Spaces:
Sleeping
Sleeping
| import os | |
| import json | |
| from dotenv import load_dotenv | |
| from google import genai | |
| from google.genai import types | |
| from typing import List, Dict, Any, Optional | |
| # 載入環境變數 | |
| load_dotenv() | |
| class GeminiService: | |
| def __init__(self): | |
| api_key = os.getenv("GEMINI_API_KEY") | |
| if not api_key: | |
| print("警告:找不到 GEMINI_API_KEY") | |
| self.client = genai.Client(api_key=api_key) if api_key else None | |
| self.model_id = os.getenv("GEMINI_MODEL_ID", "gemini-2.0-flash") | |
| def _check_client(self): | |
| if not self.client: | |
| raise ValueError("API Key 未設定") | |
| def search_companies(self, query: str, exclude_names: List[str] = []) -> List[Dict]: | |
| """ | |
| Step 1: 領域探索 -> 公司列表 | |
| """ | |
| self._check_client() | |
| exclusion_prompt = "" | |
| if exclude_names: | |
| exclusion_prompt = f"IMPORTANT: Do not include: {', '.join(exclude_names)}." | |
| # Phase 1: Google Search (廣泛探索) | |
| # 這裡的 Prompt 強調:如果使用者輸入的是「領域(如: AI)」,請列出該領域的台灣代表性公司。 | |
| search_prompt = f""" | |
| Using Google Search, find 5 to 10 prominent companies in Taiwan related to the query: "{query}". | |
| **Instructions:** | |
| 1. **Domain Search:** If "{query}" is an industry or technology (e.g., "AI", "Green Energy"), list the top representative Taiwanese companies in this field. | |
| 2. **Company Search:** If "{query}" is a specific name, list that company and its direct competitors. | |
| 3. **Target:** Focus on Taiwanese companies (or global companies with major R&D in Taiwan). | |
| {exclusion_prompt} | |
| List them (Full Name - Industry/Main Product) in Traditional Chinese. | |
| """ | |
| search_response = self.client.models.generate_content( | |
| model=self.model_id, | |
| contents=search_prompt, | |
| config=types.GenerateContentConfig( | |
| tools=[types.Tool(google_search=types.GoogleSearch())] | |
| ) | |
| ) | |
| raw_text = search_response.text | |
| # Phase 2: Extract JSON (結構化) | |
| extract_prompt = f""" | |
| From the text below, extract company names and their industry/main product. | |
| Calculate a Relevance Score (0-100) based on query: "{query}". | |
| Return ONLY a JSON array: [{{"name": "...", "industry": "...", "relevanceScore": 85}}] | |
| Text: | |
| --- | |
| {raw_text} | |
| --- | |
| """ | |
| extract_response = self.client.models.generate_content( | |
| model=self.model_id, | |
| contents=extract_prompt, | |
| config=types.GenerateContentConfig( | |
| response_mime_type='application/json' | |
| ) | |
| ) | |
| try: | |
| return json.loads(extract_response.text) | |
| except Exception as e: | |
| print(f"JSON Parse Error: {e}") | |
| return [] | |
| def get_company_details(self, company: Dict) -> Dict: | |
| """ | |
| Step 2: 進行商業徵信調查 (Deep Dive) | |
| """ | |
| self._check_client() | |
| name = company.get('name') | |
| prompt = f""" | |
| Act as a professional "Business Analyst & Investigator". | |
| Conduct a comprehensive investigation on the Taiwanese company: "{name}". | |
| **Investigation Targets:** | |
| 1. **Overview (基本盤)**: | |
| - **Tax ID (統編)** & **Capital (資本額)**. (Try to find specific numbers) | |
| - **Representative (代表人)**. | |
| - **Core Business**: What specific problem do they solve? What is their "Ace" product? | |
| 2. **Workforce & Culture (內部情報)**: | |
| - **Employee Count**. | |
| - **Reviews/Gossip**: Search **PTT (Tech_Job, Soft_Job)**, **Dcard**, **Qollie**. | |
| - Summarize the *REAL* work vibe (e.g., "Good for juniors but low ceiling", "Free snacks but forced overtime"). | |
| 3. **Legal & Risks (排雷專區)**: | |
| - Search: "{name} 勞資糾紛", "{name} 違反勞基法", "{name} 判決", "{name} 罰款". | |
| - List any red flags found in government records or news. | |
| **Format**: | |
| - Use Markdown. | |
| - Language: Traditional Chinese (繁體中文). | |
| - Be objective but don't sugarcoat potential risks. | |
| """ | |
| response = self.client.models.generate_content( | |
| model=self.model_id, | |
| contents=prompt, | |
| config=types.GenerateContentConfig( | |
| tools=[types.Tool(google_search=types.GoogleSearch())] | |
| ) | |
| ) | |
| # Extract Sources | |
| sources = [] | |
| if response.candidates[0].grounding_metadata and response.candidates[0].grounding_metadata.grounding_chunks: | |
| for chunk in response.candidates[0].grounding_metadata.grounding_chunks: | |
| if chunk.web and chunk.web.uri and chunk.web.title: | |
| sources.append({"title": chunk.web.title, "uri": chunk.web.uri}) | |
| unique_sources = {v['uri']: v for v in sources}.values() | |
| return { | |
| "text": response.text, | |
| "sources": list(unique_sources) | |
| } | |
| def chat_with_ai(self, history: List[Dict], new_message: str, context: str) -> str: | |
| self._check_client() | |
| system_instruction = f"You are an expert Business Consultant. Answer based on this company report:\n{context}" | |
| chat_history = [] | |
| for h in history: | |
| role = "user" if h["role"] == "user" else "model" | |
| chat_history.append(types.Content(role=role, parts=[types.Part(text=h["content"])])) | |
| chat = self.client.chats.create( | |
| model=self.model_id, | |
| history=chat_history, | |
| config=types.GenerateContentConfig( | |
| system_instruction=system_instruction | |
| ) | |
| ) | |
| response = chat.send_message(new_message) | |
| return response.text |