// This software is Copyright (C) 2005 by Andrew Firth, University of Otago, 
// Dunedin, New Zealand. All rights reserved.

// Programming to return tracks to zero for long gaps.
// Assumes input file is in nucleotide order.  
// Column 1 is nucleotide ID.  Columns 2 and 3 are some statistic.
// At start and end of any long gaps, writes an extra line "nuc_ID 0 0" to 
//   output file.

#include <iostream>
#include <cmath>
#include <fstream>
#include <cstdlib>
#include <iomanip>
#include<cstring>
using namespace std;

int main(int argc, char* argv[])
{
  if (2 != argc) {
    cerr << "Aborting; use '" << argv[0] << " logfile'.\n";
    exit(EXIT_FAILURE);
  }

  int k, nlines, col1prev, col1, gapsize;
  double col2, col3;

  gapsize = 3; // zeros any gaps > gapsize nucleotides long 

  // Open log file.
  ifstream logfile(argv[1]); 
  if (!logfile) {
    cerr << "Aborting: can't find log file '" << argv[1] << "'.\n";
    exit(EXIT_FAILURE);
  }

  // Number of lines in log file
  nlines = -1;
  while (logfile.ignore(1000, '\n')) {
    ++nlines;
  }
  logfile.clear();
  logfile.seekg(0);

  // Work through logfile
  if (nlines > 0) {
    logfile >> col1; logfile >> col2; logfile >> col3;
    logfile.ignore(1000, '\n');
    cout << col1 << " " << col2 << " " << col3 << "\n";
    for (k = 1; k < nlines; ++k) {
      col1prev = col1;
      logfile >> col1; logfile >> col2; logfile >> col3;
      logfile.ignore(1000, '\n');
      if (col1 - col1prev > gapsize) {
	cout << col1prev << " " << 0 << " " << 0 << "\n";
	cout << col1     << " " << 0 << " " << 0 << "\n";
      }
      cout << col1 << " " << col2 << " " << col3 << "\n";
    }
  }

}