Question

I have a program takes inputs from 2 files in c++. Then identify Is the second input file is topological sort? but somehow if i use the list.empty() in a while loop statement, it gives me a segmentation fault, but the for loop doesnt give me any error; however, the for loop only loop once, since i might need to go throught twice.

#include <list>
#include <iostream>
#include <vector>

using namespace std ;

list<unsigned> output;

list<unsigned> &
testSort ( istream & idata , istream & sdata )
{
 unsigned n,x1,x2;
 vector< list<unsigned> > successor(n);
 vector<unsigned> count(n,0);
 vector<bool> marks(n,false);

 idata >>n;

 for(int i=0;i<n;i++) {
idata>>x1>>x2;
count[x2]++;
successor[x1].push_back(x2);
if(idata.eof()) break;
 }

 for(int i=0;i<n;i++) {
 sdata>>x1;
  if(count[x1]==0) {
   marks[x1]=true;

   //for(int j=0;j<successor[x1].size();++j) {
    while(!successor[x1].empty()) { 
count[successor[x1].front()]--; 
successor[x1].pop_front();      
  }
 }
  else {
    for(int i=0;i<n;i++)
    {
     if(marks[i]==false)
     output.push_back(i);
    }
   break;
  }  

 }
  return output;
}

Answer 1

unsigned n,x1,x2;
vector< list<unsigned> > successor(n);

This is a clear error — you’re using n to specify the size of successor, but n hasn’t been initialized yet, so it contains garbage (and the same is true with count and marks since you’ve specified their size as n as well). In other words, at this point, we have no clue about the size of successor.

You have a couple of choices. You could move your idata >> n; before you define successor, count, and marks, or you could define them without a size, and then use resize to specify their size after you read n from idata.

I’ll leave off my usual rant about std::list beyond pointing out that I rarely find it an optimal choice.