Hi Benny,
From past experience, I have learned to heed Horst's advice. 🙂 I would just go with what he said above: "this wont work out", but if this requirement is absolutely critical, you can try the following:
The following suggestion applies only if you are trying to achieve this requirement in your own custom BO. This will not work with SAP's standard BOs.
It *might* work if you radically change your BO design. Whatever you have defined today as the ROOT node, you will have to move to a sub-node, and create a 'dummy' root node on top of this. In the OWL, although the data model's root is mapped to your BO's ROOT, the query and the owl list will be mapped to this sub-node.
On select of multiple records and click of button, store all their node-ids as a comma-separated structure, and pass this structure as a navigation parameter.
On doing so, when you launch the QA, you will be actually loading the solitary ROOT instance, but will populate just those selected rows (you will first have to separate the comma separated node ids and pass these as query parameters to fill your list in the QA).
Honestly though, this is breaking a lot of rules and therefore comes with lot of problems. For example, only one instance of the QA can be opened at a time by any user, because every EDIT will acquire a lock on the ROOT node, and therefore all the instances. Second, I'm not sure which data type will be able to support a comma-separated structure of indefinite length. Third, I am not sure how to resolve the comma-separated node id's (maybe on load ABSL? Or maybe ruby-script?). Fourth, you will have to ensure that for every fresh tenant, *one and only one* ROOT instance is created by default, before the tenant is ready for testing/production. Fifth, this is just too painful. 😉
Cheers,
Rohan
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.