Streamlit中Session State问题：场上球员信息重复显示

问题原因 #

核心差异在于内容渲染方式不同：

• 实时比分用score_placeholder.info()实现，Streamlit占位符（st.empty()）每次调用会替换原有内容，因此保持单行显示。
• 球员信息直接调用st.markdown()，每次执行该代码都会在页面新增元素，导致阵容变动时不断追加新行。

解决方案 #

给球员信息创建专属占位符，用占位符的替换逻辑更新内容，而非新增元素。具体修改如下：

1. 添加球员信息占位符：在创建其他占位符的位置新增一行：

# 创建比分、球员、DataFrame的占位符
score_placeholder = st.empty()
players_placeholder = st.empty()  # 新增球员信息占位符
df_placeholder = st.empty()
refresh_placeholder = st.empty()

2. 修改球员信息更新逻辑：将直接调用st.markdown()改为用占位符替换内容，同时处理首次加载显示：

# 原有的球员信息判断逻辑内修改
if 'current_players' not in st.session_state:
    st.session_state.current_players = {}

# 先更新占位符内容，确保每次都显示最新阵容
home_display = f'<b>{st.session_state.home_team}</b>: {", ".join(home_players)}'
away_display = f'<b>{st.session_state.away_team}</b>: {", ".join(away_players)}'
players_placeholder.markdown(f"{home_display}<br>{away_display}", unsafe_allow_html=True)

# 仅在阵容变化时更新Session State
if st.session_state.current_players.get('home') != home_players or st.session_state.current_players.get('away') != away_players:
    st.session_state.current_players['home'] = home_players
    st.session_state.current_players['away'] = away_players

完整修改后的关键代码片段 #

if 'selected_game_id' in st.session_state:
    if st.button("View Game Plays"):
        
        game_over = is_game_over(st.session_state.selected_game_id)

        # 创建比分、球员、DataFrame的占位符
        score_placeholder = st.empty()
        players_placeholder = st.empty()  # 新增球员占位符
        df_placeholder = st.empty()
        refresh_placeholder = st.empty()

        def update_display():
            columns = [
                        'game_id', 'action_number', 'clock', 'timeActual', 'period', 'periodType',
                        'team_id', 'teamTricode', 'actionType', 'subType', 'descriptor',
                        'qualifiers', 'personId', 'x', 'y', 'possession', 'scoreHome', 'scoreAway', 'description'
                    ]
            all_plays = get_all_plays_from_db(st.session_state.selected_game_id)
            if not game_over:
                consumer.subscribe(['nba-plays'])
                start_time = time.time()
                
                while time.time() - start_time < 3:
                    json_obj = json.loads(boxscore.BoxScore(game_id=st.session_state.selected_game_id).get_json())
                    home_players = [player['name'] for player in json_obj['game']['homeTeam']['players'] if player['oncourt'] == '1']
                    away_players = [player['name'] for player in json_obj['game']['awayTeam']['players'] if player['oncourt'] == '1']

                    if 'current_players' not in st.session_state:
                        st.session_state.current_players = {}
                    
                    # 更新占位符内容，确保显示最新阵容
                    home_display = f'<b>{st.session_state.home_team}</b>: {", ".join(home_players)}'
                    away_display = f'<b>{st.session_state.away_team}</b>: {", ".join(away_players)}'
                    players_placeholder.markdown(f"{home_display}<br>{away_display}", unsafe_allow_html=True)

                    # 仅在阵容变化时更新Session State
                    if st.session_state.current_players.get('home') != home_players or st.session_state.current_players.get('away') != away_players:
                        st.session_state.current_players['home'] = home_players
                        st.session_state.current_players['away'] = away_players

                    # Kafka消费逻辑保持不变
                    msg = consumer.poll(0.1)
                    if msg is None:
                        continue
                    if msg.error():
                        if msg.error().code() == KafkaError._PARTITION_EOF:
                            continue
                        else:
                            st.error(f"Error: {msg.error()}")
                            break

                    play_data = json.loads(msg.value().decode('utf-8'))
                    if play_data['gameId'] == st.session_state.selected_game_id:
                        play_tuple = tuple(play_data.get(col, None) for col in columns)
                        all_plays.append(play_tuple)

                # 后续DataFrame和比分更新逻辑保持不变
                if all_plays:
                    # ... 原有代码 ...

说明 #

• 占位符players_placeholder会在每次update_display执行时，用最新球员信息替换原有内容，避免重复追加。
• 修改后无论阵容是否变动，页面都会显示当前最新场上球员，且不会产生重复行。

内容的提问来源于stack exchange，提问作者Chase P.