D(ocument)e(xtractor)adSimple

This FastAPI service extracts text content from a wide variety of document formats (PDF, DOCX, PPTX, EPUB, HTML, TXT, etc.) using markitdown. It returns the content as an array of strings, one for each logical page, slide, or section.

🚀 Features

• Supports multiple document formats
• Returns page-wise content as a JSON array
• Automatically detects file type via content-type
• FastAPI + Uvicorn app, easy to deploy

📦 Supported File Types

• PDF (application/pdf)
• DOCX / Word
• PPTX / PowerPoint
• EPUB
• HTML
• Markdown
• TXT
• CSV

Example usage

• In OpenWebUI, configure for Document Extractor external and as url http://localhost:5000 (you can change this based on your needs)
• Locally, curl -X POST http://localhost:5000/process -H "Content-Type: application/pdf" --data-binary @file.pdf

Env

• PORT=5000
• LLM_TOKEN # You can use a LLM to read images
• LLM_MODEL # The LLM model
• LLM_URL # The URL of an OpenAI compatible provider

🧪 Setup (with virtual environment)

python3 -m venv .venv
source .venv/bin/activate

pip install -r requirements.txt

./main.py

Optional Backends

The service can work with optional backends that provide extra capabilities such as OCR, document linking, and content analysis. These backends are not required for basic operation; the API works out‑of‑the‑box.

Available backends

• ocrflux – OCR extraction backend
• doclings – Document linking backend
• docstrange – Specialized document processing backend
• marker – Marker based backend

Installation

Each backend is provided as an extra in pyproject.toml. Install the desired backend with pip, for example:

pip install .[ocrflux]
pip install .[doclings]
pip install .[docstrange]
pip install .[marker]

You can also install multiple extras at once:

pip install .[ocrflux,doclings,docstrange,marker]

If no extra is installed, the core service runs without these features.