Merge branch 'main' into RAGAnything

lightrag/api/lightrag_server.py

@@ -25,6 +25,7 @@ from lightrag.api.utils_api import (
     display_splash_screen,
     check_env_file,
 )
+from lightrag.llm.openai import openai_complete_if_cache, openai_embed
 from .config import (
     global_args,
     update_uvicorn_mode_config,
@@ -61,6 +62,8 @@ from lightrag.kg.shared_storage import (
 )
 from fastapi.security import OAuth2PasswordRequestForm
 from lightrag.api.auth import auth_handler
+from lightrag.ragmanager import RAGManager
+from raganything import RAGAnything, RAGAnythingConfig
 
 # use the .env that is inside the current folder
 # allows to use different .env file for each lightrag instance
@@ -619,10 +622,147 @@ def create_app(args):
         logger.error(f"Failed to initialize LightRAG: {e}")
         raise
 
+    # Initialize RAGAnything with comprehensive error handling
+    rag_anything = None
+    raganything_enabled = False
+    raganything_error_message = None
+
+    try:
+        api_key = get_env_value("LLM_BINDING_API_KEY", "", str)
+        base_url = get_env_value("LLM_BINDING_HOST", "", str)
+
+        # Validate required configuration
+        if not api_key:
+            raise ValueError(
+                "LLM_BINDING_API_KEY is required for RAGAnything functionality"
+            )
+        if not base_url:
+            raise ValueError(
+                "LLM_BINDING_HOST is required for RAGAnything functionality"
+            )
+
+        config = RAGAnythingConfig(
+            working_dir=args.working_dir or "./rag_storage",
+            parser="mineru",  # Parser selection: mineru or docling
+            parse_method="auto",  # Parse method: auto, ocr, or txt
+            enable_image_processing=True,
+            enable_table_processing=True,
+            enable_equation_processing=True,
+        )
+
+        # Define LLM model function
+        def llm_model_func(prompt, system_prompt=None, history_messages=[], **kwargs):
+            return openai_complete_if_cache(
+                "gpt-4o-mini",
+                prompt,
+                system_prompt=system_prompt,
+                history_messages=history_messages,
+                api_key=api_key,
+                base_url=base_url,
+                **kwargs,
+            )
+
+        # Define vision model function for image processing
+        def vision_model_func(
+            prompt, system_prompt=None, history_messages=[], image_data=None, **kwargs
+        ):
+            if image_data:
+                return openai_complete_if_cache(
+                    "gpt-4o",
+                    "",
+                    system_prompt=None,
+                    history_messages=[],
+                    messages=[
+                        {"role": "system", "content": system_prompt}
+                        if system_prompt
+                        else None,
+                        {
+                            "role": "user",
+                            "content": [
+                                {"type": "text", "text": prompt},
+                                {
+                                    "type": "image_url",
+                                    "image_url": {
+                                        "url": f"data:image/jpeg;base64,{image_data}"
+                                    },
+                                },
+                            ],
+                        }
+                        if image_data
+                        else {"role": "user", "content": prompt},
+                    ],
+                    api_key=api_key,
+                    base_url=base_url,
+                    **kwargs,
+                )
+            else:
+                return llm_model_func(prompt, system_prompt, history_messages, **kwargs)
+
+        # Define embedding function
+        raganything_embedding_func = EmbeddingFunc(
+            embedding_dim=3072,
+            max_token_size=8192,
+            func=lambda texts: openai_embed(
+                texts,
+                model="text-embedding-3-large",
+                api_key=api_key,
+                base_url=base_url,
+            ),
+        )
+
+        # Initialize RAGAnything with new dataclass structure
+        logger.info("Initializing RAGAnything functionality...")
+        rag_anything = RAGAnything(
+            lightrag=rag,
+            config=config,
+            llm_model_func=llm_model_func,
+            vision_model_func=vision_model_func,
+            embedding_func=raganything_embedding_func,
+        )
+
+        logger.info("Check the download status of the RAGAnything parser...")
+        rag_anything.verify_parser_installation_once()
+
+        RAGManager.set_rag(rag_anything)
+        raganything_enabled = True
+        logger.info(
+            "The RAGAnything feature has been successfully enabled, supporting multimodal document processing functionality"
+        )
+
+    except ImportError as e:
+        raganything_error_message = (
+            f"RAGAnything dependency package not installed: {str(e)}"
+        )
+        logger.warning(f"{raganything_error_message}")
+        logger.info(
+            "Please run 'pip install raganything' to install dependency packages to enable multimodal document processing functionality"
+        )
+    except ValueError as e:
+        raganything_error_message = f"RAGAnything configuration error: {str(e)}"
+        logger.warning(f"{raganything_error_message}")
+        logger.info(
+            "Please check if the environment variables LLM-BINDING_API_KEY and LLM-BINDING_HOST are set correctly"
+        )
+    except Exception as e:
+        raganything_error_message = f"RAGAnything initialization failed: {str(e)}"
+        logger.error(f" {raganything_error_message}")
+        logger.info(
+            "The system will run in basic mode and only support standard document processing functions"
+        )
+    except Exception as e:
+        logger.error(f"Failed to initialize LightRAG: {e}")
+        raise
+
+    if not raganything_enabled:
+        logger.info(
+            "The system has been downgraded to basic mode, but LightRAG core functions are still available"
+        )
+
     # Add routes
     app.include_router(
         create_document_routes(
             rag,
+            rag_anything,
             doc_manager,
             api_key,
         )

lightrag/api/routers/document_routes.py

@@ -3,6 +3,8 @@ This module contains all document-related routes for the LightRAG API.
 """
 
 import asyncio
+import json
+import uuid
 from lightrag.utils import logger, get_pinyin_sort_key
 import aiofiles
 import shutil
@@ -18,6 +20,7 @@ from fastapi import (
     File,
     HTTPException,
     UploadFile,
+    Form,
 )
 from pydantic import BaseModel, Field, field_validator
 
@@ -26,6 +29,7 @@ from lightrag.base import DeletionResult, DocProcessingStatus, DocStatus
 from lightrag.utils import generate_track_id
 from lightrag.api.utils_api import get_combined_auth_dependency
 from ..config import global_args
+from raganything import RAGAnything
 
 
 # Function to format datetime to ISO format string with timezone information
@@ -107,6 +111,81 @@ def sanitize_filename(filename: str, input_dir: Path) -> str:
     return clean_name
 
 
+class SchemeConfig(BaseModel):
+    """Configuration model for processing schemes.
+
+    Defines the processing framework and optional extractor to use for document processing.
+
+    Attributes:
+        framework (Literal['lightrag', 'raganything']): Processing framework to use.
+            - "lightrag": Standard LightRAG processing for text-based documents
+            - "raganything": Advanced multimodal processing with image/table/equation support
+        extractor (Literal['mineru', 'docling', '']): Document extraction tool to use.
+            - "mineru": MinerU parser for comprehensive document parsing
+            - "docling": Docling parser for office document processing
+            - "": Default/automatic extractor selection
+        modelSource (Literal["huggingface", "modelscope", "local", ""]): The model source used by Mineru.
+            - "huggingface": Using pre-trained models from the Hugging Face model library
+            - "modelscope": using model resources on ModelScope platform
+            - "local": Use custom models deployed locally
+            - "":Maintain the default model source configuration of the system (usually huggingface)
+    """
+
+    framework: Literal["lightrag", "raganything"]
+    extractor: Literal["mineru", "docling", ""] = ""  # 默认值
+    modelSource: Literal["huggingface", "modelscope", "local", ""] = ""
+
+
+class Scheme(BaseModel):
+    """Base model for processing schemes.
+
+    Attributes:
+        name (str): Human-readable name for the processing scheme
+        config (SchemeConfig): Configuration settings for the scheme
+    """
+
+    name: str
+    config: SchemeConfig
+
+
+class Scheme_include_id(Scheme):
+    """Scheme model with unique identifier included.
+
+    Extends the base Scheme model to include a unique ID field for
+    identification and management operations.
+
+    Attributes:
+        id (int): Unique identifier for the scheme
+        name (str): Inherited from Scheme
+        config (SchemeConfig): Inherited from Scheme
+    """
+
+    id: int
+
+
+class SchemesResponse(BaseModel):
+    """Response model for scheme management operations.
+
+    Used for all scheme-related endpoints to provide consistent response format
+    for scheme retrieval, creation, update, and deletion operations.
+
+    Attributes:
+        status (str): Operation status ("success", "error")
+        message (Optional[str]): Additional message with operation details
+        data (Optional[List[Dict[str, Any]]]): List of scheme objects when retrieving schemes
+    """
+
+    status: str = Field(..., description="Operation status")
+    message: Optional[str] = Field(None, description="Additional message")
+    data: Optional[List[Dict[str, Any]]] = Field(None, description="List of schemes")
+
+
+class ScanRequest(BaseModel):
+    """Request model for document scanning operations."""
+
+    schemeConfig: SchemeConfig = Field(..., description="Scanning scheme configuration")
+
+
 class ScanResponse(BaseModel):
     """Response model for document scanning operation
 
@@ -372,12 +451,20 @@ class DocStatusResponse(BaseModel):
         default=None, description="Additional metadata about the document"
     )
     file_path: str = Field(description="Path to the document file")
+    scheme_name: str = Field(
+        default=None, description="Name of the processing scheme used for this document"
+    )
+    multimodal_content: Optional[list[dict[str, Any]]] = Field(
+        default=None, description="Multimodal content of the document"
+    )
 
     class Config:
         json_schema_extra = {
             "example": {
                 "id": "doc_123456",
                 "content_summary": "Research paper on machine learning",
+                "scheme_name": "lightrag",
+                "multimodal_content": [],
                 "content_length": 15240,
                 "status": "PROCESSED",
                 "created_at": "2025-03-31T12:34:56",
@@ -411,6 +498,8 @@ class DocsStatusesResponse(BaseModel):
                         {
                             "id": "doc_123",
                             "content_summary": "Pending document",
+                            "scheme_name": "lightrag",
+                            "multimodal_content": [],
                             "content_length": 5000,
                             "status": "PENDING",
                             "created_at": "2025-03-31T10:00:00",
@@ -426,6 +515,8 @@ class DocsStatusesResponse(BaseModel):
                         {
                             "id": "doc_456",
                             "content_summary": "Processed document",
+                            "scheme_name": "lightrag",
+                            "multimodal_content": [],
                             "content_length": 8000,
                             "status": "PROCESSED",
                             "created_at": "2025-03-31T09:00:00",
@@ -779,7 +870,7 @@ def get_unique_filename_in_enqueued(target_dir: Path, original_name: str) -> str
 
 
 async def pipeline_enqueue_file(
-    rag: LightRAG, file_path: Path, track_id: str = None
+    rag: LightRAG, file_path: Path, track_id: str = None, scheme_name: str = None
 ) -> tuple[bool, str]:
     """Add a file to the queue for processing
 
@@ -787,6 +878,8 @@ async def pipeline_enqueue_file(
         rag: LightRAG instance
         file_path: Path to the saved file
         track_id: Optional tracking ID, if not provided will be generated
+        scheme_name (str, optional): Processing scheme name for categorization.
+            Defaults to None
     Returns:
         tuple: (success: bool, track_id: str)
     """
@@ -1159,7 +1252,10 @@ async def pipeline_enqueue_file(
 
             try:
                 await rag.apipeline_enqueue_documents(
-                    content, file_paths=file_path.name, track_id=track_id
+                    content,
+                    file_paths=file_path.name,
+                    track_id=track_id,
+                    scheme_name=scheme_name,
                 )
 
                 logger.info(
@@ -1243,17 +1339,21 @@ async def pipeline_enqueue_file(
                 logger.error(f"Error deleting file {file_path}: {str(e)}")
 
 
-async def pipeline_index_file(rag: LightRAG, file_path: Path, track_id: str = None):
+async def pipeline_index_file(
+    rag: LightRAG, file_path: Path, track_id: str = None, scheme_name: str = None
+):
     """Index a file with track_id
 
     Args:
         rag: LightRAG instance
         file_path: Path to the saved file
         track_id: Optional tracking ID
+        scheme_name (str, optional): Processing scheme name for categorization.
+            Defaults to None
     """
     try:
         success, returned_track_id = await pipeline_enqueue_file(
-            rag, file_path, track_id
+            rag, file_path, track_id, scheme_name
         )
         if success:
             await rag.apipeline_process_enqueue_documents()
@@ -1264,7 +1364,7 @@ async def pipeline_index_file(rag: LightRAG, file_path: Path, track_id: str = No
 
 
 async def pipeline_index_files(
-    rag: LightRAG, file_paths: List[Path], track_id: str = None
+    rag: LightRAG, file_paths: List[Path], track_id: str = None, scheme_name: str = None
 ):
     """Index multiple files sequentially to avoid high CPU load
 
@@ -1272,6 +1372,8 @@ async def pipeline_index_files(
         rag: LightRAG instance
         file_paths: Paths to the files to index
         track_id: Optional tracking ID to pass to all files
+        scheme_name (str, optional): Processing scheme name for categorization.
+            Defaults to None
     """
     if not file_paths:
         return
@@ -1285,7 +1387,9 @@ async def pipeline_index_files(
 
         # Process files sequentially with track_id
         for file_path in sorted_file_paths:
-            success, _ = await pipeline_enqueue_file(rag, file_path, track_id)
+            success, _ = await pipeline_enqueue_file(
+                rag, file_path, track_id, scheme_name
+            )
             if success:
                 enqueued = True
 
@@ -1297,6 +1401,61 @@ async def pipeline_index_files(
         logger.error(traceback.format_exc())
 
 
+async def pipeline_index_files_raganything(
+    rag_anything: RAGAnything,
+    file_paths: List[Path],
+    scheme_name: str = None,
+    parser: str = None,
+    source: str = None,
+):
+    """Index multiple files using RAGAnything framework for multimodal processing.
+
+    Args:
+        rag_anything (RAGAnything): RAGAnything instance for multimodal document processing
+        file_paths (List[Path]): List of file paths to be processed
+        track_id (str, optional): Tracking ID for batch monitoring. Defaults to None.
+        scheme_name (str, optional): Processing scheme name for categorization.
+            Defaults to None.
+        parser (str, optional): Document extraction tool to use.
+            Defaults to None.
+        source (str, optional): The model source used by Mineru.
+            Defaults to None.
+
+    Note:
+        - Uses RAGAnything's process_document_complete_lightrag_api method for each file
+        - Supports multimodal content processing (images, tables, equations)
+        - Files are processed with "auto" parse method and "modelscope" source
+        - Output is saved to "./output" directory
+        - Errors are logged but don't stop processing of remaining files
+    """
+    if not file_paths:
+        return
+
+    try:
+        # Use get_pinyin_sort_key for Chinese pinyin sorting
+        sorted_file_paths = sorted(
+            file_paths, key=lambda p: get_pinyin_sort_key(str(p))
+        )
+
+        # Process files sequentially with track_id
+        for file_path in sorted_file_paths:
+            success = await rag_anything.process_document_complete_lightrag_api(
+                file_path=str(file_path),
+                output_dir="./output",
+                parse_method="auto",
+                scheme_name=scheme_name,
+                parser=parser,
+                source=source,
+            )
+            if success:
+                pass
+
+    except Exception as e:
+        error_msg = f"Error indexing files: {str(e)}"
+        logger.error(error_msg)
+        logger.error(traceback.format_exc())
+
+
 async def pipeline_index_texts(
     rag: LightRAG,
     texts: List[str],
@@ -1326,24 +1485,67 @@ async def pipeline_index_texts(
 
 
 async def run_scanning_process(
-    rag: LightRAG, doc_manager: DocumentManager, track_id: str = None
+    rag: LightRAG,
+    rag_anything: RAGAnything,
+    doc_manager: DocumentManager,
+    track_id: str = None,
+    schemeConfig=None,
 ):
     """Background task to scan and index documents
 
     Args:
         rag: LightRAG instance
+        rag_anythingL: RAGAnything instance
         doc_manager: DocumentManager instance
         track_id: Optional tracking ID to pass to all scanned files
+        schemeConfig: Scanning scheme configuration.
+            Defaults to None
     """
     try:
         new_files = doc_manager.scan_directory_for_new_files()
         total_files = len(new_files)
         logger.info(f"Found {total_files} files to index.")
 
+        from lightrag.kg.shared_storage import get_namespace_data
+
+        pipeline_status = await get_namespace_data("pipeline_status")
+        is_pipeline_scan_busy = pipeline_status.get("scan_disabled", False)
+        is_pipeline_busy = pipeline_status.get("busy", False)
+
+        scheme_name = schemeConfig.framework
+        extractor = schemeConfig.extractor
+        modelSource = schemeConfig.modelSource
+
         if new_files:
             # Process all files at once with track_id
-            await pipeline_index_files(rag, new_files, track_id)
-            logger.info(f"Scanning process completed: {total_files} files Processed.")
+            if is_pipeline_busy:
+                logger.info(
+                    "Pipe is currently busy, skipping processing to avoid conflicts..."
+                )
+                return
+            if is_pipeline_scan_busy:
+                logger.info(
+                    "Pipe is currently busy, skipping processing to avoid conflicts..."
+                )
+                return
+            if scheme_name == "lightrag":
+                await pipeline_index_files(
+                    rag, new_files, track_id, scheme_name=scheme_name
+                )
+                logger.info(
+                    f"Scanning process completed with lightrag: {total_files} files Processed."
+                )
+            elif scheme_name == "raganything":
+                await pipeline_index_files_raganything(
+                    rag_anything,
+                    new_files,
+                    scheme_name=scheme_name,
+                    parser=extractor,
+                    source=modelSource,
+                )
+                logger.info(
+                    f"Scanning process completed with raganything: {total_files} files Processed."
+                )
         else:
             # No new files to index, check if there are any documents in the queue
             logger.info(
@@ -1554,15 +1756,250 @@ async def background_delete_documents(
 
 
 def create_document_routes(
-    rag: LightRAG, doc_manager: DocumentManager, api_key: Optional[str] = None
+    rag: LightRAG,
+    rag_anything: RAGAnything,
+    doc_manager: DocumentManager,
+    api_key: Optional[str] = None,
 ):
     # Create combined auth dependency for document routes
     combined_auth = get_combined_auth_dependency(api_key)
 
+    @router.get(
+        "/schemes",
+        response_model=SchemesResponse,
+        dependencies=[Depends(combined_auth)],
+    )
+    async def get_all_schemes():
+        """Get all available processing schemes.
+
+        Retrieves the complete list of processing schemes from the schemes.json file.
+        Each scheme defines a processing framework (lightrag/raganything) and
+        optional extractor configuration (mineru/docling).
+
+        Returns:
+            SchemesResponse: Response containing:
+                - status (str): Operation status ("success")
+                - message (str): Success message
+                - data (List[Dict]): List of all available schemes with their configurations
+
+        Raises:
+            HTTPException: If file reading fails or JSON parsing errors occur (500)
+        """
+        SCHEMES_FILE = Path("./examples/schemes.json")
+
+        if SCHEMES_FILE.exists():
+            with open(SCHEMES_FILE, "r", encoding="utf-8") as f:
+                try:
+                    current_data = json.load(f)
+                except json.JSONDecodeError:
+                    current_data = []
+        else:
+            current_data = []
+            SCHEMES_FILE.parent.mkdir(parents=True, exist_ok=True)
+            with open(SCHEMES_FILE, "w") as f:
+                json.dump(current_data, f)
+
+        return SchemesResponse(
+            status="success",
+            message="Schemes retrieved successfully",
+            data=current_data,
+        )
+
+    @router.post(
+        "/schemes",
+        response_model=SchemesResponse,
+        dependencies=[Depends(combined_auth)],
+    )
+    async def save_schemes(schemes: list[Scheme_include_id]):
+        """Save/update processing schemes in batch.
+
+        Updates existing schemes with new configuration data. This endpoint performs
+        a partial update by modifying existing schemes based on their IDs while
+        preserving other schemes in the file.
+
+        Args:
+            schemes (list[Scheme_include_id]): List of schemes to update, each containing:
+                - id (int): Unique identifier of the scheme to update
+                - name (str): Display name for the scheme
+                - config (SchemeConfig): Configuration object with framework and extractor settings
+
+        Returns:
+            SchemesResponse: Response containing:
+                - status (str): Operation status ("success")
+                - message (str): Success message with count of saved schemes
+                - data (List[Dict]): Updated list of all schemes after modification
+
+        Raises:
+            HTTPException: If file operations fail or JSON processing errors occur (500)
+        """
+        try:
+            SCHEMES_FILE = Path("./examples/schemes.json")
+
+            if SCHEMES_FILE.exists():
+                with open(SCHEMES_FILE, "r", encoding="utf-8") as f:
+                    try:
+                        current_data = json.load(f)
+                    except json.JSONDecodeError:
+                        current_data = []
+            else:
+                current_data = []
+
+            updated_item = {
+                "id": schemes[0].id,
+                "name": schemes[0].name,
+                "config": {
+                    "framework": schemes[0].config.framework,
+                    "extractor": schemes[0].config.extractor,
+                    "modelSource": schemes[0].config.modelSource,
+                },
+            }
+            # 保存新方案
+            for item in current_data:
+                if item["id"] == updated_item["id"]:
+                    item["name"] = updated_item["name"]
+                    item["config"]["framework"] = updated_item["config"]["framework"]
+                    item["config"]["extractor"] = updated_item["config"]["extractor"]
+                    item["config"]["modelSource"] = updated_item["config"][
+                        "modelSource"
+                    ]
+                    break
+
+            # 写回文件
+            with open(SCHEMES_FILE, "w", encoding="utf-8") as f:
+                json.dump(current_data, f, indent=4)
+
+            # 返回响应（从文件重新读取确保一致性）
+            with open(SCHEMES_FILE, "r", encoding="utf-8") as f:
+                data = json.load(f)
+
+            return SchemesResponse(
+                status="success",
+                message=f"Successfully saved {len(schemes)} schemes",
+                data=data,
+            )
+        except Exception as e:
+            raise HTTPException(status_code=500, detail=str(e))
+
+    @router.post(
+        "/schemes/add",
+        response_model=Scheme_include_id,
+        dependencies=[Depends(combined_auth)],
+    )
+    async def add_scheme(scheme: Scheme):
+        """Add a new processing scheme.
+
+        Creates a new processing scheme with auto-generated ID and saves it to the
+        schemes configuration file. The new scheme will be available for document
+        processing operations.
+
+        Args:
+            scheme (Scheme): New scheme to add, containing:
+                - name (str): Display name for the scheme
+                - config (SchemeConfig): Configuration with framework and extractor settings
+
+        Returns:
+            Scheme_include_id: The created scheme with auto-generated ID, containing:
+                - id (int): Auto-generated unique identifier
+                - name (str): Display name of the scheme
+                - config (SchemeConfig): Processing configuration
+
+        Raises:
+            HTTPException: If file operations fail or ID generation conflicts occur (500)
+        """
+        try:
+            SCHEMES_FILE = Path("./examples/schemes.json")
+
+            if SCHEMES_FILE.exists():
+                with open(SCHEMES_FILE, "r", encoding="utf-8") as f:
+                    try:
+                        current_data = json.load(f)
+                    except json.JSONDecodeError:
+                        current_data = []
+            else:
+                current_data = []
+
+            # 生成新ID（简单实现，实际项目应该用数据库自增ID）
+            new_id = uuid.uuid4().int >> 96  # 生成一个较小的整数ID
+            while new_id in current_data:
+                new_id = uuid.uuid4().int >> 96
+
+            new_scheme = {
+                "id": new_id,
+                "name": scheme.name,
+                "config": {
+                    "framework": scheme.config.framework,
+                    "extractor": scheme.config.extractor,
+                    "modelSource": scheme.config.modelSource,
+                },
+            }
+
+            current_data.append(new_scheme)
+
+            with open(SCHEMES_FILE, "w", encoding="utf-8") as f:
+                json.dump(current_data, f, ensure_ascii=False, indent=2)
+
+            return new_scheme
+        except Exception as e:
+            raise HTTPException(status_code=500, detail=str(e))
+
+    @router.delete(
+        "/schemes/{scheme_id}",
+        response_model=Dict[str, str],
+        dependencies=[Depends(combined_auth)],
+    )
+    async def delete_scheme(scheme_id: int):
+        """Delete a specific processing scheme by ID.
+
+        Removes a processing scheme from the configuration file. Once deleted,
+        the scheme will no longer be available for document processing operations.
+
+        Args:
+            scheme_id (int): Unique identifier of the scheme to delete
+
+        Returns:
+            Dict[str, str]: Success message containing:
+                - message (str): Confirmation message with the deleted scheme ID
+
+        Raises:
+            HTTPException:
+                - 404: If the scheme with the specified ID is not found
+                - 500: If file operations fail or other errors occur
+        """
+        try:
+            SCHEMES_FILE = Path("./examples/schemes.json")
+
+            if SCHEMES_FILE.exists():
+                with open(SCHEMES_FILE, "r", encoding="utf-8") as f:
+                    try:
+                        current_data = json.load(f)
+                    except json.JSONDecodeError:
+                        current_data = []
+            else:
+                current_data = []
+
+            current_data_dict = {scheme["id"]: scheme for scheme in current_data}
+
+            if scheme_id not in current_data_dict:  # 直接检查 id 是否存在
+                raise HTTPException(status_code=404, detail="Scheme not found")
+
+            for i, scheme in enumerate(current_data):
+                if scheme["id"] == scheme_id:
+                    del current_data[i]  # 直接删除列表中的元素
+                    break
+
+            with open(SCHEMES_FILE, "w", encoding="utf-8") as f:
+                json.dump(current_data, f, ensure_ascii=False, indent=2)
+
+            return {"message": f"Scheme {scheme_id} deleted successfully"}
+        except Exception as e:
+            raise HTTPException(status_code=500, detail=str(e))
+
     @router.post(
         "/scan", response_model=ScanResponse, dependencies=[Depends(combined_auth)]
     )
-    async def scan_for_new_documents(background_tasks: BackgroundTasks):
+    async def scan_for_new_documents(
+        request: ScanRequest, background_tasks: BackgroundTasks
+    ):
         """
         Trigger the scanning process for new documents.
 
@@ -1577,7 +2014,14 @@ def create_document_routes(
         track_id = generate_track_id("scan")
 
         # Start the scanning process in the background with track_id
-        background_tasks.add_task(run_scanning_process, rag, doc_manager, track_id)
+        background_tasks.add_task(
+            run_scanning_process,
+            rag,
+            rag_anything,
+            doc_manager,
+            track_id,
+            schemeConfig=request.schemeConfig,
+        )
         return ScanResponse(
             status="scanning_started",
             message="Scanning process has been initiated in the background",
@@ -1588,7 +2032,9 @@ def create_document_routes(
         "/upload", response_model=InsertResponse, dependencies=[Depends(combined_auth)]
     )
     async def upload_to_input_dir(
-        background_tasks: BackgroundTasks, file: UploadFile = File(...)
+        background_tasks: BackgroundTasks,
+        file: UploadFile = File(...),
+        schemeId: str = Form(...),
     ):
         """
         Upload a file to the input directory and index it.
@@ -1599,7 +2045,9 @@ def create_document_routes(
 
         Args:
             background_tasks: FastAPI BackgroundTasks for async processing
-            file (UploadFile): The file to be uploaded. It must have an allowed extension.
+            file (UploadFile): The file to be uploaded. It must have an allowed extension
+            schemeId (str): ID of the processing scheme to use for this file. The scheme
+                determines whether to use LightRAG or RAGAnything framework for processing.
 
         Returns:
             InsertResponse: A response object containing the upload status and a message.
@@ -1632,8 +2080,62 @@ def create_document_routes(
 
             track_id = generate_track_id("upload")
 
-            # Add to background tasks and get track_id
-            background_tasks.add_task(pipeline_index_file, rag, file_path, track_id)
+            def load_config():
+                try:
+                    SCHEMES_FILE = Path("./examples/schemes.json")
+                    with open(SCHEMES_FILE, "r") as f:
+                        schemes = json.load(f)
+                    for scheme in schemes:
+                        if str(scheme.get("id")) == schemeId:
+                            return scheme.get("config", {})
+                    return {}
+                except Exception as e:
+                    logger.error(
+                        f"Failed to load config for scheme {schemeId}: {str(e)}"
+                    )
+                    return {}
+
+            config = load_config()
+            current_framework = config.get("framework")
+            current_extractor = config.get("extractor")
+            current_modelSource = config.get("modelSource")
+            doc_pre_id = f"doc-pre-{safe_filename}"
+
+            if current_framework and current_framework == "lightrag":
+                # Add to background tasks and get track_id
+                background_tasks.add_task(
+                    pipeline_index_file,
+                    rag,
+                    file_path,
+                    track_id,
+                    scheme_name=current_framework,
+                )
+            else:
+                background_tasks.add_task(
+                    rag_anything.process_document_complete_lightrag_api,
+                    file_path=str(file_path),
+                    output_dir="./output",
+                    parse_method="auto",
+                    scheme_name=current_framework,
+                    parser=current_extractor,
+                    source=current_modelSource,
+                )
+
+            await rag.doc_status.upsert(
+                {
+                    doc_pre_id: {
+                        "status": DocStatus.READY,
+                        "content": "",
+                        "content_summary": "",
+                        "multimodal_content": [],
+                        "scheme_name": current_framework,
+                        "content_length": 0,
+                        "created_at": "",
+                        "updated_at": "",
+                        "file_path": safe_filename,
+                    }
+                }
+            )
 
             return InsertResponse(
                 status="success",
@@ -1854,6 +2356,42 @@ def create_document_routes(
                         f"Successfully dropped all {storage_success_count} storage components"
                     )
 
+            # Clean all parse_cache entries after successful storage drops
+            if storage_success_count > 0:
+                try:
+                    if "history_messages" in pipeline_status:
+                        pipeline_status["history_messages"].append(
+                            "Cleaning parse_cache entries"
+                        )
+
+                    parse_cache_result = await rag.aclean_all_parse_cache()
+                    if parse_cache_result.get("error"):
+                        cache_error_msg = f"Warning: Failed to clean parse_cache: {parse_cache_result['error']}"
+                        logger.warning(cache_error_msg)
+                        if "history_messages" in pipeline_status:
+                            pipeline_status["history_messages"].append(cache_error_msg)
+                    else:
+                        deleted_count = parse_cache_result.get("deleted_count", 0)
+                        if deleted_count > 0:
+                            cache_success_msg = f"Successfully cleaned {deleted_count} parse_cache entries"
+                            logger.info(cache_success_msg)
+                            if "history_messages" in pipeline_status:
+                                pipeline_status["history_messages"].append(
+                                    cache_success_msg
+                                )
+                        else:
+                            cache_empty_msg = "No parse_cache entries to clean"
+                            logger.info(cache_empty_msg)
+                            if "history_messages" in pipeline_status:
+                                pipeline_status["history_messages"].append(
+                                    cache_empty_msg
+                                )
+                except Exception as cache_error:
+                    cache_error_msg = f"Warning: Exception while cleaning parse_cache: {str(cache_error)}"
+                    logger.warning(cache_error_msg)
+                    if "history_messages" in pipeline_status:
+                        pipeline_status["history_messages"].append(cache_error_msg)
+
             # If all storage operations failed, return error status and don't proceed with file deletion
             if storage_success_count == 0 and storage_error_count > 0:
                 error_message = "All storage drop operations failed. Aborting document clearing process."
@@ -2025,7 +2563,7 @@ def create_document_routes(
         Get the status of all documents in the system.
 
         This endpoint retrieves the current status of all documents, grouped by their
-        processing status (PENDING, PROCESSING, PROCESSED, FAILED).
+        processing status (READY, HANDLING, PENDING, PROCESSING, PROCESSED, FAILED).
 
         Returns:
             DocsStatusesResponse: A response object containing a dictionary where keys are
@@ -2037,6 +2575,8 @@ def create_document_routes(
         """
         try:
             statuses = (
+                DocStatus.READY,
+                DocStatus.HANDLING,
                 DocStatus.PENDING,
                 DocStatus.PROCESSING,
                 DocStatus.PROCESSED,
@@ -2057,6 +2597,7 @@ def create_document_routes(
                         DocStatusResponse(
                             id=doc_id,
                             content_summary=doc_status.content_summary,
+                            multimodal_content=doc_status.multimodal_content,
                             content_length=doc_status.content_length,
                             status=doc_status.status,
                             created_at=format_datetime(doc_status.created_at),
@@ -2066,6 +2607,7 @@ def create_document_routes(
                             error_msg=doc_status.error_msg,
                             metadata=doc_status.metadata,
                             file_path=doc_status.file_path,
+                            scheme_name=doc_status.scheme_name,
                         )
                     )
             return response
@@ -2402,6 +2944,8 @@ def create_document_routes(
                         error_msg=doc.error_msg,
                         metadata=doc.metadata,
                         file_path=doc.file_path,
+                        scheme_name=doc.scheme_name,
+                        multimodal_content=doc.multimodal_content,
                     )
                 )
 

lightrag/api/webui/assets/_basePickBy-C1BlOoDW.js

@@ -0,0 +1 @@
+import{e as o,c as l,g as b,k as O,h as P,j as p,l as w,m as c,n as v,t as A,o as N}from"./_baseUniq-BZ_JDEKn.js";import{a_ as g,aw as _,a$ as $,b0 as E,b1 as F,b2 as x,b3 as M,b4 as y,b5 as B,b6 as T}from"./mermaid-vendor-CpW20EHd.js";var S=/\s/;function G(n){for(var r=n.length;r--&&S.test(n.charAt(r)););return r}var H=/^\s+/;function L(n){return n&&n.slice(0,G(n)+1).replace(H,"")}var m=NaN,R=/^[-+]0x[0-9a-f]+$/i,q=/^0b[01]+$/i,z=/^0o[0-7]+$/i,C=parseInt;function K(n){if(typeof n=="number")return n;if(o(n))return m;if(g(n)){var r=typeof n.valueOf=="function"?n.valueOf():n;n=g(r)?r+"":r}if(typeof n!="string")return n===0?n:+n;n=L(n);var t=q.test(n);return t||z.test(n)?C(n.slice(2),t?2:8):R.test(n)?m:+n}var W=1/0,X=17976931348623157e292;function Y(n){if(!n)return n===0?n:0;if(n=K(n),n===W||n===-1/0){var r=n<0?-1:1;return r*X}return n===n?n:0}function D(n){var r=Y(n),t=r%1;return r===r?t?r-t:r:0}function fn(n){var r=n==null?0:n.length;return r?l(n):[]}var I=Object.prototype,J=I.hasOwnProperty,dn=_(function(n,r){n=Object(n);var t=-1,e=r.length,i=e>2?r[2]:void 0;for(i&&$(r[0],r[1],i)&&(e=1);++t<e;)for(var f=r[t],a=E(f),s=-1,d=a.length;++s<d;){var u=a[s],h=n[u];(h===void 0||F(h,I[u])&&!J.call(n,u))&&(n[u]=f[u])}return n});function un(n){var r=n==null?0:n.length;return r?n[r-1]:void 0}function Q(n){return function(r,t,e){var i=Object(r);if(!x(r)){var f=b(t);r=O(r),t=function(s){return f(i[s],s,i)}}var a=n(r,t,e);return a>-1?i[f?r[a]:a]:void 0}}var U=Math.max;function Z(n,r,t){var e=n==null?0:n.length;if(!e)return-1;var i=t==null?0:D(t);return i<0&&(i=U(e+i,0)),P(n,b(r),i)}var hn=Q(Z);function V(n,r){var t=-1,e=x(n)?Array(n.length):[];return p(n,function(i,f,a){e[++t]=r(i,f,a)}),e}function gn(n,r){var t=M(n)?w:V;return t(n,b(r))}var j=Object.prototype,k=j.hasOwnProperty;function nn(n,r){return n!=null&&k.call(n,r)}function bn(n,r){return n!=null&&c(n,r,nn)}function rn(n,r){return n<r}function tn(n,r,t){for(var e=-1,i=n.length;++e<i;){var f=n[e],a=r(f);if(a!=null&&(s===void 0?a===a&&!o(a):t(a,s)))var s=a,d=f}return d}function mn(n){return n&&n.length?tn(n,y,rn):void 0}function an(n,r,t,e){if(!g(n))return n;r=v(r,n);for(var i=-1,f=r.length,a=f-1,s=n;s!=null&&++i<f;){var d=A(r[i]),u=t;if(d==="__proto__"||d==="constructor"||d==="prototype")return n;if(i!=a){var h=s[d];u=void 0,u===void 0&&(u=g(h)?h:B(r[i+1])?[]:{})}T(s,d,u),s=s[d]}return n}function on(n,r,t){for(var e=-1,i=r.length,f={};++e<i;){var a=r[e],s=N(n,a);t(s,a)&&an(f,v(a,n),s)}return f}export{rn as a,tn as b,V as c,on as d,mn as e,fn as f,hn as g,bn as h,dn as i,D as j,un as l,gn as m,Y as t};

lightrag/api/webui/assets/chunk-353BL4L5-0V1KVYyT.js

@@ -0,0 +1 @@
+import{_ as l}from"./mermaid-vendor-CpW20EHd.js";function m(e,c){var i,t,o;e.accDescr&&((i=c.setAccDescription)==null||i.call(c,e.accDescr)),e.accTitle&&((t=c.setAccTitle)==null||t.call(c,e.accTitle)),e.title&&((o=c.setDiagramTitle)==null||o.call(c,e.title))}l(m,"populateCommonDb");export{m as p};

lightrag/api/webui/assets/chunk-67H74DCK-bSLGaG_c.js

@@ -0,0 +1 @@
+import{_ as n,a2 as x,j as l}from"./mermaid-vendor-CpW20EHd.js";var c=n((a,t)=>{const e=a.append("rect");if(e.attr("x",t.x),e.attr("y",t.y),e.attr("fill",t.fill),e.attr("stroke",t.stroke),e.attr("width",t.width),e.attr("height",t.height),t.name&&e.attr("name",t.name),t.rx&&e.attr("rx",t.rx),t.ry&&e.attr("ry",t.ry),t.attrs!==void 0)for(const r in t.attrs)e.attr(r,t.attrs[r]);return t.class&&e.attr("class",t.class),e},"drawRect"),d=n((a,t)=>{const e={x:t.startx,y:t.starty,width:t.stopx-t.startx,height:t.stopy-t.starty,fill:t.fill,stroke:t.stroke,class:"rect"};c(a,e).lower()},"drawBackgroundRect"),g=n((a,t)=>{const e=t.text.replace(x," "),r=a.append("text");r.attr("x",t.x),r.attr("y",t.y),r.attr("class","legend"),r.style("text-anchor",t.anchor),t.class&&r.attr("class",t.class);const s=r.append("tspan");return s.attr("x",t.x+t.textMargin*2),s.text(e),r},"drawText"),h=n((a,t,e,r)=>{const s=a.append("image");s.attr("x",t),s.attr("y",e);const i=l.sanitizeUrl(r);s.attr("xlink:href",i)},"drawImage"),m=n((a,t,e,r)=>{const s=a.append("use");s.attr("x",t),s.attr("y",e);const i=l.sanitizeUrl(r);s.attr("xlink:href",`#${i}`)},"drawEmbeddedImage"),y=n(()=>({x:0,y:0,width:100,height:100,fill:"#EDF2AE",stroke:"#666",anchor:"start",rx:0,ry:0}),"getNoteRect"),p=n(()=>({x:0,y:0,width:100,height:100,"text-anchor":"start",style:"#666",textMargin:0,rx:0,ry:0,tspan:!0}),"getTextObj");export{d as a,p as b,m as c,c as d,h as e,g as f,y as g};

lightrag/api/webui/assets/chunk-AACKK3MU-DzHRGgvZ.js

@@ -0,0 +1 @@
+import{_ as s}from"./mermaid-vendor-CpW20EHd.js";var t,e=(t=class{constructor(i){this.init=i,this.records=this.init()}reset(){this.records=this.init()}},s(t,"ImperativeState"),t);export{e as I};

lightrag/api/webui/assets/chunk-BFAMUDN2-CHovHiOg.js

@@ -0,0 +1 @@
+import{_ as a,d as o}from"./mermaid-vendor-CpW20EHd.js";var d=a((t,e)=>{let n;return e==="sandbox"&&(n=o("#i"+t)),(e==="sandbox"?o(n.nodes()[0].contentDocument.body):o("body")).select(`[id="${t}"]`)},"getDiagramElement");export{d as g};

lightrag/api/webui/assets/chunk-E2GYISFI-DgamQYak.js

@@ -0,0 +1,15 @@
+import{_ as e}from"./mermaid-vendor-CpW20EHd.js";var l=e(()=>`
+  /* Font Awesome icon styling - consolidated */
+  .label-icon {
+    display: inline-block;
+    height: 1em;
+    overflow: visible;
+    vertical-align: -0.125em;
+  }
+  
+  .node .label-icon path {
+    fill: currentColor;
+    stroke: revert;
+    stroke-width: revert;
+  }
+`,"getIconStyles");export{l as g};

lightrag/api/webui/assets/chunk-SKB7J2MH-CqH3ZkpA.js

@@ -0,0 +1 @@
+import{_ as a,e as w,l as x}from"./mermaid-vendor-CpW20EHd.js";var d=a((e,t,i,o)=>{e.attr("class",i);const{width:r,height:h,x:n,y:c}=u(e,t);w(e,h,r,o);const s=l(n,c,r,h,t);e.attr("viewBox",s),x.debug(`viewBox configured: ${s} with padding: ${t}`)},"setupViewPortForSVG"),u=a((e,t)=>{var o;const i=((o=e.node())==null?void 0:o.getBBox())||{width:0,height:0,x:0,y:0};return{width:i.width+t*2,height:i.height+t*2,x:i.x,y:i.y}},"calculateDimensionsWithPadding"),l=a((e,t,i,o,r)=>`${e-r} ${t-r} ${i} ${o}`,"createViewBox");export{d as s};

lightrag/api/webui/assets/classDiagram-M3E45YP4-DbVgy_9-.js

@@ -0,0 +1 @@
+import{s as a,c as s,a as e,C as t}from"./chunk-SZ463SBG-CG1c8KxJ.js";import{_ as i}from"./mermaid-vendor-CpW20EHd.js";import"./chunk-E2GYISFI-DgamQYak.js";import"./chunk-BFAMUDN2-CHovHiOg.js";import"./chunk-SKB7J2MH-CqH3ZkpA.js";import"./feature-graph-xUsMo1iK.js";import"./react-vendor-DEwriMA6.js";import"./graph-vendor-B-X5JegA.js";import"./ui-vendor-CeCm8EER.js";import"./utils-vendor-BysuhMZA.js";var c={parser:e,get db(){return new t},renderer:s,styles:a,init:i(r=>{r.class||(r.class={}),r.class.arrowMarkerAbsolute=r.arrowMarkerAbsolute},"init")};export{c as diagram};

lightrag/api/webui/assets/classDiagram-v2-YAWTLIQI-DbVgy_9-.js

@@ -0,0 +1 @@
+import{s as a,c as s,a as e,C as t}from"./chunk-SZ463SBG-CG1c8KxJ.js";import{_ as i}from"./mermaid-vendor-CpW20EHd.js";import"./chunk-E2GYISFI-DgamQYak.js";import"./chunk-BFAMUDN2-CHovHiOg.js";import"./chunk-SKB7J2MH-CqH3ZkpA.js";import"./feature-graph-xUsMo1iK.js";import"./react-vendor-DEwriMA6.js";import"./graph-vendor-B-X5JegA.js";import"./ui-vendor-CeCm8EER.js";import"./utils-vendor-BysuhMZA.js";var c={parser:e,get db(){return new t},renderer:s,styles:a,init:i(r=>{r.class||(r.class={}),r.class.arrowMarkerAbsolute=r.arrowMarkerAbsolute},"init")};export{c as diagram};

lightrag/api/webui/assets/clone-CDvVvGlj.js

@@ -0,0 +1 @@
+import{b as r}from"./_baseUniq-BZ_JDEKn.js";var e=4;function a(o){return r(o,e)}export{a as c};

lightrag/api/webui/assets/diagram-5UYTHUR4-BB2kofV9.js

@@ -0,0 +1,24 @@
+import{p as y}from"./chunk-353BL4L5-0V1KVYyT.js";import{_ as l,s as B,g as S,t as z,q as F,a as P,b as E,F as v,K as W,e as T,z as D,G as _,H as A,l as w}from"./mermaid-vendor-CpW20EHd.js";import{p as N}from"./treemap-75Q7IDZK-CSah7hvo.js";import"./feature-graph-xUsMo1iK.js";import"./react-vendor-DEwriMA6.js";import"./graph-vendor-B-X5JegA.js";import"./ui-vendor-CeCm8EER.js";import"./utils-vendor-BysuhMZA.js";import"./_baseUniq-BZ_JDEKn.js";import"./_basePickBy-C1BlOoDW.js";import"./clone-CDvVvGlj.js";var x={packet:[]},m=structuredClone(x),L=A.packet,Y=l(()=>{const t=v({...L,..._().packet});return t.showBits&&(t.paddingY+=10),t},"getConfig"),G=l(()=>m.packet,"getPacket"),H=l(t=>{t.length>0&&m.packet.push(t)},"pushWord"),I=l(()=>{D(),m=structuredClone(x)},"clear"),u={pushWord:H,getPacket:G,getConfig:Y,clear:I,setAccTitle:E,getAccTitle:P,setDiagramTitle:F,getDiagramTitle:z,getAccDescription:S,setAccDescription:B},K=1e4,M=l(t=>{y(t,u);let e=-1,o=[],n=1;const{bitsPerRow:i}=u.getConfig();for(let{start:a,end:r,bits:c,label:f}of t.blocks){if(a!==void 0&&r!==void 0&&r<a)throw new Error(`Packet block ${a} - ${r} is invalid. End must be greater than start.`);if(a??(a=e+1),a!==e+1)throw new Error(`Packet block ${a} - ${r??a} is not contiguous. It should start from ${e+1}.`);if(c===0)throw new Error(`Packet block ${a} is invalid. Cannot have a zero bit field.`);for(r??(r=a+(c??1)-1),c??(c=r-a+1),e=r,w.debug(`Packet block ${a} - ${e} with label ${f}`);o.length<=i+1&&u.getPacket().length<K;){const[d,p]=O({start:a,end:r,bits:c,label:f},n,i);if(o.push(d),d.end+1===n*i&&(u.pushWord(o),o=[],n++),!p)break;({start:a,end:r,bits:c,label:f}=p)}}u.pushWord(o)},"populate"),O=l((t,e,o)=>{if(t.start===void 0)throw new Error("start should have been set during first phase");if(t.end===void 0)throw new Error("end should have been set during first phase");if(t.start>t.end)throw new Error(`Block start ${t.start} is greater than block end ${t.end}.`);if(t.end+1<=e*o)return[t,void 0];const n=e*o-1,i=e*o;return[{start:t.start,end:n,label:t.label,bits:n-t.start},{start:i,end:t.end,label:t.label,bits:t.end-i}]},"getNextFittingBlock"),q={parse:l(async t=>{const e=await N("packet",t);w.debug(e),M(e)},"parse")},R=l((t,e,o,n)=>{const i=n.db,a=i.getConfig(),{rowHeight:r,paddingY:c,bitWidth:f,bitsPerRow:d}=a,p=i.getPacket(),s=i.getDiagramTitle(),k=r+c,g=k*(p.length+1)-(s?0:r),b=f*d+2,h=W(e);h.attr("viewbox",`0 0 ${b} ${g}`),T(h,g,b,a.useMaxWidth);for(const[C,$]of p.entries())U(h,$,C,a);h.append("text").text(s).attr("x",b/2).attr("y",g-k/2).attr("dominant-baseline","middle").attr("text-anchor","middle").attr("class","packetTitle")},"draw"),U=l((t,e,o,{rowHeight:n,paddingX:i,paddingY:a,bitWidth:r,bitsPerRow:c,showBits:f})=>{const d=t.append("g"),p=o*(n+a)+a;for(const s of e){const k=s.start%c*r+1,g=(s.end-s.start+1)*r-i;if(d.append("rect").attr("x",k).attr("y",p).attr("width",g).attr("height",n).attr("class","packetBlock"),d.append("text").attr("x",k+g/2).attr("y",p+n/2).attr("class","packetLabel").attr("dominant-baseline","middle").attr("text-anchor","middle").text(s.label),!f)continue;const b=s.end===s.start,h=p-2;d.append("text").attr("x",k+(b?g/2:0)).attr("y",h).attr("class","packetByte start").attr("dominant-baseline","auto").attr("text-anchor",b?"middle":"start").text(s.start),b||d.append("text").attr("x",k+g).attr("y",h).attr("class","packetByte end").attr("dominant-baseline","auto").attr("text-anchor","end").text(s.end)}},"drawWord"),X={draw:R},j={byteFontSize:"10px",startByteColor:"black",endByteColor:"black",labelColor:"black",labelFontSize:"12px",titleColor:"black",titleFontSize:"14px",blockStrokeColor:"black",blockStrokeWidth:"1",blockFillColor:"#efefef"},J=l(({packet:t}={})=>{const e=v(j,t);return`
+	.packetByte {
+		font-size: ${e.byteFontSize};
+	}
+	.packetByte.start {
+		fill: ${e.startByteColor};
+	}
+	.packetByte.end {
+		fill: ${e.endByteColor};
+	}
+	.packetLabel {
+		fill: ${e.labelColor};
+		font-size: ${e.labelFontSize};
+	}
+	.packetTitle {
+		fill: ${e.titleColor};
+		font-size: ${e.titleFontSize};
+	}
+	.packetBlock {
+		stroke: ${e.blockStrokeColor};
+		stroke-width: ${e.blockStrokeWidth};
+		fill: ${e.blockFillColor};
+	}
+	`},"styles"),lt={parser:q,db:u,renderer:X,styles:J};export{lt as diagram};

lightrag/api/webui/assets/infoDiagram-LHK5PUON-BcENj9Cy.js

@@ -0,0 +1,2 @@
+import{_ as e,l as o,K as i,e as n,L as p}from"./mermaid-vendor-CpW20EHd.js";import{p as m}from"./treemap-75Q7IDZK-CSah7hvo.js";import"./feature-graph-xUsMo1iK.js";import"./react-vendor-DEwriMA6.js";import"./graph-vendor-B-X5JegA.js";import"./ui-vendor-CeCm8EER.js";import"./utils-vendor-BysuhMZA.js";import"./_baseUniq-BZ_JDEKn.js";import"./_basePickBy-C1BlOoDW.js";import"./clone-CDvVvGlj.js";var g={parse:e(async r=>{const a=await m("info",r);o.debug(a)},"parse")},v={version:p.version+""},d=e(()=>v.version,"getVersion"),c={getVersion:d},l=e((r,a,s)=>{o.debug(`rendering info diagram
+`+r);const t=i(a);n(t,100,400,!0),t.append("g").append("text").attr("x",100).attr("y",40).attr("class","version").attr("font-size",32).style("text-anchor","middle").text(`v${s}`)},"draw"),f={draw:l},L={parser:g,db:c,renderer:f};export{L as diagram};

lightrag/api/webui/assets/pieDiagram-NIOCPIFQ-DqOw1dnr.js

@@ -0,0 +1,30 @@
+import{p as N}from"./chunk-353BL4L5-0V1KVYyT.js";import{_ as i,g as B,s as U,a as q,b as H,t as K,q as V,l as C,c as Z,F as j,K as J,M as Q,N as z,O as X,e as Y,z as tt,P as et,H as at}from"./mermaid-vendor-CpW20EHd.js";import{p as rt}from"./treemap-75Q7IDZK-CSah7hvo.js";import"./feature-graph-xUsMo1iK.js";import"./react-vendor-DEwriMA6.js";import"./graph-vendor-B-X5JegA.js";import"./ui-vendor-CeCm8EER.js";import"./utils-vendor-BysuhMZA.js";import"./_baseUniq-BZ_JDEKn.js";import"./_basePickBy-C1BlOoDW.js";import"./clone-CDvVvGlj.js";var it=at.pie,D={sections:new Map,showData:!1},f=D.sections,w=D.showData,st=structuredClone(it),ot=i(()=>structuredClone(st),"getConfig"),nt=i(()=>{f=new Map,w=D.showData,tt()},"clear"),lt=i(({label:t,value:a})=>{f.has(t)||(f.set(t,a),C.debug(`added new section: ${t}, with value: ${a}`))},"addSection"),ct=i(()=>f,"getSections"),pt=i(t=>{w=t},"setShowData"),dt=i(()=>w,"getShowData"),F={getConfig:ot,clear:nt,setDiagramTitle:V,getDiagramTitle:K,setAccTitle:H,getAccTitle:q,setAccDescription:U,getAccDescription:B,addSection:lt,getSections:ct,setShowData:pt,getShowData:dt},gt=i((t,a)=>{N(t,a),a.setShowData(t.showData),t.sections.map(a.addSection)},"populateDb"),ut={parse:i(async t=>{const a=await rt("pie",t);C.debug(a),gt(a,F)},"parse")},mt=i(t=>`
+  .pieCircle{
+    stroke: ${t.pieStrokeColor};
+    stroke-width : ${t.pieStrokeWidth};
+    opacity : ${t.pieOpacity};
+  }
+  .pieOuterCircle{
+    stroke: ${t.pieOuterStrokeColor};
+    stroke-width: ${t.pieOuterStrokeWidth};
+    fill: none;
+  }
+  .pieTitleText {
+    text-anchor: middle;
+    font-size: ${t.pieTitleTextSize};
+    fill: ${t.pieTitleTextColor};
+    font-family: ${t.fontFamily};
+  }
+  .slice {
+    font-family: ${t.fontFamily};
+    fill: ${t.pieSectionTextColor};
+    font-size:${t.pieSectionTextSize};
+    // fill: white;
+  }
+  .legend text {
+    fill: ${t.pieLegendTextColor};
+    font-family: ${t.fontFamily};
+    font-size: ${t.pieLegendTextSize};
+  }
+`,"getStyles"),ft=mt,ht=i(t=>{const a=[...t.entries()].map(s=>({label:s[0],value:s[1]})).sort((s,n)=>n.value-s.value);return et().value(s=>s.value)(a)},"createPieArcs"),St=i((t,a,G,s)=>{C.debug(`rendering pie chart
+`+t);const n=s.db,y=Z(),T=j(n.getConfig(),y.pie),$=40,o=18,d=4,c=450,h=c,S=J(a),l=S.append("g");l.attr("transform","translate("+h/2+","+c/2+")");const{themeVariables:r}=y;let[A]=Q(r.pieOuterStrokeWidth);A??(A=2);const _=T.textPosition,g=Math.min(h,c)/2-$,M=z().innerRadius(0).outerRadius(g),O=z().innerRadius(g*_).outerRadius(g*_);l.append("circle").attr("cx",0).attr("cy",0).attr("r",g+A/2).attr("class","pieOuterCircle");const b=n.getSections(),v=ht(b),P=[r.pie1,r.pie2,r.pie3,r.pie4,r.pie5,r.pie6,r.pie7,r.pie8,r.pie9,r.pie10,r.pie11,r.pie12],p=X(P);l.selectAll("mySlices").data(v).enter().append("path").attr("d",M).attr("fill",e=>p(e.data.label)).attr("class","pieCircle");let E=0;b.forEach(e=>{E+=e}),l.selectAll("mySlices").data(v).enter().append("text").text(e=>(e.data.value/E*100).toFixed(0)+"%").attr("transform",e=>"translate("+O.centroid(e)+")").style("text-anchor","middle").attr("class","slice"),l.append("text").text(n.getDiagramTitle()).attr("x",0).attr("y",-400/2).attr("class","pieTitleText");const x=l.selectAll(".legend").data(p.domain()).enter().append("g").attr("class","legend").attr("transform",(e,u)=>{const m=o+d,R=m*p.domain().length/2,I=12*o,L=u*m-R;return"translate("+I+","+L+")"});x.append("rect").attr("width",o).attr("height",o).style("fill",p).style("stroke",p),x.data(v).append("text").attr("x",o+d).attr("y",o-d).text(e=>{const{label:u,value:m}=e.data;return n.getShowData()?`${u} [${m}]`:u});const W=Math.max(...x.selectAll("text").nodes().map(e=>(e==null?void 0:e.getBoundingClientRect().width)??0)),k=h+$+o+d+W;S.attr("viewBox",`0 0 ${k} ${c}`),Y(S,c,k,T.useMaxWidth)},"draw"),vt={draw:St},kt={parser:ut,db:F,renderer:vt,styles:ft};export{kt as diagram};

lightrag/api/webui/assets/stateDiagram-v2-5AN5P6BG-BvKEglP2.js

@@ -0,0 +1 @@
+import{s as r,b as e,a,S as i}from"./chunk-OW32GOEJ-DF1Nd_2F.js";import{_ as s}from"./mermaid-vendor-CpW20EHd.js";import"./chunk-BFAMUDN2-CHovHiOg.js";import"./chunk-SKB7J2MH-CqH3ZkpA.js";import"./feature-graph-xUsMo1iK.js";import"./react-vendor-DEwriMA6.js";import"./graph-vendor-B-X5JegA.js";import"./ui-vendor-CeCm8EER.js";import"./utils-vendor-BysuhMZA.js";var f={parser:a,get db(){return new i(2)},renderer:e,styles:r,init:s(t=>{t.state||(t.state={}),t.state.arrowMarkerAbsolute=t.arrowMarkerAbsolute},"init")};export{f as diagram};

lightrag/api/webui/index.html

@@ -8,20 +8,21 @@
     <link rel="icon" type="image/png" href="favicon.png" />
     <meta name="viewport" content="width=device-width, initial-scale=1.0" />
     <title>Lightrag</title>
-    <script type="module" crossorigin src="/webui/assets/index-Bjp1KQxU.js"></script>
+    <script type="module" crossorigin src="/webui/assets/index-oYZPo1xP.js"></script>
     <link rel="modulepreload" crossorigin href="/webui/assets/react-vendor-DEwriMA6.js">
     <link rel="modulepreload" crossorigin href="/webui/assets/ui-vendor-CeCm8EER.js">
     <link rel="modulepreload" crossorigin href="/webui/assets/graph-vendor-B-X5JegA.js">
     <link rel="modulepreload" crossorigin href="/webui/assets/utils-vendor-BysuhMZA.js">
-    <link rel="modulepreload" crossorigin href="/webui/assets/feature-graph-bahMe5Gt.js">
-    <link rel="modulepreload" crossorigin href="/webui/assets/feature-documents-cF4-Vta6.js">
-    <link rel="modulepreload" crossorigin href="/webui/assets/mermaid-vendor-BOzHoVUU.js">
-    <link rel="modulepreload" crossorigin href="/webui/assets/markdown-vendor-DmIvJdn7.js">
-    <link rel="modulepreload" crossorigin href="/webui/assets/feature-retrieval-CnKoygV2.js">
+    <link rel="modulepreload" crossorigin href="/webui/assets/feature-graph-xUsMo1iK.js">
+    <link rel="modulepreload" crossorigin href="/webui/assets/feature-documents-22OwnQq9.js">
+    <link rel="modulepreload" crossorigin href="/webui/assets/mermaid-vendor-CpW20EHd.js">
+    <link rel="modulepreload" crossorigin href="/webui/assets/markdown-vendor-C1oKx5V8.js">
+    <link rel="modulepreload" crossorigin href="/webui/assets/feature-retrieval-CeceOXFg.js">
     <link rel="stylesheet" crossorigin href="/webui/assets/feature-graph-BipNuM18.css">
-    <link rel="stylesheet" crossorigin href="/webui/assets/index-BvrNHAMA.css">
+    <link rel="stylesheet" crossorigin href="/webui/assets/index-CEupRNOQ.css">
   </head>
   <body>
     <div id="root"></div>
+
   </body>
 </html>

lightrag/base.py

@@ -675,6 +675,8 @@ class BaseGraphStorage(StorageNameSpace, ABC):
 class DocStatus(str, Enum):
     """Document processing status"""
 
+    READY = "ready"
+    HANDLING = "handling"
     PENDING = "pending"
     PROCESSING = "processing"
     PROCESSED = "processed"
@@ -707,6 +709,12 @@ class DocProcessingStatus:
     """Error message if failed"""
     metadata: dict[str, Any] = field(default_factory=dict)
     """Additional metadata"""
+    multimodal_content: list[dict[str, Any]] | None = None
+    """raganything: multimodal_content"""
+    multimodal_processed: bool | None = None
+    """raganything: multimodal_processed"""
+    scheme_name: str | None = None
+    """lightrag or raganything"""
 
 
 @dataclass

lightrag/lightrag.py

@@ -9,6 +9,7 @@ import warnings
 from dataclasses import asdict, dataclass, field
 from datetime import datetime, timezone
 from functools import partial
+from pathlib import Path
 from typing import (
     Any,
     AsyncIterator,
@@ -98,6 +99,7 @@ from .utils import (
 )
 from .types import KnowledgeGraph
 from dotenv import load_dotenv
+from .ragmanager import RAGManager
 
 # use the .env that is inside the current folder
 # allows to use different .env file for each lightrag instance
@@ -135,6 +137,9 @@ class LightRAG:
     doc_status_storage: str = field(default="JsonDocStatusStorage")
     """Storage type for tracking document processing statuses."""
 
+    input_dir: str = field(default_factory=lambda: os.getenv("INPUT_DIR", "./inputs"))
+    """Directory containing input documents"""
+
     # Workspace
     # ---
 
@@ -863,16 +868,22 @@ class LightRAG:
     def insert(
         self,
         input: str | list[str],
+        multimodal_content: list[dict[str, Any]]
+        | list[list[dict[str, Any]]]
+        | None = None,
         split_by_character: str | None = None,
         split_by_character_only: bool = False,
         ids: str | list[str] | None = None,
         file_paths: str | list[str] | None = None,
         track_id: str | None = None,
+        scheme_name: str | None = None,
     ) -> str:
         """Sync Insert documents with checkpoint support
 
         Args:
             input: Single document string or list of document strings
+            multimodal_content (list[dict[str, Any]] | list[list[dict[str, Any]]] | None, optional):
+                Multimodal content (images, tables, equations) associated with documents
             split_by_character: if split_by_character is not None, split the string by character, if chunk longer than
             chunk_token_size, it will be split again by token size.
             split_by_character_only: if split_by_character_only is True, split the string by character only, when
@@ -880,6 +891,7 @@ class LightRAG:
             ids: single string of the document ID or list of unique document IDs, if not provided, MD5 hash IDs will be generated
             file_paths: single string of the file path or list of file paths, used for citation
             track_id: tracking ID for monitoring processing status, if not provided, will be generated
+            scheme_name (str | None, optional): Scheme name for categorizing documents
 
         Returns:
             str: tracking ID for monitoring processing status
@@ -888,27 +900,35 @@ class LightRAG:
         return loop.run_until_complete(
             self.ainsert(
                 input,
+                multimodal_content,
                 split_by_character,
                 split_by_character_only,
                 ids,
                 file_paths,
                 track_id,
+                scheme_name,
             )
         )
 
     async def ainsert(
         self,
         input: str | list[str],
+        multimodal_content: list[dict[str, Any]]
+        | list[list[dict[str, Any]]]
+        | None = None,
         split_by_character: str | None = None,
         split_by_character_only: bool = False,
         ids: str | list[str] | None = None,
         file_paths: str | list[str] | None = None,
         track_id: str | None = None,
+        scheme_name: str | None = None,
     ) -> str:
         """Async Insert documents with checkpoint support
 
         Args:
             input: Single document string or list of document strings
+            multimodal_content (list[dict[str, Any]] | list[list[dict[str, Any]]] | None, optional):
+                Multimodal content (images, tables, equations) associated with documents
             split_by_character: if split_by_character is not None, split the string by character, if chunk longer than
             chunk_token_size, it will be split again by token size.
             split_by_character_only: if split_by_character_only is True, split the string by character only, when
@@ -916,6 +936,7 @@ class LightRAG:
             ids: list of unique document IDs, if not provided, MD5 hash IDs will be generated
             file_paths: list of file paths corresponding to each document, used for citation
             track_id: tracking ID for monitoring processing status, if not provided, will be generated
+            scheme_name (str | None, optional): Scheme name for categorizing documents
 
         Returns:
             str: tracking ID for monitoring processing status
@@ -924,13 +945,83 @@ class LightRAG:
         if track_id is None:
             track_id = generate_track_id("insert")
 
-        await self.apipeline_enqueue_documents(input, ids, file_paths, track_id)
+        paths_to_check = [file_paths] if isinstance(file_paths, str) else file_paths
+        base_input_dir = Path(self.input_dir)
+        if self.workspace:
+            current_input_dir = base_input_dir / self.workspace
+        else:
+            current_input_dir = base_input_dir
+
+        await self.apipeline_enqueue_documents(
+            input,
+            multimodal_content,
+            ids,
+            file_paths,
+            track_id,
+            scheme_name=scheme_name,
+        )
+
+        for file_path in paths_to_check:
+            current_file_path = current_input_dir / file_path
+            if current_file_path.exists():
+                self.move_file_to_enqueue(current_file_path)
+            else:
+                continue
+
         await self.apipeline_process_enqueue_documents(
             split_by_character, split_by_character_only
         )
 
         return track_id
 
+    def move_file_to_enqueue(self, file_path):
+        try:
+            enqueued_dir = file_path.parent / "__enqueued__"
+            enqueued_dir.mkdir(exist_ok=True)
+
+            # Generate unique filename to avoid conflicts
+            unique_filename = self.get_unique_filename_in_enqueued(
+                enqueued_dir, file_path.name
+            )
+            target_path = enqueued_dir / unique_filename
+
+            # Move the file
+            file_path.rename(target_path)
+            logger.debug(
+                f"Moved file to enqueued directory: {file_path.name} -> {unique_filename}"
+            )
+
+        except Exception as move_error:
+            logger.error(
+                f"Failed to move file {file_path.name} to __enqueued__ directory: {move_error}"
+            )
+            # Don't affect the main function's success status
+
+    def get_unique_filename_in_enqueued(
+        self, target_dir: Path, original_name: str
+    ) -> str:
+        from pathlib import Path
+        import time
+
+        original_path = Path(original_name)
+        base_name = original_path.stem
+        extension = original_path.suffix
+
+        # Try original name first
+        if not (target_dir / original_name).exists():
+            return original_name
+
+        # Try with numeric suffixes 001-999
+        for i in range(1, 1000):
+            suffix = f"{i:03d}"
+            new_name = f"{base_name}_{suffix}{extension}"
+            if not (target_dir / new_name).exists():
+                return new_name
+
+        # Fallback with timestamp if all 999 slots are taken
+        timestamp = int(time.time())
+        return f"{base_name}_{timestamp}{extension}"
+
     # TODO: deprecated, use insert instead
     def insert_custom_chunks(
         self,
@@ -1006,9 +1097,13 @@ class LightRAG:
     async def apipeline_enqueue_documents(
         self,
         input: str | list[str],
+        multimodal_content: list[dict[str, Any]]
+        | list[list[dict[str, Any]]]
+        | None = None,
         ids: list[str] | None = None,
         file_paths: str | list[str] | None = None,
         track_id: str | None = None,
+        scheme_name: str | None = None,
     ) -> str:
         """
         Pipeline for Processing Documents
@@ -1020,9 +1115,12 @@ class LightRAG:
 
         Args:
             input: Single document string or list of document strings
+            multimodal_content (list[dict[str, Any]] | list[list[dict[str, Any]]] | None, optional):
+                Multimodal content (images, tables, equations) associated with documents
             ids: list of unique document IDs, if not provided, MD5 hash IDs will be generated
             file_paths: list of file paths corresponding to each document, used for citation
             track_id: tracking ID for monitoring processing status, if not provided, will be generated with "enqueue" prefix
+            scheme_name (str | None, optional): Scheme name for categorizing documents
 
         Returns:
             str: tracking ID for monitoring processing status
@@ -1093,6 +1191,7 @@ class LightRAG:
             id_: {
                 "status": DocStatus.PENDING,
                 "content_summary": get_content_summary(content_data["content"]),
+                "multimodal_content": multimodal_content,
                 "content_length": len(content_data["content"]),
                 "created_at": datetime.now(timezone.utc).isoformat(),
                 "updated_at": datetime.now(timezone.utc).isoformat(),
@@ -1100,6 +1199,7 @@ class LightRAG:
                     "file_path"
                 ],  # Store file path in document status
                 "track_id": track_id,  # Store track_id in document status
+                "scheme_name": scheme_name,
             }
             for id_, content_data in contents.items()
         }
@@ -1130,6 +1230,12 @@ class LightRAG:
             if doc_id in new_docs
         }
 
+        new_docs_idList = [
+            f"doc-pre-{new_docs[doc_id]['file_path']}"
+            for doc_id in unique_new_doc_ids
+            if doc_id in new_docs
+        ]
+
         if not new_docs:
             logger.warning("No new unique documents were found.")
             return
@@ -1147,6 +1253,10 @@ class LightRAG:
         # Store document status (without content)
         await self.doc_status.upsert(new_docs)
         logger.debug(f"Stored {len(new_docs)} new unique documents")
+        await self.doc_status.index_done_callback()
+
+        await self.doc_status.delete(new_docs_idList)
+        logger.info(f"Deleted {new_docs_idList} Successful")
 
         return track_id
 
@@ -1322,6 +1432,7 @@ class LightRAG:
                     docs_to_reset[doc_id] = {
                         "status": DocStatus.PENDING,
                         "content_summary": status_doc.content_summary,
+                        "multimodal_content": status_doc.multimodal_content,
                         "content_length": status_doc.content_length,
                         "created_at": status_doc.created_at,
                         "updated_at": datetime.now(timezone.utc).isoformat(),
@@ -1330,11 +1441,15 @@ class LightRAG:
                         # Clear any error messages and processing metadata
                         "error_msg": "",
                         "metadata": {},
+                        "scheme_name": status_doc.scheme_name,
                     }
 
                     # Update the status in to_process_docs as well
                     status_doc.status = DocStatus.PENDING
                     reset_count += 1
+                    logger.info(
+                        f"Document {status_doc.file_path} from PROCESSING/FAILED to PENDING status"
+                    )
 
         # Update doc_status storage if there are documents to reset
         if docs_to_reset:
@@ -1557,6 +1672,7 @@ class LightRAG:
                                                 chunks.keys()
                                             ),  # Save chunks list
                                             "content_summary": status_doc.content_summary,
+                                            "multimodal_content": status_doc.multimodal_content,
                                             "content_length": status_doc.content_length,
                                             "created_at": status_doc.created_at,
                                             "updated_at": datetime.now(
@@ -1567,6 +1683,7 @@ class LightRAG:
                                             "metadata": {
                                                 "processing_start_time": processing_start_time
                                             },
+                                            "scheme_name": status_doc.scheme_name,
                                         }
                                     }
                                 )
@@ -1632,6 +1749,7 @@ class LightRAG:
                                         "status": DocStatus.FAILED,
                                         "error_msg": str(e),
                                         "content_summary": status_doc.content_summary,
+                                        "multimodal_content": status_doc.multimodal_content,
                                         "content_length": status_doc.content_length,
                                         "created_at": status_doc.created_at,
                                         "updated_at": datetime.now(
@@ -1643,6 +1761,7 @@ class LightRAG:
                                             "processing_start_time": processing_start_time,
                                             "processing_end_time": processing_end_time,
                                         },
+                                        "scheme_name": status_doc.scheme_name,
                                     }
                                 }
                             )
@@ -1675,10 +1794,11 @@ class LightRAG:
                                 await self.doc_status.upsert(
                                     {
                                         doc_id: {
-                                            "status": DocStatus.PROCESSED,
+                                            "status": DocStatus.PROCESSING,
                                             "chunks_count": len(chunks),
                                             "chunks_list": list(chunks.keys()),
                                             "content_summary": status_doc.content_summary,
+                                            "multimodal_content": status_doc.multimodal_content,
                                             "content_length": status_doc.content_length,
                                             "created_at": status_doc.created_at,
                                             "updated_at": datetime.now(
@@ -1690,6 +1810,32 @@ class LightRAG:
                                                 "processing_start_time": processing_start_time,
                                                 "processing_end_time": processing_end_time,
                                             },
+                                            "scheme_name": status_doc.scheme_name,
+                                        }
+                                    }
+                                )
+
+                                if (
+                                    status_doc.multimodal_content
+                                    and len(status_doc.multimodal_content) > 0
+                                ):
+                                    raganything_instance = RAGManager.get_rag()
+                                    await raganything_instance._process_multimodal_content(
+                                        status_doc.multimodal_content,
+                                        status_doc.file_path,
+                                        doc_id,
+                                        pipeline_status=pipeline_status,
+                                        pipeline_status_lock=pipeline_status_lock,
+                                    )
+
+                                current_doc_status = await self.doc_status.get_by_id(
+                                    doc_id
+                                )
+                                await self.doc_status.upsert(
+                                    {
+                                        doc_id: {
+                                            **current_doc_status,
+                                            "status": DocStatus.PROCESSED,
                                         }
                                     }
                                 )
@@ -1733,6 +1879,7 @@ class LightRAG:
                                             "status": DocStatus.FAILED,
                                             "error_msg": str(e),
                                             "content_summary": status_doc.content_summary,
+                                            "multimodal_content": status_doc.multimodal_content,
                                             "content_length": status_doc.content_length,
                                             "created_at": status_doc.created_at,
                                             "updated_at": datetime.now().isoformat(),
@@ -1742,6 +1889,7 @@ class LightRAG:
                                                 "processing_start_time": processing_start_time,
                                                 "processing_end_time": processing_end_time,
                                             },
+                                            "scheme_name": status_doc.scheme_name,
                                         }
                                     }
                                 )
@@ -2294,6 +2442,156 @@ class LightRAG:
         # Return the dictionary containing statuses only for the found document IDs
         return found_statuses
 
+    async def aclean_parse_cache_by_doc_ids(
+        self, doc_ids: str | list[str]
+    ) -> dict[str, Any]:
+        """Asynchronously clean parse_cache entries for specified document IDs
+
+        Args:
+            doc_ids: Single document ID string or list of document IDs
+
+        Returns:
+            Dictionary containing cleanup results:
+            - deleted_entries: List of deleted cache entries
+            - not_found: List of document IDs not found
+            - error: Error message (if operation fails)
+        """
+        import json
+        from pathlib import Path
+
+        # Normalize input to list
+        if isinstance(doc_ids, str):
+            doc_ids = [doc_ids]
+
+        result = {"deleted_entries": [], "not_found": [], "error": None}
+
+        try:
+            # Build parse_cache file path using class storage location variables
+            if self.workspace:
+                # If workspace exists, use workspace subdirectory
+                cache_file_path = (
+                    Path(self.working_dir)
+                    / self.workspace
+                    / "kv_store_parse_cache.json"
+                )
+            else:
+                # Default to using working_dir
+                cache_file_path = Path(self.working_dir) / "kv_store_parse_cache.json"
+
+            # Check if parse_cache file exists
+            if not cache_file_path.exists():
+                logger.warning(f"Parse cache file not found: {cache_file_path}")
+                result["not_found"] = doc_ids.copy()
+                return result
+
+            # Read current parse_cache data
+            with open(cache_file_path, "r", encoding="utf-8") as f:
+                cache_data = json.load(f)
+
+            # Find entries to delete and record found doc_ids
+            entries_to_delete = []
+            doc_ids_set = set(doc_ids)
+            found_doc_ids = set()
+
+            for cache_key, cache_entry in cache_data.items():
+                if (
+                    isinstance(cache_entry, dict)
+                    and cache_entry.get("doc_id") in doc_ids_set
+                ):
+                    entries_to_delete.append(cache_key)
+                    result["deleted_entries"].append(cache_key)
+                    found_doc_ids.add(cache_entry.get("doc_id"))
+
+            # Delete found entries
+            for cache_key in entries_to_delete:
+                del cache_data[cache_key]
+
+            # Find doc_ids not found
+            result["not_found"] = list(doc_ids_set - found_doc_ids)
+
+            # Write back updated cache data
+            with open(cache_file_path, "w", encoding="utf-8") as f:
+                json.dump(cache_data, f, indent=2, ensure_ascii=False)
+
+            logger.info(
+                f"Deleted {len(entries_to_delete)} parse_cache entries, document IDs: {doc_ids}"
+            )
+
+        except Exception as e:
+            error_msg = f"Error cleaning parse_cache: {str(e)}"
+            logger.error(error_msg)
+            result["error"] = error_msg
+
+        return result
+
+    def clean_parse_cache_by_doc_ids(self, doc_ids: str | list[str]) -> dict[str, Any]:
+        """Synchronously clean parse_cache entries for specified document IDs
+
+        Args:
+            doc_ids: Single document ID string or list of document IDs
+
+        Returns:
+            Dictionary containing cleanup results
+        """
+        loop = always_get_an_event_loop()
+        return loop.run_until_complete(self.aclean_parse_cache_by_doc_ids(doc_ids))
+
+    async def aclean_all_parse_cache(self) -> dict[str, Any]:
+        """Asynchronously clean all parse_cache entries
+
+        Returns:
+            Dictionary containing cleanup results:
+            - deleted_count: Number of deleted entries
+            - error: Error message (if operation fails)
+        """
+        import json
+        from pathlib import Path
+
+        result = {"deleted_count": 0, "error": None}
+
+        try:
+            # Build parse_cache file path
+            if self.workspace:
+                cache_file_path = (
+                    Path(self.working_dir)
+                    / self.workspace
+                    / "kv_store_parse_cache.json"
+                )
+            else:
+                cache_file_path = Path(self.working_dir) / "kv_store_parse_cache.json"
+
+            if not cache_file_path.exists():
+                logger.warning(f"Parse cache file not found: {cache_file_path}")
+                return result
+
+            # Read current cache to count entries
+            with open(cache_file_path, "r", encoding="utf-8") as f:
+                cache_data = json.load(f)
+
+            result["deleted_count"] = len(cache_data)
+
+            # Clear all entries
+            with open(cache_file_path, "w", encoding="utf-8") as f:
+                json.dump({}, f, indent=2)
+
+            logger.info(f"Cleared all {result['deleted_count']} parse_cache entries")
+
+        except Exception as e:
+            error_msg = f"Error clearing parse_cache: {str(e)}"
+            logger.error(error_msg)
+            result["error"] = error_msg
+
+        return result
+
+    def clean_all_parse_cache(self) -> dict[str, Any]:
+        """Synchronously clean all parse_cache entries
+
+        Returns:
+            Dictionary containing cleanup results
+        """
+        loop = always_get_an_event_loop()
+        return loop.run_until_complete(self.aclean_all_parse_cache())
+
     async def adelete_by_doc_id(self, doc_id: str) -> DeletionResult:
         """Delete a document and all its related data, including chunks, graph elements, and cached entries.
 

lightrag/operate.py

@@ -1936,7 +1936,29 @@ async def merge_nodes_and_edges(
     if full_entities_storage and full_relations_storage and doc_id:
         try:
             # Merge all entities: original entities + entities added during edge processing
-            final_entity_names = set()
+            existing_entites_data = None
+            existing_relations_data = None
+
+            try:
+                existing_entites_data = await full_entities_storage.get_by_id(doc_id)
+                existing_relations_data = await full_relations_storage.get_by_id(doc_id)
+            except Exception as e:
+                logger.debug(
+                    f"Could not retrieve existing entity/relation data for {doc_id}: {e}"
+                )
+
+            existing_entites_names = set()
+            if existing_entites_data and existing_entites_data.get("entity_names"):
+                existing_entites_names.update(existing_entites_data["entity_names"])
+
+            existing_relation_pairs = set()
+            if existing_relations_data and existing_relations_data.get(
+                "relation_pairs"
+            ):
+                for pair in existing_relations_data["relation_pairs"]:
+                    existing_relation_pairs.add(tuple(sorted(pair)))
+
+            final_entity_names = existing_entites_names.copy()
 
             # Add original processed entities
             for entity_data in processed_entities:
@@ -1949,7 +1971,7 @@ async def merge_nodes_and_edges(
                     final_entity_names.add(added_entity["entity_name"])
 
             # Collect all relation pairs
-            final_relation_pairs = set()
+            final_relation_pairs = existing_relation_pairs.copy()
             for edge_data in processed_edges:
                 if edge_data:
                     src_id = edge_data.get("src_id")
@@ -1959,6 +1981,12 @@ async def merge_nodes_and_edges(
                         final_relation_pairs.add(relation_pair)
 
             log_message = f"Phase 3: Updating final {len(final_entity_names)}({len(processed_entities)}+{len(all_added_entities)}) entities and  {len(final_relation_pairs)} relations from {doc_id}"
+            new_entities_count = len(final_entity_names) - len(existing_entites_names)
+            new_relation_count = len(final_relation_pairs) - len(
+                existing_relation_pairs
+            )
+
+            log_message = f"Phase 3: Merging storage - existing: {len(existing_entites_names)} entitites, {len(existing_relation_pairs)} relations; new: {new_entities_count} entities. {new_relation_count} relations; total: {len(final_entity_names)} entities, {len(final_relation_pairs)} relations"
             logger.info(log_message)
             async with pipeline_status_lock:
                 pipeline_status["latest_message"] = log_message

lightrag/prompt.py

@@ -10,9 +10,11 @@ PROMPTS["DEFAULT_COMPLETION_DELIMITER"] = "<|COMPLETE|>"
 
 PROMPTS["DEFAULT_USER_PROMPT"] = "n/a"
 
+
 PROMPTS["entity_extraction_system_prompt"] = """---Role---
 You are a Knowledge Graph Specialist responsible for extracting entities and relationships from the input text.
 
+
 ---Instructions---
 1.  **Entity Extraction & Output:**
     *   **Identification:** Identify clearly defined and meaningful entities in the input text.
@@ -58,6 +60,7 @@ You are a Knowledge Graph Specialist responsible for extracting entities and rel
 
 8.  **Completion Signal:** Output the literal string `{completion_delimiter}` only after all entities and relationships, following all criteria, have been completely extracted and outputted.
 
+
 ---Examples---
 {examples}
 

lightrag/ragmanager.py

@@ -0,0 +1,18 @@
+class RAGManager:
+    _instance = None
+    _rag = None
+
+    def __new__(cls):
+        if cls._instance is None:
+            cls._instance = super().__new__(cls)
+        return cls._instance
+
+    @classmethod
+    def set_rag(cls, rag_instance):
+        cls._rag = rag_instance
+
+    @classmethod
+    def get_rag(cls):
+        if cls._rag is None:
+            raise ValueError("RAG instance not initialized!")
+        return cls._rag

lightrag_webui/src/App.tsx

@@ -15,6 +15,7 @@ import GraphViewer from '@/features/GraphViewer'
 import DocumentManager from '@/features/DocumentManager'
 import RetrievalTesting from '@/features/RetrievalTesting'
 import ApiSite from '@/features/ApiSite'
+import { SchemeProvider } from '@/contexts/SchemeContext';
 
 import { Tabs, TabsContent } from '@/components/ui/Tabs'
 
@@ -204,9 +205,11 @@ function App() {
             >
               <SiteHeader />
               <div className="relative grow">
-                <TabsContent value="documents" className="absolute top-0 right-0 bottom-0 left-0 overflow-auto">
-                  <DocumentManager />
-                </TabsContent>
+                <SchemeProvider>
+                  <TabsContent value="documents" className="absolute top-0 right-0 bottom-0 left-0 overflow-auto">
+                    <DocumentManager />
+                  </TabsContent>
+                </SchemeProvider>
                 <TabsContent value="knowledge-graph" className="absolute top-0 right-0 bottom-0 left-0 overflow-hidden">
                   <GraphViewer />
                 </TabsContent>

lightrag_webui/src/api/lightrag.ts

@@ -161,7 +161,7 @@ export type DeleteDocResponse = {
   doc_id: string
 }
 
-export type DocStatus = 'pending' | 'processing' | 'processed' | 'failed'
+export type DocStatus = 'pending' | 'processing' | 'processed' | 'ready' | 'handling' | 'failed'
 
 export type DocStatusResponse = {
   id: string
@@ -175,6 +175,7 @@ export type DocStatusResponse = {
   error_msg?: string
   metadata?: Record<string, any>
   file_path: string
+  scheme_name: string
 }
 
 export type DocsStatusesResponse = {
@@ -252,6 +253,24 @@ export type LoginResponse = {
   webui_description?: string
 }
 
+export type Scheme = {
+  id: number;
+  name: string;
+  config: {
+    framework: 'lightrag' | 'raganything';
+    extractor?: 'mineru' | 'docling' | undefined; // Optional extractor field
+    modelSource?: 'huggingface' | 'modelscope' | 'local' | undefined; // Optional model source field
+  };
+};
+
+type AddSchemeParams = Omit<Scheme, 'id'>;
+
+export type SchemesResponse = {
+  status: string;
+  message: string;
+  data: Scheme[];
+};
+
 export const InvalidApiKeyError = 'Invalid API Key'
 export const RequireApiKeError = 'API Key required'
 
@@ -305,6 +324,32 @@ axiosInstance.interceptors.response.use(
 )
 
 // API methods
+export const getSchemes = async (): Promise<SchemesResponse> => {
+  const response = await axiosInstance.get('/documents/schemes');
+  return response.data;
+};
+
+export const saveSchemes = async (schemes: Scheme[]): Promise<{ message: string }> => {
+  const response = await axiosInstance.post('/documents/schemes', schemes);
+  return response.data;
+};
+
+export const addScheme = async (scheme: AddSchemeParams): Promise<Scheme> => {
+  try {
+    const response = await axiosInstance.post('/documents/schemes/add', scheme);
+    // 验证响应数据是否符合 Scheme 类型（可选，取决于 axios 的配置）
+    return response.data;
+  } catch (error) {
+    console.error('Failed to add scheme:', error);
+    throw error; // 重新抛出错误，由调用方处理
+  }
+};
+
+export const deleteScheme = async (schemeId: number): Promise<{ message: string }> => {
+  const response = await axiosInstance.delete(`/documents/schemes/${schemeId}`);
+  return response.data;
+};
+
 export const queryGraphs = async (
   label: string,
   maxDepth: number,
@@ -338,8 +383,10 @@ export const getDocuments = async (): Promise<DocsStatusesResponse> => {
   return response.data
 }
 
-export const scanNewDocuments = async (): Promise<ScanResponse> => {
-  const response = await axiosInstance.post('/documents/scan')
+export const scanNewDocuments = async (schemeConfig: any): Promise<ScanResponse> => {
+  const response = await axiosInstance.post('/documents/scan', {
+    schemeConfig
+  })
   return response.data
 }
 
@@ -550,10 +597,12 @@ export const insertTexts = async (texts: string[]): Promise<DocActionResponse> =
 
 export const uploadDocument = async (
   file: File,
+  schemeId: number | '',
   onUploadProgress?: (percentCompleted: number) => void
 ): Promise<DocActionResponse> => {
   const formData = new FormData()
   formData.append('file', file)
+  formData.append('schemeId', schemeId.toString())
 
   const response = await axiosInstance.post('/documents/upload', formData, {
     headers: {
@@ -573,11 +622,12 @@ export const uploadDocument = async (
 
 export const batchUploadDocuments = async (
   files: File[],
+  schemeId: number | '',
   onUploadProgress?: (fileName: string, percentCompleted: number) => void
 ): Promise<DocActionResponse[]> => {
   return await Promise.all(
     files.map(async (file) => {
-      return await uploadDocument(file, (percentCompleted) => {
+      return await uploadDocument(file, schemeId, (percentCompleted) => {
         onUploadProgress?.(file.name, percentCompleted)
       })
     })

lightrag_webui/src/components/documents/SchemeManager/SchemeManager.css

@@ -0,0 +1,165 @@
+.scheme-manager-container {
+  font-family: Arial, sans-serif;
+  max-width: 1000px;
+  margin: 0 auto;
+}
+
+.toggle-button {
+  padding: 8px 16px;
+  background-color: #4CAF50;
+  color: white;
+  border: none;
+  border-radius: 4px;
+  cursor: pointer;
+  font-size: 14px;
+  margin-bottom: 20px;
+}
+
+.toggle-button:hover {
+  background-color: #45a049;
+}
+
+.scheme-modal {
+  border: 1px solid #ddd;
+  border-radius: 8px;
+  box-shadow: 0 4px 8px rgba(0, 0, 0, 0.1);
+  overflow: hidden;
+}
+
+.modal-header {
+  display: flex;
+  justify-content: space-between;
+  align-items: center;
+  padding: 16px;
+  background-color: #f5f5f5;
+  border-bottom: 1px solid #ddd;
+}
+
+.modal-header h2 {
+  margin: 0;
+  font-size: 18px;
+}
+
+.close-button {
+  background: none;
+  border: none;
+  font-size: 20px;
+  cursor: pointer;
+  color: #999;
+}
+
+.close-button:hover {
+  color: #333;
+}
+
+.modal-content {
+  display: flex;
+  height: 500px;
+}
+
+.left-panel {
+  flex: 0 0 300px;
+  padding: 16px;
+  border-right: 1px solid #ddd;
+  overflow-y: auto;
+}
+
+.right-panel {
+  flex: 1;
+  padding: 16px;
+  overflow-y: auto;
+}
+
+.add-scheme-form {
+  display: flex;
+  margin-bottom: 16px;
+  gap: 8px;
+}
+
+.scheme-name-input {
+  flex: 1;
+  padding: 8px;
+  border: 1px solid #ddd;
+  border-radius: 4px;
+}
+
+.add-button {
+  padding: 8px 12px;
+  background-color: #2196F3;
+  color: white;
+  border: none;
+  border-radius: 4px;
+  cursor: pointer;
+}
+
+.add-button:hover {
+  background-color: #0b7dda;
+}
+
+.scheme-list ul {
+  list-style: none;
+  padding: 0;
+  margin: 0;
+}
+
+.scheme-item {
+  display: flex;
+  justify-content: space-between;
+  align-items: center;
+  padding: 12px;
+  margin-bottom: 8px;
+  background-color: #f9f9f9;
+  border-radius: 4px;
+  cursor: pointer;
+  transition: background-color 0.2s;
+}
+
+.scheme-item:hover {
+  background-color: #f0f0f0;
+}
+
+.scheme-item.active {
+  background-color: #e3f2fd;
+  border-left: 3px solid #2196F3;
+}
+
+.delete-button {
+  background: none;
+  border: none;
+  color: #f44336;
+  cursor: pointer;
+  font-size: 16px;
+  padding: 0 4px;
+}
+
+.delete-button:hover {
+  color: #d32f2f;
+}
+
+.empty-message, .select-message {
+  color: #666;
+  text-align: center;
+  padding: 20px;
+}
+
+.config-form {
+  margin-top: 16px;
+}
+
+.form-group {
+  margin-bottom: 16px;
+}
+
+.form-group label {
+  display: block;
+  margin-bottom: 8px;
+  font-weight: bold;
+}
+
+.form-group input,
+.form-group select {
+  width: 100%;
+  padding: 8px;
+  border: 1px solid #ddd;
+  border-radius: 4px;
+}

lightrag_webui/src/components/documents/SchemeManager/SchemeManager.tsx

@@ -0,0 +1,299 @@
+import React, { useRef,useState, useEffect } from "react";
+import {
+  Dialog,
+  DialogContent,
+  DialogDescription,
+  DialogHeader,
+  DialogTitle,
+  DialogTrigger
+} from '@/components/ui/Dialog';
+import Button from "@/components/ui/Button";
+import { PlusIcon } from "lucide-react";
+import { Alert, AlertDescription } from "@/components/ui/Alert";
+import { AlertCircle } from "lucide-react";
+import {
+  getSchemes,
+  saveSchemes,
+  addScheme,
+  deleteScheme,
+  Scheme
+} from '@/api/lightrag';
+import { useScheme } from '@/contexts/SchemeContext';
+import { useTranslation } from 'react-i18next';
+
+interface SchemeConfig {
+  framework: 'lightrag' | 'raganything';
+  extractor?: 'mineru' | 'docling';
+  modelSource?: 'huggingface' | 'modelscope' | 'local';
+}
+
+const SchemeManagerDialog = () => {
+  const { t } = useTranslation();
+  const [open, setOpen] = useState(false);
+  const [schemes, setSchemes] = useState<Scheme[]>([]);
+  const [newSchemeName, setNewSchemeName] = useState("");
+  const [error, _setError] = useState<string | undefined>();
+  const [isLoading, setIsLoading] = useState(true);
+  const setError = (err?: string) => _setError(err);
+  const scrollRef = useRef<HTMLDivElement>(null);
+  const { selectedScheme, setSelectedScheme } = useScheme();
+
+  // 加载方案数据
+  useEffect(() => {
+    const loadSchemes = async () => {
+      try {
+        setIsLoading(true);
+        const response = await getSchemes();
+        setSchemes(response.data);
+        localStorage.getItem('selectedSchemeId') && setSelectedScheme(response.data.find(s => s.id === Number(localStorage.getItem('selectedSchemeId'))) || undefined);
+      } catch (err) {
+        setError(err instanceof Error ? err.message : t('schemeManager.errors.loadFailed'));
+      } finally {
+        setIsLoading(false);
+      }
+    };
+    loadSchemes();
+  }, []);
+
+  // 自动滚动到底部
+  useEffect(() => {
+    handleSelectScheme(selectedScheme?.id!);
+    if (!scrollRef.current) return;
+    const scrollToBottom = () => {
+      const container = scrollRef.current!;
+      const { scrollHeight } = container;
+      container.scrollTop = scrollHeight;
+    };
+    setTimeout(scrollToBottom, 0);
+  }, [schemes]);
+
+  // 检查方案名是否已存在
+  const isNameTaken = (name: string): boolean => {
+    return schemes.some(scheme => scheme.name.trim() === name.trim());
+  };
+
+  // 选中方案（更新 Context）
+  const handleSelectScheme = (schemeId: number) => {
+    const scheme = schemes.find((s) => s.id === schemeId);
+    if (scheme) {
+      setSelectedScheme(scheme);
+      localStorage.setItem('selectedSchemeId', String(scheme.id));
+    }
+  };
+
+  // 添加新方案
+  const handleAddScheme = async () => {
+    const trimmedName = newSchemeName.trim();
+    if (!trimmedName) {
+      setError(t('schemeManager.errors.nameEmpty'));
+      return;
+    }
+    if (isNameTaken(trimmedName)) {
+      setError(t('schemeManager.errors.nameExists'));
+      return;
+    }
+
+    try {
+      const newScheme = await addScheme({
+        name: trimmedName,
+        config: { framework: 'lightrag', extractor: undefined, modelSource: undefined },
+      });
+
+      // 更新方案列表
+      setSchemes((prevSchemes) => [...prevSchemes, newScheme]);
+
+      // 选中新方案
+      setSelectedScheme(newScheme);
+
+      // 清空输入和错误
+      setNewSchemeName("");
+      setError(undefined);
+    } catch (err) {
+      setError(err instanceof Error ? err.message : t('schemeManager.errors.addFailed'));
+    }
+  };
+
+  // 删除方案
+  const handleDeleteScheme = async (schemeId: number) => {
+    try {
+      await deleteScheme(schemeId);
+      setSchemes(schemes.filter(s => s.id !== schemeId));
+      if (selectedScheme?.id === schemeId) {
+        setSelectedScheme(undefined); // 清除 Context 中的选中状态
+      }
+    } catch (err) {
+      setError(err instanceof Error ? err.message : t('schemeManager.errors.deleteFailed'));
+    }
+  };
+
+  // 更新方案配置
+  const handleConfigChange = async (updates: Partial<SchemeConfig>) => {
+    if (!selectedScheme) return;
+
+    const updatedScheme = {
+      ...selectedScheme,
+      config: {
+        ...selectedScheme.config,
+        ...updates,
+        framework: updates.framework ?? selectedScheme.config?.framework ?? 'lightrag',
+        extractor: updates.extractor || selectedScheme.config?.extractor || (updates.framework === 'raganything' ? 'mineru' : undefined),
+        modelSource: updates.modelSource || selectedScheme.config?.modelSource || (updates.extractor === 'mineru' ? 'huggingface' : undefined),
+      },
+    };
+
+    setSchemes(schemes.map(s => s.id === selectedScheme.id ? updatedScheme : s));
+    await saveSchemes([updatedScheme]);
+  };
+
+  if (isLoading) {
+    return (
+      <div className="flex justify-center items-center h-screen">
+        <div className="animate-spin rounded-full h-8 w-8 border-b-2 border-blue-500" />
+      </div>
+    );
+  }
+
+  return (
+    <Dialog open={open} onOpenChange={setOpen}>
+      <DialogTrigger asChild>
+        <Button variant="default" side="bottom" size="sm">
+          <PlusIcon className="size-4" />
+          {t('schemeManager.button')}
+        </Button>
+      </DialogTrigger>
+
+      <DialogContent className="sm:max-w-[800px]">
+        <DialogHeader>
+          <DialogTitle>{t('schemeManager.title')}</DialogTitle>
+          <DialogDescription>{t('schemeManager.description')}</DialogDescription>
+        </DialogHeader>
+
+        <div className="flex h-[500px] gap-4">
+          {/* 左侧：方案列表 */}
+          <div className="w-1/3 rounded-lg border p-4 bg-gray-50 flex flex-col dark:bg-zinc-800 dark:text-zinc-100">
+            <h3 className="mb-4 font-semibold">{t('schemeManager.schemeList')}</h3>
+
+            {/* 创建新方案输入框 */}
+            <div className="flex gap-2 mb-4">
+              <input
+                type="text"
+                value={newSchemeName}
+                onChange={(e) => {
+                  if (e.target.value.length > 50) return;
+                  setNewSchemeName(e.target.value);
+                  setError(undefined);
+                }}
+                onKeyPress={(e) => e.key === 'Enter' && handleAddScheme()}
+                placeholder={t('schemeManager.inputPlaceholder')}
+                className="w-full px-3 py-1.5 border rounded-md focus:outline-none focus:ring-2 focus:ring-blue-500"
+              />
+              <Button onClick={handleAddScheme} size="sm">
+                <PlusIcon className="size-4" />
+              </Button>
+            </div>
+
+            {/* 错误提示 */}
+            {error && (
+              <Alert variant="destructive" className="mb-4">
+                <AlertCircle className="size-4" />
+                <AlertDescription>{error}</AlertDescription>
+              </Alert>
+            )}
+
+            {/* 方案列表 */}
+            <div ref={scrollRef} className="flex-1 overflow-y-auto border rounded-md p-1 dark:bg-zinc-800 dark:text-zinc-100">
+              {schemes.length === 0 ? (
+                <p className="text-gray-500 text-center py-4">{t('schemeManager.emptySchemes')}</p>
+              ) : (
+                <div className="space-y-2">
+                  {schemes.map((scheme) => (
+                    <div
+                      key={scheme.id}
+                      className={`flex items-center justify-between p-2 rounded-md cursor-pointer transition-colors truncate ${
+                        selectedScheme?.id === scheme.id
+                          ? "bg-blue-100 text-blue-700"
+                          : "hover:bg-gray-100"
+                      }`}
+                      onClick={() => handleSelectScheme(scheme.id)}
+                    >
+                      <div className="flex-1 truncate mr-2" title={scheme.name}>
+                        {scheme.name}
+                      </div>
+                      <button
+                        onClick={(e) => {
+                          e.stopPropagation();
+                          handleDeleteScheme(scheme.id);
+                        }}
+                        className="ml-2 text-red-500 hover:text-red-700 hover:bg-red-100 rounded-full p-1 transition-colors"
+                        title={t('schemeManager.deleteTooltip')}
+                      >
+                        −
+                      </button>
+                    </div>
+                  ))}
+                </div>
+              )}
+            </div>
+          </div>
+
+          {/* 右侧：方案配置 */}
+          <div className="flex-1 rounded-lg border p-4 bg-gray-50 dark:bg-zinc-800 dark:text-zinc-100">
+            <h3 className="mb-4 font-semibold">{t('schemeManager.schemeConfig')}</h3>
+
+            {selectedScheme ? (
+              <div className="space-y-4">
+                <div>
+                  <label className="block text-sm mb-1">{t('schemeManager.processingFramework')}</label>
+                  <select
+                    value={selectedScheme.config?.framework || "lightrag"}
+                    onChange={(e) => handleConfigChange({ framework: e.target.value as 'lightrag' | 'raganything' })}
+                    className="w-full px-3 py-1.5 border rounded-md focus:outline-none dark:bg-zinc-800 dark:text-zinc-100"
+                  >
+                    <option value="lightrag">LightRAG</option>
+                    <option value="raganything">RAGAnything</option>
+                  </select>
+                </div>
+
+                {selectedScheme.config?.framework === "raganything" && (
+                  <div>
+                    <label className="block text-sm mb-1">{t('schemeManager.extractionTool')}</label>
+                    <select
+                      value={selectedScheme.config?.extractor || "mineru"}
+                      onChange={(e) => handleConfigChange({ extractor: e.target.value as 'mineru' | 'docling' })}
+                      className="w-full px-3 py-1.5 border rounded-md focus:outline-none dark:bg-zinc-800 dark:text-zinc-100"
+                    >
+                      <option value="mineru">Mineru</option>
+                      <option value="docling">DocLing</option>
+                    </select>
+                  </div>
+                )}
+
+                {selectedScheme.config?.extractor === "mineru" && (
+                  <div>
+                    <label className="block text-sm mb-1">{t('schemeManager.modelSource')}</label>
+                    <select
+                      value={selectedScheme.config?.modelSource || "huggingface"}
+                      onChange={(e) => handleConfigChange({ modelSource: e.target.value as 'huggingface' | 'modelscope' | 'local' })}
+                      className="w-full px-3 py-1.5 border rounded-md focus:outline-none dark:bg-zinc-800 dark:text-zinc-100"
+                    >
+                      <option value="huggingface">huggingface</option>
+                      <option value="modelscope">modelscope</option>
+                      <option value="local">local</option>
+                    </select>
+                  </div>
+                )}
+              </div>
+            ) : (
+              <div className="flex flex-col items-center justify-center h-[70%] text-gray-500">
+                <AlertCircle className="size-12 mb-4 opacity-50" />
+                <p>{t('schemeManager.selectSchemePrompt')}</p>
+              </div>
+            )}
+          </div>
+        </div>
+      </DialogContent>
+    </Dialog>
+  );
+};
+
+export default SchemeManagerDialog;

lightrag_webui/src/components/documents/UploadDocumentsDialog.tsx

@@ -16,6 +16,7 @@ import { uploadDocument } from '@/api/lightrag'
 
 import { UploadIcon } from 'lucide-react'
 import { useTranslation } from 'react-i18next'
+import { useScheme } from '@/contexts/SchemeContext';
 
 interface UploadDocumentsDialogProps {
   onDocumentsUploaded?: () => Promise<void>
@@ -27,6 +28,7 @@ export default function UploadDocumentsDialog({ onDocumentsUploaded }: UploadDoc
   const [isUploading, setIsUploading] = useState(false)
   const [progresses, setProgresses] = useState<Record<string, number>>({})
   const [fileErrors, setFileErrors] = useState<Record<string, string>>({})
+  const { selectedScheme } = useScheme();
 
   const handleRejectedFiles = useCallback(
     (rejectedFiles: FileRejection[]) => {
@@ -58,6 +60,11 @@ export default function UploadDocumentsDialog({ onDocumentsUploaded }: UploadDoc
 
   const handleDocumentsUpload = useCallback(
     async (filesToUpload: File[]) => {
+      if (!selectedScheme) {
+        toast.error(t('schemeManager.upload.noSchemeSelected'));
+        return;
+      }
+
       setIsUploading(true)
       let hasSuccessfulUpload = false
 
@@ -95,7 +102,7 @@ export default function UploadDocumentsDialog({ onDocumentsUploaded }: UploadDoc
               [file.name]: 0
             }))
 
-            const result = await uploadDocument(file, (percentCompleted: number) => {
+            const result = await uploadDocument(file, selectedScheme?.id, (percentCompleted: number) => {
               console.debug(t('documentPanel.uploadDocuments.single.uploading', { name: file.name, percent: percentCompleted }))
               setProgresses((pre) => ({
                 ...pre,
@@ -175,7 +182,7 @@ export default function UploadDocumentsDialog({ onDocumentsUploaded }: UploadDoc
         setIsUploading(false)
       }
     },
-    [setIsUploading, setProgresses, setFileErrors, t, onDocumentsUploaded]
+    [setIsUploading, setProgresses, setFileErrors, t, onDocumentsUploaded, selectedScheme]
   )
 
   return (
@@ -201,7 +208,11 @@ export default function UploadDocumentsDialog({ onDocumentsUploaded }: UploadDoc
         <DialogHeader>
           <DialogTitle>{t('documentPanel.uploadDocuments.title')}</DialogTitle>
           <DialogDescription>
-            {t('documentPanel.uploadDocuments.description')}
+            {selectedScheme ? (
+              <>{t('schemeManager.upload.currentScheme')}<strong>{selectedScheme.name}</strong></>
+            ) : (
+              t('schemeManager.upload.noSchemeMessage')
+            )}
           </DialogDescription>
         </DialogHeader>
         <FileUploader

lightrag_webui/src/contexts/SchemeContext.tsx

@@ -0,0 +1,28 @@
+// contexts/SchemeContext.tsx
+import React, { createContext, useContext, useState } from 'react';
+import { Scheme } from '@/api/lightrag';
+
+interface SchemeContextType {
+  selectedScheme: Scheme | undefined;
+  setSelectedScheme: (scheme: Scheme | undefined) => void;
+}
+
+const SchemeContext = createContext<SchemeContextType | undefined>(undefined);
+
+export const SchemeProvider: React.FC<{ children: React.ReactNode }> = ({ children }) => {
+  const [selectedScheme, setSelectedScheme] = useState<Scheme | undefined>();
+
+  return (
+    <SchemeContext.Provider value={{ selectedScheme, setSelectedScheme }}>
+      {children}
+    </SchemeContext.Provider>
+  );
+};
+
+export const useScheme = () => {
+  const context = useContext(SchemeContext);
+  if (context === undefined) {
+    throw new Error('useScheme must be used within a SchemeProvider');
+  }
+  return context;
+};

lightrag_webui/src/features/DocumentManager.tsx

@@ -18,6 +18,8 @@ import UploadDocumentsDialog from '@/components/documents/UploadDocumentsDialog'
 import ClearDocumentsDialog from '@/components/documents/ClearDocumentsDialog'
 import DeleteDocumentsDialog from '@/components/documents/DeleteDocumentsDialog'
 import PaginationControls from '@/components/ui/PaginationControls'
+import { SchemeProvider } from '@/contexts/SchemeContext';
+import SchemeManager from '@/components/documents/SchemeManager/SchemeManager'
 
 import {
   scanNewDocuments,
@@ -35,6 +37,8 @@ import { useBackendState } from '@/stores/state'
 import { RefreshCwIcon, ActivityIcon, ArrowUpIcon, ArrowDownIcon, RotateCcwIcon, CheckSquareIcon, XIcon, AlertTriangle, Info } from 'lucide-react'
 import PipelineStatusDialog from '@/components/documents/PipelineStatusDialog'
 
+import { useScheme } from '@/contexts/SchemeContext';
+
 type StatusFilter = DocStatus | 'all';
 
 
@@ -169,6 +173,8 @@ type SortField = 'created_at' | 'updated_at' | 'id' | 'file_path';
 type SortDirection = 'asc' | 'desc';
 
 export default function DocumentManager() {
+  const { selectedScheme } = useScheme();
+
   // Track component mount status
   const isMountedRef = useRef(true);
 
@@ -230,6 +236,8 @@ export default function DocumentManager() {
     processing: 1,
     pending: 1,
     failed: 1,
+    ready: 1,
+    handling: 1
   });
 
   // State for document selection
@@ -296,6 +304,8 @@ export default function DocumentManager() {
       processing: 1,
       pending: 1,
       failed: 1,
+      ready: 1,
+      handling: 1
     });
   };
 
@@ -441,7 +451,9 @@ export default function DocumentManager() {
   const prevStatusCounts = useRef({
     processed: 0,
     processing: 0,
+    handling: 0,
     pending: 0,
+    ready: 0,
     failed: 0
   })
 
@@ -532,6 +544,8 @@ export default function DocumentManager() {
         processed: response.documents.filter((doc: DocStatusResponse) => doc.status === 'processed'),
         processing: response.documents.filter((doc: DocStatusResponse) => doc.status === 'processing'),
         pending: response.documents.filter((doc: DocStatusResponse) => doc.status === 'pending'),
+        ready: response.documents.filter((doc: DocStatusResponse) => doc.status === 'ready'),
+        handling: response.documents.filter((doc: DocStatusResponse) => doc.status === 'handling'),
         failed: response.documents.filter((doc: DocStatusResponse) => doc.status === 'failed')
       }
     };
@@ -794,7 +808,14 @@ export default function DocumentManager() {
       // Check if component is still mounted before starting the request
       if (!isMountedRef.current) return;
 
-      const { status, message, track_id: _track_id } = await scanNewDocuments(); // eslint-disable-line @typescript-eslint/no-unused-vars
+      if (!selectedScheme) {
+        toast.error(t('documentPanel.documentManager.errors.missingSchemeId'));
+        return;
+      }
+
+      const schemeConfig = selectedScheme.config
+
+      const { status, message, track_id: _track_id } = await scanNewDocuments(schemeConfig); // eslint-disable-line @typescript-eslint/no-unused-vars
 
       // Check again if component is still mounted after the request completes
       if (!isMountedRef.current) return;
@@ -820,10 +841,10 @@ export default function DocumentManager() {
     } catch (err) {
       // Only show error if component is still mounted
       if (isMountedRef.current) {
-        toast.error(t('documentPanel.documentManager.errors.scanFailed', { error: errorMessage(err) }));
+        toast.error(t('documentPanel.documentManager.errors.scanFiled', { error: errorMessage(err) }));
       }
     }
-  }, [t, startPollingInterval, currentTab, health, statusCounts])
+  }, [t, startPollingInterval, currentTab, health, statusCounts, selectedScheme])
 
   // Handle page size change - update state and save to store
   const handlePageSizeChange = useCallback((newPageSize: number) => {
@@ -839,6 +860,8 @@ export default function DocumentManager() {
       processing: 1,
       pending: 1,
       failed: 1,
+      ready: 1,
+      handling: 1
     });
 
     setPagination(prev => ({ ...prev, page: 1, page_size: newPageSize }));
@@ -878,6 +901,8 @@ export default function DocumentManager() {
             processed: response.documents.filter(doc => doc.status === 'processed'),
             processing: response.documents.filter(doc => doc.status === 'processing'),
             pending: response.documents.filter(doc => doc.status === 'pending'),
+            ready: response.documents.filter((doc: DocStatusResponse) => doc.status === 'ready'),
+            handling: response.documents.filter((doc: DocStatusResponse) => doc.status === 'handling'),
             failed: response.documents.filter(doc => doc.status === 'failed')
           }
         };
@@ -945,7 +970,9 @@ export default function DocumentManager() {
     const newStatusCounts = {
       processed: docs?.statuses?.processed?.length || 0,
       processing: docs?.statuses?.processing?.length || 0,
+      handling: docs?.statuses?.handling?.length || 0,
       pending: docs?.statuses?.pending?.length || 0,
+      ready: docs?.statuses?.ready?.length || 0,
       failed: docs?.statuses?.failed?.length || 0
     }
 
@@ -1133,6 +1160,7 @@ export default function DocumentManager() {
               <ClearDocumentsDialog onDocumentsCleared={handleDocumentsCleared} />
             ) : null}
             <UploadDocumentsDialog onDocumentsUploaded={fetchDocuments} />
+            <SchemeManager />
             <PipelineStatusDialog
               open={showPipelineStatus}
               onOpenChange={setShowPipelineStatus}
@@ -1181,6 +1209,17 @@ export default function DocumentManager() {
                   >
                     {t('documentPanel.documentManager.status.processing')} ({statusCounts.PROCESSING || statusCounts.processing || 0})
                   </Button>
+                  <Button
+                    size="sm"
+                    variant={statusFilter === 'handling' ? 'secondary' : 'outline'}
+                    onClick={() => setStatusFilter('handling')}
+                    className={cn(
+                      documentCounts.handling > 0 ? 'text-purple-600' : 'text-gray-500',
+                      statusFilter === 'handling' && 'bg-purple-100 dark:bg-purple-900/30 font-medium border border-purple-400 dark:border-purple-600 shadow-sm'
+                    )}
+                  >
+                    {t('documentPanel.documentManager.status.handling')} ({statusCounts.HANDLING || statusCounts.handling || 0})
+                  </Button>
                   <Button
                     size="sm"
                     variant={statusFilter === 'pending' ? 'secondary' : 'outline'}
@@ -1193,6 +1232,17 @@ export default function DocumentManager() {
                   >
                     {t('documentPanel.documentManager.status.pending')} ({statusCounts.PENDING || statusCounts.pending || 0})
                   </Button>
+                  <Button
+                    size="sm"
+                    variant={statusFilter === 'ready' ? 'secondary' : 'outline'}
+                    onClick={() => setStatusFilter('ready')}
+                    className={cn(
+                      documentCounts.ready > 0 ? 'text-gray-600' : 'text-gray-500',
+                      statusFilter === 'ready' && 'bg-gray-100 dark:bg-gray-900/30 font-medium border border-gray-400 dark:border-gray-600 shadow-sm'
+                    )}
+                  >
+                    {t('documentPanel.documentManager.status.ready')} ({statusCounts.READY || statusCounts.ready || 0})
+                  </Button>
                   <Button
                     size="sm"
                     variant={statusFilter === 'failed' ? 'secondary' : 'outline'}
@@ -1273,6 +1323,7 @@ export default function DocumentManager() {
                           </div>
                         </TableHead>
                         <TableHead>{t('documentPanel.documentManager.columns.summary')}</TableHead>
+                        <TableHead>{t('documentPanel.documentManager.columns.handler')}</TableHead>
                         <TableHead>{t('documentPanel.documentManager.columns.status')}</TableHead>
                         <TableHead>{t('documentPanel.documentManager.columns.length')}</TableHead>
                         <TableHead>{t('documentPanel.documentManager.columns.chunks')}</TableHead>
@@ -1344,6 +1395,9 @@ export default function DocumentManager() {
                               </div>
                             </div>
                           </TableCell>
+                          <TableCell className="truncate max-w-[150px]">
+                            {doc.scheme_name || '-'}
+                          </TableCell>
                           <TableCell>
                             <div className="group relative flex items-center overflow-visible tooltip-container">
                               {doc.status === 'processed' && (
@@ -1358,6 +1412,12 @@ export default function DocumentManager() {
                               {doc.status === 'failed' && (
                                 <span className="text-red-600">{t('documentPanel.documentManager.status.failed')}</span>
                               )}
+                              {doc.status === 'ready' && (
+                                <span className="text-purple-600">{t('documentPanel.documentManager.status.ready')}</span>
+                              )}
+                              {doc.status === 'handling' && (
+                                <span className="text-gray-600">{t('documentPanel.documentManager.status.handling')}</span>
+                              )}
 
                               {/* Icon rendering logic */}
                               {doc.error_msg ? (
@@ -1382,10 +1442,10 @@ export default function DocumentManager() {
                           <TableCell>{doc.content_length ?? '-'}</TableCell>
                           <TableCell>{doc.chunks_count ?? '-'}</TableCell>
                           <TableCell className="truncate">
-                            {new Date(doc.created_at).toLocaleString()}
+                            {doc.created_at ? new Date(doc.created_at).toLocaleString() : '-'}
                           </TableCell>
                           <TableCell className="truncate">
-                            {new Date(doc.updated_at).toLocaleString()}
+                            {doc.updated_at ? new Date(doc.updated_at).toLocaleString() : '-'}
                           </TableCell>
                           <TableCell className="text-center">
                             <Checkbox

lightrag_webui/src/locales/ar.json

@@ -126,6 +126,7 @@
         "id": "المعرف",
         "fileName": "اسم الملف",
         "summary": "الملخص",
+        "handler": "المعالج",
         "status": "الحالة",
         "length": "الطول",
         "chunks": "الأجزاء",
@@ -138,13 +139,16 @@
         "all": "الكل",
         "completed": "مكتمل",
         "processing": "قيد المعالجة",
+        "handling": "استخراج",
         "pending": "معلق",
+        "ready": "جاهز",
         "failed": "فشل"
       },
       "errors": {
         "loadFailed": "فشل تحميل المستندات\n{{error}}",
         "scanFailed": "فشل مسح المستندات\n{{error}}",
-        "scanProgressFailed": "فشل الحصول على تقدم المسح\n{{error}}"
+        "scanProgressFailed": "فشل الحصول على تقدم المسح\n{{error}}",
+        "missingSchemeId": "الحل هو في عداد المفقودين ، حدد الحل"
       },
       "fileNameLabel": "اسم الملف",
       "showButton": "عرض",
@@ -422,5 +426,31 @@
     "prevPage": "الصفحة السابقة",
     "nextPage": "الصفحة التالية",
     "lastPage": "الصفحة الأخيرة"
+  },
+  "schemeManager": {
+    "button": "مخططات معالجة المستندات",
+    "title": "مدير المخططات",
+    "description": "إنشاء مخططات جديدة وتكوين الخيارات",
+    "schemeList": "قائمة المخططات",
+    "schemeConfig": "تكوين المخطط",
+    "inputPlaceholder": "أدخل اسم المخطط",
+    "deleteTooltip": "حذف المخطط",
+    "emptySchemes": "لا توجد مخططات متاحة",
+    "selectSchemePrompt": "يرجى تحديد أو إنشاء مخطط أولاً",
+    "processingFramework": "إطار المعالجة",
+    "extractionTool": "أداة الاستخراج",
+    "modelSource": "مصدر النموذج",
+    "errors": {
+      "loadFailed": "فشل تحميل المخططات",
+      "nameEmpty": "لا يمكن أن يكون اسم المخطط فارغًا",
+      "nameExists": "اسم المخطط موجود بالفعل",
+      "addFailed": "فشل إضافة المخطط",
+      "deleteFailed": "فشل حذف المخطط"
+    },
+    "upload": {
+      "noSchemeSelected": "لم يتم تحديد مخطط معالجة، يرجى تحديد مخطط معالجة المستندات أولاً!",
+      "currentScheme": "المخطط الحالي: ",
+      "noSchemeMessage": "لم يتم تحديد مخطط معالجة، يرجى إضافة وتحديد واحد أولاً"
+    }
   }
 }

lightrag_webui/src/locales/en.json

@@ -126,6 +126,7 @@
         "id": "ID",
         "fileName": "File Name",
         "summary": "Summary",
+        "handler": "Handler",
         "status": "Status",
         "length": "Length",
         "chunks": "Chunks",
@@ -138,13 +139,16 @@
         "all": "All",
         "completed": "Completed",
         "processing": "Processing",
+        "handling": "Handling",
         "pending": "Pending",
+        "ready": "Ready",
         "failed": "Failed"
       },
       "errors": {
         "loadFailed": "Failed to load documents\n{{error}}",
         "scanFailed": "Failed to scan documents\n{{error}}",
-        "scanProgressFailed": "Failed to get scan progress\n{{error}}"
+        "scanProgressFailed": "Failed to get scan progress\n{{error}}",
+        "missingSchemeId": "Lack of solution, please select a solution"
       },
       "fileNameLabel": "File Name",
       "showButton": "Show",
@@ -422,5 +426,31 @@
     "prevPage": "Previous Page",
     "nextPage": "Next Page",
     "lastPage": "Last Page"
+  },
+  "schemeManager": {
+    "button": "Document Processing Schemes",
+    "title": "Scheme Manager",
+    "description": "Create new schemes and configure options",
+    "schemeList": "Scheme List",
+    "schemeConfig": "Scheme Configuration",
+    "inputPlaceholder": "Enter scheme name",
+    "deleteTooltip": "Delete scheme",
+    "emptySchemes": "No schemes available",
+    "selectSchemePrompt": "Please select or create a scheme first",
+    "processingFramework": "Processing Framework",
+    "extractionTool": "Extraction Tool",
+    "modelSource": "Model Source",
+    "errors": {
+      "loadFailed": "Failed to load schemes",
+      "nameEmpty": "Scheme name cannot be empty",
+      "nameExists": "Scheme name already exists",
+      "addFailed": "Failed to add scheme",
+      "deleteFailed": "Failed to delete scheme"
+    },
+    "upload": {
+      "noSchemeSelected": "No processing scheme selected, please select a document processing scheme first!",
+      "currentScheme": "Current scheme: ",
+      "noSchemeMessage": "No processing scheme selected, please add and select one"
+    }
   }
 }

lightrag_webui/src/locales/fr.json

@@ -126,6 +126,7 @@
         "id": "ID",
         "fileName": "Nom du fichier",
         "summary": "Résumé",
+        "handler": "Gestionnaire",
         "status": "Statut",
         "length": "Longueur",
         "chunks": "Fragments",
@@ -138,13 +139,16 @@
         "all": "Tous",
         "completed": "Terminé",
         "processing": "En traitement",
+        "handling": "Extraction",
         "pending": "En attente",
+        "ready": "Prêt",
         "failed": "Échoué"
       },
       "errors": {
         "loadFailed": "Échec du chargement des documents\n{{error}}",
         "scanFailed": "Échec de la numérisation des documents\n{{error}}",
-        "scanProgressFailed": "Échec de l'obtention de la progression de la numérisation\n{{error}}"
+        "scanProgressFailed": "Échec de l'obtention de la progression de la numérisation\n{{error}}",
+        "missingSchemeId": "Schéma de traitement manquant, veuillez sélectionner un schéma de traitement"
       },
       "fileNameLabel": "Nom du fichier",
       "showButton": "Afficher",
@@ -422,5 +426,31 @@
     "prevPage": "Page précédente",
     "nextPage": "Page suivante",
     "lastPage": "Dernière page"
+  },
+  "schemeManager": {
+    "button": "Schémas de traitement de documents",
+    "title": "Gestionnaire de schémas",
+    "description": "Créer de nouveaux schémas et configurer les options",
+    "schemeList": "Liste des schémas",
+    "schemeConfig": "Configuration du schéma",
+    "inputPlaceholder": "Entrer le nom du schéma",
+    "deleteTooltip": "Supprimer le schéma",
+    "emptySchemes": "Aucun schéma disponible",
+    "selectSchemePrompt": "Veuillez d'abord sélectionner ou créer un schéma",
+    "processingFramework": "Framework de traitement",
+    "extractionTool": "Outil d'extraction",
+    "modelSource": "Source du modèle",
+    "errors": {
+      "loadFailed": "Échec du chargement des schémas",
+      "nameEmpty": "Le nom du schéma ne peut pas être vide",
+      "nameExists": "Le nom du schéma existe déjà",
+      "addFailed": "Échec de l'ajout du schéma",
+      "deleteFailed": "Échec de la suppression du schéma"
+    },
+    "upload": {
+      "noSchemeSelected": "Aucun schéma de traitement sélectionné, veuillez d'abord sélectionner un schéma de traitement de documents !",
+      "currentScheme": "Schéma actuel : ",
+      "noSchemeMessage": "Aucun schéma de traitement sélectionné, veuillez d'abord en ajouter et en sélectionner un"
+    }
   }
 }

lightrag_webui/src/locales/zh.json

@@ -126,6 +126,7 @@
         "id": "ID",
         "fileName": "文件名",
         "summary": "摘要",
+        "handler": "处理方案",
         "status": "状态",
         "length": "长度",
         "chunks": "分块",
@@ -138,13 +139,16 @@
         "all": "全部",
         "completed": "已完成",
         "processing": "处理中",
+        "handling": "提取中",
         "pending": "等待中",
+        "ready": "准备中",
         "failed": "失败"
       },
       "errors": {
         "loadFailed": "加载文档失败\n{{error}}",
         "scanFailed": "扫描文档失败\n{{error}}",
-        "scanProgressFailed": "获取扫描进度失败\n{{error}}"
+        "scanProgressFailed": "获取扫描进度失败\n{{error}}",
+        "missingSchemeId": "缺少处理方案，请选择处理方案"
       },
       "fileNameLabel": "文件名",
       "showButton": "显示",
@@ -422,5 +426,31 @@
     "prevPage": "上一页",
     "nextPage": "下一页",
     "lastPage": "末页"
+  },
+  "schemeManager": {
+    "button": "文档处理方案",
+    "title": "方案管理器",
+    "description": "创建新方案并配置选项",
+    "schemeList": "方案列表",
+    "schemeConfig": "方案配置",
+    "inputPlaceholder": "输入方案名称",
+    "deleteTooltip": "删除方案",
+    "emptySchemes": "暂无方案",
+    "selectSchemePrompt": "请先选择或创建一个方案",
+    "processingFramework": "处理框架",
+    "extractionTool": "提取工具",
+    "modelSource": "模型源",
+    "errors": {
+      "loadFailed": "加载方案失败",
+      "nameEmpty": "方案名称不能为空",
+      "nameExists": "方案名称已存在",
+      "addFailed": "添加方案失败",
+      "deleteFailed": "删除方案失败"
+    },
+    "upload": {
+      "noSchemeSelected": "未选择处理方案，请先选择文档处理方案！",
+      "currentScheme": "当前方案：",
+      "noSchemeMessage": "未选择处理方案，请先添加选择"
+    }
   }
 }

lightrag_webui/src/locales/zh_TW.json

@@ -126,6 +126,7 @@
         "id": "ID",
         "fileName": "檔案名稱",
         "summary": "摘要",
+        "handler": "處理方案",
         "status": "狀態",
         "length": "長度",
         "chunks": "分塊",
@@ -138,13 +139,16 @@
         "all": "全部",
         "completed": "已完成",
         "processing": "處理中",
+        "handling": "提取中",
         "pending": "等待中",
+        "ready": "準備中",
         "failed": "失敗"
       },
       "errors": {
         "loadFailed": "載入文件失敗\n{{error}}",
         "scanFailed": "掃描文件失敗\n{{error}}",
-        "scanProgressFailed": "取得掃描進度失敗\n{{error}}"
+        "scanProgressFailed": "取得掃描進度失敗\n{{error}}",
+        "missingSchemeId": "缺少處理方案，請選擇處理方案"
       },
       "fileNameLabel": "檔案名稱",
       "showButton": "顯示",
@@ -422,5 +426,31 @@
     "prevPage": "上一頁",
     "nextPage": "下一頁",
     "lastPage": "最後一頁"
+  },
+  "schemeManager": {
+    "button": "文件處理方案",
+    "title": "方案管理器",
+    "description": "建立新方案並設定選項",
+    "schemeList": "方案清單",
+    "schemeConfig": "方案設定",
+    "inputPlaceholder": "輸入方案名稱",
+    "deleteTooltip": "刪除方案",
+    "emptySchemes": "暫無方案",
+    "selectSchemePrompt": "請先選擇或建立一個方案",
+    "processingFramework": "處理框架",
+    "extractionTool": "提取工具",
+    "modelSource": "模型源",
+    "errors": {
+      "loadFailed": "載入方案失敗",
+      "nameEmpty": "方案名稱不能為空",
+      "nameExists": "方案名稱已存在",
+      "addFailed": "新增方案失敗",
+      "deleteFailed": "刪除方案失敗"
+    },
+    "upload": {
+      "noSchemeSelected": "未選擇處理方案，請先選擇文件處理方案！",
+      "currentScheme": "目前方案：",
+      "noSchemeMessage": "未選擇處理方案，請先新增選擇"
+    }
   }
 }

lightrag_webui/tailwind.config.js

@@ -1,4 +1,5 @@
 /** @type {import('tailwindcss').Config} */
+
 export default {
   darkMode: ['class'],
   content: [