如何替换数组中的子字符串

Question

I have a program which replaces a parts of a char array with the contents of another, however it also replaces parts of the array that should not be replaced.

The code below is the code responsible for the replacing of characters in the array. One character at a time, however when only part of what it is supposed to replace is in an array like xy of xyz it should leave that alone and look for the next occurrence of xyz which it would then replace.

i =0;
while(i < 30){
    j=0;
    while(j < k){
        if(carray[i] == aarray[j])
            carray[i] = barray[j];
        j++;
    }
    i++;
}

My problem is, that if only a part of what the program is supposed to search and replace for is present it replaces it even though it is not the full word that was meant.

for example:

if I want to replace xyz with abc in the array of xyzdefxyzghixy then the arrays have the following content:

aarray[] = {xyz}
barray[] = {abc}
carray[] = {xyzdefxyzghixy}

I get abcdefabcghiab as an output and the last two characters ab should actually remain xy

ie

I expect the output to be abcdefabcghixy

what have I done wrong here.

NB: please Note that I wish to only use stdio.h

All help is greatly appreciated.

Thank you in advance.

Answer 1

You need to replace only when you find the full substring.

There are useful standard functions :

• strstr allows to find a substring in a string
• memcpy can be used to replace the old substring by the new one

A proposal :

#include <stdio.h>
#include <string.h>

int main(int argc, char ** argv)
{
  if (argc != 4)
    printf("usage : %s <str> <old> <new>\n", *argv);
  else {
    char * old = argv[2];
    char * new = argv[3];
    size_t len = strlen(old);

    if (len != strlen(new))
      fprintf(stderr, "'%s' and '%s' do not have the same length\n", old, new);
    else {
      char * str = argv[1];
      char * p = str;

      printf("'%s' -> ", str);

      /* the loop doing the replacements */
      while ((p = strstr(p, old)) != NULL) {
        memcpy(p, new, len);
        p += len;
      }

      printf("'%s'\n", str);
    }
  }

  return 0;
}

Compilation and execution :

pi@raspberrypi:/tmp $ gcc -pedantic -Wextra -Wall r.c
pi@raspberrypi:/tmp $ ./a.out
usage : ./a.out <str> <old> <new>
pi@raspberrypi:/tmp $ ./a.out xyzdefxyzghixy xyz abc
'xyzdefxyzghixy' -> 'abcdefabcghixy'

---

Edit

NB: please Note that I wish to only use stdio.h

Can be :

#include <stdio.h>

int main(int argc, char ** argv)
{
  if (argc != 4)
    printf("usage : %s <str> <old> <new>\n", *argv);
  else {
    char * old = argv[2];
    char * new = argv[3];
    char * str = argv[1];

    /* check old and new have the same length */
    char * pold, * pnew;

    for (pold = old, pnew = new; *pold && *pnew; ++pold, ++pnew)
      ;

    if (*pold || *pnew)
      fprintf(stderr, "'%s' and '%s' do not have the same length\n", old, new);
    else {
      printf("'%s' -> ", str);

      char * pstr = str;

      /* the loop doing the replacements */
      while (*pstr) {
        /* check if substring */
        char * pold = old;
        char * psubstr = pstr;

        while (*pold && (*pold == *psubstr)) {
          pold += 1;
          psubstr += 1;
        }

        if (*pold == 0) {
          /* substring, replacement */
          pnew = new;

          while (*pnew)
            *pstr++ = *pnew++;
        }
        else
          pstr += 1;
      }

      printf("'%s'\n", str);
    }
  }

  return 0;
}

As you can see this is a bad choice, it is very easy to understand what the previous version does, this is not the case with that version