Replace vnreview with genk.vn AI (articles readable in-app via .knc-content parser)

main.py

@@ -340,31 +340,30 @@ def scrape_dantri_congnghe():
             if len(arts)>=15:break
         return arts
     except:return[]
-def scrape_vnreview_ai():
-    """Scrape AI articles from vnreview.vn"""
+def scrape_genk_ai():
+    """Scrape AI articles from genk.vn - readable in-app"""
     try:
-        r=requests.get("https://vnreview.vn/chuyen-muc/ai-pho-thong.28/",headers=HEADERS,timeout=15)
+        r=requests.get("https://genk.vn/ai.chn",headers=HEADERS,timeout=15)
         if r.status_code!=200:return[]
         r.encoding="utf-8";soup=BeautifulSoup(r.text,"lxml")
-        import re as _re
         articles=[];seen=set()
-        for img in soup.find_all("img"):
-            src=img.get("src","") or img.get("data-src","")
-            if "cdn.vnreview.vn" not in src or "avatars" in src or "logo" in src:continue
-            parent=img.parent;link=None
-            for _ in range(5):
-                if parent is None:break
-                link=parent.find("a",href=_re.compile(r"/threads/"))
-                if link:break
-                parent=parent.parent
-            if not link:continue
-            href=link.get("href","")
-            if not href.startswith("http"):href="https://vnreview.vn"+href
-            if href in seen:continue
-            title=link.get("title","") or link.get_text(strip=True) or img.get("alt","")
-            if not title or len(title)<10:continue
-            seen.add(href);articles.append({"title":title,"link":href,"img":src,"source":"vnreview"})
-            if len(articles)>=12:break
+        for a in soup.find_all("a",href=True):
+            href=a.get("href","")
+            if not href.endswith(".chn") or href=="/ai.chn":continue
+            if href.startswith("/"):href="https://genk.vn"+href
+            if href in seen or "genk.vn" not in href:continue
+            title=a.get("title","") or a.get_text(strip=True)
+            if not title or len(title)<20:continue
+            container=a.parent
+            img_src=""
+            for _ in range(4):
+                if container is None:break
+                img=container.find("img",attrs={"data-src":True})
+                if not img:img=container.find("img",src=lambda s:s and "mediacdn" in s)
+                if img:img_src=img.get("data-src","") or img.get("src","");break
+                container=container.parent
+            seen.add(href);articles.append({"title":title,"link":href,"img":img_src,"source":"genk"})
+            if len(articles)>=15:break
         return articles
     except:return[]
 
@@ -455,13 +454,30 @@ def api_categories():
     return JSONResponse(cats)
 @app.get("/api/dantri_hot")
 def api_dantri_hot():return JSONResponse(_cached("dantri_hot",scrape_dantri_hot))
-@app.get("/api/vnreview_ai")
-def api_vnreview_ai():return JSONResponse(_cached("vnreview_ai",scrape_vnreview_ai,ttl=_cache_ttl))
+@app.get("/api/genk_ai")
+def api_genk_ai():return JSONResponse(_cached("genk_ai",scrape_genk_ai,ttl=_cache_ttl))
+def scrape_genk_article(url):
+    try:
+        r=requests.get(url,headers=HEADERS,timeout=15);r.encoding="utf-8";soup=BeautifulSoup(r.text,"lxml")
+        h1=soup.find("h1");og=soup.find("meta",property="og:image");og_img=og.get("content","") if og else ""
+        desc_el=soup.find("meta",property="og:description");desc=desc_el.get("content","") if desc_el else ""
+        cd=soup.select_one(".knc-content");body=[]
+        if cd:
+            for el in cd.find_all(["p","h2","h3","figure","img"],recursive=True):
+                if el.name=="p":t=el.get_text(strip=True);(body.append({"type":"p","text":t}) if t and len(t)>15 else None)
+                elif el.name in("h2","h3"):t=el.get_text(strip=True);(body.append({"type":"heading","text":t}) if t else None)
+                elif el.name in("figure","img"):
+                    im=el if el.name=="img" else el.find("img")
+                    if im:s=im.get("data-src") or im.get("src","");(body.append({"type":"img","src":s}) if s and"base64" not in s else None)
+        return{"title":h1.get_text(strip=True) if h1 else "","summary":desc,"og_image":og_img,"body":body,"source":"genk","url":url}
+    except:return None
+
 @app.get("/api/article")
 def api_article(url:str=Query(...)):
     if"vnexpress.net" in url:data=scrape_vne_article(url)
     elif"bbc.com" in url:data=scrape_bbc_article(url)
     elif"dantri.com.vn" in url:data=scrape_dantri_article(url)
+    elif"genk.vn" in url:data=scrape_genk_article(url)
     else:data=None
     return JSONResponse(data if data else{"error":"not supported"})
 @app.get("/v")