Script JSON communication

This example shows how Script node can communicate with the outside world (host computer) using JSON serialization by sending a Buffer message. Similarly, you could use SPIIn and SPIOut to send JSON between MCU and Script node via SPI.How would this be useful:2-way communication between Script node and host/MCU could be used to eg. alter the flow of an application.For example, we might want to use 2 different NN models on the device, one until noon and the other from noon till midnight. Host (eg. RPi) could check the current time and when it's noon, it would send a simple message to the Script node which would start forwarding all frames to the other NeuralNetwork node (which has different NN model).What it does:Host creates a dictionary, serializes it, sends it to the Script node. Script node receives the Buffer message, deserializes the dictionary, changes values a bit, serializes the dictionary again and sends it to the host, which deserializes the changed dictionary and prints the new values.

Demo

Command Line

1~/depthai-python/examples/Script$ python3 script_json_communication.py
2dict {'one': 1, 'foo': 'bar'}
3[14442C1041B7EFD000] [3.496] [Script(1)] [warning] Original: {'one': 1, 'foo': 'bar'}
4[14442C1041B7EFD000] [3.496] [Script(1)] [warning] Changed: {'one': 2, 'foo': 'baz'}
5changedDict {'one': 2, 'foo': 'baz'}

Setup

Please run the install script to download all required dependencies. Please note that this script must be ran from git context, so you have to download the depthai-python repository first and then run the script

Command Line

1git clone https://github.com/luxonis/depthai-python.git
2cd depthai-python/examples
3python3 install_requirements.py

For additional information, please follow the installation guide.

Source code

Python

C++

Python

GitHub

1#!/usr/bin/env python3
2import depthai as dai
3import json
4
5pipeline = dai.Pipeline()
6
7xin = pipeline.create(dai.node.XLinkIn)
8xin.setStreamName('in')
9
10script = pipeline.create(dai.node.Script)
11xin.out.link(script.inputs['in'])
12script.setScript("""
13    import json
14
15    # Receive bytes from the host
16    data = node.io['in'].get().getData()
17    jsonStr = str(data, 'utf-8')
18    dict = json.loads(jsonStr)
19
20    # Change initial dictionary a bit
21    dict['one'] += 1
22    dict['foo'] = "baz"
23
24    b = Buffer(30)
25    b.setData(json.dumps(dict).encode('utf-8'))
26    node.io['out'].send(b)
27""")
28
29xout = pipeline.create(dai.node.XLinkOut)
30xout.setStreamName('out')
31script.outputs['out'].link(xout.input)
32
33# Connect to device with pipeline
34with dai.Device(pipeline) as device:
35    # This dict will be serialized (JSON), sent to device (Script node),
36    # edited a bit and sent back to the host
37    dict = {'one':1, 'foo': 'bar'}
38    print('dict', dict)
39    data = json.dumps(dict).encode('utf-8')
40    buffer = dai.Buffer()
41    buffer.setData(list(data))
42    device.getInputQueue("in").send(buffer)
43
44    # Wait for the script to send the changed dictionary back
45    jsonData = device.getOutputQueue("out").get()
46    jsonText = str(jsonData.getData(), 'utf-8')
47    changedDict = json.loads(jsonText)
48    print('changedDict', changedDict)

Pipeline

React Flow

Need assistance?

Head over to Discussion Forum for technical support or any other questions you might have.

ON THIS PAGE

Script JSON communication

Demo

Setup

Source code

Python

Pipeline

Need assistance?