Generic

Overview

Open the Data load editor and add the load script below to a new tab.

The load script contains a dataset which is loaded into a table named Transactions. The dataset includes a date field. The default MonthNames definition is used.

Load script

Products:
Generic
Load * inline [
Product ID, Attribute, Value
13, Status, Discontinued
13, Color, Brown
20, Color, White
13, Size, 13-15
20, Size, 16-18
2, Status, Discontinued
5, Color, Brown
2, Color, White
44, Color, Brown
45, Size, 16-18
45, Color, Brown
];

Results

Load the data and open a sheet. Create a new table and add this field as a dimension: Color.

Add this measure:

=Count([Product ID])

Now you can inspect the number of products by color.

Note the shape of the data model, where each attribute has been broken out into a separate table named according to the original target table tag Product.Each table has the attribute as a suffix. One example of this is Product.Color. The resulting Product Attribute output records are associated by the Product ID.

Overview

This example shows how to analyze EAV structured data in its original form.

Open the Data load editor and add the load script below to a new tab.

The load script contains a dataset which is loaded into a table named Products in an EAV structure.

In this example, we are still counting products by color attribute. In order to analyze data structured in this way, you will need to apply expression-level filtering of products carrying the Attribute value Color.

Furthermore, individual attributes are not available to select as dimensions or fields, making it harder to determine how to build effective visualizations.

Load script

Products:
Load * Inline
[
Product ID, Attribute, Value
13, Status, Discontinued
13, Color, Brown
20, Color, White
13, Size, 13-15
20, Size, 16-18
2, Status, Discontinued
5, Color, Brown
2, Color, White
44, Color, Brown
45, Size, 16-18
45, Color, Brown
];

Results

Load the data and open a sheet. Create a new table and add this field as a dimension: Value.

Create the following measure:

=Count({<Attribute={'Color'}>} [Product ID])

Now you can inspect the number of products by color.

Overview

In this example, we show how the normalised data structure produced by the Generic load prefix can be denormalised back into a consolidated Product dimension table. This is an advanced modeling technique which can be employed as part of data model performance tuning.

Open the Data load editor and add the load script below to a new tab.

Load script

Products:

Generic
Load * inline [
Product ID, Attribute, Value
13, Status, Discontinued
13, Color, Brown
20, Color, White
13, Size, 13-15
20, Size, 16-18
2, Status, Discontinued
5, Color, Brown
2, Color, White
44, Color, Brown
45, Size, 16-18
45, Color, Brown
];

RENAME TABLE Products.Color TO Products;

OUTER JOIN (Products)
LOAD * RESIDENT Products.Size;
						
OUTER JOIN (Products)
LOAD * RESIDENT Products.Status;
DROP TABLES Products.Size,Products.Status;

Results

Open the Data model viewer and note the shape of the resulting data model. Only one denormalized table is present. It is a combination of the three intermediary output tables: Products.Size, Products.Status, and Products.Color.

Load the data and open a sheet. Create a new table and add this field as a dimension: Color.

Add this measure:

=Count([Product ID])