async def f():
# source = Stream(asynchronous=True)
# source.rate_limit(0.500).sink(write)
# for x in range(10):
# await source.emit(x)
kafkaConfig = {'bootstrap.servers': 'localhost:9092', 'group.id': 'streamz-minas'}
topic = 'test'
kafkaSource = Stream.from_kafka([topic], kafkaConfig, start=True, asynchronous=True)
results = {}
def asSink(name):
results[name] = []
results[name + 'timestamp'] = time.time_ns()
results[name + 'elapsed'] = time.time_ns()
def f(val):
results[name].append(val)
results[name + 'elapsed'] = time.time_ns() - results[name + 'timestamp']
print(name, val)
return f
print('kafkaSource map')
kafkaSource.map(combinedMap).sink(asSink('raw'))
kafkaSource.visualize()
while len(results['raw']) < 200:
print('len(results[raw])', len(results['raw']))
await gen.sleep(1)
print(results)
avg = (sum(timeDiffs) / len(timeDiffs)) * 10**(-9)
print(avg, 'avgs', 1/avg, 'item/s')
async def minas_consumer_streamz():
kafkaSource = Stream.from_kafka([topic],
consumer_params=kafkaConfig,
start=True,
asynchronous=False).buffer(5)
kprod = confluent_kafka.Producer(kafkaConfig)
def combinedMap(jsonItem):
di = json.loads(jsonItem)
if not 'example' in di:
return
example = Example(**di['example'])
timeDiff = time.time_ns() - example.timestamp
return (minDist(clusters, example.item), timeDiff)
results = {}
def asSink(name):
results[name] = []
results[name + 'timestamp'] = time.time_ns()
results[name + 'elapsed'] = time.time_ns()
def f(val):
results[name].append(val)
results[name +
'elapsed'] = time.time_ns() - results[name + 'timestamp']
value = json.dumps({'d': val[0][0], 'label': val[0][1].label})
kprod.produce(topic=topic + '_out', value=value)
print(name, val)
return f
print('kafkaSource map')
outStream = kafkaSource.map(combinedMap).sink(asSink('raw'))
outStream = kafkaSource.scatter().map(combinedMap).gatter().sink(
asSink('raw'))
while True:
await gen.sleep(1000)
import json
from dask.distributed import Client
from time import sleep
import random
from time import time
#dask
client = Client('35.180.242.51:8786')
import os
client.run(lambda: os.system("pip install cassandra-driver"))
print('befor upload')
#streamz
source = Stream.from_kafka(['supramoteur'], {
'bootstrap.servers': '35.180.242.51:9092',
'group.id': 'mygroup1'
},
loop=client.loop)
def time_final(entry):
tt = time() - entry['time']
return tt
def ecrire(entry):
entry['time_before'] = time()
with open('data.txt', "a+") as f:
f.write("%s\r\n" % entry)
insert_to(entry)
entry['time_after'] = round(-entry['time'], 4)
