Databricks Certified Data Engineer - Associate

The correct answer is B. They need to add a CREATE LIVE TABLE table_name AS line at the beginning of the query.

Here's the detailed explanation:

In Delta Live Tables (DLT), you define tables using SQL by prefixing standard SQL queries with DLT-specific declarations. The basic syntax for creating a table is:

CREATE LIVE TABLE table_name
AS
SELECT * FROM json.`/path/to/json/file.json`;

CREATE LIVE TABLE table_name
AS
SELECT * FROM json.`/path/to/json/file.json`;

Let's examine why the other options are incorrect:

• A. They need to add a COMMENT line at the beginning of the query. - While comments are good practice, they are not required for the query to function in DLT.
• C. They need to add a live. prefix prior to json. in the FROM line. - The live. prefix is used to reference other tables within the same DLT pipeline, not to read external data sources.
• D. They need to add a CREATE DELTA LIVE TABLE table_name AS line at the beginning of the query. - The correct keyword is CREATE LIVE TABLE, not CREATE DELTA LIVE TABLE.
• E. They need to add the cloud_files(...) wrapper to the JSON file path. - While cloud_files is an excellent function for autoloader-style ingestion and is commonly used in DLT pipelines for incremental processing, it's not strictly required to convert this specific query. The original query using the json. data source will work in DLT when wrapped with the proper CREATE LIVE TABLE syntax. The cloud_files wrapper would be an enhancement for better pipeline performance and incremental processing, but the fundamental change needed is the DLT table creation syntax.

The core requirement to convert any standard Databricks SQL query for use in a DLT pipeline is to add the appropriate DLT declaration (CREATE LIVE TABLE..., CREATE LIVE VIEW..., etc.) at the beginning.*