This repository has been archived by the owner on May 24, 2022. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 2
Design and Execute Transformations
gertho edited this page Nov 28, 2014
·
10 revisions
Data transformations can be defined for existing mappings only. Every mapping in a project is represented by a tab in the transformation logic widget in the lower middle. The mapping whose transformation tab is currently activated is marked red in the mapping area.
A transformation logic can be modelled the following way:
- Select the mapping whose transformation you want to design by clicking on the respective tab in the transformation logic widget (1).
- Initially every mapping has an empty transformation in which the input attribute path and the output attribute path are directly linked with each other.
- In order to design a transformation logic, drag a function (2) from the function list widget on the lower right into the grid of the transformation logic widget (3).
- The transformation process is visualized by arrows between the components.
- Drag more functions to the desired grid position in the transformation process.
- Rearrange the order of the components in the grid as to modify the transformation process.
- You can configure a component by clicking on its representation in the transformation logic widget (4) and entering the required parameters in the configuration widget (5).
- Enter all parameters and (optionally) quit the configuration by clicking on the button “Close” (6).
- Repeat these steps for all mappings that have been defined.
- A component can be deleted by clicking the "x" (7) in the configuration widget.
- A mapping can be deleted by clicking the "x" (8) in the transformation logic widget.
Additionally, you can design complex transformations in the alpha release, e.g. concatenate two input attribute paths to one output attribute path. Within the transformation process, two data sources can be individually transformed as is suitable and eventually combined using the concat
function:
- Drag the
concat
function from the function list widget (lower right) to the grid in the transformation logic widget (lower middle). - Clicking on the "+" symbol in the
concat
component will open a dialogue for selecting the component that you want to add the attribute path:
- Select the component that you want to add to the current component by clicking in the list (1).
- Click on the button “Add selected component” to quit and concatenate the two components to a single output attribute path (2).
Any mapping (incl. transformation logic) is executable at all times:
- Click on “Run this transformation” (1) in order to execute the transformation logic of the mapping currently displayed and view the results of the mapping executed on the sample data in the target data widget on the upper right (2).
- Click on “Run all Transformations” (3) in order to execute the transformations for all mappings in the project.
- Click on “Save all transformations” (4) in order to run all transformations for the entire data model, and save the results in the data hub. You can access the results via a web GUI as described in Graph Exploration.
- next: Define Filters
- previous: Define Mappings
- Overview
- misc
- Graph Data Model
- Server-Installation (Productive Environment)
- Maintenance
- HowTos
- Use Cases