fix: /api/article supports BDP + 24h (scrape any article for in-app reading)

main.py

@@ -13,112 +13,74 @@ HEADERS = {"User-Agent":"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/5
 BASE_BDP = "https://bongdaplus.vn"
 BASE_24H = "https://www.24h.com.vn"
 SPACE_URL = "https://bep40-vnews.hf.space"
-
 _DATA_DIR = "/data" if os.path.isdir("/data") else "."
 DB_PATH = os.path.join(_DATA_DIR, "db.json")
 SHARE_PATH = os.path.join(_DATA_DIR, "shares.json")
 _db_lock = threading.Lock()
 _share_lock = threading.Lock()
-
 def _load_json(path, default):
     try:
         with open(path,"r") as f: return json.load(f)
     except: return default
-
 def _save_json(path, data):
     try:
         with open(path,"w") as f: json.dump(data, f, ensure_ascii=False)
     except: pass
-
 def db_inc(category, aid):
     with _db_lock:
         db = _load_json(DB_PATH, {"views":{},"shares":{},"comments":{}})
         db[category][aid] = db[category].get(aid, 0) + 1
-        _save_json(DB_PATH, db)
-        return db[category][aid]
-
+        _save_json(DB_PATH, db); return db[category][aid]
 def db_add_comment(aid, name, text):
     with _db_lock:
         db = _load_json(DB_PATH, {"views":{},"shares":{},"comments":{}})
         if aid not in db["comments"]: db["comments"][aid] = []
         db["comments"][aid].append({"n":name,"t":text,"d":time.strftime("%d/%m %H:%M")})
-        _save_json(DB_PATH, db)
-        return db["comments"][aid]
-
+        _save_json(DB_PATH, db); return db["comments"][aid]
 def db_get(aid):
     db = _load_json(DB_PATH, {"views":{},"shares":{},"comments":{}})
     return {"views":db["views"].get(aid,0),"shares":db["shares"].get(aid,0),"comments":db["comments"].get(aid,[])}
-
 def _slug(title):
     s = unicodedata.normalize('NFKD', title.lower())
     s = ''.join(c for c in s if not unicodedata.combining(c))
-    s = re.sub(r'[^a-z0-9\s-]','',s)
-    s = re.sub(r'[\s]+','-',s.strip())
+    s = re.sub(r'[^a-z0-9\s-]','',s); s = re.sub(r'[\s]+','-',s.strip())
     return re.sub(r'-+','-',s).strip('-')[:60]
-
-def _hash(url):
-    return hashlib.md5(url.encode()).hexdigest()[:12]
-
+def _hash(url): return hashlib.md5(url.encode()).hexdigest()[:12]
 def register_share(url, title, img=""):
-    h = _hash(url)
-    sl = _slug(title)
+    h = _hash(url); sl = _slug(title)
     with _share_lock:
-        db = _load_json(SHARE_PATH, {})
-        db[h] = {"url":url,"title":title,"img":img,"slug":sl}
-        _save_json(SHARE_PATH, db)
+        db = _load_json(SHARE_PATH, {}); db[h] = {"url":url,"title":title,"img":img,"slug":sl}; _save_json(SHARE_PATH, db)
     return sl, h
-
 def get_share(h):
-    db = _load_json(SHARE_PATH, {})
-    return db.get(h)
+    db = _load_json(SHARE_PATH, {}); return db.get(h)
 
 @app.post("/api/view")
-def api_view(aid:str=Body(...,embed=True)):
-    return JSONResponse({"views": db_inc("views", aid)})
-
+def api_view(aid:str=Body(...,embed=True)): return JSONResponse({"views": db_inc("views", aid)})
 @app.post("/api/share_inc")
-def api_share_inc(aid:str=Body(...,embed=True)):
-    return JSONResponse({"shares": db_inc("shares", aid)})
-
+def api_share_inc(aid:str=Body(...,embed=True)): return JSONResponse({"shares": db_inc("shares", aid)})
 @app.post("/api/comment")
-def api_comment(aid:str=Body(...), name:str=Body(...), text:str=Body(...)):
-    return JSONResponse({"comments": db_add_comment(aid, name, text)})
-
+def api_comment(aid:str=Body(...), name:str=Body(...), text:str=Body(...)): return JSONResponse({"comments": db_add_comment(aid, name, text)})
 @app.get("/api/stats/{aid}")
-def api_stats(aid:str):
-    return JSONResponse(db_get(aid))
-
+def api_stats(aid:str): return JSONResponse(db_get(aid))
 @app.post("/api/register_share")
 def api_register_share(url:str=Body(...), title:str=Body(...), img:str=Body("")):
-    sl, h = register_share(url, title, img)
-    db_inc("shares", _hash(url))
+    sl, h = register_share(url, title, img); db_inc("shares", _hash(url))
     return JSONResponse({"slug":sl,"hash":h,"share_url":f"{SPACE_URL}/#/{sl}/{h}","social_url":f"{SPACE_URL}/s/{sl}/{h}"})
 
-# SEO share page: crawlers see og:image, users get redirected to VNEWS article reader
 @app.get("/s/{slug}/{hash_id}")
 async def share_page(slug:str, hash_id:str):
     info = get_share(hash_id)
-    if not info:
-        return HTMLResponse(f'<meta http-equiv="refresh" content="0;url={SPACE_URL}">')
-    title = info["title"]
-    url = info["url"]
-    img = info.get("img") or "https://s1.vnecdn.net/vnexpress/restruct/i/v9505/logo_default.jpg"
-    hash_url = f"{SPACE_URL}/#/{slug}/{hash_id}"
-    # Store article URL in localStorage so VNEWS opens it in its own reader (NOT external redirect)
-    escaped_url = url.replace("'", "\\'")
-    return HTMLResponse(f'''<!DOCTYPE html><html><head><meta charset="utf-8">
-<title>{title}</title>
-<meta property="og:title" content="{title}">
-<meta property="og:image" content="{img}">
-<meta property="og:url" content="{hash_url}">
-<meta property="og:type" content="article">
-<meta property="og:site_name" content="VNEWS">
-<meta name="twitter:card" content="summary_large_image">
-<meta name="twitter:title" content="{title}">
-<meta name="twitter:image" content="{img}">
+    if not info: return HTMLResponse(f'<meta http-equiv="refresh" content="0;url={SPACE_URL}">')
+    title=info["title"]; url=info["url"]; img=info.get("img") or "https://s1.vnecdn.net/vnexpress/restruct/i/v9505/logo_default.jpg"
+    escaped_url = url.replace("\\","\\\\").replace("'","\\'")
+    return HTMLResponse(f'''<!DOCTYPE html><html><head><meta charset="utf-8"><title>{title}</title>
+<meta property="og:title" content="{title}"><meta property="og:image" content="{img}">
+<meta property="og:url" content="{SPACE_URL}/#/{slug}/{hash_id}"><meta property="og:type" content="article">
+<meta property="og:site_name" content="VNEWS"><meta name="twitter:card" content="summary_large_image">
+<meta name="twitter:title" content="{title}"><meta name="twitter:image" content="{img}">
 </head><body><script>localStorage.setItem('pending_article','{escaped_url}');window.location.replace('{SPACE_URL}');</script></body></html>''')
 
-# Scrapers
+# ═══ Scrapers ═══
 _cache={}; _cache_ttl=300
 def _cached(key,fn):
     now=time.time()
@@ -126,7 +88,6 @@ def _cached(key,fn):
     try:data=fn()
     except:data=_cache.get(key,{}).get("d",[])
     _cache[key]={"d":data,"t":now};return data
-
 def _get(url,headers=None):
     h=headers or HEADERS;r=requests.get(url,headers=h,timeout=15);r.encoding="utf-8"
     return BeautifulSoup(r.text,"lxml")
@@ -164,6 +125,53 @@ def scrape_vne_article(url):
         return{"title":h1.get_text(strip=True) if h1 else "","summary":desc.get_text(strip=True) if desc else "","og_image":og_img,"body":body,"source":"vne","url":url}
     except:return None
 
+def scrape_bdp_article(url):
+    try:
+        soup=_get(url);h1=soup.select_one(".lead-title h1") or soup.select_one("h1")
+        og=soup.find("meta",property="og:image");og_img=og.get("content","") if og else ""
+        se=soup.select_one("div.summary");body=[]
+        cd=soup.select_one("div.content#postContent") or soup.select_one("div.content")
+        if cd:
+            for ch in cd.children:
+                if not hasattr(ch,'name') or not ch.name:continue
+                if ch.name=="p":
+                    im=ch.find("img")
+                    if im:s=im.get("src") or im.get("data-src","");(body.append({"type":"img","src":s,"alt":""}) if s else None)
+                    t=ch.get_text(strip=True);(body.append({"type":"p","text":t}) if t else None)
+                elif ch.name in("h2","h3"):body.append({"type":"heading","text":ch.get_text(strip=True)})
+        return{"title":h1.get_text(strip=True) if h1 else "","summary":se.get_text(strip=True) if se else "","og_image":og_img,"body":body,"source":"bdp","url":url}
+    except:return None
+
+def scrape_24h_article(url):
+    try:
+        r=requests.get(url,headers=HEADERS,timeout=15);r.encoding="utf-8"
+        soup=BeautifulSoup(r.text,"lxml");h1=soup.select_one("h1")
+        og=soup.find("meta",property="og:image");og_img=og.get("content","") if og else ""
+        desc=soup.find("meta",property="og:description");summary=desc.get("content","")[:200] if desc else ""
+        body=[]
+        for sel in["div.the-article-body","div.nws-mainContent","div.nwsCt","div.detail-content"]:
+            cd=soup.select_one(sel)
+            if cd:
+                for ch in cd.children:
+                    if not hasattr(ch,'name') or not ch.name:continue
+                    if ch.name=="p":t=ch.get_text(strip=True);(body.append({"type":"p","text":t}) if t else None)
+                    elif ch.name in("h2","h3"):body.append({"type":"heading","text":ch.get_text(strip=True)})
+                break
+        return{"title":h1.get_text(strip=True) if h1 else "","summary":summary,"og_image":og_img,"body":body,"source":"24h","url":url}
+    except:return None
+
+def scrape_bbc_article(url):
+    try:
+        r=requests.get(url,headers={"User-Agent":"Mozilla/5.0","Accept-Language":"en-GB"},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 ""
+        body=[]
+        for p in soup.select("[data-component='text-block'] p, article p, main p"):
+            t=p.get_text(strip=True)
+            if t and len(t)>20:body.append({"type":"p","text":t})
+        return{"title":h1.get_text(strip=True) if h1 else "","summary":"","og_image":og_img,"body":body,"source":"bbc","url":url}
+    except:return None
+
 def scrape_bdp(url):
     try:
         soup=_get(url);arts=[];seen=set()
@@ -179,7 +187,6 @@ def scrape_bdp(url):
                 seen.add(lk);arts.append({"title":t,"link":lk,"img":img,"source":"bdp","summary":""})
         return arts[:15]
     except:return[]
-
 def scrape_bdp_videos():
     try:
         soup=_get(f"{BASE_BDP}/video");arts=[];seen=set()
@@ -195,7 +202,6 @@ def scrape_bdp_videos():
             seen.add(href);arts.append({"title":title,"link":href,"img":img,"source":"bdp"})
         return arts[:20]
     except:return[]
-
 def scrape_24h_highlights():
     try:
         soup=_get(f"{BASE_24H}/video-highlight-c953.html");arts=[];seen=set()
@@ -218,7 +224,6 @@ def scrape_24h_highlights():
             with ThreadPoolExecutor(5) as ex:list(ex.map(_og,need))
         return arts[:30]
     except:return[]
-
 def scrape_24h_shorts():
     try:
         r=requests.get(f"{BASE_24H}/video/video-tin-tuc-cvd769.html",headers=HEADERS,timeout=15);r.encoding="utf-8"
@@ -250,7 +255,6 @@ def scrape_24h_shorts():
                 seen.add(href);arts.append({"title":title,"link":href,"img":img_src,"source":"24h-shorts"})
         return arts[:20]
     except:return[]
-
 def scrape_bbc_vietnamese():
     try:
         bbc_h={"User-Agent":"Mozilla/5.0","Accept-Language":"en-GB,en;q=0.9"}
@@ -273,19 +277,6 @@ def scrape_bbc_vietnamese():
             if len(arts)>=15:break
         return arts
     except:return[]
-
-def scrape_bbc_article(url):
-    try:
-        r=requests.get(url,headers={"User-Agent":"Mozilla/5.0","Accept-Language":"en-GB"},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 ""
-        body=[]
-        for p in soup.select("[data-component='text-block'] p, article p, main p"):
-            t=p.get_text(strip=True)
-            if t and len(t)>20:body.append({"type":"p","text":t})
-        return{"title":h1.get_text(strip=True) if h1 else "","summary":"","og_image":og_img,"body":body,"source":"bbc","url":url}
-    except:return None
-
 def extract_video_url(article_url):
     try:
         r=requests.get(article_url,headers={**HEADERS,"Referer":"https://www.24h.com.vn/"},timeout=10);r.encoding="utf-8"
@@ -321,7 +312,6 @@ def extract_video_url(article_url):
         if len(parts)>1:return{"src":parts[0],"poster":poster,"all_parts":parts}
         return{"src":primary[0],"poster":poster}
     except:return None
-
 def extract_bdp_video(url):
     try:
         m=re.search(r'-(\d{6,})\.html',url)
@@ -333,6 +323,7 @@ def extract_bdp_video(url):
         return{"src":source.get("src","") if source else "","poster":video.get("poster","")}
     except:return None
 
+# ═══ API Routes ═══
 VNE_CATS={"thoi-su":("https://vnexpress.net/thoi-su","Thời Sự"),"the-gioi":("https://vnexpress.net/the-gioi","Thế Giới"),"kinh-doanh":("https://vnexpress.net/kinh-doanh","Kinh Doanh"),"cong-nghe":("https://vnexpress.net/so-hoa","Công Nghệ"),"the-thao":("https://vnexpress.net/the-thao","Thể Thao"),"giai-tri":("https://vnexpress.net/giai-tri","Giải Trí"),"suc-khoe":("https://vnexpress.net/suc-khoe","Sức Khỏe")}
 BDP_CATS={"ngoai-hang-anh":("https://bongdaplus.vn/ngoai-hang-anh","Ngoại Hạng Anh"),"la-liga":("https://bongdaplus.vn/la-liga","La Liga"),"champions-league":("https://bongdaplus.vn/champions-league-cup-c1","Champions League"),"bong-da-vn":("https://bongdaplus.vn/bong-da-viet-nam","Bóng Đá VN")}
 
@@ -352,7 +343,6 @@ def api_homepage():
         except:pass
         return articles
     return JSONResponse(_cached("homepage",_f))
-
 @app.get("/api/category/{cat_id}")
 def api_category(cat_id:str):
     def _f():
@@ -361,14 +351,12 @@ def api_category(cat_id:str):
         if cat_id in BDP_CATS:arts=scrape_bdp(BDP_CATS[cat_id][0]);[a.update({"group":BDP_CATS[cat_id][1]}) for a in arts];return arts
         return[]
     return JSONResponse(_cached(f"cat_{cat_id}",_f))
-
 @app.get("/api/categories")
 def api_categories():
     cats=[{"id":"bbc","name":"BBC Tiếng Việt","source":"bbc"}]
     for k,(u,n) in VNE_CATS.items():cats.append({"id":k,"name":n,"source":"vne"})
     for k,(u,n) in BDP_CATS.items():cats.append({"id":k,"name":n,"source":"bdp"})
     return JSONResponse(cats)
-
 @app.get("/api/highlights")
 def api_highlights():return JSONResponse(_cached("highlights",scrape_24h_highlights))
 @app.get("/api/shorts")
@@ -381,15 +369,17 @@ def api_video_url(url:str=Query(...)):
     elif"bongdaplus.vn" in url:v=extract_bdp_video(url)
     else:v=None
     return JSONResponse(v if v else{"error":"not found"})
+
+# ★ FIX: /api/article now supports ALL sources (VNE, BBC, BDP, 24h)
 @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)
-    else:data=None
-    return JSONResponse(data if data else{"error":"not supported"})
+    data=None
+    if "vnexpress.net" in url: data=scrape_vne_article(url)
+    elif "bbc.com" in url: data=scrape_bbc_article(url)
+    elif "bongdaplus.vn" in url: data=scrape_bdp_article(url)
+    elif "24h.com.vn" in url: data=scrape_24h_article(url)
+    return JSONResponse(data if data else {"error":"not supported"})
 
 @app.get("/")
-async def index():
-    return FileResponse("/app/static/index.html")
-
+async def index(): return FileResponse("/app/static/index.html")
 app.mount("/static",StaticFiles(directory="/app/static"),name="static")