iot_devices.py

#!/usr/bin/python3

# imports
from kafka import KafkaProducer # pip install kafka-python
import numpy as np              # pip install numpy
from sys import argv, exit
from time import time, sleep

# different device "profiles" with different 
# distributions of values to make things interesting
# tuple --> (mean, std.dev)
DEVICE_PROFILES = {
	"boston": {'temp': (51.3, 17.7), 'humd': (77.4, 18.7), 'pres': (1019.9, 9.5) },
	"denver": {'temp': (49.5, 19.3), 'humd': (33.0, 13.9), 'pres': (1012.0, 41.3) },
	"losang": {'temp': (63.9, 11.7), 'humd': (62.8, 21.8), 'pres': (1015.9, 11.3) },
}

# check for arguments, exit if wrong
if len(argv) != 2 or argv[1] not in DEVICE_PROFILES.keys():
	print("please provide a valid device name:")
	for key in DEVICE_PROFILES.keys():
		print(f"  * {key}")
	print(f"\nformat: {argv[0]} DEVICE_NAME")
	exit(1)

profile_name = argv[1]
profile = DEVICE_PROFILES[profile_name]

# set up the producer
producer = KafkaProducer(bootstrap_servers='localhost:9092')

count = 1

# until ^C
while True:
	# get random values within a normal distribution of the value
	temp = np.random.normal(profile['temp'][0], profile['temp'][1])
	humd = max(0, min(np.random.normal(profile['humd'][0], profile['humd'][1]), 100))
	pres = np.random.normal(profile['pres'][0], profile['pres'][1])
	
	# create CSV structure
	msg = f'{time()},{profile_name},{temp},{humd},{pres}'

	# send to Kafka
	producer.send('weather', bytes(msg, encoding='utf8'))
	print(f'sending data to kafka, #{count}')

	count += 1
	sleep(.5)