修复「采集热点」长时间无响应：Google Trends 不可达/限流时快速回退缓存（连通性探测+逐词抓取时间预算+收紧超时），避免 24 个种子词干等 10 分钟

graph/sources/google_trends_source.py

@@ -119,20 +119,42 @@ def _retry(func, max_attempts=3, base_delay=3):
     raise last if last else RuntimeError("retry failed")
 
 
-def fetch_related(keywords, geo="US", timeframe="today 3-m"):
-    """逐关键词串行请求 related_queries（单关键词接口，避免 429）。"""
+def _probe_google(timeout: float = 3.0) -> bool:
+    """快速探测 trends.google.com 是否可达（连接+首字节）。
+
+    不可达时逐词 related_queries 会每个词超时 ~24s（10s 连接 ×2 重试 + 3s 节流），
+    24 个种子词要干等 10 分钟；探测失败直接跳过逐词抓取，回退缓存快速返回。
+    """
+    try:
+        resp = requests.get("https://trends.google.com/trending/rss?geo=US",
+                            timeout=timeout, headers={"User-Agent": "Mozilla/5.0"})
+        return resp.status_code < 500
+    except Exception:
+        return False
+
+
+def fetch_related(keywords, geo="US", timeframe="today 3-m", time_budget=40):
+    """逐关键词串行请求 related_queries（单关键词接口，避免 429）。
+
+    time_budget：整批逐词抓取的总时间预算（秒）。网络不稳/被限流时，每个词都可能
+    超时 ~10s+，24 个种子词会干等 10 分钟；超预算提前结束，回退缓存快速返回。
+    """
     merged = {}
+    t0 = time.time()
     for kw in keywords:
+        if time.time() - t0 > time_budget:
+            print(f"[GoogleTrends] {geo} 逐词抓取超过 {time_budget}s 预算，提前结束，回退缓存")
+            break
         time.sleep(3)  # 节流
 
         def _call(kw=kw):
-            # timeout=(connect, read)：pytrends 默认 connect=2s 太短，网络波动即全挂，放宽到 10/30s
-            pytrends = TrendReq(hl="en-US", tz=360, retries=2, backoff_factor=0.5, timeout=(10, 30))
+            # timeout=(connect, read)：pytrends 默认 connect=2s 太短，网络波动即全挂，放宽到 5/15s
+            pytrends = TrendReq(hl="en-US", tz=360, retries=1, backoff_factor=0.5, timeout=(5, 15))
             pytrends.build_payload(kw_list=[kw], timeframe=timeframe, geo=geo)
             return pytrends.related_queries()
 
         try:
-            data = _retry(_call, max_attempts=2, base_delay=1)
+            data = _retry(_call, max_attempts=1, base_delay=1)
         except Exception as e:  # noqa: BLE001
             print(f"[GoogleTrends] {geo} 种子「{kw}」抓取失败(跳过): {e}")
             continue
@@ -180,11 +202,17 @@ def _parse_traffic(desc):
     return None
 
 
-def fetch_trending(geo="US", limit=40):
+def fetch_trending(geo="US", limit=40, skip_network=False):
     key = f"trending|{geo}|{limit}"
     cached = _cache_get(key)
     if cached is not None:
         return cached
+    if skip_network:
+        latest = _cache_latest(key)
+        if latest is not None:
+            print(f"[GoogleTrends 趋势] {geo} 网络不可达，回退最新缓存({len(latest)}条)")
+            return latest
+        return []
     url = f"https://trends.google.com/trending/rss?geo={geo}"
     try:
         resp = requests.get(url, timeout=15, headers={"User-Agent": "Mozilla/5.0"})
@@ -213,12 +241,12 @@ def fetch_trending(geo="US", limit=40):
         return []
 
 
-def get_rows(keywords, geo="US", timeframe="today 3-m", source="gt_related"):
+def get_rows(keywords, geo="US", timeframe="today 3-m", source="gt_related", time_budget=40):
     key = f"{','.join(keywords)}|{geo}|{timeframe}|{source}|rows"
     cached = _cache_get(key)
     if cached is not None:
         return cached
-    raw = fetch_related(keywords, geo=geo, timeframe=timeframe)
+    raw = fetch_related(keywords, geo=geo, timeframe=timeframe, time_budget=time_budget)
     rows = parse_related(raw, geo, source=source)
     if raw:  # 有结果才写当日新缓存
         _cache_set(key, rows)
@@ -243,6 +271,17 @@ class GoogleTrendsSource(DataSource):
 
         rows: List[Dict[str, Any]] = []
 
+        # 0) 快速连通性探测：trends.google.com 不可达 → 跳过逐词抓取（每个词~24s 超时，
+        #    24 个种子词要干等 10 分钟），只回退 trending 缓存；无缓存则返回空，
+        #    由 fetch_node 回退 collected_keywords.json，让「采集热点」快速返回。
+        if not _probe_google():
+            print(f"[GoogleTrends] {country} trends.google.com 不可达，跳过逐词抓取，回退缓存")
+            if trending_cfg.get("enabled", True):
+                rows.extend(fetch_trending(geo=country,
+                                           limit=int(trending_cfg.get("limit", 40)),
+                                           skip_network=True))
+            return rows
+
         # 1) 国家实时趋势榜（主源）
         if trending_cfg.get("enabled", True):
             limit = int(trending_cfg.get("limit", 40))
@@ -252,12 +291,14 @@ class GoogleTrendsSource(DataSource):
         if style_cfg.get("enabled", True):
             seeds = style_cfg.get("seeds", []) or []
             if seeds:
-                rows.extend(get_rows(seeds, geo=country, timeframe=tf, source="gt_style"))
+                rows.extend(get_rows(seeds, geo=country, timeframe=tf, source="gt_style",
+                                     time_budget=40))
 
         # 3) 行业种子词
         if related_cfg.get("enabled", True):
             seeds = related_cfg.get("seed_keywords", []) or []
             if seeds:
-                rows.extend(get_rows(seeds, geo=country, timeframe=tf, source="gt_related"))
+                rows.extend(get_rows(seeds, geo=country, timeframe=tf, source="gt_related",
+                                     time_budget=40))
 
         return rows