BigQuery has announced a significant enhancement to its search indexing capabilities, now supporting INT64 and TIMESTAMP data types in search index creation. This feature has reached general availability (GA) status.

New Capabilities

• Support for INT64 data type indexing
• Support for TIMESTAMP data type indexing
• Optimized predicate processing
• Generally Available status

Supported Data Types

The complete list of supported data types now includes:

• STRING (primitive)
• INT64 (primitive)
• TIMESTAMP (primitive)
• ARRAY
• STRUCT (with supported nested fields)
• JSON (with supported nested fields)

Implementation Details

Users can create search indexes using the following syntax:

CREATE SEARCH INDEX [IF NOT EXISTS] index_name
ON table_name({ALL COLUMNS | column_name [, ...]})
[OPTIONS(index_option_list)]

Key Features

• Optional 'IF NOT EXISTS' clause
• Support for multiple column indexing
• Table-level index management
• Flexible options configuration

This update enhances BigQuery's search capabilities, providing more efficient querying options for numerical and temporal data.