Compiler flag -d=aliasConflicts takes no affect

[Oliver Lenord]

In a model I received this error message:
[1] 15:33:36 Symbolic Warning
The model contains alias variables with conflicting start and/or nominal values. It is recommended to resolve the conflicts, because otherwise the system could be hard to solve. To print the conflicting alias sets and the chosen candidates please use -d=aliasConflicts.

For the sake of deeper exploration of the problem I assigned the
Additional Translation flags in OMEdit: -d=aliasConflicts

Nothing changed. Where am I supposed to find the extra information?

[Karim Adbdelhak]

Consider following model:

  model M
    Real a(start = 0.0), b(start= 1.0);
  equation
    a = b;
    b = sin(time);
  end M;

This is such a conflict since the model states that a=b but the start values do not match.
If i run this i get the warning you described and using that translation flag yields:

[2] 16:55:54 Translation Warning
Alias set with conflicting start values
 * Candidate: a(start = 0.0, confidence number = 6)
 * Candidate: b(start = 1.0, confidence number = 6)
=> Select value from b(start = 1.0) for variable: a

At the same place where the warning popped up.

Isn't that the case for your model?

[Oliver Lenord]

Thanks for sharing the test case. This helped to narrow down the problem.

I used the Tools/Options dialog to define the "Additional Translation Flags" and expected that these are applied always.

Now I recognized the disclaimer in the dialogue box saying: "Global flags applied to the Simulation Setup dialog upon the first simulation of a model."

Changing the flag in the simulation setup dialogue solved my problem.

Neither simulate nor unload+reload+simulate, nor re-simulate, nor delete results+simulate did work. Only a restart of OMEdit would carry over the settings from the global settings to the simulation.

Is this by design? - At least from my user-perspective this is an unexpected behavior which caused a significant amount of frustration when trying out the compiler flags.

This ticket can be closed. But I'll open a change request for proper handling of global settings.

[Karim Adbdelhak]

Okay, glad i could be of help. I will close this ticket.

Unfortunately i am a backend developer and do not know much about the OMEdit stuff, i guess @adeas31 can have a look at it once you opened the new ticket!

[Francesco Casella]

Replying to olivleno:

Now I recognized the disclaimer in the dialogue box saying: "Global flags applied to the Simulation Setup dialog upon the first simulation of a model."

Changing the flag in the simulation setup dialogue solved my problem.

Neither simulate nor unload+reload+simulate, nor re-simulate, nor delete results+simulate did work. Only a restart of OMEdit would carry over the settings from the global settings to the simulation.

Is this by design? - At least from my user-perspective this is an unexpected behavior which caused a significant amount of frustration when trying out the compiler flags.

It is. Let's continue the discussion on #6062.