Add sponsor recognition for all 42 news sources

refresh_task.py

@@ -55,6 +55,7 @@ def serialize_clusters(clusters):
                     "summary": a.analysis.summary if a.analysis else "",
                     "topics": a.analysis.topics if a.analysis else [],
                     "trust": f"{a.analysis.trustworthiness_score:.0%}" if a.analysis else "",
+                    "sponsor": a.analysis.sponsor if a.analysis else "",
                     "score": a.post.score,
                     "comments": a.post.num_comments,
                     "url": a.post.url,
@@ -84,6 +85,7 @@ def serialize_clusters(clusters):
                     "summary": a.analysis.summary if a.analysis else "",
                     "topics": a.analysis.topics if a.analysis else [],
                     "trust": round(a.analysis.trustworthiness_score, 2) if a.analysis else 0,
+                    "sponsor": a.analysis.sponsor if a.analysis else "",
                     "score": a.post.score,
                     "comments": a.post.num_comments,
                     "image": a.article.image_url if a.article else a.post.image_url,

src/analyzer.py

@@ -20,6 +20,56 @@ UNRELIABLE_DOMAINS = {
     "theonion.com": -0.25, "naturalnews.com": -0.30, "zerohedge.com": -0.15,
 }
 
+SPONSORS: dict[str, str] = {
+    "bbc.com": "BBC (publicly funded)",
+    "bbc.co.uk": "BBC (publicly funded)",
+    "nytimes.com": "The New York Times Company",
+    "npr.org": "NPR (publicly funded)",
+    "aljazeera.com": "Al Jazeera Media Network (Qatar-funded)",
+    "theguardian.com": "Guardian Media Group",
+    "statnews.com": "Boston Globe Media",
+    "sciencedaily.com": "ScienceDaily LLC",
+    "techcrunch.com": "Yahoo (Apollo Global Management)",
+    "wired.com": "Condé Nast (Advance Publications)",
+    "theverge.com": "Vox Media",
+    "arstechnica.com": "Condé Nast (Advance Publications)",
+    "krebsonsecurity.com": "Independent (Brian Krebs)",
+    "bleepingcomputer.com": "Independent",
+    "threatpost.com": "Independent",
+    "therecord.media": "Recorded Future",
+    "theonion.com": "Global Tetrahedron",
+    "thedailymash.co.uk": "Independent",
+    "babylonbee.com": "Independent",
+    "ign.com": "Ziff Davis",
+    "eurogamer.net": "ReedPop (Gamer Network)",
+    "pcgamer.com": "Future plc",
+    "kotaku.com": "G/O Media",
+    "gamespot.com": "Fandom",
+    "polygon.com": "Vox Media",
+    "variety.com": "Penske Media Corporation",
+    "hollywoodreporter.com": "Penske Media Corporation",
+    "deadline.com": "Penske Media Corporation",
+    "screenrant.com": "Valnet Inc.",
+    "arabnews.com": "Saudi Research & Publishing",
+    "middleeasteye.net": "Independent (London-based)",
+    "newarab.com": "Fikra Publishing",
+    "france24.com": "France Médias Monde (government-funded)",
+    "tunisiaonlinenews.com": "Independent",
+    "northafricapost.com": "Independent",
+    "africanews.com": "Independent",
+    "reuters.com": "Thomson Reuters",
+    "apnews.com": "Associated Press (non-profit)",
+    "wsj.com": "Dow Jones (News Corp)",
+    "economist.com": "The Economist Group",
+    "nature.com": "Springer Nature",
+    "science.org": "American Association for the Advancement of Science",
+    "washingtonpost.com": "The Washington Post (Jeff Bezos)",
+    "ft.com": "Financial Times (Nikkei)",
+    "bloomberg.com": "Bloomberg L.P. (Michael Bloomberg)",
+    "thehackernews.com": "Independent (THN)",
+    "ign.com": "Ziff Davis",
+}
+
 CLICKBAIT_PATTERNS = [
     r"you won'?t believe", r"shocked?", r"gobsmacked",
     r"this is what happens", r"number \d+ will",
@@ -199,6 +249,11 @@ class NewsAnalyzer:
         except Exception as exc:
             logger.warning("Model loading failed: %s — using rule-based", exc)
 
+    @staticmethod
+    def _detect_sponsor(article: Article) -> str:
+        domain = re.sub(r"^www\.", "", (article.source_domain or ""))
+        return SPONSORS.get(domain, "")
+
     def analyze(self, article: Article) -> Analysis:
         summary = self._summarise(article)
         topics = self._classify_topics(article, summary)
@@ -207,6 +262,7 @@ class NewsAnalyzer:
         leaning = self._detect_political_leaning(article.text or "")
 
         category = self._map_category(topics)
+        sponsor = self._detect_sponsor(article)
 
         return Analysis(
             summary=summary,
@@ -215,6 +271,7 @@ class NewsAnalyzer:
             is_opinion=is_opinion,
             political_leaning=leaning,
             category=category,
+            sponsor=sponsor,
         )
 
     def _summarise(self, article: Article) -> str:

src/models.py

@@ -34,6 +34,7 @@ class Analysis:
     is_opinion: bool
     political_leaning: str = "centrist"
     category: str = "General"
+    sponsor: str = ""
 
 
 @dataclass

templates/index.html

@@ -131,6 +131,7 @@ body {
 }
 .card-meta { display: flex; align-items: center; gap: 10px; font-size: 12px; color: #64748b; margin-bottom: 10px; flex-wrap: wrap; }
 .card-meta .source { color: #94a3b8; }
+.sponsor { font-size: 10px; color: #fbbf24; background: rgba(251, 191, 36, 0.1); padding: 2px 6px; border-radius: 4px; }
 .score-bar-wrap { height: 3px; background: #334155; border-radius: 3px; margin-bottom: 3px; }
 .score-bar-fill { height: 100%; border-radius: 3px; background: linear-gradient(90deg, #60a5fa, #a78bfa); transition: width .5s; }
 .score-label { font-size: 10px; color: #64748b; display: flex; justify-content: space-between; }
@@ -243,10 +244,11 @@ body {
                {% endfor %}
              </div>
              <div class="card-meta">
-               <span class="source">{{ c.articles[0].domain }}</span>
-               <span>🛡️ {{ c.trust }}</span>
-               {% if c.coverage > 1 %}<span>📰 {{ c.coverage }}</span>{% endif %}
-             </div>
+                <span class="source">{{ c.articles[0].domain }}</span>
+                <span>🛡️ {{ c.trust }}</span>
+                {% if c.coverage > 1 %}<span>📰 {{ c.coverage }}</span>{% endif %}
+                {% if c.articles[0].sponsor %}<span class="sponsor">🏢 {{ c.articles[0].sponsor }}</span>{% endif %}
+              </div>
              <div class="score-bar-wrap">
                <div class="score-bar-fill" style="width:{{ c.score * 100 }}%"></div>
              </div>

webapp.py

@@ -55,6 +55,7 @@ def _cluster_to_html_dict(c):
                 "summary": a.analysis.summary if a.analysis else "",
                 "topics": a.analysis.topics if a.analysis else [],
                 "trust": f"{a.analysis.trustworthiness_score:.0%}" if a.analysis else "",
+                "sponsor": a.analysis.sponsor if a.analysis else "",
                 "score": a.post.score,
                 "comments": a.post.num_comments,
                 "url": a.post.url,
@@ -89,6 +90,7 @@ def _cluster_to_api_dict(c):
                 "summary": a.analysis.summary if a.analysis else "",
                 "topics": a.analysis.topics if a.analysis else [],
                 "trust": round(a.analysis.trustworthiness_score, 2) if a.analysis else 0,
+                "sponsor": a.analysis.sponsor if a.analysis else "",
                 "score": a.post.score,
                 "comments": a.post.num_comments,
                 "image": a.article.image_url if a.article else a.post.image_url,