Upload personalized_ht4.py

personalized_ht4.py

@@ -10,6 +10,12 @@ import pandas as pd
 import matplotlib.pyplot as plt
 from typing import Dict, List, Any
 
+import uuid  
+import os 
+import shutil  
+from datetime import datetime, timedelta 
+import hashlib
+
 # LangChain imports
 from langchain_openai import ChatOpenAI, OpenAIEmbeddings
 from langchain_community.vectorstores import Chroma
@@ -30,6 +36,19 @@ st.set_page_config(
     initial_sidebar_state="expanded",
 )
 
+# ===== 用戶隔離系統初始化 =====
+# 為每個用戶生成唯一 ID
+if 'user_id' not in st.session_state:
+    st.session_state.user_id = str(uuid.uuid4())
+    st.session_state.session_start = datetime.now()
+
+# Session 過期檢查（2小時後過期）
+if 'session_start' in st.session_state:
+    session_duration = datetime.now() - st.session_state.session_start
+    if session_duration > timedelta(hours=2):
+        st.warning("⏱️ Your session has expired (2 hours). Please refresh the page.")
+        st.stop()
+
 # Initialize session state for chat histories
 if 'assistant_messages' not in st.session_state:
     st.session_state.assistant_messages = []
@@ -43,6 +62,33 @@ if 'vectorstore' not in st.session_state:
     st.session_state.vectorstore = None
 if 'cea_results' not in st.session_state:
     st.session_state.cea_results = None
+if 'uploaded_files' not in st.session_state:
+    st.session_state.uploaded_files = []  # ✅ 新增
+if 'api_call_count' not in st.session_state:
+    st.session_state.api_call_count = 0  # ✅ 新增
+
+
+# ✅ 啟動時清理超過 24 小時的舊資料
+def cleanup_old_data():
+    """Delete vector stores older than 24 hours"""
+    data_dir = "./data/chroma"
+    if not os.path.exists(data_dir):
+        return
+    
+    cutoff_time = datetime.now() - timedelta(hours=24)
+    
+    try:
+        for user_folder in os.listdir(data_dir):
+            folder_path = os.path.join(data_dir, user_folder)
+            if os.path.isdir(folder_path):
+                mod_time = datetime.fromtimestamp(os.path.getmtime(folder_path))
+                if mod_time < cutoff_time:
+                    shutil.rmtree(folder_path)
+    except Exception as e:
+        pass  # 靜默失敗，不影響用戶
+
+cleanup_old_data()
+
 
 # Header with OpenAI API Key input
 col1, col2 = st.columns([3, 1])
@@ -62,20 +108,29 @@ with col2:
         st.success("✓ API Key set")
     else:
         st.warning("⚠️ Enter API key")
+        
 
-
-# Check if API key is provided
+# ali Check if API key is provided
 def get_llm():
-    """Initialize LangChain LLM with OpenAI"""
+    """Initialize LangChain LLM with OpenAI - with rate limiting"""
     if not openai_api_key:
         return None
 
+    # ✅ API 調用限制（每 session 最多 100 次）
+    if st.session_state.api_call_count >= 100:
+        st.error("⚠️ API call limit reached (100 calls per session). Please start a new session.")
+        st.stop()
+
     try:
         llm = ChatOpenAI(
             model="gpt-4o-mini",
             temperature=0.7,
             openai_api_key=openai_api_key
         )
+        
+        # ✅ 記錄 API 調用
+        st.session_state.api_call_count += 1
+        
         return llm
     except Exception as e:
         st.error(f"Error initializing OpenAI: {str(e)}")
@@ -84,11 +139,16 @@ def get_llm():
 
 # Create vector store from patient data
 def create_patient_vectorstore(patients_df: pd.DataFrame):
-    """Create vector store from patient dataframe for RAG retrieval"""
+    """Create vector store from patient dataframe for RAG retrieval - USER ISOLATED"""
     if not openai_api_key:
         return None
 
     try:
+        # ✅ 建立用戶專屬目錄
+        user_id = st.session_state.user_id
+        persist_dir = f"./data/chroma/{user_id}"
+        os.makedirs(persist_dir, exist_ok=True)
+        
         documents = []
         for idx, row in patients_df.iterrows():
             patient_text = f"""Patient ID: {row['patient_id']}
@@ -102,12 +162,36 @@ Betel: {row.get('betel', 'No')}, Family History: {row['family_history']}"""
 
             doc = Document(
                 page_content=patient_text,
-                metadata={"patient_id": row['patient_id']}
+                metadata={
+                    "patient_id": row['patient_id'],
+                    "user_id": user_id  # ✅ 加入 user_id 標記
+                }
             )
             documents.append(doc)
 
         embeddings = OpenAIEmbeddings(openai_api_key=openai_api_key)
-        vectorstore = Chroma.from_documents(documents=documents, embedding=embeddings)
+        
+        # ✅ 使用用戶專屬的 collection 和目錄
+        vectorstore = Chroma(
+            collection_name=f"user_{user_id}_patients",
+            embedding_function=embeddings,
+            persist_directory=persist_dir
+        )
+        
+        # 清空舊資料（如果有）
+        try:
+            vectorstore.delete_collection()
+            vectorstore = Chroma(
+                collection_name=f"user_{user_id}_patients",
+                embedding_function=embeddings,
+                persist_directory=persist_dir
+            )
+        except:
+            pass
+        
+        # 加入新文件
+        vectorstore.add_documents(documents)
+        
         return vectorstore
 
     except Exception as e:
@@ -115,6 +199,7 @@ Betel: {row.get('betel', 'No')}, Family History: {row['family_history']}"""
         return None
 
 
+
 # Retrieve patient by ID
 def retrieve_patient_by_id(patient_id: str):
     """Retrieve patient from dataframe by ID"""
@@ -134,31 +219,44 @@ def retrieve_patient_by_id(patient_id: str):
 # Sidebar for patient information
 st.sidebar.header("👤 Patient Information")
 
-# ===== NEW: Patient Data Upload Section =====
-with st.sidebar.expander("📁 Upload Patient Database (Optional)", expanded=False):
+
+# ali ===== NEW: Patient Data Upload Section =====
+with st.sidebar.expander("📂 Upload Patient Database (Optional)", expanded=False):
     uploaded_file = st.file_uploader(
         "Upload CSV/Excel with patient data",
         type=['csv', 'xlsx'],
-        help="Upload a file with multiple patients to enable quick retrieval by ID"
+        help="Upload a file with multiple patients (Max 10MB)"
     )
 
     if uploaded_file is not None:
         try:
+            # ✅ 檔案大小限制
+            file_size_mb = uploaded_file.size / (1024 * 1024)
+            if file_size_mb > 10:
+                st.error(f"❌ File too large ({file_size_mb:.1f}MB). Max 10MB.")
+                st.stop()
+            
             if uploaded_file.name.endswith('.csv'):
                 df = pd.read_csv(uploaded_file)
             else:
                 df = pd.read_excel(uploaded_file)
 
             st.session_state.patients_df = df
-            st.success(f"✅ Loaded {len(df)} patients")
+            
+            # ✅ 記錄上傳的檔案
+            file_hash = hashlib.md5(uploaded_file.getvalue()).hexdigest()
+            if file_hash not in st.session_state.uploaded_files:
+                st.session_state.uploaded_files.append(file_hash)
+            
+            st.success(f"✅ Loaded {len(df)} patients ({file_size_mb:.1f}MB)")
 
             # Optionally create vector store
             if openai_api_key and st.button("🔄 Create Vector Store for Smart Search"):
-                with st.spinner("Creating vector store..."):
+                with st.spinner("Creating isolated vector store..."):
                     vectorstore = create_patient_vectorstore(df)
                     if vectorstore:
                         st.session_state.vectorstore = vectorstore
-                        st.success("✅ Vector store created!")
+                        st.success("✅ Vector store created! (Isolated to your session)")
 
         except Exception as e:
             st.error(f"Error loading file: {str(e)}")
@@ -180,6 +278,7 @@ with st.sidebar.expander("📁 Upload Patient Database (Optional)", expanded=Fal
 
 st.sidebar.markdown("---")
 
+
 # Check if we have loaded patient data
 if 'loaded_patient' in st.session_state:
     # Use loaded patient data