Logo Questions Linux Laravel Mysql Ubuntu Git Menu
 

Parsing a comma-delimited std::string [duplicate]

Input one number at a time, and check whether the following character is ,. If so, discard it.

#include <vector>
#include <string>
#include <sstream>
#include <iostream>

int main()
{
    std::string str = "1,2,3,4,5,6";
    std::vector<int> vect;

    std::stringstream ss(str);

    for (int i; ss >> i;) {
        vect.push_back(i);    
        if (ss.peek() == ',')
            ss.ignore();
    }

    for (std::size_t i = 0; i < vect.size(); i++)
        std::cout << vect[i] << std::endl;
}

Something less verbose, std and takes anything separated by a comma.

stringstream ss( "1,1,1,1, or something else ,1,1,1,0" );
vector<string> result;

while( ss.good() )
{
    string substr;
    getline( ss, substr, ',' );
    result.push_back( substr );
}

Yet another, rather different, approach: use a special locale that treats commas as white space:

#include <locale>
#include <vector>

struct csv_reader: std::ctype<char> {
    csv_reader(): std::ctype<char>(get_table()) {}
    static std::ctype_base::mask const* get_table() {
        static std::vector<std::ctype_base::mask> rc(table_size, std::ctype_base::mask());

        rc[','] = std::ctype_base::space;
        rc['\n'] = std::ctype_base::space;
        rc[' '] = std::ctype_base::space;
        return &rc[0];
    }
}; 

To use this, you imbue() a stream with a locale that includes this facet. Once you've done that, you can read numbers as if the commas weren't there at all. Just for example, we'll read comma-delimited numbers from input, and write then out one-per line on standard output:

#include <algorithm>
#include <iterator>
#include <iostream>

int main() {
    std::cin.imbue(std::locale(std::locale(), new csv_reader()));
    std::copy(std::istream_iterator<int>(std::cin), 
              std::istream_iterator<int>(),
              std::ostream_iterator<int>(std::cout, "\n"));
    return 0;
}

The C++ String Toolkit Library (Strtk) has the following solution to your problem:

#include <string>
#include <deque>
#include <vector>
#include "strtk.hpp"
int main()
{ 
   std::string int_string = "1,2,3,4,5,6,7,8,9,10,11,12,13,14,15";
   std::vector<int> int_list;
   strtk::parse(int_string,",",int_list);

   std::string double_string = "123.456|789.012|345.678|901.234|567.890";
   std::deque<double> double_list;
   strtk::parse(double_string,"|",double_list);

   return 0;
}

More examples can be found Here


Alternative solution using generic algorithms and Boost.Tokenizer:

struct ToInt
{
    int operator()(string const &str) { return atoi(str.c_str()); }
};

string values = "1,2,3,4,5,9,8,7,6";

vector<int> ints;
tokenizer<> tok(values);

transform(tok.begin(), tok.end(), back_inserter(ints), ToInt());

Lots of pretty terrible answers here so I'll add mine (including test program):

#include <string>
#include <iostream>
#include <cstddef>

template<typename StringFunction>
void splitString(const std::string &str, char delimiter, StringFunction f) {
  std::size_t from = 0;
  for (std::size_t i = 0; i < str.size(); ++i) {
    if (str[i] == delimiter) {
      f(str, from, i);
      from = i + 1;
    }
  }
  if (from <= str.size())
    f(str, from, str.size());
}


int main(int argc, char* argv[]) {
    if (argc != 2)
        return 1;

    splitString(argv[1], ',', [](const std::string &s, std::size_t from, std::size_t to) {
        std::cout << "`" << s.substr(from, to - from) << "`\n";
    });

    return 0;
}

Nice properties:

  • No dependencies (e.g. boost)
  • Not an insane one-liner
  • Easy to understand (I hope)
  • Handles spaces perfectly fine
  • Doesn't allocate splits if you don't want to, e.g. you can process them with a lambda as shown.
  • Doesn't add characters one at a time - should be fast.
  • If using C++17 you could change it to use a std::stringview and then it won't do any allocations and should be extremely fast.

Some design choices you may wish to change:

  • Empty entries are not ignored.
  • An empty string will call f() once.

Example inputs and outputs:

""      ->   {""}
","     ->   {"", ""}
"1,"    ->   {"1", ""}
"1"     ->   {"1"}
" "     ->   {" "}
"1, 2," ->   {"1", " 2", ""}
" ,, "  ->   {" ", "", " "}

You could also use the following function.

void tokenize(const string& str, vector<string>& tokens, const string& delimiters = ",")
{
  // Skip delimiters at beginning.
  string::size_type lastPos = str.find_first_not_of(delimiters, 0);

  // Find first non-delimiter.
  string::size_type pos = str.find_first_of(delimiters, lastPos);

  while (string::npos != pos || string::npos != lastPos) {
    // Found a token, add it to the vector.
    tokens.push_back(str.substr(lastPos, pos - lastPos));

    // Skip delimiters.
    lastPos = str.find_first_not_of(delimiters, pos);

    // Find next non-delimiter.
    pos = str.find_first_of(delimiters, lastPos);
  }
}