Merge branch 'ingest-settings' of github.com:phact/gendb into ingest-settings

src/main.py

@@ -1,8 +1,7 @@
-
 # Configure structured logging early
-from services.flows_service import FlowsService
 from connectors.langflow_connector_service import LangflowConnectorService
 from connectors.service import ConnectorService
+from services.flows_service import FlowsService
 from utils.logging_config import configure_from_env, get_logger
 
 configure_from_env()
@@ -23,24 +22,28 @@ from starlette.routing import Route
 multiprocessing.set_start_method("spawn", force=True)
 
 # Create process pool FIRST, before any torch/CUDA imports
-from utils.process_pool import process_pool
+from utils.process_pool import process_pool  # isort: skip
-
 import torch
 
 # API endpoints
 from api import (
-    router,
     auth,
     chat,
     connectors,
+    flows,
     knowledge_filter,
     langflow_files,
+    nudges,
     oidc,
+    router,
     search,
     settings,
     tasks,
     upload,
 )
+
+# Existing services
+from api.connector_router import ConnectorRouter
 from auth_middleware import optional_auth, require_auth
 
 # Configuration and setup
@@ -53,9 +56,6 @@ from config.settings import (
     clients,
     is_no_auth_mode,
 )
-
-# Existing services
-from api.connector_router import ConnectorRouter
 from services.auth_service import AuthService
 from services.chat_service import ChatService
 
@@ -70,24 +70,6 @@ from services.monitor_service import MonitorService
 from services.search_service import SearchService
 from services.task_service import TaskService
 from session_manager import SessionManager
-from utils.process_pool import process_pool
-
-# API endpoints
-from api import (
-    flows,
-    router,
-    nudges,
-    upload,
-    search,
-    chat,
-    auth,
-    connectors,
-    tasks,
-    oidc,
-    knowledge_filter,
-    settings,
-)
-
 
 logger.info(
     "CUDA device information",
@@ -246,7 +228,10 @@ async def init_index_when_ready():
 async def ingest_default_documents_when_ready(services):
     """Scan the local documents folder and ingest files like a non-auth upload."""
     try:
-        logger.info("Ingesting default documents when ready", disable_langflow_ingest=DISABLE_INGEST_WITH_LANGFLOW)
+        logger.info(
+            "Ingesting default documents when ready",
+            disable_langflow_ingest=DISABLE_INGEST_WITH_LANGFLOW,
+        )
         base_dir = os.path.abspath(os.path.join(os.getcwd(), "documents"))
         if not os.path.isdir(base_dir):
             logger.info(
@@ -295,7 +280,7 @@ async def _ingest_default_documents_langflow(services, file_paths):
             logger.debug("Processing file with Langflow pipeline", file_path=file_path)
 
             # Read file content
-            with open(file_path, 'rb') as f:
+            with open(file_path, "rb") as f:
                 content = f.read()
 
             # Create file tuple for upload
@@ -303,12 +288,13 @@ async def _ingest_default_documents_langflow(services, file_paths):
             # Determine content type based on file extension
             content_type, _ = mimetypes.guess_type(filename)
             if not content_type:
-                content_type = 'application/octet-stream'
+                content_type = "application/octet-stream"
 
             file_tuple = (filename, content, content_type)
 
             # Use AnonymousUser details for default documents
             from session_manager import AnonymousUser
+
             anonymous_user = AnonymousUser()
 
             # Get JWT token using same logic as DocumentFileProcessor
@@ -322,7 +308,7 @@ async def _ingest_default_documents_langflow(services, file_paths):
                     anonymous_user.user_id, effective_jwt
                 )
                 # Get the JWT that was created by session manager
-                if hasattr(session_manager, '_anonymous_jwt'):
+                if hasattr(session_manager, "_anonymous_jwt"):
                     effective_jwt = session_manager._anonymous_jwt
 
             # Prepare tweaks for default documents with anonymous user metadata
@@ -332,7 +318,7 @@ async def _ingest_default_documents_langflow(services, file_paths):
                         {"key": "owner", "value": None},
                         {"key": "owner_name", "value": anonymous_user.name},
                         {"key": "owner_email", "value": anonymous_user.email},
-                        {"key": "connector_type", "value": "system_default"}
+                        {"key": "connector_type", "value": "system_default"},
                     ]
                 }
             }
@@ -447,7 +433,7 @@ async def initialize_services():
     # Create connector router that chooses based on configuration
     connector_service = ConnectorRouter(
         langflow_connector_service=langflow_connector_service,
-        openrag_connector_service=openrag_connector_service
+        openrag_connector_service=openrag_connector_service,
     )
 
     # Initialize auth service

uv.lock

@@ -1405,7 +1405,7 @@ wheels = [
 
 [[package]]
 name = "openrag"
-version = "0.1.1"
+version = "0.1.2"
 source = { editable = "." }
 dependencies = [
     { name = "agentd" },