doc/graph-adaptors.dox
author ladanyi
Wed, 02 Jul 2008 12:37:47 +0000
changeset 2615 2bf1f6e3d5ae
parent 2391 14a343be7a5a
permissions -rw-r--r--
Fix bug caused by m4 consuming pairs of square brackets (#108).
     1 /* -*- C++ -*-
     2  *
     3  * This file is a part of LEMON, a generic C++ optimization library
     4  *
     5  * Copyright (C) 2003-2008
     6  * Egervary Jeno Kombinatorikus Optimalizalasi Kutatocsoport
     7  * (Egervary Research Group on Combinatorial Optimization, EGRES).
     8  *
     9  * Permission to use, modify and distribute this software is granted
    10  * provided that this copyright notice appears in all copies. For
    11  * precise terms see the accompanying LICENSE file.
    12  *
    13  * This software is provided "AS IS" with no warranty of any kind,
    14  * express or implied, and with no claim as to its suitability for any
    15  * purpose.
    16  *
    17  */
    18 
    19 /**
    20    @defgroup graph_adaptors Adaptor Classes for Graphs
    21    @ingroup graphs
    22    \brief This group contains several adaptor classes for graphs
    23    
    24    The main parts of LEMON are the different graph structures, 
    25    generic graph algorithms, graph concepts which couple these, and 
    26    graph adaptors. While the previous notions are more or less clear, the 
    27    latter one needs further explanation. 
    28    Graph adaptors are graph classes which serve for considering graph 
    29    structures in different ways. 
    30 
    31    A short example makes this much 
    32    clearer. 
    33    Suppose that we have an instance \c g of a directed graph
    34    type say ListGraph and an algorithm 
    35    \code template<typename Graph> int algorithm(const Graph&); \endcode 
    36    is needed to run on the reversed oriented graph. 
    37    It may be expensive (in time or in memory usage) to copy 
    38    \c g with the reversed orientation. 
    39    In this case, an adaptor class is used, which 
    40    (according to LEMON graph concepts) works as a graph. 
    41    The adaptor uses 
    42    the original graph structure and graph operations when methods of the 
    43    reversed oriented graph are called. 
    44    This means that the adaptor have minor memory usage, 
    45    and do not perform sophisticated algorithmic actions. 
    46    The purpose of it is to give a tool for the cases when 
    47    a graph have to be used in a specific alteration. 
    48    If this alteration is obtained by a usual construction 
    49    like filtering the edge-set or considering a new orientation, then 
    50    an adaptor is worthwhile to use. 
    51    To come back to the reversed oriented graph, in this situation 
    52    \code template<typename Graph> class RevGraphAdaptor; \endcode 
    53    template class can be used. 
    54    The code looks as follows 
    55    \code
    56    ListGraph g;
    57    RevGraphAdaptor<ListGraph> rga(g);
    58    int result=algorithm(rga);
    59    \endcode
    60    After running the algorithm, the original graph \c g 
    61    is untouched. 
    62    This techniques gives rise to an elegant code, and 
    63    based on stable graph adaptors, complex algorithms can be 
    64    implemented easily. 
    65 
    66    In flow, circulation and bipartite matching problems, the residual 
    67    graph is of particular importance. Combining an adaptor implementing 
    68    this, shortest path algorithms and minimum mean cycle algorithms, 
    69    a range of weighted and cardinality optimization algorithms can be 
    70    obtained. 
    71    For other examples, 
    72    the interested user is referred to the detailed documentation of 
    73    particular adaptors. 
    74 
    75    The behavior of graph adaptors can be very different. Some of them keep 
    76    capabilities of the original graph while in other cases this would be 
    77    meaningless. This means that the concepts that they are models of depend 
    78    on the graph adaptor, and the wrapped graph(s). 
    79    If an edge of \c rga is deleted, this is carried out by 
    80    deleting the corresponding edge of \c g, thus the adaptor modifies the 
    81    original graph. 
    82    But for a residual 
    83    graph, this operation has no sense. 
    84    Let us stand one more example here to simplify your work. 
    85    RevGraphAdaptor has constructor 
    86    \code 
    87    RevGraphAdaptor(Graph& _g);
    88    \endcode
    89    This means that in a situation, 
    90    when a <tt> const ListGraph& </tt> reference to a graph is given, 
    91    then it have to be instantiated with <tt>Graph=const ListGraph</tt>.
    92    \code
    93    int algorithm1(const ListGraph& g) {
    94    RevGraphAdaptor<const ListGraph> rga(g);
    95    return algorithm2(rga);
    96    }
    97    \endcode
    98 */