fix: black format linting

Author: BLMgithub

data_pipeline/contract/contract_logic.py

@@ -125,7 +125,6 @@ def remove_rows_with_null_constraint(
         tuple: (Filtered DataFrame, Count of dropped rows, Set of invalid order_ids)
     """
 
-
     initial_count = len(df)
     invalid_ids = set()
 

tests/test_semantic_stage.py

@@ -53,22 +53,30 @@ def valid_assembled_df():
             "order_revenue": [12.34, 56.78],
             "product_id": ["prod1", "prod2"],
             "order_status": ["delivered", "delivered"],
-            "order_purchase_timestamp": pd.to_datetime([
-                "2023-01-02 09:00:00",
-                "2023-01-10 14:00:00",
-            ]),
-            "order_approved_at": pd.to_datetime([
-                "2023-01-03 09:00:00",
-                "2023-01-11 14:00:00",
-            ]),
-            "order_delivered_timestamp": pd.to_datetime([
-                "2023-01-06 09:00:00",
-                "2023-01-16 14:00:00",
-            ]),
-            "order_estimated_delivery_date": pd.to_datetime([
-                "2023-01-05",
-                "2023-01-15",
-            ]),
+            "order_purchase_timestamp": pd.to_datetime(
+                [
+                    "2023-01-02 09:00:00",
+                    "2023-01-10 14:00:00",
+                ]
+            ),
+            "order_approved_at": pd.to_datetime(
+                [
+                    "2023-01-03 09:00:00",
+                    "2023-01-11 14:00:00",
+                ]
+            ),
+            "order_delivered_timestamp": pd.to_datetime(
+                [
+                    "2023-01-06 09:00:00",
+                    "2023-01-16 14:00:00",
+                ]
+            ),
+            "order_estimated_delivery_date": pd.to_datetime(
+                [
+                    "2023-01-05",
+                    "2023-01-15",
+                ]
+            ),
             "lead_time_days": [3, 5],
             "approval_lag_days": [1, 1],
             "delivery_delay_days": [1, 1],
@@ -77,14 +85,16 @@ def valid_assembled_df():
             "order_year_week": ["2023-W01", "2023-W01"],
             "run_id": ["20230101T120000", "20230101T120000"],
         }
-    ).astype({
-        "order_status": "category",
-        "lead_time_days": "int16",
-        "approval_lag_days": "int16",
-        "delivery_delay_days": "int16",
-        "order_year": "int16",
-        "order_revenue": "float32",
-    })
+    ).astype(
+        {
+            "order_status": "category",
+            "lead_time_days": "int16",
+            "approval_lag_days": "int16",
+            "delivery_delay_days": "int16",
+            "order_year": "int16",
+            "order_revenue": "float32",
+        }
+    )
 
 
 # =============================================================================
@@ -116,8 +126,10 @@ def test_log_info_appends_only_to_info(empty_report):
 def test_seller_semantic_model_grain_preserved_success(tmp_path, valid_assembled_df):
     run_context = RunContext.create(base=tmp_path, run_id="20230101T120000")
     seller_semantic = build_seller_semantic(valid_assembled_df, run_context)
-    
-    expected_fact_len = valid_assembled_df[["seller_id", "order_year_week"]].drop_duplicates().shape[0]
+
+    expected_fact_len = (
+        valid_assembled_df[["seller_id", "order_year_week"]].drop_duplicates().shape[0]
+    )
     assert len(seller_semantic["seller_weekly_fact"]) == expected_fact_len
 
     expected_dim_len = valid_assembled_df["seller_id"].nunique()
@@ -149,14 +161,20 @@ def test_build_semantic_layer_success(
     run_context.initialize_directories()
 
     # Setup Assembled layer
-    valid_assembled_df.to_parquet(run_context.assembled_path / "assembled_events_2023_01_01.parquet")
-    valid_customers_df.to_parquet(run_context.assembled_path / "df_customers_2023_01_01.parquet")
-    valid_products_df.to_parquet(run_context.assembled_path / "df_products_2023_01_01.parquet")
+    valid_assembled_df.to_parquet(
+        run_context.assembled_path / "assembled_events_2023_01_01.parquet"
+    )
+    valid_customers_df.to_parquet(
+        run_context.assembled_path / "df_customers_2023_01_01.parquet"
+    )
+    valid_products_df.to_parquet(
+        run_context.assembled_path / "df_products_2023_01_01.parquet"
+    )
 
     report = build_semantic_layer(run_context)
 
     assert report["status"] == "success"
-    
+
     for module_name, module_config in SEMANTIC_MODULES.items():
         for table_name in module_config["tables"]:
             outputs_path = (
@@ -175,14 +193,18 @@ def test_build_semantic_layer_fails_on_multiple_ids(tmp_path, valid_assembled_df
     broken_assembled = valid_assembled_df.copy()
     broken_assembled.loc[1, "run_id"] = "another_run"
 
-    broken_assembled.to_parquet(run_context.assembled_path / "assembled_events_2023_01_01.parquet")
+    broken_assembled.to_parquet(
+        run_context.assembled_path / "assembled_events_2023_01_01.parquet"
+    )
 
     report = build_semantic_layer(run_context)
 
     assert report["status"] == "failed"
     assert report["modules"]["seller_semantic"]["build_stage"]["status"] == "failed"
-    assert any("Multiple run_ids detected" in error 
-               for error in report["modules"]["seller_semantic"]["build_stage"]["errors"])
+    assert any(
+        "Multiple run_ids detected" in error
+        for error in report["modules"]["seller_semantic"]["build_stage"]["errors"]
+    )
 
 
 def test_build_semantic_layer_fails_on_missing_columns(tmp_path, valid_assembled_df):
@@ -191,9 +213,13 @@ def test_build_semantic_layer_fails_on_missing_columns(tmp_path, valid_assembled
     run_context.initialize_directories()
 
     broken_assembled = valid_assembled_df.copy()
-    broken_assembled.drop(columns="order_revenue", inplace=True) # Used in seller_weekly_fact
+    broken_assembled.drop(
+        columns="order_revenue", inplace=True
+    )  # Used in seller_weekly_fact
 
-    broken_assembled.to_parquet(run_context.assembled_path / "assembled_events_2023_01_01.parquet")
+    broken_assembled.to_parquet(
+        run_context.assembled_path / "assembled_events_2023_01_01.parquet"
+    )
 
     report = build_semantic_layer(run_context)
 
@@ -210,11 +236,15 @@ def test_build_semantic_layer_fails_on_missing_or_empty_df(tmp_path):
     run_context.initialize_directories()
 
     empty_df = pd.DataFrame()
-    empty_df.to_parquet(run_context.assembled_path / "assembled_events_2023_01_01.parquet")
+    empty_df.to_parquet(
+        run_context.assembled_path / "assembled_events_2023_01_01.parquet"
+    )
 
     report = build_semantic_layer(run_context)
 
     assert report["status"] == "failed"
     # load_single_delta itself succeeds even if DF is empty
-    assert any("missing or empty" in error 
-               for error in report["steps"]["load_tables"]["errors"])
+    assert any(
+        "missing or empty" in error
+        for error in report["steps"]["load_tables"]["errors"]
+    )