Data Types

Spectrum Technology Platform supports a variety of numeric, string, and complex data types. Depending on the type of processing you want to perform you may use one or more of these. For an address validation flow you might only use string data. For flows that involve the mathematical computations you may use numeric or Boolean data types. For flows that perform spatial processing you may use a complex data type. For flows that combine these, you may use a variety of data types.

Specifying a Field's Data Type

You can specify the data type for a field in these situations:
  • Source stages: Specifying data types allows you to set the data type at the beginning of a flow, eliminating the need for data type conversions later in the flow. Note that for Read from DB, the data type is selected automatically and cannot be changed.
  • Sink stages: Specifying data types allows you to control the data format returned by the flow. Note that for Write to DB, the data type is selected automatically and cannot be changed.
  • Transformer stage: You can specify data types in this stage if you use a custom script.
  • Math stage and Group Statistics stage: Since these stages perform mathematical calculations, choosing to use a particular numeric data type can have an effect on the results of the calculations, such as the precision of a division operation. If you specify a data type for a field that is different than the data type of the field coming into the stage, the downstream channel will automatically convert the field to the data type you specify, as described in Automatic Data Type Conversion.
Note: Each stage supports different data types. For a description of the supported data types for each stage, see the documentation for a specific stage.