%%%%%%%%%%%%%%%%%%%%%%%%
	% Breadth-First Search %
	%%%%%%%%%%%%%%%%%%%%%%%%
   
   %%   'dist' gives shortest distance in terms of # arcs.
   %%   dist is initialized to infinity for all nodes. 
   %%   Then dist(s) is set to zero. Then, as we increase 
   %%   breadth of the search by one level, the distance
   %%   of the nodes marked in the current level is set to 
   %%   dist(i) + 1, where i is the node from which the new
   %%   nodes in the current level are accessed from.
   
   %%   Rather than adding nodes to and deleting nodes from LIST,
   %%   we use two variables first and last to keep track ot it.

	% s---starting node for BFS Search
	s = input('give the starting node for the BFS search   ');
	
	LIST=[s];
	first=1;
	last=length(LIST);
    
	mark=zeros(1,N);
	mark(s)=1;
        pred=zeros(1,N);
    
	bigM = 10000; %% acts like infinity
        dist=bigM*ones(1,N); 
	dist(s)=0;

	while first <= last
   	   i=LIST(first);
	   if ~isempty(A{i})
	      jj=H(A{i});          %A{i} stores arc_names
	      admit=jj( find( mark(jj)==0 ) );
	      mark(admit)=1;
	      pred(admit)=i;	
	      dist(admit) = dist(i) + 1;
	      LIST=[LIST admit'];
	      last=last+length(admit);
	   end%if
	   first=first+1;
	end%while
	
	dist
   	BFS=LIST;
	fprintf(1,'\nBFS visits nodes 1 to %d in the sequence ',N);
	fprintf(1, '  %d  ',BFS);
	fprintf(1,'\n\n');
        order=zeros(1,N);
	order(BFS)=(1:length(BFS))
        pred