Fix: DanTri article parser uses <main> tag, proxy images in articles

main.py

@@ -345,16 +345,24 @@ def scrape_dantri_article(url):
         r=requests.get(url,headers=HEADERS,timeout=15);r.encoding="utf-8";soup=BeautifulSoup(r.text,"lxml")
         for tag in soup.find_all(["script","style","nav","footer","aside"]):tag.decompose()
         h1=soup.find("h1");og=soup.find("meta",property="og:image");og_img=og.get("content","") if og else ""
-        content=soup.select_one("div.singular-content") or soup.select_one("article");body=[]
+        if og_img and "cdnphoto.dantri" in og_img:og_img="/api/proxy/img?url="+quote(og_img,safe="")
+        content=soup.select_one("main") or soup.select_one("div.singular-content") or soup.select_one("article");body=[]
         if content:
             for el in content.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)
-        desc="";sapo=soup.select_one("h2.singular-sapo")
-        if sapo:desc=sapo.get_text(strip=True)
+                    if im:
+                        s=im.get("data-src") or im.get("src","")
+                        if s and"base64" not in s:
+                            if "cdnphoto.dantri" in s:s="/api/proxy/img?url="+quote(s,safe="")
+                            body.append({"type":"img","src":s})
+        desc="";sapo=soup.select_one("h2.singular-sapo") or soup.select_one("h2[class*=sapo]")
+        if not sapo:
+            og_desc=soup.find("meta",property="og:description")
+            if og_desc:desc=og_desc.get("content","")
+        else:desc=sapo.get_text(strip=True)
         return{"title":h1.get_text(strip=True) if h1 else "","summary":desc,"og_image":og_img,"body":body,"source":"dantri","url":url}
     except:return None
 def scrape_bbc_vietnamese():