feat: inline article reader + BBC Vietnamese + fix shorts/highlights sliders

main.py

@@ -1,8 +1,7 @@
 """VNEWS - FastAPI backend with news scrapers and API."""
-import hashlib, re, time, threading
-from urllib.parse import quote as urlquote
+import hashlib, re, time
 from concurrent.futures import ThreadPoolExecutor, as_completed
-from fastapi import FastAPI, Request, Query
+from fastapi import FastAPI, Query
 from fastapi.responses import HTMLResponse, JSONResponse
 from fastapi.staticfiles import StaticFiles
 import requests
@@ -25,15 +24,11 @@ def _cached(key, fn):
     now = time.time()
     if key in _cache and now - _cache[key]["t"] < _cache_ttl:
         return _cache[key]["d"]
-    try:
-        data = fn()
-    except:
-        data = _cache.get(key, {}).get("d", [])
+    try: data = fn()
+    except: data = _cache.get(key, {}).get("d", [])
     _cache[key] = {"d": data, "t": now}
     return data
 
-def make_id(url): return hashlib.md5(url.encode()).hexdigest()[:12]
-
 def _get(url):
     r = requests.get(url, headers=HEADERS, timeout=15); r.encoding="utf-8"
     return BeautifulSoup(r.text, "lxml")
@@ -53,14 +48,43 @@ def scrape_vne(cat_url):
                 src = it.find("source")
                 if src: img = src.get("srcset","").split(",")[0].strip().split(" ")[0]
             desc = it.select_one("p.description")
-            arts.append({"title":t,"link":lk,"img":img,"summary":(desc.get_text(strip=True)[:120] if desc else ""),"source":"vne"})
+            arts.append({"title":t,"link":lk,"img":img,"summary":(desc.get_text(strip=True)[:150] if desc else ""),"source":"vne"})
         return arts
     except: return []
 
+def scrape_vne_article(url):
+    """Scrape full VnExpress article content."""
+    try:
+        soup = _get(url)
+        h1 = soup.select_one("h1.title-detail")
+        title = h1.get_text(strip=True) if h1 else ""
+        desc = soup.select_one("p.description")
+        summary = desc.get_text(strip=True) if desc else ""
+        og = soup.find("meta",property="og:image")
+        og_img = og.get("content","") if og else ""
+        cd = soup.select_one("article.fck_detail")
+        body = []
+        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)
+                    if t: body.append({"type":"p","text":t})
+                elif ch.name == "figure":
+                    im = ch.find("img")
+                    if im:
+                        s = im.get("data-src") or im.get("src","")
+                        cap = ch.find("figcaption")
+                        if s: body.append({"type":"img","src":s,"alt":cap.get_text(strip=True) if cap else ""})
+                elif ch.name in ("h2","h3"):
+                    body.append({"type":"heading","text":ch.get_text(strip=True)})
+        return {"title":title,"summary":summary,"og_image":og_img,"body":body,"source":"vne","url":url}
+    except: return None
+
 def scrape_bdp(url):
     try:
         soup = _get(url); arts = []; seen = set()
-        for sel,feat in [("div.news.fst",True),("div.sld-itm.news",True),("li.news",False)]:
+        for sel in ["div.news.fst","div.sld-itm.news","li.news"]:
             for it in soup.select(sel):
                 tag = it.find("a",class_="title") or it.find("a",href=True)
                 if not tag: continue
@@ -69,8 +93,7 @@ def scrape_bdp(url):
                 if lk and not lk.startswith("http"): lk=BASE_BDP+lk
                 if lk in seen: continue
                 im=it.find("img"); img=(im.get("data-src") or im.get("src","")) if im else ""
-                seen.add(lk)
-                arts.append({"title":t,"link":lk,"img":img,"source":"bdp"})
+                seen.add(lk); arts.append({"title":t,"link":lk,"img":img,"source":"bdp","summary":""})
         return arts[:15]
     except: return []
 
@@ -106,7 +129,7 @@ def scrape_24h_highlights():
                 img_src = img.get("data-original") or img.get("data-src") or img.get("src","")
                 if "base64" in img_src: img_src = ""
             arts.append({"title":title,"link":href,"img":img_src,"source":"24h"})
-        return arts[:25]
+        return arts[:30]
     except: return []
 
 def scrape_24h_shorts():
@@ -129,6 +152,51 @@ def scrape_24h_shorts():
         return arts[:20]
     except: return []
 
+def scrape_bbc_vietnamese():
+    """Scrape BBC Vietnamese news."""
+    try:
+        soup = _get("https://www.bbc.com/vietnamese")
+        arts = []; seen = set()
+        for a in soup.find_all("a", href=True):
+            href = a.get("href","")
+            if not href.startswith("/vietnamese/"): continue
+            if href.count("/") < 3: continue  # skip section links
+            full = "https://www.bbc.com" + href
+            if full in seen: continue
+            # Get title from text
+            title = a.get_text(strip=True)
+            if not title or len(title) < 15: continue
+            # Find nearby image
+            parent = a.parent
+            img = ""
+            if parent:
+                im = parent.find("img")
+                if im: img = im.get("src","") or im.get("data-src","")
+            seen.add(full)
+            arts.append({"title":title,"link":full,"img":img,"source":"bbc","summary":""})
+        return arts[:15]
+    except: return []
+
+def scrape_bbc_article(url):
+    """Scrape BBC Vietnamese article content."""
+    try:
+        soup = _get(url)
+        h1 = soup.find("h1")
+        title = h1.get_text(strip=True) if h1 else ""
+        og = soup.find("meta",property="og:image")
+        og_img = og.get("content","") if og else ""
+        # BBC article body
+        body = []
+        for block in soup.select("main [data-component='text-block'], main p"):
+            t = block.get_text(strip=True)
+            if t and len(t) > 20: body.append({"type":"p","text":t})
+        for img in soup.select("main img[src]"):
+            src = img.get("src","")
+            if src and "ichef" in src:
+                body.append({"type":"img","src":src,"alt":img.get("alt","")})
+        return {"title":title,"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)
@@ -160,110 +228,98 @@ def extract_bdp_video(url):
         return {"src":source.get("src","") if source else "","poster":video.get("poster","")}
     except: return None
 
-# ═══ API Routes ═══
-VNE_CATEGORIES = {
-    "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"),
+# ═══ API ═══
+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_CATEGORIES = {
-    "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"),
+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"),
 }
 
 @app.get("/api/homepage")
 def api_homepage():
-    def _fetch():
-        articles = []
-        sources = [("thoi-su",),("the-gioi",),("kinh-doanh",),("the-thao",),("giai-tri",)]
-        with ThreadPoolExecutor(5) as ex:
-            futs = {ex.submit(scrape_vne, VNE_CATEGORIES[s[0]][0]): VNE_CATEGORIES[s[0]][1] for s in sources}
+    def _f():
+        articles=[]
+        with ThreadPoolExecutor(6) as ex:
+            futs={ex.submit(scrape_vne,VNE_CATS[k][0]):VNE_CATS[k][1] for k in ["thoi-su","the-gioi","kinh-doanh","the-thao","giai-tri"]}
+            futs[ex.submit(scrape_bbc_vietnamese)]="BBC"
             for f in as_completed(futs):
                 try:
                     for a in f.result(): a["group"]=futs[f]; articles.append(a)
                 except: pass
-        # Add BDP
         try:
-            bdp = scrape_bdp(f"{BASE_BDP}/tin-moi")
-            for a in bdp[:6]: a["group"]="Bóng Đá"; articles.append(a)
+            for a in scrape_bdp(f"{BASE_BDP}/tin-moi")[:6]: a["group"]="Bóng Đá"; articles.append(a)
         except: pass
         return articles
-    return JSONResponse(_cached("homepage", _fetch))
+    return JSONResponse(_cached("homepage",_f))
 
 @app.get("/api/category/{cat_id}")
-def api_category(cat_id: str):
-    def _fetch():
-        if cat_id in VNE_CATEGORIES:
-            url, name = VNE_CATEGORIES[cat_id]
-            arts = scrape_vne(url)
-            for a in arts: a["group"] = name
+def api_category(cat_id:str):
+    def _f():
+        if cat_id=="bbc":return scrape_bbc_vietnamese()
+        if cat_id in VNE_CATS:
+            arts=scrape_vne(VNE_CATS[cat_id][0])
+            for a in arts:a["group"]=VNE_CATS[cat_id][1]
             return arts
-        elif cat_id in BDP_CATEGORIES:
-            url, name = BDP_CATEGORIES[cat_id]
-            arts = scrape_bdp(url)
-            for a in arts: a["group"] = name
+        if cat_id in BDP_CATS:
+            arts=scrape_bdp(BDP_CATS[cat_id][0])
+            for a in arts:a["group"]=BDP_CATS[cat_id][1]
             return arts
-        return []
-    return JSONResponse(_cached(f"cat_{cat_id}", _fetch))
+        return[]
+    return JSONResponse(_cached(f"cat_{cat_id}",_f))
 
 @app.get("/api/categories")
 def api_categories():
-    cats = []
-    for k,(u,n) in VNE_CATEGORIES.items():
-        cats.append({"id":k,"name":n,"source":"vne"})
-    for k,(u,n) in BDP_CATEGORIES.items():
-        cats.append({"id":k,"name":n,"source":"bdp"})
+    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))
+    return JSONResponse(_cached("highlights",scrape_24h_highlights))
 
 @app.get("/api/shorts")
 def api_shorts():
-    return JSONResponse(_cached("shorts", scrape_24h_shorts))
+    return JSONResponse(_cached("shorts",scrape_24h_shorts))
 
 @app.get("/api/bdp_videos")
 def api_bdp_videos():
-    return JSONResponse(_cached("bdp_videos", scrape_bdp_videos))
+    return JSONResponse(_cached("bdp_videos",scrape_bdp_videos))
 
 @app.get("/api/video_url")
-def api_video_url(url: str = Query(...)):
-    if "24h.com.vn" in url:
-        v = extract_video_url(url)
-    elif "bongdaplus.vn" in url:
-        v = extract_bdp_video(url)
-    else:
-        v = None
+def api_video_url(url:str=Query(...)):
+    if "24h.com.vn" in url: v=extract_video_url(url)
+    elif "bongdaplus.vn" in url: v=extract_bdp_video(url)
+    else: v=None
     return JSONResponse(v if v else {"error":"not found"})
 
-# ═══ Share page with og:image ═══
+@app.get("/api/article")
+def api_article(url:str=Query(...)):
+    """Read article content inline."""
+    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"})
+
 @app.get("/share/{path:path}")
-async def share_page(path: str, img: str = Query(default=""), title: str = Query(default="VNEWS")):
-    og_image = img or "https://s1.vnecdn.net/vnexpress/restruct/i/v9505/logo_default.jpg"
-    html = f"""<!DOCTYPE html><html><head>
-<meta charset="utf-8"><title>{title}</title>
-<meta property="og:title" content="{title}">
-<meta property="og:image" content="{og_image}">
-<meta property="og:url" content="{SPACE_URL}/share/{path}">
-<meta property="og:type" content="article">
-<meta property="og:site_name" content="VNEWS">
-<meta name="twitter:card" content="summary_large_image">
-<meta name="twitter:image" content="{og_image}">
-<meta http-equiv="refresh" content="0;url={SPACE_URL}">
-</head><body><p>Redirecting...</p></body></html>"""
+async def share_page(path:str,img:str=Query(default=""),title:str=Query(default="VNEWS")):
+    og_image=img or "https://s1.vnecdn.net/vnexpress/restruct/i/v9505/logo_default.jpg"
+    html=f'<!DOCTYPE html><html><head><meta charset="utf-8"><title>{title}</title><meta property="og:title" content="{title}"><meta property="og:image" content="{og_image}"><meta property="og:type" content="article"><meta name="twitter:card" content="summary_large_image"><meta name="twitter:image" content="{og_image}"><meta http-equiv="refresh" content="0;url={SPACE_URL}"></head><body>Redirecting...</body></html>'
     return HTMLResponse(content=html)
 
-# ═══ Serve frontend ═══
 @app.get("/")
 async def index():
-    with open("/app/static/index.html", "r", encoding="utf-8") as f:
+    with open("/app/static/index.html","r",encoding="utf-8") as f:
         return HTMLResponse(content=f.read())
 
-app.mount("/static", StaticFiles(directory="/app/static"), name="static")
+app.mount("/static",StaticFiles(directory="/app/static"),name="static")