Wires the TCPLoRaRadio and USBLoRaRadio drivers that landed in pyMC_core
on 2026-05-13 (PR pyMC-dev/pyMC_core#68) into get_radio_for_board() so
they can be selected from a repeater config file without any code change
in main.py / api_endpoints.
Both branches follow the existing pattern: read host/port (TCP) or
serial port (USB) plus auth/LBT options from their own config section,
share the LoRa parameters from the common `radio` section, fall back to
the firmware-default sync word 0x12, and surface ImportError as a clear
RuntimeError if the installed pymc_core is too old to ship the drivers.
config.yaml.example documents both sections and updates the radio_type
header comment with the full supported list. Five new tests in
tests/test_radio_config.py monkeypatch the radio classes and verify the
section/parameter wiring + missing-required-field errors.
No web UI / endpoint changes — the deployment this targets edits the
config file directly. A GUI wizard for these radio types can land
separately if there's appetite.
Bit 0 of byte 2 is the T_VALID/H_VALID flag (datasheet Figure 32/33,
page 23 example: t_valid = (t_val>>16) & 0x1). The previous code checked
bit 1 (CRC LSB), which caused sporadic timeouts when the CRC happened to
have a 0 in that position.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Use multi-line ensure_python_modules list format in ens210.py,
matching the established pattern from ina219.py
- Fix auto_install_packages indentation in ina219.py docstring
- Remove smbus2 from pyproject.toml core dependencies; sensor packages
are handled at runtime via ensure_python_modules/auto_install_packages
- Update docs/adding_sensors.md template and guidance to match
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Add smbus2>=0.4.0 to pyproject.toml core dependencies so it is always
present in the venv rather than relying on runtime auto-install
- Add i2c-tools to apt-get installs in both install and upgrade paths
so /dev/i2c-* devices are accessible and i2cdetect is available for
diagnostics (service user was already being added to the i2c group)
- Switch ENS210 config examples to hex I2C address notation (0x43) to
match datasheets and i2cdetect output; update contributor docs guidance
accordingly
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds support for the ENS210 relative humidity and temperature sensor
as a new plug-in under repeater/sensors/ens210.py. Also adds a
commented configuration example to config.yaml.example and a
contributor guide at docs/adding_sensors.md explaining how to add
further sensor plug-ins.
## Implementation notes
### Why smbus2 instead of an Adafruit/CircuitPython library
The ENS210 has no maintained Adafruit CircuitPython driver. The
available third-party options are either unmaintained or bring in the
full Blinka/CircuitPython hardware-abstraction stack as a dependency.
smbus2 is a thin, widely-packaged wrapper around the Linux i2c-dev
kernel interface that is already present on Raspberry Pi OS and most
Debian-based systems. It has no transitive dependencies and adds no
abstraction cost.
The ENS210 protocol is simple enough that direct register access is
preferable: two writes to start a measurement (REG_SENS_RUN + REG_SENS_START)
and two three-byte block reads to retrieve temperature and humidity.
The status/validity bit is checked inline rather than relying on a
library to surface it. There is no value a higher-level driver would
add here.
### Read strategy
A fixed post-trigger delay is unreliable — the sensor datasheet quotes
~130 ms typical conversion time but the actual ready time varies. The
implementation instead polls the data-valid status bit (bit 1 of the
third byte in each register block) every 50 ms for up to
read_timeout_seconds (default 1.0 s), breaking as soon as both T and
H report valid data. This is the same approach used in the validated
reference script.
The I2C bus is opened and closed on every read rather than kept open
across poll cycles. Keeping a persistent SMBus handle caused subsequent
reads to time out, consistent with the Linux i2c-dev file descriptor
accumulating state between transactions.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Every MQTT-published packet has shipped with duration="0" since the
PacketRecord factory was introduced. The repeater already computes LoRa
time-on-air via AirtimeManager.calculate_airtime() (the canonical
Semtech reference formula) for duty-cycle gating and TX delay, but the
result was thrown away after each packet - never stored on the
packet_record dict that flows to MQTT/SQLite/Glass/websocket.
What changes
- engine.py: RepeaterHandler._build_packet_record() now computes
airtime_ms once per packet (Semtech formula via AirtimeManager) and
stores it as packet_record['airtime_ms']. Single source of truth for
every downstream consumer.
- storage_utils.py: PacketRecord.from_packet_record() reads the new
airtime_ms field and serializes it as a rounded integer in the
'duration' field of the published JSON. Falls back to 0 if the field
is missing (backward compatibility for any older code path).
- storage_collector.py: _publish_packet_to_mqtt() simplified - no
recomputation, no helper. The publish path is now a passthrough.
Why
MQTT consumers (firmware-compatible analyzers, dashboards, the upstream
meshcoretomqtt project) expect the same time-on-air value the firmware
emits. Hard-coded "0" makes airtime/utilization charts derived from the
mqtt stream useless and silently diverges from firmware behavior.
Plumbing the value through packet_record (instead of recomputing in the
publish path) means any future consumer - SQLite schema, web UI charts,
Glass telemetry - reads the same number without separate calculations.
Tests
tests/test_packet_duration.py - 5 tests covering:
- backward compat (legacy packet_record without airtime_ms => '0')
- airtime_ms field flows through to duration as rounded integer string
- explicit zero stays '0'
- AirtimeManager output matches an independently-implemented Semtech
reference for typical MeshCore EU settings (SF8/62.5kHz/CR4-8)
- low-data-rate optimization branch (SF12/125kHz triggers DE=1)
Co-Authored-By: Oz <oz-agent@warp.dev>
Introduces a 'set format and forget' workflow for MQTT brokers. Users
reference a bundled preset by name inside the existing brokers: list,
and the package supplies the endpoints, audiences, and TLS settings.
Endpoint changes ship via 'pip install -U' instead of manual edits.
What changes
- New repeater/presets/ package with a tiny lazy YAML loader and two
bundled presets: waev (mqtt-{a,b}.waev.app) and letsmesh (EU + US).
- New format-family constant MC2MQTT_FORMATS = ('meshcoretomqtt',
'letsmesh', 'waev') replaces the inline tuple in topic resolution.
The legacy 'mqtt' format keeps its custom-topic semantics unchanged.
- Two-pass broker assembly in mqtt_handler.py: pass 1 expands every
{preset: <name>} entry inline; pass 2 collapses duplicates by name
with later-wins semantics. Place override entries AFTER preset
entries.
- Hard-coded LETSMESH_BROKERS constant deleted; its data now lives in
repeater/presets/letsmesh.yaml.
- convert_letsmesh_to_broker_config() collapsed from ~70 to ~25 lines
by emitting {preset: letsmesh} plus disable overrides for unwanted
brokers. Honors broker_index in (-1, 0, 1), additional_brokers, and
enabled flag exactly as before.
- update_mqtt_config API endpoint accepts {preset: <name>} entries and
passes them through unchanged so the web UI can author them when the
frontend is updated.
- config.yaml.example documents the preset entry shape, the override
rule, and the format family hierarchy.
- pyproject.toml ships presets/*.yaml as package data.
How to use
mqtt_brokers:
iata_code: "LAX"
brokers:
- preset: waev
# Override a single preset broker:
brokers:
- preset: waev
- name: waev-b
enabled: false
Tests
- tests/test_presets.py: 9 tests covering loader, expand/merge,
MC2MQTT topic-family parity, and parametrized legacy migration.
Co-Authored-By: Oz <oz-agent@warp.dev>
yellowcooln