Question 120
For a university data platform, this issue must not recur: A bulk load of customer CSV records splits addresses into wrong columns because the delimiter differs from the source contract. Which review control fits best?
Explanation
Correct Answer
Use sample files with quoted delimiters and escaped values in ingestion acceptance tests. This reduces recurrence of: CSV parsing depends on the declared file format, not just a .csv extension. The verification should demonstrate: A representative record maps its fields and header correctly without column shifts.
Incorrect Answers
Audit cloud trust relationships and allowed stage prefixes before onboarding new data. That safeguard is aimed at ANOTHER risk: A data lake S3 bucket is accessed using embedded access keys in a COPY script shared with contractors. It is less targeted than the control for: CSV parsing depends on the declared file format, not just a .csv extension.
Validate representative nested arrays, nulls and malformed JSON before production loads. That safeguard is aimed at ANOTHER risk: An engineer sets JSON FILE FORMAT to CSV and loads a nested event record into a one-column raw table. It is less targeted than the control for: CSV parsing depends on the declared file format, not just a .csv extension.
Monitor event-to-commit delay and producer file shapes. That safeguard is aimed at ANOTHER risk: A Snowpipe ingestion shows increasing time-to-availability after upstream sends huge compressed archives once daily. It is less targeted than the control for: CSV parsing depends on the declared file format, not just a .csv extension.