![]() |
Telelogic Rhapsody (steve huntington) | ![]() |
Topic Title: Forward Declaration Problem on target Topic Summary: Created On: 23-Oct-2008 11:14 Status: Read Only |
Linear : Threading : Single : Branch |
![]() |
![]()
|
![]() |
|
I am using RiC 7.0 and i am having a problem when compiling the sourcecode for OSEepsilon with the tasking compiler.
Rhapsody generates a forward decleration of the "object" for every aggregation. eg. if I aggregate a Display with a HMI Class in the OMD then Rhapsody creates a the output: struct HMI; in the Display.h file In C this would not be necessary because by including the HMI.h file in the Display.h, the object already knows the struct HMI because its declared in HMI.h via a typdef. Normally this forward declaration is no problem, but the tasking compiler does not allow it. is there a possibility to keep Rhapsody from creating it? regards mike |
|
![]() |
|
![]() |
|
Mike.
I'm primarily a RiC++ user but I think the mechanism is the same in RiC. The property CG::Relation::Dependency controls whether a forward declaration or #include is generated in the .h file of the aggregator. Set to "Strong" to remove the forward declaration. Also check out the CG::Dependency::Usage type property. Which one of these properties is invoked depends on how you've drawn the relationship on your OMD. Cheers, Alistair |
|
![]() |
|
![]() |
|
Seems like setting the property to strong was the solution!
Many thanx for that quick reply. best regards, Mike |
|
![]() |
Telelogic Rhapsody
» Rhapsody Category » Rhapsody
»
Forward Declaration Problem on target
|
![]() |
FuseTalk Standard Edition v3.2 - © 1999-2009 FuseTalk Inc. All rights reserved.