spcompack

spcompack — converts a compressed adjacency representation

Parameters

xadj : integer vector of length (n+1).
xlindx : integer vector of length n+1 (pointers).
lindx : integer vector
adjncy : integer vector

Description



Utility fonction spcompak is used to convert a compressed adjacency
representation into standard adjacency representation.
   
    

Examples



// A is the sparse matrix:
A=[1,0,0,0,0,0,0;
   0,1,0,0,0,0,0;
   0,0,1,0,0,0,0;
   0,0,1,1,0,0,0;
   0,0,1,1,1,0,0;
   0,0,1,1,0,1,0;
   0,0,1,1,0,1,1];
A=sparse(A);
//For this matrix, the standard adjacency representation is given by:
xadj=[1,2,3,8,12,13,15,16];
adjncy=[1, 2, 3,4,5,6,7, 4,5,6,7, 5, 6,7, 7];
//(see sp2adj).
// increments in vector xadj give the number of non zero entries in each column
// ie there is 2-1=1 entry in the column 1
//    there is 3-2=1 entry in the column 2
//    there are 8-3=5 entries in the column 3
//              12-8=4                      4
//etc
//The row index of these entries is given by the adjncy vector
// for instance, 
// adjncy (3:7)=adjncy(xadj(3):xadj(4)-1)=[3,4,5,6,7] 
// says that the 5=xadj(4)-xadj(3) entries in column 3 have row
// indices 3,4,5,6,7.
//In the compact representation, the repeated sequences in adjncy
//are eliminated.
//Here in adjncy the sequences 4,5,6,7  and 7 are eliminated.
//The standard structure (xadj,adjncy) takes the compressed form (lindx,xlindx)
lindx=[1, 2, 3,4,5,6,7, 5, 6,7];
xlindx=[1,2,3,8,9,11];
//(Columns 4 and 7 of A are eliminated).
//A can be reconstructed from (xadj,xlindx,lindx).
[xadj,adjncy,anz]= sp2adj(A);
adjncy -  spcompack(xadj,xlindx,lindx)
 
  

See also

sp2adj, adj2sp, spget