Now don't have to have join column in reference df. Can remove input files without error message.

tools/helper_functions.py

@@ -50,6 +50,9 @@ def initial_data_load(in_file):
     output_message = ""
     results_df = pd.DataFrame()
     df = pd.DataFrame()
+
+    if not in_file:
+        return "No files provided.", gr.Dropdown(choices=[]), gr.Dropdown(choices=[]), df, results_df
     
     file_list = [string.name for string in in_file]
 

tools/matcher_funcs.py

@@ -355,10 +355,12 @@ def query_addressbase_api(in_api_key:str, Matcher:MatcherClass, query_type:str,
                 # Matcher.ref_df = Matcher.ref_df.loc[Matcher.ref_df["LOCAL_CUSTODIAN_CODE"] != 7655,:] 
 
         if save_file:
-            final_api_output_file_name = output_folder + api_ref_save_loc[:-5] + ".parquet"
+            final_api_output_file_name_pq = output_folder + api_ref_save_loc[:-5] + ".parquet"
+            final_api_output_file_name = output_folder + api_ref_save_loc[:-5] + ".csv"
             print("Saving reference file to: " + api_ref_save_loc[:-5] + ".parquet")
             Matcher.ref_df.to_parquet(output_folder + api_ref_save_loc + ".parquet", index=False) # Save checkpoint as well
-            Matcher.ref_df.to_parquet(final_api_output_file_name, index=False)
+            Matcher.ref_df.to_parquet(final_api_output_file_name_pq, index=False)
+            Matcher.ref_df.to_csv(final_api_output_file_name)
 
         if Matcher.ref_df.empty:
             print ("No reference data found with API")

tools/preparation.py

@@ -162,7 +162,7 @@ def create_full_address(df):
     
     return df["full_address"]
 
-def prepare_ref_address(ref_df, ref_address_cols, new_join_col = ['UPRN'], standard_cols = True):
+def prepare_ref_address(ref_df, ref_address_cols, new_join_col = [], standard_cols = True):
     
     if ('SaoText' in ref_df.columns) | ("Secondary_Name_LPI" in ref_df.columns): standard_cols = True
     else: standard_cols = False