Calling an IP from a non-persistent procedure

[MrGinge]

trying to get into GUI and have a (hopefully basic) issue that i can't get my head round.

the application i'm using is written and supplied by a third party but we have the ability to bolt new devs onto it. however, we supposedly cannot run any new persistent procedures which gives me a bit of a headache...

so the 'standard' approach of running the business logic persistently and calling the IPs whenever i need to can't be done. How do i get round this? Do i stick all my business logic procs in a .i and include it in my main program? The app is essentially thick client, with users connecting remotely via propalms.
Or is there a better way?

any help much appreciated!

MrG

 

[tamhas]

The first question is why can't you run any new persistent procedures ... is this just some rule the vendor has given you? It makes no sense.

 

[jp-uk]

The first question is why can't you run any new persistent procedures ... is this just some rule the vendor has given you? It makes no sense.

its just a rule i believe - but obviously straying from the rule will invalidate the service contract - the only thing i can think of is that its to ensure the integrity of their product

 

[tamhas]

But, how does forbidding PPs, but allowing other code changes insure integrity. If anything, the use of PPs and SPs helps to keep code changes to the base product to a minimum.

 

[jp-uk]

But, how does forbidding PPs, but allowing other code changes insure integrity. If anything, the use of PPs and SPs helps to keep code changes to the base product to a minimum.

i believe - and i do stress that i'm not sure - that its incase you create a new super procedure that is named similar to one already existing - this is a fairly big system and everything is encrypted so you aren't even sure what all their pp/sp are...

at the moment all our code is going into include files and bunged into the method library

 

[tamhas]

The irony, of course, is that is exactly the way to do the equivalent of overriding a method, preserving the code in the super and being able to add your own. Yes, they would need to document the API to some degree, but it is exactly the right approach.