes keys identified as config args from the test entry
-        # dictionary. The keys remaining in the 'args' dictionary will be
-        # "kwargs", or keyword args that are passed to the test macro.
-        # The "kwargs" are not rendered into strings until compilation time.
-        # The "configs" are rendered here (since they were not rendered back
-        # in the 'get_key_dicts' methods in the schema parsers).
+        if self.namespace is not None:
+            self.package_name = self.namespace
+
+        # If the user has provided a custom name for this generic test, use it
+        # Then delete the "name" argument to avoid passing it into the test macro
+        # Otherwise, use an auto-generated name synthesized from test inputs
+        self.compiled_name: str = ""
+        self.fqn_name: str = ""
+
+        if "name" in self.args:
+            # Assign the user-defined name here, which will be checked for uniqueness later
+            # we will raise an error if two tests have same name for same model + column combo
+            self.compiled_name = self.args["name"]
+            self.fqn_name = self.args["name"]
+            del self.args["name"]
+        else:
+            short_name, full_name = self.get_synthetic_test_names()
+            self.compiled_name = short_name
+            self.fqn_name = full_name
+            # use hashed name as alias if full name is too long
+            if short_name != full_name and "alias" not in self.config:
+                self.config["alias"] = short_name
+
+    def initialize_config(self, render_ctx, column_name):
         for key in self.CONFIG_ARGS:
             value = self.args.pop(key, None)
-            # 'modifier' config could be either top level arg or in config
             if value and "config" in self.args and key in self.args["config"]:
                 raise SameKeyNestedError()
             if not value and "config" in self.args:
                 value = self.args["config"].pop(key, None)
             if isinstance(value, str):
-
                 try:
                     value = get_rendered(value, render_ctx, native=True)
                 except UndefinedMacroError as e:
-
                     raise CustomMacroPopulatingConfigValueError(
                         target_name=self.target.name,
                         column_name=column_name,
                         name=self.name,
                         key=key,
                         err_msg=e.msg,
                     )
-
             if value is not None:
                 self.config[key] = value
 
         if "config" in self.args:
+            self.process_config_args(self.args["config"], render_ctx, column_name)
             del self.args["config"]
 
-        if self.namespace is not None:
-            self.package_name = self.namespace
-
-        # If the user has provided a custom name for this generic test, use it
-        # Then delete the "name" argument to avoid passing it into the test macro
-        # Otherwise, use an auto-generated name synthesized from test inputs
-        self.compiled_name: str = ""
-        self.fqn_name: str = ""
-
-        if "name" in self.args:
-            # Assign the user-defined name here, which will be checked for uniqueness later
-            # we will raise an error if two tests have same name for same model + column combo
-            self.compiled_name = self.args["name"]
-            self.fqn_name = self.args["name"]
-            del self.args["name"]
-        else:
-            short_name, full_name = self.get_synthetic_test_names()
-            self.compiled_name = short_name
-            self.fqn_name = full_name
-            # use hashed name as alias if full name is too long
-            if short_name != full_name and "alias" not in self.config:
-                self.config["alias"] = short_name
+    def process_config_args(self, config_dict, render_ctx, column_name):
+        for key, value in config_dict.items():
+            if isinstance(value, str):
+                try:
+                    value = get_rendered(value, render_ctx, native=True)
+                except UndefinedMacroError as e:
+                    raise CustomMacroPopulatingConfigValueError(
+                        target_name=self.target.name,
+                        column_name=column_name,
+                        name=self.name,
+                        key=key,
+                        err_msg=e.msg,
+                    )
+            if value is not None:
+                self.config[key] = value
 
     def _bad_type(self) -> TypeError:
         return TypeError('invalid target type "{}"'.format(type(self.target)))

tests/functional/schema_tests/data_test_config.py

@@ -0,0 +1,83 @@
+import pytest
+
+from dbt.exceptions import CompilationError
+from dbt.tests.util import get_manifest, run_dbt
+from tests.functional.schema_tests.fixtures import (
+    custom_config_yml,
+    mixed_config_yml,
+    same_key_error_yml,
+    seed_csv,
+    table_sql,
+)
+
+
+class BaseDataTestsConfig:
+    @pytest.fixture(scope="class")
+    def seeds(self):
+        return {"seed.csv": seed_csv}
+
+    @pytest.fixture(scope="class")
+    def project_config_update(self):
+        return {
+            "config-version": 2,
+        }
+
+    @pytest.fixture(scope="class", autouse=True)
+    def setUp(self, project):
+        run_dbt(["seed"])
+
+
+class TestCustomDataTestConfig(BaseDataTestsConfig):
+    @pytest.fixture(scope="class")
+    def models(self):
+        return {"table.sql": table_sql, "custom_config.yml": custom_config_yml}
+
+    def test_custom_config(self, project):
+        run_dbt(["parse"])
+        manifest = get_manifest(project.project_root)
+        test_id = "test.test.accepted_values_table_color__blue__red.9482147132"
+        assert test_id in manifest.nodes
+        test_node = manifest.nodes[test_id]
+        assert "custom_config_key" in test_node.config
+        assert test_node.config["custom_config_key"] == "some_value"
+
+
+class TestMixedDataTestConfig(BaseDataTestsConfig):
+    @pytest.fixture(scope="class")
+    def models(self):
+        return {"table.sql": table_sql, "mixed_config.yml": mixed_config_yml}
+
+    def test_mixed_config(self, project):
+        run_dbt(["parse"])
+        manifest = get_manifest(project.project_root)
+        test_id = "test.test.accepted_values_table_color__blue__red.9482147132"
+        assert test_id in manifest.nodes
+        test_node = manifest.nodes[test_id]
+        assert "custom_config_key" in test_node.config
+        assert test_node.config["custom_config_key"] == "some_value"
+        assert "severity" in test_node.config
+        assert test_node.config["severity"] == "warn"
+
+
+class TestSameKeyErrorDataTestConfig:
+    @pytest.fixture(scope="class")
+    def models(self):
+        return {"table.sql": table_sql, "same_key_error.yml": same_key_error_yml}
+
+    def test_same_key_error(self, project):
+        """
+        Test that verifies dbt raises a CompilationError when the test configuration
+        contains the same key at the top level and inside the config dictionary.
+        """
+        # Run dbt and expect a CompilationError due to the invalid configuration
+        with pytest.raises(CompilationError) as exc_info:
+            run_dbt(["parse"])
+
+        # Extract the exception message
+        exception_message = str(exc_info.value)
+
+        # Assert that the error message contains the expected text
+        assert "Test cannot have the same key at the top-level and in config" in exception_message
+
+        # Assert that the error message contains the context of the error
+        assert "models/same_key_error.yml" in exception_message

tests/functional/schema_tests/fixtures.py

@@ -1273,3 +1273,63 @@
     data_tests:
       - my_custom_test
 """
+
+custom_config_yml = """
+version: 2
+models:
+  - name: table
+    columns:
+      - name: color
+        data_tests:
+          - accepted_values:
+              values: ['blue', 'red']
+              config:
+                custom_config_key: some_value
+"""
+
+mixed_config_yml = """
+version: 2
+models:
+  - name: table
+    columns:
+      - name: color
+        data_tests:
+          - accepted_values:
+              values: ['blue', 'red']
+              severity: warn
+              config:
+                custom_config_key: some_value
+"""
+
+same_key_error_yml = """
+version: 2
+models:
+  - name: table
+    columns:
+      - name: color
+        data_tests:
+          - accepted_values:
+              values: ['blue', 'red']
+              severity: warn
+              config:
+                severity: error
+"""
+
+seed_csv = """
+id,color,value
+1,blue,10
+2,red,20
+3,green,30
+4,yellow,40
+5,blue,50
+6,red,60
+7,blue,70
+8,green,80
+9,yellow,90
+10,blue,100
+"""
+
+table_sql = """
+-- content of the table.sql
+select * from {{ ref('seed') }}
+"""