첫 커밋

FIXED.py

@@ -0,0 +1,202 @@
+import requests
+import json
+import os
+from geopy.distance import geodesic
+from flask import Flask, request, jsonify
+from flask_cors import CORS
+import logging
+from langchain.chains.question_answering import load_qa_chain
+from langchain.prompts import PromptTemplate
+from langchain_openai import ChatOpenAI, OpenAIEmbeddings
+from langchain.vectorstores import FAISS
+from langchain.text_splitter import CharacterTextSplitter
+from langchain.document_loaders import PyPDFLoader
+import openai
+import wikipedia
+
+import API_KEYS
+
+# Initialize Flask app and CORS
+app = Flask(__name__)
+CORS(app)  # 모든 출처에서의 요청 허용
+
+# Set up logging
+logging.basicConfig(level=logging.DEBUG)  # Change to logging.INFO or logging.ERROR in production
+logger = logging.getLogger(__name__)
+
+# OpenAI and Google API keys
+OPENAI_API_KEY = API_KEYS.OPENAI_API_KEY # 환경 변수에서 API 키를 가져옵니다.
+GOOGLE_PLACES_API_KEY = API_KEYS.GOOGLE_MAPS_API_KEY  # 환경 변수에서 Google Places API 키를 가져옵니다.
+
+# PDF Loading and Text Splitting
+pdf_path = 'seoul.pdf'
+if not os.path.exists(pdf_path):
+    logger.error(f"{pdf_path} file not found.")
+    raise FileNotFoundError(f"{pdf_path} file not found.")
+
+pdf_loader = PyPDFLoader(pdf_path)
+pdf_docs = pdf_loader.load()
+
+# Split text and prepare FAISS vector store
+text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
+texts = []
+for doc in pdf_docs:
+    if hasattr(doc, 'page_content') and isinstance(doc.page_content, str):
+        texts.extend(text_splitter.split_text(doc.page_content))
+    else:
+        logger.warning("Document does not have valid page content.")
+
+embeddings = OpenAIEmbeddings(openai_api_key=OPENAI_API_KEY)
+vectorstore = FAISS.from_texts(texts, embeddings)
+
+# Set up Wikipedia API
+wikipedia.set_lang('ko')
+
+
+# RAG-based description generation function
+def generate_description_with_rag(name, place_type):
+    try:
+        relevant_docs = vectorstore.similarity_search(f"{name} {place_type}")
+        qa_chain = load_qa_chain(
+            llm=ChatOpenAI(openai_api_key=OPENAI_API_KEY, temperature=0.7),
+            chain_type="stuff"
+        )
+
+        prompt = PromptTemplate(
+            input_variables=["name", "place_type"],
+            template=f"당신은 유능한 관광지 전문가입니다. '{name}'({place_type})에 대해 두 문장 이내로, 그곳에 가고싶게 설명해주세요. 정보가 없으면, 고객을 절대 속이지 말고 'False'만 반환하세요."
+        )
+
+        formatted_prompt = prompt.format(name=name, place_type=place_type)
+        response = qa_chain.run(input_documents=relevant_docs, question=formatted_prompt)
+
+        logger.debug(f"Generated description for {name}: {response}")
+        return response
+    except openai.APIStatusError as e:
+        logger.error(f"OpenAI API error: {str(e)}")
+        return f"OpenAI API error: {str(e)}"
+
+
+# Google Places API function
+def search_nearby_places(lat, lon, user_preferences):
+    results = []
+    category_mapping = {
+        "culturalTourism": ["museum", "art_gallery"],
+        "historicalTourism": ["church", "historical_place"],
+        "shopping": ["shopping_mall"],
+        "leisureSports": ["gym", "stadium"],
+        "restaurant": ["restaurant", "cafe"],
+        "natureTourism": ["park"]
+    }
+    excluded_types = {"lodging", "hotel", "motel"}
+
+    for preference, types in user_preferences.items():
+        if types > 0.5:
+            for place_type in category_mapping.get(preference, []):
+                url = f"https://maps.googleapis.com/maps/api/place/nearbysearch/json?location={lat},{lon}&radius=1500&type={place_type}&language=ko&key={GOOGLE_PLACES_API_KEY}"
+
+                # Logging the request URL for debugging
+                logger.debug(f"Request URL for {place_type}: {url}")
+
+                # Send the request
+                response = requests.get(url)
+
+                # Log response status and content
+                logger.debug(f"Response status: {response.status_code}")
+                logger.debug(f"Response content: {response.text}")
+
+                if response.status_code != 200:
+                    logger.error(f"Google Places API returned an error for {place_type}: {response.status_code}")
+                    continue
+
+                data = response.json()
+                if data.get("status") != "OK":
+                    logger.warning(f"Google Places API returned status {data.get('status')} for {place_type}")
+                    continue
+
+                for place in data.get('results', []):
+                    if any(ptype in excluded_types for ptype in place.get("types", [])):
+                        continue
+                    if place.get('user_ratings_total', 0) >= 15:
+                        results.append({
+                            'name': place['name'],
+                            'type': place_type,
+                            'distance': geodesic((lat, lon), (
+                                place['geometry']['location']['lat'],
+                                place['geometry']['location']['lng']
+                            )).kilometers,
+                            'source': 'Google',
+                            'place_id': place['place_id']
+                        })
+    return results
+
+
+# Recommendation function
+def recommend_spots(user_location, user_preferences, max_distance=2, top_n=5):
+    google_places = search_nearby_places(user_location[0], user_location[1], user_preferences)
+    logger.info(f"Google Places Results: {google_places}")
+
+    recommendations = sorted(google_places, key=lambda x: x.get('distance'))
+
+    for place in recommendations[:top_n]:
+        description = generate_description_with_rag(place['name'], place['type'])
+        if description == "False":
+            try:
+                candidate_description = wikipedia.page(place['name'])
+                description = candidate_description.summary + " (from Wikipedia)"
+            except (wikipedia.exceptions.PageError, wikipedia.exceptions.DisambiguationError):
+                description = "정보가 업데이트 중입니다!"
+        place['description'] = description
+        logger.debug(f"Final description for {place['name']}: {description}")
+
+    return recommendations[:top_n]
+
+
+@app.route('/recommend', methods=['POST'])
+def recommend():
+    try:
+        logger.debug("Received a new request for recommendation.")
+
+        # JSON 데이터 수신 확인
+        data = request.get_json()
+        if data is None:
+            logger.error("No JSON data received.")
+            return jsonify({"error": "No JSON data received."}), 400
+        logger.debug(f"Request Data: {data}")
+
+        # 사용자 위치 데이터 확인
+        user_latitude = data.get('latitude')
+        user_longitude = data.get('longitude')
+        if not user_latitude or not user_longitude:
+            logger.error("User location (latitude/longitude) is missing.")
+            return jsonify({"error": "User location is missing."}), 400
+
+        # 사용자 위치 및 기본 사용자 선호도 설정
+        user_location = (user_latitude, user_longitude)
+        user_location = (37.496486063, 127.028361548)
+        user_preferences = data.get('user_preferences', {
+            "culturalTourism": 0.2,
+            "historicalTourism": 0.5,
+            "shopping": 0.3,
+            "leisureSports": 0.5,
+            "restaurant": 0.1,
+            "natureTourism": 0.7
+        })
+        logger.debug(f"User Location: {user_location}")
+        logger.debug(f"User Preferences: {user_preferences}")
+
+        # 추천 장소 생성 함수 호출
+        recommended_spots = recommend_spots(user_location, user_preferences)
+        logger.info(f"Recommended Spots: {recommended_spots}")
+
+        # 추천 결과 반환
+        return jsonify({"results": recommended_spots})
+
+    except Exception as e:
+        logger.error(f"Error occurred during recommendation processing: {str(e)}")
+        return jsonify({"error": str(e)}), 500
+
+
+
+if __name__ == '__main__':
+    app.run(debug=False, port=5000)

crawler.py

@@ -0,0 +1,73 @@
+import requests
+from sympy.integrals.meijerint_doc import category
+
+"""
+from bs4 import BeautifulSoup
+
+headers = {
+    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64)AppleWebKit/537.36 (KHTML, like Gecko) Chrome/73.0.3683.86 Safari/537.36'}
+
+# 이 이후로 필요한 부분 추출
+data = requests.get('https://korean.visitseoul.net/attractions?srchType=&srchOptnCode=&srchCtgry=68&sortOrder=&srchWord=&radioOptionLike=TURSM_AREA_8',headers=headers)
+soup = BeautifulSoup(data.text, 'html.parser')
+
+print(soup.prettify())
+
+title = soup.select_one('#li.item:nth-child(1) > a:nth-child(1) > div:nth-child(2) > div:nth-child(1) > span:nth-child(1)')
+print(title)
+"""
+
+from bs4 import BeautifulSoup
+
+headers = {
+    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64)AppleWebKit/537.36 (KHTML, like Gecko) Chrome/73.0.3683.86 Safari/537.36'
+}
+
+category = ["카페&디저트", "주점", "한식", "양식", "중식", "일식", "아시아식", "서양식", "채식", "할랄"]
+
+# BeautifulSoup 객체 생성
+for i in range(1, 60):
+    try:
+        html = requests.get(f'https://korean.visitseoul.net/area?curPage={i}&srchType=&srchOptnCode=&srchCtgry=98&sortOrder=&srchWord=&radioOptionLike=TURSM_AREA_8',headers=headers)
+    except requests.RequestException as e:
+        print(f'Error: {e}')
+        break
+
+    soup = BeautifulSoup(html.text, 'html.parser')
+
+    # 모든 `li` 요소 찾기
+    items = soup.find_all('li', class_='item')
+
+    # 정보 추출
+    for item in items:
+        title = item.find('span', class_='title')
+        description = item.find('span', class_='small-text text-dot-d')
+
+        # 텍스트가 있으면 출력
+        if title and description:
+            print("Title:", title.get_text(strip=True))
+            print("Location: 기타")
+            print("Description:", description.get_text(strip=True))
+            print('-' * 30)
+"""
+headers = {
+    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64)AppleWebKit/537.36 (KHTML, like Gecko) Chrome/73.0.3683.86 Safari/537.36'
+}
+
+html = requests.get(f'https://korean.visitseoul.net/attractions?curPage=2&srchType=&srchOptnCode=&srchCtgry=69&sortOrder=&srchWord=&radioOptionLike=TURSM_AREA_8',headers=headers)
+soup = BeautifulSoup(html.text, 'html.parser')
+
+# 모든 `li` 요소 찾기
+items = soup.find_all('li', class_='item')
+
+# 정보 추출
+for item in items:
+    title = item.find('span', class_='title')
+    description = item.find('span', class_='small-text text-dot-d')
+
+    # 텍스트가 있으면 출력
+    if title and description:
+        print("Title:", title.get_text(strip=True))
+        print("Description:", description.get_text(strip=True))
+        print('-' * 30)
+"""