intake-document

A Python application to convert documents into markdown format using the OCR capabilities provided by Mistral.ai.

Features

• Uses Mistral.ai's OCR API to analyze document content and convert into markdown
• Processes document using Mistral batch mode to save cost
• Uploads documents to Mistral as a separate step
• Maintains document structure and hierarchy
• Preserves formatting including:
  • Headers and subheadings
  • Paragraphs
  • Lists (ordered and unordered)
  • Tables with headers and data
• The Mistral API returns results in clean markdown format for easy parsing and rendering
• Handles complex layouts including multi-column text and mixed content
• Maintains non-text images as references that can be downloaded separately

Installation

Requirements

• Python 3.10 or higher
• Mistral.ai API key

Install from source

# Clone the repository
git clone https://github.com/yourusername/intake-document.git
cd intake-document

# Install using pip in development mode
uv pip install -e .

Environment setup

Set your Mistral.ai API key as an environment variable:

export MISTRAL_API_KEY="your-api-key-here"

Configuration

The application follows the XDG Base Directory specification for configuration:

• Configuration: ~/.config/intake-document/init.cfg
• Data: ~/.local/share/intake-document/
• Cache: ~/.cache/intake-document/
• State: ~/.local/state/intake-document/

Configuration file example

[mistral]
api_key = your-api-key-here
batch_size = 5
max_retries = 3
timeout = 60

[app]
output_dir = ./output
log_level = INFO

Usage

The application can be used as a command-line tool to process documents:

# Process a single file
intake-document --input path/to/document.pdf --output-dir path/to/output

# Process all documents in a directory
intake-document --input path/to/document/folder --output-dir path/to/output

# Show the current configuration
intake-document --show-config

# Display help
intake-document --help

Command Line Arguments

-i, --input PATH                  Path to input file or directory
-o, --output-dir PATH             Output directory (default: ./output)
-c, --config PATH                 Path to config file 
--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}  Set logging level (default: ERROR)
--show-config                     Show current configuration and exit
-h, --help                        Show help message and exit

Dependencies

• mistralai - Official Python library for Mistral.ai
• pydantic - Data validation and settings management
• rich - Rich text and formatting for the terminal
• structlog - Structured logging
• typer - CLI creation with type hints
• configparser - Configuration file parsing
• xdg-base-dirs - XDG Base Directory specification

Development

Testing

# Install development dependencies
uv pip install pytest

# Run tests
pytest

Formatting and Linting

# Install development tools
pip install ruff mypy

# Run formatter
ruff format src/ tests/

# Run linter
ruff check src/ tests/

# Run type checking
mypy src/

License

This project is licensed under the MIT License - see the LICENSE file for details.

API Documentation

The Mistral.ai OCR API is documented at: https://docs.mistral.ai/capabilities/document/