MQTT samples

[s-celles]

MQTT sample should be add in this repository.
subscriber should be able to display data in console or plot matplotlib graph or pyqtgraph.
see https://github.com/scls19fr/numpy-buffer/tree/master/samples
vispy may also be considered http://vispy.org/

If we want to publish SenseHat sensors state on several topics, this could help

# import mock
# sense = mock.Mock() # Mock SenseHat for testing

from collections import OrderedDict

# Define a dictionary for sensors
#   key: sensor name
#   value: a "callable" to get value of a sensor
d_sensors = OrderedDict([
    ('pixels', sense.get_pixels),
    ('humidity', sense.get_humidity),
    ('temperature', sense.get_temperature_from_humidity),
    ('temperature_from_pressure', sense.get_temperature_from_pressure),
    ('pressure', sense.get_pressure),
    ('orientation', sense.get_orientation),
    ('compass', sense.get_compass),
    ('gyroscope', sense.get_gyroscope),
    ('accelerometer', sense.get_accelerometer),
])

sensors = d_sensors.keys() # all sensors
#sensors = ['temperature', 'pressure']

for sensor in sensors:
    get_value = d_sensors[sensor] # get a "callable"
    print(sensor, get_value)

    data = {
        'ts': now.isoformat(),
        'd': {
            sensor: get_value()
        }
    }
    payload = json.dumps(data) # serialization
    cli.publish(topic='/sensors/SenseHat01/%s' % sensor, payload=payload, qos=0, retain=False)

    ...

We can (should?) also consider that we have only one topic per SenseHat and all sensors state are publish on the same topic so data (d) should be create accordingly.

d = {sensor: get_value() for sensor, get_value in d_sensors.items()}
data = {
    'ts': now.isoformat(),
    'd': d
}
payload = json.dumps(data) # serialization
cli.publish(topic='/sensors/SenseHat01', payload=payload, qos=0, retain=False)