MySQL HeatWave GenAI includes a new schema retrieval routine that enables you to find the most relevant tables based on natural-language input. This innovative feature provides a compact and easy-to-read DDL output, focusing on essential elements such as column names, simplified data types, comments, and basic foreign keys, making it ideal for large language model (LLM) prompts and related workflows.

For more information, see ML_RETRIEVE_SCHEMA_METADATA. (WL #17201)

MySQL HeatWave Lakehouse now supports the inference and loading of unique columns that are matched by name. With the introduction of the allow_missing_columns parameter, you can choose to disable or enable the ability to include all unique columns by name found during inference. When you enable allow_missing_columns, MySQL HeatWave Lakehouse loads all unique columns even if they are not present in every file, and loads data with missing or extra columns without errors. This feature enhances the overall data loading experience, providing a more robust and adaptable solution for handling diverse data sets.

For more information, see Allow Missing Columns Option. (WL #17165)