Tutorial: UML Part 1 -- Class Diagrams.
Robert C. Martin My next several columns will be a running tutorial of UML. The 1.0 version of UML was released on the 13th of January, 1997. The 1.1 release should be out before the end of the year. This column will track the progress of UML and present the issues that the three amigos (Grady Booch, Jim Rumbaugh, and Ivar Jacobson) are dealing with.
Introduction
UML stands for Unied Modeling Language. It represents a unication of the concepts and notations presented by the three amigos in their respective books1. The goal is for UML to become a common language for creating models of object oriented computer software. In its current form UML is comprised of two major components: a Meta-model and a notation. In the future, some form of method or process may also be added to; or associated with, UML. The Meta-model UML is unique in that it has a standard data representation. This representation is called the metamodel. The meta-model is a description of UML in UML. It describes the objects, attributes, and relationships necessary to represent the concepts of UML within a software application. This provides CASE manufacturers with a standard and unambiguous way to represent UML models. Hopefully it will allow for easy transport of UML models between tools. It may also make it easier to write ancillary tools for browsing, summarizing, and modifying UML models. A deeper discussion of the metamodel is beyond the scope of this column. Interested readers can learn more about it by downloading the UML documents from the rational web site2. The Notation The UML notation is rich and full bodied. It is comprised of two major subdivisions. There is a notation for modeling the static elements of a design such as classes, attributes, and relationships. There is also a notation for modeling the dynamic elements of a design such as objects, messages, and nite state machines. In this article we will present some of the aspects of the static modeling notation. Static models are presented in diagrams called: Class Diagrams.
Class Diagrams.
The purpose of a class diagram is to depict the classes within a model. In an object oriented application, classes have attributes (member variables), operations (member functions) and relation1. Object Oriented Analysis and Design, Grady Booch, Benjamin Cummings, 1994. Object Oriented Modeling and Design, James Rumbaugh, et. al., Prentice Hall, 1991 Object Oriented Software Engineering, Ivar Jacobson, et. al., Addison Wesley, 1992 2. http://www.rational.com
ships with other classes. The UML class diagram can depict all these things quite easily. The fundamental element of the class diagram is an icon the represents a class. This icon is shown in Figure 1. Figure 1: The Class Icon
Class
Attribute operation()
A class icon is simply a rectangle divided into three compartments. The topmost compartment contains the name of the class. The middle compartment contains a list of attributes (member variables), and the bottom compartment contains a list of operations (member functions). In many diagrams, the bottom two compartments are omitted. Even when they are present, they typically do not show every attribute and operations. The goal is to show only those attributes and operations that are useful for the particular diagram. This ability to abbreviate an icon is one of the hallmarks of UML. Each diagram has a particular purpose. That purpose may be to highlight on particular part of the system, or it may be to illuminate the system in general. The class icons in such diagrams are abbreviated as necessary. There is typically never a need to show every attribute and operation of a class on any diagram. Figure 2 shows a typical UML description of a class that represents a circle. Figure 2: Circle class
Circle
itsRadius:double itsCenter:Point Area():double Circumference():double SetCenter(Point) SetRadius(double)
Notice that each member variable is followed by a colon and by the type of the variable. If the type is redundant, or otherwise unnecessary, it can be omitted. Notice also that the return values follow the member functions in a similar fashion. Again, these can be omitted. Finally, notice that the member function arguments are just types. I could have named them too, and used colons to separate them from their types; or I could have omitted the arguments altogether. Composition Relationships Each instance of type Circle seems to contain an instance of type Point. This is a relationship known as composition. It can be depicted in UML using a class relationship. Figure 3 shows the composition relationship.
Figure 3: Circle contains Point
Circle
Point
The black diamond represents composition. It is placed on the Circle class because it is the Circle that is composed of a Point. The arrowhead on the other end of the relationship denotes that the relationship is navigable in only one direction. That is, Point does not know about Circle. In UML relationships are presumed to be bidirectional unless the arrowhead is present to restrict them. Had I omitted the arrowhead, it would have meant that Point knew about Circle. At the code level, this would imply a #include circle.h within point.h. For this reason, I tend to use a lot of arrowheads. Composition relationships are a strong form of containment or aggregation. Aggregation is a whole/part relationship. In this case, Circle is the whole, and Point is part of Circle. However, composition is more than just aggregation. Composition also indicates that the lifetime of Point is dependent upon Circle. This means that if Circle is destroyed, Point will be destroyed with it. For those of you who are familiar with the Booch-94 notation, this is the Hasby-value relationship. In C++ we would represent this as shown in Listing 1. Listing 1: Circle class
class Circle { public: void SetCenter(const Point&); void SetRadius(double); double Area() const; double Circumference() const; private: double itsRadius; Point itsCenter; };
In this case we have represented the composition relationship as a member We variable. could also have used a pointer so long as the destructor of Circle deleted the pointer. Inheritance The inheritance relationship in UML is depicted by a peculiar triangular arrowhead. This arrowhead, that looks rather like a slice of pizza, points to the base class. One or more lines proceed from the base of the arrowhead connecting it to the derived classes. Figure 4 shows the form of the inheritance relationship. In this diagram we see that Circle and Square both derive from Shape. Note that the name of class Shape is shown in italics. This indicates that Shape is an abstract class. Note also that the operations, Draw() and Erase() are also shown in italics. This indicates that they are pure virtual.
Figure 4: Inheritance
Shape
{abstract}
Draw() Erase()
Circle
Square
Italics are not always very easy to see. Therefore, as shown in Figure 4, an abstract class can also be marked with the {abstract} property. Whats more, though it is not a standard part of UML, I will often write Draw()=0 in the operations compartment to denote a pure virtual function. Aggregation / Association The weak form of aggregation is denoted with an open diamond. This relationship denotes that the aggregate class (the class with the white diamond touching it) is in some way the whole, and the other class in the relationship is somehow part of that whole. Figure 5: Aggregation
Window
*
itsShapes
Shape
{abstrac
Figure 5 shows an aggregation relationship. In this case, the Window class contains many Shape instances. In UML the ends of a relationship are referred to as its roles. Notice that the role at the Shape end of the aggregation is marked with a *. This indicates that the Window contains many Shape instances. Notice also that the role has been named. This is the name that Window knows its Shape instances by. i.e. it is the name of the instance variable within Window that holds all the Shapes.
Listing 2 shows how Figure 5 might be implemented in C++ Listing 2: Window contains Shapes class Window
{ public: //... private: vector<Shape*> itsShapes; };
There are other forms of containment that do not have whole / part implications. For example, Each Window refers back to its parent Frame. This is not aggregation since it is not reasonable to consider a parent Frame to be part of a child Window. We use the association relationship to depict this. Figure 6: Associations
itsParent
Frame
Window
Figure 6 shows how we draw an association. An association is nothing but a line drawn between the participating classes. In Figure 6 the association has an arrowhead to denote that Frame does not know anything about Window. Once again note the name on the role. This relationship will almost certainly be implemented with a pointer of some kind. What is the difference between an aggregation and an association? The difference is one of implication. Aggregation denotes whole/part relationships whereas associations do not. However, there is not likely to be much difference in the way that the two relationships are implemented. That is, it would be very difcult to look at the code and determine whether a particular relationship ought to be aggregation or association. For this reason, it is pretty safe to ignore the aggregation relationship altogether. As the amigos said in the UML 0.8 document: ...if you dont understand [aggregation] dont use it. Aggregation and Association both correspond to the Has-by-reference relationship from the Booch-94 notation.
Dependency Sometimes the relationship between a two classes is very weak. They are not implemented with member variables at all. Rather they might be implemented a...