Update app.py

app.py

@@ -1,109 +1,7 @@
 import os, re, base64
 from langchain_core.documents import Document
 from langchain_chroma import Chroma
-from openai import OpenAI
-from langchain.embeddings.base import Embeddings
-from langchain_google_genai import ChatGoogleGenerativeAI
-from langchain_community.vectorstores import FAISS
-import gradio as gr
-from langchain.memory import ConversationBufferMemory
-
-# =============================================
-# 1️⃣ 內建 Embedding：使用 Gemini embedding API
-# =============================================
-from langchain_community.embeddings import HuggingFaceEmbeddings
-
-embedding = HuggingFaceEmbeddings(model_name="BAAI/bge-small-zh-v1.5")
-
-# =============================================
-# 2️⃣ 載入 QA 檔案並分類
-# =============================================
-BASE_DIR = os.path.dirname(os.path.abspath(__file__))
-path = os.path.join(BASE_DIR, "QA_v2.txt")
-
-if not os.path.exists(path):
-    raise FileNotFoundError(f"❌ 找不到 QA 檔案：{path}")
-
-with open(path, "r", encoding="utf-8") as f:
-    text = f.read()
-
-pattern = r"(Ｑ[:：].*?)(?=Ｑ[:：]|$)"
-qas = re.findall(pattern, text, flags=re.S)
-
-qa_docs = {"證券": [], "期貨": [], "複委託": []}
-for qa in qas:
-    if "證券" in qa:
-        qa_docs["證券"].append(Document(page_content=qa.strip(), metadata={"source": path}))
-    elif "期貨" in qa:
-        qa_docs["期貨"].append(Document(page_content=qa.strip(), metadata={"source": path}))
-    elif "複委託" in qa:
-        qa_docs["複委託"].append(Document(page_content=qa.strip(), metadata={"source": path}))
-
-print("✅ 已成功讀取 QA 並完成分類：", {k: len(v) for k, v in qa_docs.items()})
-
-# =============================================
-# 3️⃣ 建立向量資料庫（使用 FAISS，記憶體型）
-# =============================================
-vectordbs = {}
-for k, docs in qa_docs.items():
-    vectordbs[k] = FAISS.from_documents(docs, embedding)
-
-# =============================================
-# 4️⃣ 初始化 Gemini LLM
-# =============================================
-API_KEY = os.getenv("GOOGLE_API_KEY")
-if not API_KEY:
-    raise ValueError("⚠️ 未設定 GOOGLE_API_KEY，請在 Hugging Face Secrets 中新增。")
-
-llm = ChatGoogleGenerativeAI(model='gemini-2.5-flash', google_api_key=API_KEY)
-memory = ConversationBufferMemory(memory_key="chat_history", return_messages=True)
-
-# =============================================
-# 5️⃣ 對話邏輯
-# =============================================
-def auto_detect_category(text):
-    if any(k in text for k in ["股票", "證券", "開戶", "下單", "交割", "現股"]):
-        return "證券"
-    elif any(k in text for k in ["期貨", "選擇權", "結算", "保證金", "契約"]):
-        return "期貨"
-    elif any(k in text for k in ["複委託", "海外", "美股", "港股", "國外"]):
-        return "複委託"
-    else:
-        return "證券"
-
-def chat_fn(message, history):
-    category = auto_detect_category(message)
-    vectordb = vectordbs.get(category)
-    if not vectordb:
-        return "目前尚無此類別的知識庫。"
-
-    docs = vectordb.similarity_search(message, k=2)
-    context = "\n\n".join([d.page_content for d in docs]) if docs else "查無相關內容。"
-
-    prompt = f"""
-我是一位金融客服人員。根據以下公司規章內容回答使用者問題：
----
-{context}
----
-使用者問題：{message}
-    """
-
-    try:
-        response = llm.invoke(prompt)
-        reply = response.content.strip()
-    except Exception as e:
-        reply = f"⚠️ 生成錯誤：{e}"
-
-    return reply or "請洽營業員"
-
-# =============================================
-# 6️⃣ Gradio 介面
-# =============================================
-import os, re, base64
-from langchain_core.documents import Document
-from langchain_chroma import Chroma
-from openai import OpenAI
-from langchain.embeddings.base import Embeddings
+from langchain_huggingface import HuggingFaceEmbeddings   # ✅ 雲端可直接用
 from langchain_google_genai import ChatGoogleGenerativeAI
 import chromadb
 import gradio as gr
@@ -119,21 +17,9 @@ except ImportError:
 
 
 # =============================================
-# 1️⃣ 自訂 LM Studio Embedding 類別
+# 1️⃣ 使用 Hugging Face 雲端 embedding 模型
 # =============================================
-class LmStudioEmbeddings(Embeddings):
-    def __init__(self, model_name, url):
-        self.model_name = model_name
-        self.client = OpenAI(base_url=url, api_key="lm-studio")
-
-    def embed_query(self, text: str):
-        res = self.client.embeddings.create(input=text, model=self.model_name)
-        return res.data[0].embedding
-
-    def embed_documents(self, texts: list[str]):
-        res = self.client.embeddings.create(input=texts, model=self.model_name)
-        return [x.embedding for x in res.data]
-
+embedding = HuggingFaceEmbeddings(model_name="BAAI/bge-small-zh-v1.5")
 
 # =============================================
 # 2️⃣ 載入 QA 檔案並分類
@@ -166,13 +52,9 @@ for k, v in qa_docs.items():
 # =============================================
 # 3️⃣ 建立向量資料庫
 # =============================================
-embedding = LmStudioEmbeddings(
-    model_name="text-embedding-bge-large-zh-v1.5",
-    url="http://127.0.0.1:1234/v1"
-)
 client = chromadb.PersistentClient(path="./chroma_db")
-
 collection_names = {"證券": "stocks", "期貨": "futures", "複委託": "overseas"}
+
 vectordbs = {}
 for cat, docs in qa_docs.items():
     vectordbs[cat] = Chroma(
@@ -186,7 +68,7 @@ print("✅ 各類別向量資料庫建立完成")
 
 
 # =============================================
-# 4️⃣ 初始化 Gemini LLM
+# 4️⃣ 初始化 Gemini LLM（雲端）
 # =============================================
 API_KEY = os.getenv("GOOGLE_API_KEY")
 if not API_KEY:
@@ -233,7 +115,7 @@ def chat_fn(message, history):
 
 
 # =============================================
-# 6️⃣ 介面（LINE風格 + 純白footer + 小輸入按鈕）
+# 6️⃣ Gradio 介面（純雲端版）
 # =============================================
 logo_path = os.path.join(BASE_DIR, "mega.png")
 logo_base64 = ""
@@ -268,16 +150,10 @@ with gr.Blocks(
 
         /* ====== footer（純白背景） ====== */
         #footer {
-            position: fixed;
-            bottom: 40px;
-            left: 0;
-            width: 100%;
-            text-align: center;
-            font-size: 13px;
-            color: #aaa;                  /* ✅ 更淡的灰色 */
-            border-top: 1px solid #ddd;   /* ✅ 細灰分隔線 */
-            padding-top: 8px;
-            background-color: transparent; /* ✅ 移除反白 */
+            position: fixed; bottom: 40px; left: 0; width: 100%;
+            text-align: center; font-size: 13px; color: #aaa;
+            border-top: 1px solid #ddd; padding-top: 8px;
+            background-color: transparent;
         }
         @media (max-width: 768px) {
             #footer { position: relative; margin-top: 40px; }
@@ -309,13 +185,11 @@ with gr.Blocks(
         #send-btn:active { transform: scale(0.95); }
     """
 ) as demo:
-    # 左上角 logo
     if logo_base64:
         gr.HTML(f"""
             <div id="logo-top"><img src="data:image/png;base64,{logo_base64}" alt="logo"></div>
         """)
 
-    # 標題區
     gr.HTML("""
         <div id="main-title-wrapper" style="text-align:center; margin-top:20px;">
             <h1 id='main-title'><span>👨‍💼 我是小智&nbsp;&nbsp;您的金融好幫手 🫰</span></h1>
@@ -323,11 +197,9 @@ with gr.Blocks(
         </div>
     """)
 
-    # 聊天介面
     with gr.Row():
         with gr.Column(scale=4):
             chatbox = gr.Chatbot(label="💬 對話紀錄", type="messages")
-
             with gr.Row(elem_id="input-row"):
                 user_input = gr.Textbox(elem_id="user-input", show_label=False, placeholder="輸入訊息...", scale=8)
                 send_btn = gr.Button("輸入", elem_id="send-btn", scale=1)
@@ -363,10 +235,8 @@ with gr.Blocks(
                 return [], gr.update(value="", placeholder="輸入訊息...")
             gr.Button("🧹 整理畫面").click(clear_memory, outputs=[chatbox, user_input])
 
-    # 底部 footer（純白）
     gr.HTML("<div id='footer'>© Fintech Assistant — 僅業務使用，非官方授權</div>")
 
-    # 手機鍵盤彈出自動捲動
     demo.load(None, None, None, js="""
         window.addEventListener('focusin', () => {
             document.querySelector('textarea')?.scrollIntoView({ behavior: 'smooth', block: 'center' });