OpenRAG 0.5.1__tar.gz → 0.5.1.dev1__tar.gz

{openrag-0.5.1/src/openrag.egg-info → openrag-0.5.1.dev1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: openrag
-Version: 0.5.1
+Version: 0.5.1.dev1
 Summary: OpenRAG is a comprehensive Retrieval-Augmented Generation platform that enables intelligent document search and AI-powered conversations.
 Classifier: Development Status :: 4 - Beta
 Classifier: Environment :: Console

{openrag-0.5.1 → openrag-0.5.1.dev1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "openrag"
-version = "0.5.1"
+version = "0.5.1.dev1"
 description = "OpenRAG is a comprehensive Retrieval-Augmented Generation platform that enables intelligent document search and AI-powered conversations."
 readme = "README.md"
 requires-python = ">=3.13"

{openrag-0.5.1 → openrag-0.5.1.dev1}/src/agent.py

@@ -171,16 +171,8 @@ async def async_response_stream(
             try:
                 # Try to serialize the chunk object
                 if hasattr(chunk, "model_dump"):
-                    # Pydantic model. Exclude "delta" from serialization since Langflow's
-                    # /response SSE wraps text deltas as {"content": "..."} for every provider but
-                    # openai SDK's event models declare/expect delta to be str so model_dump() will
-                    # emit noisy logs (PydanticSerializationUnexpectedValue) for every chunk. Basically
-                    # just reattach the raw (unvalidated) value afterwards to preserve same shape.
-                    # TODO: replace this with just chunk.model_dump() if langflow's /response SSE matches openai's 
-                    # SDK's delta: str schema
-                    chunk_data = chunk.model_dump(exclude={"delta"})
-                    if hasattr(chunk, "delta"):
-                        chunk_data["delta"] = chunk.delta
+                    # Pydantic model
+                    chunk_data = chunk.model_dump()
                 elif hasattr(chunk, "__dict__"):
                     chunk_data = chunk.__dict__
                 else:

{openrag-0.5.1 → openrag-0.5.1.dev1}/src/api/connectors.py

@@ -734,8 +734,8 @@ async def connector_sync(
             {
                 "task_ids": task_ids,
                 "status": "sync_started",
-                "message": f"Started syncing files from 1 {connector_type} connection",
-                "connections_synced": len(task_ids),
+                "message": f"Started syncing files from {len(active_connections)} {connector_type} connection(s)",
+                "connections_synced": len(active_connections),
             },
             status_code=201,
         )

{openrag-0.5.1 → openrag-0.5.1.dev1}/src/api/settings.py

@@ -391,27 +391,16 @@ async def get_settings(
             environment=ENVIRONMENT or None,
         )
 
-    except Exception:
-        logger.exception("Failed to retrieve settings")
+    except Exception as e:
+        logger.error(f"Failed to retrieve settings: {str(e)}")
         return JSONResponse(
-            {"error": "Failed to retrieve settings"}, status_code=500
+            {"error": f"Failed to retrieve settings: {str(e)}"}, status_code=500
         )
 
 
-# Provider names in priority order. LLM supports anthropic; embeddings do not.
-_LLM_PROVIDER_NAMES = ("openai", "anthropic", "watsonx", "ollama")
-_EMBEDDING_PROVIDER_NAMES = ("openai", "watsonx", "ollama")
-
-
-def _configured_provider_names(config, provider_names) -> list:
-    """Return the provider names from `provider_names` marked configured in the OpenRAG config."""
-    providers = config.providers
-    return [name for name in provider_names if getattr(providers, name).configured]
-
-
 def _first_configured_llm_provider(config, excluding: str) -> str:
     """Return the first configured LLM provider that isn't `excluding`."""
-    for p in _LLM_PROVIDER_NAMES:
+    for p in ["openai", "anthropic", "watsonx", "ollama"]:
         if p != excluding and getattr(config.providers, p).configured:
             return p
     return "openai"
@@ -419,7 +408,7 @@ def _first_configured_llm_provider(config, excluding: str) -> str:
 
 def _first_configured_embedding_provider(config, excluding: str) -> str:
     """Return the first configured embedding provider (openai/watsonx/ollama) that isn't `excluding`."""
-    for p in _EMBEDDING_PROVIDER_NAMES:
+    for p in ["openai", "watsonx", "ollama"]:
         if p != excluding and getattr(config.providers, p).configured:
             return p
     return "openai"
@@ -1014,14 +1003,14 @@ async def update_settings(
         )
         return SettingsUpdateResponse(message="Configuration updated successfully")
 
-    except Exception:
-        logger.exception("Failed to update settings")
+    except Exception as e:
+        logger.error("Failed to update settings", error=str(e))
         await TelemetryClient.send_event(
             Category.SETTINGS_OPERATIONS,
             MessageId.ORB_SETTINGS_UPDATE_FAILED
         )
         return JSONResponse(
-            {"error": "Failed to update settings"}, status_code=500
+            {"error": f"Failed to update settings: {str(e)}"}, status_code=500
         )
 
 
@@ -1283,10 +1272,15 @@ async def onboarding(
                 admin_username = user.user_id if IBM_AUTH_ENABLED and user else None
                 await init_index(opensearch_client, admin_username=admin_username)
                 logger.info("OpenSearch index initialization completed successfully")
-            except Exception:
-                logger.exception("Failed to initialize OpenSearch index after onboarding")
+            except Exception as e:
+                logger.error(
+                    "Failed to initialize OpenSearch index after onboarding",
+                    error=str(e),
+                )
                 return JSONResponse(
-                    {"Failed to initialize OpenSearch index after onboarding"},
+                    {
+                        "error": str(e),
+                    },
                     status_code=500,
                 )
 
@@ -1649,31 +1643,14 @@ async def _update_langflow_model_values(config, flows_service, llm_model=None, l
             )
 
         if not (embedding_model or embedding_provider or llm_model or llm_provider):
-            # 1. Update ALL configured LLM providers.
-            # Regression fix (#1587): the no-argument fallback used by
-            # reapply_all_settings previously only reapplied embedding providers,
-            # leaving LLM model values unset whenever flows were reset.
-            llm_providers = _configured_provider_names(config, _LLM_PROVIDER_NAMES)
-
-            current_llm_provider = config.agent.llm_provider.lower()
-            for provider in llm_providers:
-                # Use configured model for current provider, or None (first available) for others
-                provider_llm_model = (
-                    config.agent.llm_model
-                    if provider == current_llm_provider
-                    else None
-                )
-                await flows_service.change_langflow_model_value(
-                        provider,
-                        llm_model=provider_llm_model,
-                        force_llm_update=True
-                    )
-                logger.info(
-                    f"Successfully updated Langflow flows for LLM provider {provider}"
-                )
-
             # 2. Update ALL configured embedding providers
-            embedding_providers = _configured_provider_names(config, _EMBEDDING_PROVIDER_NAMES)
+            embedding_providers = []
+            if config.providers.openai.configured:
+                embedding_providers.append("openai")
+            if config.providers.watsonx.configured:
+                embedding_providers.append("watsonx")
+            if config.providers.ollama.configured:
+                embedding_providers.append("ollama")
 
             current_embedding_provider = config.knowledge.embedding_provider.lower()
             for provider in embedding_providers:
@@ -2019,10 +1996,10 @@ async def rollback_onboarding(
             deleted_conversations=deleted_conversations_count
         )
 
-    except Exception:
-        logger.error("Failed to rollback onboarding configuration")
+    except Exception as e:
+        logger.error("Failed to rollback onboarding configuration", error=str(e))
         return JSONResponse(
-            {"error": "Failed to rollback onboarding"}, status_code=500
+            {"error": f"Failed to rollback onboarding: {str(e)}"}, status_code=500
         )
 
 
@@ -2090,12 +2067,11 @@ async def update_docling_preset(
             settings=settings_toggles,
             preset_config=preset_config,
         )
-    except HTTPException:
-        # Preserve intended HTTP status codes (e.g. 400 for an invalid preset)
-        raise
+
     except Exception as e:
         logger.error("Failed to update docling settings", error=str(e))
-        raise HTTPException(status_code=500, detail="Failed to update docling settings") from e
+        raise HTTPException(status_code=500, detail=f"Failed to update docling settings: {str(e)}")
+
 
 async def refresh_openrag_docs(
     document_service=Depends(get_document_service),
@@ -2131,4 +2107,4 @@ async def refresh_openrag_docs(
         raise HTTPException(
             status_code=500,
             detail=f"Failed to refresh OpenRAG docs: {str(e)}",
-        )
+        )

{openrag-0.5.1 → openrag-0.5.1.dev1}/src/connectors/onedrive/connector.py

@@ -636,7 +636,7 @@ class OneDriveConnector(BaseConnector):
 
         # Check if ID contains '!' which indicates driveId!itemId format
         if "!" in file_id:
-            parts = file_id.split("!", 1)
+            parts = file_id.rsplit("!", 1)
             if len(parts) == 2:
                 drive_id = parts[0]
                 item_id = parts[1]
@@ -735,7 +735,7 @@ class OneDriveConnector(BaseConnector):
 
             # Build URL based on ID format
             if "!" in file_id:
-                parts = file_id.split("!", 1)
+                parts = file_id.rsplit("!", 1)
                 if len(parts) == 2:
                     drive_id = parts[0]
                     item_id = parts[1]
@@ -747,8 +747,11 @@ class OneDriveConnector(BaseConnector):
                             return content
 
                     # Try drives endpoint for driveId!itemId format
-                    url = f"{self._graph_base_url}/drives/{drive_id}/items/{file_id}/content"
-                    logger.info(f"Downloading via drives endpoint: {url}")
+                    if not item_id.startswith("s"):
+                        url = f"{self._graph_base_url}/drives/{drive_id}/items/{item_id}/content"
+                        logger.info(f"Downloading via drives endpoint: {url}")
+                    else:
+                        url = f"{self._graph_base_url}/me/drive/items/{file_id}/content"
                 else:
                     url = f"{self._graph_base_url}/me/drive/items/{file_id}/content"
             else:
@@ -883,7 +886,18 @@ class OneDriveConnector(BaseConnector):
         files: list[dict[str, Any]] = []
 
         try:
-            url = f"{self._graph_base_url}/me/drive/items/{folder_id}/children"
+            drive_id = None
+            if "!" in folder_id:
+                parts = folder_id.rsplit("!", 1)
+                if len(parts) == 2:
+                    potential_drive_id, item_id = parts
+                    if not item_id.startswith("s"):
+                        drive_id = potential_drive_id
+                        url = f"{self._graph_base_url}/drives/{drive_id}/items/{item_id}/children"
+
+            if not drive_id:
+                url = f"{self._graph_base_url}/me/drive/items/{folder_id}/children"
+
             params = dict(self._default_params)
 
             response = await self._make_graph_request(url, params=params)

{openrag-0.5.1 → openrag-0.5.1.dev1}/src/models/processors.py

@@ -27,15 +27,6 @@ from .tasks import FileTask, TaskStatus, UploadTask
 
 logger = get_logger(__name__)
 
-# Image file extensions that require OCR
-IMAGE_EXTENSIONS = {"bmp", "jpeg", "jpg", "png", "tiff", "tif", "webp"}
-
-
-def is_image_file(filename: str) -> bool:
-    """Check if a file is an image based on its extension."""
-    ext = filename.split(".")[-1].lower() if "." in filename else ""
-    return ext in IMAGE_EXTENSIONS
-
 if TYPE_CHECKING:
     from connectors.base import DocumentACL
 
@@ -464,18 +455,6 @@ class DocumentFileProcessor(TaskProcessor):
         file_task.updated_at = time.time()
 
         try:
-            # Check if OCR is disabled and file is an image
-            filename = os.path.basename(item)
-            config = get_openrag_config()
-            ocr_enabled = config.knowledge.ocr
-            
-            if not ocr_enabled and is_image_file(filename):
-                file_task.status = TaskStatus.FAILED
-                file_task.error = f"The file '{filename}' is an image file and cannot be ingested because OCR is disabled."
-                file_task.updated_at = time.time()
-                upload_task.failed_files += 1
-                return
-
             # Compute hash
             file_hash = hash_id(item)
 
@@ -599,17 +578,6 @@ class ConnectorFileProcessor(TaskProcessor):
                     file_task.updated_at = time.time()
                     upload_task.failed_files += 1
                     return
-                
-                # Check if OCR is disabled and file is an image
-                config = get_openrag_config()
-                ocr_enabled = config.knowledge.ocr
-                
-                if not ocr_enabled and is_image_file(file_task.filename):
-                    file_task.status = TaskStatus.FAILED
-                    file_task.error = f"The file '{file_task.filename}' is an image file and cannot be ingested because OCR is disabled."
-                    file_task.updated_at = time.time()
-                    upload_task.failed_files += 1
-                    return
 
             # Get file content from connector
             try:
@@ -674,17 +642,6 @@ class ConnectorFileProcessor(TaskProcessor):
                 file_task.updated_at = time.time()
                 upload_task.failed_files += 1
                 return
-            
-            # Check if OCR is disabled and file is an image
-            config = get_openrag_config()
-            ocr_enabled = config.knowledge.ocr
-            
-            if not ocr_enabled and is_image_file(name):
-                file_task.status = TaskStatus.FAILED
-                file_task.error = f"The file '{name}' is an image file and cannot be ingested because OCR is disabled."
-                file_task.updated_at = time.time()
-                upload_task.failed_files += 1
-                return
 
             if not self.user_id:
                 raise ValueError("user_id not provided to ConnectorFileProcessor")

{openrag-0.5.1 → openrag-0.5.1.dev1/src/openrag.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: openrag
-Version: 0.5.1
+Version: 0.5.1.dev1
 Summary: OpenRAG is a comprehensive Retrieval-Augmented Generation platform that enables intelligent document search and AI-powered conversations.
 Classifier: Development Status :: 4 - Beta
 Classifier: Environment :: Console

{openrag-0.5.1 → openrag-0.5.1.dev1}/src/services/api_key_service.py

@@ -205,7 +205,7 @@ class APIKeyService:
         jwt_token: str = None,
     ) -> Dict[str, Any]:
         """
-        List all active (non-revoked) API keys for a user (without the actual keys).
+        List all API keys for a user (without the actual keys).
 
         Args:
             user_id: The user's ID
@@ -220,12 +220,7 @@ class APIKeyService:
             # Search for user's keys
             search_body = {
                 "query": {
-                    "bool": {
-                        "must": [
-                            {"term": {"user_id": user_id}},
-                            {"term": {"revoked": False }},
-                        ]
-                    }
+                    "term": {"user_id": user_id}
                 },
                 "sort": [{"created_at": {"order": "desc"}}],
                 "_source": [

{openrag-0.5.1 → openrag-0.5.1.dev1}/src/services/search_service.py

@@ -2,7 +2,6 @@ import asyncio
 import copy
 import os
 import json
-import re
 from collections import Counter
 from typing import Any, Dict
 from agentd.tool_decorator import tool
@@ -23,19 +22,6 @@ EMBED_RETRY_MAX_DELAY = 8.0
 _global_search_service = None
 
 
-def _is_exact_token_query(query: str) -> bool:
-    """Return True for code/token-like queries that should not allow partial fuzzy matches."""
-    if not query or len(query.strip()) < 3:
-        return False
-
-    query = query.strip()
-
-    if re.search(r"[^a-zA-Z0-9\s]", query):
-        return True
-
-    return bool(re.search(r"[a-zA-Z]", query) and re.search(r"\d", query))
-
-
 def register_search_service(service: "SearchService") -> None:
     """
     Explicitly register the active search service for the @tool wrapper.
@@ -598,17 +584,8 @@ class SearchService:
                     )
                 )
             }
-            
-            # Determine if we should apply exact-token filtering
-            should_filter_exact = bool(exact_files) or _is_exact_token_query(query)
-            
-            if should_filter_exact:
-                if exact_files:
-                    # Filter to only chunks from files with exact matches
-                    chunks = [chunk for chunk in chunks if chunk.get("filename") in exact_files]
-                else:
-                    # No exact matches found for a token-like query - return empty results
-                    chunks = []
+            if exact_files:
+                chunks = [chunk for chunk in chunks if chunk.get("filename") in exact_files]
 
                 def _build_terms_agg(field: str) -> Dict[str, Any]:
                     counts = Counter(

{openrag-0.5.1 → openrag-0.5.1.dev1}/src/tui/_assets/docker-compose.yml

@@ -1,6 +1,6 @@
 services:
   opensearch:
-    image: docker.io/langflowai/openrag-opensearch:${OPENRAG_VERSION:-latest}
+    image: langflowai/openrag-opensearch:${OPENRAG_VERSION:-latest}
     # build:
     #   context: .
     #   dockerfile: Dockerfile
@@ -26,7 +26,7 @@ services:
     restart: unless-stopped
 
   dashboards:
-    image: docker.io/opensearchproject/opensearch-dashboards:3.0.0
+    image: opensearchproject/opensearch-dashboards:3.0.0
     container_name: osdash
     depends_on:
       - opensearch
@@ -38,7 +38,7 @@ services:
       - "5601:5601"
 
   openrag-backend:
-    image: docker.io/langflowai/openrag-backend:${OPENRAG_VERSION:-latest}
+    image: langflowai/openrag-backend:${OPENRAG_VERSION:-latest}
     build:
       context: .
       dockerfile: Dockerfile.backend
@@ -116,7 +116,7 @@ services:
       - "host.docker.internal:host-gateway"
 
   openrag-frontend:
-    image: docker.io/langflowai/openrag-frontend:${OPENRAG_VERSION:-latest}
+    image: langflowai/openrag-frontend:${OPENRAG_VERSION:-latest}
     build:
       context: .
       dockerfile: Dockerfile.frontend
@@ -129,7 +129,7 @@ services:
       - "${FRONTEND_PORT:-3000}:3000"
 
   langflow:
-    image: docker.io/langflowai/openrag-langflow:${OPENRAG_VERSION:-latest}
+    image: langflowai/openrag-langflow:${OPENRAG_VERSION:-latest}
     build:
       context: .
       dockerfile: Dockerfile.langflow