A stimulus is a communication between two instances and
is generated by an action. On a sequence diagram a stimulus is
associated with a link—an instance of an association linking
two object instances. In the UML metamodel
Stimulus is a sub-class of
The link (see Section 19.9, “ Link ”) associated with a stimulus is represented on a sequence diagram in ArgoUML as an arrow between the time lines of the object instances (or the object head in the case of stimulus create, described below) labeled with the name of the action (if any), and the action, separated by a colon (:). The type of line and arrowhead depends on the type of action that generated the stimulus:
Stimulus Call. Generated by a call
action, itself the result of an operation of a class. Shown
as a solid line with a solid arrowhead to the time line of
the object instance receiving the stimulus.
Stimulus Create. Generated by a
create action for the class for which an instance is to be
created Shown as a solid line with a solid arrowhead to the
object head of the object instance being created.
Stimulus Destroy. Generated by a
destroy action of the originating object. Shown as a solid
line with an open arrowhead terminating in a diagonal cross
at the end of the time line of the receiving (destroyed)
Stimulus Send. Generated by a send
action, the result of a signal raised by an operation of
the sending object instance and handled by the receiving
object instance. Shown as a solid line with half an open
Stimulus Return. Generated by an
object instance that has received an earlier call stimulus
and is returning a result to the calling object instance.
Shown as a dotted line with an open arrowhead.
ArgoUML does not allow you to create stimuli directly, but instead provides tools to create stimuli of each of the five types above.
In the current release of ArgoUML there is no way to show a terminate action where an object instance destroys itself. One way is to draw a destroy action that loops back to the object itself, give it an action with no name and use the style tab to set an invisible line, but this still leaves the arrow head showing, which is unsightly. It is also semantically incorrect anyway to use a destroy action to represent a terminate action.
The details tabs that are active for stimuli are as follows.
Standard tab. The values for the bounds of the stimulus notionally define the bounding box of the stimulus and its time line. However if you change them it will have no effect, and the original values will be reset when you next revisit the tab.
Shadow entries has no effect. Rather
bizarrely you can set the
and it will draw a line around the signal, which is not
a standard UML representation.
To change the color of the line, you should select the associated link (click on it a little way from the stimulus) and use its style tab (see Section 19.9, “ Link ”).
In the current release of ArgoUML changing the
values of the
Standard tab, but with no contents.
A stimulus should not generate any code, so having this tab active is probably a mistake.
Standard tab. ArgoUML only supports constraints on Classes and Features (Attributes, Operations, Receptions, and Methods), so this tab is grayed out.
Standard tab. In the UML metamodel,
Stimulus has the following standard tagged
derived (from the
true, meaning the
stimulus is redundant—it can be formally derived
from other elements, or
meaning it cannot.
Derived stimuli still have their value in analysis and design to introduce useful names or concepts, and in design to avoid re-computation.
Navigate up through the package structure.
This deletes the stimulus from the model
This is a deletion from the model
not just the diagram. To delete
an stimulus from the diagram, but keep it within the
model, use the main menu
Text box. There is no convention for naming stimuli, and it is quite normal to leave them unnamed. The action is sufficient identification.
It is sometimes useful to give simple names to stimuli, so they can be referred to in attached notes giving timing constraints.
Text box. This is used to identify the action that generated the stimulus.
The current release of ArgoUML only implements actions as textual descriptions.
As a practical convention it is suggested that call actions are shown as the name of the operation generating the action with any arguments in parentheses and that send actions are shown as the name of the signal generating the action with any arguments in parentheses. Return actions should be shown as the expression for the value they return, or empty otherwise. Create and destroy actions should be left empty, since they are implied by their representation.
Drop down selector. Stimulus has no stereotypes
by default in the UML standard, but ArgoUML provides
ArgoUML also provides the stereotype
icon. If a stereotype has been selected, this will navigate to the stereotype property panel (see Section 18.5, “ Stereotype ”).
Text box. Identifies the instance which sent this stimulus.
Button 1 click navigates to the sender instance, button 2 gives a pop up menu with one entry.
Open. Navigate to the
selected sender instance.
Text box. Identifies the instance which receives this stimulus.
Button 1 click navigates to the receiver instance, button 2 gives a pop up menu with one entry.
Open. Navigate to the
selected receiver instance.
In the current release of ArgoUML this field is
broken. It always shows the entry
Text box. Records the namespace for the stimulus. This is the package hierarchy.
Button 1 click on the entry will navigate to the package defining this namespace (or the model for the top level namespace).