using boost tokenizer to parse CSV files into Eigen Matrices

test_csv_parse.cpp

#include <iostream>
#include <iomanip>
#include <fstream>
#include "mkl.h"
#include "math.h"
#include <vector>
#include <cmath>
#include <string>
#include <cstdlib>

#include <ctype.h>
#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>
#include <getopt.h>

#define EIGEN_USE_MKL_BLAS
#define EIGEN_USE_MKL_LAPACKE
#define NDEBUG

#ifndef PI
#define PI 3.141592653589793
#endif


#include <Eigen/Dense>

#include <boost/tokenizer.hpp>


using namespace boost;
using namespace Eigen;
using namespace std;

// THIS IS A TYPEDEF FOR A ROWMAJOR MATRIX
typedef Eigen::Matrix<double,Eigen::Dynamic, Eigen::Dynamic, Eigen::RowMajor> MatrixRMXd;



////////////////////////////////////////
// PARSE THE OPTIONS
void parseTheOptions(int argc, char *argv[], 
                     string& fname) {
  cout << "This is the argument parsing function" << endl;

    // process the command line arguments
  while(1)
  {
      int c;
      int digit_optind = 0;
      int this_option_optind = optind ? optind : 1;
      int option_index = 0;
      static struct option long_options[] = {
        {"input"    , 1, 0, 'u' },
        {0,0,0,0}
      };

      c = getopt_long_only(argc, argv, "abc:d:012",
                           long_options, &option_index);

      if (c == -1)
        break;

      switch(c)
      {
        case 'u':
          fname = optarg;
          break;
      }

  } // end while 
}  // end opt parse


////////////////////////////////////////
// double round(double number)
// {
//     return number < 0.0 ? ceil(number - 0.5) : floor(number + 0.5);
// }

////////////////////////////////////////
// parse csv file into a Matrix of Integers
// assumes no header
// (yes this is ugly)
// NOTE: assumes that the "right" number of columns is the number
//       of columns in the first row
MatrixXi parseCSVfile_int(string infilename) {

    ifstream in(infilename.c_str());
    if (!in.is_open()) return MatrixXi(1,1);

    typedef tokenizer< escaped_list_separator<char> > Tokenizer;

    vector< string > vec;
    string line;
    vector< vector< string > > matrows;


    while (getline(in,line))
    {
        Tokenizer tok(line);
        vec.assign(tok.begin(),tok.end());

	// // Print each row
        // copy(vec.begin(), vec.end(),
        //      ostream_iterator<string>(cout, "|"));
        // cout << "\n----------------------" << endl;
	
	matrows.push_back(vec);
    }
    in.close();

    // FIGURE OUT HOW MANY OF THE ROWS HAVE THE RIGHT NUMBER
    // OF COLUMNS
    int Nrows = matrows.size();
    int Ncols = matrows[0].size();
    int Ngoodrows = 0;
    for(int i = 0; i < Nrows; i++) {
      if(matrows[i].size() == Ncols) {
	Ngoodrows++;
      }
    }


    // TRANSFORM THE VECTOR OF ROWS INTO AN EIGEN INTEGER MATRIX
    MatrixXi xmat = MatrixXi(Ngoodrows, Ncols);
    cout << "INPUT MATRIX: " << Nrows << "x" << Ncols << endl;
    int rc = 0;
    for(int i = 0; i < Nrows; i++) {
      int rowsize = matrows[i].size();
      if(rowsize != Ncols) {
	cout << "Row " << i << " has bad column count" << endl;
	continue;
      } 
      for(int j = 0; j < Ncols; j++) {
	xmat(rc,j) = int(round(strtod(matrows[i][j].c_str(), NULL)));
      }
      rc++;
    }


    return(xmat);
}



int main(int argc, char **argv) 
{

    string fname;

    parseTheOptions(argc, argv, fname);

    
    MatrixXi parsed = parseCSVfile_int(fname);
    cout << "PARSED: " << endl;
    cout << parsed << endl << endl;


}