Update app.py

app.py

@@ -8,25 +8,18 @@ import sys
 from pathlib import Path
 from PIL import Image
 import gradio as gr
-from fastapi import FastAPI
-from fastapi.middleware.cors import CORSMiddleware
-from pydantic import BaseModel
-from contextlib import asynccontextmanager
 
-# إعداد Logging شامل
+# إعداد Logging
 logging.basicConfig(
     level=logging.INFO,
     format='%(asctime)s - %(name)s - %(levelname)s - %(message)s',
-    handlers=[
-        logging.StreamHandler(sys.stdout)
-    ]
+    handlers=[logging.StreamHandler(sys.stdout)]
 )
 logger = logging.getLogger(__name__)
 
 logger.info("🚀 بدء تشغيل DermaScan AI...")
 
 try:
-    logger.info("📦 جاري استيراد الوحدات...")
     from rag import build_default_vectorstore
     from agent import (
         build_agent,
@@ -42,7 +35,6 @@ except Exception as e:
 
 # Configuration
 MODEL_NAME = "llama-3.1-8b-instant"
-FALLBACK_MODEL_NAME = "llama-3.3-70b-versatile"
 TEMPERATURE = 0.25
 TOP_K = 3
 EXTERNAL_API_URL = "https://omarelrayes-api.hf.space"
@@ -50,7 +42,6 @@ EXTERNAL_API_URL = "https://omarelrayes-api.hf.space"
 # Global variables
 SESSIONS = {}
 agent = None
-fallback_agent = None
 api_client = None
 
 def get_session(thread_id: str) -> dict:
@@ -59,225 +50,64 @@ def get_session(thread_id: str) -> dict:
             "role": "patient",
             "last_analysis": None,
             "current_image": None,
-            "matched_patient_id": None,
-            "matched_patient_report": None,
             "_image_blobs": {},
             "_session_image_names": [],
-            "_turn_count": 0,
-            "_summary": "",
         }
     return SESSIONS[thread_id]
 
 def initialize_system():
-    """Initialize agent and API client"""
-    global agent, fallback_agent, api_client
+    global agent, api_client
     
     try:
         logger.info("🔧 بدء تهيئة النظام...")
         
-        # Check Groq API Key
         groq_key = os.environ.get("GROQ_API_KEY")
         if not groq_key:
-            logger.warning("️ GROQ_API_KEY not set in environment")
+            logger.warning("⚠️ GROQ_API_KEY not set")
         else:
             logger.info("✅ GROQ_API_KEY found")
         
-        # Initialize API client
-        logger.info("🌐 جاري تهيئة API Client...")
         api_client = DermaScanAPIClient(EXTERNAL_API_URL)
         set_api_client(api_client)
         
-        # Check external API health
-        try:
-            if api_client.health_check():
-                logger.info("✅ External API is healthy")
-            else:
-                logger.warning("⚠️ External API not reachable")
-        except Exception as e:
-            logger.error(f"❌ External API health check failed: {e}")
-        
-        # Build vectorstore and agent
-        logger.info("🗄️ جاري بناء Vectorstore...")
-        try:
-            vectorstore = build_default_vectorstore()
-            if vectorstore:
-                logger.info("✅ Vectorstore loaded successfully")
-                retriever = vectorstore.as_retriever(search_kwargs={"k": TOP_K})
-            else:
-                logger.warning("⚠️ Vectorstore not available")
-                retriever = None
-        except Exception as e:
-            logger.error(f" Failed to build vectorstore: {e}", exc_info=True)
-            retriever = None
+        if api_client.health_check():
+            logger.info("✅ External API is healthy")
+        else:
+            logger.warning("⚠️ External API not reachable")
         
-        # Build agents
-        logger.info("🤖 جاري بناء Agent...")
-        try:
-            agent = build_agent(retriever, model_name=MODEL_NAME, temperature=TEMPERATURE)
-            logger.info("✅ Primary agent built successfully")
-        except Exception as e:
-            logger.error(f"❌ Failed to build primary agent: {e}", exc_info=True)
-            agent = None
+        vectorstore = build_default_vectorstore()
+        retriever = vectorstore.as_retriever(search_kwargs={"k": TOP_K}) if vectorstore else None
         
-        try:
-            fallback_agent = build_agent(retriever, model_name=FALLBACK_MODEL_NAME, temperature=TEMPERATURE)
-            logger.info("✅ Fallback agent built successfully")
-        except Exception as e:
-            logger.error(f" Failed to build fallback agent: {e}", exc_info=True)
-            fallback_agent = None
+        if retriever is None:
+            logger.warning("⚠️ Vectorstore not available")
         
+        agent = build_agent(retriever, model_name=MODEL_NAME, temperature=TEMPERATURE)
+        logger.info("✅ Agent built successfully")
         logger.info("🎉 تم تهيئة النظام بنجاح!")
         
     except Exception as e:
-        logger.error(f"❌ فشل تهيئة النظام: {e}", exc_info=True)
+        logger.error(f"❌ فشل التهيئة: {e}", exc_info=True)
         raise
 
-# FastAPI app with lifespan
-@asynccontextmanager
-async def lifespan(app: FastAPI):
-    # Startup
-    logger.info("�� FastAPI startup...")
-    initialize_system()
-    yield
-    # Shutdown
-    logger.info("🛑 FastAPI shutdown...")
-
-app = FastAPI(title="DermaScan AI - RAG System", lifespan=lifespan)
-
-app.add_middleware(
-    CORSMiddleware,
-    allow_origins=["*"],
-    allow_methods=["*"],
-    allow_headers=["*"],
-)
-
-# Pydantic models
-class ChatRequest(BaseModel):
-    thread_id: str
-    message: str
+# Initialize on startup
+initialize_system()
 
-class BookingRequest(BaseModel):
-    thread_id: str
-    city: str = ""
-
-# FastAPI endpoints
-@app.get("/health")
-def health():
-    return {
-        "agent_ready": agent is not None,
-        "api_client_ready": api_client is not None,
-        "external_api_healthy": api_client.health_check() if api_client else False,
-    }
-
-@app.post("/chat")
-def chat(req: ChatRequest):
-    """Chat endpoint"""
-    global agent
-    
-    if agent is None:
-        return {"ok": False, "error": "Agent not initialized yet"}
-    
-    session = get_session(req.thread_id)
-    
-    # Set session context
-    set_session_storage(SESSIONS)
-    set_current_thread_id(req.thread_id)
-    
-    try:
-        # Build prompt with context
-        last_analysis = session.get("last_analysis")
-        analysis_ctx = ""
-        if last_analysis:
-            analysis_ctx = (
-                f"\n\n[CURRENT IMAGE ANALYSIS — Classification: {last_analysis.get('label', 'N/A')}, "
-                f"Confidence: {last_analysis.get('confidence_pct', 0):.1f}%, "
-                f"Affected Area: {last_analysis.get('infection_pct', 0):.1f}%]"
-            )
-        
-        full_prompt = req.message + analysis_ctx
-        
-        # Run agent
-        config = {"configurable": {"thread_id": req.thread_id}, "recursion_limit": 14}
-        
-        final_text = ""
-        tool_calls = []
-        
-        for event in agent.stream(
-            {"messages": [("user", full_prompt)]},
-            config=config,
-            stream_mode="values"
-        ):
-            last_msg = event["messages"][-1]
-            if hasattr(last_msg, "tool_calls") and last_msg.tool_calls:
-                for tc in last_msg.tool_calls:
-                    tool_calls.append({"name": tc["name"], "args": tc["args"]})
-            if last_msg.type == "ai" and last_msg.content:
-                final_text = last_msg.content
-        
-        # Clean response
-        cleaned_text = re.sub(r'\[ANALYSIS_RESULT:\{.*?\}\]', '', final_text).strip()
-        
-        session["_turn_count"] = session.get("_turn_count", 0) + 1
-        
-        return {
-            "ok": True,
-            "reply": cleaned_text,
-            "tool_calls": tool_calls,
-        }
-        
-    except Exception as e:
-        logger.error(f"Chat error: {e}", exc_info=True)
-        return {"ok": False, "error": str(e)}
-
-@app.post("/book-appointment")
-def book_appointment(req: BookingRequest):
-    """Book appointment via external API"""
-    if api_client is None:
-        return {"ok": False, "error": "API client not initialized"}
-    
-    result = api_client.book_appointment(req.thread_id, req.city)
-    return result
-
-@app.get("/image")
-def get_image(name: str = "", thread_id: str = ""):
-    """Get image from session"""
-    try:
-        session = get_session(thread_id) if thread_id else {}
-        blobs = session.get("_image_blobs", {})
-        
-        if name not in blobs:
-            return {"ok": False, "error": "Image not found"}
-        
-        return {"ok": True, "base64": blobs[name], "media_type": "image/png"}
-    except Exception as e:
-        return {"ok": False, "error": str(e)}
-
-# Gradio Interface
+# Gradio Functions
 def gradio_chat(message: str, history: list, thread_id: str = "default") -> str:
-    """Gradio chat function"""
     if agent is None:
-        return "System is not initialized yet. Please wait..."
+        return "System not initialized yet."
     
     session = get_session(thread_id)
-    
-    # Set session context
     set_session_storage(SESSIONS)
     set_current_thread_id(thread_id)
     
     try:
-        # Build prompt with context
         last_analysis = session.get("last_analysis")
         analysis_ctx = ""
         if last_analysis:
-            analysis_ctx = (
-                f"\n\n[CURRENT IMAGE ANALYSIS — Classification: {last_analysis.get('label', 'N/A')}, "
-                f"Confidence: {last_analysis.get('confidence_pct', 0):.1f}%, "
-                f"Affected Area: {last_analysis.get('infection_pct', 0):.1f}%]"
-            )
+            analysis_ctx = f"\n\n[Analysis: {last_analysis.get('label', 'N/A')} - {last_analysis.get('confidence_pct', 0):.1f}%]"
         
         full_prompt = message + analysis_ctx
-        
-        # Run agent
         config = {"configurable": {"thread_id": thread_id}, "recursion_limit": 14}
         
         final_text = ""
@@ -290,30 +120,25 @@ def gradio_chat(message: str, history: list, thread_id: str = "default") -> str:
             if last_msg.type == "ai" and last_msg.content:
                 final_text = last_msg.content
         
-        # Clean response
         cleaned_text = re.sub(r'\[ANALYSIS_RESULT:\{.*?\}\]', '', final_text).strip()
-        
         session["_turn_count"] = session.get("_turn_count", 0) + 1
         
-        return cleaned_text if cleaned_text else "No response generated."
+        return cleaned_text if cleaned_text else "No response."
         
     except Exception as e:
-        logger.error(f"Gradio chat error: {e}", exc_info=True)
+        logger.error(f"Chat error: {e}", exc_info=True)
         return f"Error: {str(e)}"
 
 def gradio_upload_image(image: Image.Image, thread_id: str = "default") -> tuple:
-    """Handle image upload"""
     if api_client is None:
         return "API client not initialized", None, None, None
     
     try:
-        # Upload and analyze
         result = api_client.upload_and_analyze_image(image, thread_id)
         
         if 'error' in result:
             return f"Error: {result['error']}", None, None, None
         
-        # Update session
         session = get_session(thread_id)
         session["last_analysis"] = {
             "label": result.get("label"),
@@ -321,41 +146,27 @@ def gradio_upload_image(image: Image.Image, thread_id: str = "default") -> tuple
             "infection_pct": result.get("infection_pct"),
         }
         
-        # Store images
         if 'images' in result:
             session["_image_blobs"].update(result['images'])
         
-        # Prepare display
         analysis_text = f"""
-✅ **تم تحليل الصورة بنجاح!**
+✅ **تم التحليل بنجاح!**
+
 **النتائج:**
 - التصنيف: {result.get('label')}
 - نسبة الثقة: {result.get('confidence_pct')}%
 - المساحة المصابة: {result.get('infection_pct')}%
-يمكنك الآن طرح أسئلة عن هذه الصورة.
+
+يمكنك الآن طرح أسئلة عن الصورة.
 """
         
-        # Get images for display
         orig_b64 = result.get('images', {}).get('original', '')
         mask_b64 = result.get('images', {}).get('mask', '')
         overlay_b64 = result.get('images', {}).get('overlay', '')
         
-        # Convert base64 to numpy arrays
-        orig_img = None
-        mask_img = None
-        overlay_img = None
-        
-        if orig_b64:
-            orig_data = base64.b64decode(orig_b64)
-            orig_img = Image.open(io.BytesIO(orig_data))
-        
-        if mask_b64:
-            mask_data = base64.b64decode(mask_b64)
-            mask_img = Image.open(io.BytesIO(mask_data))
-        
-        if overlay_b64:
-            overlay_data = base64.b64decode(overlay_b64)
-            overlay_img = Image.open(io.BytesIO(overlay_data))
+        orig_img = Image.open(io.BytesIO(base64.b64decode(orig_b64))) if orig_b64 else None
+        mask_img = Image.open(io.BytesIO(base64.b64decode(mask_b64))) if mask_b64 else None
+        overlay_img = Image.open(io.BytesIO(base64.b64decode(overlay_b64))) if overlay_b64 else None
         
         return analysis_text, orig_img, mask_img, overlay_img
         
@@ -364,84 +175,72 @@ def gradio_upload_image(image: Image.Image, thread_id: str = "default") -> tuple
         return f"Error: {str(e)}", None, None, None
 
 # Build Gradio UI
-logger.info(" جاري بناء واجهة Gradio...")
-try:
-    with gr.Blocks(title="DermaScan AI - Patient Portal") as demo:
-        gr.Markdown("""
-        # 🏥 DermaScan AI - مساعد أمراض الجلدية
-        
-        ارفع صورة الجلد واحصل على تحليل بالذكاء الاصطناعي + اسأل أي أسئلة.
-        """)
-        
-        thread_id_state = gr.State(value="default")
-        
-        with gr.Row():
-            with gr.Column(scale=1):
-                gr.Markdown("### 📤 رفع الصورة")
-                image_input = gr.Image(type="pil", label="ارفع صورة الجلد")
-                upload_btn = gr.Button("📎 ارفع وحلّل", variant="primary")
-                analysis_output = gr.Markdown(label="نتيجة التحليل")
-            
-            with gr.Column(scale=2):
-                gr.Markdown("###  الدردشة مع DermaScan AI")
-                chatbot = gr.Chatbot(label="المحادثة", height=400)
-                msg_input = gr.Textbox(
-                    label="رسالتك",
-                    placeholder="اسأل عن حالتك الجلدية...",
-                    lines=2
-                )
-                send_btn = gr.Button("🚀 إرسال", variant="primary")
-        
-        with gr.Row():
-            with gr.Column():
-                gr.Markdown("### 🖼️ الصور المحللة")
-                orig_display = gr.Image(label="الصورة الأصلية", type="pil")
-            with gr.Column():
-                mask_display = gr.Image(label="قناع التجزئة", type="pil")
-            with gr.Column():
-                overlay_display = gr.Image(label="الصورة مع التغطية", type="pil")
-        
-        # Event handlers
-        def user_message(user_msg, history):
-            return "", history + [[user_msg, None]]
-        
-        def bot_response(history, thread_id):
-            if not history:
-                return history
-            
-            user_msg = history[-1][0]
-            response = gradio_chat(user_msg, history[:-1], thread_id)
-            
-            history[-1][1] = response
+with gr.Blocks(title="DermaScan AI - Patient Portal") as demo:
+    gr.Markdown("""
+    #  DermaScan AI - مساعد أمراض الجلدية
+    
+    ارفع صورة الجلد واحصل على تحليل بالذكاء الاصطناعي.
+    """)
+    
+    thread_id_state = gr.State(value="default")
+    
+    with gr.Row():
+        with gr.Column(scale=1):
+            gr.Markdown("###  رفع الصورة")
+            image_input = gr.Image(type="pil", label="ارفع صورة الجلد")
+            upload_btn = gr.Button(" ارفع وحلّل", variant="primary")
+            analysis_output = gr.Markdown(label="نتيجة التحليل")
+        
+        with gr.Column(scale=2):
+            gr.Markdown("### 💬 الدردشة")
+            chatbot = gr.Chatbot(label="المحادثة", height=400)
+            msg_input = gr.Textbox(
+                label="رسالتك",
+                placeholder="اسأل عن حالتك...",
+                lines=2
+            )
+            send_btn = gr.Button("🚀 إرسال", variant="primary")
+    
+    with gr.Row():
+        with gr.Column():
+            orig_display = gr.Image(label="الأصلية", type="pil")
+        with gr.Column():
+            mask_display = gr.Image(label="القناع", type="pil")
+        with gr.Column():
+            overlay_display = gr.Image(label="التغطية", type="pil")
+    
+    def user_message(user_msg, history):
+        return "", history + [[user_msg, None]]
+    
+    def bot_response(history, thread_id):
+        if not history:
             return history
-        
-        upload_btn.click(
-            fn=gradio_upload_image,
-            inputs=[image_input, thread_id_state],
-            outputs=[analysis_output, orig_display, mask_display, overlay_display]
-        )
-        
-        send_btn.click(
-            fn=user_message,
-            inputs=[msg_input, chatbot],
-            outputs=[msg_input, chatbot]
-        ).then(
-            fn=bot_response,
-            inputs=[chatbot, thread_id_state],
-            outputs=[chatbot]
-        )
+        user_msg = history[-1][0]
+        response = gradio_chat(user_msg, history[:-1], thread_id)
+        history[-1][1] = response
+        return history
     
-    logger.info("✅ تم بناء واجهة Gradio بنجاح")
+    upload_btn.click(
+        fn=gradio_upload_image,
+        inputs=[image_input, thread_id_state],
+        outputs=[analysis_output, orig_display, mask_display, overlay_display]
+    )
     
-except Exception as e:
-    logger.error(f"❌ فشل بناء واجهة Gradio: {e}", exc_info=True)
-    raise
-
-logger.info("🎉 التطبيق جاهز للتشغيل!")
+    send_btn.click(
+        fn=user_message,
+        inputs=[msg_input, chatbot],
+        outputs=[msg_input, chatbot]
+    ).then(
+        fn=bot_response,
+        inputs=[chatbot, thread_id_state],
+        outputs=[chatbot]
+    )
 
-# ✅ الحل الصحيح لـ HF Spaces:
-# Mount Gradio to FastAPI
-app = gr.mount_gradio_app(app, demo, path="/")
-logger.info("✅ تم ربط Gradio بـ FastAPI")
+logger.info("🎉 التطبيق جاهز!")
 
-# HF Spaces بيتكفل بتشغيل التطبيق - مش محتاج demo.launch()
+# ✅ تشغيل Gradio مباشرة - ده الحل الصحيح لـ HF Spaces
+demo.launch(
+    server_name="0.0.0.0",
+    server_port=7860,
+    debug=False
+)