Create app-backup2.py

app-backup2.py

@@ -0,0 +1,574 @@
+import gradio as gr
+import requests
+import json
+import os
+from datetime import datetime, timedelta
+from huggingface_hub import InferenceClient
+
+MAX_COUNTRY_RESULTS = 100    # 국가별 최대 결과 수
+MAX_GLOBAL_RESULTS = 1000    # 전세계 최대 결과 수
+
+def create_article_components(max_results):
+    article_components = []
+    for i in range(max_results):
+        with gr.Group(visible=False) as article_group:
+            title = gr.Markdown()
+            image = gr.Image(width=200, height=150)
+            snippet = gr.Markdown()
+            info = gr.Markdown()
+            
+            article_components.append({
+                'group': article_group,
+                'title': title,
+                'image': image,
+                'snippet': snippet,
+                'info': info,
+                'index': i,
+            })
+    return article_components
+    
+API_KEY = os.getenv("SERPHOUSE_API_KEY")
+# hf_client = InferenceClient("CohereForAI/c4ai-command-r-plus-08-2024", token=os.getenv("HF_TOKEN"))
+
+# 국가별 언어 코드 매핑
+COUNTRY_LANGUAGES = {
+    "United States": "en",
+    "United Kingdom": "en",
+    "Taiwan": "zh-TW",  # 대만어(번체 중국어)
+    "Canada": "en",
+    "Australia": "en",
+    "Germany": "de",
+    "France": "fr",
+    "Japan": "ja",
+    "South Korea": "ko",
+    "China": "zh",
+    "India": "hi",
+    "Brazil": "pt",
+    "Mexico": "es",
+    "Russia": "ru",
+    "Italy": "it",
+    "Spain": "es",
+    "Netherlands": "nl",
+    "Singapore": "en",
+    "Hong Kong": "zh-HK",
+    "Indonesia": "id",
+    "Malaysia": "ms",
+    "Philippines": "tl",
+    "Thailand": "th",
+    "Vietnam": "vi",
+    "Belgium": "nl",
+    "Denmark": "da",
+    "Finland": "fi",
+    "Ireland": "en",
+    "Norway": "no",
+    "Poland": "pl",
+    "Sweden": "sv",
+    "Switzerland": "de",
+    "Austria": "de",
+    "Czech Republic": "cs",
+    "Greece": "el",
+    "Hungary": "hu",
+    "Portugal": "pt",
+    "Romania": "ro",
+    "Turkey": "tr",
+    "Israel": "he",
+    "Saudi Arabia": "ar",
+    "United Arab Emirates": "ar",
+    "South Africa": "en",
+    "Argentina": "es",
+    "Chile": "es",
+    "Colombia": "es",
+    "Peru": "es",
+    "Venezuela": "es",
+    "New Zealand": "en",
+    "Bangladesh": "bn",
+    "Pakistan": "ur",
+    "Egypt": "ar",
+    "Morocco": "ar",
+    "Nigeria": "en",
+    "Kenya": "sw",
+    "Ukraine": "uk",
+    "Croatia": "hr",
+    "Slovakia": "sk",
+    "Bulgaria": "bg",
+    "Serbia": "sr",
+    "Estonia": "et",
+    "Latvia": "lv",
+    "Lithuania": "lt",
+    "Slovenia": "sl",
+    "Luxembourg": "fr",
+    "Malta": "mt",
+    "Cyprus": "el",
+    "Iceland": "is"
+}
+
+COUNTRY_LOCATIONS = {
+    "United States": "United States",
+    "United Kingdom": "United Kingdom",
+    "Taiwan": "Taiwan",  # 국가명 사용
+    "Canada": "Canada",
+    "Australia": "Australia",
+    "Germany": "Germany",
+    "France": "France",
+    "Japan": "Japan",
+    "South Korea": "South Korea",
+    "China": "China",
+    "India": "India",
+    "Brazil": "Brazil",
+    "Mexico": "Mexico",
+    "Russia": "Russia",
+    "Italy": "Italy",
+    "Spain": "Spain",
+    "Netherlands": "Netherlands",
+    "Singapore": "Singapore",
+    "Hong Kong": "Hong Kong",
+    "Indonesia": "Indonesia",
+    "Malaysia": "Malaysia",
+    "Philippines": "Philippines",
+    "Thailand": "Thailand",
+    "Vietnam": "Vietnam",
+    "Belgium": "Belgium",
+    "Denmark": "Denmark",
+    "Finland": "Finland",
+    "Ireland": "Ireland",
+    "Norway": "Norway",
+    "Poland": "Poland",
+    "Sweden": "Sweden",
+    "Switzerland": "Switzerland",
+    "Austria": "Austria",
+    "Czech Republic": "Czech Republic",
+    "Greece": "Greece",
+    "Hungary": "Hungary",
+    "Portugal": "Portugal",
+    "Romania": "Romania",
+    "Turkey": "Turkey",
+    "Israel": "Israel",
+    "Saudi Arabia": "Saudi Arabia",
+    "United Arab Emirates": "United Arab Emirates",
+    "South Africa": "South Africa",
+    "Argentina": "Argentina",
+    "Chile": "Chile",
+    "Colombia": "Colombia",
+    "Peru": "Peru",
+    "Venezuela": "Venezuela",
+    "New Zealand": "New Zealand",
+    "Bangladesh": "Bangladesh",
+    "Pakistan": "Pakistan",
+    "Egypt": "Egypt",
+    "Morocco": "Morocco",
+    "Nigeria": "Nigeria",
+    "Kenya": "Kenya",
+    "Ukraine": "Ukraine",
+    "Croatia": "Croatia",
+    "Slovakia": "Slovakia",
+    "Bulgaria": "Bulgaria",
+    "Serbia": "Serbia",
+    "Estonia": "Estonia",
+    "Latvia": "Latvia",
+    "Lithuania": "Lithuania",
+    "Slovenia": "Slovenia",
+    "Luxembourg": "Luxembourg",
+    "Malta": "Malta",
+    "Cyprus": "Cyprus",
+    "Iceland": "Iceland"
+}
+
+MAJOR_COUNTRIES = list(COUNTRY_LOCATIONS.keys())
+
+def translate_query(query, country):
+    try:
+        # 영어 입력 확인
+        if is_english(query):
+            print(f"영어 검색어 감지 - 원본 사용: {query}")
+            return query
+            
+        # 선택된 국가가 번역 지원 국가인 경우    
+        if country in COUNTRY_LANGUAGES:
+            # South Korea 선택시 한글 입력은 그대로 사용    
+            if country == "South Korea":
+                print(f"한국 선택 - 원본 사용: {query}")
+                return query
+                
+            target_lang = COUNTRY_LANGUAGES[country]
+            print(f"번역 시도: {query} -> {country}({target_lang})")
+            
+            url = f"https://translate.googleapis.com/translate_a/single"
+            params = {
+                "client": "gtx",
+                "sl": "auto",
+                "tl": target_lang,
+                "dt": "t",
+                "q": query
+            }
+            
+            response = requests.get(url, params=params)
+            translated_text = response.json()[0][0][0]
+            print(f"번역 완료: {query} -> {translated_text} ({country})")
+            return translated_text
+            
+        return query
+        
+    except Exception as e:
+        print(f"번역 오류: {str(e)}")
+        return query
+
+def translate_to_korean(text):
+    try:
+        url = "https://translate.googleapis.com/translate_a/single"
+        params = {
+            "client": "gtx",
+            "sl": "auto",
+            "tl": "ko",
+            "dt": "t",
+            "q": text
+        }
+        
+        response = requests.get(url, params=params)
+        translated_text = response.json()[0][0][0]
+        return translated_text
+    except Exception as e:
+        print(f"한글 번역 오류: {str(e)}")
+        return text
+
+def is_english(text):
+    return all(ord(char) < 128 for char in text.replace(' ', '').replace('-', '').replace('_', ''))
+    
+def is_korean(text):
+    return any('\uAC00' <= char <= '\uD7A3' for char in text)
+    
+def search_serphouse(query, country, page=1, num_result=10):
+    url = "https://api.serphouse.com/serp/live"
+    
+    now = datetime.utcnow()
+    yesterday = now - timedelta(days=1)
+    date_range = f"{yesterday.strftime('%Y-%m-%d')},{now.strftime('%Y-%m-%d')}"
+    
+    translated_query = translate_query(query, country)
+    print(f"Original query: {query}")
+    print(f"Translated query: {translated_query}")
+    
+    payload = {
+        "data": {
+            "q": translated_query,
+            "domain": "google.com",
+            "loc": COUNTRY_LOCATIONS.get(country, "United States"),
+            "lang": COUNTRY_LANGUAGES.get(country, "en"),
+            "device": "desktop",
+            "serp_type": "news",
+            "page": "1",
+            "num": "10",
+            "date_range": date_range,
+            "sort_by": "date"
+        }
+    }
+
+    headers = {
+        "accept": "application/json",
+        "content-type": "application/json",
+        "authorization": f"Bearer {API_KEY}"
+    }
+
+    try:
+        response = requests.post(url, json=payload, headers=headers)
+        print("Request payload:", json.dumps(payload, indent=2, ensure_ascii=False))
+        print("Response status:", response.status_code)
+        
+        response.raise_for_status()
+        return {"results": response.json(), "translated_query": translated_query}
+    except requests.RequestException as e:
+        return {"error": f"Error: {str(e)}", "translated_query": query}
+
+def format_results_from_raw(response_data):
+    if "error" in response_data:
+        return "Error: " + response_data["error"], []
+
+    try:
+        results = response_data["results"]
+        translated_query = response_data["translated_query"]
+        
+        news_results = results.get('results', {}).get('results', {}).get('news', [])
+        if not news_results:
+            return "검색 결과가 없습니다.", []
+
+        articles = []
+        for idx, result in enumerate(news_results, 1):
+            articles.append({
+                "index": idx,
+                "title": result.get("title", "제목 없음"),
+                "link": result.get("url", result.get("link", "#")),
+                "snippet": result.get("snippet", "내용 없음"),
+                "channel": result.get("channel", result.get("source", "알 수 없음")),
+                "time": result.get("time", result.get("date", "알 수 없는 시간")),
+                "image_url": result.get("img", result.get("thumbnail", "")),
+                "translated_query": translated_query
+            })
+        return "", articles
+    except Exception as e:
+        return f"결과 처리 중 오류 발생: {str(e)}", []
+
+def serphouse_search(query, country):
+    response_data = search_serphouse(query, country)
+    return format_results_from_raw(response_data)
+
+css = """
+footer {visibility: hidden;}
+"""
+
+with gr.Blocks(theme="Nymbo/Nymbo_Theme", css=css, title="NewsAI 서비스") as iface:
+    with gr.Tabs():
+        # 국가별 탭
+        with gr.Tab("국가별"):
+            gr.Markdown("검색어를 입력하고 원하는 국가(67개국)를 선택하면, 검색어와 일치하는 24시간 이내 뉴스를 최대 100개 출력합니다.")
+            gr.Markdown("국가 선택후 검색어에 '한글'을 입력하면 현지 언어로 번역되어 검색합니다. 예: 'Taiwan' 국가 선택후 '삼성' 입력시 '三星'으로 자동 검색")
+            
+            with gr.Column():
+                with gr.Row():
+                    query = gr.Textbox(label="검색어")
+                    country = gr.Dropdown(MAJOR_COUNTRIES, label="국가", value="South Korea")
+                
+                # 검색 상태 메시지    
+                status_message = gr.Markdown("", visible=True)
+                
+                # 번역 결과를 보여주는 컴포넌트
+                translated_query_display = gr.Markdown(visible=False)
+                
+                search_button = gr.Button("검색", variant="primary")
+                
+                progress = gr.Progress()
+                articles_state = gr.State([])
+
+                article_components = []
+                for i in range(100):
+                    with gr.Group(visible=False) as article_group:
+                        title = gr.Markdown()
+                        image = gr.Image(width=200, height=150)
+                        snippet = gr.Markdown()
+                        info = gr.Markdown()
+                        
+                        article_components.append({
+                            'group': article_group,
+                            'title': title,
+                            'image': image,
+                            'snippet': snippet,
+                            'info': info,
+                            'index': i,
+                        })
+
+        # 전세계 탭
+        with gr.Tab("전세계"):
+            gr.Markdown("검색어를 입력하면 67개국의 24시간 이내 뉴스를 최대 1000개 출력합니다.")
+            
+            with gr.Column():
+                query_global = gr.Textbox(label="검색어")
+                status_message_global = gr.Markdown("", visible=True)
+                translated_query_display_global = gr.Markdown(visible=False)
+                search_button_global = gr.Button("전세계 검색", variant="primary")
+                
+                progress_global = gr.Progress()
+                articles_state_global = gr.State([])
+
+                global_article_components = []
+                for i in range(1000):
+                    with gr.Group(visible=False) as article_group:
+                        title = gr.Markdown()
+                        image = gr.Image(width=200, height=150)
+                        snippet = gr.Markdown()
+                        info = gr.Markdown()
+                        
+                        global_article_components.append({
+                            'group': article_group,
+                            'title': title,
+                            'image': image,
+                            'snippet': snippet,
+                            'info': info,
+                            'index': i,
+                        })
+
+    def search_and_display(query, country, articles_state, progress=gr.Progress()):
+        # 검색 상태 메시지 업데이트
+        status_msg = "검색을 진행중입니다. 잠시만 기다리세요..."
+        
+        progress(0, desc="검색어 번역 중...")
+        
+        # 검색어 번역
+        translated_query = translate_query(query, country)
+        translated_display = f"**원본 검색어:** {query}\n**번역된 검색어:** {translated_query}" if translated_query != query else f"**검색어:** {query}"
+        
+        progress(0.2, desc="검색 시작...")
+        error_message, articles = serphouse_search(query, country)
+        progress(0.5, desc="결과 처리 중...")
+        
+        outputs = []
+        outputs.append(gr.update(value=status_msg, visible=True))  # 상태 메시지
+        outputs.append(gr.update(value=translated_display, visible=True))  # 번역 결과
+        
+        if error_message:
+            outputs.append(gr.update(value=error_message, visible=True))
+            for comp in article_components:
+                outputs.extend([
+                    gr.update(visible=False), gr.update(), gr.update(),
+                    gr.update(), gr.update()
+                ])
+            articles_state = []
+        else:
+            outputs.append(gr.update(value="", visible=False))
+            total_articles = len(articles)
+            for idx, comp in enumerate(article_components):
+                progress((idx + 1) / total_articles, desc=f"결과 표시 중... {idx + 1}/{total_articles}")
+                if idx < len(articles):
+                    article = articles[idx]
+                    image_url = article['image_url']
+                    image_update = gr.update(value=image_url, visible=True) if image_url and not image_url.startswith('data:image') else gr.update(value=None, visible=False)
+                    
+                    # 요약 내용 한글 번역
+                    korean_summary = translate_to_korean(article['snippet'])
+                    
+                    outputs.extend([
+                        gr.update(visible=True),
+                        gr.update(value=f"### [{article['title']}]({article['link']})"),
+                        image_update,
+                        gr.update(value=f"**요약:** {article['snippet']}\n\n**한글 요약:** {korean_summary}"),
+                        gr.update(value=f"**출처:** {article['channel']} | **시간:** {article['time']}")
+                    ])
+                else:
+                    outputs.extend([
+                        gr.update(visible=False), gr.update(), gr.update(),
+                        gr.update(), gr.update()
+                    ])
+            articles_state = articles
+
+        progress(1.0, desc="완료!")
+        outputs.append(articles_state)
+        
+        # 검색 완료 후 상태 메시지 숨김
+        outputs[0] = gr.update(value="", visible=False)
+        
+        return outputs
+
+    def search_global(query, articles_state_global, progress=gr.Progress()):
+        status_msg = "전세계 검색을 진행중입니다. 실시간으로 결과가 출력됩니다..."
+        all_results = []
+        displayed_count = 0
+        
+        # 초기 출력 설정
+        outputs = []
+        outputs.append(gr.update(value=status_msg, visible=True))
+        outputs.append(gr.update(value=f"**검색어:** {query}", visible=True))
+        
+        # 1000개의 결과 컴포넌트를 초기에 모두 숨김 처리
+        for _ in global_article_components:
+            outputs.extend([
+                gr.update(visible=False), gr.update(), gr.update(),
+                gr.update(), gr.update()
+            ])
+        outputs.append([])  # articles_state_global
+        
+        yield outputs  # 초기 상태 출력
+        
+        # 국가별 검색 및 실시간 결과 출력
+        for idx, (country, location) in enumerate(COUNTRY_LOCATIONS.items()):
+            progress(idx / len(COUNTRY_LOCATIONS), f"{country} 검색 중... ({idx + 1}/{len(COUNTRY_LOCATIONS)})")
+            
+            try:
+                error_message, articles = serphouse_search(query, country)
+                if not error_message and articles:
+                    # 해당 국가의 결과에 국가 정보 추가
+                    for article in articles:
+                        article['source_country'] = country
+                    
+                    # 새로운 결과 추가 및 정렬
+                    all_results.extend(articles)
+                    sorted_results = sorted(all_results, key=lambda x: x.get('time', ''), reverse=True)
+                    
+                    # 중복 제거
+                    seen_urls = set()
+                    unique_results = []
+                    for article in sorted_results:
+                        url = article.get('link', '')
+                        if url not in seen_urls:
+                            seen_urls.add(url)
+                            unique_results.append(article)
+                    
+                    # 최대 1000개까지만 유지
+                    unique_results = unique_results[:1000]
+                    
+                    # 새로운 결과 출력
+                    outputs = []
+                    outputs.append(gr.update(value=f"전세계 검색 진행 중... ({idx + 1}/{len(COUNTRY_LOCATIONS)} 국가 완료)", visible=True))
+                    outputs.append(gr.update(value=f"**검색어:** {query}\n**현재까지 발견된 뉴스:** {len(unique_results)}건", visible=True))
+                    
+                    # 결과 컴포넌트 업데이트
+                    for idx, comp in enumerate(global_article_components):
+                        if idx < len(unique_results):
+                            article = unique_results[idx]
+                            image_url = article.get('image_url', '')
+                            image_update = gr.update(value=image_url, visible=True) if image_url and not image_url.startswith('data:image') else gr.update(value=None, visible=False)
+                            
+                            korean_summary = translate_to_korean(article['snippet'])
+                            
+                            outputs.extend([
+                                gr.update(visible=True),
+                                gr.update(value=f"### [{article['title']}]({article['link']})"),
+                                image_update,
+                                gr.update(value=f"**요약:** {article['snippet']}\n\n**한글 요약:** {korean_summary}"),
+                                gr.update(value=f"**출처:** {article['channel']} | **국가:** {article['source_country']} | **시간:** {article['time']}")
+                            ])
+                        else:
+                            outputs.extend([
+                                gr.update(visible=False), gr.update(), gr.update(),
+                                gr.update(), gr.update()
+                            ])
+                    
+                    outputs.append(unique_results)
+                    yield outputs
+                    
+            except Exception as e:
+                print(f"Error searching {country}: {str(e)}")
+                continue
+        
+        # 최종 상태 업데이트
+        final_status = f"검색 완료! 총 {len(unique_results)}개의 뉴스가 발견되었습니다."
+        outputs[0] = gr.update(value=final_status, visible=True)
+        yield outputs
+
+    search_outputs = [
+        status_message,
+        translated_query_display,
+        gr.Markdown(visible=False)
+    ]
+    
+    for comp in article_components:
+        search_outputs.extend([
+            comp['group'], comp['title'], comp['image'],
+            comp['snippet'], comp['info']
+        ])
+    search_outputs.append(articles_state)
+
+    search_button.click(
+        search_and_display,
+        inputs=[query, country, articles_state],
+        outputs=search_outputs,
+        show_progress=True
+    )
+
+    # 전세계 검색 이벤트 연결
+    global_search_outputs = [
+        status_message_global,
+        translated_query_display_global,
+    ]
+    
+    for comp in global_article_components:
+        global_search_outputs.extend([
+            comp['group'], comp['title'], comp['image'],
+            comp['snippet'], comp['info']
+        ])
+    global_search_outputs.append(articles_state_global)
+
+    search_button_global.click(
+        search_global,
+        inputs=[query_global, articles_state_global],
+        outputs=global_search_outputs,
+        show_progress=True
+    )
+
+iface.launch(auth=("it1","chosun1"))