fix(news): robust gnews URL extraction + smarter lazy trigger

- Replace fragile entry.get("link") with _gnews_entry_url() helper that
  checks entry.link attribute then falls back to entry.links[].href,
  fixing cases where feedparser puts the URL in a non-standard location
- Lazy news re-fetch on bill detail now only triggers when the stored
  trend score confirms gnews_count > 0, preventing endless re-queuing
  for bills with genuinely no news coverage

Co-Authored-By: Jack Levy

backend/app/api/bills.py

@@ -110,8 +110,11 @@ async def get_bill(bill_id: str, db: AsyncSession = Depends(get_db)):
     if bill.trend_scores:
         detail.latest_trend = bill.trend_scores[0]
 
-    # Trigger a background news refresh if no articles are stored yet
-    if not bill.news_articles:
+    # Trigger a background news refresh if no articles are stored but trend
+    # data shows there are gnews results out there waiting to be fetched.
+    latest_trend = bill.trend_scores[0] if bill.trend_scores else None
+    has_gnews = latest_trend and (latest_trend.gnews_count or 0) > 0
+    if not bill.news_articles and has_gnews:
         try:
             from app.workers.news_fetcher import fetch_news_for_bill
             fetch_news_for_bill.delay(bill_id)

backend/app/services/news_service.py

@@ -89,6 +89,20 @@ def fetch_gnews_count(query: str, days: int = 30) -> int:
         return 0
 
 
+def _gnews_entry_url(entry) -> str:
+    """Extract the article URL from a feedparser Google News RSS entry."""
+    # Primary: entry.link attribute
+    link = getattr(entry, "link", None) or entry.get("link", "")
+    if link:
+        return link
+    # Fallback: scan entry.links list for rel=alternate
+    for lnk in getattr(entry, "links", []):
+        href = lnk.get("href", "")
+        if href:
+            return href
+    return ""
+
+
 def fetch_gnews_articles(query: str, days: int = 30) -> list[dict]:
     """Fetch articles from Google News RSS. No rate limit — unlimited source."""
     import time as time_mod
@@ -100,25 +114,28 @@ def fetch_gnews_articles(query: str, days: int = 30) -> list[dict]:
         articles = []
         for entry in feed.entries[:20]:
             pub_at = None
-            if entry.get("published_parsed"):
+            if getattr(entry, "published_parsed", None):
                 try:
                     pub_at = datetime.fromtimestamp(
                         time_mod.mktime(entry.published_parsed), tz=timezone.utc
                     ).isoformat()
                 except Exception:
                     pass
+            # Source: feedparser puts it in entry.source.title for Google News
             source = ""
-            if hasattr(entry, "source") and isinstance(entry.source, dict):
-                source = entry.source.get("title", "")
-            elif entry.get("tags"):
-                source = entry.tags[0].get("term", "") if entry.tags else ""
+            src = getattr(entry, "source", None)
+            if src:
+                source = getattr(src, "title", "") or src.get("title", "")
+            headline = entry.get("title", "") or getattr(entry, "title", "")
+            article_url = _gnews_entry_url(entry)
+            if article_url and headline:
                 articles.append({
                     "source": source or "Google News",
-                "headline": entry.get("title", ""),
-                "url": entry.get("link", ""),
+                    "headline": headline,
+                    "url": article_url,
                     "published_at": pub_at,
                 })
-        return [a for a in articles if a["url"] and a["headline"]]
+        return articles
     except Exception as e:
         logger.error(f"Google News RSS article fetch failed: {e}")
         return []