spark_read_from_topic_and_show.py

from pyspark.context import SparkContext
from pyspark.sql.session import SparkSession
sc = SparkContext('local')
spark = SparkSession(sc)

TOPIC_NAME = 'SAMPLE_TOPIC_NAME'
df = spark \
    .read \
    .format("kafka") \
    .option("kafka.bootstrap.servers", "172.25.0.12:9092,172.25.0.13:9092") \
    .option("startingOffsets", "earliest") \
    .option("subscribe", TOPIC_NAME) \
    .load()

df1 = df.selectExpr("CAST(key AS STRING)", "CAST(value AS STRING)")

df1.show()