Documentation

Import Columns in Blend Data feature expands the capabilities of the Blend Data operator, allowing users to bulk add columns from existing tables or previous steps directly. This feature provides greater efficiency and streamlines data integration during unification.

How Does It Work?

The core functionality of this feature is the ability to bulk add columns from existing tables or previous steps:

1. To begin click the icon next to the output schema, select the desired tables, and choose the columns you want to add.
2. Add columns from all existing tables and transformation results from previous steps.
3. Don’t worry about duplications, the automatic handler ensures there are no duplicate columns in the output schema, simplifying the process.

Use Cases

Starting with a blank recipe

Goal

Create a new Blend Data step from scratch and map data sources to the output schema.

Steps

1. Add a new Blend Data step to the Recipe.
2. Use the column import feature to bring fields from existing tables.
3. Bulk add columns from the selected tables into the output schema.
4. Add new data sources and map their columns to the output schema.
5. Click Run and Save to update the transformation.

Adding custom fields based on previous steps

Goal

Enhance a completed transformation, such as a join or group operation, by adding calculated fields.

Steps

1. Start with an existing Recipe where a transformation (e.g., join or group) has been performed.
2. Add a new Blend Data step to the Recipe.
3. Use the column import feature to bring columns from previous steps.
4. Add a calculated column to the dataset, such as a new metric or derived field.
5. Click Run and Save to update the transformation.

Conclusion

By leveraging bulk column addition and seamless integration with previous transformations, you can streamline data unification and focus on deriving actionable insights.