Cross Model Mapping
You may want to create interactions between content items (data discoveries, slicers, dynamic content, or custom formulations) that are querying different data models. For such interactions to work, the hierarchies and member elements being used as a filter must also exist in the data model of the target visualization. Otherwise, when the hierarchy is used as a filter when running the publication, you will receive the following error:
If the corresponding hierarchies and member elements are identically labeled in both data models, then you can create an interaction as usual. However, if the corresponding hierarchies and/ or member elements are not identically labeled in each data model, then it's necessary to create a model mapping.
In such a scenario, the Model Mapping wizard is used to map the corresponding hierarchies and/ or member elements from the two data models, enabling the user to create interactions between them.
For instance, Model 1 may contain a hierarchy called State, while Model 2’s hierarchy is called State Code. Say that a data discovery displaying Manufacturer Sales from Model 1 is added to the publication, along with a State slicer from Model 2. When the publication is executed, Pyramid will search for the given State member in the target visualization; but because Model 1 doesn't contain a hierarchy called State, the user will get an error.
To enable the interaction from the dataKey slicer to the Product Sales chart, the user must configure a new model mapping which tells Pyramid that the two differently labeled hierarchies are the same.
Model Mapping Wizard
- Open the Model Mapping wizard from the Publication ribbon.
- Click the blue plus sign in the Mapping tab to create a new model mapping.
- The new mapping will be listed; from the Settings panel name the mapping, and select the two models to be mapped (the selection order is not important).
- Click the blue plus sign in the Map Hierarchies panel
- From the drop down under each model, select the hierarchy to be mapped.
- To map multiple hierarchies from the same models, click the plus sign in the Map Hierarchies panel.
- From the Map Members panel, select the mapping type: either Simple or Advanced. Continue reading for details.
Ignore Unmapped Hierarchies
The models being mapped may contain multiple hierarchies that are similar, yet not identical. For instance, they may both contain hierarchies listing Dates, where the naming convention differs between the models. In this case, if you have no need to map these hierarchies, you can simply select the 'Ignore Unmapped Hierarchies' option. This will stop you from receiving an error when applying interactions between visuals from the mapped models.
Member Types
Unique Name
The unique name describes the member element path within the data model. It is written [Dimension].[Hierarchy].[Member]. The unique name of the Bikes member element, for instance, is [Products].[Product Category].[Bikes].
Caption
Caption describes the member element's label. The caption for the Bikes member element is Bikes.
Mapping Types
There are 2 types of model mapping: simple and advanced. The type that you will need to configure will depend on what differences exist between the hierarchies being mapped.
Simple Mapping
If the hierarchies to be mapped are labeled differently, but each corresponding member element is identically labelled, choose Simple.
For instance, say the country hierarchy in Model 1 is called Country, while in Model 2 it's called Location. Within both hierarchies, the members are labeled identically. In this case, choose Simple and select the hierarchy name type (either Unique Name or Caption).
- Click here to review a simple mapping example.
Model 1 |
Model 2 |
---|---|
Country |
Location |
Australia |
Australia |
United States |
United States |
Canada |
Canada |
France |
France |
Germany |
Germany |
Advanced Mapping
Advanced Mapping is required to map the member elements within two hierarchies, if the member elements are not identically labelled. In this case, the Advanced option must be selected in order to configure the dynamic functions required to map the differently labelled hierarchy members.
Click here to review the advanced mapping functions; for a detailed examples of advanced mapping for date hierarchies, review the MemberCaption function.
Model 1 |
Model 2 |
---|---|
Country |
Country |
Australia |
AUS |
United States |
USA |
Canada |
CAN |
United Kingdom |
UK |
France |
FRA |
Germany |
GER |
- Click here to review an advanced mapping example.