Is it advisable to change the direct inheritance of a framework's data class to point to divisional's data class
Hi All,
I have seen few pega implementations where the direct interitance of the framework data class(MyOrg-FW-MyFW1-Data-) is pointing to the divisional data class(MyOrg-Div1-Data-) .As per my understanding this should never be the case as implementation layer will always have access to all the rules available in the divisional layer .In what scenario's would we go for this kind of approach and what are the advantages/disadvantages of it .
Thanks
Dinesh
***Updated by Moderator: Vidyaranjan. Removed user added #helpme and Ask the Expert tags. Apologies for confusion, shouldn't have been an end-user option; added Category***
Pegasystems
US
It may be useful to know what tool you used to generate the divisional data class ancestry, as I assume it wasn't a personal choice.
/Eric
JP Morgan Chase
IN
Its been generated by the Wizard and then the direct inheritance was modified.Probable reason why they must have done it is to re-use the data model created in the divisional layer which i feel is incorrect as framework is tightly coupled with divison and there by cant work standalone unless it has the divisional layer .
Pegasystems Inc.
US
Dinesh,
Without knowing about the specific implementation, it's hard to say. I could speculate that the implementation in question had no divisional layer. Perhaps it was built before the wizard existed and was upgraded, or they just put nothing in it and decided it was cleaner to bypass it? It's hard to say exactly how one should implement their class structure withhout knowing the specific business requirements the application needs to address. You want to make sure your inheritance path contains everything you need and leaves room for growth and future expansion (which would be my argument for leaving a divisional layer in there that is nothing but a few empty classes).
Thanks,
Mike
Pegasystems Inc.
US
Dinesh,
Without knowing about the specific implementation, it's hard to say. I could speculate that the implementation in question had no divisional layer. Perhaps it was built before the wizard existed and was upgraded, or they just put nothing in it and decided it was cleaner to bypass it? It's hard to say exactly how one should implement their class structure withhout knowing the specific business requirements the application needs to address. You want to make sure your inheritance path contains everything you need and leaves room for growth and future expansion (which would be my argument for leaving a divisional layer in there that is nothing but a few empty classes).
Thanks,
Mike