Ensure LightRAG instance is initialized, create if necessary

Legacy method - now handled by _extract_all_json_candidates

Legacy method - now handled by progressive strategies

Process inline markdown formatting (bold, italic, code, links) Args: text: Raw text with markdown formatting Re

Cancel a running session

Main chat endpoint that receives files and instructions Args: message: User's text message content: 'paper' or 'general'

Check if MinerU 2.0 is properly installed Returns: bool: True if installation is valid, False otherwise

Check if Docling is properly installed Returns: bool: True if installation is valid, False otherwise

Check if the configured parser is properly installed Returns: bool: True if the configured parser is properly installed

Create an enhanced summary by separating summary text and original elements. Args: summary_content: The RAG-generated summary co

Download generated file (supports subdirectories)

Filter file paths to only include supported file types Args: file_paths: List of file paths to filter recurs

Fix common invalid escape sequences in JSON strings.

Create from environment variables.

Wrap an existing LightRAG instance.

Load from existing storage directory.

Get current configuration information.

Get detailed processing status for a document. Args: doc_id: Document ID to check Returns: Dict wit

Get list of full paths to figure images.

Get merged content for display or saving.

Get processor information

Get processor information.

Get queries for document analysis. Returns: If use_predefined_paper_queries=True: Dict with query configs Otherwise: Lis

Get the final result for a completed session

Check if there is a session currently running

Get (FigureInfo, focus) pairs for a section.

Get (TableInfo, extract) pairs for a section.

Get processing status for a session

Get supported file extensions.

Identify all table blocks in content. Useful for debugging or selective table handling. Args: content: Text content to

Index all documents in a folder. Args: folder_path: Path to folder containing documents. output_dir:

Insert content list directly without document parsing Args: content_list: Pre-parsed content list containing text and mu

Check if a document is fully processed (both text and multimodal content). Args: doc_id: Document ID to check R

Backward compatibility property for old code. .. deprecated:: Use `parse_method` instead. This property will be removed i

Parse document using MinerU 2.0 based on file extension Args: file_path: Path to the file to be parsed metho

Parse document using Docling based on file extension Args: file_path: Path to the file to be parsed method:

Abstract method to parse PDF document. Must be implemented by subclasses. Args: pdf_path: Path to the PDF file

Complete document processing workflow Args: file_path: Path to the file to process output_dir: output direct

API exclusively for LightRAG calls: Complete document processing workflow Args: file_path: Path to the file to process

Asynchronously process multiple documents in batch Args: file_paths: List of file paths or directories to process

Process table content with context support

Process equation content with context support

Process generic modal content with context support

Process a single file Args: file_path: Path to the file to process output_dir: Output directory

Process single item using the correct processor for its type

Synchronous version of pure text query Args: query: Query text mode: Query mode ("local", "global", "hybrid"

Execute a single query with semaphore control.

Synchronous version of multimodal query Args: query: Base query text multimodal_content: List of multimodal

Run pipeline in background and store results

Set content source for context extraction in all modal processors. Args: content_source: Source content for cont

Calculate success rate as percentage

Convert to dictionary for JSON serialization.

Convert to dictionary for JSON serialization.

Convert to markdown format with caption before image.

Update configuration with new values

Update RAG configuration with new values.

Update context extraction configuration Args: **context_kwargs: Context configuration parameters to update (conte

Update context extraction configuration. Args: **context_kwargs: Context configuration parameters