Your Snowflake Native App includes a JavaScript UDF that processes data from a secure view provided by the provider account. During consumer testing, the UDF fails with an 'Authorization error' even though the consumer has been granted the necessary USAGE privileges on the application package. What is the MOST likely cause of this error, and how can you resolve it within the context of a Snowflake Native App?

• A. The application package needs to be explicitly authorized to access secure views in the provider account. This requires a manifest update and resubmission of the package.
• B. The consumer lacks the necessary permissions on the underlying tables referenced by the secure view in the provider account. Grant the consumer direct access to these tables.
• C. The secure view in the provider account has data masking policies applied that prevent the UDF from accessing the data. Remove the masking policies.
• D. The UDF is not declared as SECURE, preventing it from accessing objects in a different security context. Declare the UDF as SECURE.
• E. The consumer's role does not have the OWNERSHIP privilege on the application object. Assign the OWNERSHIP privilege to the consumer's role.

An application provider wants to grant the 'EXECUTE MANAGED TASK' privilege to a consumer account for a specific managed task within their Snowflake Native App. The provider uses the following SQL statement:

However, when the consumer account attempts to execute the managed task, they encounter an error indicating insufficient privileges. Which of the following reasons could explain why the consumer account is unable to execute the task, and how could the provider resolve the issue? Select two correct answers.

• A. The 'EXECUTE MANAGED TASK' privilege needs to be granted with the 'WITH GRANT OPTION' clause to allow the consumer to further grant the privilege to other roles within their account.
• B. The consumer account needs to explicitly grant the 'USAGE' privilege on the application database and schema to the application role 'app_public'. The provider needs to provide SQL to facilitate this in their setup script for consumer.
• C. The managed task owner (the application) does not have the OWNERSHIP privilege on the database and schema where the managed task resides. The application must have OWNERSHIP on these objects.
• D. The application role 'app_public' might not be activated within the consumer account's session. The consumer needs to ensure the application role is active using 'USE ROLE app_public;' before attempting to execute the managed task.
• E. The 'EXECUTE MANAGED TASK' privilege must be granted directly to the consumer account's user role, not the application role. The provider needs to identify the user role used by the consumer and grant the privilege to that specific role.

You've deployed a Snowflake Native Application to the Marketplace. A consumer reports an issue, and after investigating, you suspect a vulnerability related to SQL injection within a stored procedure. You want to implement best practices to mitigate this risk. Which of the following strategies are recommended to secure the stored procedure against SQL injection?

• A. Encode all user input before inserting it into SQL queries to neutralize potentially harmful characters.
• B. Grant 'OWNERSHIP' privilege to the stored procedure to prevent unauthorized modifications.
• C. Implement strict input validation and sanitization to ensure that user input conforms to expected data types and patterns.
• D. Use parameterized queries or prepared statements whenever possible to prevent user input from being directly interpreted as SQL code.
• E. Disable all external network access from within the stored procedure environment.

You are developing a Snowflake Native Application that performs complex data transformations. As part of your development lifecycle, you need to ensure that your code behaves idempotently even when facing intermittent network errors or unexpected disruptions. Consider the following function written in Snowpark Python that updates a table:

What modification is necessary to make this function idempotent against disruptions during the writing process to Snowflake and unexpected execution?

• A.
  
• B. Wrap the 'df.write' operation in a retry mechanism with exponential backoff to handle intermittent network errors.
• C. Create a temporary table, write the transformed data to the temporary table, and then atomically rename the temporary table to the original table name.
• D. Leverage Snowflake's 'TRANSACTIONAL_REPLICATION' to replicate the operations to minimize data loss in between failures.
• E. Use a 'MERGE statement to selectively update only changed rows, while ignoring existing ones.