openccu-data

Extract and distribute Homematic CCU configuration metadata (translations, easymodes, link profiles) from OCCU / OpenCCU / RaspberryMatic.

This repository is the single source of truth for the data artifacts that are consumed by aiohomematic and aiohomematic-config. Both projects vendor the produced JSON archives at runtime.

What this provides

All three read from either:

• a local OCCU/OpenCCU/RaspberryMatic checkout (OCCU_PATH=/path/to/occu), or
• a running CCU instance over HTTP/HTTPS (CCU_URL=https://my-ccu.local).

If both are set, the easymode/translation extractors merge results; the profile extractor prefers the running CCU and falls back to local.

Repository layout

openccu-data/
├── LICENSE                MIT (covers the code)
├── NOTICE.md              Data-artifact licensing (EQ-3/OCCU)
├── README.md              this file
├── CLAUDE.md              guide for AI assistants
├── changelog.md
├── pyproject.toml
├── openccu_data/
│   ├── const.py
│   ├── easymodes/extractor.py        easymode metadata parser
│   ├── translations/extractor.py     CCU WebUI translation parser
│   ├── profiles/extractor.py         easymode link-profile parser
│   └── data/                         committed, vendored output
│       ├── easymode_extract.json.gz
│       ├── translation_extract.json.gz
│       ├── translation_custom/*.json
│       └── profiles/*.json.gz (+ _receiver_type_aliases.json)
├── script/                           CLI wrappers
└── tests/

Installation

python -m pip install -e .[test]

No third-party runtime dependencies; only the standard library.

Usage

Console scripts

After installation, three console scripts are available on the PATH:

OCCU_PATH=/path/to/occu openccu-extract-easymodes
OCCU_PATH=/path/to/occu openccu-extract-translations
CCU_URL=https://my-ccu.local openccu-extract-profiles

Output lands in openccu_data/data/ by default. Override via OUTPUT_DIR.

Without installation

OCCU_PATH=/path/to/occu python script/extract_easymodes.py
OCCU_PATH=/path/to/occu python script/extract_translations.py
CCU_URL=https://my-ccu.local python script/extract_profiles.py

Environment variables

.env files at the repository root are auto-loaded (existing env vars win).

Vendoring into consumer projects

The committed artifacts in openccu_data/data/ are the source of truth. Consumers maintain their own runtime copies:

After regenerating any artifact, copy the relevant files into the consumer repository and open a PR there as well.

Development

python -m pip install -e .[test]
pytest tests/
ruff check openccu_data/ tests/
mypy

Parts of openccu-data are developed with agentic AI assistance, primarily Claude Code. Submitted issues are also triaged and analyzed with agentic help. Every change is still reviewed by a human maintainer and must pass the project's tests before it lands — AI accelerates the work, it does not replace the review gate.

License

• Code: MIT.
• Data artifacts under openccu_data/data/: derivative of OCCU/RaspberryMatic and subject to the EQ-3 license (see NOTICE.md). The curated translation_custom/ overrides are MIT.

"Homematic" and "HomematicIP" are trademarks of eQ-3 AG. This project is not affiliated with or endorsed by eQ-3 AG.