May 05-12, 2025 — 9.12 Release Notes (with behavior changes)¶

Release channels for Snowflake Native Apps — General availability¶

With this release, the release channels feature in Snowflake Native Apps is generally available.

Release channels allow providers to publish apps at different stages of the app development lifecycle. For example, a provider can use release channels to perform the following tasks for a version or patch of an app:

• Test an app.

• Publish an app to consumers as a preview or for UAT (user acceptance testing).

• Publish the app to a production environment.

For more information, see Publish an app using release channels.

Improved error messages for Data Manipulation Language (DML) commands¶

In past releases, error messages for DML commands didn’t include the column name for errors that involved a specific column. With this release, some error messages for DML commands include the column name. Note that the column name isn’t included in all DML error messages.

For example, the following SQL statements return a DML error message:

CREATE OR REPLACE TABLE demo_dml_error_message (v VARCHAR);

INSERT INTO demo_dml_error_message (v) VALUES
  (3),
  ('d');

In past releases, the following error message was returned:

100038 (22018): Numeric value 'd' is not recognized

With this release, the following error message is returned:

100038 (22018): DML operation to table DEMO_INSERT_TYPE_MISMATCH failed on
column V with error: Numeric value 'd' is not recognized

New SQL functions¶

The following functions are now available with this release:

Built-in code profiler for Python stored procedures — General availability¶

With this release, we are pleased to announce the general availability of built-in code profiling for stored procedure handler code written in Python. Using the profiler, you can discover how much time or memory was spent executing your handler code. The profiler generates information describing how much time or memory was spent executing each line of the procedure handler.

For procedures written in SQL, see Profiling Python procedure handler code.

For procedures written with the Snowpark API, see Profiling Snowpark Python stored procedure handlers.

Support for internal stage cloning — General availability¶

With this release, we are pleased to announce the general availability of support for internal stage cloning when you clone a database or schema.

For more information, see CREATE <object> … CLONE.

Vectorized scanner now available without ON_ERROR restrictions¶

Previously, enabling the vectorized scanner required the ON_ERROR option to be set to either ABORT_STATEMENT or SKIP_FILE. This limitation has been removed.

You can now leverage the performance benefits of the vectorized scanner regardless of the ON_ERROR setting you choose, including CONTINUE, SKIP_FILE_num, and 'SKIP_FILE_num%'. This provides greater flexibility in configuring your data loading processes while still taking advantage of optimized scanning.

For more information, see USE_VECTORIZED_SCANNER.

Sensitive data classification: New classifiers for India¶

The following sensitive data classifiers now support the protection of sensitive data in India:

• NATIONAL_IDENTIFIER (Permanent account number (PAN), Aadhaar, and Voter ID)

• DRIVERS_LICENSE

• TAX_IDENTIFIER (Goods and Service Tax Identification Number (GSTIN))

Using caller’s rights to connect to Snowflake — General availability¶

With this release, we are pleased to announce the general availability of connecting to Snowflake from inside a container using caller’s rights.