Transformation types
Transformation types define how logic is implemented and executed in AnalyticsCreator. They determine whether a transformation is generated automatically, maintained manually, executed externally, or used for a specific modeling purpose such as data marts or unions.
Use this section to understand the available transformation types and choose the appropriate one for your modeling and execution scenario.
Available transformation types
Datamart transformation
Used to expose data in a form intended for analytical consumption in the data mart layer.
- Consumption-oriented structure
- Typical basis for facts and dimensions
- Used in reporting-facing models
Direct-connector transformation
Used when transformation logic is tied directly to a connector-based source or source-side access pattern.
- Connector-oriented processing
- Useful for direct source integration scenarios
- Bridges source access and transformation logic
External transformation
Used when transformation logic is implemented outside the standard generated SQL transformation flow.
- External execution logic
- Useful for integration with external processing
- Suitable when logic is not managed as standard generated SQL
Manual transformation
Used when transformation logic is written manually instead of being generated automatically from metadata.
- Manual SQL definition
- Maximum flexibility
- Useful for special-case logic
Regular transformation
Standard generated transformation type used for typical SQL-based transformation logic.
- Common default transformation type
- Usually generated as SQL view logic
- Suitable for most modeling scenarios
Script transformation
Used when transformation logic is implemented through script-based execution instead of standard generated SQL view logic.
- Script-driven transformation behavior
- Useful for non-standard processing steps
- Supports custom execution logic
Union transformation
Used to combine multiple compatible inputs into a single transformation result.
- Combines multiple sources or transformations
- Useful for consolidation scenarios
- Supports structurally aligned inputs
How to choose a transformation type
- Use Regular transformation for standard generated SQL-based transformation logic
- Use Datamart transformation for reporting and analytical output structures
- Use Manual transformation when the logic must be written explicitly
- Use Script transformation or External transformation for non-standard execution patterns
- Use Union transformation when multiple inputs must be merged into one output
- Use Direct-connector transformation when the transformation is tightly coupled to connector-based source access
Key takeaway
Transformation types define how transformation logic is generated, maintained, and executed in AnalyticsCreator, and should be selected based on the required level of automation, flexibility, and execution pattern.
