app.py

import gradio as gr
import requests
import json
import os
from datetime import datetime, timedelta
from huggingface_hub import InferenceClient


    
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 = {
    "South Korea": "ko",
    "Japan": "ja",
    "China": "zh",
    "Russia": "ru",
    "France": "fr",
    "Germany": "de",
    "Spain": "es",
    "Italy": "it",
    "Netherlands": "nl",
    "Portugal": "pt",
    "Thailand": "th",
    "Vietnam": "vi",
    "Indonesia": "id",
    "Malaysia": "ms",
    "Saudi Arabia": "ar",
    "United Arab Emirates": "ar",
    "Egypt": "ar",
    "Morocco": "ar",
    "Greece": "el",
    "Poland": "pl",
    "Czech Republic": "cs",
    "Hungary": "hu",
    "Turkey": "tr",
    "Romania": "ro",
    "Bulgaria": "bg",
    "Croatia": "hr",
    "Serbia": "sr",
    "Slovakia": "sk",
    "Slovenia": "sl",
    "Estonia": "et",
    "Latvia": "lv",
    "Lithuania": "lt"
}

COUNTRY_LOCATIONS = {
    "United States": "United States",
    "United Kingdom": "United Kingdom",
    "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 입력은 모든 국가에 대해 그대로 사용
        if is_english(query):
            print(f"영어 검색어 감지 - 원본 사용: {query}")
            return query
            
        # South Korea 선택시 한글 입력은 그대로 사용    
        if country == "South Korea" and is_korean(query):
            print(f"한국어 검색어 - 원본 사용: {query}")
            return query
            
        # 선택된 국가가 번역 지원 국가인 경우    
        if country in COUNTRY_LANGUAGES:
            target_lang = COUNTRY_LANGUAGES[country]
            
            # Google Translate API URL
            url = f"https://translate.googleapis.com/translate_a/single"
            
            params = {
                "client": "gtx",
                "sl": "ko",  # 소스 언어: 한국어
                "tl": target_lang,  # 대상 언어
                "dt": "t",
                "q": query
            }
            
            response = requests.get(url, params=params)
            translated_text = response.json()[0][0][0]
            
            print(f"번역 결과: {query} -> {translated_text} ({country})")
            return translated_text
            
        return query
        
    except Exception as e:
        print(f"번역 오류: {str(e)}")
        return query

def is_english(text):
    return all(ord(char) < 128 for char in text.replace(' ', '').replace('-', ''))
    
def is_korean(text):
    return any('\uAC00' <= char <= '\uD7A3' for char in text)

# 사용 예시:
# query = "삼성전자"
# country = "Japan"
# translated = translate_query(query, country)
# print(f"입력: {query}, 국가: {country}, 번역: {translated}")

def is_english(text):
    return all(ord(char) < 128 for char in text.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"
    
    # 24시간 이내 날짜 범위 설정
    now = datetime.utcnow()
    yesterday = now - timedelta(days=1)
    date_range = f"{yesterday.strftime('%Y-%m-%d')},{now.strftime('%Y-%m-%d')}"
    
    # 검색어 번역
    translated_query = translate_query(query, country)
    print(f"Original query: {query}")
    print(f"Translated query: {translated_query}")
    
    payload = {
        "data": {
            "q": query,
            "domain": "google.com",
            "loc": COUNTRY_LOCATIONS.get(country, "United States"),
            "lang": "en",
            "device": "desktop",
            "serp_type": "news",
            "page": "1",
            "num": "10",
            "date_range": date_range,  # 날짜 범위 추가
            "sort_by": "date"  # 날짜순 정렬 추가
        }
    }

    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))
        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;}
"""

# Gradio 인터페이스 수정
with gr.Blocks(theme="Nymbo/Nymbo_Theme", css=css, title="NewsAI 서비스") as iface:
    gr.Markdown("검색어를 입력하고 원하는 국가를 선택하면, 검색어와 일치하는 24시간 이내 뉴스를 최대 100개 출력합니다.")

    with gr.Column():
        with gr.Row():
            query = gr.Textbox(label="검색어")
            country = gr.Dropdown(MAJOR_COUNTRIES, label="국가", value="South Korea")
            
        # 번역 결과를 보여주는 컴포넌트 추가
        translated_query_display = gr.Markdown(visible=False)
            
        search_button = gr.Button("검색", variant="primary")

        progress = gr.Progress()
        status_message = gr.Markdown(visible=False)
        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,
                })

    def search_and_display(query, country, articles_state, progress=gr.Progress()):
        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 = [gr.update(value=translated_display, visible=True)]  # 번역 결과 표시
        
        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)
                    
                    outputs.extend([
                        gr.update(visible=True),
                        gr.update(value=f"### [{article['title']}]({article['link']})"),
                        image_update,
                        gr.update(value=f"**요약:** {article['snippet']}"),
                        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)
        outputs.append(gr.update(visible=False))
        return outputs

    search_outputs = [translated_query_display, gr.Markdown(visible=False)]  # 번역 결과 컴포넌트 추가
    for comp in article_components:
        search_outputs.extend([comp['group'], comp['title'], comp['image'],
                             comp['snippet'], comp['info']])
    search_outputs.extend([articles_state, status_message])

    search_button.click(
        search_and_display,
        inputs=[query, country, articles_state],
        outputs=search_outputs,
        show_progress=True
    )

iface.launch()