Spaces:
Running
Running
Restore to commit 8f37e0efb93a31689e91942abbc2edb6f7493bf5 - v18.11 (add /api/proxy/img endpoint)
Browse files- .rebuild +1 -1
- app_v2_entry.py +1047 -74
- static/app_v2.js +4 -9
- static/index_v2.html +2 -1
.rebuild
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
|
|
|
|
| 1 |
+
1783786956
|
app_v2_entry.py
CHANGED
|
@@ -508,13 +508,18 @@ def _st():return JSONResponse({'persistent':os.path.isdir('/data') and os.access
|
|
| 508 |
# ===== SHARE HELPERS: render content pages for shared links =====
|
| 509 |
def _render_slides_page(post, safe_title, safe_img, safe_url):
|
| 510 |
slides = post.get('slides', [])
|
|
|
|
| 511 |
if not safe_img and slides and slides[0].get('image'):
|
| 512 |
safe_img = slides[0].get('image', '')
|
|
|
|
| 513 |
description = _clean((post.get('text') or '')[:200]) or "Tin tức tóm tắt, AI rewrite, World Cup 2026"
|
|
|
|
|
|
|
| 514 |
if safe_url and safe_url != '/':
|
| 515 |
canonical_url = f"{SPACE}/s?url={quote(safe_url)}&title={quote(safe_title[:100])}"
|
| 516 |
else:
|
| 517 |
canonical_url = f"{SPACE}/s?post_id={post.get('id') or ''}"
|
|
|
|
| 518 |
h = f'''<!DOCTYPE html>
|
| 519 |
<html lang="vi">
|
| 520 |
<head>
|
|
@@ -547,11 +552,15 @@ def _render_slides_page(post, safe_title, safe_img, safe_url):
|
|
| 547 |
|
| 548 |
def _render_video_page(post, safe_title, safe_img, safe_url):
|
| 549 |
video_url = post.get('video', '')
|
|
|
|
| 550 |
description = _clean((post.get('text') or '')[:200]) or "Tin tức tóm tắt, AI rewrite, World Cup 2026"
|
|
|
|
|
|
|
| 551 |
if safe_url and safe_url != '/':
|
| 552 |
canonical_url = f"{SPACE}/s?url={quote(safe_url)}&title={quote(safe_title[:100])}"
|
| 553 |
else:
|
| 554 |
canonical_url = f"{SPACE}/s?post_id={post.get('id') or ''}"
|
|
|
|
| 555 |
h = f'''<!DOCTYPE html>
|
| 556 |
<html lang="vi">
|
| 557 |
<head>
|
|
@@ -583,9 +592,14 @@ video{{width:100%;height:100%;max-height:100vh;object-fit:contain;background:#00
|
|
| 583 |
|
| 584 |
@app.get('/s/{slug}')
|
| 585 |
async def _sh_slug(slug: str, request: Request, url: str = '', title: str = '', img: str = ''):
|
|
|
|
|
|
|
|
|
|
| 586 |
safe_title = _clean(title) if title else 'VNEWS - Tin tức'
|
| 587 |
safe_img = _clean(img) if img else ''
|
| 588 |
safe_url = _clean(url) if url else '/'
|
|
|
|
|
|
|
| 589 |
post = None
|
| 590 |
try:
|
| 591 |
if url:
|
|
@@ -597,22 +611,44 @@ async def _sh_slug(slug: str, request: Request, url: str = '', title: str = '',
|
|
| 597 |
safe_img = p.get('img', safe_img) or safe_img
|
| 598 |
safe_url = p.get('url', safe_url) or safe_url
|
| 599 |
break
|
|
|
|
| 600 |
if not post and url:
|
| 601 |
for p in posts:
|
| 602 |
if p.get('url') == url:
|
| 603 |
-
post = p
|
| 604 |
-
|
| 605 |
-
|
| 606 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 607 |
return HTMLResponse(f'''<!DOCTYPE html>
|
| 608 |
-
<html lang="vi">
|
| 609 |
-
<
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 610 |
<meta property="og:description" content="Tin tức tóm tắt, AI rewrite, World Cup 2026">
|
| 611 |
-
<meta property="og:url" content="{SPACE}/s/{slug}">
|
| 612 |
-
<
|
|
|
|
|
|
|
|
|
|
|
|
|
| 613 |
|
| 614 |
@app.get('/api/proxy/img')
|
| 615 |
def proxy_img(url: str = Query(default=""), max_size: int = Query(default=1200)):
|
|
|
|
| 616 |
from urllib.parse import unquote
|
| 617 |
safe_url = unquote(url)
|
| 618 |
import requests as _req
|
|
@@ -620,41 +656,79 @@ def proxy_img(url: str = Query(default=""), max_size: int = Query(default=1200))
|
|
| 620 |
r = _req.get(safe_url, timeout=10, headers={"User-Agent": "Mozilla/5.0"})
|
| 621 |
if r.status_code == 200 and r.content:
|
| 622 |
return Response(r.content, media_type=r.headers.get("content-type", "image/jpeg"))
|
| 623 |
-
except Exception as e:
|
|
|
|
| 624 |
return Response(status_code=502)
|
| 625 |
-
|
| 626 |
@app.get('/s')
|
| 627 |
async def _sh(url:str='',title:str='',img:str='',post_id:str=''):
|
| 628 |
safe_title = _clean(title) if title else 'VNEWS - Tin tức'
|
| 629 |
safe_img = _clean(img) if img else ''
|
| 630 |
safe_url = _clean(url) if url else '/'
|
|
|
|
|
|
|
| 631 |
post = None
|
| 632 |
try:
|
| 633 |
posts = _load_wall_posts()
|
| 634 |
if post_id:
|
| 635 |
for p in posts:
|
| 636 |
-
if p.get('id') == post_id:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 637 |
elif url:
|
|
|
|
| 638 |
for p in posts:
|
| 639 |
-
if p.get('url') == url and p.get('slides'):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 640 |
if not post:
|
|
|
|
| 641 |
for p in posts:
|
| 642 |
-
if p.get('url') == url:
|
| 643 |
-
|
| 644 |
-
|
| 645 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 646 |
if url and not safe_img:
|
| 647 |
try:
|
| 648 |
art = _scrape_article_fast(url)
|
| 649 |
-
if art and art.get('og_image'):
|
| 650 |
-
|
| 651 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 652 |
return HTMLResponse(f'''<!DOCTYPE html>
|
| 653 |
-
<html lang="vi">
|
| 654 |
-
<
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 655 |
<meta property="og:description" content="Tin tức tóm tắt, AI rewrite, World Cup 2026">
|
| 656 |
-
<meta property="og:url" content="{SPACE}/s?url={quote(safe_url)}">
|
| 657 |
-
<
|
|
|
|
|
|
|
| 658 |
|
| 659 |
from wc2026_scraper import scrape_summary,scrape_fixtures,scrape_standings,scrape_stats,scrape_wc_news,scrape_road_to_wc,get_wc2026_all,scrape_history,scrape_h2h,scrape_lineups,scrape_match_detail
|
| 660 |
|
|
@@ -664,38 +738,56 @@ _xlb_lock = threading.Lock()
|
|
| 664 |
def _xlb_scrape(path):
|
| 665 |
url = f"https://xemlaibongda.top/{path}"
|
| 666 |
r = req.get(url, headers={"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36"}, timeout=15, allow_redirects=True)
|
| 667 |
-
if r.status_code != 200:
|
|
|
|
| 668 |
soup = BeautifulSoup(r.text, 'lxml')
|
| 669 |
-
vids = []
|
|
|
|
| 670 |
for a in soup.select('a[href*="/video/"]'):
|
| 671 |
href = a.get('href', '')
|
| 672 |
-
if not href or href in seen:
|
|
|
|
| 673 |
seen.add(href)
|
| 674 |
-
if not href.startswith('http'):
|
|
|
|
| 675 |
img = a.select_one('img')
|
| 676 |
p = a.parent
|
| 677 |
for _ in range(4):
|
| 678 |
-
if img:
|
| 679 |
-
|
|
|
|
|
|
|
|
|
|
| 680 |
img_src = ''
|
| 681 |
-
if img:
|
| 682 |
-
|
| 683 |
-
|
|
|
|
|
|
|
|
|
|
| 684 |
title = ''
|
| 685 |
for sel in ['.title', 'h3', 'h2', '.name', '.post-title', '.entry-title', '.video-title']:
|
| 686 |
t = a.select_one(sel)
|
| 687 |
-
if t:
|
| 688 |
-
|
|
|
|
|
|
|
|
|
|
| 689 |
if not title:
|
| 690 |
img_alt = a.select_one('img')
|
| 691 |
-
if img_alt:
|
|
|
|
| 692 |
if not title:
|
| 693 |
parent = a.parent
|
| 694 |
-
if parent:
|
| 695 |
-
|
| 696 |
-
|
|
|
|
|
|
|
|
|
|
| 697 |
vids.append({"link": href, "img": img_src, "title": title})
|
| 698 |
-
if len(vids) >= 30:
|
|
|
|
| 699 |
return vids
|
| 700 |
|
| 701 |
@app.get('/api/proxy/xlb')
|
|
@@ -704,11 +796,13 @@ def proxy_xlb(path: str = Query(default="")):
|
|
| 704 |
cache_key = f"xlb:{path}"
|
| 705 |
with _xlb_lock:
|
| 706 |
cached = _xlb_cache.get(cache_key)
|
| 707 |
-
if cached and now - cached['t'] < 120:
|
|
|
|
| 708 |
try:
|
| 709 |
vids = _xlb_scrape(path)
|
| 710 |
result = {"videos": vids, "count": len(vids)}
|
| 711 |
-
with _xlb_lock:
|
|
|
|
| 712 |
return JSONResponse(result)
|
| 713 |
except Exception as e:
|
| 714 |
return JSONResponse({"videos": [], "count": 0, "error": str(e)}, status_code=500)
|
|
@@ -747,7 +841,7 @@ def _lj(p):
|
|
| 747 |
try:
|
| 748 |
if os.path.exists(p):return json.load(open(p,'r',encoding='utf-8'))
|
| 749 |
except:pass
|
| 750 |
-
return []
|
| 751 |
def _sj(p,d):
|
| 752 |
try:open(p+'.tmp','w',encoding='utf-8').write(json.dumps(d,ensure_ascii=False));os.replace(p+'.tmp',p)
|
| 753 |
except:pass
|
|
@@ -776,23 +870,28 @@ async def _pc(request:Request):
|
|
| 776 |
return JSONResponse({'comments':cms})
|
| 777 |
|
| 778 |
def _load_wall_posts():
|
| 779 |
-
with _wl_lock:
|
|
|
|
| 780 |
|
| 781 |
def _save_wall_posts(posts):
|
| 782 |
-
with _wl_lock:
|
|
|
|
| 783 |
|
| 784 |
@app.get('/api/wall')
|
| 785 |
def api_wall():
|
| 786 |
posts = _load_wall_posts()
|
| 787 |
-
if not posts:
|
|
|
|
| 788 |
return JSONResponse({"posts": posts})
|
| 789 |
|
| 790 |
@app.post('/api/wall')
|
| 791 |
async def api_wall_post(request: Request):
|
| 792 |
content_type = request.headers.get('content-type', '')
|
| 793 |
if 'multipart/form-data' in content_type:
|
| 794 |
-
try:
|
| 795 |
-
|
|
|
|
|
|
|
| 796 |
title = form.get('title', 'Video mới') or 'Video mới'
|
| 797 |
text = form.get('text', '') or ''
|
| 798 |
source = form.get('source', 'vtv_recorder') or 'vtv_recorder'
|
|
@@ -801,40 +900,80 @@ async def api_wall_post(request: Request):
|
|
| 801 |
video_url = None
|
| 802 |
if video_file and hasattr(video_file, 'filename') and video_file.filename:
|
| 803 |
fname = video_file.filename.lower()
|
| 804 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 805 |
video_filename = f"wall_{post_id}{ext}"
|
| 806 |
video_path = os.path.join(WALL_VIDEO_DIR, video_filename)
|
| 807 |
try:
|
| 808 |
content = await video_file.read()
|
| 809 |
-
if not content:
|
| 810 |
-
|
|
|
|
|
|
|
| 811 |
file_size_mb = len(content) / 1024 / 1024
|
| 812 |
-
if file_size_mb > 50:
|
|
|
|
|
|
|
| 813 |
video_url = f"/api/wall/video/{video_filename}"
|
| 814 |
-
except Exception as e:
|
| 815 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 816 |
posts = _load_wall_posts()
|
| 817 |
-
if not isinstance(posts, list):
|
| 818 |
-
|
|
|
|
|
|
|
|
|
|
| 819 |
return JSONResponse({"post": post, "ok": True})
|
| 820 |
-
try:
|
| 821 |
-
|
|
|
|
|
|
|
| 822 |
title = body.get('title', 'Bài mới') or 'Bài mới'
|
| 823 |
text = body.get('text', '') or ''
|
| 824 |
img = body.get('img', None)
|
| 825 |
source = body.get('source', 'user') or 'user'
|
| 826 |
post_id = str(uuid.uuid4())[:12]
|
| 827 |
-
post = {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 828 |
posts = _load_wall_posts()
|
| 829 |
-
if not isinstance(posts, list):
|
| 830 |
-
|
|
|
|
|
|
|
|
|
|
| 831 |
return JSONResponse({"post": post, "ok": True})
|
| 832 |
|
| 833 |
@app.get('/api/wall/video/{filename}')
|
| 834 |
def api_wall_video(filename: str):
|
| 835 |
-
if '..' in filename or '/' in filename:
|
|
|
|
| 836 |
video_path = os.path.join(WALL_VIDEO_DIR, filename)
|
| 837 |
-
if not os.path.exists(video_path):
|
|
|
|
| 838 |
ext = os.path.splitext(filename)[1].lower()
|
| 839 |
media_type = 'video/mp4' if ext == '.mp4' else 'video/webm'
|
| 840 |
return FileResponse(video_path, media_type=media_type)
|
|
@@ -842,14 +981,17 @@ def api_wall_video(filename: str):
|
|
| 842 |
@app.delete('/api/wall/{post_id}')
|
| 843 |
def api_wall_delete(post_id: str):
|
| 844 |
posts = _load_wall_posts()
|
| 845 |
-
if not isinstance(posts, list):
|
|
|
|
| 846 |
for i, p in enumerate(posts):
|
| 847 |
if p.get('id') == post_id:
|
| 848 |
if p.get('video'):
|
| 849 |
video_name = p['video'].split('/')[-1]
|
| 850 |
video_path = os.path.join(WALL_VIDEO_DIR, video_name)
|
| 851 |
-
if os.path.exists(video_path):
|
| 852 |
-
|
|
|
|
|
|
|
| 853 |
return JSONResponse({"ok": True})
|
| 854 |
return JSONResponse({"error": "Post not found"}, status_code=404)
|
| 855 |
|
|
@@ -859,6 +1001,7 @@ from urllib.parse import quote as _quote2
|
|
| 859 |
|
| 860 |
_UA_RW = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36', 'Accept-Language': 'vi-VN,vi;q=0.9'}
|
| 861 |
|
|
|
|
| 862 |
_UNIQUE_CHARS = {
|
| 863 |
'vietnamese': set('đăâêôơưàảãạáằẳẵặắầẩẫậấèẻẽẹéềễểệếìỉĩịíòỏõọóồổỗộốờởỡợớùủũụúừửữựứỳỷỹỵý'),
|
| 864 |
'spanish': set('ñáéíóúü¿¡'),
|
|
@@ -873,23 +1016,853 @@ _STOPWORDS = {
|
|
| 873 |
}
|
| 874 |
|
| 875 |
def detect_language(text):
|
| 876 |
-
|
| 877 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 878 |
vn_chars = len(text_chars & _UNIQUE_CHARS['vietnamese'])
|
| 879 |
-
if vn_chars >= 2:
|
|
|
|
|
|
|
|
|
|
| 880 |
es_chars = len(text_chars & _UNIQUE_CHARS['spanish'])
|
| 881 |
pt_chars = len(text_chars & _UNIQUE_CHARS['portuguese'])
|
|
|
|
|
|
|
| 882 |
words = set(re.findall(r'\b\w+\b', text_lower))
|
| 883 |
scores = {}
|
| 884 |
-
for lang, stops in _STOPWORDS.items():
|
|
|
|
|
|
|
|
|
|
| 885 |
pt_markers = {'não', 'pelo', 'pela', 'isso', 'há', 'estão', 'num', 'numa', 'tenho', 'posso', 'você', 'nós', 'eles', 'elas', 'também', 'muito', 'já', 'só', 'até', 'entre', 'depois', 'sem', 'mesmo', 'aos', 'serão'}
|
| 886 |
es_markers = {'pero', 'está', 'están', 'porque', 'también', 'hasta', 'donde', 'quien', 'fue', 'son', 'fueron', 'ese', 'eso', 'ante', 'ellos', 'ella', 'nosotros', 'usted', 'ustedes', 'tú', 'él', 'desde', 'todo', 'durante', 'todos', 'uno', 'les', 'ni', 'contra', 'otros', 'fueron'}
|
| 887 |
-
|
| 888 |
-
|
| 889 |
-
|
| 890 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 891 |
best = max(scores, key=scores.get)
|
| 892 |
return best if scores[best] > 0.05 else 'vietnamese'
|
| 893 |
|
|
|
|
| 894 |
_EMOTION_KEYWORDS = {
|
| 895 |
-
'happy': {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 508 |
# ===== SHARE HELPERS: render content pages for shared links =====
|
| 509 |
def _render_slides_page(post, safe_title, safe_img, safe_url):
|
| 510 |
slides = post.get('slides', [])
|
| 511 |
+
# Get image from post.img or first slide's image
|
| 512 |
if not safe_img and slides and slides[0].get('image'):
|
| 513 |
safe_img = slides[0].get('image', '')
|
| 514 |
+
# Use text for description if available
|
| 515 |
description = _clean((post.get('text') or '')[:200]) or "Tin tức tóm tắt, AI rewrite, World Cup 2026"
|
| 516 |
+
|
| 517 |
+
# Build canonical URL preserving original query format if url was provided
|
| 518 |
if safe_url and safe_url != '/':
|
| 519 |
canonical_url = f"{SPACE}/s?url={quote(safe_url)}&title={quote(safe_title[:100])}"
|
| 520 |
else:
|
| 521 |
canonical_url = f"{SPACE}/s?post_id={post.get('id') or ''}"
|
| 522 |
+
|
| 523 |
h = f'''<!DOCTYPE html>
|
| 524 |
<html lang="vi">
|
| 525 |
<head>
|
|
|
|
| 552 |
|
| 553 |
def _render_video_page(post, safe_title, safe_img, safe_url):
|
| 554 |
video_url = post.get('video', '')
|
| 555 |
+
# Use text for description if available
|
| 556 |
description = _clean((post.get('text') or '')[:200]) or "Tin tức tóm tắt, AI rewrite, World Cup 2026"
|
| 557 |
+
|
| 558 |
+
# Build canonical URL preserving original query format if url was provided
|
| 559 |
if safe_url and safe_url != '/':
|
| 560 |
canonical_url = f"{SPACE}/s?url={quote(safe_url)}&title={quote(safe_title[:100])}"
|
| 561 |
else:
|
| 562 |
canonical_url = f"{SPACE}/s?post_id={post.get('id') or ''}"
|
| 563 |
+
|
| 564 |
h = f'''<!DOCTYPE html>
|
| 565 |
<html lang="vi">
|
| 566 |
<head>
|
|
|
|
| 592 |
|
| 593 |
@app.get('/s/{slug}')
|
| 594 |
async def _sh_slug(slug: str, request: Request, url: str = '', title: str = '', img: str = ''):
|
| 595 |
+
"""SEO-friendly share endpoint with slug in URL path.
|
| 596 |
+
Shows slide content when url matches a wall post, otherwise redirects.
|
| 597 |
+
"""
|
| 598 |
safe_title = _clean(title) if title else 'VNEWS - Tin tức'
|
| 599 |
safe_img = _clean(img) if img else ''
|
| 600 |
safe_url = _clean(url) if url else '/'
|
| 601 |
+
|
| 602 |
+
# Try to find post by URL first (most reliable)
|
| 603 |
post = None
|
| 604 |
try:
|
| 605 |
if url:
|
|
|
|
| 611 |
safe_img = p.get('img', safe_img) or safe_img
|
| 612 |
safe_url = p.get('url', safe_url) or safe_url
|
| 613 |
break
|
| 614 |
+
# Fallback: any matching URL
|
| 615 |
if not post and url:
|
| 616 |
for p in posts:
|
| 617 |
if p.get('url') == url:
|
| 618 |
+
post = p
|
| 619 |
+
safe_title = p.get('title', safe_title) or safe_title
|
| 620 |
+
safe_img = p.get('img', safe_img) or safe_img
|
| 621 |
+
safe_url = p.get('url', safe_url) or safe_url
|
| 622 |
+
break
|
| 623 |
+
except:
|
| 624 |
+
pass
|
| 625 |
+
|
| 626 |
+
if post and post.get('slides'):
|
| 627 |
+
return _render_slides_page(post, safe_title, safe_img, safe_url)
|
| 628 |
+
|
| 629 |
+
if post and post.get('video'):
|
| 630 |
+
return _render_video_page(post, safe_title, safe_img, safe_url)
|
| 631 |
+
|
| 632 |
+
# Otherwise redirect
|
| 633 |
return HTMLResponse(f'''<!DOCTYPE html>
|
| 634 |
+
<html lang="vi">
|
| 635 |
+
<head>
|
| 636 |
+
<meta charset="utf-8">
|
| 637 |
+
<meta name="viewport" content="width=device-width,initial-scale=1">
|
| 638 |
+
<title>{_clean(safe_title)}</title>
|
| 639 |
+
<meta property="og:title" content="{_clean(safe_title)}">
|
| 640 |
+
<meta property="og:image" content="{_clean(safe_img)}">
|
| 641 |
<meta property="og:description" content="Tin tức tóm tắt, AI rewrite, World Cup 2026">
|
| 642 |
+
<meta property="og:url" content="{SPACE}/s/{slug}">
|
| 643 |
+
<link rel="canonical" href="{SPACE}/s/{slug}">
|
| 644 |
+
<meta http-equiv="refresh" content="0;url={safe_url}">
|
| 645 |
+
</head><body></body></html>''')
|
| 646 |
+
|
| 647 |
+
|
| 648 |
|
| 649 |
@app.get('/api/proxy/img')
|
| 650 |
def proxy_img(url: str = Query(default=""), max_size: int = Query(default=1200)):
|
| 651 |
+
"""Proxy image from blocked CDN to public URL."""
|
| 652 |
from urllib.parse import unquote
|
| 653 |
safe_url = unquote(url)
|
| 654 |
import requests as _req
|
|
|
|
| 656 |
r = _req.get(safe_url, timeout=10, headers={"User-Agent": "Mozilla/5.0"})
|
| 657 |
if r.status_code == 200 and r.content:
|
| 658 |
return Response(r.content, media_type=r.headers.get("content-type", "image/jpeg"))
|
| 659 |
+
except Exception as e:
|
| 660 |
+
pass
|
| 661 |
return Response(status_code=502)
|
|
|
|
| 662 |
@app.get('/s')
|
| 663 |
async def _sh(url:str='',title:str='',img:str='',post_id:str=''):
|
| 664 |
safe_title = _clean(title) if title else 'VNEWS - Tin tức'
|
| 665 |
safe_img = _clean(img) if img else ''
|
| 666 |
safe_url = _clean(url) if url else '/'
|
| 667 |
+
|
| 668 |
+
# Try to find wall post by post_id or URL (prioritize posts with slides/video)
|
| 669 |
post = None
|
| 670 |
try:
|
| 671 |
posts = _load_wall_posts()
|
| 672 |
if post_id:
|
| 673 |
for p in posts:
|
| 674 |
+
if p.get('id') == post_id:
|
| 675 |
+
post = p
|
| 676 |
+
safe_title = p.get('title', safe_title) or safe_title
|
| 677 |
+
safe_img = p.get('img', safe_img) or safe_img
|
| 678 |
+
safe_url = p.get('url', safe_url) or safe_url
|
| 679 |
+
break
|
| 680 |
elif url:
|
| 681 |
+
# Find matching URL - prioritize posts with slides or video
|
| 682 |
for p in posts:
|
| 683 |
+
if p.get('url') == url and p.get('slides'):
|
| 684 |
+
post = p
|
| 685 |
+
safe_title = p.get('title', safe_title) or safe_title
|
| 686 |
+
safe_img = p.get('img', safe_img) or safe_img
|
| 687 |
+
safe_url = p.get('url', safe_url) or safe_url
|
| 688 |
+
break
|
| 689 |
if not post:
|
| 690 |
+
# Fallback: find any matching URL
|
| 691 |
for p in posts:
|
| 692 |
+
if p.get('url') == url:
|
| 693 |
+
post = p
|
| 694 |
+
safe_title = p.get('title', safe_title) or safe_title
|
| 695 |
+
safe_img = p.get('img', safe_img) or safe_img
|
| 696 |
+
safe_url = p.get('url', safe_url) or safe_url
|
| 697 |
+
break
|
| 698 |
+
except:
|
| 699 |
+
pass
|
| 700 |
+
|
| 701 |
+
if post and post.get('slides'):
|
| 702 |
+
return _render_slides_page(post, safe_title, safe_img, safe_url)
|
| 703 |
+
|
| 704 |
+
if post and post.get('video'):
|
| 705 |
+
return _render_video_page(post, safe_title, safe_img, safe_url)
|
| 706 |
+
|
| 707 |
+
# Fallback: redirect to original URL
|
| 708 |
+
# Fetch og:image for better rich preview
|
| 709 |
if url and not safe_img:
|
| 710 |
try:
|
| 711 |
art = _scrape_article_fast(url)
|
| 712 |
+
if art and art.get('og_image'):
|
| 713 |
+
safe_img = art.get('og_image', '')
|
| 714 |
+
if art and art.get('title'):
|
| 715 |
+
safe_title = art.get('title', safe_title)
|
| 716 |
+
except:
|
| 717 |
+
pass
|
| 718 |
+
|
| 719 |
return HTMLResponse(f'''<!DOCTYPE html>
|
| 720 |
+
<html lang="vi">
|
| 721 |
+
<head>
|
| 722 |
+
<meta charset="utf-8">
|
| 723 |
+
<meta name="viewport" content="width=device-width,initial-scale=1">
|
| 724 |
+
<title>{safe_title}</title>
|
| 725 |
+
<meta property="og:title" content="{safe_title}">
|
| 726 |
+
<meta property="og:image" content="{safe_img}">
|
| 727 |
<meta property="og:description" content="Tin tức tóm tắt, AI rewrite, World Cup 2026">
|
| 728 |
+
<meta property="og:url" content="{SPACE}/s?url={quote(safe_url)}">
|
| 729 |
+
<link rel="canonical" href="{SPACE}/s?url={quote(safe_url)}">
|
| 730 |
+
<meta http-equiv="refresh" content="0;url={safe_url}">
|
| 731 |
+
</head><body></body></html>''')
|
| 732 |
|
| 733 |
from wc2026_scraper import scrape_summary,scrape_fixtures,scrape_standings,scrape_stats,scrape_wc_news,scrape_road_to_wc,get_wc2026_all,scrape_history,scrape_h2h,scrape_lineups,scrape_match_detail
|
| 734 |
|
|
|
|
| 738 |
def _xlb_scrape(path):
|
| 739 |
url = f"https://xemlaibongda.top/{path}"
|
| 740 |
r = req.get(url, headers={"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36"}, timeout=15, allow_redirects=True)
|
| 741 |
+
if r.status_code != 200:
|
| 742 |
+
return []
|
| 743 |
soup = BeautifulSoup(r.text, 'lxml')
|
| 744 |
+
vids = []
|
| 745 |
+
seen = set()
|
| 746 |
for a in soup.select('a[href*="/video/"]'):
|
| 747 |
href = a.get('href', '')
|
| 748 |
+
if not href or href in seen:
|
| 749 |
+
continue
|
| 750 |
seen.add(href)
|
| 751 |
+
if not href.startswith('http'):
|
| 752 |
+
href = 'https://xemlaibongda.top' + href
|
| 753 |
img = a.select_one('img')
|
| 754 |
p = a.parent
|
| 755 |
for _ in range(4):
|
| 756 |
+
if img:
|
| 757 |
+
break
|
| 758 |
+
if p:
|
| 759 |
+
img = p.select_one('img')
|
| 760 |
+
p = p.parent
|
| 761 |
img_src = ''
|
| 762 |
+
if img:
|
| 763 |
+
img_src = img.get('data-src','') or img.get('src','') or img.get('data-lazy','') or img.get('data-original','')
|
| 764 |
+
if img_src.startswith('//'):
|
| 765 |
+
img_src = 'https:' + img_src
|
| 766 |
+
elif img_src.startswith('/'):
|
| 767 |
+
img_src = 'https://xemlaibongda.top' + img_src
|
| 768 |
title = ''
|
| 769 |
for sel in ['.title', 'h3', 'h2', '.name', '.post-title', '.entry-title', '.video-title']:
|
| 770 |
t = a.select_one(sel)
|
| 771 |
+
if t:
|
| 772 |
+
title = _clean(t.get_text())
|
| 773 |
+
break
|
| 774 |
+
if not title:
|
| 775 |
+
title = _clean(a.get('title',''))
|
| 776 |
if not title:
|
| 777 |
img_alt = a.select_one('img')
|
| 778 |
+
if img_alt:
|
| 779 |
+
title = _clean(img_alt.get('alt',''))
|
| 780 |
if not title:
|
| 781 |
parent = a.parent
|
| 782 |
+
if parent:
|
| 783 |
+
pt = _clean(parent.get_text(' ',strip=True))
|
| 784 |
+
if 5 < len(pt) < 120:
|
| 785 |
+
title = pt
|
| 786 |
+
if not title or len(title) < 3:
|
| 787 |
+
continue
|
| 788 |
vids.append({"link": href, "img": img_src, "title": title})
|
| 789 |
+
if len(vids) >= 30:
|
| 790 |
+
break
|
| 791 |
return vids
|
| 792 |
|
| 793 |
@app.get('/api/proxy/xlb')
|
|
|
|
| 796 |
cache_key = f"xlb:{path}"
|
| 797 |
with _xlb_lock:
|
| 798 |
cached = _xlb_cache.get(cache_key)
|
| 799 |
+
if cached and now - cached['t'] < 120:
|
| 800 |
+
return JSONResponse(cached['d'])
|
| 801 |
try:
|
| 802 |
vids = _xlb_scrape(path)
|
| 803 |
result = {"videos": vids, "count": len(vids)}
|
| 804 |
+
with _xlb_lock:
|
| 805 |
+
_xlb_cache[cache_key] = {'t': now, 'd': result}
|
| 806 |
return JSONResponse(result)
|
| 807 |
except Exception as e:
|
| 808 |
return JSONResponse({"videos": [], "count": 0, "error": str(e)}, status_code=500)
|
|
|
|
| 841 |
try:
|
| 842 |
if os.path.exists(p):return json.load(open(p,'r',encoding='utf-8'))
|
| 843 |
except:pass
|
| 844 |
+
return [] # Return empty list instead of dict for wall posts
|
| 845 |
def _sj(p,d):
|
| 846 |
try:open(p+'.tmp','w',encoding='utf-8').write(json.dumps(d,ensure_ascii=False));os.replace(p+'.tmp',p)
|
| 847 |
except:pass
|
|
|
|
| 870 |
return JSONResponse({'comments':cms})
|
| 871 |
|
| 872 |
def _load_wall_posts():
|
| 873 |
+
with _wl_lock:
|
| 874 |
+
return _lj(WALL_FILE)
|
| 875 |
|
| 876 |
def _save_wall_posts(posts):
|
| 877 |
+
with _wl_lock:
|
| 878 |
+
_sj(WALL_FILE, posts)
|
| 879 |
|
| 880 |
@app.get('/api/wall')
|
| 881 |
def api_wall():
|
| 882 |
posts = _load_wall_posts()
|
| 883 |
+
if not posts:
|
| 884 |
+
return JSONResponse({"posts": []})
|
| 885 |
return JSONResponse({"posts": posts})
|
| 886 |
|
| 887 |
@app.post('/api/wall')
|
| 888 |
async def api_wall_post(request: Request):
|
| 889 |
content_type = request.headers.get('content-type', '')
|
| 890 |
if 'multipart/form-data' in content_type:
|
| 891 |
+
try:
|
| 892 |
+
form = await request.form()
|
| 893 |
+
except Exception as e:
|
| 894 |
+
return JSONResponse({"error": f"Form parse error: {str(e)}"}, status_code=400)
|
| 895 |
title = form.get('title', 'Video mới') or 'Video mới'
|
| 896 |
text = form.get('text', '') or ''
|
| 897 |
source = form.get('source', 'vtv_recorder') or 'vtv_recorder'
|
|
|
|
| 900 |
video_url = None
|
| 901 |
if video_file and hasattr(video_file, 'filename') and video_file.filename:
|
| 902 |
fname = video_file.filename.lower()
|
| 903 |
+
if fname.endswith('.mp4'):
|
| 904 |
+
ext = '.mp4'
|
| 905 |
+
elif fname.endswith('.webm'):
|
| 906 |
+
ext = '.webm'
|
| 907 |
+
else:
|
| 908 |
+
ext = '.webm'
|
| 909 |
video_filename = f"wall_{post_id}{ext}"
|
| 910 |
video_path = os.path.join(WALL_VIDEO_DIR, video_filename)
|
| 911 |
try:
|
| 912 |
content = await video_file.read()
|
| 913 |
+
if not content:
|
| 914 |
+
return JSONResponse({"error": "Empty video file"}, status_code=400)
|
| 915 |
+
with open(video_path, 'wb') as f:
|
| 916 |
+
f.write(content)
|
| 917 |
file_size_mb = len(content) / 1024 / 1024
|
| 918 |
+
if file_size_mb > 50:
|
| 919 |
+
os.remove(video_path)
|
| 920 |
+
return JSONResponse({"error": f"Video quá lớn ({file_size_mb:.1f}MB). Tối đa 50MB."}, status_code=400)
|
| 921 |
video_url = f"/api/wall/video/{video_filename}"
|
| 922 |
+
except Exception as e:
|
| 923 |
+
return JSONResponse({"error": f"Lỗi lưu video: {str(e)}"}, status_code=500)
|
| 924 |
+
post = {
|
| 925 |
+
"id": post_id,
|
| 926 |
+
"title": title[:200],
|
| 927 |
+
"text": text[:2000],
|
| 928 |
+
"source": source,
|
| 929 |
+
"video": video_url,
|
| 930 |
+
"img": None,
|
| 931 |
+
"images": [],
|
| 932 |
+
"created": int(time.time()),
|
| 933 |
+
"created_str": time.strftime('%H:%M %d/%m/%Y', time.localtime()),
|
| 934 |
+
}
|
| 935 |
posts = _load_wall_posts()
|
| 936 |
+
if not isinstance(posts, list):
|
| 937 |
+
posts = []
|
| 938 |
+
posts.insert(0, post)
|
| 939 |
+
posts = posts[:200]
|
| 940 |
+
_save_wall_posts(posts)
|
| 941 |
return JSONResponse({"post": post, "ok": True})
|
| 942 |
+
try:
|
| 943 |
+
body = await request.json()
|
| 944 |
+
except:
|
| 945 |
+
body = {}
|
| 946 |
title = body.get('title', 'Bài mới') or 'Bài mới'
|
| 947 |
text = body.get('text', '') or ''
|
| 948 |
img = body.get('img', None)
|
| 949 |
source = body.get('source', 'user') or 'user'
|
| 950 |
post_id = str(uuid.uuid4())[:12]
|
| 951 |
+
post = {
|
| 952 |
+
"id": post_id,
|
| 953 |
+
"title": title[:200],
|
| 954 |
+
"text": text[:2000],
|
| 955 |
+
"source": source,
|
| 956 |
+
"video": None,
|
| 957 |
+
"img": img,
|
| 958 |
+
"images": [],
|
| 959 |
+
"created": int(time.time()),
|
| 960 |
+
"created_str": time.strftime('%H:%M %d/%m/%Y', time.localtime()),
|
| 961 |
+
}
|
| 962 |
posts = _load_wall_posts()
|
| 963 |
+
if not isinstance(posts, list):
|
| 964 |
+
posts = []
|
| 965 |
+
posts.insert(0, post)
|
| 966 |
+
posts = posts[:200]
|
| 967 |
+
_save_wall_posts(posts)
|
| 968 |
return JSONResponse({"post": post, "ok": True})
|
| 969 |
|
| 970 |
@app.get('/api/wall/video/{filename}')
|
| 971 |
def api_wall_video(filename: str):
|
| 972 |
+
if '..' in filename or '/' in filename:
|
| 973 |
+
return Response(status_code=403)
|
| 974 |
video_path = os.path.join(WALL_VIDEO_DIR, filename)
|
| 975 |
+
if not os.path.exists(video_path):
|
| 976 |
+
return Response(status_code=404)
|
| 977 |
ext = os.path.splitext(filename)[1].lower()
|
| 978 |
media_type = 'video/mp4' if ext == '.mp4' else 'video/webm'
|
| 979 |
return FileResponse(video_path, media_type=media_type)
|
|
|
|
| 981 |
@app.delete('/api/wall/{post_id}')
|
| 982 |
def api_wall_delete(post_id: str):
|
| 983 |
posts = _load_wall_posts()
|
| 984 |
+
if not isinstance(posts, list):
|
| 985 |
+
return JSONResponse({"error": "No posts"}, status_code=404)
|
| 986 |
for i, p in enumerate(posts):
|
| 987 |
if p.get('id') == post_id:
|
| 988 |
if p.get('video'):
|
| 989 |
video_name = p['video'].split('/')[-1]
|
| 990 |
video_path = os.path.join(WALL_VIDEO_DIR, video_name)
|
| 991 |
+
if os.path.exists(video_path):
|
| 992 |
+
os.remove(video_path)
|
| 993 |
+
posts.pop(i)
|
| 994 |
+
_save_wall_posts(posts)
|
| 995 |
return JSONResponse({"ok": True})
|
| 996 |
return JSONResponse({"error": "Post not found"}, status_code=404)
|
| 997 |
|
|
|
|
| 1001 |
|
| 1002 |
_UA_RW = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36', 'Accept-Language': 'vi-VN,vi;q=0.9'}
|
| 1003 |
|
| 1004 |
+
# Unique character markers for language detection
|
| 1005 |
_UNIQUE_CHARS = {
|
| 1006 |
'vietnamese': set('đăâêôơưàảãạáằẳẵặắầẩẫậấèẻẽẹéềễểệếìỉĩịíòỏõọóồổỗộốờởỡợớùủũụúừửữựứỳỷỹỵý'),
|
| 1007 |
'spanish': set('ñáéíóúü¿¡'),
|
|
|
|
| 1016 |
}
|
| 1017 |
|
| 1018 |
def detect_language(text):
|
| 1019 |
+
"""Detect language from text content using stopword + character analysis."""
|
| 1020 |
+
if not text:
|
| 1021 |
+
return 'vietnamese'
|
| 1022 |
+
text_lower = text.lower()
|
| 1023 |
+
text_chars = set(text_lower)
|
| 1024 |
+
|
| 1025 |
+
# Strong signal: Vietnamese unique characters
|
| 1026 |
vn_chars = len(text_chars & _UNIQUE_CHARS['vietnamese'])
|
| 1027 |
+
if vn_chars >= 2:
|
| 1028 |
+
return 'vietnamese'
|
| 1029 |
+
|
| 1030 |
+
# Spanish unique chars (ñ, ¿, ¡)
|
| 1031 |
es_chars = len(text_chars & _UNIQUE_CHARS['spanish'])
|
| 1032 |
pt_chars = len(text_chars & _UNIQUE_CHARS['portuguese'])
|
| 1033 |
+
|
| 1034 |
+
# Stopword scoring
|
| 1035 |
words = set(re.findall(r'\b\w+\b', text_lower))
|
| 1036 |
scores = {}
|
| 1037 |
+
for lang, stops in _STOPWORDS.items():
|
| 1038 |
+
scores[lang] = len(words & stops) / max(len(stops), 1)
|
| 1039 |
+
|
| 1040 |
+
# Disambiguate Portuguese vs Spanish
|
| 1041 |
pt_markers = {'não', 'pelo', 'pela', 'isso', 'há', 'estão', 'num', 'numa', 'tenho', 'posso', 'você', 'nós', 'eles', 'elas', 'também', 'muito', 'já', 'só', 'até', 'entre', 'depois', 'sem', 'mesmo', 'aos', 'serão'}
|
| 1042 |
es_markers = {'pero', 'está', 'están', 'porque', 'también', 'hasta', 'donde', 'quien', 'fue', 'son', 'fueron', 'ese', 'eso', 'ante', 'ellos', 'ella', 'nosotros', 'usted', 'ustedes', 'tú', 'él', 'desde', 'todo', 'durante', 'todos', 'uno', 'les', 'ni', 'contra', 'otros', 'fueron'}
|
| 1043 |
+
|
| 1044 |
+
pt_overlap = len(words & pt_markers)
|
| 1045 |
+
es_overlap = len(words & es_markers)
|
| 1046 |
+
|
| 1047 |
+
if scores.get('portuguese', 0) > 0 and pt_overlap > es_overlap:
|
| 1048 |
+
return 'portuguese'
|
| 1049 |
+
if scores.get('spanish', 0) > 0 and es_overlap > pt_overlap:
|
| 1050 |
+
return 'spanish'
|
| 1051 |
+
if scores.get('english', 0) > 0.15:
|
| 1052 |
+
return 'english'
|
| 1053 |
+
|
| 1054 |
best = max(scores, key=scores.get)
|
| 1055 |
return best if scores[best] > 0.05 else 'vietnamese'
|
| 1056 |
|
| 1057 |
+
# Emotion keyword-based detection
|
| 1058 |
_EMOTION_KEYWORDS = {
|
| 1059 |
+
'happy': {
|
| 1060 |
+
'en': ['happy', 'joy', 'wonderful', 'great', 'amazing', 'fantastic', 'love', 'excellent', 'beautiful', 'glad', 'delighted', 'pleased', 'cheerful', 'celebrate', 'victory', 'win', 'success'],
|
| 1061 |
+
'pt': ['feliz', 'alegria', 'maravilhoso', 'ótimo', 'incrível', 'fantástico', 'amor', 'excelente', 'lindo', 'contente', 'encantado', 'vitória', 'sucesso'],
|
| 1062 |
+
'es': ['feliz', 'alegria', 'maravilloso', 'genial', 'increíble', 'fantástico', 'amor', 'excelente', 'hermoso', 'contento', 'encantado', 'victoria', 'éxito'],
|
| 1063 |
+
'vi': ['vui', 'hạnh phúc', 'tuyệt vời', 'tuyệt', 'ý nghĩa', 'đẹp', 'thích', 'yêu', 'vui vẻ', 'hân hoan', 'phấn khích', 'chiến thắng', 'thành công'],
|
| 1064 |
+
},
|
| 1065 |
+
'sad': {
|
| 1066 |
+
'en': ['sad', 'unhappy', 'terrible', 'awful', 'horrible', 'miserable', 'depressed', 'grief', 'sorrow', 'tragic', 'unfortunate', 'painful', 'death', 'die', 'kill'],
|
| 1067 |
+
'pt': ['triste', 'infeliz', 'terrível', 'horrível', 'miserável', 'deprimido', 'dor', 'trágico', 'infelizmente', 'penoso', 'morte', 'morrer'],
|
| 1068 |
+
'es': ['triste', 'infeliz', 'terrible', 'horrible', 'miserable', 'deprimido', 'dolor', 'trágico', 'desafortunado', 'penoso', 'muerte', 'morir'],
|
| 1069 |
+
'vi': ['buồn', 'không vui', 'tồi tệ', 'kinh khủng', '��au khổ', 'đau buồn', 'bi thương', 'khốn nạn', 'đau đớn', 'thảm họa', 'chết', 'mất'],
|
| 1070 |
+
},
|
| 1071 |
+
'excited': {
|
| 1072 |
+
'en': ['excited', 'thrilling', 'amazing', 'wow', 'incredible', 'unbelievable', 'awesome', 'exhilarating', 'electrifying', 'breathtaking', 'breakthrough', 'record'],
|
| 1073 |
+
'pt': ['animado', 'emocionante', 'incrível', 'impressionante', 'sensacional', 'eletrizante', 'empolgante', 'recorde'],
|
| 1074 |
+
'es': ['emocionante', 'increíble', 'impresionante', 'sensacional', 'electrizante', 'emocionado', 'entusiasmado', 'récord'],
|
| 1075 |
+
'vi': ['hào hứng', 'phấn khích', 'thú vị', 'tuyệt cú mèo', 'đỉnh cao', 'ngoạn mục', 'sục sôi', 'kỷ lục', 'đột phá'],
|
| 1076 |
+
},
|
| 1077 |
+
'humorous': {
|
| 1078 |
+
'en': ['funny', 'hilarious', 'joke', 'laugh', 'comedy', 'humor', 'amusing', 'witty', 'sarcastic', 'ironic', 'ridiculous', 'absurd', 'lol', 'haha'],
|
| 1079 |
+
'pt': ['engraçado', 'hilário', 'piada', 'rir', 'comédia', 'humor', 'divertido', 'irônico', 'ridículo', 'absurdo', 'kkk'],
|
| 1080 |
+
'es': ['gracioso', 'hilarante', 'broma', 'risa', 'comedia', 'humor', 'divertido', 'irónico', 'ridículo', 'absurdo', 'jaja'],
|
| 1081 |
+
'vi': ['hài hước', 'buồn cười', 'đùa', 'cười', 'hài', 'vui nhộn', 'hóm hỉnh', 'mỉa mai', 'lố bịch', 'vô lý', 'haha'],
|
| 1082 |
+
},
|
| 1083 |
+
'serious': {
|
| 1084 |
+
'en': ['serious', 'critical', 'important', 'urgent', 'severe', 'grave', 'significant', 'crucial', 'vital', 'essential', 'alarming', 'concerning', 'crisis', 'war', 'conflict'],
|
| 1085 |
+
'pt': ['sério', 'crítico', 'importante', 'urgente', 'grave', 'significativo', 'crucial', 'vital', 'essencial', 'preocupante', 'crise', 'guerra', 'conflito'],
|
| 1086 |
+
'es': ['serio', 'crítico', 'importante', 'urgente', 'grave', 'significativo', 'crucial', 'vital', 'esencial', 'preocupante', 'crisis', 'guerra', 'conflicto'],
|
| 1087 |
+
'vi': ['nghiêm trọng', 'quan trọng', 'khẩn cấp', 'nghiêm túc', 'đáng kể', 'thiết yếu', 'cần thiết', 'báo động', 'lo ngại', 'khủng hoảng', 'chiến tranh', 'xung đột'],
|
| 1088 |
+
},
|
| 1089 |
+
}
|
| 1090 |
+
|
| 1091 |
+
def detect_emotion(text, language='vietnamese'):
|
| 1092 |
+
"""Detect emotion from text using keyword matching."""
|
| 1093 |
+
if not text:
|
| 1094 |
+
return 'neutral'
|
| 1095 |
+
text_lower = text.lower()
|
| 1096 |
+
|
| 1097 |
+
scores = {}
|
| 1098 |
+
for emotion, lang_keywords in _EMOTION_KEYWORDS.items():
|
| 1099 |
+
keywords = lang_keywords.get(language, lang_keywords.get('en', []))
|
| 1100 |
+
score = sum(1 for kw in keywords if kw in text_lower)
|
| 1101 |
+
scores[emotion] = score
|
| 1102 |
+
|
| 1103 |
+
if max(scores.values()) == 0:
|
| 1104 |
+
return 'neutral'
|
| 1105 |
+
|
| 1106 |
+
return max(scores, key=scores.get)
|
| 1107 |
+
|
| 1108 |
+
def detect_language_and_emotion(title, text):
|
| 1109 |
+
"""Detect both language and emotion from article content."""
|
| 1110 |
+
combined = f"{title} {text}"
|
| 1111 |
+
lang = detect_language(combined)
|
| 1112 |
+
emotion = detect_emotion(combined, lang)
|
| 1113 |
+
return lang, emotion
|
| 1114 |
+
|
| 1115 |
+
# Voice selection based on language and emotion (using MultilingualNeural voices)
|
| 1116 |
+
VOICE_BY_LANG_EMOTION = {
|
| 1117 |
+
'vietnamese': {
|
| 1118 |
+
'happy': ('vi-VN-HoaiMyNeural', 'vui'),
|
| 1119 |
+
'sad': ('vi-VN-NamMinhNeural', 'buồn'),
|
| 1120 |
+
'excited': ('vi-VN-HoaiMyNeural', 'hào hứng'),
|
| 1121 |
+
'humorous': ('vi-VN-HoaiMyNeural', 'vui'),
|
| 1122 |
+
'serious': ('vi-VN-NamMinhNeural', 'nghiêm túc'),
|
| 1123 |
+
'neutral': ('vi-VN-HoaiMyNeural', 'trung_tinh'),
|
| 1124 |
+
},
|
| 1125 |
+
'portuguese': {
|
| 1126 |
+
'happy': ('pt-BR-ThalitaMultilingualNeural', 'feliz'),
|
| 1127 |
+
'sad': ('pt-BR-ThalitaMultilingualNeural', 'triste'),
|
| 1128 |
+
'excited': ('pt-BR-ThalitaMultilingualNeural', 'animado'),
|
| 1129 |
+
'humorous': ('pt-BR-ThalitaMultilingualNeural', 'engraçado'),
|
| 1130 |
+
'serious': ('pt-BR-ThalitaMultilingualNeural', 'sério'),
|
| 1131 |
+
'neutral': ('pt-BR-ThalitaMultilingualNeural', 'neutro'),
|
| 1132 |
+
},
|
| 1133 |
+
'english': {
|
| 1134 |
+
'happy': ('en-US-AndrewMultilingualNeural', 'happy'),
|
| 1135 |
+
'sad': ('en-AU-WilliamMultilingualNeural', 'sad'),
|
| 1136 |
+
'excited': ('en-US-AndrewMultilingualNeural', 'excited'),
|
| 1137 |
+
'humorous': ('en-US-AndrewMultilingualNeural', 'funny'),
|
| 1138 |
+
'serious': ('en-AU-WilliamMultilingualNeural', 'serious'),
|
| 1139 |
+
'neutral': ('en-US-AndrewMultilingualNeural', 'neutral'),
|
| 1140 |
+
},
|
| 1141 |
+
'french': {
|
| 1142 |
+
'happy': ('fr-FR-VivienneMultilingualNeural', 'heureux'),
|
| 1143 |
+
'sad': ('fr-FR-RemyMultilingualNeural', 'triste'),
|
| 1144 |
+
'excited': ('fr-FR-VivienneMultilingualNeural', 'excité'),
|
| 1145 |
+
'humorous': ('fr-FR-VivienneMultilingualNeural', 'drôle'),
|
| 1146 |
+
'serious': ('fr-FR-RemyMultilingualNeural', 'sérieux'),
|
| 1147 |
+
'neutral': ('fr-FR-VivienneMultilingualNeural', 'neutre'),
|
| 1148 |
+
},
|
| 1149 |
+
'german': {
|
| 1150 |
+
'happy': ('de-DE-SeraphinaMultilingualNeural', 'glücklich'),
|
| 1151 |
+
'sad': ('de-DE-FlorianMultilingualNeural', 'traurig'),
|
| 1152 |
+
'excited': ('de-DE-SeraphinaMultilingualNeural', 'aufgeregt'),
|
| 1153 |
+
'humorous': ('de-DE-SeraphinaMultilingualNeural', 'lustig'),
|
| 1154 |
+
'serious': ('de-DE-FlorianMultilingualNeural', 'ernst'),
|
| 1155 |
+
'neutral': ('de-DE-SeraphinaMultilingualNeural', 'neutral'),
|
| 1156 |
+
},
|
| 1157 |
+
'korean': {
|
| 1158 |
+
'happy': ('ko-KR-HyunsuMultilingualNeural', '행복'),
|
| 1159 |
+
'sad': ('ko-KR-HyunsuMultilingualNeural', '슬픔'),
|
| 1160 |
+
'excited': ('ko-KR-HyunsuMultilingualNeural', '흥분'),
|
| 1161 |
+
'humorous': ('ko-KR-HyunsuMultilingualNeural', '유쾌'),
|
| 1162 |
+
'serious': ('ko-KR-HyunsuMultilingualNeural', '진지'),
|
| 1163 |
+
'neutral': ('ko-KR-HyunsuMultilingualNeural', '중립'),
|
| 1164 |
+
},
|
| 1165 |
+
'italian': {
|
| 1166 |
+
'happy': ('it-IT-GiuseppeMultilingualNeural', 'felice'),
|
| 1167 |
+
'sad': ('it-IT-GiuseppeMultilingualNeural', 'triste'),
|
| 1168 |
+
'excited': ('it-IT-GiuseppeMultilingualNeural', 'emozionato'),
|
| 1169 |
+
'humorous': ('it-IT-GiuseppeMultilingualNeural', 'divertente'),
|
| 1170 |
+
'serious': ('it-IT-GiuseppeMultilingualNeural', 'serio'),
|
| 1171 |
+
'neutral': ('it-IT-GiuseppeMultilingualNeural', 'neutro'),
|
| 1172 |
+
},
|
| 1173 |
+
}
|
| 1174 |
+
|
| 1175 |
+
# All valid voice IDs (new MultilingualNeural format)
|
| 1176 |
+
VALID_VOICES = {
|
| 1177 |
+
'vi-VN-HoaiMyNeural', 'vi-VN-NamMinhNeural',
|
| 1178 |
+
'en-US-AndrewMultilingualNeural', 'en-AU-WilliamMultilingualNeural',
|
| 1179 |
+
'pt-BR-ThalitaMultilingualNeural',
|
| 1180 |
+
'fr-FR-VivienneMultilingualNeural', 'fr-FR-RemyMultilingualNeural',
|
| 1181 |
+
'de-DE-SeraphinaMultilingualNeural', 'de-DE-FlorianMultilingualNeural',
|
| 1182 |
+
'ko-KR-HyunsuMultilingualNeural',
|
| 1183 |
+
'it-IT-GiuseppeMultilingualNeural',
|
| 1184 |
+
}
|
| 1185 |
+
|
| 1186 |
+
def get_voice_for_content(title, text, preferred_voice=None):
|
| 1187 |
+
"""Get appropriate voice based on content language and emotion."""
|
| 1188 |
+
# Accept the new MultilingualNeural voices directly
|
| 1189 |
+
if preferred_voice and preferred_voice in VALID_VOICES:
|
| 1190 |
+
return preferred_voice
|
| 1191 |
+
|
| 1192 |
+
# Also accept old shorthand voice IDs and map them to new format
|
| 1193 |
+
old_voice_map = {
|
| 1194 |
+
'hoaimy': 'vi-VN-HoaiMyNeural',
|
| 1195 |
+
'namminh': 'vi-VN-NamMinhNeural',
|
| 1196 |
+
'andrew': 'en-US-AndrewMultilingualNeural',
|
| 1197 |
+
'jenny': 'en-US-AndrewMultilingualNeural',
|
| 1198 |
+
'thalita': 'pt-BR-ThalitaMultilingualNeural',
|
| 1199 |
+
'pt_thalita': 'pt-BR-ThalitaMultilingualNeural',
|
| 1200 |
+
'pt_francisco': 'pt-BR-ThalitaMultilingualNeural',
|
| 1201 |
+
'ela': 'en-US-AndrewMultilingualNeural',
|
| 1202 |
+
'es_carlos': 'en-US-AndrewMultilingualNeural',
|
| 1203 |
+
'denise': 'fr-FR-VivienneMultilingualNeural',
|
| 1204 |
+
'katja': 'de-DE-SeraphinaMultilingualNeural',
|
| 1205 |
+
'nanami': 'en-US-AndrewMultilingualNeural',
|
| 1206 |
+
'sunhee': 'ko-KR-HyunsuMultilingualNeural',
|
| 1207 |
+
'xiaochen': 'en-US-AndrewMultilingualNeural',
|
| 1208 |
+
}
|
| 1209 |
+
if preferred_voice and preferred_voice in old_voice_map:
|
| 1210 |
+
return old_voice_map[preferred_voice]
|
| 1211 |
+
|
| 1212 |
+
lang, emotion = detect_language_and_emotion(title, text)
|
| 1213 |
+
lang_map = VOICE_BY_LANG_EMOTION.get(lang, VOICE_BY_LANG_EMOTION['vietnamese'])
|
| 1214 |
+
voice, _ = lang_map.get(emotion, lang_map['neutral'])
|
| 1215 |
+
return voice
|
| 1216 |
+
|
| 1217 |
+
|
| 1218 |
+
def _is_relevant_image(img_url, title, text):
|
| 1219 |
+
"""Check if an image is relevant to the article content."""
|
| 1220 |
+
if not img_url:
|
| 1221 |
+
return False
|
| 1222 |
+
skip_patterns = ['pixel', 'analytics', 'tracking', '1x1.gif', 'spacer.gif',
|
| 1223 |
+
'logo', 'icon', 'avatar', 'emoji', 'smiley', 'sprite',
|
| 1224 |
+
'advertisement', 'ad-banner', 'sponsored', 'banner-ads']
|
| 1225 |
+
img_lower = img_url.lower()
|
| 1226 |
+
for p in skip_patterns:
|
| 1227 |
+
if p in img_lower:
|
| 1228 |
+
return False
|
| 1229 |
+
if not any(img_lower.endswith(ext) for ext in ['.jpg', '.jpeg', '.png', '.webp', '.gif']):
|
| 1230 |
+
return False
|
| 1231 |
+
return True
|
| 1232 |
+
|
| 1233 |
+
|
| 1234 |
+
def _filter_relevant_images(images, title, text, max_images=8):
|
| 1235 |
+
"""Filter and rank images by relevance to article content."""
|
| 1236 |
+
if not images:
|
| 1237 |
+
return []
|
| 1238 |
+
seen = set()
|
| 1239 |
+
relevant = []
|
| 1240 |
+
for img in images:
|
| 1241 |
+
if img in seen:
|
| 1242 |
+
continue
|
| 1243 |
+
seen.add(img)
|
| 1244 |
+
if _is_relevant_image(img, title, text):
|
| 1245 |
+
relevant.append(img)
|
| 1246 |
+
return relevant[:max_images]
|
| 1247 |
+
|
| 1248 |
+
|
| 1249 |
+
def _scrape_article_for_rewrite(url):
|
| 1250 |
+
"""Scrape article: extract title, paragraphs, RELEVANT images, OG image."""
|
| 1251 |
+
try:
|
| 1252 |
+
r = req.get(url, headers=_UA_RW, timeout=15, allow_redirects=True)
|
| 1253 |
+
r.encoding = 'utf-8'
|
| 1254 |
+
soup = BeautifulSoup(r.text, 'lxml')
|
| 1255 |
+
for tag in soup.find_all(['script', 'style', 'nav', 'footer', 'aside', 'form']):
|
| 1256 |
+
tag.decompose()
|
| 1257 |
+
h1 = soup.find('h1')
|
| 1258 |
+
ogt = soup.find('meta', property='og:title')
|
| 1259 |
+
title = (h1.get_text(strip=True) if h1 else '') or (ogt.get('content', '') if ogt else '')
|
| 1260 |
+
ogi = soup.find('meta', property='og:image')
|
| 1261 |
+
og_img = ogi.get('content', '') if ogi else ''
|
| 1262 |
+
if og_img and og_img.startswith('//'):
|
| 1263 |
+
og_img = 'https:' + og_img
|
| 1264 |
+
block = None
|
| 1265 |
+
for sel in ['article', '.singular-content', '.detail-content', '.fck_detail', '.content-detail', '.knc-content', 'main', '.cms-body', '.article__body']:
|
| 1266 |
+
el = soup.select_one(sel)
|
| 1267 |
+
if el and len(el.find_all('p')) >= 2:
|
| 1268 |
+
block = el
|
| 1269 |
+
break
|
| 1270 |
+
if not block:
|
| 1271 |
+
block = soup.body or soup
|
| 1272 |
+
paragraphs = []
|
| 1273 |
+
all_images = []
|
| 1274 |
+
seen_imgs = set()
|
| 1275 |
+
if og_img and og_img not in seen_imgs:
|
| 1276 |
+
all_images.append(og_img)
|
| 1277 |
+
seen_imgs.add(og_img)
|
| 1278 |
+
for el in block.find_all(['p', 'h2', 'h3', 'figure', 'img'], recursive=True):
|
| 1279 |
+
if el.name == 'p':
|
| 1280 |
+
t = _clean(el.get_text(strip=True))
|
| 1281 |
+
if t and len(t) > 40:
|
| 1282 |
+
paragraphs.append(t)
|
| 1283 |
+
elif el.name in ('figure', 'img'):
|
| 1284 |
+
im = el if el.name == 'img' else el.find('img')
|
| 1285 |
+
if im:
|
| 1286 |
+
src = im.get('data-src') or im.get('src') or im.get('data-original') or ''
|
| 1287 |
+
if src and 'base64' not in src:
|
| 1288 |
+
if src.startswith('//'):
|
| 1289 |
+
src = 'https:' + src
|
| 1290 |
+
if src not in seen_imgs:
|
| 1291 |
+
all_images.append(src)
|
| 1292 |
+
seen_imgs.add(src)
|
| 1293 |
+
# Filter to relevant images only
|
| 1294 |
+
relevant_images = _filter_relevant_images(all_images, title, ' '.join(paragraphs[:5]))
|
| 1295 |
+
return {'title': _clean(title), 'paragraphs': paragraphs, 'images': relevant_images, 'og_img': og_img}
|
| 1296 |
+
except Exception:
|
| 1297 |
+
return None
|
| 1298 |
+
|
| 1299 |
+
|
| 1300 |
+
def _extract_key_points_rw(paragraphs, max_points=5):
|
| 1301 |
+
r"""Extract key points from paragraphs - extracts ALL sentences, not just first one.
|
| 1302 |
+
|
| 1303 |
+
Fixes: Original regex `^(.+?[.!?])\s` only captured first sentence per paragraph.
|
| 1304 |
+
Now splits on all sentence boundaries and takes valid sentences until max_points.
|
| 1305 |
+
"""
|
| 1306 |
+
points = []
|
| 1307 |
+
|
| 1308 |
+
for p in paragraphs:
|
| 1309 |
+
if len(points) >= max_points:
|
| 1310 |
+
break
|
| 1311 |
+
|
| 1312 |
+
p = _clean(p)
|
| 1313 |
+
if not p:
|
| 1314 |
+
continue
|
| 1315 |
+
|
| 1316 |
+
# Split paragraph into sentences using Vietnamese + English punctuation
|
| 1317 |
+
sentences = re.split(r'(?<=[.!?])\s+(?=[A-ZÀ-Ỹ0-9])', p)
|
| 1318 |
+
sentences = [s.strip() for s in sentences if s.strip()]
|
| 1319 |
+
|
| 1320 |
+
for sentence in sentences:
|
| 1321 |
+
if len(points) >= max_points:
|
| 1322 |
+
break
|
| 1323 |
+
|
| 1324 |
+
# Clean sentence - remove extra whitespace
|
| 1325 |
+
sentence = _clean(sentence)
|
| 1326 |
+
|
| 1327 |
+
if len(sentence) < 30:
|
| 1328 |
+
continue
|
| 1329 |
+
|
| 1330 |
+
# Check for duplicates
|
| 1331 |
+
if any(sentence[:60] in existing for existing in points):
|
| 1332 |
+
continue
|
| 1333 |
+
|
| 1334 |
+
# Ensure sentence ends with punctuation
|
| 1335 |
+
if not sentence.endswith(('.', '!', '?')):
|
| 1336 |
+
sentence = sentence + '.'
|
| 1337 |
+
|
| 1338 |
+
points.append(sentence)
|
| 1339 |
+
|
| 1340 |
+
# If no valid sentences found, take chunks from raw text
|
| 1341 |
+
if not points:
|
| 1342 |
+
raw = '\n'.join(paragraphs)
|
| 1343 |
+
for i in range(0, min(len(raw), max_points * 300), 280):
|
| 1344 |
+
chunk = _clean(raw[i:i+280])
|
| 1345 |
+
if len(chunk) >= 30 and chunk not in points:
|
| 1346 |
+
points.append(chunk + ('.' if not chunk.endswith('.') else ''))
|
| 1347 |
+
if len(points) >= max_points:
|
| 1348 |
+
break
|
| 1349 |
+
|
| 1350 |
+
return points
|
| 1351 |
+
|
| 1352 |
+
|
| 1353 |
+
@app.post("/api/rewrite_slide")
|
| 1354 |
+
async def api_rewrite_slide(request: Request):
|
| 1355 |
+
"""Fast rewrite as SLIDES - no AI needed, instant response."""
|
| 1356 |
+
body = await request.json()
|
| 1357 |
+
url = _clean(body.get("url", ""))
|
| 1358 |
+
context = body.get("context", "")
|
| 1359 |
+
preferred_voice = body.get("voice", "") # Accept custom voice selection
|
| 1360 |
+
if not url and not context:
|
| 1361 |
+
return JSONResponse({"error": "Cần URL hoặc nội dung"}, status_code=400)
|
| 1362 |
+
data = None
|
| 1363 |
+
if url and url.startswith("http"):
|
| 1364 |
+
data = _scrape_article_for_rewrite(url)
|
| 1365 |
+
if not data and context:
|
| 1366 |
+
paragraphs = [_clean(p) for p in context.split('\n') if len(_clean(p)) > 40]
|
| 1367 |
+
data = {'title': paragraphs[0][:80] if paragraphs else 'Bài viết', 'paragraphs': paragraphs, 'images': [], 'og_img': ''}
|
| 1368 |
+
if not data or not data.get('paragraphs'):
|
| 1369 |
+
return JSONResponse({"error": "Không đọc được bài viết"}, status_code=422)
|
| 1370 |
+
points = _extract_key_points_rw(data['paragraphs'], max_points=12)
|
| 1371 |
+
if not points:
|
| 1372 |
+
return JSONResponse({"error": "Không tìm được ý chính"}, status_code=422)
|
| 1373 |
+
images = data.get('images', [])
|
| 1374 |
+
slides = []
|
| 1375 |
+
for i, point in enumerate(points):
|
| 1376 |
+
img = images[i] if i < len(images) else (images[-1] if images else '')
|
| 1377 |
+
if img and 'cdnphoto.dantri' in img:
|
| 1378 |
+
img = '/api/proxy/img?url=' + _quote2(img, safe='')
|
| 1379 |
+
slides.append({'text': point, 'image': img, 'index': i + 1})
|
| 1380 |
+
summary_text = '\n\n'.join([f"• {s['text']}" for s in slides])
|
| 1381 |
+
|
| 1382 |
+
# Auto-detect language and emotion
|
| 1383 |
+
lang, emotion = detect_language_and_emotion(data['title'], summary_text)
|
| 1384 |
+
# Use preferred voice if provided, otherwise auto-detect
|
| 1385 |
+
voice = preferred_voice if preferred_voice else get_voice_for_content(data['title'], summary_text)
|
| 1386 |
+
|
| 1387 |
+
post = {
|
| 1388 |
+
"id": str(int(time.time() * 1000)) + str(_random2.randint(100, 999)),
|
| 1389 |
+
"title": data['title'],
|
| 1390 |
+
"text": summary_text,
|
| 1391 |
+
"img": images[0] if images else '',
|
| 1392 |
+
"url": url,
|
| 1393 |
+
"kind": "slide_summary",
|
| 1394 |
+
"slides": slides,
|
| 1395 |
+
"images": images[:10],
|
| 1396 |
+
"video": "",
|
| 1397 |
+
"voice": voice,
|
| 1398 |
+
"emotion": emotion,
|
| 1399 |
+
"language": lang,
|
| 1400 |
+
"ts": int(time.time())
|
| 1401 |
+
}
|
| 1402 |
+
posts = _load_wall_posts()
|
| 1403 |
+
posts.insert(0, post)
|
| 1404 |
+
_save_wall_posts(posts)
|
| 1405 |
+
return JSONResponse({"post": post, "slides": slides})
|
| 1406 |
+
|
| 1407 |
+
|
| 1408 |
+
@app.post("/api/rewrite_share")
|
| 1409 |
+
async def api_rewrite_share(request: Request):
|
| 1410 |
+
"""Rewrite article and post to Tường AI with SLIDES + AI text."""
|
| 1411 |
+
body = await request.json()
|
| 1412 |
+
url = _clean(body.get("url", ""))
|
| 1413 |
+
ctx = _clean(body.get("context", ""))
|
| 1414 |
+
preferred_voice = body.get("voice", "") # Accept custom voice selection
|
| 1415 |
+
if not url and not ctx:
|
| 1416 |
+
return JSONResponse({"error": "Cần URL hoặc nội dung"}, status_code=400)
|
| 1417 |
+
data = None
|
| 1418 |
+
if url and url.startswith("http"):
|
| 1419 |
+
data = _scrape_article_for_rewrite(url)
|
| 1420 |
+
if not data and ctx:
|
| 1421 |
+
paragraphs = [_clean(p) for p in ctx.split('\n') if len(_clean(p)) > 40]
|
| 1422 |
+
data = {'title': paragraphs[0][:80] if paragraphs else 'Bài viết', 'paragraphs': paragraphs, 'images': [], 'og_img': ''}
|
| 1423 |
+
if not data or not data.get('paragraphs'):
|
| 1424 |
+
return JSONResponse({"error": "Không đọc được bài viết"}, status_code=422)
|
| 1425 |
+
raw_text = '\n'.join(data['paragraphs'])
|
| 1426 |
+
if len(raw_text) < 50:
|
| 1427 |
+
raw_text = ctx[:14000]
|
| 1428 |
+
if len(raw_text) < 50:
|
| 1429 |
+
return JSONResponse({"error": "Bài viết quá ngắn"}, status_code=422)
|
| 1430 |
+
domain = ''
|
| 1431 |
+
try:
|
| 1432 |
+
from urllib.parse import urlparse
|
| 1433 |
+
domain = urlparse(url).netloc.replace('www.', '')
|
| 1434 |
+
except:
|
| 1435 |
+
pass
|
| 1436 |
+
|
| 1437 |
+
# Generate AI summary text
|
| 1438 |
+
ai_text = None
|
| 1439 |
+
try:
|
| 1440 |
+
import ai_ext
|
| 1441 |
+
if hasattr(ai_ext, 'qwen_generate'):
|
| 1442 |
+
prompt = f'Tóm tắt đăng Tường AI:\nTiêu đề: {data["title"]}\n{raw_text[:14000]}\n\n4-6 ý chính. Cuối ghi nguồn.'
|
| 1443 |
+
ai_text = await ai_ext.qwen_generate(prompt, max_tokens=1000)
|
| 1444 |
+
except Exception:
|
| 1445 |
+
pass
|
| 1446 |
+
if not ai_text or len(ai_text) < 80:
|
| 1447 |
+
key_pts = _extract_key_points_rw(data['paragraphs'], max_points=12)
|
| 1448 |
+
if key_pts:
|
| 1449 |
+
ai_text = '\n\n'.join([f"• {p}" for p in key_pts])
|
| 1450 |
+
else:
|
| 1451 |
+
ai_text = f"Tóm tắt: {data['title']}\n\n{raw_text[:1200]}\n\nNguồn: {domain}"
|
| 1452 |
+
|
| 1453 |
+
# Build slides from key points (FIX: include slides in rewrite_share too!)
|
| 1454 |
+
points = _extract_key_points_rw(data['paragraphs'], max_points=12)
|
| 1455 |
+
images = data.get('images', [])
|
| 1456 |
+
slides = []
|
| 1457 |
+
for i, point in enumerate(points):
|
| 1458 |
+
img = images[i] if i < len(images) else (images[-1] if images else '')
|
| 1459 |
+
if img and 'cdnphoto.dantri' in img:
|
| 1460 |
+
img = '/api/proxy/img?url=' + _quote2(img, safe='')
|
| 1461 |
+
slides.append({'text': point, 'image': img, 'index': i + 1})
|
| 1462 |
+
|
| 1463 |
+
# Auto-detect language and emotion
|
| 1464 |
+
lang, emotion = detect_language_and_emotion(data['title'], ai_text)
|
| 1465 |
+
# Use preferred voice if provided, otherwise auto-detect
|
| 1466 |
+
voice = preferred_voice if preferred_voice else get_voice_for_content(data['title'], ai_text)
|
| 1467 |
+
|
| 1468 |
+
post = {
|
| 1469 |
+
"id": str(int(time.time() * 1000)) + str(_random2.randint(100, 999)),
|
| 1470 |
+
"title": data['title'],
|
| 1471 |
+
"text": ai_text,
|
| 1472 |
+
"img": images[0] if images else '',
|
| 1473 |
+
"url": url,
|
| 1474 |
+
"kind": "rewrite",
|
| 1475 |
+
"slides": slides,
|
| 1476 |
+
"images": images[:10],
|
| 1477 |
+
"video": "",
|
| 1478 |
+
"voice": voice,
|
| 1479 |
+
"emotion": emotion,
|
| 1480 |
+
"language": lang,
|
| 1481 |
+
"ts": int(time.time())
|
| 1482 |
+
}
|
| 1483 |
+
posts = _load_wall_posts()
|
| 1484 |
+
posts.insert(0, post)
|
| 1485 |
+
_save_wall_posts(posts)
|
| 1486 |
+
return JSONResponse({"post": post, "slides": slides})
|
| 1487 |
+
|
| 1488 |
+
|
| 1489 |
+
@app.post("/api/url_wall")
|
| 1490 |
+
async def api_url_wall(request: Request):
|
| 1491 |
+
"""Submit URL to add to Tường AI."""
|
| 1492 |
+
body = await request.json()
|
| 1493 |
+
url = _clean(body.get("url", ""))
|
| 1494 |
+
if not url or not url.startswith('http'):
|
| 1495 |
+
return JSONResponse({"error": "URL không hợp lệ"}, status_code=400)
|
| 1496 |
+
# Reuse rewrite_share logic
|
| 1497 |
+
req._body = json.dumps({"url": url}).encode()
|
| 1498 |
+
return await api_rewrite_share(request)
|
| 1499 |
+
|
| 1500 |
+
|
| 1501 |
+
def _bg():
|
| 1502 |
+
time.sleep(15)
|
| 1503 |
+
while True:
|
| 1504 |
+
try:get_wc2026_all()
|
| 1505 |
+
except:pass
|
| 1506 |
+
time.sleep(90)
|
| 1507 |
+
threading.Thread(target=_bg,daemon=True).start()
|
| 1508 |
+
|
| 1509 |
+
# ===== AUTO SCHEDULER: rewrite AI + short at 7/13/19 VN time =====
|
| 1510 |
+
_AUTO_SCHEDULE_TIMES = [(7, '07:00'), (13, '13:00'), (19, '19:00')]
|
| 1511 |
+
_AUTO_LOG = os.path.join(DATA_DIR, 'auto_rewrite_log.json')
|
| 1512 |
+
|
| 1513 |
+
def _load_auto_log():
|
| 1514 |
+
try:
|
| 1515 |
+
if os.path.exists(_AUTO_LOG):
|
| 1516 |
+
with open(_AUTO_LOG, 'r') as f:
|
| 1517 |
+
return json.load(f)
|
| 1518 |
+
except: pass
|
| 1519 |
+
return {}
|
| 1520 |
+
|
| 1521 |
+
def _save_auto_log(log):
|
| 1522 |
+
try:
|
| 1523 |
+
tmp = _AUTO_LOG + '.tmp'
|
| 1524 |
+
with open(tmp, 'w') as f:
|
| 1525 |
+
json.dump(log, f)
|
| 1526 |
+
os.replace(tmp, _AUTO_LOG)
|
| 1527 |
+
except: pass
|
| 1528 |
+
|
| 1529 |
+
async def _auto_fetch_short(post_id):
|
| 1530 |
+
"""Try to auto-generate a short for a post."""
|
| 1531 |
+
try:
|
| 1532 |
+
import httpx
|
| 1533 |
+
async with httpx.AsyncClient(timeout=180) as cl:
|
| 1534 |
+
r = await cl.post(
|
| 1535 |
+
f"http://localhost:7860/api/ai/short/{post_id}",
|
| 1536 |
+
json={"voice":"vi-VN-HoaiMyNeural","emotion":"neutral","speed":1.2},
|
| 1537 |
+
headers={"Content-Type":"application/json"}
|
| 1538 |
+
)
|
| 1539 |
+
if r.status_code < 300:
|
| 1540 |
+
sj = r.json()
|
| 1541 |
+
if sj.get('video'):
|
| 1542 |
+
posts = _load_wall_posts()
|
| 1543 |
+
for p in posts:
|
| 1544 |
+
if p.get('id') == post_id:
|
| 1545 |
+
p['video'] = sj['video']
|
| 1546 |
+
break
|
| 1547 |
+
_save_wall_posts(posts)
|
| 1548 |
+
return True
|
| 1549 |
+
except: pass
|
| 1550 |
+
return False
|
| 1551 |
+
|
| 1552 |
+
async def _auto_rewrite_one(topic, slot_label, used_urls=None, post_index=0):
|
| 1553 |
+
"""Rewrite one topic: find articles, summarize, post to wall, trigger short.
|
| 1554 |
+
used_urls: shared set to avoid duplicate articles across topics.
|
| 1555 |
+
post_index: 0-based index to create multiple posts per topic (0,1,2 = up to 3 posts)."""
|
| 1556 |
+
from urllib.parse import quote as _q
|
| 1557 |
+
# Get MORE items to support 1-3 posts per topic
|
| 1558 |
+
items = _search_all(topic, limit=12)
|
| 1559 |
+
# Skip URLs already used by another topic
|
| 1560 |
+
if used_urls is not None:
|
| 1561 |
+
filtered = [it for it in items if it.get('url') not in used_urls]
|
| 1562 |
+
if filtered:
|
| 1563 |
+
items = filtered
|
| 1564 |
+
if not items or post_index >= len(items):
|
| 1565 |
+
return False
|
| 1566 |
+
|
| 1567 |
+
# Get article at post_index (0,1,2 for multiple posts)
|
| 1568 |
+
item = items[post_index] # post_index allows multiple articles per topic
|
| 1569 |
+
url = item.get('url', '')
|
| 1570 |
+
title = item.get('title', topic)
|
| 1571 |
+
if url and used_urls is not None:
|
| 1572 |
+
used_urls.add(url)
|
| 1573 |
+
if not url.startswith('http'):
|
| 1574 |
+
return False
|
| 1575 |
+
|
| 1576 |
+
data = _scrape_article_for_rewrite(url)
|
| 1577 |
+
if not data or not data.get('paragraphs'):
|
| 1578 |
+
return False
|
| 1579 |
+
|
| 1580 |
+
raw_text = '\n'.join(data['paragraphs'])
|
| 1581 |
+
ai_text = None
|
| 1582 |
+
|
| 1583 |
+
# Try AI generation
|
| 1584 |
+
try:
|
| 1585 |
+
import ai_ext
|
| 1586 |
+
prompt = f"Tóm tắt tin tức (tự động {slot_label}):\nTiêu đề: {data['title']}\n{raw_text[:10000]}\n\n4-6 ý chính dạng bullet. Cuối ghi nguồn."
|
| 1587 |
+
ai_text = await ai_ext.qwen_generate(prompt, max_tokens=1000)
|
| 1588 |
+
except: pass
|
| 1589 |
+
|
| 1590 |
+
if not ai_text or len(ai_text) < 80:
|
| 1591 |
+
pts = data['paragraphs'][:6]
|
| 1592 |
+
ai_text = '\n\n'.join([f"• {p[:300]}" for p in pts])
|
| 1593 |
+
via = item.get('via', '') or urlparse(url).netloc.replace('www.', '')
|
| 1594 |
+
ai_text += f"\n\nNguồn tham khảo: {via}"
|
| 1595 |
+
|
| 1596 |
+
# Build slides
|
| 1597 |
+
images = data.get('images', [])
|
| 1598 |
+
pts = data['paragraphs'][:10]
|
| 1599 |
+
slides = []
|
| 1600 |
+
for i, p in enumerate(pts[:8]):
|
| 1601 |
+
img = images[i] if i < len(images) else (images[-1] if images else data.get('og_img', ''))
|
| 1602 |
+
slides.append({'text': p[:300], 'image': img, 'index': i + 1})
|
| 1603 |
+
|
| 1604 |
+
post_id = str(int(time.time() * 1000)) + str(_random2.randint(100, 999))
|
| 1605 |
+
post = {
|
| 1606 |
+
"id": post_id, "title": data.get('title', title)[:200],
|
| 1607 |
+
"text": ai_text, "img": images[0] if images else data.get('og_img', ''),
|
| 1608 |
+
"url": url, "kind": "auto_rewrite", "slides": slides,
|
| 1609 |
+
"images": images[:10], "video": "",
|
| 1610 |
+
"voice": "vi-VN-HoaiMyNeural", "emotion": "neutral",
|
| 1611 |
+
"language": "vietnamese", "ts": int(time.time()),
|
| 1612 |
+
"auto_scheduled": True, "slot": slot_label,
|
| 1613 |
+
}
|
| 1614 |
+
|
| 1615 |
+
posts = _load_wall_posts()
|
| 1616 |
+
posts.insert(0, post)
|
| 1617 |
+
_save_wall_posts(posts)
|
| 1618 |
+
|
| 1619 |
+
# Trigger short generation async
|
| 1620 |
+
threading.Thread(target=lambda: asyncio.run(_auto_fetch_short(post_id)), daemon=True).start()
|
| 1621 |
+
return True
|
| 1622 |
+
|
| 1623 |
+
async def _do_scheduled_run(slot_label):
|
| 1624 |
+
"""Main scheduled run: 1-3 posts from 3 different HOT topics (3-9 total), no duplicates."""
|
| 1625 |
+
print(f"[auto] Starting scheduled rewrite for {slot_label}")
|
| 1626 |
+
|
| 1627 |
+
# Get top hot topics, skip duplicates
|
| 1628 |
+
all_topics = _get_hot_topics()
|
| 1629 |
+
seen_topics = set()
|
| 1630 |
+
unique_topics = []
|
| 1631 |
+
for t in all_topics:
|
| 1632 |
+
kw = t.get('topic', '').lower().strip()
|
| 1633 |
+
if kw and len(kw) > 5 and kw not in seen_topics:
|
| 1634 |
+
is_dup = False
|
| 1635 |
+
for s in seen_topics:
|
| 1636 |
+
# Check if one topic is substring of another
|
| 1637 |
+
if kw in s or s in kw:
|
| 1638 |
+
is_dup = True
|
| 1639 |
+
break
|
| 1640 |
+
if not is_dup:
|
| 1641 |
+
seen_topics.add(kw)
|
| 1642 |
+
unique_topics.append(t)
|
| 1643 |
+
if len(unique_topics) >= 3:
|
| 1644 |
+
break
|
| 1645 |
+
|
| 1646 |
+
job_topics = [t['topic'] for t in unique_topics[:3] if t.get('topic')]
|
| 1647 |
+
if not job_topics:
|
| 1648 |
+
print(f"[auto] No hot topics found, skipping")
|
| 1649 |
+
return
|
| 1650 |
+
|
| 1651 |
+
print(f"[auto] Running 3 topics: {job_topics}")
|
| 1652 |
+
|
| 1653 |
+
# Track used URLs to avoid cross-topic duplicates
|
| 1654 |
+
_used_urls = set()
|
| 1655 |
+
results = []
|
| 1656 |
+
|
| 1657 |
+
# Process each topic, create 1-3 posts per topic
|
| 1658 |
+
for jt in job_topics:
|
| 1659 |
+
for post_idx in range(3): # Try up to 3 posts per topic
|
| 1660 |
+
try:
|
| 1661 |
+
ok = await asyncio.wait_for(_auto_rewrite_one(jt, slot_label, _used_urls, post_idx), timeout=120)
|
| 1662 |
+
if ok:
|
| 1663 |
+
results.append((jt, post_idx, True))
|
| 1664 |
+
print(f"[auto] Created post {post_idx+1} for '{jt}'")
|
| 1665 |
+
else:
|
| 1666 |
+
# No more articles for this topic
|
| 1667 |
+
break
|
| 1668 |
+
except Exception as e:
|
| 1669 |
+
print(f"[auto] Error on '{jt}' post {post_idx}: {e}")
|
| 1670 |
+
results.append((jt, post_idx, False))
|
| 1671 |
+
await asyncio.sleep(1) # Small delay between posts
|
| 1672 |
+
|
| 1673 |
+
# Ensure at least 3 posts total (fallback if needed)
|
| 1674 |
+
successful_posts = sum(1 for _, _, ok in results if ok)
|
| 1675 |
+
print(f"[auto] Done {slot_label}: {successful_posts} posts created")
|
| 1676 |
+
|
| 1677 |
+
# Log
|
| 1678 |
+
from datetime import datetime, timezone, timedelta
|
| 1679 |
+
VN_TZ_SCHED = timezone(timedelta(hours=7))
|
| 1680 |
+
today_str = datetime.now(VN_TZ_SCHED).strftime('%Y-%m-%d')
|
| 1681 |
+
log = _load_auto_log()
|
| 1682 |
+
if today_str not in log: log[today_str] = {}
|
| 1683 |
+
log[today_str][slot_label] = {
|
| 1684 |
+
'time': datetime.now(VN_TZ_SCHED).strftime('%H:%M:%S'),
|
| 1685 |
+
'count': successful_posts,
|
| 1686 |
+
'total': len(job_topics),
|
| 1687 |
+
}
|
| 1688 |
+
_save_auto_log(log)
|
| 1689 |
+
|
| 1690 |
+
def _scheduler_loop():
|
| 1691 |
+
"""Check every 60s; trigger at 7:00, 13:00, 19:00 VN time.
|
| 1692 |
+
On startup, check for any missed slots today and run them immediately."""
|
| 1693 |
+
time.sleep(35)
|
| 1694 |
+
from datetime import datetime, timezone, timedelta
|
| 1695 |
+
VN_TZ_SCHED = timezone(timedelta(hours=7))
|
| 1696 |
+
|
| 1697 |
+
_last_run_date = ""
|
| 1698 |
+
_last_run_slots = set()
|
| 1699 |
+
|
| 1700 |
+
# On startup: check log for missed slots today
|
| 1701 |
+
try:
|
| 1702 |
+
start_now = datetime.now(VN_TZ_SCHED)
|
| 1703 |
+
today_str = start_now.strftime('%Y-%m-%d')
|
| 1704 |
+
current_hour = start_now.hour
|
| 1705 |
+
current_minute = start_now.minute
|
| 1706 |
+
log = _load_auto_log()
|
| 1707 |
+
today_log = log.get(today_str, {})
|
| 1708 |
+
for h, label in _AUTO_SCHEDULE_TIMES:
|
| 1709 |
+
# Run if slot is past (either strictly earlier hour, or same hour but window has passed)
|
| 1710 |
+
should_run = False
|
| 1711 |
+
if h < current_hour:
|
| 1712 |
+
should_run = True
|
| 1713 |
+
elif h == current_hour and current_minute > 10:
|
| 1714 |
+
should_run = True
|
| 1715 |
+
if should_run and label not in today_log:
|
| 1716 |
+
print(f"[auto] Detected missed slot {label} (h={h} < now={current_hour}:{current_minute}), running catch-up now")
|
| 1717 |
+
_run_scheduled_sync(label)
|
| 1718 |
+
_last_run_slots.add(label)
|
| 1719 |
+
except Exception as e:
|
| 1720 |
+
print(f"[auto] Catch-up check error: {e}")
|
| 1721 |
+
|
| 1722 |
+
while True:
|
| 1723 |
+
try:
|
| 1724 |
+
now = datetime.now(VN_TZ_SCHED)
|
| 1725 |
+
today = now.strftime('%Y-%m-%d')
|
| 1726 |
+
hour = now.hour
|
| 1727 |
+
minute = now.minute
|
| 1728 |
+
|
| 1729 |
+
if today != _last_run_date:
|
| 1730 |
+
_last_run_date = today
|
| 1731 |
+
_last_run_slots = set()
|
| 1732 |
+
|
| 1733 |
+
slot = None
|
| 1734 |
+
for h, label in _AUTO_SCHEDULE_TIMES:
|
| 1735 |
+
if hour == h and 0 <= minute < 5:
|
| 1736 |
+
slot = label
|
| 1737 |
+
break
|
| 1738 |
+
|
| 1739 |
+
if slot and slot not in _last_run_slots:
|
| 1740 |
+
_last_run_slots.add(slot)
|
| 1741 |
+
_run_scheduled_sync(slot)
|
| 1742 |
+
except Exception as e:
|
| 1743 |
+
print(f"[auto] Loop error: {e}")
|
| 1744 |
+
|
| 1745 |
+
time.sleep(60)
|
| 1746 |
+
|
| 1747 |
+
threading.Thread(target=_scheduler_loop, daemon=True, name='auto-rewrite-scheduler').start()
|
| 1748 |
+
|
| 1749 |
+
@app.get('/api/debug/auto_schedule')
|
| 1750 |
+
async def debug_auto_schedule(slot: str = '07:00'):
|
| 1751 |
+
"""Manually trigger auto scheduler for debugging."""
|
| 1752 |
+
try:
|
| 1753 |
+
# Check if we can access the data directory
|
| 1754 |
+
log = _load_auto_log()
|
| 1755 |
+
topics = _get_hot_topics()[:3]
|
| 1756 |
+
job_topics = [t['topic'] for t in topics if t.get('topic')]
|
| 1757 |
+
return JSONResponse({
|
| 1758 |
+
"slot": slot,
|
| 1759 |
+
"log": log,
|
| 1760 |
+
"hot_topics": job_topics,
|
| 1761 |
+
"wall_posts_count": len(_load_wall_posts()),
|
| 1762 |
+
"data_dir_writable": os.access(DATA_DIR, os.W_OK) if os.path.isdir(DATA_DIR) else False,
|
| 1763 |
+
"data_dir_exists": os.path.isdir(DATA_DIR),
|
| 1764 |
+
})
|
| 1765 |
+
except Exception as e:
|
| 1766 |
+
return JSONResponse({"error": str(e)}, status_code=500)
|
| 1767 |
+
|
| 1768 |
+
def _run_scheduled_sync(slot):
|
| 1769 |
+
"""Run _do_scheduled_run in a separate event loop (for background thread)."""
|
| 1770 |
+
loop = asyncio.new_event_loop()
|
| 1771 |
+
asyncio.set_event_loop(loop)
|
| 1772 |
+
try:
|
| 1773 |
+
loop.run_until_complete(_do_scheduled_run(slot))
|
| 1774 |
+
except Exception as e:
|
| 1775 |
+
print(f"[auto] Background run error: {e}")
|
| 1776 |
+
finally:
|
| 1777 |
+
loop.close()
|
| 1778 |
+
|
| 1779 |
+
@app.get('/api/debug/trigger_auto')
|
| 1780 |
+
async def debug_trigger_auto(slot: str = '19:00'):
|
| 1781 |
+
"""Trigger _do_scheduled_run in background thread (non-blocking)."""
|
| 1782 |
+
threading.Thread(target=_run_scheduled_sync, args=(slot,), daemon=True).start()
|
| 1783 |
+
return JSONResponse({"status": "started", "slot": slot})
|
| 1784 |
+
|
| 1785 |
+
# ===== SHORTS RSS PROXY ENDPOINT =====
|
| 1786 |
+
@app.get("/api/shorts/rss")
|
| 1787 |
+
def shorts_rss():
|
| 1788 |
+
"""Get shorts from YouTube RSS feeds server-side"""
|
| 1789 |
+
import xml.etree.ElementTree as ET
|
| 1790 |
+
import html as html_lib2
|
| 1791 |
+
import re as re2
|
| 1792 |
+
|
| 1793 |
+
YOUTUBE_CHANNELS = {
|
| 1794 |
+
"baodantri7941": "UC_x5TKhOgd6GhYvv5z4I3jg",
|
| 1795 |
+
"baosuckhoedoisongboyte": "UCBsY5fXTQLkF_JnH9kLkL4g",
|
| 1796 |
+
}
|
| 1797 |
+
|
| 1798 |
+
shorts = []
|
| 1799 |
+
seen = set()
|
| 1800 |
+
|
| 1801 |
+
for handle, channel_id in YOUTUBE_CHANNELS.items():
|
| 1802 |
+
try:
|
| 1803 |
+
rss_url = f"https://www.youtube.com/feeds/videos.xml?channel_id={channel_id}"
|
| 1804 |
+
r = req.get(rss_url, headers=HEADERS, timeout=15)
|
| 1805 |
+
if r.status_code != 200:
|
| 1806 |
+
continue
|
| 1807 |
+
|
| 1808 |
+
root = ET.fromstring(r.text)
|
| 1809 |
+
ns = {
|
| 1810 |
+
'atom': 'http://www.w3.org/2005/Atom',
|
| 1811 |
+
'yt': 'http://www.youtube.com/xml/schemas/2015',
|
| 1812 |
+
'media': 'http://search.yahoo.com/mrss/'
|
| 1813 |
+
}
|
| 1814 |
+
|
| 1815 |
+
for entry in root.findall('atom:entry', ns)[:30]:
|
| 1816 |
+
title_el = entry.find('atom:title', ns)
|
| 1817 |
+
title = html_lib2.unescape(title_el.text) if title_el is not None and title_el.text else ''
|
| 1818 |
+
|
| 1819 |
+
link_el = entry.find('atom:link', ns)
|
| 1820 |
+
link = link_el.get('href', '') if link_el is not None else ''
|
| 1821 |
+
|
| 1822 |
+
vid_el = entry.find('yt:videoId', ns)
|
| 1823 |
+
vid = vid_el.text if vid_el is not None else ''
|
| 1824 |
+
|
| 1825 |
+
if not vid or vid in seen:
|
| 1826 |
+
continue
|
| 1827 |
+
|
| 1828 |
+
# Check if it's a short
|
| 1829 |
+
is_short = '#shorts' in title.lower() or '#short' in title.lower() or '/shorts/' in link
|
| 1830 |
+
|
| 1831 |
+
if not is_short:
|
| 1832 |
+
desc_el = entry.find('media:description', ns)
|
| 1833 |
+
if desc_el is not None and desc_el.text:
|
| 1834 |
+
if '#shorts' in desc_el.text.lower():
|
| 1835 |
+
is_short = True
|
| 1836 |
+
|
| 1837 |
+
if not is_short:
|
| 1838 |
+
continue
|
| 1839 |
+
|
| 1840 |
+
seen.add(vid)
|
| 1841 |
+
|
| 1842 |
+
# Get thumbnail
|
| 1843 |
+
thumb = f"https://i.ytimg.com/vi/{vid}/hqdefault.jpg"
|
| 1844 |
+
media_group = entry.find('media:group', ns)
|
| 1845 |
+
if media_group is not None:
|
| 1846 |
+
thumb_el = media_group.find('media:thumbnail', ns)
|
| 1847 |
+
if thumb_el is not None:
|
| 1848 |
+
thumb = thumb_el.get('url', thumb)
|
| 1849 |
+
|
| 1850 |
+
shorts.append({
|
| 1851 |
+
'id': vid,
|
| 1852 |
+
'title': title.replace('#shorts', '').replace('#short', '').strip()[:120],
|
| 1853 |
+
'img': thumb,
|
| 1854 |
+
'link': f'https://www.youtube.com/shorts/{vid}',
|
| 1855 |
+
'channel': handle,
|
| 1856 |
+
'source': 'yt'
|
| 1857 |
+
})
|
| 1858 |
+
|
| 1859 |
+
if len(shorts) >= 40:
|
| 1860 |
+
break
|
| 1861 |
+
|
| 1862 |
+
except Exception as e:
|
| 1863 |
+
print(f"RSS error for {handle}: {e}")
|
| 1864 |
+
continue
|
| 1865 |
+
|
| 1866 |
+
return {"shorts": shorts, "count": len(shorts)}
|
| 1867 |
+
|
| 1868 |
+
app.mount('/static',StaticFiles(directory=STATIC_DIR),name='vnews_static')
|
static/app_v2.js
CHANGED
|
@@ -5,7 +5,6 @@
|
|
| 5 |
* readSlidePost has share button, /s + /s/{slug} render slides/video inline
|
| 6 |
* v2.5 - Added 'Tạo lại' button on wall cards with video
|
| 7 |
* v2.4 - Fixed: prependWallPost detached-element bug, makeShortVideo UI update, slide viewer for rewrite posts
|
| 8 |
-
* v2.9 - ADDED: rewriteUrl() function for the URL rewrite button on homepage (was missing!)
|
| 9 |
*/
|
| 10 |
function _proxyImg(url){
|
| 11 |
if(!url || typeof url !== 'string') return '';
|
|
@@ -147,8 +146,8 @@ function makeWallItem(p,i){
|
|
| 147 |
+'</select>'
|
| 148 |
+'</div>';
|
| 149 |
var makeBtn = hasVideo
|
| 150 |
-
|
| 151 |
-
|
| 152 |
return '<div class="wall-item" id="wall-item-'+esc(vid)+'"><div class="wall-thumb">'+thumbContent+videoBadge+'</div><div class="wall-title">'+esc(p.title)+'</div><div class="wall-text">'+esc((p.text||'').slice(0,180))+'</div>'+voiceBar+'<div class="wall-actions"><button class="primary" onclick="readWallPost('+i+')">Xem</button>'+makeBtn+'</div></div>';
|
| 153 |
}
|
| 154 |
|
|
@@ -386,10 +385,6 @@ function searchTopic(topic){if(!topic){topic=document.getElementById('topic-inpu
|
|
| 386 |
async function showHashtagSources(topic,page){const box=document.getElementById('hashtag-box');if(!box)return;if(page===0)box.innerHTML=`<div class="hashtag-sources"><h3>🔍 ${esc(topic)}</h3><div class="hashtag-loading"><div class="hashtag-spinner"></div>Đang tìm...</div></div>`;try{const r=await fetch(`/api/hashtag/sources?topic=${encodeURIComponent(topic)}&page=${page}`);const j=await r.json();const sources=j.sources||[];if(!sources.length&&page===0){box.innerHTML=`<div class="hashtag-sources"><h3>🔍 ${esc(topic)}</h3><div style="color:#888;padding:8px">Không tìm được bài viết liên quan</div></div>`;return;}let h='';if(page===0)h=`<div class="hashtag-sources"><h3>🔍 ${esc(topic)} <span style="font-size:10px;color:#888">(${j.total} bài từ 8 nguồn)</span></h3><div id="ht-list">`;sources.forEach((s,i)=>{const idx=page*8+i;h+=`<div class="hashtag-src-item" onclick="readArticle('${esc(s.url)}')"><div class="hashtag-src-img" id="ht-img-${idx}"></div><div class="hashtag-src-text"><div class="hashtag-src-title">${esc(s.title)}</div><div class="hashtag-src-via">${esc(s.via||'')}</div></div></div>`;});if(page===0){h+=`</div><button class="hashtag-rewrite-btn" onclick="rewriteHashtag('${esc(topic).replace(/'/g,"\\'")}')">🤖 Rewrite AI tổng hợp & đăng tường</button>`;if(j.has_more)h+=`<button class="hashtag-load-more" id="ht-more" onclick="loadMoreHashtag()">Tải thêm ▼</button>`;h+=`</div>`;box.innerHTML=h;}else{document.getElementById('ht-list')?.insertAdjacentHTML('beforeend',h);const btn=document.getElementById('ht-more');if(btn){if(!j.has_more)btn.remove();else{btn.disabled=false;btn.textContent='Tải thêm ▼';}}}sources.forEach((s,i)=>{const idx=page*8+i;if(!s.url)return;const ctrl=new AbortController();setTimeout(()=>ctrl.abort(),4000);fetch('/api/article?url='+encodeURIComponent(s.url),{signal:ctrl.signal}).then(r=>r.json()).then(d=>{if(d&&(d.og_image||d.img)){const el=document.getElementById('ht-img-'+idx);if(el)el.innerHTML=`<img src="${_proxyImg(esc(d.og_image||d.img))}" onerror="this.style.display='none'">`;}}).catch(()=>{});});}catch(e){box.innerHTML=`<div class="hashtag-sources"><h3>🔍 ${esc(topic)}</h3><div style="color:#e74c3c;padding:8px">Lỗi: ${esc(e.message)}</div></div>`;}}
|
| 387 |
function loadMoreHashtag(){_htPage++;const btn=document.getElementById('ht-more');if(btn){btn.disabled=true;btn.textContent='Đang tải...';}showHashtagSources(_htTopic,_htPage);}
|
| 388 |
async function rewriteHashtag(topic){const btn=event?.target;if(btn){btn.disabled=true;btn.textContent='Đang tổng hợp...';}try{const r=await fetch('/api/topic_post',{method:'POST',headers:{'Content-Type':'application/json'},body:JSON.stringify({topic})});const j=await r.json();if(!r.ok||j.error)throw new Error(j.error||'Lỗi');toast('✅ Đã đăng Tường AI!');if(btn)btn.textContent='✅ Đăng thành công!';if(j.post)prependWallPost(j.post);}catch(e){toast('❌ '+e.message);if(btn){btn.disabled=false;btn.textContent='🤖 Rewrite AI';}}}
|
| 389 |
-
|
| 390 |
-
/** rewriteUrl: called when user clicks "Rewrite" button next to URL input on homepage */
|
| 391 |
-
async function rewriteUrl(){const inp=document.getElementById('url-input');if(!inp)return;const url=inp.value.trim();if(!url){alert('Dán URL bài viết trước');return}if(!/^https?:\/\//i.test(url)){alert('URL cần bắt đầu bằng http:// hoặc https://');return}inp.value='';let btn=event?.target;if(btn){btn.disabled=true;btn.textContent='⏳ Đang xử lý...';}toast('⏳ Đang phân tích URL...');try{const r=await fetch('/api/rewrite_share',{method:'POST',headers:{'Content-Type':'application/json'},body:JSON.stringify({url})});const j=await r.json();if(!r.ok||j.error)throw new Error(j.error||'Lỗi');toast('✅ Đã tạo rewrite! Đang tạo Short AI...');if(btn)btn.textContent='✅ Đang tạo Short AI...';if(j.post){j.post.slides=j.slides||[];prependWallPost(j.post);}if(j.slides&&j.slides.length){let slideDiv=document.createElement('div');slideDiv.style.cssText='margin:8px 4px;background:#141414;border:1px solid #2b2b2b;border-radius:10px;padding:10px';let sh='<div style="font-size:13px;color:#5cb87a;font-weight:800;margin-bottom:8px">📊 Slides đã tạo ('+j.slides.length+' ý)</div>';j.slides.forEach(function(s){sh+='<div style="padding:6px 0;border-bottom:1px solid #222;font-size:11px;color:#ccc;line-height:1.4">• '+esc(s.text.slice(0,120))+'</div>';});slideDiv.innerHTML=sh;const target=document.getElementById('hashtag-box');if(target)target.parentNode.insertBefore(slideDiv,target.nextSibling);}if(btn)btn.textContent='✅ Hoàn tất!';if(j.post&&j.post.id){setTimeout(async function(){try{const sr=await fetch('/api/ai/short/'+encodeURIComponent(j.post.id),{method:'POST',headers:{'Content-Type':'application/json'},body:JSON.stringify({voice:'vi-VN-HoaiMyNeural',emotion:'neutral',speed:1.2})});const sj=await sr.json();if(sj&&sj.video){toast('✅ Đã tạo Short AI!');const p=_wallPosts.find(x=>String(x.id)===String(j.post.id));if(p){p.video=sj.video;const idx=_wallPosts.indexOf(p);const itemId='wall-item-'+j.post.id;const el=document.getElementById(itemId);if(el){el.insertAdjacentHTML('afterend',makeWallItem(p,idx));el.remove();}}}else{toast('✅ Rewrite xong! Chưa tạo được Short, thử lại sau.')}}catch(e){toast('✅ Rewrite xong! Chưa tạo được Short.')}},1000);}setTimeout(function(){if(btn){btn.disabled=false;btn.textContent='Rewrite';}},5000);}catch(e){toast('❌ '+e.message);if(btn){btn.disabled=false;btn.textContent='Rewrite';}}}
|
| 392 |
-
|
| 393 |
async function loadLivescore(tab){document.querySelectorAll('.ls-tab').forEach(t=>t.classList.remove('active'));document.querySelector(`.ls-tab[data-tab="${tab}"]`)?.classList.add('active');const el=document.getElementById('ls-content');if(!el)return;el.innerHTML='<div class="loading">Đang tải...</div>';let ep='/api/livescore/'+tab;if(tab.startsWith('bxh_'))ep='/api/livescore/standings/'+tab.replace('bxh_','');try{const r=await fetch(ep);const d=await r.json();el.innerHTML=d.html&&d.html.length>50?d.html:'<div class="loading">Không có dữ liệu</div>';bindMatchClicks(el);}catch(e){el.innerHTML='<div class="loading">Lỗi</div>';}}
|
| 394 |
function bindMatchClicks(el){el.querySelectorAll('.match-detail').forEach(md=>{md.style.cursor='pointer';md.addEventListener('click',function(e){const statusA=this.querySelector('.status a');const teamA=this.querySelector('.teams a[href*="/tran-dau/"]');const a = statusA || teamA;if(a){e.preventDefault();e.stopPropagation();const href=a.getAttribute('href')||'';const m=href.match(/\/tran-dau\/(\d+)\//);if(m){const fullUrl=href.startsWith('http')?href:'https://bongda.com.vn'+href;openMatch(m[1],fullUrl);}}});});el.querySelectorAll('a').forEach(a=>{a.addEventListener('click',e=>{e.preventDefault();e.stopPropagation()});});}
|
| 395 |
function openMatch(id,url){if(!id)return;_currentEventId=id;if(url)_currentMatchUrl=url;document.getElementById('match-overlay').classList.add('active');document.body.style.overflow='hidden';loadMatchTab('detail')}
|
|
@@ -460,9 +455,9 @@ function readWallPost(idx){const p=_wallPosts&&_wallPosts[idx];if(!p)return;if(p
|
|
| 460 |
/** Show rewrite slide viewer - vertical slides with text+image */
|
| 461 |
function readSlidePost(idx){const p=_wallPosts[idx];if(!p||!p.slides)return;showView('view-article');const el=document.getElementById('view-article');let h=`<button class="back-btn" onclick="switchCat('home')">← Tường AI</button><div class="slide-viewer" style="padding:12px;max-width:600px;margin:0 auto">`;p.slides.forEach((s,i)=>{h+=`<div class="slide-card" style="background:#1a1a1a;border:1px solid #2a2a2a;border-radius:12px;padding:16px;margin-bottom:12px"><div class="slide-num" style="color:#5cb87a;font-size:12px;font-weight:700;margin-bottom:6px">Slide ${s.index||i+1}/${p.slides.length}</div>${s.image?`<img src="${_proxyImg(s.image)}" style="width:100%;max-height:300px;object-fit:cover;border-radius:8px;margin-bottom:8px" loading="lazy" onerror="this.style.display=\'none\'">`:''}<p style="color:#ddd;font-size:14px;line-height:1.6;margin:0">${esc(s.text)}</p></div>`;});h+=`</div><div class="article-actions" style="padding:12px;text-align:center"><button onclick="doShare('${esc(p.title)}','${esc(p.url||'')}','${esc(p.img||'')}','${esc(p.id||'')}')" style="background:#2d8659;border:0;color:#fff;padding:8px 16px;border-radius:10px;cursor:pointer;font-size:12px">📤 Chia sẻ bài viết này</button></div>`;el.innerHTML=h;}
|
| 462 |
function readNewsTab(tab){loadNewsTab();}
|
| 463 |
-
function loadNewsTab(){const el=document.getElementById('view-cat');if(!el)return;el.innerHTML='<div class="loading">Đang tải tin tức...</div>';fetch('/api/homepage').then(r=>r.json()).then(articles=>{if(!articles||!articles.length){el.innerHTML='<div class="loading">Không có tin</div>';return}let h='<div class="grid">';articles.forEach(a=>{const src=a.source||'vne';const badge=a.group||a.source||'';h+=`<div class="card" onclick="readArticle('${esc(a.link)}')"><div class="card-img">${a.img?`<img src="${_proxyImg(a.img)}" loading="lazy" onerror="this.style.display='none'">`:''}</div><div class="card-body"><span class="badge badge-${src}">${esc(badge)}</span><div class="card-title">${esc(a.title)}</div></div></div>`;});h+='</div>';el.innerHTML=h;}).catch(()=>{el.innerHTML='<div class="loading">Lỗi tải</div>';});}
|
| 464 |
|
| 465 |
-
function readArticle(url,title,img,presetTitle,presetText){showView('view-article');const el=document.getElementById('view-article');el.innerHTML='<div class="loading">Đang tải...</div><button class="back-btn" onclick="switchCat(\'home\')">← Quay lại</button>';if(presetTitle){el.innerHTML=`<button class="back-btn" onclick="switchCat('home')">← Quay lại</button><div class="article-view"><h1 class="article-title">${esc(presetTitle)}</h1>${presetText?`<div class="article-summary" style="white-space:pre-wrap">${esc(presetText)}</div>`:''}</div><div class="article-actions"><button onclick="doShare('${esc(presetTitle)}','${esc(url||'')}','')">📤 Chia sẻ</button></div>`;return;}if(!url)return;fetch('/api/article?url='+encodeURIComponent(url)).then(r=>r.json()).then(d=>{let h=`<button class="back-btn" onclick="switchCat('home')">← Quay lại</button><div class="article-view">`;if(d.title)h+=`<h1 class="article-title">${esc(d.title)}</h1>`;if(d.summary)h+=`<div class="article-summary">${esc(d.summary)}</div>`;if(d.body)d.body.forEach(b=>{if(b.type==='p')h+=`<p class="article-p">${esc(b.text)}</p>`;else if(b.type==='heading')h+=`<h2 class="article-h2">${esc(b.text)}</h2>`;else if(b.type==='img'&&b.src)h+=`<img class="article-img" src="${_proxyImg(b.src)}" loading="lazy" onerror="this.style.display='none'">`;});h+=`</div><div class="article-actions"><button onclick="doShare('${esc(d.title||'')}','${esc(url)}','${esc(d.og_image||'')}')">📤 Chia sẻ</button><button class="primary" onclick="rewriteSlide('${esc(url)}')">🤖 Slide Rewrite AI</button></div>`;el.innerHTML=h;}).catch(()=>{el.innerHTML=`<button class="back-btn" onclick="switchCat('home')">← Quay lại</button><div class="article-view"><p>Không thể tải bài viết</p></div>`;});}
|
| 466 |
async function rewriteSlide(url){if(!url)return;const btn=document.querySelector('.article-actions .primary')||event?.target;if(btn){btn.disabled=true;btn.textContent='⏳ Đang tạo slides...';}toast('⏳ Đang tạo slide rewrite...');try{const r=await fetch('/api/rewrite_share',{method:'POST',headers:{'Content-Type':'application/json'},body:JSON.stringify({url})});const j=await r.json();if(!r.ok||j.error)throw new Error(j.error||'Lỗi');toast('✅ Đã tạo slide! Đang tạo Short AI...');if(btn)btn.textContent='✅ Đang tạo Short AI...';if(j.post){j.post.slides=j.slides||[];prependWallPost(j.post);}// Show slides immediately if available
|
| 467 |
if(j.slides && j.slides.length){showView('view-article');const el=document.getElementById('view-article');let h=`<button class="back-btn" onclick="switchCat('home')">← Tường AI</button><div class="slide-viewer" style="padding:12px;max-width:600px;margin:0 auto">`;j.slides.forEach(s=>{h+=`<div class="slide-card" style="background:#1a1a1a;border:1px solid #2a2a2a;border-radius:12px;padding:16px;margin-bottom:12px"><div class="slide-num" style="color:#5cb87a;font-size:12px;font-weight:700;margin-bottom:6px">Slide ${s.index}/${j.slides.length}</div>${s.image?`<img src="${s.image.startsWith('/api/')?s.image:'/api/proxy/img?url='+encodeURIComponent(s.image)}" style="width:100%;max-height:300px;object-fit:cover;border-radius:8px;margin-bottom:8px" loading="lazy" onerror="this.style.display='none'">`:''}<p style="color:#ddd;font-size:14px;line-height:1.6;margin:0">${esc(s.text)}</p></div>`;});h+=`</div><div style="text-align:center;padding:12px"><div id="short-ai-status" style="color:#888;font-size:12px">⏳ Đang tạo video Short AI...</div></div>`;el.innerHTML=h;}
|
| 468 |
const postId=j.post&&j.post.id;if(postId){setTimeout(async()=>{const sr=await fetch('/api/ai/short/'+encodeURIComponent(postId),{method:'POST',headers:{'Content-Type':'application/json'},body:JSON.stringify({voice:'vi-VN-HoaiMyNeural',emotion:'neutral',speed:1.2})});const sj=await sr.json();if(sr.ok&&sj.video&&typeof _wallPosts!=='undefined'){const p=_wallPosts.find(x=>String(x.id)===String(postId));if(p){p.video=sj.video;const itemId='wall-item-'+postId;const el2=document.getElementById(itemId);if(el2){const idx=_wallPosts.indexOf(p);el2.insertAdjacentHTML('afterend',makeWallItem(p,idx));el2.remove();}}toast('✅ Short AI đã sẵn sàng!');const statusEl=document.getElementById('short-ai-status');if(statusEl)statusEl.innerHTML='✅ Short AI đẵn sàng! <button class="primary" onclick="openShortAIFeed(0)" style="background:#2d8659;border:0;color:#fff;padding:6px 12px;border-radius:10px;font-size:11px;cursor:pointer;margin-left:8px">▶ Xem ngay</button>';}},500);}if(btn)btn.textContent='✅ Hoàn tất';}catch(e){toast('❌ '+e.message);if(btn){btn.disabled=false;btn.textContent='🤖 Slide Rewrite AI';}}}
|
|
|
|
| 5 |
* readSlidePost has share button, /s + /s/{slug} render slides/video inline
|
| 6 |
* v2.5 - Added 'Tạo lại' button on wall cards with video
|
| 7 |
* v2.4 - Fixed: prependWallPost detached-element bug, makeShortVideo UI update, slide viewer for rewrite posts
|
|
|
|
| 8 |
*/
|
| 9 |
function _proxyImg(url){
|
| 10 |
if(!url || typeof url !== 'string') return '';
|
|
|
|
| 146 |
+'</select>'
|
| 147 |
+'</div>';
|
| 148 |
var makeBtn = hasVideo
|
| 149 |
+
? '<button class="wall-btn-video" onclick="event.stopPropagation();openShortAIFeed('+i+')">▶ Xem Short</button><button class="wall-btn-make" style="margin-left:4px" onclick="event.stopPropagation();makeShortVideo(\''+esc(vid)+'\',this,_ttsSelections[\'inline-'+esc(vid)+'\'].voice,parseFloat(document.querySelector(\'.wss[data-selkey=inline-'+esc(vid)+']\').value)||1.2,_ttsSelections[\'inline-'+esc(vid)+'\'].emotion)">🎬 Tạo lại</button>'
|
| 150 |
+
: '<button class="wall-btn-make" onclick="event.stopPropagation();makeShortVideo(\''+esc(vid)+'\',this,_ttsSelections[\'inline-'+esc(vid)+'\'].voice,parseFloat(document.querySelector(\'.wss[data-selkey=inline-'+esc(vid)+']\').value)||1.2,_ttsSelections[\'inline-'+esc(vid)+'\'].emotion)">🎬 Tạo Short</button>';
|
| 151 |
return '<div class="wall-item" id="wall-item-'+esc(vid)+'"><div class="wall-thumb">'+thumbContent+videoBadge+'</div><div class="wall-title">'+esc(p.title)+'</div><div class="wall-text">'+esc((p.text||'').slice(0,180))+'</div>'+voiceBar+'<div class="wall-actions"><button class="primary" onclick="readWallPost('+i+')">Xem</button>'+makeBtn+'</div></div>';
|
| 152 |
}
|
| 153 |
|
|
|
|
| 385 |
async function showHashtagSources(topic,page){const box=document.getElementById('hashtag-box');if(!box)return;if(page===0)box.innerHTML=`<div class="hashtag-sources"><h3>🔍 ${esc(topic)}</h3><div class="hashtag-loading"><div class="hashtag-spinner"></div>Đang tìm...</div></div>`;try{const r=await fetch(`/api/hashtag/sources?topic=${encodeURIComponent(topic)}&page=${page}`);const j=await r.json();const sources=j.sources||[];if(!sources.length&&page===0){box.innerHTML=`<div class="hashtag-sources"><h3>🔍 ${esc(topic)}</h3><div style="color:#888;padding:8px">Không tìm được bài viết liên quan</div></div>`;return;}let h='';if(page===0)h=`<div class="hashtag-sources"><h3>🔍 ${esc(topic)} <span style="font-size:10px;color:#888">(${j.total} bài từ 8 nguồn)</span></h3><div id="ht-list">`;sources.forEach((s,i)=>{const idx=page*8+i;h+=`<div class="hashtag-src-item" onclick="readArticle('${esc(s.url)}')"><div class="hashtag-src-img" id="ht-img-${idx}"></div><div class="hashtag-src-text"><div class="hashtag-src-title">${esc(s.title)}</div><div class="hashtag-src-via">${esc(s.via||'')}</div></div></div>`;});if(page===0){h+=`</div><button class="hashtag-rewrite-btn" onclick="rewriteHashtag('${esc(topic).replace(/'/g,"\\'")}')">🤖 Rewrite AI tổng hợp & đăng tường</button>`;if(j.has_more)h+=`<button class="hashtag-load-more" id="ht-more" onclick="loadMoreHashtag()">Tải thêm ▼</button>`;h+=`</div>`;box.innerHTML=h;}else{document.getElementById('ht-list')?.insertAdjacentHTML('beforeend',h);const btn=document.getElementById('ht-more');if(btn){if(!j.has_more)btn.remove();else{btn.disabled=false;btn.textContent='Tải thêm ▼';}}}sources.forEach((s,i)=>{const idx=page*8+i;if(!s.url)return;const ctrl=new AbortController();setTimeout(()=>ctrl.abort(),4000);fetch('/api/article?url='+encodeURIComponent(s.url),{signal:ctrl.signal}).then(r=>r.json()).then(d=>{if(d&&(d.og_image||d.img)){const el=document.getElementById('ht-img-'+idx);if(el)el.innerHTML=`<img src="${_proxyImg(esc(d.og_image||d.img))}" onerror="this.style.display='none'">`;}}).catch(()=>{});});}catch(e){box.innerHTML=`<div class="hashtag-sources"><h3>🔍 ${esc(topic)}</h3><div style="color:#e74c3c;padding:8px">Lỗi: ${esc(e.message)}</div></div>`;}}
|
| 386 |
function loadMoreHashtag(){_htPage++;const btn=document.getElementById('ht-more');if(btn){btn.disabled=true;btn.textContent='Đang tải...';}showHashtagSources(_htTopic,_htPage);}
|
| 387 |
async function rewriteHashtag(topic){const btn=event?.target;if(btn){btn.disabled=true;btn.textContent='Đang tổng hợp...';}try{const r=await fetch('/api/topic_post',{method:'POST',headers:{'Content-Type':'application/json'},body:JSON.stringify({topic})});const j=await r.json();if(!r.ok||j.error)throw new Error(j.error||'Lỗi');toast('✅ Đã đăng Tường AI!');if(btn)btn.textContent='✅ Đăng thành công!';if(j.post)prependWallPost(j.post);}catch(e){toast('❌ '+e.message);if(btn){btn.disabled=false;btn.textContent='🤖 Rewrite AI';}}}
|
|
|
|
|
|
|
|
|
|
|
|
|
| 388 |
async function loadLivescore(tab){document.querySelectorAll('.ls-tab').forEach(t=>t.classList.remove('active'));document.querySelector(`.ls-tab[data-tab="${tab}"]`)?.classList.add('active');const el=document.getElementById('ls-content');if(!el)return;el.innerHTML='<div class="loading">Đang tải...</div>';let ep='/api/livescore/'+tab;if(tab.startsWith('bxh_'))ep='/api/livescore/standings/'+tab.replace('bxh_','');try{const r=await fetch(ep);const d=await r.json();el.innerHTML=d.html&&d.html.length>50?d.html:'<div class="loading">Không có dữ liệu</div>';bindMatchClicks(el);}catch(e){el.innerHTML='<div class="loading">Lỗi</div>';}}
|
| 389 |
function bindMatchClicks(el){el.querySelectorAll('.match-detail').forEach(md=>{md.style.cursor='pointer';md.addEventListener('click',function(e){const statusA=this.querySelector('.status a');const teamA=this.querySelector('.teams a[href*="/tran-dau/"]');const a = statusA || teamA;if(a){e.preventDefault();e.stopPropagation();const href=a.getAttribute('href')||'';const m=href.match(/\/tran-dau\/(\d+)\//);if(m){const fullUrl=href.startsWith('http')?href:'https://bongda.com.vn'+href;openMatch(m[1],fullUrl);}}});});el.querySelectorAll('a').forEach(a=>{a.addEventListener('click',e=>{e.preventDefault();e.stopPropagation()});});}
|
| 390 |
function openMatch(id,url){if(!id)return;_currentEventId=id;if(url)_currentMatchUrl=url;document.getElementById('match-overlay').classList.add('active');document.body.style.overflow='hidden';loadMatchTab('detail')}
|
|
|
|
| 455 |
/** Show rewrite slide viewer - vertical slides with text+image */
|
| 456 |
function readSlidePost(idx){const p=_wallPosts[idx];if(!p||!p.slides)return;showView('view-article');const el=document.getElementById('view-article');let h=`<button class="back-btn" onclick="switchCat('home')">← Tường AI</button><div class="slide-viewer" style="padding:12px;max-width:600px;margin:0 auto">`;p.slides.forEach((s,i)=>{h+=`<div class="slide-card" style="background:#1a1a1a;border:1px solid #2a2a2a;border-radius:12px;padding:16px;margin-bottom:12px"><div class="slide-num" style="color:#5cb87a;font-size:12px;font-weight:700;margin-bottom:6px">Slide ${s.index||i+1}/${p.slides.length}</div>${s.image?`<img src="${_proxyImg(s.image)}" style="width:100%;max-height:300px;object-fit:cover;border-radius:8px;margin-bottom:8px" loading="lazy" onerror="this.style.display=\'none\'">`:''}<p style="color:#ddd;font-size:14px;line-height:1.6;margin:0">${esc(s.text)}</p></div>`;});h+=`</div><div class="article-actions" style="padding:12px;text-align:center"><button onclick="doShare('${esc(p.title)}','${esc(p.url||'')}','${esc(p.img||'')}','${esc(p.id||'')}')" style="background:#2d8659;border:0;color:#fff;padding:8px 16px;border-radius:10px;cursor:pointer;font-size:12px">📤 Chia sẻ bài viết này</button></div>`;el.innerHTML=h;}
|
| 457 |
function readNewsTab(tab){loadNewsTab();}
|
| 458 |
+
function loadNewsTab(){const el=document.getElementById('view-cat');if(!el)return;el.innerHTML='<div class="loading">Đang tải tin tức...</div>';fetch('/api/homepage').then(r=>r.json()).then(articles=>{if(!articles||!articles.length){el.innerHTML='<div class="loading">Không có tin</div>';return}let h='<div class="grid">';articles.forEach(a=>{const src=a.source||'vne';const badge=a.group||a.source||'';h+=`<div class="card" onclick="readArticle('${esc(a.link)}')"><div class="card-img">${a.img?`<img src="${_proxyImg(a.img)}" loading="lazy" onerror="this.style.display=\'none\'">`:''}</div><div class="card-body"><span class="badge badge-${src}">${esc(badge)}</span><div class="card-title">${esc(a.title)}</div></div></div>`;});h+='</div>';el.innerHTML=h;}).catch(()=>{el.innerHTML='<div class="loading">Lỗi tải</div>';});}
|
| 459 |
|
| 460 |
+
function readArticle(url,title,img,presetTitle,presetText){showView('view-article');const el=document.getElementById('view-article');el.innerHTML='<div class="loading">Đang tải...</div><button class="back-btn" onclick="switchCat(\'home\')">← Quay lại</button>';if(presetTitle){el.innerHTML=`<button class="back-btn" onclick="switchCat('home')">← Quay lại</button><div class="article-view"><h1 class="article-title">${esc(presetTitle)}</h1>${presetText?`<div class="article-summary" style="white-space:pre-wrap">${esc(presetText)}</div>`:''}</div><div class="article-actions"><button onclick="doShare('${esc(presetTitle)}','${esc(url||'')}','')">📤 Chia sẻ</button></div>`;return;}if(!url)return;fetch('/api/article?url='+encodeURIComponent(url)).then(r=>r.json()).then(d=>{let h=`<button class="back-btn" onclick="switchCat('home')">← Quay lại</button><div class="article-view">`;if(d.title)h+=`<h1 class="article-title">${esc(d.title)}</h1>`;if(d.summary)h+=`<div class="article-summary">${esc(d.summary)}</div>`;if(d.body)d.body.forEach(b=>{if(b.type==='p')h+=`<p class="article-p">${esc(b.text)}</p>`;else if(b.type==='heading')h+=`<h2 class="article-h2">${esc(b.text)}</h2>`;else if(b.type==='img'&&b.src)h+=`<img class="article-img" src="${_proxyImg(b.src)}" loading="lazy" onerror="this.style.display=\'none\'">`;});h+=`</div><div class="article-actions"><button onclick="doShare('${esc(d.title||'')}','${esc(url)}','${esc(d.og_image||'')}')">📤 Chia sẻ</button><button class="primary" onclick="rewriteSlide('${esc(url)}')">🤖 Slide Rewrite AI</button></div>`;el.innerHTML=h;}).catch(()=>{el.innerHTML=`<button class="back-btn" onclick="switchCat('home')">← Quay lại</button><div class="article-view"><p>Không thể tải bài viết</p></div>`;});}
|
| 461 |
async function rewriteSlide(url){if(!url)return;const btn=document.querySelector('.article-actions .primary')||event?.target;if(btn){btn.disabled=true;btn.textContent='⏳ Đang tạo slides...';}toast('⏳ Đang tạo slide rewrite...');try{const r=await fetch('/api/rewrite_share',{method:'POST',headers:{'Content-Type':'application/json'},body:JSON.stringify({url})});const j=await r.json();if(!r.ok||j.error)throw new Error(j.error||'Lỗi');toast('✅ Đã tạo slide! Đang tạo Short AI...');if(btn)btn.textContent='✅ Đang tạo Short AI...';if(j.post){j.post.slides=j.slides||[];prependWallPost(j.post);}// Show slides immediately if available
|
| 462 |
if(j.slides && j.slides.length){showView('view-article');const el=document.getElementById('view-article');let h=`<button class="back-btn" onclick="switchCat('home')">← Tường AI</button><div class="slide-viewer" style="padding:12px;max-width:600px;margin:0 auto">`;j.slides.forEach(s=>{h+=`<div class="slide-card" style="background:#1a1a1a;border:1px solid #2a2a2a;border-radius:12px;padding:16px;margin-bottom:12px"><div class="slide-num" style="color:#5cb87a;font-size:12px;font-weight:700;margin-bottom:6px">Slide ${s.index}/${j.slides.length}</div>${s.image?`<img src="${s.image.startsWith('/api/')?s.image:'/api/proxy/img?url='+encodeURIComponent(s.image)}" style="width:100%;max-height:300px;object-fit:cover;border-radius:8px;margin-bottom:8px" loading="lazy" onerror="this.style.display='none'">`:''}<p style="color:#ddd;font-size:14px;line-height:1.6;margin:0">${esc(s.text)}</p></div>`;});h+=`</div><div style="text-align:center;padding:12px"><div id="short-ai-status" style="color:#888;font-size:12px">⏳ Đang tạo video Short AI...</div></div>`;el.innerHTML=h;}
|
| 463 |
const postId=j.post&&j.post.id;if(postId){setTimeout(async()=>{const sr=await fetch('/api/ai/short/'+encodeURIComponent(postId),{method:'POST',headers:{'Content-Type':'application/json'},body:JSON.stringify({voice:'vi-VN-HoaiMyNeural',emotion:'neutral',speed:1.2})});const sj=await sr.json();if(sr.ok&&sj.video&&typeof _wallPosts!=='undefined'){const p=_wallPosts.find(x=>String(x.id)===String(postId));if(p){p.video=sj.video;const itemId='wall-item-'+postId;const el2=document.getElementById(itemId);if(el2){const idx=_wallPosts.indexOf(p);el2.insertAdjacentHTML('afterend',makeWallItem(p,idx));el2.remove();}}toast('✅ Short AI đã sẵn sàng!');const statusEl=document.getElementById('short-ai-status');if(statusEl)statusEl.innerHTML='✅ Short AI đẵn sàng! <button class="primary" onclick="openShortAIFeed(0)" style="background:#2d8659;border:0;color:#fff;padding:6px 12px;border-radius:10px;font-size:11px;cursor:pointer;margin-left:8px">▶ Xem ngay</button>';}},500);}if(btn)btn.textContent='✅ Hoàn tất';}catch(e){toast('❌ '+e.message);if(btn){btn.disabled=false;btn.textContent='🤖 Slide Rewrite AI';}}}
|
static/index_v2.html
CHANGED
|
@@ -41,6 +41,7 @@ function doShare(title,url,img,postId){
|
|
| 41 |
} else {
|
| 42 |
shareUrl = SPACE+'/s?url='+encodeURIComponent(url)+'&title='+encodeURIComponent(title)+'&img='+encodeURIComponent(img||'');
|
| 43 |
}
|
|
|
|
| 44 |
if(navigator.clipboard && navigator.clipboard.writeText){
|
| 45 |
navigator.clipboard.writeText(shareUrl).then(function(){
|
| 46 |
toast('📋 Đã sao chép link!');
|
|
@@ -67,7 +68,7 @@ function doShare(title,url,img,postId){
|
|
| 67 |
async function init(){_cats=await fetch('/api/categories').then(r=>r.json()).catch(()=>[]);let bar='<div class="cat active" data-cat="home">🏠</div><div class="cat" data-cat="news-all">📰 Tin tức</div>';_cats.forEach(c=>{bar+='<div class="cat" data-cat="'+c.id+'">'+c.name+'</div>'});document.getElementById('cat-bar').innerHTML=bar;document.querySelectorAll('.cat').forEach(t=>{t.onclick=()=>switchCat(t.dataset.cat)});}
|
| 68 |
var SPACE=location.origin;
|
| 69 |
</script>
|
| 70 |
-
<script src="/static/app_v2.js?v=
|
| 71 |
<script src="/static/yt_live.js"></script>
|
| 72 |
<script src="/static/vtv_init.js"></script>
|
| 73 |
<script src="/static/hot_multi.js?v=1"></script>
|
|
|
|
| 41 |
} else {
|
| 42 |
shareUrl = SPACE+'/s?url='+encodeURIComponent(url)+'&title='+encodeURIComponent(title)+'&img='+encodeURIComponent(img||'');
|
| 43 |
}
|
| 44 |
+
// Use clipboard API first (modern, works on Chrome mobile 85+)
|
| 45 |
if(navigator.clipboard && navigator.clipboard.writeText){
|
| 46 |
navigator.clipboard.writeText(shareUrl).then(function(){
|
| 47 |
toast('📋 Đã sao chép link!');
|
|
|
|
| 68 |
async function init(){_cats=await fetch('/api/categories').then(r=>r.json()).catch(()=>[]);let bar='<div class="cat active" data-cat="home">🏠</div><div class="cat" data-cat="news-all">📰 Tin tức</div>';_cats.forEach(c=>{bar+='<div class="cat" data-cat="'+c.id+'">'+c.name+'</div>'});document.getElementById('cat-bar').innerHTML=bar;document.querySelectorAll('.cat').forEach(t=>{t.onclick=()=>switchCat(t.dataset.cat)});}
|
| 69 |
var SPACE=location.origin;
|
| 70 |
</script>
|
| 71 |
+
<script src="/static/app_v2.js?v=20260711"></script>
|
| 72 |
<script src="/static/yt_live.js"></script>
|
| 73 |
<script src="/static/vtv_init.js"></script>
|
| 74 |
<script src="/static/hot_multi.js?v=1"></script>
|