Summary

This PR adds per-query query tags functionality that allows users to pass tags as a dictionary to the execute.*() methods.

Changes

• Added serialize_query_tags() utility function in utils.py to convert dict to the required format "key1:value1,key2:value2"
• Added query_tags parameter to Cursor.execute() and Cursor.execute_async() methods
• Updated DatabricksClient interface and ThriftDatabricksClient implementation
• Query tags are serialized and passed via internal confOverlay with key "query_tags"
• Added comprehensive unit tests for query_tags serialization (8 new tests)
• Updated query_tags example to demonstrate new per-query tags usage

Key Features

• None value handling: If value is None, omits colon and value (e.g., "key1:value1,key2,key3:value3")
• Special character escaping: Escapes :, ,, \ in values with leading backslash
• Keys not escaped: Assumes keys are controlled identifiers

Usage Example

cursor.execute(
    "SELECT * FROM table",
    query_tags={"team": "data-eng", "application": "etl"}
)

Testing

• All existing tests pass (122/122 client tests, 66/66 thrift backend tests)
• Added 8 new unit tests for query_tags serialization
• All tests pass with no regressions

The branch is ready and all tests are passing! 🎉

What type of PR is this?

• Refactor
• Feature
• Bug Fix
• Other

Description

How is this tested?

• Unit tests
• E2E Tests
• Manually
• N/A

Related Tickets & Documents