feat: Add Yahoo Finance as fallback data source for overseas environments

backend/requirements.txt

@@ -6,3 +6,4 @@ akshare>=1.18.25
 huggingface_hub>=0.26.0
 pydantic>=2.10.0
 python-dotenv>=1.0.0
+yfinance>=0.2.36

backend/scripts/sync_data.py

@@ -85,117 +85,107 @@ def get_stock_list() -> pd.DataFrame:
         raise
 
 
+import yfinance as yf
+
 def get_stock_daily(code: str, years: int = YEARS_OF_DATA) -> Optional[pd.DataFrame]:
     """
     获取单只股票的历史日线数据（后复权）
-    
-    Args:
-        code: 股票代码
-        years: 获取最近几年的数据
-    
-    Returns:
-        DataFrame with OHLCV data or None if failed
+    优先使用 AkShare，海外环境失败则尝试 Yahoo Finance
     """
+    # 1. 尝试 AkShare (国内源)
     try:
-        # 计算日期范围
         end_date = datetime.now().strftime('%Y%m%d')
         start_date = (datetime.now() - timedelta(days=years * 365)).strftime('%Y%m%d')
         
-        # 使用 AkShare 获取后复权数据
         df = ak.stock_zh_a_hist(
             symbol=code,
             period="daily",
             start_date=start_date,
             end_date=end_date,
-            adjust="hfq"  # 后复权
+            adjust="hfq"
         )
         
-        if df is None or df.empty:
-            return None
-        
-        # 标准化列名
-        df = df.rename(columns={
-            '日期': 'trade_date',
-            '开盘': 'open',
-            '最高': 'high',
-            '最低': 'low',
-            '收盘': 'close',
-            '成交量': 'volume',
-            '成交额': 'amount',
-            '涨跌幅': 'pct_chg',
-            '换手率': 'turnover_rate'
-        })
-        
-        # 添加股票代码
-        df['code'] = code
-        
-        # 转换日期格式
-        df['trade_date'] = pd.to_datetime(df['trade_date'])
-        
-        # 选择需要的列
-        columns = ['code', 'trade_date', 'open', 'high', 'low', 'close', 'volume', 'amount', 'pct_chg', 'turnover_rate']
-        df = df[[col for col in columns if col in df.columns]]
+        if df is not None and not df.empty:
+            df = df.rename(columns={
+                '日期': 'trade_date', '开盘': 'open', '最高': 'high', '最低': 'low',
+                '收盘': 'close', '成交量': 'volume', '成交额': 'amount',
+                '涨跌幅': 'pct_chg', '换手率': 'turnover_rate'
+            })
+            df['code'] = code
+            df['trade_date'] = pd.to_datetime(df['trade_date'])
+            return df[['code', 'trade_date', 'open', 'high', 'low', 'close', 'volume', 'amount', 'pct_chg', 'turnover_rate']]
+    except Exception as e:
+        logger.warning(f"AkShare failed for {code}, trying Yahoo Finance: {e}")
+
+    # 2. 尝试 Yahoo Finance (海外源)
+    try:
+        # 转换代码格式: 600519 -> 600519.SS, 000001 -> 000001.SZ
+        yf_code = f"{code}.SS" if code.startswith('6') or code.startswith('9') else f"{code}.SZ"
         
-        return df
+        ticker = yf.Ticker(yf_code)
+        # 获取最近3年数据
+        df_yf = ticker.history(period=f"{years}y", interval="1d", auto_adjust=True)
         
+        if df_yf is not None and not df_yf.empty:
+            df_yf = df_yf.reset_index()
+            df_yf = df_yf.rename(columns={
+                'Date': 'trade_date', 'Open': 'open', 'High': 'high', 'Low': 'low',
+                'Close': 'close', 'Volume': 'volume'
+            })
+            df_yf['code'] = code
+            df_yf['trade_date'] = pd.to_datetime(df_yf['trade_date']).dt.tz_localize(None)
+            df_yf['amount'] = df_yf['close'] * df_yf['volume'] # 估算成交额
+            df_yf['pct_chg'] = df_yf['close'].pct_change() * 100
+            df_yf['turnover_rate'] = 0 # Yahoo 不直接提供换手率
+            
+            return df_yf[['code', 'trade_date', 'open', 'high', 'low', 'close', 'volume', 'amount', 'pct_chg', 'turnover_rate']]
     except Exception as e:
-        logger.debug(f"Failed to fetch data for {code}: {e}")
-        return None
+        logger.error(f"Yahoo Finance also failed for {code}: {e}")
+        
+    return None
 
 
 def get_index_daily(code: str, years: int = YEARS_OF_DATA) -> Optional[pd.DataFrame]:
     """
     获取指数的历史日线数据
-    
-    Args:
-        code: 指数代码（如 '000300' 沪深300）
-        years: 获取最近几年的数据
-    
-    Returns:
-        DataFrame with OHLCV data or None if failed
+    优先使用 AkShare，海外环境失败则尝试 Yahoo Finance
     """
+    # 1. 尝试 AkShare
     try:
-        # 计算日期范围
         end_date = datetime.now().strftime('%Y%m%d')
         start_date = (datetime.now() - timedelta(days=years * 365)).strftime('%Y%m%d')
-        
-        # 使用 AkShare 获取指数数据
         df = ak.stock_zh_index_daily(symbol=f"sh{code}")
-        
-        if df is None or df.empty:
-            return None
-        
-        # 过滤日期范围
-        df['date'] = pd.to_datetime(df['date'])
-        df = df[(df['date'] >= start_date) & (df['date'] <= end_date)]
-        
-        # 标准化列名
-        df = df.rename(columns={
-            'date': 'trade_date',
-            'open': 'open',
-            'high': 'high',
-            'low': 'low',
-            'close': 'close',
-            'volume': 'volume',
-        })
-        
-        # 添加代码
-        df['code'] = code
-        
-        # 添加缺失的列
-        df['amount'] = 0
-        df['pct_chg'] = df['close'].pct_change() * 100
-        df['turnover_rate'] = 0
-        
-        # 选择需要的列
-        columns = ['code', 'trade_date', 'open', 'high', 'low', 'close', 'volume', 'amount', 'pct_chg', 'turnover_rate']
-        df = df[[col for col in columns if col in df.columns]]
-        
-        return df
-        
+        if df is not None and not df.empty:
+            df['date'] = pd.to_datetime(df['date'])
+            df = df[(df['date'] >= start_date) & (df['date'] <= end_date)]
+            df = df.rename(columns={'date': 'trade_date'})
+            df['code'] = code
+            df['amount'] = 0
+            df['pct_chg'] = df['close'].pct_change() * 100
+            df['turnover_rate'] = 0
+            return df[['code', 'trade_date', 'open', 'high', 'low', 'close', 'volume', 'amount', 'pct_chg', 'turnover_rate']]
     except Exception as e:
-        logger.debug(f"Failed to fetch index data for {code}: {e}")
-        return None
+        logger.warning(f"AkShare index failed for {code}, trying Yahoo Finance: {e}")
+
+    # 2. 尝试 Yahoo Finance
+    try:
+        # 沪深300在 Yahoo 的代码是 000300.SS
+        yf_code = f"{code}.SS"
+        ticker = yf.Ticker(yf_code)
+        df_yf = ticker.history(period=f"{years}y", interval="1d")
+        if df_yf is not None and not df_yf.empty:
+            df_yf = df_yf.reset_index()
+            df_yf = df_yf.rename(columns={'Date': 'trade_date', 'Open': 'open', 'High': 'high', 'Low': 'low', 'Close': 'close', 'Volume': 'volume'})
+            df_yf['code'] = code
+            df_yf['trade_date'] = pd.to_datetime(df_yf['trade_date']).dt.tz_localize(None)
+            df_yf['amount'] = 0
+            df_yf['pct_chg'] = df_yf['close'].pct_change() * 100
+            df_yf['turnover_rate'] = 0
+            return df_yf[['code', 'trade_date', 'open', 'high', 'low', 'close', 'volume', 'amount', 'pct_chg', 'turnover_rate']]
+    except Exception as e:
+        logger.error(f"Yahoo Finance index also failed for {code}: {e}")
+
+    return None
 
 
 def sync_stock_list(df: pd.DataFrame) -> None: