Update app.py

app.py

@@ -108,6 +108,44 @@ def get_songs_from_spotify(playlist_link, songs_df, previews_df):
     songs_df = pd.concat([songs_df,new_df])
     return songs_df, previews_df
 
+def get_songs_from_spotify(playlist_link, songs_df, previews_df):
+    if match := re.match(r"https://open.spotify.com/playlist/(.*)\?", playlist_link):
+        playlist_uri = match.groups()[0]
+    else:
+        raise ValueError("Expected format: https://open.spotify.com/playlist/...")
+    
+    # Get playlist track information
+    tracks = sp.playlist_tracks(playlist_uri)["items"]
+    
+    # For efficiency, collect new rows in a list before converting it to a DataFrame
+    new_rows = []
+    for track in tracks:
+        track_info = {
+            'artist': track['track']['artists'][0]['name'], 
+            'song_title': track['track']['name'], 
+            'preview_url': track['track']['preview_url'],
+            'elo_score': 1000
+        }
+        new_rows.append(track_info)
+    
+    new_df = pd.DataFrame(new_rows)
+    
+    # Merge original DataFrames to ensure we don't have duplicates across them
+    consolidated_df = pd.concat([songs_df[['artist', 'song_title']], previews_df[['artist', 'song_title']]]).drop_duplicates()
+    
+    # Identify new, unique songs not in the original datasets
+    unique_songs = pd.merge(new_df, consolidated_df, on=['artist', 'song_title'], how='left', indicator=True).query('_merge=="left_only"').drop('_merge', axis=1)
+    
+    # Update previews_df if there is a 'preview_url' column in the new unique songs DataFrame
+    if 'preview_url' in unique_songs:
+        previews_df = pd.concat([previews_df, unique_songs])
+    
+    # Drop the 'preview_url' for songs_df update
+    unique_songs_simplified = unique_songs.drop(['preview_url'], axis=1, errors='ignore')
+    songs_df = pd.concat([songs_df, unique_songs_simplified])
+    
+    return songs_df, previews_df
+
 def update_scores(winner, loser, k_factor=100):
   score_difference = int(k_factor/(winner/loser))
   winner += score_difference
@@ -261,6 +299,37 @@ def import_csv(file, songs_df, previews_df):
         songs_df = pd.concat([songs_df,new_df])
     return songs_df, previews_df
 
+def import_csv(file, songs_df, previews_df):
+    if file is not None:
+        new_df = pd.read_csv(file)
+        
+        # Ensure the input DataFrames have the correct structure, creating them if they don't exist
+        if 'elo_score' not in songs_df.columns or 'artist' not in songs_df.columns or 'song_title' not in songs_df.columns:
+            songs_df = pd.DataFrame(columns=['elo_score', 'artist', 'song_title'])
+        if 'preview_url' not in previews_df.columns:
+            previews_df = pd.DataFrame(columns=["elo_score", "artist", "song_title", "preview_url"])
+        
+        # Ensure the new DataFrame has the correct columns
+        if all(column in new_df.columns for column in ["elo_score", "artist", "song_title", "preview_url"]):
+            new_df = new_df[["elo_score", "artist", "song_title", "preview_url"]]
+        else:
+            raise ValueError("New CSV file does not contain required columns.")
+        
+        # Merge original DataFrames to ensure we don't have duplicates across them
+        consolidated_df = pd.concat([songs_df[['artist', 'song_title']], previews_df[['artist', 'song_title']]]).drop_duplicates()
+        
+        # Identify new, unique songs not in the original datasets
+        unique_songs = pd.merge(new_df, consolidated_df, on=['artist', 'song_title'], how='left', indicator=True).query('_merge=="left_only"').drop('_merge', axis=1)
+        
+        # Update previews_df with new unique songs including their preview URLs
+        previews_df = pd.concat([previews_df, unique_songs])
+        
+        # Separate the unique songs data without the 'preview_url' for songs_df
+        unique_songs_simplified = unique_songs.drop(['preview_url'], axis=1, errors='ignore')
+        songs_df = pd.concat([songs_df, unique_songs_simplified])
+        
+    return songs_df, previews_df
+    
 # Function to remove a song
 def remove_song(artist, song_title, songs_df):
     # Find and remove the song from the DataFrame