Roadmap and Vision

Vision

User Interface (GUI): Develop an intuitive and visually appealing interface to replace the current Node-RED setup.
Middleware Integration: Enable seamless coupling with platforms like Home Assistant or similar solutions to simplify device connections and management.
Ready-to-Use Images: Offer pre-configured Docker images and long-term solutions with pre-installed hardware for easy deployment.
Device Anbindung (Integration): Standardize and expand support for various devices and protocols to ensure scalability and flexibility.

First release packages, almost stable API, user + dev documentation.

API
- translate endpoints
- Prepare for stable API
  - Support for multiple devices (API only)
  - All components optional (where applicable) (API only)
- Drop unused endpoints
Documentation
- Documented Workflow for HA(/OpenHab) integration + evcc
- Documentation about general system usage on README (not for development, forward to contributing.md)
  - hardware requirements (pi)
  - platforms specifics? (linux/windows/mac)
  - setup guide with python package
  - setup guide with docker image
- Swagger API-docs with examples, more detailed
- Review contributions guide for new developers (where to start, setup, how to test, PR process)
CI/CD
- Release process
  - PyPi Packaging
  - Dockerhub image (PR #xxx)
  - Publish documentation on ReadTheDocs
Others
- Get feature/config-overhaul merged
- Prepare for stable config
  - nested config for convenience/readability (pydantic.SettingsModel)

refactor ems into new class layout
optimize doesn't change global config state (PR #xxx)
load balancer fasthtml/fastapi (PR #xxx)
tests don't affect dev/user config (PR #xxx)
hardcoded pvforecast attributes
mypy works with pre-commit but still has a few issues on venv execution
float <-> int changes of individual
missing API documentation of all the new pydantic models (description fields), documentation/examples about usage

translate German code to English
1. inverter (done #xxx)
2. battery (done #xxx)
3. ems (partially #xxx)
4. devices
improve optimization by reducing generations needed (partially #xxx)
fix pdf generation, parts of graphs do intersect (done #xxx)
PDF shows actual dates/time (x-axis)
PDF from old run still accessible (maybe optimize returns id and that id can be taken for pdf endpoint or latest without id, last n ids or last n days are preserved)
Tests:
- all endpoint are tested for deployed server (docker-compose)
HA/OpenHab interface (e.g. MQTT)
MR workflow required? (approvals who, how many, when to use feature branch, when to use main)
User management (access control, e.g. config change, prediction update, read-only etc.)
- Database
GUI
- Dashboard
- Update predictions
- Update optimization
- Update config
- Mobile support
Middleware (take inspiration from evcc)
- MQTT
- HA-Addon
- OpenHab-Addon
Extendable interface (documentation how to add new device with its own mutation/optimization into system)
Support for multiple devices
- pv/inverter
- ev
- batteries
- household devices
All components optional (where applicable)
Systems runs out-of-the-box without config changes (maybe in GUI setup wizard)
stochastic optimization (encounter variance)