muscle::core::kernel::Submodel Class Reference

Implements muscle::core::kernel::RawInstance.

Initializes the submodel, determining at what time it will start. It will be called once at the start. Override to implement the f_init operator in MML. May only receive messages. The default implementation is to start at origin 0, and to increase this with getScale().getOmegaT() every next run.

Parameters:

previousOrigin

the previous time origin. This is null if there was no previous origin.

Returns:

the new time origin. This can be determined for instance by the timestamp of a conduitExit.receiveObservation().getTimestamp() or simply 0.

Perform one step of the submodel. It will be called while the model is active. Override to implement the S and B operators in MML. During this step, messages may only received.

Send messages from within the loop. It will be called while the model is active. Override to implement the O_i operator in MML. During this step, messages may only be sent.

Send messages and clean up as a final operation. It will be called at the end of a submodel. Override to implement the O_f operator in MML. During this step, messages may only be sent.

Determines whether the submodel will exit the submodel execution loop, and jump to the final observation.

Returns:

by default, the value of willStop().

Determines whether a submodel will be restarted, once it is finished.

Returns:

false by default.

currently returns true if all portals of this kernel have passed a maximum timestep given in the cxa properties (this is a temporary implementation because this mechanism is going to change anyway) note: if init portals are being used, they do increment their time only once! do not change signature! (used from native code)

Reimplemented from muscle::core::kernel::RawInstance.