11.3.1 A graph representing cities and roads

Next we describe a graph representing cities and roads between cities. This graph is represented by a class RoadAndCityGraph, which is a subclass of Graph:

   class RoadAndCityGraph: Graph 
      class Node::< 
         cityName -> cn: ref String:
            cn := id	
         display::< 
             ...
      class Edge::<
         roadLength: var Integer
         display::
            ...
      addCity(nm: var String) -> n: ref Node: 
         :::
      addRoad(from, to: ref Node, dist: var integer) -> e: ref Edge:
         :::	 
      inner(RoadAndCityGraph)

RoadAndCityGraph has the following attributes:

• A further binding of the virtual classes extended the descriptions of Node and Edge from Graph. A Node represents a city and an Edge represents a road.
• A method addCity and a method addRoad.

The further binding of Node adds the following attributes:

• A method cityName which returns the id of the node – the id is used to represent the name of the city.
• A further binding of display, we which extends the description of display from Node.
• Note that no further binding of addEdge is included here.

The further binding of Edge adds the following attributes:

• A integer variable roadLength holding the length of the road.
• A further binding of the display method.

Below we show the details of addCity and addNode:

      addCity(nm: var String) -> n: ref Node:
         n := addNode(nm)
      addRoad(from, to: ref Node, dist: var integer) -> e: ref Edge:
         e := from.addEdge(to)
         e.roadLength := dist

• AddCity has the name, nm, of the city as parameter and adds a Node with nm as argument. The new Node is returned as the value of addNode.
• AddRoad has the two cities being connected by the road and the distance between them as parameters: from, to, and dist.

We may now use class RoadAndCityGraph to define a region with some cities and roads. Herre we use Europe and Paris, London and Berlin:

   Europe: obj RoadAndCityGraph
      setUp:
         n1, n2: ref Node
         e: ref Edge
         n1 := addCity("Paris")	 
         n2 := addCity("London")
         e := addRoad(n1,n2,291) -- 291 miles
         e := addRoad(n2,n1,291)
         n2 := addCity("Berlin")
         e := addRoad(n1,n2,1054) -- 1054 km
   Europe.setUp

The example should be self explanatory.

Note, however, as a remainder of the importance of being aware of units of quantities, we have deliberately shown the distance from Paris to London in miles and the distance from Paris to Berlin in kilometers. When you use Google Maps (at least at the location of the authors) you get the distance from Paris to London in miles and the one from Paris to Berlin in kilometers. In section , it is show to be explicit about the units represented by numbers.

Exerises

1. Shortest path. Given two cities A and B, write a method that computes the shorts route from A to B.
2. Traveling salesman. For a given graph, write a method that computes the shortest path that visits/included all nodes in the graph.

(2) is called the travelling salesman since it is inspired by a salesman that needs to visit all cities in a given region and use the shortest route between them. It is due to the famous computer scientist E. Dijkstra +++ reference.

These methods are not straight forward to write, and are examples of complex algorithms that may require the reader to study the topic of algorithms and data structures as mentioned in the introduction.

For small graphs, simple algorithms may be usable, but for large graph, it is necessary to find a algorithms that are efficient with respect to the time it takes to compute them.