Databricks Certified Data Engineer - Associate

Explanation

The correct answer is B. spark.table("sales").

Here's why:

1. spark.table("sales") is the standard PySpark method to access a table registered in the Spark catalog. This method returns a DataFrame representing the table, which can then be used for data processing, transformations, and testing in Python.

2. Option A (SELECT * FROM sales) is incorrect because this is SQL syntax, not PySpark Python code. While you could use spark.sql("SELECT * FROM sales") to execute SQL in PySpark, the option shows only the SQL statement without the spark.sql() wrapper.

3. Option C (spark.sql("sales")) is incorrect because spark.sql() expects a complete SQL query string, not just a table name. This would result in a syntax error.

4. Option D (spark.delta.table("sales")) is incorrect because there is no spark.delta.table() method in PySpark. While Databricks provides Delta Lake functionality, the correct way to access a Delta table is through spark.table() or spark.read.format("delta").table("sales").

Additional context:

• In PySpark, once a table is registered in the Spark catalog (which happens automatically when creating Delta tables in Databricks), it can be accessed using spark.table("table_name").
• The data engineering team can use this DataFrame to implement data quality tests in Python using PySpark's DataFrame API.
• This approach allows them to leverage Python's testing frameworks while working with the same Delta table that the analytics team uses.