October 28-30, 2024 — 8.41 Release Notes¶

Outbound private connectivity for Snowflake features¶

By default, Snowflake features that generate outbound network traffic from Snowflake to a cloud platform traverse the public Internet. With this release, we are pleased to announce that you can create private endpoints in Snowflake to access the cloud platform using the platform’s private connectivity solution rather than the Internet. This lets you access cloud platform services privately and securely from Snowflake.

With this release, outbound private connectivity is now available for the following Snowflake features:

EXECUTE IMMEDIATE FROM: Preview SQL rendered from Jinja2 templates¶

With this release, we are pleased to announce support for previewing the SQL statements rendered by Jinja2 templates.

If you are using the EXECUTE IMMEDIATE FROM command to render and execute SQL statements from a Jinja2 template, you can preview the rendered statements without executing them by specifying DRY_RUN=TRUE. This parameter is useful for debugging templating code and for previewing SQL statements from staged files not intended for execution.

For more information, see EXECUTE IMMEDIATE FROM.

GENERATE_SYNTHETIC_DATA: New system stored procedure for generating synthetic data — Preview¶

With this release, we are pleased to announce the preview of the GENERATE_SYNTHETIC_DATA system stored procedure. With the synthetic data generation feature, you can now programmatically create realistic datasets that closely mirror your original data. This allows you to represent sensitive, confidential, or restricted information across various workloads, such as testing and validation.

For more information, see GENERATE_SYNTHETIC_DATA.

Increased limits for network policies on internal stages¶

With this release, we are pleased to announce the general availability of using network policies to restrict incoming network traffic to the internal stages of AWS accounts.

For Business Critical or higher customers, this release increases the limits on how many network identifiers can be included in a network policy.

For more information, see Protecting internal stages on AWS.

Extended support for bind variables¶

You can use bind variables to replace literals in SQL statements, which allows applications to dynamically construct SQL statements based on user input. Bind variables are commonly used with Snowflake drivers, Snowflake Scripting, and the SQL REST API.

With this release, Snowflake extends support for bind variables so that you can use them for more use cases. The extended support includes the use of bind variables for stage names and other parameters in COPY INTO <table> statements.

For more information, see a Example that uses bind variables to set parameters in a command.

Writing files from Snowpark Python UDFs and UDTFs — Preview¶

With this release, we are pleased to announce the preview of Writing files from Snowpark Python UDFs and UDTFs. With this feature in Snowpark Python, you can now write files to stages using user-defined functions (UDFs), vectorized UDFs, user-defined table functions (UDTFs), and vectorized UDTFs.

For more information, see Writing files from Snowpark Python UDFs and UDTFs.