Implementation:Infiniflow Ragflow Parser Config Utils

Knowledge Sources	Infiniflow_Ragflow
Domains	Document_Processing, Configuration
Last Updated	2026-02-12 06:00 GMT

Overview

Concrete tool for normalizing layout recognizer configuration strings by extracting parser model names provided by the RAGFlow common library.

Description

The parser_config_utils module provides normalize_layout_recognizer which parses layout recognizer strings in the format "model_name@ParserType" (e.g., "yolo@MinerU") and returns a tuple of the normalized recognizer string and the extracted parser model name.

Usage

Import this function when processing dataset configuration to extract and normalize the layout recognizer selection from user input.

Code Reference

Source Location

• Repository: Infiniflow_Ragflow
• File: common/parser_config_utils.py
• Lines: 1-34

Signature

def normalize_layout_recognizer(layout_recognizer_raw: str) -> tuple:
    """Parse layout recognizer string, returns (layout_recognizer, parser_model_name)."""

Import

from common.parser_config_utils import normalize_layout_recognizer

I/O Contract

Inputs

Name	Type	Required	Description
layout_recognizer_raw	str	Yes	Raw layout recognizer string (e.g., "yolo@MinerU")

Outputs

Name	Type	Description
returns	tuple	(layout_recognizer: str, parser_model_name: str)

Usage Examples

from common.parser_config_utils import normalize_layout_recognizer

recognizer, model = normalize_layout_recognizer("yolo@MinerU")
# recognizer = "yolo@MinerU", model = "yolo"

recognizer2, model2 = normalize_layout_recognizer("paddle")
# recognizer2 = "paddle", model2 = "paddle"

Related Pages

• Environment:Infiniflow_Ragflow_Python_Runtime