COIN-OR::LEMON - Graph Library

source: lemon-0.x/doc/graphs.dox @ 666:410a1419e86b

Last change on this file since 666:410a1419e86b was 666:410a1419e86b, checked in by Akos Ladanyi, 16 years ago

Added a short tutorial on using graphs.

File size: 3.9 KB
Line 
1/*!
2
3\page graphs How to use graphs
4
5The following program demonstrates the basic features of HugoLib's graph
6structures.
7
8\code
9#include <iostream>
10#include <hugo/list_graph.h>
11
12using namespace hugo;
13
14int main()
15{
16  typedef ListGraph Graph;
17\endcode
18
19ListGraph is one of HugoLib's graph classes. It is based on linked lists,
20therefore iterating throuh its edges and nodes is fast.
21
22\code
23  typedef Graph::Edge Edge;
24  typedef Graph::InEdgeIt InEdgeIt;
25  typedef Graph::OutEdgeIt OutEdgeIt;
26  typedef Graph::EdgeIt EdgeIt;
27  typedef Graph::Node Node;
28  typedef Graph::NodeIt NodeIt;
29
30  Graph g;
31 
32  for (int i = 0; i < 3; i++)
33    g.addNode();
34 
35  for (NodeIt i(g); g.valid(i); g.next(i))
36    for (NodeIt j(g); g.valid(j); g.next(j))
37      if (i != j) g.addEdge(i, j);
38\endcode
39
40After some convenience typedefs we create a graph and add three nodes to it.
41Then we add edges to it to form a full graph.
42
43\code
44  std::cout << "Nodes:";
45  for (NodeIt i(g); g.valid(i); g.next(i))
46    std::cout << " " << g.id(i);
47  std::cout << std::endl;
48\endcode
49
50Here we iterate through all nodes of the graph. We use a constructor of the
51node iterator to initialize it to the first node. The next member function is
52used to step to the next node, and valid is used to check if we have passed the
53last one.
54
55\code
56  std::cout << "Nodes:";
57  NodeIt n;
58  for (g.first(n); n != INVALID; g.next(n))
59    std::cout << " " << g.id(n);
60  std::cout << std::endl;
61\endcode
62
63Here you can see an alternative way to iterate through all nodes. Here we use a
64member function of the graph to initialize the node iterator to the first node
65of the graph. Using next on the iterator pointing to the last node invalidates
66the iterator i.e. sets its value to INVALID. Checking for this value is
67equivalent to using the valid member function.
68
69Both of the previous code fragments print out the same:
70
71\code
72Nodes: 2 1 0
73\endcode
74
75\code
76  std::cout << "Edges:";
77  for (EdgeIt i(g); g.valid(i); g.next(i))
78    std::cout << " (" << g.id(g.tail(i)) << "," << g.id(g.head(i)) << ")";
79  std::cout << std::endl;
80\endcode
81
82\code
83Edges: (0,2) (1,2) (0,1) (2,1) (1,0) (2,0)
84\endcode
85
86We can also iterate through all edges of the graph very similarly. The head and
87tail member functions can be used to access the endpoints of an edge.
88
89\code
90  NodeIt first_node(g);
91
92  std::cout << "Out-edges of node " << g.id(first_node) << ":";
93  for (OutEdgeIt i(g, first_node); g.valid(i); g.next(i))
94    std::cout << " (" << g.id(g.tail(i)) << "," << g.id(g.head(i)) << ")";
95  std::cout << std::endl;
96
97  std::cout << "In-edges of node " << g.id(first_node) << ":";
98  for (InEdgeIt i(g, first_node); g.valid(i); g.next(i))
99    std::cout << " (" << g.id(g.tail(i)) << "," << g.id(g.head(i)) << ")";
100  std::cout << std::endl;
101\endcode
102
103\code
104Out-edges of node 2: (2,0) (2,1)
105In-edges of node 2: (0,2) (1,2)
106\endcode
107
108We can also iterate through the in and out-edges of a node. In the above
109example we print out the in and out-edges of the first node of the graph.
110
111\code
112  Graph::EdgeMap<int> m(g);
113
114  for (EdgeIt e(g); g.valid(e); g.next(e))
115    m.set(e, 10 - g.id(e));
116 
117  std::cout << "Id Edge  Value" << std::endl;
118  for (EdgeIt e(g); g.valid(e); g.next(e))
119    std::cout << g.id(e) << "  (" << g.id(g.tail(e)) << "," << g.id(g.head(e))
120      << ") " << m[e] << std::endl;
121\endcode
122
123\code
124Id Edge  Value
1254  (0,2) 6
1262  (1,2) 8
1275  (0,1) 5
1280  (2,1) 10
1293  (1,0) 7
1301  (2,0) 9
131\endcode
132
133In generic graph optimization programming graphs are not containers rather
134incidence structures which are iterable in many ways. HugoLib introduces
135concepts that allow us to attach containers to graphs. These containers are
136called maps.
137
138In the example above we create an EdgeMap which assigns an int value to all
139edges of the graph. We use the set member function of the map to write values
140into the map and the operator[] to retrieve them.
141
142Here we used the maps provided by the ListGraph class, but you can also write
143your own maps. You can read more about using maps \ref maps "here".
144
145*/
Note: See TracBrowser for help on using the repository browser.