#!/usr/bin/env python3 # AKShare-vnPy数据适配器 - 赵云数据工程工具 # 将AKShare数据格式转换为vnPy兼容格式 import sys import os import json import pandas as pd import numpy as np from datetime import datetime, timedelta from typing import Dict, List, Optional, Union, Any import logging # 配置日志 logging.basicConfig( level=logging.INFO, format='%(asctime)s - %(name)s - %(levelname)s - %(message)s' ) logger = logging.getLogger(__name__) class AKShareDataAdapter: """AKShare到vnPy的数据适配器""" def __init__(self, config_path: str = None): """初始化适配器 Args: config_path: 配置文件路径 """ self.config = self._load_config(config_path) self.data_cache = {} # 尝试导入akshare（可选） try: import akshare as ak self.ak = ak self.akshare_available = True logger.info("AKShare已成功导入") except ImportError: self.ak = None self.akshare_available = False logger.warning("AKShare未安装，将使用模拟数据") def _load_config(self, config_path: str) -> Dict: """加载配置文件 Args: config_path: 配置文件路径 Returns: Dict: 配置信息 """ default_config = { 'data_sources': { 'stock': { 'provider': 'akshare', 'fields_mapping': { 'date': 'date', 'open': 'open', 'high': 'high', 'low': 'low', 'close': 'close', 'volume': 'volume', 'amount': 'amount', 'turnover': 'turnover' } }, 'index': { 'provider': 'akshare', 'fields_mapping': { 'date': 'date', 'open': 'open', 'high': 'high', 'low': 'low', 'close': 'close', 'volume': 'volume', 'amount': 'amount' } } }, 'vnpy_format': { 'datetime_format': '%Y-%m-%d', 'numeric_precision': 6, 'null_value': 0.0 }, 'cache_settings': { 'enabled': True, 'ttl_hours': 24, 'cache_dir': './data/running_data/cache' } } if config_path and os.path.exists(config_path): try: with open(config_path, 'r', encoding='utf-8') as f: user_config = json.load(f) default_config.update(user_config) except Exception as e: logger.error(f"加载配置文件失败 {config_path}: {e}") return default_config def get_stock_daily(self, symbol: str, start_date: str, end_date: str) -> pd.DataFrame: """获取股票日线数据 Args: symbol: 股票代码（如：000001） start_date: 开始日期（格式：YYYY-MM-DD） end_date: 结束日期（格式：YYYY-MM-DD） Returns: pd.DataFrame: 转换后的vnPy格式数据 """ logger.info(f"获取股票日线数据: {symbol} [{start_date} - {end_date}]") try: if self.akshare_available: # 使用akshare获取数据 df = self.ak.stock_zh_a_hist( symbol=symbol, period="daily", start_date=start_date, end_date=end_date, adjust="qfq" # 前复权 ) else: # 模拟数据 df = self._generate_mock_stock_data(symbol, start_date, end_date) # 转换数据格式 vnpy_df = self._convert_to_vnpy_format(df, 'stock') logger.info(f"股票数据获取成功: {symbol}, 数据量: {len(vnpy_df)}") return vnpy_df except Exception as e: logger.error(f"获取股票数据失败 {symbol}: {e}") # 返回空DataFrame return pd.DataFrame() def _generate_mock_stock_data(self, symbol: str, start_date: str, end_date: str) -> pd.DataFrame: """生成模拟股票数据（当akshare不可用时） Args: symbol: 股票代码 start_date: 开始日期 end_date: 结束日期 Returns: pd.DataFrame: 模拟数据 """ # 生成日期范围 dates = pd.date_range(start=start_date, end=end_date, freq='D') # 生成模拟数据 data = { '日期': dates, '开盘': np.random.uniform(10, 100, len(dates)), '收盘': np.random.uniform(10, 100, len(dates)), '最高': np.random.uniform(10, 100, len(dates)), '最低': np.random.uniform(10, 100, len(dates)), '成交量': np.random.uniform(10000, 1000000, len(dates)), '成交额': np.random.uniform(100000, 10000000, len(dates)), '振幅': np.random.uniform(0.1, 5.0, len(dates)), '涨跌幅': np.random.uniform(-5.0, 5.0, len(dates)), '涨跌额': np.random.uniform(-5.0, 5.0, len(dates)), '换手率': np.random.uniform(0.1, 10.0, len(dates)) } df = pd.DataFrame(data) return df def _convert_to_vnpy_format(self, df: pd.DataFrame, data_type: str) -> pd.DataFrame: """转换为vnPy格式 Args: df: 原始数据DataFrame data_type: 数据类型（stock, index等） Returns: pd.DataFrame: 转换后的数据 """ if df.empty: return df # 获取字段映射 mapping = self.config['data_sources'].get(data_type, {}).get('fields_mapping', {}) # 创建新的DataFrame vnpy_data = {} for vnpy_field, source_field in mapping.items(): if source_field in df.columns: vnpy_data[vnpy_field] = df[source_field] else: # 如果字段不存在，填充默认值 logger.warning(f"字段 {source_field} 不存在，使用默认值填充 {vnpy_field}") vnpy_data[vnpy_field] = np.nan vnpy_df = pd.DataFrame(vnpy_data) # 确保日期列为datetime类型 if 'date' in vnpy_df.columns: vnpy_df['date'] = pd.to_datetime(vnpy_df['date']) # 处理空值 null_value = self.config['vnpy_format'].get('null_value', 0.0) vnpy_df = vnpy_df.fillna(null_value) # 设置数值精度 numeric_precision = self.config['vnpy_format'].get('numeric_precision', 6) for col in vnpy_df.select_dtypes(include=[np.number]).columns: vnpy_df[col] = vnpy_df[col].round(numeric_precision) # 按日期排序 if 'date' in vnpy_df.columns: vnpy_df = vnpy_df.sort_values('date').reset_index(drop=True) return vnpy_df def get_index_daily(self, index_symbol: str, start_date: str, end_date: str) -> pd.DataFrame: """获取指数日线数据 Args: index_symbol: 指数代码（如：000001.SH） start_date: 开始日期 end_date: 结束日期 Returns: pd.DataFrame: 转换后的vnPy格式数据 """ logger.info(f"获取指数日线数据: {index_symbol} [{start_date} - {end_date}]") try: if self.akshare_available: # 使用akshare获取数据 df = self.ak.index_zh_a_hist( symbol=index_symbol, period="daily", start_date=start_date, end_date=end_date ) else: # 模拟数据 df = self._generate_mock_index_data(index_symbol, start_date, end_date) # 转换数据格式 vnpy_df = self._convert_to_vnpy_format(df, 'index') logger.info(f"指数数据获取成功: {index_symbol}, 数据量: {len(vnpy_df)}") return vnpy_df except Exception as e: logger.error(f"获取指数数据失败 {index_symbol}: {e}") return pd.DataFrame() def _generate_mock_index_data(self, index_symbol: str, start_date: str, end_date: str) -> pd.DataFrame: """生成模拟指数数据 Args: index_symbol: 指数代码 start_date: 开始日期 end_date: 结束日期 Returns: pd.DataFrame: 模拟数据 """ # 生成日期范围 dates = pd.date_range(start=start_date, end=end_date, freq='D') # 生成模拟数据 data = { '日期': dates, '开盘': np.random.uniform(3000, 4000, len(dates)), '收盘': np.random.uniform(3000, 4000, len(dates)), '最高': np.random.uniform(3000, 4000, len(dates)), '最低': np.random.uniform(3000, 4000, len(dates)), '成交量': np.random.uniform(1000000, 10000000, len(dates)), '成交额': np.random.uniform(10000000, 100000000, len(dates)) } df = pd.DataFrame(data) return df def export_to_vnpy_csv(self, df: pd.DataFrame, symbol: str, output_dir: str = None) -> str: """导出为vnPy CSV格式 Args: df: 数据DataFrame symbol: 标的代码 output_dir: 输出目录 Returns: str: 输出文件路径 """ if df.empty: logger.warning(f"数据为空，跳过导出: {symbol}") return "" if output_dir is None: output_dir = './data/running_data/vnpy_import' os.makedirs(output_dir, exist_ok=True) # 生成文件名 timestamp = datetime.now().strftime('%Y%m%d_%H%M%S') filename = f"vnpy_{symbol}_{timestamp}.csv" output_path = os.path.join(output_dir, filename) # 保存为CSV df.to_csv(output_path, index=False, encoding='utf-8-sig') logger.info(f"数据已导出为vnPy CSV格式: {output_path}") return output_path def export_to_vnpy_database(self, df: pd.DataFrame, symbol: str, table_name: str = None) -> bool: """导出到vnPy数据库格式（模拟） Args: df: 数据DataFrame symbol: 标的代码 table_name: 数据库表名 Returns: bool: 是否成功 """ if df.empty: logger.warning(f"数据为空，跳过数据库导出: {symbol}") return False # 这里可以集成vnPy的数据库接口 # 示例：保存为JSON文件 if table_name is None: table_name = f"vnpy_data_{symbol}" output_dir = './data/running_data/vnpy_database' os.makedirs(output_dir, exist_ok=True) output_path = os.path.join(output_dir, f"{table_name}.json") # 转换为字典格式 data_dict = { 'symbol': symbol, 'table_name': table_name, 'export_time': datetime.now().isoformat(), 'data_count': len(df), 'data': df.to_dict(orient='records') } with open(output_path, 'w', encoding='utf-8') as f: json.dump(data_dict, f, ensure_ascii=False, indent=2) logger.info(f"数据已导出为vnPy数据库格式: {output_path}") return True def main(): """示例使用""" adapter = AKShareDataAdapter() # 示例：获取股票数据 stock_data = adapter.get_stock_daily( symbol='000001', start_date='2024-01-01', end_date='2024-01-31' ) if not stock_data.empty: print(f"股票数据获取成功，数据量: {len(stock_data)}") print(stock_data.head()) # 导出为vnPy CSV格式 csv_path = adapter.export_to_vnpy_csv(stock_data, '000001') print(f"CSV导出路径: {csv_path}") else: print("股票数据获取失败") # 示例：获取指数数据 index_data = adapter.get_index_daily( index_symbol='000001.SH', start_date='2024-01-01', end_date='2024-01-31' ) if not index_data.empty: print(f"\n指数数据获取成功，数据量: {len(index_data)}") print(index_data.head()) if __name__ == "__main__": main()