temp: patch script (will be replaced)

patch_shorts.py

@@ -0,0 +1,159 @@
+"""
+Patch script: adds 24h news shorts to homepage.
+Run this to generate the patched app.py.
+"""
+import re
+
+# Read current app.py
+with open('app.py', 'r', encoding='utf-8') as f:
+    content = f.read()
+
+# 1. Add scrape_24h_news_video_list() after scrape_24h_video_list()
+new_scraper = '''
+def scrape_24h_news_video_list():
+    """Scrape 24h.com.vn video tin tuc (shorts/news) page."""
+    try:
+        url = f"{BASE_24H}/video/video-tin-tuc-cvd769.html"
+        r = requests.get(url, headers=HEADERS, timeout=15); r.encoding="utf-8"
+        soup = BeautifulSoup(r.text, "lxml")
+        articles, seen = [], set()
+        for art in soup.find_all("article"):
+            a = art.find("a", href=True)
+            if not a: continue
+            href = a.get("href","")
+            if not href.startswith("http"): href = BASE_24H + href
+            if href in seen: continue
+            img_tag = art.find("img")
+            title = ""
+            if img_tag: title = img_tag.get("alt","")
+            if not title: title = a.get("title","") or a.get_text(strip=True)
+            if not title or len(title) < 10: continue
+            img_src = None
+            if img_tag:
+                for attr in ["data-original","data-src","src"]:
+                    v = img_tag.get(attr,"")
+                    if v and "base64" not in v and len(v) > 20:
+                        img_src = v; break
+            seen.add(href)
+            articles.append({"title": title, "link": href, "img": img_src, "summary": "", "time": "",
+                "featured": False, "source": "24h", "group": "Shorts Tin T\\u1ee9c", "is_video": True})
+        return articles[:20]
+    except:
+        return []
+'''
+
+# Insert after scrape_24h_video_list function (find the end of it)
+marker = "def scrape_24h_article(url):"
+content = content.replace(marker, new_scraper + "\n" + marker)
+
+# 2. Modify fetch_homepage to also fetch news shorts
+old_fetch_homepage = '''def fetch_homepage():
+    all_articles=[]
+    h24_videos=[]
+    def _fetch(src,url,group):
+        arts=scrape_bdp_list(url) if src=="bdp" else scrape_vne_list(url)
+        for a in arts: a["group"]=group
+        return arts
+    def _fetch_24h():
+        nonlocal h24_videos
+        try: h24_videos=scrape_24h_video_list()[:15]
+        except: pass
+    with ThreadPoolExecutor(max_workers=6) as ex:
+        ex.submit(_fetch_24h)
+        futures={ex.submit(_fetch,s,u,g):g for s,u,g in HOMEPAGE_SOURCES}
+        for f in as_completed(futures):
+            try: all_articles.extend(f.result())
+            except: pass
+    return all_articles, h24_videos'''
+
+new_fetch_homepage = '''def fetch_homepage():
+    all_articles=[]
+    h24_videos=[]
+    h24_shorts=[]
+    def _fetch(src,url,group):
+        arts=scrape_bdp_list(url) if src=="bdp" else scrape_vne_list(url)
+        for a in arts: a["group"]=group
+        return arts
+    def _fetch_24h():
+        nonlocal h24_videos
+        try: h24_videos=scrape_24h_video_list()[:15]
+        except: pass
+    def _fetch_shorts():
+        nonlocal h24_shorts
+        try: h24_shorts=scrape_24h_news_video_list()[:15]
+        except: pass
+    with ThreadPoolExecutor(max_workers=7) as ex:
+        ex.submit(_fetch_24h)
+        ex.submit(_fetch_shorts)
+        futures={ex.submit(_fetch,s,u,g):g for s,u,g in HOMEPAGE_SOURCES}
+        for f in as_completed(futures):
+            try: all_articles.extend(f.result())
+            except: pass
+    return all_articles, h24_videos, h24_shorts'''
+
+content = content.replace(old_fetch_homepage, new_fetch_homepage)
+
+# 3. Update fetch_news_list to pass shorts
+old_home_call = '''    if src=="mix" and url_or_key=="home":
+        articles, h24_videos = fetch_homepage()
+        return render_homepage_html(articles, h24_videos)'''
+new_home_call = '''    if src=="mix" and url_or_key=="home":
+        articles, h24_videos, h24_shorts = fetch_homepage()
+        return render_homepage_html(articles, h24_videos, h24_shorts)'''
+content = content.replace(old_home_call, new_home_call)
+
+# 4. Add render_shorts_carousel_html function before render_video_carousel_html
+shorts_carousel_fn = '''
+def render_shorts_carousel_html(shorts):
+    """Carousel shorts tin tuc 24h - displayed at very top of homepage."""
+    vids_with_img = [v for v in shorts if v.get("img")]
+    if not vids_with_img: return ""
+    items = []
+    for i, v in enumerate(vids_with_img[:15]):
+        img = safe_url(v.get("img",""))
+        link = v.get("link","#")
+        title = v.get("title","")
+        aid = make_id(link); sl = slug(title)
+        click_js = f"window.bdpOpenTikTok(\\'{esc(link)}\\',\\'{aid}\\')"
+        items.append(f\\'\\'\\'<div class="vslide-item vslide-item-shorts" onclick="{click_js}">
+<div class="vslide-thumb vslide-thumb-shorts"><img src="{img}" alt="" class="bdp-lazy-img">
+<div class="vslide-play">\\u25b6</div><span class="vslide-badge vslide-badge-shorts">Shorts</span></div>
+<p class="vslide-title">{title}</p></div>\\'\\'\\'\\')
+    return f\\'\\'\\'<div class="vslide-wrap vslide-wrap-shorts">
+<div class="vslide-header"><span class="vslide-label">\\U0001f4f1 Shorts Tin T\\u1ee9c</span>
+<div class="vslide-nav"><button class="vslide-btn" onclick="window.bdpSlideScroll(-1,\\'vslide-shorts\\')">\\u25c0</button>
+<button class="vslide-btn" onclick="window.bdpSlideScroll(1,\\'vslide-shorts\\')">\\u25b6</button></div></div>
+<div class="vslide-track" id="vslide-shorts">{\\'\\'.join(items)}</div></div>\\'\\'\\'
+'''
+
+# Actually, the triple-quote escaping above is too complex. Let me write the function directly.
+# I'll insert it properly.
+
+# Let me take a different approach - insert raw code
+content = content.replace(
+    "def render_video_carousel_html(videos):",
+    '''def render_shorts_carousel_html(shorts):
+    """Carousel shorts tin tuc 24h - displayed at very top of homepage."""
+    vids_with_img = [v for v in shorts if v.get("img")]
+    if not vids_with_img: return ""
+    items = []
+    for i, v in enumerate(vids_with_img[:15]):
+        img = safe_url(v.get("img",""))
+        link = v.get("link","#")
+        title = v.get("title","")
+        aid = make_id(link); sl = slug(title)
+        click_js = f"window.bdpOpenTikTok('{esc(link)}','{aid}')"
+        items.append(f\'\'\'<div class="vslide-item vslide-item-shorts" onclick="{click_js}">
+<div class="vslide-thumb vslide-thumb-shorts"><img src="{img}" alt="" class="bdp-lazy-img">
+<div class="vslide-play">\u25b6</div><span class="vslide-badge vslide-badge-shorts">Shorts</span></div>
+<p class="vslide-title">{title}</p></div>\'\'\')
+    return f\'\'\'<div class="vslide-wrap vslide-wrap-shorts">
+<div class="vslide-header"><span class="vslide-label">\U0001f4f1 Shorts Tin T\u1ee9c</span>
+<div class="vslide-nav"><button class="vslide-btn" onclick="window.bdpSlideScroll(-1,\'vslide-shorts\')">\u25c0</button>
+<button class="vslide-btn" onclick="window.bdpSlideScroll(1,\'vslide-shorts\')">\u25b6</button></div></div>
+<div class="vslide-track" id="vslide-shorts">{\'\\'\'.join(items)}</div></div>\'\'\'
+
+def render_video_carousel_html(videos):'''
+)
+
+print("Patch approach too complex with escaping. Writing direct file instead.")