Struggling with a duplicate group model

[Legacy User]

We have an issue with a group model, which was duplicated almost a decade ago, and continued growing both copies starting from same basic objects (same id for model and ancient objects). Ten years later, both models have same features, and a good number of biased developments. We are trying to eliminate silos, looking for ways to bring both together. Some of you have experience on this subject?

[George Jeffcock]

forked is forked

 

Imagine asking these software packages to merge List of software forks - Wikipedia, the free encyclopedia

 

Have tried before merging two xml group model exports to create a new group model but I'm sure with time and many many many edits latter you could do it but have given up in the past

[Legacy User]

indeed, George...

moving to a third model isn't a solution, because of use of references. Maybe losing 50% of references (one of both models) could be managed, but losing 100%, not

I played thinking in some way to overwrite xml export file, but I do not know anyone working on this way.

I would consider your reply the correct answer, and I will do, but some idea would be welcome...

[George Jeffcock]

Last time I tried on a big application model was in 2007 and even after trying to get CA support to help, all concerned gave up after about a month on and off trying.

 

BUT I did not have StellaTools back then and deep knowledge about how triples and surrogates hold everything together under the covers in the plex model.

 

i Am ready to give this another go if you are. We need to export the smaller model, parse it, update it, import into the larger model. Try again, and try agin.

 

I remember it was the surrogates that in the end hurt us but this time we could use model api to replace. Let's talk off line and publish our findings latter if you want.

[Legacy User]

Thanks George,

I will e-mail you later today.

Enlighting on a method to try surely will help a lot.

Kindly

Jorge

[George Jeffcock]

Hi Jorge,

 

Further thought over night, Is you need a merge more than consolidation.

 

Where I failed previously was I had group model 1 and group model 2. Group model 2 was smaller than group model 1 and inherited from objects in group model 1. So we (Platform Home Loans) regretted having group model 2 and the separation made for more work than helping dev so we wanted to import group model 2 into model 1 (Consolidation like Plex Local Model extract duration) . I feel I could succeed now with this task where I failed back in 2007 but your merge has so many more issues.....of course it is possible, give us a desert island and all the time in the world we could create an import xml file manipulated to be a success but after millions of attempts...lets talk off line to see what are the issues with a merge.

[George Jeffcock]

A quick test and consolidation works in Plex 6.1

                                                                       

GroupModel1

FncA

                                                                      

GroupModel2

FncB calls GroupModel1/FncA

FncC is a GroupModel1/FncA

                                                                      

 

• Exported GroupModel2
• Imported it into GroupModel1

And what did not work in 2007 with a much more complicated proper application model worked in 2015 for this proof of concept

GroupModel1

FncA

FncB calls FncA

FncC is a FncA

Enabling me to throw away GroupModel2

[Legacy User]

Hi George

Excuse me for not replying yet. Very busy these days. I hope I will have time enough this weekend.

Kindly

Jorge

[Legacy User]

George, done: sent a note to your e-mail account.

[George Jeffcock]

This is going to help us a lot CPLEX-700-1863-001:NEW PLEXAPI METHODS

 

bUT no one ever got back to me was this avaliable for plex 6.1