deba@2034: /* -*- C++ -*-
deba@2034:  *
deba@2034:  * This file is a part of LEMON, a generic C++ optimization library
deba@2034:  *
deba@2034:  * Copyright (C) 2003-2006
deba@2034:  * Egervary Jeno Kombinatorikus Optimalizalasi Kutatocsoport
deba@2034:  * (Egervary Research Group on Combinatorial Optimization, EGRES).
deba@2034:  *
deba@2034:  * Permission to use, modify and distribute this software is granted
deba@2034:  * provided that this copyright notice appears in all copies. For
deba@2034:  * precise terms see the accompanying LICENSE file.
deba@2034:  *
deba@2034:  * This software is provided "AS IS" with no warranty of any kind,
deba@2034:  * express or implied, and with no claim as to its suitability for any
deba@2034:  * purpose.
deba@2034:  *
deba@2034:  */
deba@2034: 
deba@2034: #ifndef LEMON_EDMONDS_KARP_H
deba@2034: #define LEMON_EDMONDS_KARP_H
deba@2034: 
deba@2034: /// \file
deba@2034: /// \ingroup flowalgs
deba@2034: /// \brief Implementation of the Edmonds-Karp algorithm.
deba@2034: 
deba@2034: #include <lemon/graph_adaptor.h>
deba@2034: #include <lemon/tolerance.h>
deba@2034: #include <lemon/bfs.h>
deba@2034: 
deba@2034: namespace lemon {
deba@2034: 
deba@2034:   /// \ingroup flowalgs
deba@2034:   /// \brief Edmonds-Karp algorithms class.
deba@2034:   ///
deba@2034:   /// This class provides an implementation of the \e Edmonds-Karp \e
deba@2034:   /// algorithm producing a flow of maximum value in a directed
deba@2034:   /// graph. The Edmonds-Karp algorithm is slower than the Preflow algorithm
deba@2034:   /// but it has an advantage of the step-by-step execution control with
deba@2034:   /// feasible flow solutions. The \e source node, the \e target node, the \e
deba@2034:   /// capacity of the edges and the \e starting \e flow value of the
deba@2034:   /// edges should be passed to the algorithm through the
deba@2036:   /// constructor.
deba@2034:   ///
deba@2059:   /// The time complexity of the algorithm is \f$ O(n * e^2) \f$ in
deba@2059:   /// worst case.  Always try the preflow algorithm instead of this if
deba@2113:   /// you just want to compute the optimal flow.
deba@2034:   ///
deba@2034:   /// \param _Graph The directed graph type the algorithm runs on.
deba@2034:   /// \param _Number The number type of the capacities and the flow values.
deba@2034:   /// \param _CapacityMap The capacity map type.
deba@2034:   /// \param _FlowMap The flow map type.
deba@2034:   /// \param _Tolerance The tolerance class to handle computation problems.
deba@2034:   ///
deba@2034:   /// \author Balazs Dezso 
deba@2059: #ifdef DOXYGEN
deba@2059:   template <typename _Graph, typename _Number,
deba@2059: 	    typename _CapacityMap, typename _FlowMap, typename _Tolerance>
deba@2059: #else
deba@2034:   template <typename _Graph, typename _Number,
deba@2034: 	    typename _CapacityMap = typename _Graph::template EdgeMap<_Number>,
deba@2034:             typename _FlowMap = typename _Graph::template EdgeMap<_Number>,
deba@2034: 	    typename _Tolerance = Tolerance<_Number> >
deba@2059: #endif
deba@2034:   class EdmondsKarp {
deba@2034:   public:
deba@2034: 
deba@2034:     /// \brief \ref Exception for the case when the source equals the target.
deba@2034:     ///
deba@2034:     /// \ref Exception for the case when the source equals the target.
deba@2034:     ///
deba@2034:     class InvalidArgument : public lemon::LogicError {
deba@2034:     public:
alpar@2151:       virtual const char* what() const throw() {
deba@2034: 	return "lemon::EdmondsKarp::InvalidArgument";
deba@2034:       }
deba@2034:     };
deba@2034: 
deba@2034: 
deba@2034:     /// \brief The graph type the algorithm runs on. 
deba@2034:     typedef _Graph Graph;
deba@2034:     /// \brief The value type of the algorithms.
deba@2034:     typedef _Number Number;
deba@2034:     /// \brief The capacity map on the edges.
deba@2034:     typedef _CapacityMap CapacityMap;
deba@2034:     /// \brief The flow map on the edges.
deba@2034:     typedef _FlowMap FlowMap;
deba@2034:     /// \brief The tolerance used by the algorithm.
deba@2034:     typedef _Tolerance Tolerance;
deba@2034: 
deba@2034:     typedef ResGraphAdaptor<Graph, Number, CapacityMap, 
deba@2034:                             FlowMap, Tolerance> ResGraph;
deba@2034: 
deba@2034:   private:
deba@2034: 
deba@2034:     typedef typename Graph::Node Node;
deba@2034:     typedef typename Graph::Edge Edge;
deba@2034:     
deba@2034:     typedef typename Graph::NodeIt NodeIt;
deba@2034:     typedef typename Graph::EdgeIt EdgeIt;
deba@2034:     typedef typename Graph::InEdgeIt InEdgeIt;
deba@2034:     typedef typename Graph::OutEdgeIt OutEdgeIt;
deba@2034:     
deba@2034:   public:
deba@2034: 
deba@2034:     /// \brief The constructor of the class.
deba@2034:     ///
deba@2034:     /// The constructor of the class. 
deba@2037:     /// \param graph The directed graph the algorithm runs on. 
deba@2037:     /// \param source The source node.
deba@2037:     /// \param target The target node.
deba@2037:     /// \param capacity The capacity of the edges. 
deba@2037:     /// \param flow The flow of the edges. 
deba@2037:     /// \param tolerance Tolerance class.
deba@2034:     EdmondsKarp(const Graph& graph, Node source, Node target,
deba@2034:                 const CapacityMap& capacity, FlowMap& flow, 
deba@2034:                 const Tolerance& tolerance = Tolerance())
deba@2034:       : _graph(graph), _capacity(capacity), _flow(flow), 
deba@2034:         _tolerance(tolerance), _resgraph(graph, capacity, flow, tolerance), 
deba@2034:         _source(source), _target(target) 
deba@2034:     {
deba@2034:       if (_source == _target) {
deba@2034:         throw InvalidArgument();
deba@2034:       }
deba@2034:     }
deba@2034: 
deba@2034:     /// \brief Initializes the algorithm
deba@2034:     /// 
deba@2034:     /// It sets the flow to empty flow.
deba@2034:     void init() {
deba@2034:       for (EdgeIt it(_graph); it != INVALID; ++it) {
deba@2034:         _flow.set(it, 0);
deba@2034:       }
deba@2034:       _value = 0;
deba@2034:     }
deba@2034:     
deba@2034:     /// \brief Initializes the algorithm
deba@2034:     /// 
deba@2034:     /// If the flow map initially flow this let the flow map
deba@2034:     /// unchanged but the flow value will be set by the flow
deba@2034:     /// on the outedges from the source.
deba@2034:     void flowInit() {
deba@2034:       _value = 0;
deba@2034:       for (OutEdgeIt jt(_graph, _source); jt != INVALID; ++jt) {
deba@2034:         _value += _flow[jt];
deba@2034:       }
deba@2034:       for (InEdgeIt jt(_graph, _source); jt != INVALID; ++jt) {
deba@2034:         _value -= _flow[jt];
deba@2034:       }
deba@2034:     }
deba@2034: 
deba@2034:     /// \brief Initializes the algorithm
deba@2034:     /// 
deba@2034:     /// If the flow map initially flow this let the flow map
deba@2034:     /// unchanged but the flow value will be set by the flow
deba@2034:     /// on the outedges from the source. It also checks that
deba@2034:     /// the flow map really contains a flow.
deba@2034:     /// \return %True when the flow map really a flow.
deba@2034:     bool checkedFlowInit() {
deba@2034:       _value = 0;
deba@2034:       for (OutEdgeIt jt(_graph, _source); jt != INVALID; ++jt) {
deba@2034:         _value += _flow[jt];
deba@2034:       }
deba@2034:       for (InEdgeIt jt(_graph, _source); jt != INVALID; ++jt) {
deba@2034:         _value -= _flow[jt];
deba@2034:       }
deba@2034:       for (NodeIt it(_graph); it != INVALID; ++it) {
deba@2034:         if (it == _source || it == _target) continue;
deba@2034:         Number outFlow = 0;
deba@2034:         for (OutEdgeIt jt(_graph, it); jt != INVALID; ++jt) {
deba@2034:           outFlow += _flow[jt];
deba@2034:         }
deba@2034:         Number inFlow = 0;
deba@2034:         for (InEdgeIt jt(_graph, it); jt != INVALID; ++jt) {
deba@2034:           inFlow += _flow[jt];
deba@2034:         }
deba@2034:         if (_tolerance.different(outFlow, inFlow)) {
deba@2034:           return false;
deba@2034:         }
deba@2034:       }
deba@2034:       for (EdgeIt it(_graph); it != INVALID; ++it) {
deba@2034:         if (_tolerance.less(_flow[it], 0)) return false;
deba@2034:         if (_tolerance.less(_capacity[it], _flow[it])) return false;
deba@2034:       }
deba@2034:       return true;
deba@2034:     }
deba@2034: 
deba@2034:     /// \brief Augment the solution on an edge shortest path.
deba@2034:     /// 
deba@2034:     /// Augment the solution on an edge shortest path. It search an
deba@2034:     /// edge shortest path between the source and the target
deba@2034:     /// in the residual graph with the bfs algoritm.
deba@2034:     /// Then it increase the flow on this path with the minimal residual
deba@2034:     /// capacity on the path. If there is not such path it gives back
deba@2034:     /// false.
deba@2034:     /// \return %False when the augmenting is not success so the
deba@2034:     /// current flow is a feasible and optimal solution.
deba@2034:     bool augment() {
deba@2034:       typename Bfs<ResGraph>
deba@2034:       ::template DefDistMap<NullMap<Node, int> >
deba@2034:       ::Create bfs(_resgraph);
deba@2034: 
deba@2034:       NullMap<Node, int> distMap;
deba@2034:       bfs.distMap(distMap);
deba@2034:       
deba@2034:       bfs.init();
deba@2034:       bfs.addSource(_source);
deba@2034:       bfs.start(_target);
deba@2034: 
deba@2034:       if (!bfs.reached(_target)) {
deba@2034:         return false;
deba@2034:       }
deba@2034:       Number min = _resgraph.rescap(bfs.predEdge(_target));
deba@2034:       for (Node it = bfs.predNode(_target); it != _source; 
deba@2034:            it = bfs.predNode(it)) {
deba@2034:         if (min > _resgraph.rescap(bfs.predEdge(it))) {
deba@2034:           min = _resgraph.rescap(bfs.predEdge(it));
deba@2034:         }
deba@2034:       } 
deba@2034:       for (Node it = _target; it != _source; it = bfs.predNode(it)) {
deba@2034:         _resgraph.augment(bfs.predEdge(it), min);
deba@2034:       }
deba@2034:       _value += min;
deba@2034:       return true;
deba@2034:     }
deba@2034: 
deba@2034:     /// \brief Executes the algorithm
deba@2034:     ///
deba@2034:     /// It runs augmenting phases until the optimal solution is reached. 
deba@2034:     void start() {
deba@2034:       while (augment()) {}
deba@2034:     }
deba@2034: 
deba@2034:     /// \brief Gives back the current flow value.
deba@2034:     ///
deba@2034:     /// Gives back the current flow _value.
deba@2034:     Number flowValue() const {
deba@2034:       return _value;
deba@2034:     }
deba@2034: 
deba@2034:     /// \brief runs the algorithm.
deba@2034:     /// 
deba@2034:     /// It is just a shorthand for:
deba@2059:     ///
deba@2059:     ///\code 
deba@2034:     /// ek.init();
deba@2034:     /// ek.start();
deba@2059:     ///\endcode
deba@2034:     void run() {
deba@2034:       init();
deba@2034:       start();
deba@2034:     }
deba@2034: 
deba@2034:     /// \brief Returns a minimum value cut.
deba@2034:     ///
deba@2034:     /// Sets \c cut to the characteristic vector of a minimum value cut
deba@2034:     /// It simply calls the minMinCut member.
deba@2037:     /// \retval cut Write node bool map. 
deba@2034:     template <typename CutMap>
deba@2034:     void minCut(CutMap& cut) const {
deba@2034:       minMinCut(cut);
deba@2034:     }
deba@2034: 
deba@2034:     /// \brief Returns the inclusionwise minimum of the minimum value cuts.
deba@2034:     ///
deba@2034:     /// Sets \c cut to the characteristic vector of the minimum value cut
deba@2034:     /// which is inclusionwise minimum. It is computed by processing a
deba@2034:     /// bfs from the source node \c source in the residual graph.  
deba@2037:     /// \retval cut Write node bool map. 
deba@2034:     template <typename CutMap>
deba@2034:     void minMinCut(CutMap& cut) const {
deba@2034: 
deba@2034:       typename Bfs<ResGraph>
deba@2034:       ::template DefDistMap<NullMap<Node, int> >
deba@2034:       ::template DefProcessedMap<CutMap>
deba@2034:       ::Create bfs(_resgraph);
deba@2034: 
deba@2034:       NullMap<Node, int> distMap;
deba@2034:       bfs.distMap(distMap);
deba@2034: 
deba@2034:       bfs.processedMap(cut);
deba@2034:      
deba@2034:       bfs.run(_source);
deba@2034:     }
deba@2034: 
deba@2034:     /// \brief Returns the inclusionwise minimum of the minimum value cuts.
deba@2034:     ///
deba@2034:     /// Sets \c cut to the characteristic vector of the minimum value cut
deba@2034:     /// which is inclusionwise minimum. It is computed by processing a
deba@2034:     /// bfs from the source node \c source in the residual graph.  
deba@2037:     /// \retval cut Write node bool map. 
deba@2034:     template <typename CutMap>
deba@2034:     void maxMinCut(CutMap& cut) const {
deba@2034: 
deba@2034:       typedef RevGraphAdaptor<const ResGraph> RevGraph;
deba@2034: 
deba@2034:       RevGraph revgraph(_resgraph);
deba@2034: 
deba@2034:       typename Bfs<RevGraph>
deba@2034:       ::template DefDistMap<NullMap<Node, int> >
deba@2034:       ::template DefPredMap<NullMap<Node, Edge> >
deba@2034:       ::template DefProcessedMap<NotWriteMap<CutMap> >
deba@2034:       ::Create bfs(revgraph);
deba@2034: 
deba@2034:       NullMap<Node, int> distMap;
deba@2034:       bfs.distMap(distMap);
deba@2034: 
deba@2034:       NullMap<Node, Edge> predMap;
deba@2034:       bfs.predMap(predMap);
deba@2034: 
deba@2034:       NotWriteMap<CutMap> notcut(cut);
deba@2034:       bfs.processedMap(notcut);
deba@2034:      
deba@2034:       bfs.run(_target);
deba@2034:     }
deba@2034: 
deba@2034:     /// \brief Returns the source node.
deba@2034:     ///
deba@2034:     /// Returns the source node.
deba@2034:     /// 
deba@2034:     Node source() const { 
deba@2034:       return _source;
deba@2034:     }
deba@2034: 
deba@2034:     /// \brief Returns the target node.
deba@2034:     ///
deba@2034:     /// Returns the target node.
deba@2034:     /// 
deba@2034:     Node target() const { 
deba@2034:       return _target;
deba@2034:     }
deba@2034: 
deba@2034:     /// \brief Returns a reference to capacity map.
deba@2034:     ///
deba@2034:     /// Returns a reference to capacity map.
deba@2034:     /// 
deba@2034:     const CapacityMap &capacityMap() const { 
deba@2034:       return *_capacity;
deba@2034:     }
deba@2034:      
deba@2034:     /// \brief Returns a reference to flow map.
deba@2034:     ///
deba@2034:     /// Returns a reference to flow map.
deba@2034:     /// 
deba@2034:     const FlowMap &flowMap() const { 
deba@2034:       return *_flow;
deba@2034:     }
deba@2034: 
deba@2036:     /// \brief Returns the tolerance used by algorithm.
deba@2036:     ///
deba@2036:     /// Returns the tolerance used by algorithm.
deba@2036:     const Tolerance& tolerance() const {
deba@2036:       return tolerance;
deba@2036:     } 
deba@2036:     
deba@2034:   private:
deba@2034:     
deba@2034:     const Graph& _graph;
deba@2034:     const CapacityMap& _capacity;
deba@2034:     FlowMap& _flow;
deba@2034:     Tolerance _tolerance;
deba@2034:     
deba@2034:     ResGraph _resgraph;
deba@2034:     Node _source, _target;
deba@2034:     Number _value;
deba@2034:     
deba@2034:   };
deba@2034: 
deba@2034: }
deba@2034: 
deba@2034: #endif