Attribute columns
A column must be an attribute to be used to group or filter measures. When an attribute is created for a selected column, sub-properties become active.
Dimension and attribute keys
Column reporting design properties
Display Type - Defines how the attribute is displayed when in reports. The following options are available:
Attribute - A single-level hierarchy. This option represents a simple way to filter using just this column.
Hidden - Does not display the attribute in the dimension browser. This option is recommended when the current column is to be used as a level in a multi-level hierarchy.
Hierarchy - A two-level hierarchy that includes the current column and the key attribute. (This option is not available in dimensions where Stand-alone is enabled.
Member Property - Displayed as a property of the key attribute. This option is useful when the current column contains extra information about the key attribute and is not required for filtering.
Parent-child Hierarchy - Displayed as a multi-level parent-child hierarchy. This option should be used when the current column relates to the key attribute of the pipeline (e.g. EmployeeReportsTo. (This option is not available in dimensions where Stand-alone is enabled.
Name Column - If a column is specified here, members in the selected attribute are identified with the value of this column, rather than by the value of the key column. This means that members display meaningful identifiers rather than with the key column values themselves, which are typically numeric values that are not immediately recognizable.
An attribute may be ordered by its name column using the Order By setting.
Tip
For example, when using a CRM data source, and viewing the key attribute column (highlighted below), this property is automatically specified, based on the corresponding name column found in CRM.
Note
If the pipeline is a dimension, the primary key attribute will be named for the main key column. If the pipeline is marked as a dimension in the cube, these keys will also be used as the primary key attribute in the dimension.
Order By - This allows you to specify how each attribute is listed in a cube. An attribute is ordered by either key value (the value of the attribute column itself) or by its name column, as assigned by the Name Column setting. If the attribute does not have a name column assigned, the setting has no effect and the attribute is assigned by key value.
Note
The order of rows in the Preview window will not alter in response to this setting as it is applied to the cube, not to the warehouse table displayed by the preview.
Order By Attribute - This allows you to specify another attribute in your model by which this attribute is ordered. By default, an attribute is ordered by its own key or name, i.e. the (none) option is selected. You can only order properties of this attribute.
Property Of - This allows you to specify which column a given column is a member property of, instead of all properties automatically being properties of the pipeline's key.
Tip
Example In a Customer pipeline, you can make the state abbreviation column (for example, CA) a member property of the state name column (e.g. California), rather than the CustomerID key column.
Tip
Example To create a map chart with bubbles representing sales by state, you need the latitude and longitude for each state in order to position the bubbles.
Note
The Property Of setting makes an attribute relationship.
For columns that are to be member properties, you should set the column's Display Type setting to Member Property before selecting the Property Of setting.
If this feature is used with an attribute not marked as a member property, ensure that there is a one-to-one or one-to-many relationship between entries in the attribute column selected in the pipeline schema and the column selected in the Property Of drop-down box. For example, if the State column is made a property of City column, there must be a one-to-one or one-to-many relationship (but not a many-to-many relationship) between State and City.
The Order By drop-down list allows an attribute to be ordered either by its key or by its name. If a different order is required, change the order of the attribute's key using a calculated column. Make the calculated column the attribute's key, then order by the key.
Attribute relationships are also used to optimize aggregation in a hierarchy. This feature may be used for that purpose, although the Force Natural check box in the hierarchy's Properties pane is often a faster way of achieving this result.
Reporting Type - This allows you to provide additional details about the information stored in the current column. This property is used for levels of date hierarchies when using relative time functions.
Additional attribute key columns - This allows you to increase the granularity of an attribute (other than the main key column) by adding additional columns to the attribute's key.