Explanation:
In Databricks Auto Loader, setting the schema evolution mode to 'rescue' (or using the default behavior which includes a rescued data column) captures any unexpected columns, mismatched data types, or malformed data into a dedicated JSON column (usually _rescued_data). This prevents data loss and stream failures while handling unexpected schema changes.
Ultimate access to all questions.
A
Ignored and dropped.
B
Saved in a dedicated rescued data column.
C
Instantly added to the target schema.
D
Rejected and causes a stream failure.
No comments yet.