To derive a hierarchy from a given structure there is ideally a parent-child relationship of the individual tuples. In such a case, the function BuildHierarchy can be used to create the hierarchy.

If a direct parent-child relationship is only possible via the sequence of tuples and unique parent or child characteristics¹⁾, the following approach can be chosen:

We use the following functions and components:

1. $$ Variables
2. SplitToArray
3. Selektionsmodus: concatenate
4. XPATH Filter

First we store the tupples all in a $$ Variables with name $$Position to run through them in the next step. We then filter out the tuples by parent or child characteristic to form a string containing the hierarchy²⁾. By cleverly applying the selection mode “concatenate” on the rule sets and controlling the separators³⁾ a string can be formed according to the following pattern:

MainPositionID1,MainPositionID2#SubPositionID1#SubPositionID2,MainPositionID3#SubPositionID1

This hierarchy string can finally be split with SplitToArray and the separator “,” into the main positions, which trigger Positions in the target object and then for sub-items are again individually resolved after the separator “#” into the sub-items, which are triggered in the shown example below Positions as CustomItems in the MiddleWare⁴⁾.

Note: If the same item numbers are expected to come in different positions, then the original position number must be taken into account when creating the hierarchy!

We avoid an intermediate mapping⁵⁾ which could contain the parent-child relationship.