diff --git a/datasource/universal_fetcher.py b/datasource/universal_fetcher.py index 4cf799b..9c8aa4c 100644 --- a/datasource/universal_fetcher.py +++ b/datasource/universal_fetcher.py @@ -107,16 +107,30 @@ class UniversalDataFetcher: # 统一入口（自动路由） # ============================================================ + # 各资产类型支持的 adj 参数 + VALID_ADJ_BY_TYPE = { + AssetType.CHINA_INDEX: ['raw'], # 指数无复权 + AssetType.CHINA_ETF: ['raw', 'hfq'], # ETF 仅支持后复权 + AssetType.CHINA_STOCK: ['raw', 'qfq', 'hfq'], + AssetType.US_INDEX: ['raw'], # 指数无复权 + AssetType.US_STOCK: ['raw', 'qfq', 'hfq'], + AssetType.HK_INDEX: ['raw'], # 指数无复权 + AssetType.HK_STOCK: ['raw', 'qfq', 'hfq'], + AssetType.FUTURES: ['raw'], # 期货无复权 + AssetType.CRYPTO: ['raw'], # 加密货币无复权 + } + def fetch( self, code: str, start_date: str, end_date: str, + adj: str = 'raw', retry: int = 3, timeframe: str = '1d' ) -> Optional[pd.DataFrame]: """ - 统一数据获取入口 + 统一数据获取入口（支持 adj 参数） 自动识别资产类型并路由到对应方法 @@ -124,31 +138,60 @@ class UniversalDataFetcher: code: 标的代码 start_date: 开始日期 'YYYY-MM-DD' end_date: 结束日期 'YYYY-MM-DD' + adj: 复权类型 'raw'(原始) / 'qfq'(前复权) / 'hfq'(后复权)，默认 'raw' retry: 重试次数 timeframe: K线周期（仅加密货币需要，默认1d） Returns: DataFrame with columns: date, open, high, low, close, volume + adj='hfq' 时 A股 ETF 会额外返回 adj_factor, close_hfq + + 示例: + # 原始价格 + df = fetcher.fetch("000300.SH", "2020-01-01", "2024-12-31") + + # A股股票后复权 + df = fetcher.fetch("000001.SZ", "2020-01-01", "2024-12-31", adj='hfq') + + # 美股股票前复权 + df = fetcher.fetch("AAPL", "2020-01-01", "2024-12-31", adj='qfq') """ + # 校验 adj 参数 + if adj not in ['raw', 'qfq', 'hfq']: + raise ValueError(f"adj 参数必须是 'raw', 'qfq' 或 'hfq'，当前: {adj}") + asset_type = AssetTypeDetector.detect(code) + # 校验 adj 是否适用于该资产类型 + valid_adj = self.VALID_ADJ_BY_TYPE.get(asset_type, ['raw']) + if adj not in valid_adj: + raise ValueError( + f"adj='{adj}' 不适用于 {asset_type.value}，支持的类型: {valid_adj}" + ) + for attempt in range(retry): try: - # 路由到具体方法 + # 路由到具体方法（传递 adj 参数） if asset_type == AssetType.CHINA_INDEX: - return self._fetch_china_index(code, start_date, end_date) + return self._tushare.fetch(code, start_date, end_date, adj) elif asset_type == AssetType.CHINA_ETF: - return self._fetch_china_etf(code, start_date, end_date) + return self._tushare.fetch(code, start_date, end_date, adj) + elif asset_type == AssetType.CHINA_STOCK: + return self._tushare.fetch(code, start_date, end_date, adj) elif asset_type == AssetType.US_INDEX: - return self._fetch_us_index(code, start_date, end_date) + self._start_tunnel() + return self._yfinance.fetch(code, start_date, end_date, adj) elif asset_type == AssetType.US_STOCK: - return self._fetch_us_stock(code, start_date, end_date) + self._start_tunnel() + return self._yfinance.fetch(code, start_date, end_date, adj) elif asset_type == AssetType.HK_INDEX: - return self._fetch_hk_index(code, start_date, end_date) + self._start_tunnel() + return self._yfinance.fetch(code, start_date, end_date, adj) elif asset_type == AssetType.HK_STOCK: - return self._fetch_hk_stock(code, start_date, end_date) + self._start_tunnel() + return self._yfinance.fetch(code, start_date, end_date, adj) elif asset_type == AssetType.FUTURES: - return self._fetch_futures(code, start_date, end_date) + return self._fetch_futures(code, start_date, end_date, adj) elif asset_type == AssetType.CRYPTO: return self._fetch_crypto(code, start_date, end_date, timeframe) else: @@ -159,7 +202,7 @@ class UniversalDataFetcher: if attempt < retry - 1: time.sleep(2) else: - print(f"✗ 获取 {code} 失败 (尝试 {attempt+1}/{retry}): {e}") + print(f"✗ 获取 {code} adj={adj} 失败 (尝试 {attempt+1}/{retry}): {e}") return None return None @@ -359,7 +402,8 @@ class UniversalDataFetcher: self, code: str, start_date: str, - end_date: str + end_date: str, + adj: str = 'raw' ) -> Optional[pd.DataFrame]: """ 获取期货数据 @@ -367,11 +411,16 @@ class UniversalDataFetcher: 特点： - 中国期货(.SHF/.DCE/.CZC): Tushare - NYMEX(.NYM): YFinance + - 期货不支持复权（adj 只能为 'raw'） """ + # 期货不支持复权 + if adj != 'raw': + raise ValueError(f"期货不支持复权，adj='{adj}' 仅适用于股票/ETF") + if code.endswith('.NYM'): # NYMEX期货走YFinance self._start_tunnel() - return self._yfinance.fetch(code, start_date, end_date) + return self._yfinance.fetch(code, start_date, end_date, adj='raw') else: # 中国期货走Tushare return self._tushare.fetch_futures(code, start_date, end_date) @@ -493,12 +542,13 @@ class UniversalDataFetcher: self, code: str, start_date: str, - end_date: str + end_date: str, + adj: str = 'qfq' ) -> Optional[pd.DataFrame]: """ 获取美股复权价格 - 使用 YFinance auto_adjust=True + 使用 YFinance，支持前复权(qfq)和后复权(hfq) - 消除拆分(split)和分红(dividend)对价格的影响 - 适用于美股股票/ETF @@ -506,13 +556,98 @@ class UniversalDataFetcher: code: 美股代码，如 'AAPL', 'TSLA', 'QQQ' start_date: 开始日期 'YYYY-MM-DD' end_date: 结束日期 'YYYY-MM-DD' + adj: 复权类型 'qfq'(前复权) 或 'hfq'(后复权)，默认 'qfq' Returns: DataFrame with columns: date, open, high, low, close, volume (复权后) 示例: # 苹果复权价格（包含分红和拆分调整） - df = fetcher.fetch_us_adj("AAPL", "2020-01-01", "2024-12-31") + df = fetcher.fetch_us_adj("AAPL", "2020-01-01", "2024-12-31", adj='qfq') """ self._start_tunnel() - return self._yfinance.fetch_adj(code, start_date, end_date) \ No newline at end of file + return self._yfinance.fetch_adj(code, start_date, end_date, adj) + + def fetch_hk_adj( + self, + code: str, + start_date: str, + end_date: str, + adj: str = 'qfq' + ) -> Optional[pd.DataFrame]: + """ + 获取港股股票复权价格 + + 使用 YFinance，支持前复权(qfq)和后复权(hfq) + + Args: + code: 港股代码，如 '00700.HK', '00941.HK' + start_date: 开始日期 'YYYY-MM-DD' + end_date: 结束日期 'YYYY-MM-DD' + adj: 复权类型 'qfq'(前复权) 或 'hfq'(后复权)，默认 'qfq' + + Returns: + DataFrame with columns: date, open, high, low, close, volume (复权后) + """ + self._start_tunnel() + return self._yfinance.fetch_adj(code, start_date, end_date, adj) + + def fetch_stock_adj( + self, + code: str, + start_date: str, + end_date: str, + adj: str = 'hfq' + ) -> Optional[pd.DataFrame]: + """ + 获取 A股股票复权价格 + + 使用 Tushare pro_bar 接口，支持前复权(qfq)和后复权(hfq) + + Args: + code: A股股票代码，如 '000001.SZ', '600000.SH' + start_date: 开始日期 'YYYY-MM-DD' + end_date: 结束日期 'YYYY-MM-DD' + adj: 复权类型 'qfq'(前复权) 或 'hfq'(后复权)，默认 'hfq' + + Returns: + DataFrame with columns: date, open, high, low, close, volume, adj_factor + """ + return self._tushare.fetch_stock_adj(code, start_date, end_date, adj) + + # ============================================================ + # 统一复权入口（简化版，直接调用 fetch） + # ============================================================ + + def fetch_with_adj( + self, + code: str, + start_date: str, + end_date: str, + adj: str = 'raw', + retry: int = 3 + ) -> Optional[pd.DataFrame]: + """ + 统一复权入口（简化版） + + 直接调用 fetch(adj=adj)，无需重复实现路由逻辑。 + + Args: + code: 标的代码 + start_date: 开始日期 'YYYY-MM-DD' + end_date: 结束日期 'YYYY-MM-DD' + adj: 复权类型，默认 'raw' + retry: 重试次数 + + Returns: + DataFrame，结构因资产类型和 adj 参数略有不同 + + 示例: + # A股股票后复权 + df = fetcher.fetch_with_adj("000001.SZ", "2020-01-01", "2024-12-31", adj='hfq') + + # 美股股票前复权 + df = fetcher.fetch_with_adj("AAPL", "2020-01-01", "2024-12-31", adj='qfq') + """ + # 直接调用 fetch，传递 adj 参数 + return self.fetch(code, start_date, end_date, adj, retry) \ No newline at end of file