STRUCT data type

Syntax for STRUCT

column_name STRUCT < name : type [COMMENT 'comment_string'], ... >
type ::= primitive_type | complex_type

A STRUCT has a fixed number of named fields, and each field can be a different type. A field within a STRUCT can also be another STRUCT, an ARRAY, or a MAP

STRUCTs in the Dataset Field interface

In the Dataset Fields interface, a basic STRUCT data type may look something like the following example. You can see that each level of a complex data type can be expanded to show component details, or collapsed for simplicity.

In the example of the dataset Complex Type - Struct, you can see that the Dimensions Customer (String, marked with the symbol A), Orderid (Integer, marked with the symbol #), and overalldiscount (Real, marked with the symbol 1.2) are primitive types. However, orderinfo is a Struct data type, marked with the symbol [S].

When you click Edit Fields, you can see that while primitive types can be cast as alternate data types (such as Integer into Real), the complex data type STRUCT cannot be changed to another type. However, the primitive components of the array may be cast as other primitive data types. Additionally, unlike other data types, Cloudera Data Visualization uses complex data types only as Dimensions. They or their components cannot be redefined as Measurements of the dataset.

STRUCTs in visuals

When building a visual with complex data, you cannot use the complex type directly as a whole. However, you can add the primitive components of the complex type to the shelves of the visual.

For example, in a Pie visual, you might place Customer Name on the X Trellis shelf, the orderinfo.category component on the Dimensions shelf, and the orderinfo.amount component on the Measures shelf.

Changing field properties

It is very simple to change field properties for a component of a complex data type.

For example, you can change the orderinfor.qty component on the Tooltips shelf from the default max() aggregation to the count() function.

STRUCT in Expression Editor

The expression editor fully supports the use of STRUCTs, both in the Dataset and Visual interfaces. This enables advanced data manipulation and customization, leveraging the capabilities of STRUCT data type in your data visualization tasks.