source: lemon-0.x/doc/maps.dox @ 621:2db02d4a9e6e

/*!

\page maps How to write your own maps

\section read-maps Readable Maps

The readable maps are very frequently used as the input of the
algorithms.  For this purpose the most straightforward is to use the
maps provided by Hugo's graph structres. Very often however, it is more
convenient and/or more efficient to write your own readable map.

You can find some example below.

This simple map assigns \f$\pi\f$ to each edge.

\code
struct MyMap 
{
  typedef double ValueType;
  double operator[](Graph::Edge e) const { return M_PI;}
};
\endcode

An alternative way to define maps. For this, \c MapBase seems to
be a better name then \c NullMap

\code
struct MyMap : public MapBase<Edge,double>
{
  double operator[](Graph::Edge e) const { return M_PI;}
};
\endcode

Or, if we had \c KeyType and \c ValueType

\code
struct MyMap : public MapBase<Edge,double>
{
  ValueType operator[](KeyType e) const { return M_PI;}
};
\endcode


Here is a more complex example. It provides a length function which is obtained
from a base length function modified by a potential difference.

\code
class MyLengthMap 
{
  const Graph::EdgeMap &ol;
  const Graph::NodeMap &pot;
  
public:
  typedef double ValueType;

  double operator[](Graph::Edge e) const {
    return ol.get(e)-pot.get(v)-pot.get(u);
  }
  
  MyComplexMap(const Graph::EdgeMap &o,const Graph::NodeMap &p) :
    ol(o), pot(p);
};
\endcode

\todo Please improve on the english. 
\todo Don't we need \e to \e require a 'typedef xxx KeyType' tag, as well?
*/