fix: deploy scripts/ files properly (correct directory)

scripts/capital_flow_collector.py

@@ -6,14 +6,39 @@
 
 API: push2his.eastmoney.com 个股资金流日线
 """
-import json, os, sys, time
+import json, os, sys, time, urllib.request
 from datetime import datetime
-from urllib.request import urlopen
+from urllib.request import urlopen, Request
+from concurrent.futures import ThreadPoolExecutor, as_completed
+from threading import Semaphore
 
 DATA_DIR = "/home/hmo/web-dashboard/data"
 DECISIONS_PATH = f"{DATA_DIR}/decisions.json"
 CACHE_PATH = f"{DATA_DIR}/capital_flow_cache.json"
 
+UA = "Mozilla/5.0"
+# 限速器：最多5个并发，每请求后强制间隔0.3s
+RATE_LIMIT = Semaphore(5)
+MIN_INTERVAL = 0.3
+_last_req = 0
+
+def _rate_limited_request(url):
+    """带速率限制的HTTP GET，用Semaphore控制并发数"""
+    global _last_req
+    with RATE_LIMIT:
+        elapsed = time.time() - _last_req
+        if elapsed < MIN_INTERVAL:
+            time.sleep(MIN_INTERVAL - elapsed)
+        proxy_handler = urllib.request.ProxyHandler({})
+        opener = urllib.request.build_opener(proxy_handler)
+        req = Request(url, headers={"User-Agent": UA, "Referer": "https://data.eastmoney.com/"})
+        try:
+            resp = opener.open(req, timeout=8)
+            _last_req = time.time()
+            return json.loads(resp.read().decode("utf-8"))
+        except Exception:
+            return None
+
 # eastmoney secid: 1=上海 0=深圳
 def secid(code):
     code = str(code).strip()
@@ -22,30 +47,28 @@ def secid(code):
     return f"0.{code}"
 
 def fetch_flow(code, days=5):
-    """拉取个股近N日资金流"""
+    """拉取个股近N日资金流（带限速+代理绕过）"""
     sid = secid(code)
     url = f"http://push2his.eastmoney.com/api/qt/stock/fflow/daykline/get?secid={sid}&fields1=f1,f2,f3,f7&fields2=f51,f52,f53,f54,f55,f56,f57&lmt={days}"
-    try:
-        resp = urlopen(url, timeout=5)
-        data = json.loads(resp.read().decode("utf-8"))
-        klines = data.get("data", {}).get("klines", [])
-        if not klines:
-            return None
-        result = []
-        for k in klines:
-            p = k.split(",")
-            if len(p) >= 7:
-                result.append({
-                    "date": p[0],
-                    "main_net": float(p[1]),     # 主力净流入(元)
-                    "super_large": float(p[2]),   # 超大单净流入(元)
-                    "large": float(p[3]),          # 大单净流入(元)
-                    "medium": float(p[4]),         # 中单净流入(元)
-                    "small": float(p[5]),          # 小单净流入(元)
-                })
-        return result
-    except Exception as e:
+    data = _rate_limited_request(url)
+    if not data:
         return None
+    klines = data.get("data", {}).get("klines", [])
+    if not klines:
+        return None
+    result = []
+    for k in klines:
+        p = k.split(",")
+        if len(p) >= 7:
+            result.append({
+                "date": p[0],
+                "main_net": float(p[1]),     # 主力净流入(元)
+                "super_large": float(p[2]),   # 超大单净流入(元)
+                "large": float(p[3]),          # 大单净流入(元)
+                "medium": float(p[4]),         # 中单净流入(元)
+                "small": float(p[5]),          # 小单净流入(元)
+            })
+    return result
 
 def fetch_flow_intraday(code):
     """拉取当日分时资金流（用于盘中判断）"""
@@ -119,27 +142,40 @@ def main():
                 codes.add(c)
     except:
         pass
-    
+
     all_flows = {}
-    
-    for code in sorted(codes):
+
+    # 并行抓取：ThreadPoolExecutor + 内置限速器（Semaphore 5 + 0.3s间隔）
+    code_list = sorted(codes)
+    if not code_list:
+        print("[capital_flow] 无代码需要采集")
+        return
+
+    def fetch_one(code):
         flow = fetch_flow(code, days=5)
         if flow:
             analysis = analyze_flow(flow)
-            all_flows[code] = {
+            return (code, {
                 "updated_at": datetime.now().strftime("%Y-%m-%d %H:%M"),
                 "flow": flow,
                 "analysis": analysis,
-            }
-        time.sleep(0.3)  # API限流
-    
+            })
+        return (code, None)
+
+    with ThreadPoolExecutor(max_workers=5) as pool:
+        futures = {pool.submit(fetch_one, c): c for c in code_list}
+        for f in as_completed(futures):
+            code, result = f.result()
+            if result:
+                all_flows[code] = result
+
     # 写缓存
     cache = {
         "updated_at": datetime.now().strftime("%Y-%m-%d %H:%M"),
         "stocks": all_flows,
     }
     json.dump(cache, open(CACHE_PATH, "w"), indent=2, ensure_ascii=False)
-    print(f"[capital_flow] {len(all_flows)}只更新完成")
+    print(f"[capital_flow] {len(all_flows)}/{len(code_list)}只更新完成")
 
 if __name__ == "__main__":
     main()