Quickstart¶
From install to live observations in about thirty lines. This assumes a running OSH node with the Connected Systems API and MQTT services enabled — see Prerequisites if you don’t have one yet.
Install the current alpha (details and uv equivalents in Installation):
pip install --pre oshconnect
Connect, discover, and stream:
import time
from oshconnect import OSHConnect, Node, StreamableModes
# one app, one node — adjust host/port/credentials to your OSH server
app = OSHConnect(name='QuickstartApp')
node = Node(protocol='http', address='localhost', port=8585,
username='admin', password='admin',
enable_mqtt=True, mqtt_port=1883)
app.add_node(node)
# find every system and datastream the node serves
app.discover_systems()
app.discover_datastreams()
# start real-time streaming on each datastream
for ds in app.get_datastreams():
ds.set_connection_mode(StreamableModes.PULL)
ds.initialize()
ds.start()
# observations accumulate on each datastream's inbound deque
time.sleep(2)
for ds in app.get_datastreams():
while ds.get_inbound_deque():
print(ds.get_inbound_deque().popleft())
That’s the whole read path: one OSHConnect app, one Node with the
MQTT transport enabled, discovery, then a deque of live observations per
datastream.
Where to go next¶
Connecting to a Node — nodes, transports (MQTT / NATS), and authentication in depth
Consuming Sensor Data — discovery, streaming, topic conventions, and event subscriptions
Publishing Sensor Data — register your own systems and datastreams and publish observations
Commanding Systems — control streams, commands, and status
Architecture — how the pieces fit together