Update app.py

app.py

@@ -1,12 +1,13 @@
 # ================================================================
-# 🔬 Scientific Paper Discovery Bot — v7.2
-# FIX: CrossRef garbage dates (2048, 2116, 2117...) now rejected
+# 🔬 Scientific Paper Discovery Bot — v7.3
+# NEW: PDF Export for Explanation + 🔎 Global Paper Search Tab
 # ================================================================
 import os, re, time, json, pickle, threading
 import requests
 import xml.etree.ElementTree as ET
 from datetime import datetime, timedelta
 from collections import Counter
+from io import BytesIO
 
 import numpy as np
 import faiss
@@ -19,6 +20,14 @@ from sentence_transformers import SentenceTransformer
 from groq import Groq
 from gtts import gTTS
 from langdetect import detect, DetectorFactory
+from reportlab.lib.pagesizes import A4
+from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle
+from reportlab.lib.units import cm
+from reportlab.lib import colors
+from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer, HRFlowable
+from reportlab.pdfbase import pdfmetrics
+from reportlab.pdfbase.ttfonts import TTFont
+
 DetectorFactory.seed = 0
 
 GROQ_API_KEY = os.environ.get("GROQ_API_KEY", "")
@@ -35,7 +44,7 @@ ACTIVE_PAPERS = []
 FAISS_INDEX   = None
 AUTO_RUNNING  = False
 AUTO_LOG      = []
-CURRENT_YEAR  = datetime.now().year   # 2026
+CURRENT_YEAR  = datetime.now().year
 
 PERSIST_DIR    = "/tmp"
 FAVORITES_PATH = f"{PERSIST_DIR}/favorites.pkl"
@@ -122,31 +131,24 @@ def build_table(papers_list):
     return rows, choices
 
 def s2_headers():
-    h = {"User-Agent": "ScientificPaperBot/7.2"}
+    h = {"User-Agent": "ScientificPaperBot/7.3"}
     if S2_API_KEY: h["x-api-key"] = S2_API_KEY
     return h
 
 def cr_headers():
-    return {"User-Agent": "ScientificPaperBot/7.2 (mailto:researcher@example.com)"}
+    return {"User-Agent": "ScientificPaperBot/7.3 (mailto:researcher@example.com)"}
 
 # ================================================================
-# ✅ FIXED: CrossRef date parser — rejects garbage years
+# ✅ CrossRef date parser — rejects garbage years
 # ================================================================
 def parse_crossref_date(item: dict) -> str:
-    """
-    CrossRef sometimes stores volume/issue/page numbers in date-parts
-    causing years like 2048, 2103, 2116, 2117 etc.
-    We try 5 date fields in priority order and reject any year
-    outside the range [1900, CURRENT_YEAR+1].
-    """
     for field in ["issued", "published", "published-print", "published-online", "created"]:
         dp = (item.get(field) or {}).get("date-parts", [[]])
         if not dp or not dp[0]: continue
         pts = dp[0]
         try:
             year = int(pts[0])
-            if not (1900 <= year <= CURRENT_YEAR + 1):
-                continue                          # ← reject garbage year
+            if not (1900 <= year <= CURRENT_YEAR + 1): continue
             month = max(1, min(12, int(pts[1]) if len(pts) >= 2 else 1))
             day   = max(1, min(31, int(pts[2]) if len(pts) >= 3 else 1))
             return f"{year:04d}-{month:02d}-{day:02d}"
@@ -190,6 +192,86 @@ def export_favorites_csv():
 
 def gr_export_fav(): return export_favorites_csv()
 
+# ================================================================
+# ✅ NEW: PDF EXPORT
+# ================================================================
+def export_explanation_pdf(explanation_text, paper_title="paper"):
+    if not explanation_text or len(explanation_text) < 30:
+        return None
+    safe_title = re.sub(r"[^\w\s-]", "", paper_title)[:50].strip().replace(" ", "_")
+    path = f"{PERSIST_DIR}/explanation_{safe_title}.pdf"
+
+    doc    = SimpleDocTemplate(path, pagesize=A4,
+                               rightMargin=2*cm, leftMargin=2*cm,
+                               topMargin=2*cm, bottomMargin=2*cm)
+    styles = getSampleStyleSheet()
+
+    title_style = ParagraphStyle("Title2", parent=styles["Title"],
+                                 fontSize=14, textColor=colors.HexColor("#1e3a5f"),
+                                 spaceAfter=12)
+    h2_style    = ParagraphStyle("H2",     parent=styles["Heading2"],
+                                 fontSize=11, textColor=colors.HexColor("#2563eb"),
+                                 spaceBefore=14, spaceAfter=6)
+    body_style  = ParagraphStyle("Body",   parent=styles["Normal"],
+                                 fontSize=10, leading=16, spaceAfter=8)
+    meta_style  = ParagraphStyle("Meta",   parent=styles["Normal"],
+                                 fontSize=9,  textColor=colors.HexColor("#64748b"),
+                                 spaceAfter=10)
+
+    story = []
+    lines = explanation_text.split("\n")
+    for line in lines:
+        line = line.strip()
+        if not line: story.append(Spacer(1, 6)); continue
+
+        # strip markdown artifacts for PDF
+        clean = re.sub(r"\*\*(.+?)\*\*", r"\1", line)
+        clean = re.sub(r"\*(.+?)\*",     r"\1", clean)
+        clean = re.sub(r"`(.+?)`",       r"\1", clean)
+        clean = re.sub(r"^#{1,6}\s*",    "",    clean)
+        clean = re.sub(r"[🎯❓🔧📊🌟🔗📄👥📅📡🤖#*_~]", "", clean).strip()
+
+        if line.startswith("## ") or line.startswith("# "):
+            story.append(HRFlowable(width="100%", thickness=0.5,
+                                    color=colors.HexColor("#e2e8f0"), spaceAfter=4))
+            story.append(Paragraph(clean, h2_style))
+        elif line.startswith("**") and line.endswith("**"):
+            story.append(Paragraph(f"<b>{clean}</b>", body_style))
+        elif line.startswith(">"):
+            quote = line.lstrip(">").strip()
+            quote = re.sub(r"[🎯❓🔧📊🌟🔗📄👥📅📡🤖#*_~]", "", quote).strip()
+            q_style = ParagraphStyle("Quote", parent=styles["Normal"],
+                                     fontSize=9, leftIndent=20, rightIndent=10,
+                                     textColor=colors.HexColor("#475569"),
+                                     borderPadding=(4,4,4,8),
+                                     leading=14, spaceAfter=8)
+            story.append(Paragraph(quote, q_style))
+        else:
+            if clean:
+                story.append(Paragraph(clean, body_style))
+
+    # Footer
+    story.append(Spacer(1, 20))
+    story.append(HRFlowable(width="100%", thickness=0.5, color=colors.HexColor("#e2e8f0")))
+    story.append(Paragraph(
+        f"Generated by 🔬 Scientific Paper Discovery v7.3 — {datetime.now().strftime('%Y-%m-%d %H:%M')}",
+        meta_style))
+
+    try:
+        doc.build(story)
+        return path
+    except Exception as e:
+        print(f"PDF error: {e}")
+        return None
+
+def gr_export_pdf(explanation_text, choice):
+    if not explanation_text or len(explanation_text) < 50:
+        return None, "⚠️ اشرح الورقة أولاً ثم صدّر PDF."
+    title = choice.split(". ", 1)[-1] if choice else "paper"
+    path  = export_explanation_pdf(explanation_text, title)
+    if path: return path, "✅ تم إنشاء PDF!"
+    return None, "❌ فشل إنشاء PDF."
+
 # ================================================================
 # SOURCE 1 — arXiv
 # ================================================================
@@ -235,14 +317,14 @@ def fetch_arxiv_papers(query, category, max_results=20, days_back=365):
     return papers
 
 # ================================================================
-# SOURCE 2 — CrossRef (with fixed date parser + title filter)
+# SOURCE 2 — CrossRef
 # ================================================================
 def fetch_crossref_papers(query, category_label="", max_results=20, days_back=365):
     subject    = CROSSREF_SUBJECTS.get(category_label, "")
     full_query = f"{query} {subject}".strip() if subject else query
     params = {
         "query":  full_query,
-        "rows":   min(max_results * 3, 200),  # fetch more, filter bad dates after
+        "rows":   min(max_results * 3, 200),
         "sort":   "relevance",
         "select": "title,author,abstract,published,published-print,published-online,issued,created,DOI,is-referenced-by-count,link,subject",
     }
@@ -261,34 +343,27 @@ def fetch_crossref_papers(query, category_label="", max_results=20, days_back=36
     papers, seen_ids = [], set()
     for item in items:
         if len(papers) >= max_results: break
-
         title_list = item.get("title", [])
         if not title_list: continue
         title = title_list[0].strip()
-        if not title or title.lower().startswith("title pending"): continue  # skip junk
-
-        # ✅ Use fixed date parser
+        if not title or title.lower().startswith("title pending"): continue
         pub = parse_crossref_date(item)
-        if pub == "N/A": continue   # skip items with no valid date at all
-
+        if pub == "N/A": continue
         cit     = int(item.get("is-referenced-by-count", 0) or 0)
         authors = [f"{a.get('given','').strip()} {a.get('family','').strip()}".strip()
                    for a in item.get("author",[])[:6]]
         authors = [a for a in authors if a.strip()]
         if not authors: authors = ["Unknown"]
-
         abstract = re.sub(r"<[^>]+>","", item.get("abstract","No abstract.")).strip()[:1200]
         doi      = item.get("DOI","")
         url      = f"https://doi.org/{doi}" if doi else "#"
         pid      = doi or re.sub(r"\W","",title)[:40]
         if pid in seen_ids: continue
         seen_ids.add(pid)
-
         pdf_url  = next((l.get("URL","") for l in item.get("link",[])
                          if "pdf" in l.get("content-type","").lower()), "")
         try:    recent = datetime.strptime(pub[:10],"%Y-%m-%d") >= cutoff
         except: recent = False
-
         papers.append({
             "id": pid, "title": title, "authors": authors,
             "abstract": abstract, "published": pub[:10],
@@ -296,10 +371,42 @@ def fetch_crossref_papers(query, category_label="", max_results=20, days_back=36
             "citations": cit, "url": url, "pdf_url": pdf_url,
             "recent": recent, "source": "CrossRef",
         })
-
     papers.sort(key=lambda x: x["citations"], reverse=True)
     return papers
 
+# ================================================================
+# ✅ NEW: GLOBAL PAPER SEARCH (independent of local index)
+# ================================================================
+def global_paper_search(query, source_choice, max_results=10):
+    if not query or not query.strip():
+        return "⚠️ أدخل عنوان أو كلمات مفتاحية للبحث."
+    papers = []
+    q = query.strip()
+    if source_choice in ("arXiv", "كلاهما / Both"):
+        papers += fetch_arxiv_papers(q, "", int(max_results), 3650)
+    if source_choice in ("CrossRef", "كلاهما / Both"):
+        papers += fetch_crossref_papers(q, "", int(max_results), 3650)
+    if not papers:
+        return f"❌ لا نتائج لـ `{q}`. جرب كلمات مختلفة."
+
+    seen, unique = set(), []
+    for p in papers:
+        key = re.sub(r"\W","",p["title"].lower())[:60]
+        if key not in seen: seen.add(key); unique.append(p)
+
+    md = f"## 🔎 نتائج البحث العالمي — `{q}` — **{len(unique)}** ورقة\n\n---\n\n"
+    for i, p in enumerate(unique, 1):
+        cit  = f" | {cit_badge(p.get('citations'))}" if p.get("citations") else ""
+        cats = " · ".join(p.get("categories",[])[:2])
+        md += (f"### {i}. {p['title']}\n\n"
+               f"👥 {', '.join(p['authors'][:3])} | 📅 {p['published']}{cit} | "
+               f"📡 {p.get('source','')} | 🏷️ {cats}\n\n"
+               f"> {p['abstract'][:400]}...\n\n"
+               f"🔗 [View]({p['url']})"
+               +(f"  📥 [PDF]({p['pdf_url']})" if p.get("pdf_url") else "")
+               +"\n\n---\n\n")
+    return md
+
 # ================================================================
 # CITATION ENGINE — 3-layer
 # ================================================================
@@ -310,7 +417,6 @@ def enrich_citations(papers):
         for p in papers:
             if p.get("citations") is None: p["citations"] = 0
         return papers
-
     id_map, batch_ids = {}, []
     for p in arxiv_papers:
         clean = re.sub(r"v\d+$","", p["id"].split("/")[-1].strip())
@@ -639,7 +745,6 @@ def gr_fetch(query, category_label, max_results, days_back, source_choice,
     progress(1.0)
     return md, upd, upd, upd, upd, f"✅ {len(papers)} | 🆕 {recent} | 📊 {tot_cit:,}"
 
-
 def gr_filter_papers(year_from, year_to, cit_min, cit_max, sort_by):
     global ACTIVE_PAPERS
     if not PAPERS: return "⚠️ اجلب الأوراق أولاً.", gr.update(), "⚠️"
@@ -667,7 +772,6 @@ def gr_filter_papers(year_from, year_to, cit_min, cit_max, sort_by):
            f"📊 اقتباسات {cit_min}–{cit_max} &nbsp;|&nbsp; مجموع {tot:,}\n\n---\n\n{tbl}")
     return md, gr.update(choices=choices, value=choices[0] if choices else None), f"🔽 {len(filtered)}/{len(PAPERS)}"
 
-
 def gr_search_fetched(query):
     if not query or not query.strip(): return "⚠️ أدخل كلمة بحث."
     if not PAPERS: return "⚠️ اجلب الأوراق أولاً."
@@ -766,14 +870,15 @@ h1{text-align:center}
 .status-bar{font-size:.85rem;color:#94a3b8;padding:2px 0}
 .legend{font-size:.8rem;color:#cbd5e1;background:#1e293b;border-radius:8px;padding:6px 14px;margin-bottom:6px}
 .filter-box{background:#1e293b;border-radius:10px;padding:12px 16px;margin-top:8px}
+.global-search-box{background:#1e293b;border-radius:10px;padding:14px 18px;margin-bottom:10px;border:1px solid #334155}
 """
 
 with gr.Blocks(
     theme=gr.themes.Soft(primary_hue="blue", secondary_hue="purple"),
-    title="🔬 Paper Discovery v7.2", css=CSS
+    title="🔬 Paper Discovery v7.3", css=CSS
 ) as demo:
 
-    gr.Markdown("# 🔬 Scientific Paper Discovery v7.2\n**arXiv · CrossRef · Llama-3.3-70B · FAISS**")
+    gr.Markdown("# 🔬 Scientific Paper Discovery v7.3\n**arXiv · CrossRef · Llama-3.3-70B · FAISS**")
     gr.Markdown(
         "📊 **الاقتباسات:** &nbsp; 🥇 ≥ 1,000 &nbsp;|&nbsp; "
         "🏆 ≥ 100 &nbsp;|&nbsp; ⭐ ≥ 10 &nbsp;|&nbsp; 📄 < 10 &nbsp;|&nbsp; · = 0",
@@ -781,6 +886,7 @@ with gr.Blocks(
     status_bar = gr.Markdown("_لم يتم جلب أوراق بعد_", elem_classes="status-bar")
 
     with gr.Tabs():
+        # ── TAB 1: SEARCH ──────────────────────────────────
         with gr.Tab("🔍 البحث / Search"):
             with gr.Row():
                 with gr.Column(scale=3):
@@ -804,24 +910,46 @@ with gr.Blocks(
                 with gr.Row():
                     f_sort     = gr.Dropdown(choices=SORT_CHOICES, value=SORT_CHOICES[2], label="🔃 الترتيب", scale=3)
                     btn_filter = gr.Button("✅ تطبيق", variant="primary", scale=1)
-            gr.Markdown("---\n### 🔍 بحث دلالي")
+            gr.Markdown("---\n### 🔍 بحث دلالي (في الأوراق المحملة)")
             with gr.Row():
                 search_in_box = gr.Textbox(label="🔍 ابحث", placeholder="ARIMA, transformer...", scale=5)
                 btn_search_in = gr.Button("بحث 🔍", scale=1)
             search_in_out = gr.Markdown()
 
+        # ── TAB 2: GLOBAL SEARCH ✅ NEW ────────────────────
+        with gr.Tab("🌐 بحث عالمي / Global Search"):
+            gr.Markdown("### 🌐 ابحث عن أي ورقة مباشرة — بدون جلب مسبق")
+            with gr.Group(elem_classes="global-search-box"):
+                with gr.Row():
+                    gs_query  = gr.Textbox(
+                        label="🔎 ابحث بالعنوان أو الكلمات المفتاحية",
+                        placeholder="مثال: ARIMA forecasting inflation Algeria ...",
+                        scale=4)
+                    gs_source = gr.Radio(
+                        label="📡 المصدر",
+                        choices=["arXiv","CrossRef","كلاهما / Both"],
+                        value="كلاهما / Both", scale=2)
+                    gs_max    = gr.Slider(5, 30, value=10, step=5, label="📊 عدد النتائج", scale=1)
+                btn_gs = gr.Button("🔎 بحث عالمي", variant="primary", size="lg")
+            gs_out = gr.Markdown("_أدخل موضوعاً للبحث..._")
+
+        # ── TAB 3: EXPLAIN ─────────────────────────────────
         with gr.Tab("📖 الشرح / Explain"):
             with gr.Row():
                 paper_sel2 = gr.Dropdown(label="📄 اختر الورقة", choices=[], interactive=True, scale=4)
                 lang_exp   = gr.Radio(LANG_CHOICES, value=LANG_CHOICES[0], label="🌐 اللغة", scale=1)
             with gr.Row():
-                btn_explain = gr.Button("📖 اشرح", variant="primary")
-                btn_fav     = gr.Button("⭐ حفظ")
-                btn_audio   = gr.Button("🔊 استمع")
+                btn_explain    = gr.Button("📖 اشرح", variant="primary")
+                btn_fav        = gr.Button("⭐ حفظ")
+                btn_audio      = gr.Button("🔊 استمع")
+                btn_export_pdf = gr.Button("📄 تصدير PDF", variant="secondary")   # ✅ NEW
             fav_status      = gr.Markdown()
+            pdf_status      = gr.Markdown()                                        # ✅ NEW
             explanation_out = gr.Markdown("_اجلب الأوراق واختر ورقة._")
             audio_out       = gr.Audio(label="🔊", type="filepath")
+            pdf_out         = gr.File(label="📄 تحميل PDF")                       # ✅ NEW
 
+        # ── TAB 4: COMPARE ─────────────────────────────────
         with gr.Tab("⚖️ المقارنة / Compare"):
             with gr.Row():
                 cmp_a    = gr.Dropdown(label="📄 الأولى",  choices=[], interactive=True)
@@ -830,29 +958,34 @@ with gr.Blocks(
             btn_compare = gr.Button("⚖️ قارن الآن", variant="primary")
             compare_out = gr.Markdown("_اختر ورقتين._")
 
+        # ── TAB 5: OVERVIEW ────────────────────────────────
         with gr.Tab("🌐 نظرة عامة"):
             with gr.Row():
                 lang_ov      = gr.Radio(LANG_CHOICES, value=LANG_CHOICES[0], label="🌐 اللغة", scale=1)
                 btn_overview = gr.Button("🤖 توليد التقرير", variant="primary", scale=3)
             overview_out = gr.Markdown("_اجلب الأوراق أولاً._")
 
+        # ── TAB 6: TRENDS ──────────────────────────────────
         with gr.Tab("📊 الاتجاهات / Trends"):
             btn_trends  = gr.Button("📊 تحليل الاتجاهات", variant="primary", size="lg")
             trend_chart = gr.Image(label="📊 لوحة الاتجاهات", type="filepath")
             trend_stats = gr.Markdown("_اجلب الأوراق أولاً._")
 
+        # ── TAB 7: BIBLIOGRAPHY ────────────────────────────
         with gr.Tab("📚 المراجع"):
             bib_style = gr.Radio(["APA","IEEE","Chicago","BibTeX"], value="APA", label="📐 النمط")
             btn_bib   = gr.Button("📚 توليد المراجع", variant="primary")
             bib_out   = gr.Markdown()
             bib_file  = gr.File(label="📥 تحميل")
 
+        # ── TAB 8: FAVORITES ───────────────────────────────
         with gr.Tab("⭐ المفضلة"):
             btn_show_fav   = gr.Button("📋 عرض المفضلة")
             favs_md        = gr.Markdown("_اضغط عرض._")
             btn_export_fav = gr.Button("📥 تصدير CSV", variant="secondary")
             fav_csv_file   = gr.File(label="📄 CSV")
 
+        # ── TAB 9: AUTO-FETCH ─────────────���────────────────
         with gr.Tab("🔔 تحديث تلقائي"):
             with gr.Row():
                 auto_q        = gr.Textbox(label="🔎 الموضوع", value="economic forecasting", scale=3)
@@ -865,6 +998,7 @@ with gr.Blocks(
             auto_status = gr.Markdown()
             auto_log_md = gr.Markdown("_لا يوجد سجل._")
 
+        # ── TAB 10: CHAT ───────────────────────────────────
         with gr.Tab("💬 محادثة / Chat"):
             chatbot_ui = gr.Chatbot(label="مساعد الأبحاث", height=480, bubble_full_width=False)
             with gr.Row():
@@ -872,26 +1006,22 @@ with gr.Blocks(
                 btn_send = gr.Button("إرسال ✉️", variant="primary", scale=1)
             btn_clear = gr.Button("🗑️ مسح", size="sm")
 
+        # ── TAB 11: ABOUT ──────────────────────────────────
         with gr.Tab("ℹ️ حول"):
             gr.Markdown("""
-## 🔬 Scientific Paper Discovery — v7.2
+## 🔬 Scientific Paper Discovery — v7.3
+
+### ✅ جديد في v7.3
+| الميزة | التفاصيل |
+|---|---|
+| 📄 تصدير PDF | تحويل الشرح الكامل إلى PDF احترافي بزر واحد |
+| 🌐 بحث عالمي | تبويب مستقل للبحث عن أي ورقة بدون جلب مسبق |
 
-### ✅ إصلاحات v7.2
+### 🔧 إصلاحات v7.2 (مُضمَّنة)
 | المشكلة | الإصلاح |
 |---|---|
-| تواريخ CrossRef مزيفة (2048، 2116، 2117...) | `parse_crossref_date()` ترفض أي سنة خارج `[1900, 2027]` |
-| `published-print` قد يحوي رقم المجلد | أولوية الحقول: `issued → published → published-print → published-online → created` |
-| عناوين "Title Pending" | فلتر `title.lower().startswith("title pending")` يحذفها |
-| إعادة تشغيل عند الفلتر | Filter → 3 outputs فقط (لا cascade) ✅ |
-
-### 🔧 دالة parse_crossref_date
-```python
-for field in ["issued", "published", "published-print", "published-online", "created"]:
-    year = int(pts[0])
-    if not (1900 <= year <= CURRENT_YEAR + 1):
-        continue   # ← reject 2048, 2116, 2117...
-    return f"{year:04d}-{month:02d}-{day:02d}"
-```
+| تواريخ CrossRef مزيفة (2048، 2116...) | `parse_crossref_date()` ترفض أي سنة خارج `[1900, 2027]` |
+| إعادة تشغيل عند الفلتر | Filter → 3 outputs فقط (لا cascade) |
 """)
 
     # ── WIRING ──────────────────────────────────────────────
@@ -902,16 +1032,24 @@ for field in ["issued", "published", "published-print", "published-online", "cre
                     outputs=FETCH_OUT)
     btn_filter.click(gr_filter_papers,
                      inputs=[f_year_from, f_year_to, f_cit_min, f_cit_max, f_sort],
-                     outputs=[papers_table_md, paper_selector, status_bar])  # 3 only ✅
+                     outputs=[papers_table_md, paper_selector, status_bar])
     paper_selector.change(lambda x: [gr.update(value=x)]*3,
                           inputs=[paper_selector], outputs=[paper_sel2, cmp_a, cmp_b])
 
     btn_search_in.click(gr_search_fetched, inputs=[search_in_box], outputs=[search_in_out])
     search_in_box.submit(gr_search_fetched, inputs=[search_in_box], outputs=[search_in_out])
 
+    # Global Search wiring
+    btn_gs.click(global_paper_search, inputs=[gs_query, gs_source, gs_max], outputs=[gs_out])
+    gs_query.submit(global_paper_search, inputs=[gs_query, gs_source, gs_max], outputs=[gs_out])
+
+    # Explain wiring
     btn_explain.click(gr_explain,   inputs=[paper_sel2, lang_exp],      outputs=[explanation_out])
     btn_fav.click(gr_save_fav,      inputs=[paper_sel2],                outputs=[fav_status])
     btn_audio.click(gr_audio,       inputs=[explanation_out, lang_exp], outputs=[audio_out])
+    btn_export_pdf.click(gr_export_pdf,
+                         inputs=[explanation_out, paper_sel2],
+                         outputs=[pdf_out, pdf_status])  # ✅ NEW
 
     btn_compare.click(gr_compare,   inputs=[cmp_a, cmp_b, lang_cmp],    outputs=[compare_out])
     btn_overview.click(gr_overview, inputs=[t_query, lang_ov],          outputs=[overview_out])