Boost example graph program

A topological sort works on a directed acyclic graph DAG. If a cyclic graph is passed in, then an exception is thrown, thus indicating that the graph has a circular dependency. Using the Boost Graph Library. These code samples listed on here appear reasonably up to date and appear to compile and work fine. I am finding that some of the online documentation concerning the use of the Boost Graph Library seems to be out of date or produces compilation errors.

There are a number of working examples here including creating directed and undirected graphs, printing the weights of edges, finding minimal spanning trees using Kruskal's algorithm, and maximum flow problems. Stack Overflow for Teams — Collaborate and share knowledge with a private group.

Create a free Team What is Teams? Collectives on Stack Overflow. Learn more. Asked 10 years ago. Active 2 years, 7 months ago. Viewed 47k times. How do you make a graph, and add vertices and edges as you go? Improve this question. Jim Jim 3, 7 7 gold badges 31 31 silver badges 41 41 bronze badges. Possible duplicate of stackoverflow.

Add a comment. Active Oldest Votes. Improve this answer. Richard Liam M Liam M 5, 2 2 gold badges 38 38 silver badges 52 52 bronze badges. I've put that code on ideone I had to tweak it a little for some typos. It keeps track of how many objects point to a particular resource and deletes the resource automatically when nothing is pointing to it.

The following example has two classes, A and B. If we had used pointers to achieve such sharing of an int , each class would have had a hard time figuring out when it should be deleted. In the example, until the end of main , the reference count is 3. If all of the pointers go out of scope, the reference count reaches 0 , allowing the shared instance of int to be deleted.

When the reference count reaches zero, the internal pointer is deleted and the memory is released. For most dynamically allocated objects, delete is used. However, some resources require more complex cleanup. The deleter determines how to destroy the resource. For dynamically allocated arrays , we shouldn't use either of them because they use delete in their destructor but not delete[]. We can use vector instead. A pointer to a single object with one owner.

As we see in the code above, we allocate Implementation object inside the Timer constructor, and initialize the private member, pImpl of the class. This approach is error-prone, and every now an then we may forget to delete the object in our destructor. Using smart pointers make easier. In other words, we could use a shared pointer or a scoped pointer to hold the Implementation object pointer.

There are two ways to do this. In the example below, we define classes Singer and Song. Each class has a pointer to an instance of the other class. This creates a circular reference between the two classes. Classes Singer and Song define destructors that each display a message to indicate when an instance of either class is destroyed.

Each class also defines a member function to print the title of the Song and Singer's name. Therefore, you must initialize all elements in the array distances before you start. This is the shortest path from top left to bottom right. It leads over the top right field although the path over the bottom left field would be equally short. As you already know, this algorithm just visits points. To get a description of the shortest path, an appropriate visitor must be used.

To find the first field when travelling from the top left to the bottom right, the element with the index 0 — the index of the top left field — is accessed in predecessors. The value found in predecessors is 1, which means the next field is at the top right. Accessing predecessors with the index 1 returns the next field. If more than two visitors are needed, the pairs have to be nested.

This means the time taken to cross any line between points is always the same. If lines are weighted, meaning that each line may require a different amount of time to traverse, then you need to use a different algorithm to find the shortest path. This algorithm is used if lines are weighted. This is done with the array weights. The elements in the array correspond to the lines in the graph. Because the line from the top left to the top right is first, the first element in weights is set to a value twice as big as all others.

To assign weights to lines, the iterator to the beginning of the array weights is passed as the third parameter to the constructor of the graph. This third parameter can be used to initialize properties of lines. This only works if properties have been defined for lines. The fourth and fifth template parameter specify if points and lines have properties and what those properties are. You can assign properties to both lines and points.

The fifth parameter uses boost::property to define a bundled property. Bundled properties are properties that are stored internally in a graph. The second template parameter passed to boost::property is the type of the property.

You only need to pass a container to store the predecessor of every point. This is a helper function which expects a pointer or an iterator to the beginning of an array. The path over the top right field has a greater weight than the other possibilities. It is possible to add more member variables if other properties are required.