This small Python (bottle) server receives analytics events from Dolphin instances in the wild, deserializes them, and writes records to ClickHouse. It dynamically maintains the ClickHouse table schema in order to support some amount of schema agility, where new analytics fields can be added in the client without requiring server side changes.
- Python 3 and Poetry
- ClickHouse
Note: this requires Nix Flakes to be enabled on your system.
nix run github:dolphin-emu/analytics-ingestThis project uses Poetry for dependency management.
# Install dependencies (use --no-dev to skip optional dev dependencies).
poetry install
# Run the server
poetry run analytics-ingestLicensed under the MIT License. See LICENSE.