Exam Certified Associate Developer for Apache Spark topic 1 question 43 discussion

B is correct. 'storeDF' has not attribute or method `sql` Test code below: from pyspark.sql import SparkSession spark = SparkSession.builder.appName("MyApp").getOrCreate() data = [ (0, 3, "John"), (1, 1, "Jane"), (2, 2, "Jack"), ] storesDF = spark.createDataFrame(data, ["storeID", "customerSatisfaction", "managerName"]) storesDF.createOrReplaceTempView("stores") try: storesDF.sql("SELECT storeId, managerName FROM stores") except AttributeError as e: print(e) finally: spark.sql("SELECT storeId, managerName FROM stores").show()

B is correct: storesDF = spark.createDataFrame([('1', 'juan'), ('2', 'perez')], ['storeId', 'managerName']) storesDF.createOrReplaceTempView("stores") spark.sql("SELECT storeId, managerName FROM stores").show()

Option B is correct because the sql() function is not a method of a DataFrame object. It is actually a method of the SparkSession object spark. Therefore, the correct way to execute a SQL statement using Spark SQL is to call sql() on the SparkSession object as follows: spark.sql("SELECT storeId, managerName FROM stores") In the code block provided in the question, sql() is called on a DataFrame object, which will result in a DataFrame object without executing the SQL statement. Therefore, option B correctly identifies the error in the code block.