Hi Nicolas,
I see that you reverted Kernel-ct.1382 (which has been merged into the Trunk earlier via Kernel-mt.1383). Was this a merge error or an intended rollback?
Assuming the latter, I would like to discuss this again, please. :-) I already noticed by myself that the control primitive warning occurred a bit too often in situations where you would not expect it (for instance, when stepping through "Object sourceCodeAt: #at:"). Still, do we indeed do not want to get informed about any attempt by the debugged process to execute logic in another process? Also, the simulator is not only used by the debugger but also in miscellaneous non-interactive situations, including MessageTally, some KernelTests, and SimulationStudio where the warning might be even more relevant because the assumption or the contract is that the passed logic will be simulated completely. Would it be a compromise for you to signal a custom EscapeFromSimulationWarning (better name proposals are welcome :-) ) instead and to ignore them in the Debugger?
I am looking forward to your reply! :-)
Best, Christoph
----- Carpe Squeak! -- Sent from: http://forum.world.st/Squeak-Dev-f45488.html