Update streamlit_app.py

streamlit_app.py

@@ -38,9 +38,9 @@ UI_TEXT = {
         "metric_source": "출처",
         "expander_news": "📰 뉴스 원문 보기",
         "insight_title": "🧠 AI 종합 분석 리포트",
-        "scanner_title": "📡 글로벌 시장 전광판 (Max 100)",
-        "scanner_caption": "※ 약 100개 기업을 분석하여 상위 10개를 선별합니다.",
-        "btn_scan": "100개 종목 스캔 시작",
+        "scanner_title": "📡 글로벌 시장 전광판 (Full Version)",
+        "scanner_caption": "※ 약 100개 기업을 정밀 분석하여 상위 10개를 선별합니다.",
+        "btn_scan": "전체 종목 스캔 시작",
         "tab_gainers": "🔥 급등 Top 10 (상승장)",
         "tab_losers": "💧 급락 Top 10 (하락장)",
         "col_name": "기업명",
@@ -63,9 +63,9 @@ UI_TEXT = {
         "metric_source": "Source",
         "expander_news": "📰 View Source News",
         "insight_title": "🧠 AI Comprehensive Report",
-        "scanner_title": "📡 Global Market Scanner (Max 100)",
-        "scanner_caption": "※ Analyzing 100 companies to show Top 10.",
-        "btn_scan": "Scan 100 Stocks",
+        "scanner_title": "📡 Global Market Scanner (Full Version)",
+        "scanner_caption": "※ Analyzing 100+ companies to show Top 10.",
+        "btn_scan": "Start Full Scan",
         "tab_gainers": "🔥 Top 10 Gainers",
         "tab_losers": "💧 Top 10 Losers",
         "col_name": "Company",
@@ -78,35 +78,87 @@ UI_TEXT = {
 }
 
 # ==============================================================================
-# 💾 [데이터] 티커 매핑 (스캐너용) - 검색은 AI가 알아서 함
+# 💾 [데이터] 대규모 확장 티커 (Full List) - 절대 줄이지 않음!
 # ==============================================================================
-# (스캐너용 리스트는 기존 유지 - 너무 길어서 생략하지 않고 그대로 둡니다)
 TICKER_NAMES = {
+    # 🇻🇳 베트남 (부동산/건설/증권/소매 포함)
     "VIC.VN": "Vingroup", "VHM.VN": "Vinhomes", "VCB.VN": "Vietcombank", "VNM.VN": "Vinamilk", 
-    "HPG.VN": "Hoa Phat", "MSN.VN": "Masan Group", "GAS.VN": "PV Gas", "NVL.VN": "Novaland",
-    "BBCA.JK": "BCA Bank", "BBRI.JK": "BRI Bank", "TLKM.JK": "Telkom", "GOTO.JK": "GoTo",
-    "005930.KS": "Samsung Elec", "000660.KS": "SK Hynix", "035420.KS": "NAVER", "207940.KS": "Samsung Bio",
+    "HPG.VN": "Hoa Phat", "MSN.VN": "Masan Group", "GAS.VN": "PV Gas", "NVL.VN": "Novaland", 
+    "PDR.VN": "Phat Dat", "DIG.VN": "DIC Corp", "CEO.VN": "CEO Group", "SSI.VN": "SSI Sec", 
+    "VND.VN": "VNDirect", "MWG.VN": "Mobile World", "FRT.VN": "FPT Retail", "FPT.VN": "FPT Corp",
+    "STB.VN": "Sacombank", "MBB.VN": "MB Bank", "TCB.VN": "Techcombank", "VPB.VN": "VPBank",
+    "VRE.VN": "Vincom Retail", "DGC.VN": "Duc Giang Chem", "VHC.VN": "Vinh Hoan", "KBC.VN": "Kinh Bac",
+    "GVR.VN": "Vietnam Rubber", "SAB.VN": "Sabeco", "BID.VN": "BIDV", "CTG.VN": "VietinBank",
+
+    # 🇮🇩 인도네시아
+    "BBCA.JK": "BCA Bank", "BBRI.JK": "BRI Bank", "TLKM.JK": "Telkom", "BMRI.JK": "Mandiri", 
+    "ASII.JK": "Astra Intl", "GOTO.JK": "GoTo", "UNVR.JK": "Unilever", "ADRO.JK": "Adaro Energy", 
+    "MDKA.JK": "Merdeka Copper", "ANTM.JK": "Aneka Tambang", "ICBP.JK": "Indofood CBP", 
+    "KLBF.JK": "Kalbe Farma", "BUKA.JK": "Bukalapak", "PGAS.JK": "Perusahaan Gas",
+
+    # 🇰🇷 한국
+    "005930.KS": "Samsung Elec", "000660.KS": "SK Hynix", "035420.KS": "NAVER", "035720.KS": "Kakao", 
+    "005380.KS": "Hyundai Motor", "207940.KS": "Samsung Bio", "068270.KS": "Celltrion",
+    "086520.KQ": "Ecopro", "247540.KQ": "Ecopro BM", "005490.KS": "POSCO Holdings", "010120.KS": "LS ELECTRIC",
+    "012450.KS": "Hanwha Aero", "042700.KS": "Hanmi Semi", "196170.KQ": "Alteogen",
+    "352820.KS": "HYBE", "035900.KQ": "JYP Ent", "000270.KS": "Kia", "010950.KS": "S-Oil",
+    "032830.KQ": "Samsung Life", "329180.KS": "Hyundai Heavy", "015760.KS": "KEPCO",
+
+    # 🇺🇸 미국 (M7 + 성장 + 바이오 + 에너지)
     "AAPL": "Apple", "NVDA": "NVIDIA", "TSLA": "Tesla", "AMZN": "Amazon", "MSFT": "Microsoft", 
     "GOOGL": "Google", "META": "Meta", "AMD": "AMD", "NFLX": "Netflix", "INTC": "Intel",
+    "PLTR": "Palantir", "COIN": "Coinbase", "U": "Unity", "RBLX": "Roblox", "SOFI": "SoFi", 
+    "PYPL": "PayPal", "SQ": "Block", "UBER": "Uber", "ABNB": "Airbnb", "HOOD": "Robinhood",
+    "LLY": "Eli Lilly", "NVO": "Novo Nordisk", "PFE": "Pfizer", "MRK": "Merck",
+    "XOM": "Exxon", "CVX": "Chevron", "KO": "Coca-Cola", "PEP": "Pepsi", "COST": "Costco",
+    "DIS": "Disney", "NKE": "Nike", "SBUX": "Starbucks", "MCD": "McDonalds",
+    
+    # 🇯🇵 일본
     "7203.T": "Toyota", "6758.T": "Sony", "9984.T": "SoftBank", "8035.T": "Tokyo Elec", 
-    "9988.HK": "Alibaba", "0700.HK": "Tencent", "2330.TW": "TSMC", "RELIANCE.NS": "Reliance"
+    "6861.T": "Keyence", "7974.T": "Nintendo", "8306.T": "MUFG", "6501.T": "Hitachi",
+    "8058.T": "Mitsubishi", "8001.T": "Itochu", "8031.T": "Mitsui", "6920.T": "Lasertec",
+    "4063.T": "Shin-Etsu", "7741.T": "HOYA", "6146.T": "Disco", "7267.T": "Honda",
+
+    # 🇨🇳 중국/홍콩/대만/인도
+    "9988.HK": "Alibaba", "0700.HK": "Tencent", "3690.HK": "Meituan", "1211.HK": "BYD", 
+    "1810.HK": "Xiaomi", "0941.HK": "China Mobile", "9888.HK": "Baidu", "9618.HK": "JD.com",
+    "2015.HK": "Li Auto", "9868.HK": "Xpeng", "0981.HK": "SMIC",
+    "2330.TW": "TSMC", "2454.TW": "MediaTek", "2317.TW": "Foxconn", "2308.TW": "Delta Elec",
+    "RELIANCE.NS": "Reliance", "TCS.NS": "TCS", "HDFCBANK.NS": "HDFC Bank", "INFY.NS": "Infosys"
 }
-# (스캐너용 MARKET_SAMPLES도 기존 코드 유지)
+
 MARKET_SAMPLES = {
-    "🇻🇳 Vietnam (Max 100)": ["VIC.VN", "VHM.VN", "VCB.VN", "VNM.VN", "HPG.VN", "MSN.VN", "GAS.VN", "NVL.VN", "PDR.VN", "DIG.VN", "SSI.VN", "VND.VN", "MWG.VN", "FRT.VN", "FPT.VN", "STB.VN", "TCB.VN", "VPB.VN", "VRE.VN", "DGC.VN"],
-    "🇺🇸 USA (Max 100)": ["AAPL", "NVDA", "TSLA", "AMZN", "MSFT", "GOOGL", "META", "AMD", "INTC", "PLTR", "COIN", "U", "RBLX", "SOFI", "PYPL", "SQ", "UBER", "LLY", "NVO", "JNJ", "PFE", "MRK", "XOM", "CVX", "KO", "PEP", "COST", "DIS"],
-    "🇰🇷 Korea (Max 100)": ["005930.KS", "000660.KS", "005380.KS", "000270.KS", "035420.KS", "035720.KS", "005490.KS", "086520.KQ", "247540.KQ", "207940.KS", "068270.KS", "196170.KQ", "012450.KS", "352820.KS", "032830.KQ"],
-    "🇮🇩 Indonesia (Max 50)": ["BBCA.JK", "BBRI.JK", "BMRI.JK", "BBNI.JK", "TLKM.JK", "ASII.JK", "GOTO.JK", "ADRO.JK", "MDKA.JK", "UNVR.JK", "ICBP.JK"],
-    "🇯🇵 Japan (Max 50)": ["7203.T", "6758.T", "9984.T", "8035.T", "6861.T", "7974.T", "8306.T", "6501.T", "8058.T", "6920.T"],
-    "🇨🇳 China/HK (Max 50)": ["9988.HK", "0700.HK", "3690.HK", "1211.HK", "1810.HK", "0941.HK", "9888.HK", "9618.HK"],
+    "🇻🇳 Vietnam (Max 100)": [
+        "VIC.VN", "VHM.VN", "VRE.VN", "VNM.VN", "MSN.VN", "GAS.VN", "HPG.VN", "HSG.VN",
+        "NVL.VN", "PDR.VN", "DIG.VN", "CEO.VN", "DXG.VN", "KBC.VN", "GVR.VN",
+        "VCB.VN", "TCB.VN", "VPB.VN", "MBB.VN", "STB.VN", "ACB.VN", "BID.VN", "CTG.VN",
+        "SSI.VN", "VND.VN", "VCI.VN", "MWG.VN", "FRT.VN", "DGW.VN", "FPT.VN", "DGC.VN", "VHC.VN", "SAB.VN"
+    ],
+    "🇺🇸 USA (Max 100)": [
+        "AAPL", "MSFT", "GOOGL", "AMZN", "NVDA", "META", "TSLA", "AMD", "INTC", "QCOM", 
+        "AVGO", "MU", "PLTR", "COIN", "U", "RBLX", "SOFI", "UBER", "ABNB", "HOOD", "PYPL", "SQ",
+        "LLY", "NVO", "JNJ", "PFE", "MRK", "UNH", "XOM", "CVX", "JPM", "V", "MA", 
+        "KO", "PEP", "COST", "WMT", "DIS", "NKE", "SBUX", "MCD"
+    ],
+    "🇰🇷 Korea (Max 100)": [
+        "005930.KS", "000660.KS", "005380.KS", "000270.KS", "035420.KS", "035720.KS", 
+        "005490.KS", "086520.KQ", "247540.KQ", "010120.KS", "207940.KS", "068270.KS", 
+        "196170.KQ", "012450.KS", "042700.KS", "352820.KS", "010950.KS", "032830.KQ", "329180.KS", "015760.KS"
+    ],
+    "🇮🇩 Indonesia (Max 50)": ["BBCA.JK", "BBRI.JK", "BMRI.JK", "BBNI.JK", "TLKM.JK", "ASII.JK", "GOTO.JK", "ADRO.JK", "MDKA.JK", "ANTM.JK", "UNVR.JK", "ICBP.JK", "KLBF.JK", "BUKA.JK", "PGAS.JK"],
+    "🇯🇵 Japan (Max 50)": ["7203.T", "6758.T", "9984.T", "8035.T", "6861.T", "7974.T", "8306.T", "6501.T", "8058.T", "8001.T", "8031.T", "6920.T", "4063.T", "7741.T", "6146.T", "7267.T"],
+    "🇨🇳 China/HK (Max 50)": ["9988.HK", "0700.HK", "3690.HK", "1211.HK", "1810.HK", "0941.HK", "9888.HK", "9618.HK", "2015.HK", "9868.HK", "0981.HK"],
     "🇹🇼 Taiwan (Max 30)": ["2330.TW", "2454.TW", "2317.TW", "2308.TW", "2603.TW", "2881.TW", "1301.TW"],
-    "🇮🇳 India (Max 30)": ["RELIANCE.NS", "TCS.NS", "HDFCBANK.NS", "INFY.NS", "ICICIBANK.NS", "TATAMOTORS.NS"]
+    "🇮🇳 India (Max 30)": ["RELIANCE.NS", "TCS.NS", "HDFCBANK.NS", "INFY.NS", "ICICIBANK.NS", "TATAMOTORS.NS", "BHARTIARTL.NS"]
 }
 
 # ==============================================================================
 # 🛠️ [엔진] 데이터 수집 함수들 (Robust)
 # ==============================================================================
-FAKE_HEADERS = {'User-Agent': 'Mozilla/5.0 ...'}
+FAKE_HEADERS = {
+    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36',
+    'Referer': 'https://www.google.com/'
+}
 
 def fetch_news_robust(keyword):
     summary = ""
@@ -143,6 +195,7 @@ def scrape_price_from_web(name, ticker):
     return None, None
 
 def get_price_data_robust(ticker, name):
+    # FDR
     try:
         fdr_symbol = ticker
         if ".VN" in ticker: fdr_symbol = ticker.split('.')[0]
@@ -156,12 +209,14 @@ def get_price_data_robust(ticker, name):
             if prev != 0: pct = ((curr - prev)/prev)*100
             return curr, pct, "FDR"
     except: pass
+    # Yahoo
     try:
         stock = yf.Ticker(ticker)
         price = stock.fast_info.last_price
         prev = stock.fast_info.previous_close
         if price and prev: return price, ((price - prev)/prev)*100, "Yahoo"
     except: pass
+    # Web
     price, pct = scrape_price_from_web(name, ticker)
     if price is not None: return price, pct, "Web"
     return None, None, "Fail"
@@ -182,23 +237,12 @@ def get_chart_data(ticker):
     except: pass
     return pd.DataFrame(), "None"
 
-# ==============================================================================
-# 🧠 [엔진 1] 멀티 타겟 AI 인식기 (핵심 업그레이드)
-# ==============================================================================
 def identify_targets_with_ai(user_query):
     try:
         prompt = f"""
-        [ROLE] Financial Entity Resolver. [QUERY] "{user_query}"
-        [TASK]
-        1. Identify ALL companies mentioned in the query. (e.g., "Samsung vs Apple" -> 2 companies)
-        2. If no specific company is mentioned but a sector is (e.g., "Vietnam Banks"), pick top 2-3 representative companies.
-        3. For EACH company, determine Name, Ticker, English Keyword, Native Keyword.
-        
-        [OUTPUT JSON LIST]
-        [
-            {{ "name": "Eng Name 1", "ticker": "TICKER1", "eng_key": "...", "native_key": "..." }},
-            {{ "name": "Eng Name 2", "ticker": "TICKER2", "eng_key": "...", "native_key": "..." }}
-        ]
+        [ROLE] Entity Resolver. [QUERY] "{user_query}"
+        [TASK] Identify ALL companies. [OUTPUT JSON LIST]
+        [ {{ "name": "EngName", "ticker": "TICKER", "eng_key": "Name news", "native_key": "LocalName news" }} ]
         """
         messages = [{"role": "user", "content": prompt}]
         response = client.chat.completions.create(model="Qwen/Qwen2.5-72B-Instruct", messages=messages, max_tokens=300)
@@ -235,7 +279,6 @@ with st.sidebar:
     st.divider()
     menu = st.radio("MENU", [T['menu_search'], T['menu_scanner']], index=0)
 
-# --- 1. AI 멀티 스마트 검색 ---
 if menu == T['menu_search']:
     st.subheader(T['search_title'])
     c1, c2 = st.columns([3, 1])
@@ -244,46 +287,30 @@ if menu == T['menu_search']:
 
     if btn:
         with st.status(T['status_thinking'], expanded=True) as status:
-            # 1. 다중 타겟 인식
             targets = identify_targets_with_ai(query)
-            
             if targets:
-                collected_data = [] # AI 분석용 통합 데이터
-                
-                # 탭 생성 (기업 수만큼)
-                tab_names = [t.get('name', 'Unknown') for t in targets]
-                tabs = st.tabs(tab_names)
+                collected_data = []
+                tabs = st.tabs([t.get('name', 'Unknown') for t in targets])
                 
                 for i, target in enumerate(targets):
-                    name = target.get('name')
-                    ticker = target.get('ticker')
-                    eng_key = target.get('eng_key')
-                    native_key = target.get('native_key')
+                    name, ticker = target.get('name'), target.get('ticker')
+                    eng_key, native_key = target.get('eng_key'), target.get('native_key')
                     
                     with tabs[i]:
                         st.info(f"📍 **{name} ({ticker})**")
-                        
-                        # 데이터 수집
                         h, src = get_chart_data(ticker)
                         news_data = get_polyglot_news(eng_key, native_key)
-                        
                         curr, pct = "N/A", 0
-                        has_data = False
                         
                         if not h.empty:
                             curr = h['Close'].iloc[-1]
                             if len(h) >= 2:
                                 prev = h['Close'].iloc[-2]
                                 if prev != 0: pct = ((curr - prev)/prev)*100
-                            has_data = True
                         else:
-                            # 차트 없으면 웹 가격 시도
                             c_web, p_web, s_web = get_price_data_robust(ticker, name)
-                            if c_web: 
-                                curr, pct, src = c_web, p_web, s_web
-                                has_data = True
+                            if c_web: curr, pct, src = c_web, p_web, s_web
                         
-                        # UI 표시
                         m1, m2 = st.columns(2)
                         m1.metric(T['metric_price'], f"{curr:,.0f}" if isinstance(curr, (int, float)) else curr, f"{pct:.2f}%")
                         m2.metric(T['metric_source'], src)
@@ -293,42 +320,23 @@ if menu == T['menu_search']:
                         
                         with st.expander(T['expander_news']): st.text(news_data)
                         
-                        # AI 분석용 데이터 축적
-                        collected_data.append(f"""
-                        [TARGET {i+1}] {name} ({ticker})
-                        - Price: {curr}, Change: {pct:.2f}%
-                        - News Summary: {news_data[:500]}...
-                        """)
-                
-                status.update(label="✅ 분석 완료", state="complete", expanded=False)
+                        collected_data.append(f"[TARGET {i+1}] {name} ({ticker})\nPrice: {curr}, Change: {pct:.2f}%\nNews: {news_data[:500]}...")
                 
-                # 3. 종합 분석 리포트
+                status.update(label="✅ OK", state="complete", expanded=False)
                 st.divider()
                 st.subheader(T['insight_title'])
                 
                 target_lang = T['llm_lang_instruction']
-                context_str = "\n".join(collected_data)
-                
                 prompt = f"""
-                [ROLE] Global Investment Analyst.
-                [USER QUERY] "{query}"
-                [COLLECTED DATA OF MULTIPLE TARGETS]
-                {context_str}
-                
-                [TASK]
-                1. Provide a synthesized analysis or comparison based on the user's query.
-                2. If multiple companies, compare their recent sentiment and performance.
-                3. Provide investment verdicts for each.
-                
-                [OUTPUT LANGUAGE] **{target_lang}** (Strictly output in {target_lang})
+                [ROLE] Analyst. [QUERY] "{query}"
+                [DATA] {chr(10).join(collected_data)}
+                [TASK] Comparative Analysis & Verdict. [LANG] {target_lang}.
                 """
                 msg = [{"role": "user", "content": prompt}]
                 stream = client.chat.completions.create(model="Qwen/Qwen2.5-72B-Instruct", messages=msg, stream=True)
                 st.write_stream(parse_stream(stream))
-                
-            else: st.error("AI가 기업을 찾지 못했습니다.")
+            else: st.error("AI Error")
 
-# --- 2. 시장 스캐너 ---
 elif menu == T['menu_scanner']:
     st.subheader(T['scanner_title'])
     st.caption(T['scanner_caption'])
@@ -342,11 +350,9 @@ elif menu == T['menu_scanner']:
         results = []
         bar = st.progress(0)
         
-        # Phase 1: Batch
         try: batch_data = yf.download(tickers, period="5d", progress=False)['Close']
         except: batch_data = pd.DataFrame()
         
-        # Phase 2: Loop
         for i, t in enumerate(tickers):
             name_display = TICKER_NAMES.get(t, t)
             p, c, s = None, None, None