I would like to get your opinions on separation of SDO generated models
from our code.
Currently on the client side we need to work with project container.
There are SDO interfaces generated from XML Schema, which can be used
for now, but adding and additional layer makes us more independent. So
it would look like this:
module1
|
module2
|
module3
|
...
|
Our separation interfaces
|
SDO generated interfaces
|
Advantages:
- if we change SDO to any other model (eg. XMLBeans or whatever),
we do not have to change code in moduleX
- we can provide some convenient methods in our interface (eg.
getRepositoryByKind() )
- if the model is changed (eg. repositories goes out from container
definition) we can still handle it without changes in moduleX
Disadvantages:
- additional effort to make adapters (not a big effort)
- possibility that SDO generated models get out of sync with our
separation interfaces
- people may fill confused with two kinds of interfaces for the
same thing
- an additional layer is always some overhead
Personally I'm for adding the additional layer. What is your opinion
about it?
Marcin
The contents of this e-mail are intended for the named addressee only. It contains information that may be confidential. Unless you are the named addressee or an authorized designee, you may not copy or use it, or disclose it to anyone else. If you received it in error please notify us immediately and then destroy it.
|
