feat: websocket kline fetch and enforce proxy routing

data/fetcher.py

@@ -1,15 +1,19 @@
 """从Binance获取加密货币OHLCV数据"""
 
 import os
+import json
 import time
 from datetime import datetime, timedelta
 from pathlib import Path
-from typing import Optional, Literal
+from typing import Optional, Literal, Any
+import logging
+from urllib.parse import urlparse
 
 import pandas as pd
 from binance.client import Client
 import pytz
 from dotenv import load_dotenv
+from websocket import create_connection
 
 from data.database import DatabaseManager
 from utils.send_msg import lark_notifier
@@ -18,6 +22,19 @@ from utils.send_msg import lark_notifier
 load_dotenv()
 
 
+logger = logging.getLogger("polymarket_trader")
+
+
+def _apply_global_proxy(proxy_url: str) -> None:
+    """统一设置代理环境变量，确保第三方网络库也走代理。"""
+    os.environ['HTTP_PROXY'] = proxy_url
+    os.environ['HTTPS_PROXY'] = proxy_url
+    os.environ['ALL_PROXY'] = proxy_url
+    os.environ['http_proxy'] = proxy_url
+    os.environ['https_proxy'] = proxy_url
+    os.environ['all_proxy'] = proxy_url
+
+
 class BinanceDataFetcher:
     """从Binance公开API获取和缓存OHLCV数据"""
 
@@ -26,6 +43,13 @@ class BinanceDataFetcher:
         '15m': Client.KLINE_INTERVAL_15MINUTE,
     }
 
+    INTERVAL_MS = {
+        '1h': 60 * 60 * 1000,
+        '15m': 15 * 60 * 1000,
+    }
+
+    WS_API_ENDPOINT = "wss://ws-api.binance.com/ws-api/v3"
+
     def __init__(self, cache_dir: str = "./data/cache"):
         """初始化Binance客户端和数据库管理器
         
@@ -33,33 +57,24 @@ class BinanceDataFetcher:
             cache_dir: 缓存/数据库目录
         """
         # 配置代理
-        proxies = None
-        use_proxy = os.getenv('USE_PROXY', 'false').lower() == 'true'
-
-        if use_proxy:
-            socks5_proxy = os.getenv('SOCKS5_PROXY', 'socks5://127.0.0.1:7890')
-            proxies = {
-                'http': socks5_proxy,
-                'https': socks5_proxy
-            }
-            print(f"✓ 使用SOCKS5代理: {socks5_proxy}")
+        socks5_proxy = os.getenv('SOCKS5_PROXY', 'socks5://127.0.0.1:7890')
+        _apply_global_proxy(socks5_proxy)
+        proxies = {
+            'http': socks5_proxy,
+            'https': socks5_proxy
+        }
+        logger.info(f"强制使用SOCKS5代理: {socks5_proxy}")
 
         try:
-            # 初始化Binance客户端（带代理）
-            if proxies:
-                self.client = Client(requests_params={'proxies': proxies})
-            else:
-                self.client = Client()
+            # 初始化Binance客户端（强制走代理）
+            self.client = Client(requests_params={'proxies': proxies})
 
             # 测试连接
             self.client.ping()
-            print("✓ 已成功连接到Binance API")
+            logger.info("已成功连接到Binance API")
         except Exception as e:
-            print(f"⚠ Binance API连接失败: {e}")
-            print("提示：")
-            print("  1. 检查代理设置 (.env文件)")
-            print("  2. 确保代理服务正在运行")
-            print("  3. 或者使用已缓存的数据")
+            logger.warning(f"Binance API连接失败: {e}")
+            logger.warning("提示: 1) 检查代理设置 2) 确保代理服务运行 3) 或者使用缓存数据")
             self.client = None
 
         # 初始化数据库管理器
@@ -112,13 +127,13 @@ class BinanceDataFetcher:
                 
                 # 简单判断覆盖范围 (容忍少量缺失)
                 if min_ts <= start_dt + timedelta(days=2) and max_ts >= end_dt - timedelta(days=2):
-                    print(f"从数据库加载数据：{symbol} {timeframe} ({len(df)} rows)")
+                    logger.info(f"从数据库加载数据: {symbol} {timeframe} ({len(df)} rows)")
                     return df
 
         # 2. 从API获取数据
         if self.client is None:
             if not df.empty:
-                 print(f"⚠ API不可用，返回数据库现有数据 ({len(df)} rows)")
+                 logger.warning(f"API不可用，返回数据库现有数据 ({len(df)} rows)")
                  return df
             
             raise ConnectionError(
@@ -128,32 +143,35 @@ class BinanceDataFetcher:
         # Binance使用USDT交易对
         pair = f"{symbol}USDT"
         interval = self.INTERVALS[timeframe]
+        interval_ms = self.INTERVAL_MS[timeframe]
 
-        print(f"获取 {symbol} {timeframe} 数据，时间范围：{start_dt.date()} 至 {end_dt.date()}...")
+        logger.info(f"获取 {symbol} {timeframe} 数据，时间范围: {start_dt.date()} 至 {end_dt.date()}")
 
         klines = []
         for attempt in range(max_retries):
             try:
-                klines = self.client.get_historical_klines(
+                # 使用 Binance WebSocket API 的 klines 方法获取历史K线
+                klines = self._get_historical_klines_via_ws(
                     symbol=pair,
                     interval=interval,
-                    start_str=str(int(start_dt.timestamp() * 1000)),
-                    end_str=str(int(end_dt.timestamp() * 1000))
+                    start_ts_ms=int(start_dt.timestamp() * 1000),
+                    end_ts_ms=int(end_dt.timestamp() * 1000),
+                    interval_ms=interval_ms
                 )
                 # 如果成功获取，跳出循环
                 break
             except Exception as e:
                 # 如果是最后一次尝试，且依然失败
                 if attempt == max_retries - 1:
-                    print(f"❌ 第 {attempt + 1} 次尝试失败，已达到最大重试次数。错误: {e}")
+                    logger.error(f"第 {attempt + 1} 次尝试失败，已达到最大重试次数。错误: {e}")
                     if not df.empty:
-                        print(f"⚠ API请求最终失败，返回数据库现有数据")
+                        logger.warning("WebSocket请求最终失败，返回数据库现有数据")
                         return df
                     raise e  # 如果没有缓存数据，抛出异常
                 
                 # 计算等待时间 (指数退避: 2s, 4s, 8s...)
                 wait_time = 2 * (2 ** attempt) 
-                print(f"⚠ 请求失败 (尝试 {attempt + 1}/{max_retries})，{wait_time}秒后重试... 错误: {e}")
+                logger.warning(f"请求失败 (尝试 {attempt + 1}/{max_retries})，{wait_time}秒后重试。错误: {e}")
                 time.sleep(wait_time)
 
         if not klines:
@@ -179,13 +197,94 @@ class BinanceDataFetcher:
         # 按时间排序
         new_df = new_df.sort_values('timestamp').reset_index(drop=True)
 
-        print(f"已获取 {len(new_df)} 行数据")
+        logger.info(f"已获取 {len(new_df)} 行数据")
 
         # 保存到数据库
         self.db_manager.save_ohlcv(new_df, symbol, timeframe)
         
         return new_df
 
+    def _get_historical_klines_via_ws(
+        self,
+        symbol: str,
+        interval: str,
+        start_ts_ms: int,
+        end_ts_ms: int,
+        interval_ms: int,
+        limit: int = 1000
+    ):
+        """通过 Binance WebSocket API 分页拉取历史K线。"""
+        if end_ts_ms <= start_ts_ms:
+            return []
+
+        all_klines = []
+        request_id = 1
+        current_start = start_ts_ms
+
+        # WebSocket API 单次最多返回 1000 条，这里按窗口分页拉取。
+        while current_start < end_ts_ms:
+            max_window_end = current_start + (limit * interval_ms)
+            current_end = min(end_ts_ms, max_window_end)
+
+            payload = {
+                "id": request_id,
+                "method": "klines",
+                "params": {
+                    "symbol": symbol,
+                    "interval": interval,
+                    "startTime": current_start,
+                    "endTime": current_end,
+                    "limit": limit
+                }
+            }
+
+            response_data = self._ws_request(payload)
+            batch = response_data.get("result", [])
+            if not batch:
+                break
+
+            all_klines.extend(batch)
+            last_open_time = int(batch[-1][0])
+            next_start = last_open_time + interval_ms
+            if next_start <= current_start:
+                break
+            current_start = next_start
+            request_id += 1
+
+        return all_klines
+
+    def _ws_request(self, payload: dict) -> dict:
+        """发送单次 WebSocket API 请求并返回 JSON 响应。"""
+        ws_kwargs: dict[str, Any] = {
+            "timeout": 30
+        }
+
+        socks5_proxy = os.getenv('SOCKS5_PROXY', 'socks5://127.0.0.1:7890')
+        parsed = urlparse(socks5_proxy)
+        if parsed.hostname and parsed.port:
+            scheme = (parsed.scheme or 'socks5').lower()
+            if scheme not in {'socks4', 'socks5', 'http'}:
+                scheme = 'socks5'
+            if scheme == 'socks5h':
+                scheme = 'socks5'
+            ws_kwargs.update({
+                "http_proxy_host": parsed.hostname,
+                "http_proxy_port": parsed.port,
+                "proxy_type": scheme
+            })
+
+        ws = create_connection(self.WS_API_ENDPOINT, **ws_kwargs)
+        try:
+            ws.send(json.dumps(payload))
+            raw = ws.recv()
+            data = json.loads(raw)
+            status = data.get("status", 200)
+            if status != 200:
+                raise RuntimeError(f"Binance WS API返回异常状态: {data}")
+            return data
+        finally:
+            ws.close()
+
     def get_latest_price(self, symbol: Literal['BTC', 'ETH', 'SOL']) -> float:
         """获取符号的当前价格
 
@@ -195,6 +294,9 @@ class BinanceDataFetcher:
         Returns:
             当前USDT价格
         """
+        if self.client is None:
+            raise ConnectionError("Binance客户端不可用，无法获取最新价格")
+
         pair = f"{symbol}USDT"
         ticker = self.client.get_symbol_ticker(symbol=pair)
         return float(ticker['price'])