Eclipse Community Forums
Forum Search:

Search      Help    Register    Login    Home
Home » Eclipse Projects » EPF » processes and configurations relationship
processes and configurations relationship [message #36245] Tue, 31 July 2007 18:48 Go to next message
Eclipse User
Originally posted by: carlos.ribeiro.edisoft.pt

Hi there,
One thing I still do not understand is the reason to have to associate a
process (capability pattern/ delivery process) to a configuration at
creation time.

I thought that I could:
1. define all the Method Content and,
2. define all the Processes (Capability Patterns and Delivery Processes)
without caring about the configuration.
3. Lastly, define the configurations and determine which Processes to
include on them.

I see no reason to have to associate processes to configurations at
creation time.

Moreover, I read that the choice of default configuration, once done,
cannot be changed!

Can anyone explain me this or point to some document where it is explained?

Thanks in advance,
André
Re: processes and configurations relationship [message #36861 is a reply to message #36245] Mon, 06 August 2007 20:51 Go to previous message
Peter Haumer is currently offline Peter Haumer
Messages: 228
Registered: July 2009
Senior Member
The reason is variability. When you apply a task to a process it needs to
evaluate all variability relationships (e.g. contributions, replace) to
create the respective task and role descriptors as well and link them. The
descriptor manages its own relationships, i.e. the user could then
afterwards go ahead and still remove and add more elements. The tool then
later cannot remember if a relationship was added or removed manually or
via some variability relationships, that might also not exist anymore or
have been changed. Hence, applying a descriptor creates is snaphot of the
configuration state at the point of time of creating it. It does not know
about other changes until you tell it to synchronize, in which it will
restore to the current state of the config, which might now be completely
different and might override your local changes. To not accidentally loose
your changes we recommend switching off synchronization for the descriptor
using the check box in the properties view.

Peter.


"Carlos "Andr
Re: processes and configurations relationship [message #581209 is a reply to message #36245] Mon, 06 August 2007 20:51 Go to previous message
Peter Haumer is currently offline Peter Haumer
Messages: 228
Registered: July 2009
Senior Member
The reason is variability. When you apply a task to a process it needs to
evaluate all variability relationships (e.g. contributions, replace) to
create the respective task and role descriptors as well and link them. The
descriptor manages its own relationships, i.e. the user could then
afterwards go ahead and still remove and add more elements. The tool then
later cannot remember if a relationship was added or removed manually or
via some variability relationships, that might also not exist anymore or
have been changed. Hence, applying a descriptor creates is snaphot of the
configuration state at the point of time of creating it. It does not know
about other changes until you tell it to synchronize, in which it will
restore to the current state of the config, which might now be completely
different and might override your local changes. To not accidentally loose
your changes we recommend switching off synchronization for the descriptor
using the check box in the properties view.

Peter.


"Carlos "Andr
Previous Topic:how to know which tasks are used by which processes
Next Topic:Compatibilty with RMC 7.x ?
Goto Forum:
  


Current Time: Tue Oct 21 03:47:30 GMT 2014

Powered by FUDForum. Page generated in 0.06291 seconds
.:: Contact :: Home ::.

Powered by: FUDforum 3.0.2.
Copyright ©2001-2010 FUDforum Bulletin Board Software