feat(strategy): rotation策略支持Flask API数据获取

- 新增 flask_api_source.py: Flask API远程数据源模块 - 修改 strategy.py: get_data() 支持通过Flask API获取数据使用方式: strategy.get_data(use_flask_api=True) # 通过部署服务获取 strategy.get_data(use_flask_api=False) # 本地HybridDataSource 配置项: flask_api_url: 可在config.yaml中指定API地址
2026-05-13 23:49:26 +08:00
parent 416f708d53
commit 0a9795febb
2 changed files with 418 additions and 3 deletions
--- a/datasource/flask_api_source.py
+++ b/datasource/flask_api_source.py
@@ -0,0 +1,293 @@
+"""
+Flask API 数据源
+
+通过部署后的 Flask API 服务获取 OHLCV 数据
+支持远程调用，无需本地 SSH 隧道
+"""
+
+import os
+import json
+import requests
+import pandas as pd
+from typing import Optional, Dict, List
+from datetime import datetime
+from pathlib import Path
+from dotenv import load_dotenv
+
+load_dotenv()
+
+
+class FlaskAPIDataSource:
+    """
+    Flask API 数据源
+    
+    通过 HTTP API 获取数据，无需本地配置 SSH 隧道
+    适用于远程调用或生产环境
+    
+    用法:
+        source = FlaskAPIDataSource(base_url="https://k3s.tokenpluse.xyz")
+        df = source.fetch("000300.SH", "2024-01-01", "2024-12-31")
+    """
+    
+    def __init__(
+        self,
+        base_url: str = None,
+        api_path: str = "/api/v1/ohlcv",
+        timeout: int = 120,
+        retries: int = 3
+    ):
+        """
+        初始化
+        
+        Args:
+            base_url: API 服务基础地址，默认从环境变量读取
+            api_path: API 路径
+            timeout: 请求超时时间（秒）
+            retries: 重试次数
+        """
+        self.base_url = base_url or os.getenv(
+            'FLASK_API_URL', 
+            'https://k3s.tokenpluse.xyz'
+        )
+        self.api_path = api_path
+        self.timeout = timeout
+        self.retries = retries
+        
+        # 确保 base_url 不以 / 结尾
+        self.base_url = self.base_url.rstrip('/')
+    
+    def fetch(
+        self,
+        code: str,
+        start_date: str,
+        end_date: str,
+        asset_type: str = None,
+        timeframe: str = '1d'
+    ) -> Optional[pd.DataFrame]:
+        """
+        获取单只标的 OHLCV 数据
+        
+        Args:
+            code: 标的代码
+            start_date: 开始日期 YYYY-MM-DD
+            end_date: 结束日期 YYYY-MM-DD
+            asset_type: 资产类型（可选，用于覆盖自动检测）
+            timeframe: K线周期（加密货币需要）
+        
+        Returns:
+            DataFrame with columns: date, open, high, low, close, volume
+        """
+        # 构建请求 URL
+        url = f"{self.base_url}{self.api_path}"
+        
+        # 构建请求参数
+        params = {
+            'code': code,
+            'start': start_date,
+            'end': end_date,
+        }
+        
+        # 加密货币需要 timeframe 参数
+        if asset_type == 'crypto' or code.upper() in ['BTC', 'ETH']:
+            params['timeframe'] = timeframe
+        
+        # 可选：强制指定 asset_type
+        if asset_type:
+            params['asset_type'] = asset_type
+        
+        for attempt in range(self.retries):
+            try:
+                response = requests.get(
+                    url, 
+                    params=params, 
+                    timeout=self.timeout
+                )
+                
+                if response.status_code != 200:
+                    if attempt < self.retries - 1:
+                        continue
+                    print(f"✗ API请求失败: {response.status_code} - {response.text[:100]}")
+                    return None
+                
+                data = response.json()
+                
+                # 检查错误
+                if 'error' in data:
+                    print(f"✗ API返回错误: {data['error']}")
+                    return None
+                
+                # 解析数据
+                records = data.get('data', [])
+                if not records:
+                    print(f"⚠ {code}: 无数据返回")
+                    return None
+                
+                # 转换为 DataFrame
+                df = pd.DataFrame(records)
+                
+                # 处理日期列
+                if 'date' in df.columns:
+                    df['date'] = pd.to_datetime(df['date'])
+                    df = df.set_index('date')
+                
+                # 确保列名标准化
+                df = df[['open', 'high', 'low', 'close', 'volume']]
+                
+                # 缓存 info 信息（如果有）
+                if 'info' in data:
+                    df.attrs['info'] = data['info']
+                
+                print(f"✓ {code}: {len(df)} 条数据 ({start_date} ~ {end_date})")
+                return df
+                
+            except requests.exceptions.Timeout:
+                if attempt < self.retries - 1:
+                    print(f"⚠ {code}: 请求超时，重试 {attempt + 2}/{self.retries}")
+                    continue
+                print(f"✗ {code}: 请求超时")
+                return None
+                
+            except requests.exceptions.RequestException as e:
+                if attempt < self.retries - 1:
+                    continue
+                print(f"✗ {code}: 请求异常 - {e}")
+                return None
+                
+            except json.JSONDecodeError as e:
+                print(f"✗ {code}: JSON解析失败 - {e}")
+                return None
+        
+        return None
+    
+    def fetch_batch(
+        self,
+        codes: List[str],
+        start_date: str,
+        end_date: str,
+        asset_types: Dict[str, str] = None
+    ) -> Dict[str, Optional[pd.DataFrame]]:
+        """
+        批量获取多只标的数据
+        
+        Args:
+            codes: 标的代码列表
+            start_date: 开始日期
+            end_date: 结束日期
+            asset_types: 资产类型映射 {code: asset_type}
+        
+        Returns:
+            {code: DataFrame}
+        """
+        results = {}
+        asset_types = asset_types or {}
+        
+        print(f"从 Flask API 获取 {len(codes)} 只标的...")
+        
+        for i, code in enumerate(codes, 1):
+            asset_type = asset_types.get(code)
+            df = self.fetch(code, start_date, end_date, asset_type)
+            results[code] = df
+            
+            # 显示进度
+            if i % 5 == 0 or i == len(codes):
+                success = sum(1 for v in results.values() if v is not None)
+                print(f"  进度: {i}/{len(codes)} (成功: {success})")
+        
+        return results
+    
+    def fetch_etf_nav(
+        self,
+        code: str,
+        start_date: str,
+        end_date: str
+    ) -> Optional[pd.DataFrame]:
+        """
+        获取 ETF 净值数据
+        
+        Args:
+            code: ETF代码
+            start_date: 开始日期
+            end_date: 结束日期
+        
+        Returns:
+            DataFrame with nav column
+        """
+        url = f"{self.base_url}/api/v1/etf/nav"
+        
+        params = {
+            'code': code,
+            'start': start_date,
+            'end': end_date
+        }
+        
+        try:
+            response = requests.get(url, params=params, timeout=self.timeout)
+            
+            if response.status_code != 200:
+                return None
+            
+            data = response.json()
+            
+            if 'error' in data or 'nav' not in data:
+                return None
+            
+            nav_data = data.get('nav', {})
+            records = nav_data.get('data', [])
+            
+            if not records:
+                return None
+            
+            df = pd.DataFrame(records)
+            if 'date' in df.columns:
+                df['date'] = pd.to_datetime(df['date'])
+                df = df.set_index('date')
+            
+            return df
+            
+        except Exception as e:
+            print(f"✗ {code} 净值获取失败: {e}")
+            return None
+    
+    def get_health(self) -> Dict:
+        """获取服务健康状态"""
+        # 先尝试 ohlcv 端点检查服务是否可用
+        url = f"{self.base_url}{self.api_path}"
+        params = {'code': '000300.SH', 'start': '2024-01-01', 'end': '2024-01-05'}
+            
+        try:
+            response = requests.get(url, params=params, timeout=self.timeout)
+            if response.status_code == 200:
+                data = response.json()
+                return {
+                    'status': 'healthy',
+                    'ssh_configured': True,
+                    'available': True
+                }
+            else:
+                return {'status': 'error', 'available': False}
+        except Exception as e:
+            return {'status': 'error', 'message': str(e), 'available': False}
+    
+    def get_service_info(self) -> Dict:
+        """获取服务信息"""
+        url = f"{self.base_url}/"
+        
+        try:
+            response = requests.get(url, timeout=10)
+            return response.json()
+        except Exception as e:
+            return {"error": str(e)}
+
+
+# 全局实例
+_flask_api_source: Optional[FlaskAPIDataSource] = None
+
+
+def get_flask_api_source(base_url: str = None) -> FlaskAPIDataSource:
+    """获取 Flask API 数据源实例"""
+    global _flask_api_source
+    
+    if _flask_api_source is None:
+        _flask_api_source = FlaskAPIDataSource(base_url=base_url)
+    
+    return _flask_api_source