mqttcommander 0.0.1

MQTTCommander

MQTTCommander

Convenience tools for discovering and commanding Tasmota devices over MQTT, plus a small CLI and a ready-to-use Docker image. This project now depends on the external library mqttstuff for the generic MQTT client/wrapper utilities — the mqttstuff source is no longer embedded in this repo and is pulled via the package manager.

Key capabilities:

• Connecting/subscribing and publishing via the mqttstuff wrapper

• Reading “last/most recent” messages with timeout-based collection

• Inspecting and commanding Tasmota devices via their MQTT topics

• Pydantic-based configuration (YAML + environment overrides)

• Developer helpers for JSON pretty-printing, deep updates, and logging configuration

• This repository: mqttcommander

• External dependency: mqttstuff (https://github.com/vroomfondel/mqttstuff)

Overview

MQTTCommander builds on top of paho-mqtt through the external mqttstuff library to simplify common patterns:

• A mqttstuff.mosquittomqttwrapper.MosquittoClientWrapper to configure, connect, subscribe, and publish with minimal boilerplate
• A mqttstuff.mosquittomqttwrapper.MQTTLastDataReader utility to retrieve the most recent messages quickly
• A mqttcommander.tasmotacommander.MqttCommander toolkit to discover Tasmota devices from retained topics and interact with them in bulk
• Pydantic settings in config.py to load credentials and broker details from config.yaml/config.local.yaml and/or environment

The project also includes a Dockerfile for a batteries-included container image useful for testing and running these tools in a consistent environment.

Repository layout

.
├─ mqttcommander/
│  ├─ __init__.py
│  ├─ cli.py
│  ├─ tasmotacommander.py
│  ├─ models.py
│  ├─ Helper.py
│  └─ py.typed
├─ config.py                    # Pydantic settings + logging setup
├─ main.py                      # Entry example/utility
├─ requirements*.txt
├─ pyproject.toml
├─ Makefile
├─ Dockerfile
├─ config.yaml                  # default config (example)
├─ config.local.yaml            # local overrides (git-ignored; example provided)
└─ tests/

Installation

Options:

• From PyPI:

  • python -m pip install mqttcommander
  • This will also install mqttstuff as a dependency.

• From source (editable):

  • python -m venv .venv && source .venv/bin/activate
  • pip install -r requirements-dev.txt
  • pip install -e .

• Build distributions with Hatch:

  • make pypibuild
  • Artifacts are created under dist/

Quick Start

Simple publish and subscribe using the wrapper:

from mqttstuff.mosquittomqttwrapper import MosquittoClientWrapper

client = MosquittoClientWrapper(
    host="localhost", port=1883, username="user", password="pass",
    topics=["test/topic"],
)

def on_any_message(msg, userdata):
    # msg is an instance of MWMqttMessage with convenient fields
    print(msg.topic, msg.value)

client.set_on_msg_callback(on_any_message, rettype="valuemsg")
client.connect_and_start_loop_forever()

# elsewhere or in another process
client.publish_one("test/topic", {"hello": "world"}, retain=False)

Read last retained or recent messages with a timeout:

from mqttstuff.mosquittomqttwrapper import MQTTLastDataReader

data = MQTTLastDataReader.get_most_recent_data_with_timeout(
    host="localhost", port=1883, username="user", password="pass",
    topics=["tele/+/STATE", "stat/+/STATUS"],
    retained="only",  # "yes" | "no" | "only"
    rettype="str_raw", # or "json", "valuemsg", "str", "int", "float"
)
print(data)

Configuration

Configuration is defined with Pydantic Settings in config.py and loaded from:

1. Environment variables
2. config.local.yaml (if present)
3. config.yaml

You can override paths with environment variables:

• MQTTSTUFF_CONFIG_DIR_PATH – base config dir
• MQTTSTUFF_CONFIG_PATH – path to main YAML config
• MQTTSTUFF_CONFIG_LOCAL_PATH – path to local override YAML

The Mqtt section is expected to contain common fields like host, port, username, password, and optional topic lists. See the file headers in config.py for details.

Python Modules

Each Python module provided by this repository is documented here with a focused explanation of its purpose and usage.

Module: mqttstuff.mosquittomqttwrapper (external dependency)

Key classes and responsibilities:

• MWMqttMessage (Pydantic model)

  • Normalized container for incoming/outgoing MQTT messages
  • Helpers like from_pahomsg(...) and fields for topic, qos, retain, payload, value, created_at, and optional metadata

• MosquittoClientWrapper

  • Thin wrapper around paho.mqtt.client.Client
  • Simplifies connection setup and topic subscriptions via set_topics([...])
  • Register callbacks per-topic (add_message_callback(topic, callback, rettype=...)) or a global callback (set_on_msg_callback)
  • Publish utilities:
    • publish_one(topic, value, created_at=None, metadata=None, rettype="valuemsg", retain=False, timeout=None)
    • publish_multiple(list[MWMqttMessage], timeout=None)
  • Connection loop helpers:
    • connect_and_start_loop_forever(topics=None, timeout_connect_seconds=None)
    • wait_for_connect_and_start_loop()
  • Convenience: automatic payload conversion for int/float/str/JSON/valuemsg

• MQTTLastDataReader

  • Static helper to retrieve the most recent messages within a configurable timeout window
  • Supports retained-only, no-retained, or mixed operation via retained parameter
  • Returns results in different representations via rettype and fallback_rettype

Example – per-topic callback with type conversion:

from mqttstuff.mosquittomqttwrapper import MosquittoClientWrapper

client = MosquittoClientWrapper(
    host="localhost", port=1883, username="user", password="pass",
    topics=["home/+/temperature"],
)

def on_temperature(msg, userdata):
    # msg.value is already a number if rettype="int"/"float"
    print("Temp:", msg.value)

client.add_message_callback("home/+/temperature", on_temperature, rettype="float")
client.connect_and_start_loop_forever()

Module: mqttcommander.tasmotacommander

Tools to discover and command Tasmota devices using their MQTT topics.

Highlights:

• Data models for timers, timezone/DST config, device config and sensors
• MqttCommander to collect retained messages across topics, filter noisy subtrees, and start processing loops
• Discovery helpers:
  • get_all_tasmota_devices_from_retained(...)
  • filter_online_tasmotas_from_retained(...)
  • update_online_tasmotas(...)
• Command helpers to send one or many commands to all online devices:
  • send_cmds_to_online_tasmotas(tasmotas, to_be_used_commands=[...], values_to_send=[...])
• Timezone utilities to ensure consistent device settings:
  • ensure_correct_timezone_settings_for_tasmotas(online_tasmotas)
• JSON utilities and pretty-printers for review and persistence:
  • write_tasmota_devices_file(...)
  • read_tasmotas_from_latest_file(...)

Example – list online devices and send a command:

from mqttcommander.tasmotacommander import MqttCommander

comm = MqttCommander(host="localhost", port=1883, username="user", password="pass")
all_devs = comm.get_all_tasmota_devices_from_retained(topics=["tele/+/STATE"], noisy=False)
online = comm.filter_online_tasmotas_from_retained(all_devs)
comm.send_cmds_to_online_tasmotas(online, to_be_used_commands=["Power"], values_to_send=[["Toggle"]])

Module: config

• Centralized configuration and Loguru logging setup
• Uses pydantic-settings to read from environment and YAML
• Timezone helpers and constants (e.g., TZBERLIN)
• Environment variables LOGURU_LEVEL, MQTTSTUFF_CONFIG_* are respected

Typical usage:

from config import Settings

settings = Settings()  # loads from env + config.local.yaml + config.yaml
print(settings.MQTT.host, settings.MQTT.port)

Module: Helper

Small utilities used across the project:

• ComplexEncoder for JSON serialization of complex types (UUID, datetimes, dict/list pretty rendering)
• print_pretty_dict_json, get_pretty_dict_json, get_pretty_dict_json_no_sort
• update_deep(base, u) for deep dict/list merge/update
• get_exception_tb_as_string(exc) for converting exception tracebacks to strings
• get_loguru_logger_info() to introspect Loguru handlers and filters

Docker

The repository contains a ready-to-use Dockerfile at the repository root designed for local development and CI usage.

What the Docker image includes

• Base: python:${python_version}-${debian_version} (defaults 3.14-trixie)
• Useful packages: htop, procps, iputils-ping, locales, vim, tini
• Python dependencies from requirements.txt and requirements-dev.txt
• Source code copied into /app and PYTHONPATH=/app
• Loguru-friendly environment with tini as entrypoint

Build arguments

• python_version – default 3.14
• debian_version – default trixie
• UID, GID, UNAME – user configuration in the image (defaults: 1234/1234/pythonuser)
• TARGETOS, TARGETARCH, TARGETPLATFORM – auto-populated by BuildKit/buildx for multi-arch
• gh_ref, gh_sha, buildtime – injected into environment variables (GITHUB_REF, GITHUB_SHA, BUILDTIME)
• forwarded_allow_ips – forwarded IPs for proxied setups, default *

Building the image

Basic build:

docker build -t mqttcommander:local .

Pass custom Python/Debian versions:

docker build \
  --build-arg python_version=3.12 \
  --build-arg debian_version=bookworm \
  -t mqttcommander:py312 .

Embed source metadata (useful in CI):

docker build \
  --build-arg gh_ref="${GITHUB_REF}" \
  --build-arg gh_sha="${GITHUB_SHA}" \
  --build-arg buildtime="$(date -u +%Y-%m-%dT%H:%M:%SZ)" \
  -t mqttcommander:with-meta .

Multi-architecture build with buildx (example for amd64 and arm64):

docker buildx build \
  --platform linux/amd64,linux/arm64 \
  -t ghcr.io/youruser/mqttcommander:latest \
  --push .

Note: This repository already provides a docker-config/ buildx context. You can reuse your existing builder or create a new one:

docker buildx create --name mbuilder --use || true
docker buildx inspect --bootstrap

Running the image

The image uses tini as entrypoint and defaults to a no-op tail -f /dev/null command, so you can exec into it or run your own command.

Examples:

# Run interactively and inspect
docker run --rm -it \
  -e LOGURU_LEVEL=INFO \
  -e MQTTSTUFF_CONFIG_DIR_PATH=/app \
  -v $(pwd)/config.yaml:/app/config.yaml:ro \
  mqttcommander:local bash

# Run a Python one-liner using the mqttstuff wrapper (installed as dependency)
docker run --rm -it mqttcommander:local \
  python -c "from mqttstuff.mosquittomqttwrapper import MQTTLastDataReader as R; print(R.get_most_recent_data_with_timeout('broker',1883,'user','pass',['tele/+/STATE'], retained='only'))"

Why Docker here is useful

• Ensures consistent Python/dependency versions across dev machines and CI
• Provides a preconfigured environment for quick experiments against an MQTT broker
• Makes multi-arch builds straightforward with buildx

Development

Helpful Makefile targets:

• make help – list available targets with short descriptions
• make install – create virtualenv and install development requirements
• make venv – ensure .venv exists and dev requirements are installed
• make tests – run pytest
• make lint – run Black code formatter
• make isort – fix and check import order
• make tcheck – run mypy type checks over *.py, scripts/, and mqttcommander/
• make commit-checks – run pre-commit hooks on all files
• make prepare – run tests and commit-checks (useful before committing/PRs)
• make pypibuild – build sdist and wheel with Hatch into dist/
• make pypipush – publish built artifacts with Hatch (configure credentials first)
• make build – build the Docker image via ./build.sh
• make dstart – start ephemeral container (host network), mapping config.local.yaml into /app

Testing

Tests live under tests/. Run all tests with:

pytest -q

License

This project is licensed under the LGPL where applicable/possible — see LICENSE.md. Some files/parts may be governed by other licenses and/or licensors, such as MIT | GPL | LGPL. Please also check file headers/comments.

Acknowledgments

See inline comments in the codebase for inspirations and references.

⚠️ Disclaimer

This is a development/experimental project. For production use, review security settings, customize configurations, and test thoroughly in your environment. Provided "as is" without warranty of any kind, express or implied, including but not limited to the warranties of merchantability, fitness for a particular purpose and noninfringement. In no event shall the authors or copyright holders be liable for any claim, damages or other liability, whether in an action of contract, tort or otherwise, arising from, out of or in connection with the software or the use or other dealings in the software. Use at your own risk.