allow pipeline to run without manual matches provided

Author: sposerina

src/server/pipeline/clean_and_load_data.py

@@ -13,7 +13,8 @@
 def start(connection, pdp_contacts_df, file_path_list):
     result = pd.DataFrame(columns=pdp_contacts_df.columns)
     json_rows = pd.DataFrame(columns=["source_type", "source_id", "json"])
-
+    manual_matches_df = None
+    
     for uploaded_file in file_path_list:
         file_path = os.path.join(CURRENT_SOURCE_FILES_PATH, uploaded_file)
         table_name = file_path.split('/')[-1].split('-')[0]

src/server/pipeline/match_data.py

@@ -72,11 +72,13 @@ def start(connection, added_or_updated_rows, manual_matches_df):
             )
         ]
         #collect other linked ids from manual matches source
-        linked_ids = manual_matches_df[(manual_matches_df[row["source_type"]] == row["source_id"])]
-        ids = linked_ids.to_dict(orient="records")
-        for id_num, row_dict in enumerate(ids):
-            for column, value in row_dict.items():
-                row_matches = row_matches.append(pdp_contacts[(pdp_contacts["source_type"] == column) & (pdp_contacts["source_id"] == value)])
+        if manual_matches_df != None:
+            linked_ids = manual_matches_df[(manual_matches_df[row["source_type"]] == row["source_id"])]
+            ids = linked_ids.to_dict(orient="records")
+            for row_dict in enumerate(ids):
+                for column, value in row_dict.items():
+                    row_matches = row_matches.append(pdp_contacts[(pdp_contacts["source_type"] == column) & (pdp_contacts["source_id"] == value)])
+
 
         if row_matches.empty:  # new record, no matching rows
             max_matching_group += 1