TSP (libtsp) - MAfSubMat

MAfSubMat

Routine

float **MAfSubMat (const float *A[], int n, int m, int Nrow)

Purpose

Set up pointers to a submatrix of a matrix of float values

Description

This routine allocates an array of row pointers. These pointers access a submatrix of a previously allocated matrix. The returned value is a pointer to an array of pointers to the rows of the submatrix. The calling routine can treat the returned value as a two-dimensional matrix of float values. Consider the invocation

  float **A;
  float **sA;
  ...
  A = MAfAllocMat (Nrow, Ncol);
  sA = MAfSubMat ((const float **) A, n, m, Nrow-n);
      ...  now sA[0][0] == A[n][m]

Then sA[i][j] is a matrix of float values for 0 <= i < N-n and 0 <= j < Ncol-m. The array of pointers to the submatrix should be deallocated using the routine UTfree.

Parameters

<- float **MAfSubMat: Returned pointer to array of row pointers for the submatrix
-> const float *A[]: A is an array of pointers to the rows of an N by N matrix. Note that with ANSI C, if the actual parameter is not declared to have the const attribute, an explicit cast to (const float **) is required.
-> int n: Index for the first row of the submatrix
-> int m: Index for the first column of the submatrix
-> int Nrow: Number of rows in the submatrix

Author / revision