Trying to code a C-style function as a C++ one

Good morning. I'm trying to convert this into a C++ style program:

Code:

---


#include <iostream>
#include <string>

const int MAXSIZE = 15;

using namespace std;

unsigned int c_in_str(const char *str, char ch);

int main(void)
{
  char mmm[MAXSIZE] = "minimum";
  char *wail = "ululate";
 
  unsigned int ms = c_in_str(mmm, 'm');
  unsigned int us = c_in_str(wail, 'u');
 
  cout << ms << " m characters in " << mmm << endl;
  cout << us << " u characters in " << wail << endl;     
   
  return 0;   
}   

unsigned int c_in_str(const char *str, char ch)
{
  int count = 0;
 
  while(*str)
  {
    if(*str == ch)
      count++;
    str++;
  }
  return count;                 
}

---

g++ complains I'm using char * in this case

Code:

---


#include <iostream>
#include <string>

using namespace std;

unsigned int c_in_str(const string str, char ch);

int main(void)
{
  string mmm  = "minimum";
  string wail = "ululate";
 
  unsigned int ms = c_in_str(mmm, 'm');
  unsigned int us = c_in_str(wail, 'u');
 
  cout << ms << " m characters in " << mmm << endl;
  cout << us << " u characters in " << wail << endl;     
   
  return 0;   
}   

unsigned int c_in_str(const string str, char ch)
{
  int count = 0;
  int i;
  while(str[i] != '\0')
  {
    if(str[i] == ch)
      count++;
    str[i++];
  }
  return count;                 
}

---

in this case, valgrind complains I'm not setting a size to mmm and wail.

Code:

---


thames@semaht ~/C++/Strings $ valgrind ./strgfun
==3259== Memcheck, a memory error detector
==3259== Copyright (C) 2002-2011, and GNU GPL'd, by Julian Seward et al.
==3259== Using Valgrind-3.7.0 and LibVEX; rerun with -h for copyright info
==3259== Command: ./strgfun
==3259==
==3259== Use of uninitialised value of size 8
==3259==    at 0x400E58: c_in_str(std::string, char) (strgfun.cpp:26)
==3259==    by 0x400CA2: main (strgfun.cpp:13)
==3259==
==3259== Use of uninitialised value of size 8
==3259==    at 0x400E19: c_in_str(std::string, char) (strgfun.cpp:28)
==3259==    by 0x400CA2: main (strgfun.cpp:13)
==3259==
==3259== Use of uninitialised value of size 8
==3259==    at 0x400E58: c_in_str(std::string, char) (strgfun.cpp:26)
==3259==    by 0x400CD5: main (strgfun.cpp:14)
==3259==
==3259== Use of uninitialised value of size 8
==3259==    at 0x400E19: c_in_str(std::string, char) (strgfun.cpp:28)
==3259==    by 0x400CD5: main (strgfun.cpp:14)
==3259==
3 m characters in minimum
2 u characters in ululate
==3259==
==3259== HEAP SUMMARY:
==3259==    in use at exit: 0 bytes in 0 blocks
==3259==  total heap usage: 2 allocs, 2 frees, 64 bytes allocated
==3259==
==3259== All heap blocks were freed -- no leaks are possible
==3259==
==3259== For counts of detected and suppressed errors, rerun with: -v
==3259== Use --track-origins=yes to see where uninitialised values come from
==3259== ERROR SUMMARY: 30 errors from 4 contexts (suppressed: 2 from 2)

---

In the first code the following needs to be changed:

Code:

---

  char *wail = "ululate";
  // To:
  const char *wail = "ululate";

---

You need to match the const of the char pointer to the const of the string.

In the second code the following line:

Code:

---

unsigned int c_in_str(const string str, char ch)

---

The const on the string parameter is actually not required since you are sending a copy of the string to the function. However I would change the function to send a const string reference for the string instead of the copy.

Code:

---

unsigned int c_in_str(const string& str, char ch)
{
  int count = 0;
  int i;
  while(str[i] != '\0')

---

Now in the above code you never initialize i so when you use it in the while loop you get undefined behavior. This what valgrind is complaining about.

Also since this is a std::string, I'm not sure the use of the end of string character is guaranteed to work. I suggest changing to a for loop for this and use the size of the string.

Code:

---

  for(int i = 0; i < str.length(); ++i)
  {
      if(str[i] == ch)
        count++;
    }

---

Jim

Many thanks Jim. I fixed the code.

Code:

---

#include <iostream>
#include <string>

using namespace std;

unsigned int c_in_str(const string& str, char ch);

int main(void)
{
  const string mmm  = "minimum";
  const string wail = "ululate";
 
  unsigned int ms = c_in_str(mmm, 'm');
  unsigned int us = c_in_str(wail, 'u');
 
  cout << ms << " m characters in " << mmm << endl;
  cout << us << " u characters in " << wail << endl;     
   
  return 0;   
}   

unsigned int c_in_str(const string& str, char ch)
{
  int count = 0;
  unsigned int i;
  for(i = 0; i < str.length(); i++)
    if(str[i] == ch)
    ++count;
  return count;                 
}

---

You could do without that c_in_str function by using std::count from <algorithm>, e.g.,

Code:

---

#include <iostream>
#include <string>
#include <algorithm>

using namespace std;

int main(void)
{
  const string mmm  = "minimum";
  const string wail = "ululate";

  unsigned int ms = count(mmm.begin(), mmm.end(), 'm');
  unsigned int us = count(wail.begin(), wail.end(), 'u');

  cout << ms << " m characters in " << mmm << endl;
  cout << us << " u characters in " << wail << endl;

  return 0;
}

---

Quote:

---

You could do without that c_in_str function by using std::count from <algorithm>, e.g.

---

cool! I didn't know that. Many thanks laserlight.