mtrview 0.2.0

Human-friendly dashboard for mtr2mqtt summary topics

mtrview

mtrview is a small standalone web dashboard for retained JSON summaries published by mtr2mqtt. mtr2mqtt writes the measurement summaries to MQTT; mtrview subscribes to those topics and turns the latest values into a simple human-friendly dashboard.

The app subscribes to summary topics such as summary/<receiver>, keeps the latest summary per receiver in memory, normalizes transmitter readings, and serves a dark auto-updating dashboard at http://localhost:8000/.

MQTT input

By default mtrview subscribes to:

summary/#

Each retained message should be a JSON object with a receiver, an updated_at timestamp, and a transmitters mapping. Transmitter metadata may be partial; missing location, zone, quantity, description, or unit fields are displayed with sensible placeholders instead of failing.

Configuration

Settings are read from environment variables:

Variable	Default
MTRVIEW_MQTT_HOST	localhost
MTRVIEW_MQTT_PORT	1883
MTRVIEW_MQTT_USERNAME	unset
MTRVIEW_MQTT_PASSWORD	unset
MTRVIEW_MQTT_TOPICS	summary/#
MTRVIEW_HTTP_HOST	0.0.0.0
MTRVIEW_HTTP_PORT	8000
MTRVIEW_REFRESH_INTERVAL_SECONDS	20
MTRVIEW_DISPLAY_TIMEZONE	UTC

See .env.example for a copyable starting point.

Local run

uv sync --group dev
uv run mtrview

Then open http://localhost:8000/.

Useful endpoints:

• GET / - dashboard
• GET /api/summary - normalized JSON for all receivers and transmitters
• GET /health - health status

Docker

docker build -t mtrview .
docker run --rm -p 8000:8000 \
  -e MTRVIEW_MQTT_HOST=host.docker.internal \
  -e MTRVIEW_MQTT_TOPICS=summary/# \
  mtrview

Dashboard

The UI is server-rendered first, with a small vanilla JavaScript refresh loop that fetches /api/summary. It includes summary counts, MQTT connection state, search, status/zone/receiver filters, card and compact table views, and a priority section for non-online readings.

Development

make install
make verify

Individual checks:

uv run pytest
uv run ruff check .

Releases

Releases are managed by GitHub Actions using Conventional Commits and semantic-release. Merges to master can create GitHub releases, publish the Python package to PyPI, and publish multi-arch Docker images to Docker Hub as tvallas/mtrview.