ladanyi@666
|
1 |
/*!
|
ladanyi@666
|
2 |
|
ladanyi@1638
|
3 |
\page graphs Graphs
|
ladanyi@666
|
4 |
|
alpar@921
|
5 |
The primary data structures of LEMON are the graph classes. They all
|
alpar@756
|
6 |
provide a node list - edge list interface, i.e. they have
|
alpar@756
|
7 |
functionalities to list the nodes and the edges of the graph as well
|
athos@1168
|
8 |
as incoming and outgoing edges of a given node.
|
alpar@756
|
9 |
|
alpar@756
|
10 |
|
alpar@873
|
11 |
Each graph should meet the
|
klao@959
|
12 |
\ref lemon::concept::StaticGraph "StaticGraph" concept.
|
alpar@873
|
13 |
This concept does not
|
athos@1168
|
14 |
make it possible to change the graph (i.e. it is not possible to add
|
alpar@756
|
15 |
or delete edges or nodes). Most of the graph algorithms will run on
|
alpar@756
|
16 |
these graphs.
|
alpar@756
|
17 |
|
alpar@873
|
18 |
The graphs meeting the
|
klao@959
|
19 |
\ref lemon::concept::ExtendableGraph "ExtendableGraph"
|
alpar@873
|
20 |
concept allow node and
|
athos@1168
|
21 |
edge addition. You can also "clear" such a graph (i.e. erase all edges and nodes ).
|
alpar@756
|
22 |
|
alpar@873
|
23 |
In case of graphs meeting the full feature
|
klao@959
|
24 |
\ref lemon::concept::ErasableGraph "ErasableGraph"
|
alpar@873
|
25 |
concept
|
athos@1168
|
26 |
you can also erase individual edges and nodes in arbitrary order.
|
alpar@756
|
27 |
|
alpar@756
|
28 |
The implemented graph structures are the following.
|
alpar@921
|
29 |
\li \ref lemon::ListGraph "ListGraph" is the most versatile graph class. It meets
|
klao@959
|
30 |
the \ref lemon::concept::ErasableGraph "ErasableGraph" concept
|
athos@1168
|
31 |
and it also has some convenient extra features.
|
alpar@921
|
32 |
\li \ref lemon::SmartGraph "SmartGraph" is a more memory
|
alpar@921
|
33 |
efficient version of \ref lemon::ListGraph "ListGraph". The
|
athos@1168
|
34 |
price of this is that it only meets the
|
klao@959
|
35 |
\ref lemon::concept::ExtendableGraph "ExtendableGraph" concept,
|
alpar@756
|
36 |
so you cannot delete individual edges or nodes.
|
alpar@921
|
37 |
\li \ref lemon::FullGraph "FullGraph"
|
alpar@1200
|
38 |
implements a complete graph. It is a
|
alpar@1200
|
39 |
\ref lemon::concept::StaticGraph "StaticGraph", so you cannot
|
alpar@756
|
40 |
change the number of nodes once it is constructed. It is extremely memory
|
alpar@756
|
41 |
efficient: it uses constant amount of memory independently from the number of
|
alpar@1043
|
42 |
the nodes of the graph. Of course, the size of the \ref maps-page "NodeMap"'s and
|
alpar@1043
|
43 |
\ref maps-page "EdgeMap"'s will depend on the number of nodes.
|
alpar@756
|
44 |
|
alpar@921
|
45 |
\li \ref lemon::NodeSet "NodeSet" implements a graph with no edges. This class
|
alpar@921
|
46 |
can be used as a base class of \ref lemon::EdgeSet "EdgeSet".
|
alpar@921
|
47 |
\li \ref lemon::EdgeSet "EdgeSet" can be used to create a new graph on
|
alpar@873
|
48 |
the node set of another graph. The base graph can be an arbitrary graph and it
|
alpar@921
|
49 |
is possible to attach several \ref lemon::EdgeSet "EdgeSet"'s to a base graph.
|
alpar@756
|
50 |
|
alpar@756
|
51 |
\todo Don't we need SmartNodeSet and SmartEdgeSet?
|
alpar@756
|
52 |
\todo Some cross-refs are wrong.
|
alpar@756
|
53 |
|
athos@1168
|
54 |
The graph structures themselves can not store data attached
|
alpar@756
|
55 |
to the edges and nodes. However they all provide
|
alpar@1043
|
56 |
\ref maps-page "map classes"
|
alpar@756
|
57 |
to dynamically attach data the to graph components.
|
alpar@756
|
58 |
|
alpar@921
|
59 |
The following program demonstrates the basic features of LEMON's graph
|
ladanyi@666
|
60 |
structures.
|
ladanyi@666
|
61 |
|
ladanyi@666
|
62 |
\code
|
ladanyi@666
|
63 |
#include <iostream>
|
alpar@921
|
64 |
#include <lemon/list_graph.h>
|
ladanyi@666
|
65 |
|
alpar@921
|
66 |
using namespace lemon;
|
ladanyi@666
|
67 |
|
ladanyi@666
|
68 |
int main()
|
ladanyi@666
|
69 |
{
|
ladanyi@666
|
70 |
typedef ListGraph Graph;
|
ladanyi@666
|
71 |
\endcode
|
ladanyi@666
|
72 |
|
alpar@921
|
73 |
ListGraph is one of LEMON's graph classes. It is based on linked lists,
|
ladanyi@666
|
74 |
therefore iterating throuh its edges and nodes is fast.
|
ladanyi@666
|
75 |
|
ladanyi@666
|
76 |
\code
|
ladanyi@666
|
77 |
typedef Graph::Edge Edge;
|
ladanyi@666
|
78 |
typedef Graph::InEdgeIt InEdgeIt;
|
ladanyi@666
|
79 |
typedef Graph::OutEdgeIt OutEdgeIt;
|
ladanyi@666
|
80 |
typedef Graph::EdgeIt EdgeIt;
|
ladanyi@666
|
81 |
typedef Graph::Node Node;
|
ladanyi@666
|
82 |
typedef Graph::NodeIt NodeIt;
|
ladanyi@666
|
83 |
|
ladanyi@666
|
84 |
Graph g;
|
ladanyi@666
|
85 |
|
ladanyi@666
|
86 |
for (int i = 0; i < 3; i++)
|
ladanyi@666
|
87 |
g.addNode();
|
ladanyi@666
|
88 |
|
ladanyi@875
|
89 |
for (NodeIt i(g); i!=INVALID; ++i)
|
ladanyi@875
|
90 |
for (NodeIt j(g); j!=INVALID; ++j)
|
ladanyi@666
|
91 |
if (i != j) g.addEdge(i, j);
|
ladanyi@666
|
92 |
\endcode
|
ladanyi@666
|
93 |
|
athos@1168
|
94 |
After some convenient typedefs we create a graph and add three nodes to it.
|
athos@1168
|
95 |
Then we add edges to it to form a complete graph.
|
ladanyi@666
|
96 |
|
ladanyi@666
|
97 |
\code
|
ladanyi@666
|
98 |
std::cout << "Nodes:";
|
ladanyi@875
|
99 |
for (NodeIt i(g); i!=INVALID; ++i)
|
ladanyi@666
|
100 |
std::cout << " " << g.id(i);
|
ladanyi@666
|
101 |
std::cout << std::endl;
|
ladanyi@666
|
102 |
\endcode
|
ladanyi@666
|
103 |
|
ladanyi@666
|
104 |
Here we iterate through all nodes of the graph. We use a constructor of the
|
ladanyi@875
|
105 |
node iterator to initialize it to the first node. The operator++ is used to
|
ladanyi@875
|
106 |
step to the next node. Using operator++ on the iterator pointing to the last
|
ladanyi@875
|
107 |
node invalidates the iterator i.e. sets its value to
|
alpar@921
|
108 |
\ref lemon::INVALID "INVALID". This is what we exploit in the stop condition.
|
ladanyi@666
|
109 |
|
ladanyi@875
|
110 |
The previous code fragment prints out the following:
|
ladanyi@666
|
111 |
|
ladanyi@666
|
112 |
\code
|
ladanyi@666
|
113 |
Nodes: 2 1 0
|
ladanyi@666
|
114 |
\endcode
|
ladanyi@666
|
115 |
|
ladanyi@666
|
116 |
\code
|
ladanyi@666
|
117 |
std::cout << "Edges:";
|
ladanyi@875
|
118 |
for (EdgeIt i(g); i!=INVALID; ++i)
|
alpar@986
|
119 |
std::cout << " (" << g.id(g.source(i)) << "," << g.id(g.target(i)) << ")";
|
ladanyi@666
|
120 |
std::cout << std::endl;
|
ladanyi@666
|
121 |
\endcode
|
ladanyi@666
|
122 |
|
ladanyi@666
|
123 |
\code
|
ladanyi@666
|
124 |
Edges: (0,2) (1,2) (0,1) (2,1) (1,0) (2,0)
|
ladanyi@666
|
125 |
\endcode
|
ladanyi@666
|
126 |
|
athos@1168
|
127 |
We can also iterate through all edges of the graph very similarly. The
|
athos@1168
|
128 |
\c target and
|
athos@1168
|
129 |
\c source member functions can be used to access the endpoints of an edge.
|
ladanyi@666
|
130 |
|
ladanyi@666
|
131 |
\code
|
ladanyi@666
|
132 |
NodeIt first_node(g);
|
ladanyi@666
|
133 |
|
ladanyi@666
|
134 |
std::cout << "Out-edges of node " << g.id(first_node) << ":";
|
ladanyi@875
|
135 |
for (OutEdgeIt i(g, first_node); i!=INVALID; ++i)
|
alpar@986
|
136 |
std::cout << " (" << g.id(g.source(i)) << "," << g.id(g.target(i)) << ")";
|
ladanyi@666
|
137 |
std::cout << std::endl;
|
ladanyi@666
|
138 |
|
ladanyi@666
|
139 |
std::cout << "In-edges of node " << g.id(first_node) << ":";
|
ladanyi@875
|
140 |
for (InEdgeIt i(g, first_node); i!=INVALID; ++i)
|
alpar@986
|
141 |
std::cout << " (" << g.id(g.source(i)) << "," << g.id(g.target(i)) << ")";
|
ladanyi@666
|
142 |
std::cout << std::endl;
|
ladanyi@666
|
143 |
\endcode
|
ladanyi@666
|
144 |
|
ladanyi@666
|
145 |
\code
|
ladanyi@666
|
146 |
Out-edges of node 2: (2,0) (2,1)
|
ladanyi@666
|
147 |
In-edges of node 2: (0,2) (1,2)
|
ladanyi@666
|
148 |
\endcode
|
ladanyi@666
|
149 |
|
ladanyi@666
|
150 |
We can also iterate through the in and out-edges of a node. In the above
|
ladanyi@666
|
151 |
example we print out the in and out-edges of the first node of the graph.
|
ladanyi@666
|
152 |
|
ladanyi@666
|
153 |
\code
|
ladanyi@666
|
154 |
Graph::EdgeMap<int> m(g);
|
ladanyi@666
|
155 |
|
ladanyi@875
|
156 |
for (EdgeIt e(g); e!=INVALID; ++e)
|
ladanyi@666
|
157 |
m.set(e, 10 - g.id(e));
|
ladanyi@666
|
158 |
|
ladanyi@666
|
159 |
std::cout << "Id Edge Value" << std::endl;
|
ladanyi@875
|
160 |
for (EdgeIt e(g); e!=INVALID; ++e)
|
alpar@986
|
161 |
std::cout << g.id(e) << " (" << g.id(g.source(e)) << "," << g.id(g.target(e))
|
ladanyi@666
|
162 |
<< ") " << m[e] << std::endl;
|
ladanyi@666
|
163 |
\endcode
|
ladanyi@666
|
164 |
|
ladanyi@666
|
165 |
\code
|
ladanyi@666
|
166 |
Id Edge Value
|
ladanyi@666
|
167 |
4 (0,2) 6
|
ladanyi@666
|
168 |
2 (1,2) 8
|
ladanyi@666
|
169 |
5 (0,1) 5
|
ladanyi@666
|
170 |
0 (2,1) 10
|
ladanyi@666
|
171 |
3 (1,0) 7
|
ladanyi@666
|
172 |
1 (2,0) 9
|
ladanyi@666
|
173 |
\endcode
|
ladanyi@666
|
174 |
|
alpar@873
|
175 |
As we mentioned above, graphs are not containers rather
|
alpar@921
|
176 |
incidence structures which are iterable in many ways. LEMON introduces
|
ladanyi@666
|
177 |
concepts that allow us to attach containers to graphs. These containers are
|
ladanyi@666
|
178 |
called maps.
|
ladanyi@666
|
179 |
|
athos@1168
|
180 |
In the example above we create an EdgeMap which assigns an integer value to all
|
ladanyi@666
|
181 |
edges of the graph. We use the set member function of the map to write values
|
ladanyi@666
|
182 |
into the map and the operator[] to retrieve them.
|
ladanyi@666
|
183 |
|
ladanyi@666
|
184 |
Here we used the maps provided by the ListGraph class, but you can also write
|
alpar@1043
|
185 |
your own maps. You can read more about using maps \ref maps-page "here".
|
ladanyi@666
|
186 |
|
ladanyi@666
|
187 |
*/
|