uploading pdfs...5

app.py

@@ -9,7 +9,8 @@ from fastapi import FastAPI, Request
 from fastapi.staticfiles import StaticFiles
 from fastapi.responses import FileResponse
 from fastapi.middleware.cors import CORSMiddleware
-from backend.main import app as backend_app
+from dotenv import load_dotenv
+from backend.main import app as backend_app, initialize_rag_system
 
 @asynccontextmanager
 async def lifespan(app: FastAPI):
@@ -18,6 +19,58 @@ async def lifespan(app: FastAPI):
     print("[root_app] Startup event triggered")
     print("[root_app] Mounting backend application at /api")
     
+    # Initialize RAG system here to ensure it runs
+    # The backend's lifespan might not run when mounted, so initialize explicitly
+    print("[root_app] Initializing RAG system...")
+    try:
+        # Set up paths (same as backend/main.py)
+        # Calculate paths relative to app.py (project root)
+        project_root = Path(__file__).parent
+        documents_dir = project_root / "documents"
+        processed_json = project_root / "processed_documents.json"
+        
+        # Load environment variables from .env file (if it exists)
+        env_path = project_root / ".env"
+        print(f"[root_app] .env file path: {env_path}")
+        print(f"[root_app] .env file exists? {env_path.exists()}")
+        
+        if env_path.exists():
+            load_dotenv(env_path, override=True)
+            print("[root_app] Loaded .env file")
+        else:
+            # Try loading anyway in case it's in a different location
+            load_dotenv(env_path, override=True)
+            print("[root_app] .env file not found, using environment variables")
+        
+        # Check for API keys (from .env or environment variables)
+        openai_key = os.getenv("OPENAI_API_KEY")
+        hf_token = os.getenv("HF_TOKEN")
+        
+        if openai_key:
+            print(f"[root_app] OPENAI_API_KEY found (length: {len(openai_key)} characters)")
+        else:
+            print("[root_app] WARNING: OPENAI_API_KEY not found in environment")
+        
+        if hf_token:
+            print(f"[root_app] HF_TOKEN found (length: {len(hf_token)} characters)")
+        
+        # Set global variables in backend.main module so initialize_rag_system can use them
+        import backend.main as backend_main
+        backend_main.PROJECT_ROOT = project_root
+        backend_main.DOCUMENTS_DIR = documents_dir
+        backend_main.PROCESSED_JSON = processed_json
+        
+        print(f"[root_app] PROJECT_ROOT: {project_root}")
+        print(f"[root_app] DOCUMENTS_DIR: {documents_dir} (exists: {documents_dir.exists()})")
+        print(f"[root_app] PROCESSED_JSON: {processed_json} (exists: {processed_json.exists()})")
+        
+        initialize_rag_system()
+        print("[root_app] RAG system initialization completed")
+    except Exception as e:
+        print(f"[root_app] WARNING: RAG system initialization failed: {e}")
+        import traceback
+        traceback.print_exc()
+    
     yield
     
     # Shutdown (if needed in the future)
@@ -75,6 +128,6 @@ else:
 
 if __name__ == "__main__":
     import uvicorn
-    port = int(os.getenv("PORT", 7860))
+    port = int(os.getenv("PORT", 7860))# 7860
     uvicorn.run(app, host="0.0.0.0", port=port)
 

backend/main.py

@@ -15,9 +15,10 @@ except ModuleNotFoundError:
     from rag_system import RAGSystem
 
 # Load environment variables from .env file in project root
-# PROJECT_ROOT = Path(__file__).resolve().parents[1]
-# DOCUMENTS_DIR = PROJECT_ROOT / "documents"
-# PROCESSED_JSON = PROJECT_ROOT / "processed_documents.json"
+# Initialize paths at module level (will be updated in lifespan if needed)
+PROJECT_ROOT = Path(__file__).resolve().parents[1]
+DOCUMENTS_DIR = PROJECT_ROOT / "documents"
+PROCESSED_JSON = PROJECT_ROOT / "processed_documents.json"
 
 # env_path = PROJECT_ROOT / ".env"
 # print(f"[Module Init] .env file path: {env_path}")