LibGraph

LibGraph is Graph library for C++. It can be used to solve various graph problems.

Usage

• Clone the repository

  git clone http://github.com/Hrily/LibGraph

• Include header in your code

  #include "graph.h"

• And use...

Features

• Graph
  • BFS
  • DFS
  • Cycle Detection
  • Single Source Shortest Path
  • All Pair Shortest Path
  • Minimum Spanning Tree
• Tree
  • Topological sort

Graph

Initialization

Graph can be initialized as follows

Graph<int, int> G;

where first int is type of vertices and second int is type of edge weight

Adding edges

Edges are added as follows

G.addEdge(1, 2, 3);

where edge 1->2 has weight 3

BFS

BFS of graph generates a mapping of vertices to their levels

map<int, int> bfsLevel = G.bfs(1);

where 1 is source of bfs tree

DFS

DFS of graph generates vector of vertices which are in sequence of their dfs visit

vector<int> dfsVisitSequence = G.dfs(1);

Again 1 here is source

Cycle Detection

Cycle can be detected as follows

if(G.hasCycle())
  cout<<"Graph has cycle\n";
else
  cout<<"Graph does not have any cycle\n";

Single Source Shortest Path

Shortest paths from a source generates mapping of vertices to shortest distance

map<int, int> shortestPaths = G.getShortestPathsFrom(1, false);

where first parameter is source and second parameter is whether graph has negative edges, false by default.

All Pair Shortest Path

All Pair Shortest Path generates two dimensional mapping of pair of vertices to their shortest paths

map<int, map<int, int> > allPairShortestPath = G.getAllPairShortestPath();

Minimum Spanning Tree

Minimum Spanning Tree returns Object of type Tree, which is defined in tree.h

Tree<int, int> mst = G.minimumSpanningTree(1);

where 1 is the source

Tree

Tree is a class derived from Graph. It has same functionalities as Graph, but can't have any cycles

Usage

Tree is defined in tree.h. Use it as follows

#include "tree.h"

NOTE: Graph is already included in tree. So you need not to include graph when you are including tree.

Initialization

Tree is initialized as follows

Tree<int, int> T;

Same rules apply here, first int is vertex type, while second is edge weight type

Adding Edges

Adding edges to Tree is same as Graph, but adding edge here can generate exception (of type Exception) if adding the edge to tree makes it invalid

try{
  T.addEdge(1, 2, 3);
}catch(Exception e){
  cout<<"Exception: "<<e.getMessage()<<endl;
}

Topological Sort

Topological sort of the Tree vector of the vertices listed topologically

vector<int> tpSort = T.topologicalSort();

Contributors

• Dhaval Khandla
• Hrishikesh Hiraskar
• Manish Kumar