[MAINTENANCE] In ExecutionEngine: Make variable names and usage more …

…descriptive of their purpose. (#6342)
great-expectations · Nov 10, 2022 · 1669a8b · 1669a8b
1 parent 2d5ba40
commit 1669a8b
Show file tree

Hide file tree

Showing 5 changed files with 47 additions and 93 deletions.
diff --git a/great_expectations/execution_engine/execution_engine.py b/great_expectations/execution_engine/execution_engine.py
@@ -335,16 +335,16 @@ def resolve_metrics(  # noqa: C901 - 16
         accessor_domain_kwargs: dict
         metric_provider_kwargs: dict
         metric_to_resolve: MetricConfiguration
-        metric_dependencies: dict
+        resolved_metrics_by_metric_name: Dict[str, Any]
         k: str
         v: MetricConfiguration
         for metric_to_resolve in metrics_to_resolve:
-            metric_dependencies = {}
+            resolved_metrics_by_metric_name = {}
             for k, v in metric_to_resolve.metric_dependencies.items():
                 if v.id in metrics:
-                    metric_dependencies[k] = metrics[v.id]
+                    resolved_metrics_by_metric_name[k] = metrics[v.id]
                 elif self._caching and v.id in self._metric_cache:  # type: ignore[operator] # TODO: update NoOpDict
-                    metric_dependencies[k] = self._metric_cache[v.id]
+                    resolved_metrics_by_metric_name[k] = self._metric_cache[v.id]
                 else:
                     raise ge_exceptions.MetricError(
                         message=f'Missing metric dependency: {str(k)} for metric "{metric_to_resolve.metric_name}".'
@@ -358,7 +358,7 @@ def resolve_metrics(  # noqa: C901 - 16
                 "execution_engine": self,
                 "metric_domain_kwargs": metric_to_resolve.metric_domain_kwargs,
                 "metric_value_kwargs": metric_to_resolve.metric_value_kwargs,
-                "metrics": metric_dependencies,
+                "metrics": resolved_metrics_by_metric_name,
                 "runtime_configuration": runtime_configuration,
             }
             if metric_fn is None:
@@ -367,7 +367,7 @@ def resolve_metrics(  # noqa: C901 - 16
                         metric_fn,
                         compute_domain_kwargs,
                         accessor_domain_kwargs,
-                    ) = metric_dependencies.pop("metric_partial_fn")
+                    ) = resolved_metrics_by_metric_name.pop("metric_partial_fn")
                 except KeyError as e:
                     raise ge_exceptions.MetricError(
                         message=f'Missing metric dependency: {str(e)} for metric "{metric_to_resolve.metric_name}".'

diff --git a/great_expectations/execution_engine/pandas_execution_engine.py b/great_expectations/execution_engine/pandas_execution_engine.py
@@ -22,6 +22,7 @@
 from great_expectations.core.metric_domain_types import MetricDomainTypes
 from great_expectations.core.util import AzureUrl, GCSUrl, S3Url, sniff_s3_compression
 from great_expectations.execution_engine import ExecutionEngine
+from great_expectations.execution_engine.execution_engine import SplitDomainKwargs
 from great_expectations.execution_engine.pandas_batch_data import PandasBatchData
 from great_expectations.execution_engine.split_and_sample.pandas_data_sampler import (
     PandasDataSampler,
@@ -602,15 +603,15 @@ class MetricDomainTypes.
               - a dictionary of accessor_domain_kwargs, describing any accessors needed to
                 identify the domain within the compute domain
         """
-        data = self.get_domain_records(domain_kwargs)
-
-        table = domain_kwargs.get("table", None)
+        table: str = domain_kwargs.get("table", None)
         if table:
             raise ValueError(
                 "PandasExecutionEngine does not currently support multiple named tables."
             )
 
-        split_domain_kwargs = self._split_domain_kwargs(
+        data: pd.DataFrame = self.get_domain_records(domain_kwargs=domain_kwargs)
+
+        split_domain_kwargs: SplitDomainKwargs = self._split_domain_kwargs(
             domain_kwargs, domain_type, accessor_keys
         )
 

diff --git a/great_expectations/execution_engine/sparkdf_execution_engine.py b/great_expectations/execution_engine/sparkdf_execution_engine.py
@@ -36,6 +36,7 @@
 from great_expectations.execution_engine.bundled_metric_configuration import (
     BundledMetricConfiguration,
 )
+from great_expectations.execution_engine.execution_engine import SplitDomainKwargs
 from great_expectations.execution_engine.sparkdf_batch_data import SparkDFBatchData
 from great_expectations.execution_engine.split_and_sample.sparkdf_data_sampler import (
     SparkDataSampler,
@@ -614,15 +615,15 @@ class MetricDomainTypes.
               - a dictionary of accessor_domain_kwargs, describing any accessors needed to
                 identify the domain within the compute domain
         """
-        data = self.get_domain_records(domain_kwargs)
-
-        table = domain_kwargs.get("table", None)
+        table: str = domain_kwargs.get("table", None)
         if table:
             raise ValueError(
                 "SparkDFExecutionEngine does not currently support multiple named tables."
             )
 
-        split_domain_kwargs = self._split_domain_kwargs(
+        data: DataFrame = self.get_domain_records(domain_kwargs=domain_kwargs)
+
+        split_domain_kwargs: SplitDomainKwargs = self._split_domain_kwargs(
             domain_kwargs, domain_type, accessor_keys
         )
 
@@ -721,9 +722,7 @@ def resolve_metric_bundle(
 
         for aggregate in aggregates.values():
             domain_kwargs: dict = aggregate["domain_kwargs"]
-            df: Optional[DataFrame] = self.get_domain_records(
-                domain_kwargs=domain_kwargs,
-            )
+            df: DataFrame = self.get_domain_records(domain_kwargs=domain_kwargs)
 
             assert len(aggregate["column_aggregates"]) == len(aggregate["ids"])
 

diff --git a/great_expectations/execution_engine/sqlalchemy_execution_engine.py b/great_expectations/execution_engine/sqlalchemy_execution_engine.py
@@ -788,12 +788,12 @@ class MetricDomainTypes.
         Returns:
             SqlAlchemy column
         """
-        selectable = self.get_domain_records(domain_kwargs)
-
-        split_domain_kwargs = self._split_domain_kwargs(
+        split_domain_kwargs: SplitDomainKwargs = self._split_domain_kwargs(
             domain_kwargs, domain_type, accessor_keys
         )
 
+        selectable: Selectable = self.get_domain_records(domain_kwargs=domain_kwargs)
+
         return selectable, split_domain_kwargs.compute, split_domain_kwargs.accessor
 
     def _split_column_metric_domain_kwargs(  # type: ignore[override] # ExecutionEngine method is static
@@ -993,8 +993,8 @@ def resolve_metric_bundle(
 
         for query in queries.values():
             domain_kwargs: dict = query["domain_kwargs"]
-            selectable: Any = self.get_domain_records(
-                domain_kwargs=domain_kwargs,
+            selectable: Selectable = self.get_domain_records(
+                domain_kwargs=domain_kwargs
             )
 
             assert len(query["select"]) == len(query["ids"])