Question 51

You are building a **Lakeflow Spark Declarative Pipeline** that ingests JSON files from a Unity Catalog volume into a bronze **streaming table** using Auto Loader. You author the following SQL: ```sql CREATE OR REFRESH STREAMING TABLE bronze_orders AS SELECT * FROM STREAM read_files( '/Volumes/main/raw/landing/orders', format => 'json', inferColumnTypes => true ); ``` Which **three** statements are true about this declarative streaming-table ingestion pattern? (Choose THREE.)

• AThe `STREAM` keyword and the `read_files` table-valued function together invoke Auto Loader (`cloudFiles`) functionality for incremental file ingestion.
• BWhen `read_files` is used inside a `CREATE OR REFRESH STREAMING TABLE` statement in a pipeline, Lakeflow automatically configures and manages the checkpoint and schema locations.
• CYou must manually call `.option("checkpointLocation", ...)` and `.writeStream` for this streaming table, because pipelines do not manage triggers or checkpoints.
• DTo load files with Auto Loader in a Unity Catalog-enabled pipeline, the source path must be governed by a Unity Catalog external location (or a volume).
• EStreaming tables are designed for append-only sources; new rows are inserted into the target with each pipeline update.
• F`read_files` can only be used in batch materialized views and is not supported for streaming tables.