Upload 4 files

app.py

@@ -30,13 +30,15 @@ load_dotenv()
 from llm_fallback import get_groq_fallback_response
 from rag_system import initialize_and_get_rag_system
 from rag_components import KnowledgeRAG
-from utils import download_and_unzip_gdrive_file # MODIFIED: Import the new utility
+from utils import download_and_unzip_gdrive_file, download_gdrive_file # MODIFIED: Import the new utility
 from config import (
     RAG_SOURCES_DIR,
     RAG_STORAGE_PARENT_DIR,
     RAG_CHUNKED_SOURCES_FILENAME,
-    GDRIVE_INDEX_ENABLED,           # MODIFIED: Import new config
-    GDRIVE_INDEX_ID_OR_URL            # MODIFIED: Import new config
+    GDRIVE_INDEX_ENABLED,
+    GDRIVE_INDEX_ID_OR_URL,
+    GDRIVE_USERS_CSV_ENABLED,      # NEW
+    GDRIVE_USERS_CSV_ID_OR_URL     # NEW
 )
 
 # Setup logging (remains global for the app)
@@ -1082,7 +1084,23 @@ if __name__ == '__main__':
                         TEXT_EXTRACTIONS_DIR]:
         os.makedirs(folder_path, exist_ok=True)
 
-    # MODIFIED: Load users from CSV at startup
+    # --- NEW: Download users.csv from GDrive if enabled ---
+    if GDRIVE_USERS_CSV_ENABLED:
+        logger.info("[GDRIVE_USERS_DOWNLOAD] Google Drive users.csv download is ENABLED.")
+        if GDRIVE_USERS_CSV_ID_OR_URL:
+            users_csv_target_path = os.path.join(_APP_BASE_DIR, 'assets', 'users.csv')
+            logger.info(f"[GDRIVE_USERS_DOWNLOAD] Attempting to download users.csv to: {users_csv_target_path}")
+            download_successful = download_gdrive_file(GDRIVE_USERS_CSV_ID_OR_URL, users_csv_target_path)
+            if download_successful:
+                logger.info("[GDRIVE_USERS_DOWNLOAD] Successfully downloaded users.csv.")
+            else:
+                logger.error("[GDRIVE_USERS_DOWNLOAD] Failed to download users.csv from Google Drive. Will use existing file or fallback.")
+        else:
+            logger.warning("[GDRIVE_USERS_DOWNLOAD] GDRIVE_USERS_CSV_ENABLED is True, but GDRIVE_USERS_CSV_URL is not set.")
+    else:
+        logger.info("[GDRIVE_USERS_DOWNLOAD] Google Drive users.csv download is DISABLED.")
+
+    # Load users from CSV at startup (will use the downloaded file if successful)
     load_users_from_csv()
 
     load_qa_data_on_startup()

config.py

@@ -48,13 +48,18 @@ RAG_CHUNK_OVERLAP = int(os.getenv("RAG_CHUNK_OVERLAP", 150))
 RAG_RERANKER_MODEL_NAME = os.getenv("RAG_RERANKER_MODEL", "jinaai/jina-reranker-v2-base-multilingual")
 RAG_RERANKER_ENABLED = os.getenv("RAG_RERANKER_ENABLED", "True").lower() == "true"
 
+# GDrive configuration for RAG sources
 GDRIVE_SOURCES_ENABLED = os.getenv("GDRIVE_SOURCES_ENABLED", "False").lower() == "true"
 GDRIVE_FOLDER_ID_OR_URL = os.getenv("GDRIVE_FOLDER_URL")
 
-# MODIFIED: New configuration for downloading a pre-built FAISS index
+# GDrive configuration for downloading a pre-built FAISS index
 GDRIVE_INDEX_ENABLED = os.getenv("GDRIVE_INDEX_ENABLED", "False").lower() == "true"
 GDRIVE_INDEX_ID_OR_URL = os.getenv("GDRIVE_INDEX_URL")
 
+# --- NEW: GDrive configuration for downloading users.csv ---
+GDRIVE_USERS_CSV_ENABLED = os.getenv("GDRIVE_USERS_CSV_ENABLED", "False").lower() == "true"
+GDRIVE_USERS_CSV_ID_OR_URL = os.getenv("GDRIVE_USERS_CSV_URL")
+
 
 # Detailed logging configuration
 RAG_DETAILED_LOGGING = os.getenv("RAG_DETAILED_LOGGING", "True").lower() == "true"
@@ -67,4 +72,5 @@ logger.info(f"Reranker Model: {RAG_RERANKER_MODEL_NAME}")
 logger.info(f"Retrieval Pipeline: Initial Fetch K={RAG_INITIAL_FETCH_K}, Reranker Final K={RAG_RERANKER_K}")
 logger.info(f"Detailed Logging: {'ENABLED' if RAG_DETAILED_LOGGING else 'DISABLED'}")
 logger.info(f"GDrive Sources Download: {'ENABLED' if GDRIVE_SOURCES_ENABLED else 'DISABLED'}")
-logger.info(f"GDrive Pre-built Index Download: {'ENABLED' if GDRIVE_INDEX_ENABLED else 'DISABLED'}")
+logger.info(f"GDrive Pre-built Index Download: {'ENABLED' if GDRIVE_INDEX_ENABLED else 'DISABLED'}")
+logger.info(f"GDrive users.csv Download: {'ENABLED' if GDRIVE_USERS_CSV_ENABLED else 'DISABLED'}")

utils.py

@@ -66,6 +66,36 @@ def get_id_from_gdrive_input(url_or_id: str) -> Optional[str]:
     logger.warning(f"Could not reliably extract Google Drive ID from input: {url_or_id}")
     return None
 
+def download_gdrive_file(file_id_or_url: str, target_path: str) -> bool:
+    """
+    Downloads a single file from Google Drive to a specific path.
+    """
+    logger.info(f"[GDRIVE_SINGLE_FILE] Attempting to download file. Input: {file_id_or_url}")
+
+    file_id = get_id_from_gdrive_input(file_id_or_url)
+    if not file_id:
+        logger.error(f"[GDRIVE_SINGLE_FILE] Invalid Google Drive File ID or URL provided: {file_id_or_url}")
+        return False
+
+    try:
+        # Ensure the target directory exists before downloading
+        target_dir = os.path.dirname(target_path)
+        os.makedirs(target_dir, exist_ok=True)
+
+        logger.info(f"[GDRIVE_SINGLE_FILE] Downloading file ID: {file_id} to path: {target_path}")
+        # Use gdown to download directly to the target file path, fuzzy=True helps with some permissions
+        gdown.download(id=file_id, output=target_path, quiet=False, fuzzy=True)
+
+        if not os.path.exists(target_path) or os.path.getsize(target_path) == 0:
+            logger.error("[GDRIVE_SINGLE_FILE] Download failed or the resulting file is empty.")
+            return False
+
+        logger.info(f"[GDRIVE_SINGLE_FILE] Download successful.")
+        return True
+
+    except Exception as e:
+        logger.error(f"[GDRIVE_SINGLE_FILE] An error occurred during download: {e}", exc_info=True)
+        return False
 
 def download_and_unzip_gdrive_file(file_id_or_url: str, target_extraction_dir: str) -> bool:
     """