E2M: Convert multiple file formats to Markdown for easy document formatting unification

General Introduction

E2M (Everything to Markdown) is an open source Python library designed to convert a wide range of file formats to Markdown format. The tool supports a wide range of file types including doc, docx, epub, html, htm, url, pdf, ppt, pptx, mp3, and m4a. E2M utilizes a parser-transformer architecture that efficiently parses and transforms file content, providing flexible configuration options for data retrieval enhanced generation (RAG) and model training or fine-tuning. E2M's goal is to provide users with high-quality data conversion services that simplify the process of document format harmonization. Each format has a specialized parser and converter, using the Parser parser to extract text and images from the file, and the Converter converter to convert the extracted content to Markdown.

Function List

• file parsing: Supports parsing of multiple file types, including text and image data.
• format conversion: Convert the parsed data into Markdown format.
• Multiple parsers and converters: Parsers and converters that support different engines and strategies.
• Open source and flexible configuration: Provides open source code and flexible configuration options that can be customized by the user.
• API Services: Provides API services for easy integration into other applications.

Using Help

Installation process

1. Creating the Environment::

   conda create -n e2m python=3.10
conda activate e2m

2. Update pip::

   pip install --upgrade pip

3. Installation of E2M::
   • Install via git (recommended): bash
pip install git+https://github.com/wisupai/e2m.git --index-url https://pypi.org/simple

   • Installation via pip: bash
pip install --upgrade wisup_e2m

   • Manual installation: bash
git clone https://github.com/wisupai/e2m.git
cd e2m
pip install poetry
poetry build
pip install dist/wisup_e2m-0.1.63-py3-none-any.whl


Usage

1. Starting the API service::

   gunicorn wisup_e2m.api.main:app --workers 4 --worker-class uvicorn.workers.UvicornWorker --bind 0.0.0.0:8000

2. Access to API Documentation: Open your browser and visithttp://127.0.0.1:8000/docsTo view the API documentation and usage examples, click here.

Main function operation flow

1. File parsing and conversion::
   • Parses the contents of the file using a parser:

    from wisup_e2m.parsers import PdfParser
   parser = PdfParser()
   text_data = parser.parse('example.pdf')
   

   • Use a converter to convert the parsed content to Markdown format:

    from wisup_e2m.converters import TextConverter
   converter = TextConverter()
   markdown_data = converter.convert(text_data)
   

2. Customized Configuration::
   • Modify the configuration fileconfig.yaml, adjust the parameters of the parser and converter according to the needs:

    parsers.
   pdf.
   engine: 'unstructured'
   converters.
   engine: 'unstructured' converters: text.
   engine: 'litellm'
   

3. Integration into other applications::
   • Integrate E2M into other applications using API services to send HTTP requests for file parsing and conversion: python
import requests
response = requests.post('http://127.0.0.1:8000/convert', files={'file': open('example.pdf', 'rb')})
markdown_data = response.text