Refactor name matching process in app.py: optimize the matching logic by processing names individually with extractOne for improved efficiency, while maintaining the handling of name-to-ID mapping and ensuring consistent debug output for better user feedback.

app.py

@@ -145,56 +145,47 @@ with tab1:
             
             # Update projections_df with any new matches
             st.session_state['projections_df'] = find_name_mismatches(st.session_state['portfolio'], st.session_state['projections_df'])
-            try:
-                name_id_map = dict(zip(
-                    st.session_state['csv_file']['Name'], 
-                    st.session_state['csv_file']['Name + ID']
-                ))
-                print("Using Name + ID mapping")
-            except:
-                name_id_map = dict(zip(
-                    st.session_state['csv_file']['Nickname'], 
-                    st.session_state['csv_file']['Id']
-                ))
-                print("Using Nickname + Id mapping")
-            
-            print(f"Number of names in name_id_map: {len(name_id_map)}")
-            print("Sample of name_id_map:", list(name_id_map.items())[:3])
-
-            # Get all names at once
-            names = projections['player_names'].tolist()
-            choices = list(name_id_map.keys())
-            
-            print(f"Number of names to match: {len(names)}")
-            print("Sample of names to match:", names[:3])
-            print("Sample of choices:", choices[:3])
-
-            # Process all names in one batch
-            matches = process.extract(
-                names,
-                choices,
-                scorer=fuzz.ratio,
-                score_cutoff=85,
-                limit=1  # Only get the best match
-            )
-            
-            print(f"Number of matches found: {len(matches)}")
-            print("Sample of matches:", matches[:3])
-
-            # Convert matches to dictionary - Fixed the match handling
-            match_dict = {}
-            for name, match_list in zip(names, matches):
-                if match_list and match_list[0][1] >= 85:  # Check if we have a match and it meets threshold
-                    match_dict[name] = name_id_map[match_list[0][0]]  # Use the matched name to get the ID
-                else:
-                    match_dict[name] = name  # Keep original name if no good match
-            
-            print(f"Number of entries in match_dict: {len(match_dict)}")
-            print("Sample of match_dict:", list(match_dict.items())[:3])
+            if 'export_dict' not in st.session_state and csv_file is not None:
+                try:
+                    name_id_map = dict(zip(
+                        st.session_state['csv_file']['Name'], 
+                        st.session_state['csv_file']['Name + ID']
+                    ))
+                    print("Using Name + ID mapping")
+                except:
+                    name_id_map = dict(zip(
+                        st.session_state['csv_file']['Nickname'], 
+                        st.session_state['csv_file']['Id']
+                    ))
+                    print("Using Nickname + Id mapping")
+                
+                # Get all names at once
+                names = projections['player_names'].tolist()
+                choices = list(name_id_map.keys())
+                
+                # Create a dictionary to store matches
+                match_dict = {}
+                
+                # Process each name individually but more efficiently
+                for name in names:
+                    # Use extractOne with score_cutoff for efficiency
+                    match = process.extractOne(
+                        name,
+                        choices,
+                        score_cutoff=85
+                    )
+                    
+                    if match:
+                        match_dict[name] = name_id_map[match[0]]
+                    else:
+                        match_dict[name] = name
+                
+                print(f"Number of entries in match_dict: {len(match_dict)}")
+                print("Sample of match_dict:", list(match_dict.items())[:3])
 
-            # Apply the matches
-            projections['upload_match'] = projections['player_names'].map(match_dict)
-            st.session_state['export_dict'] = match_dict  # Use match_dict directly
+                # Apply the matches
+                projections['upload_match'] = projections['player_names'].map(match_dict)
+                st.session_state['export_dict'] = match_dict
 
             st.write("Export Dictionary Contents:")
             st.write(st.session_state['export_dict'])