client_twitter.py

from pyspark.sql import SparkSession as ss
from pyspark.sql import functions as f

spark = ss.builder.appName('SparkStreaming').getOrCreate()

lines = spark.readStream.format('socket').option('host', 'localhost').option('port', 9009).load()

words = lines.select(f.explode(f.split(lines.value, ' ')).alias('word'))

wordCounts = words.groupBy('word').count()

query = wordCounts.writeStream.outputMode('complete').format('console').start()

query.awaitTermination()