I have a very large CSV file. 75+ million rows.
I have to "Publish" this .csv file into a Redis Cluster once every hour. (hourly)
Script:
import csv
import redis
import random
from redis import StrictRedis
import multiprocessing as mp
import itertools
import time
def worker(chunk):
return len(chunk)
def keyfunc(row):
return row[0]
def main():
client = redis.StrictRedis(host='XXXXXXXXX.XX.XXX.X.XXX', port=6379, db=0)
client1 = redis.StrictRedis(host='XXXXXXXXX.XX.XXX.X.XXX', port=6379, db=0)
client2 = redis.StrictRedis(host='XXXXXXXXX.XX.XXX.X.XXX', port=6379, db=0)
list1 =(client, client1, client2)
pool = mp.Pool()
largefile = 'Example.csv'
num_chunks = 10
results = []
with open(largefile) as f:
reader = csv.reader(f)
chunks = itertools.groupby(reader, keyfunc)
while True:
# make a list of num_chunks chunks
groups = [list(chunk) for key, chunk in
itertools.islice(chunks, num_chunks)]
if groups:
result = pool.map(worker, groups)
results.extend(result)
else:
break
key1 = 'AAM_CDF_Traits'
doc = chunk
random.choice(list1).publish(key1, pool)
pool.close()
pool.join()
print(results)
if __name__ == '__main__':
main()
Questions:
Is this the right approach to take to solve this problem? What other ways could I solve this problem.
Why do I have this error?
Traceback (most recent call last):
File "./AAM_Redis4.sh", line 47, in
main()
File "./AAM_Redis4.sh", line 33, in main
itertools.islice(chunks, num_chunks)]
TypeError: 'tuple' object is not callabl