![]() Might not want all flows stopped and all tokens destroyed. Issue in UML 1.5 because of the run-to-completion semantics, but with the unrestricted parallelism of UML 2.0, you Flow Final simply terminates the flow to which it is attached. It is needed because in UML 2.0, when control reaches any instance of Activity Final node, the entire activity UML 2.0 has an additional control node type called Flow Final (shown below in a diagram taken from ) that is used as an alternative to the Activity Final node to terminate aįlow. ![]() Shown with an arrowed line, a visual analog to the UML 1.x transition arrow. Nodes - decision, merge, fork, join, initial and final - look like their UML 1.x equivalents, and the control flows are No semantic significance), and this model has the same execution result in UML 1.x and UML 2.0. They are similar in appearance (allowing for the differing orientation and color conventions used-these have Compare this diagram with the UML 1.x version shownīelow it. Rectangular frame and a name in a compartment at the upper left. The diagram below illustrates many of the UML 2.0 elements, and is presented in the usual way for UML 2.0, with a This puts the onus on the modeler to be aware of race conditions and interactions.Īlso, see the section Semantic Differences below for another example of the effect on concurrency Invocations of an activity to be handled by a single execution with multiple streams of tokens moving through the nodesĪnd flow connectors of the activity. (activity) performed a run-to-completion step, the UML 2.0 capability, in its most complete form, permits multiple The modeling capability of UML 2.0 allows unrestricted parallelism: whereas in UML 1.x, the entire state machine This contrasts with UML 1.x, where the nodes were states (or pseudo states) with transitions between them, which The flows connecting nodes are further refined into control and data or object flows and, as you mightĮxpect, control tokens move across control flows and object or data tokens pass across object flows. Tokens are met, and when it completes execution, it offers tokens on its output flows, so that downstream nodes mayīegin execution. A node is allowed to begin execution when specified conditions on its input Tokens, containing objects or a locus of control, flowīetween nodes across these connections. In UML 2.0,Īctivities are composed of nodes, of which actions are one kind others, described further below, areĪctivities now have Petri Net-like semantics, based on token flow, where the execution of one node affects theĮxecution of another through directed connections called flows. The connotation of state hasĭisappeared in UML 2.0 because an activity is no longer a kind of state machine, as it was in UML 1.x. We may have informally referred to the individual executable steps in a UML 1.x activityĭiagram as activities or activity states or, correctly, as action states: now these steps in a UML 2.0 activity areĬalled actions - and these actions are not decomposed further within the activity. Refer to for moreĪs defines it, an activity (which will be shown in an activityĭiagram) is the specification of behavior as the coordinated sequencing of subordinate units whose individualĮlements are actions. In the case of more complex models involving concurrency. Therefore we caution the modeler that even when a UML 1.xĪctivity model appears to be acceptable to UML 2.0 without change, it might not execute in the same way - particularly Use, the effect and appearance might be very similar, although depending on the formality of modeling in UML 1.5 (andĮarlier versions), it is possible that the strict interpretation and the execution result of a model constructedĪccording to UML 1.x rules would not be the same in UML 2.0. It is fair to say that, at least for casual The modeling of activities has undergone a complete revision in UML 2.0. The most significant diagrammatical changes in the UML 2.0 feature set are in the behavioral diagrams, specifically theĪctivity diagram and the set of interaction diagrams (see Activity Diagram, Sequence Diagram and Communication Diagram below).Ĭomposite Structure Diagram and Structured Class are also new UML 2.0 features (see Composite Structure Diagram below). Note that "UML 1.x" refers to UML 1.0 to UML 1.5 versions. Infrastructure and Superstructure Specifications, but to give an overview of relevant UML capabilities instead. This page describes some differences from UML 1.x and UML 2.0 that are relevant to RUP context.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |