fricon

Data collection automation framework.

Documentation

GitHub Pages

Overview

Fricon is a data collection automation framework designed for managing datasets in scientific and research workflows. It provides:

• Workspace Management: Organized data storage with metadata tracking
• Dataset Operations: Arrow-format data tables with UUID and incremental ID tracking
• Desktop UI: Cross-platform Tauri application with React 19 frontend
• Server Architecture: IPC-based server for workspace operations

Quick Start

Installation

For Python users (recommended):

pip install fricon

For development or building from source:

Building from source requires several dependencies including Rust, protoc, uv, pnpm, and platform-specific libraries. See CONTRIBUTING.md for complete setup instructions.

git clone https://github.com/kahojyun/fricon.git
cd fricon
python3 scripts/setup-dev.py

Basic Usage

Create a workspace:

fricon init path/to/workspace

Launch the desktop UI:

fricon gui path/to/workspace

Or connect from Python to a workspace with a running server:

from pathlib import Path

from fricon import Workspace

workspace_path = Path("path/to/workspace")
ws = Workspace.connect(workspace_path)

# Initialize a new dataset (schema is automatically inferred)
writer = ws.dataset_manager.create("my_dataset", description="My test dataset")

# Write data - schema is inferred from the first row
# MVP currently supports float and complex types only
writer.write(id=1, value=42.0, measurement=3.14 + 2j)
writer.write(id=2, value=84.0, measurement=1.618 - 1j)
writer.close()

# List all datasets
datasets = ws.dataset_manager.list_all()
print(datasets)

Contributing

We welcome contributions! Please see CONTRIBUTING.md for detailed guidelines on development environment setup, building components, testing, and contribution workflow.

License

fricon is distributed under the terms of the MIT OR Apache-2.0 license.