Enhance name matching functionality in app.py and find_name_mismatches.py

- Updated the find_name_mismatches function to accept ownership and actual dictionaries, allowing for more comprehensive name matching.
- Modified app.py to handle the additional outputs from the updated function, ensuring proper session state management for ownership and actual data.
- Improved data integrity by copying ownership and actual dictionaries before modifications, maintaining original data consistency.

app.py

@@ -79,7 +79,7 @@ with tab1:
                 st.session_state['projections_df'] = projections.copy()
                 st.session_state['projections_df']['salary'] = (st.session_state['projections_df']['salary'].astype(str).str.replace(',', '').astype(float).astype(int))
                 # Run name matching only once when first loading the files
-                st.session_state['Contest'], st.session_state['projections_df'] = find_name_mismatches(st.session_state['Contest'], st.session_state['projections_df'])
+                st.session_state['Contest'], st.session_state['projections_df'], st.session_state['ownership_dict'], st.session_state['actual_dict'] = find_name_mismatches(st.session_state['Contest'], st.session_state['projections_df'], st.session_state['ownership_dict'], st.session_state['actual_dict'])
 
 with tab2:
     if st.button('Clear data', key='reset3'):

global_func/find_name_mismatches.py

@@ -4,10 +4,12 @@ import pandas as pd
 import time
 from fuzzywuzzy import process
 
-def find_name_mismatches(contest_df, projections_df):
+def find_name_mismatches(contest_df, projections_df, ownership_dict, fpts_dict):
     # Create a copy of the projections dataframe to avoid modifying the original
     projections_df = projections_df.copy()
     contest_df = contest_df.copy()
+    ownership_dict = ownership_dict.copy()
+    fpts_dict = fpts_dict.copy()
     
     name_columns = [col for col in contest_df.columns if not col in ['BaseName', 'EntryCount']]
     
@@ -79,12 +81,23 @@ def find_name_mismatches(contest_df, projections_df):
                     for col in name_columns:
                         contest_df[col] = contest_df[col].replace(contest_name, projection_name)
                 
+                if contest_name in ownership_dict:
+                    ownership_dict[projection_name] = ownership_dict.pop(contest_name)
+                if contest_name in fpts_dict:
+                    fpts_dict[projection_name] = fpts_dict.pop(contest_name)
+                        
                 # Process manual selections
                 for projection_name, selection in selections.items():
                     if selection != "None of these":
                         selected_name = selection.split(" (")[0]
                         for col in name_columns:
                             contest_df[col] = contest_df[col].replace(selected_name, projection_name)
+                        
+                        if selected_name in ownership_dict:
+                            ownership_dict[projection_name] = ownership_dict.pop(selected_name)
+                        if selected_name in fpts_dict:
+                            fpts_dict[projection_name] = fpts_dict.pop(selected_name)
+
                         st.success(f"Replaced '{selected_name}' with '{projection_name}'")
                 
                 st.success("All changes applied successfully!")
@@ -98,4 +111,10 @@ def find_name_mismatches(contest_df, projections_df):
         for projection_name, contest_name in auto_matches.items():
             for col in name_columns:
                 contest_df[col] = contest_df[col].replace(contest_name, projection_name)
-        return contest_df, projections_df
+            
+            if contest_name in ownership_dict:
+                ownership_dict[projection_name] = ownership_dict.pop(contest_name)
+            if contest_name in fpts_dict:
+                fpts_dict[projection_name] = fpts_dict.pop(contest_name)
+
+        return contest_df, projections_df, ownership_dict, fpts_dict