Thread: Easy way to reverse a string

As I showed in a recent post, cute XOR tricks to swap a pair of values are likely to make things worse rather than better.

> s = reverse_str("mango");
> output > segmentation fault
> why ??
"strings" can be placed in read-only memory (you can't change read-only memory)
As soon as you try and swap a pair of characters, its game over for you

char a[] = "mango";
s = reverse( a );
should work, if reverse() does as advertised.

"strings" can be placed in read-only memory (you can't change read-only memory)

hmmm......i did a small experiment.....

code-1
---------

Code:

int main()
{
char *s ="mango";
s[3] ='p';  // read only..so no change is allowed.
cout<<s<<endl;

}

output >segmentation fault.




code-2
----------

Code:

char *s =new char[6]; // giving some memory
s="mango";
s[3] ='p';
cout<<s<<endl;

code-2 is also giving segmentaion fault !!
i have given memory to the code -2 by new operator to play around. how it still remains read only . why the latter case not read-write both !!.....does it mean pointers are always read only ( using new also !!) ?

thanks

This is the C forum and not the C++ forum, soooo what you have is not valid.

So new shoud be malloc and cout should be printf. Also s = "mango" will change s to pointer to the readonly location of "mango" (if it even compiles)

Here is a proper example:

Code:

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

int main (void)
{
  char *s =  malloc(6);
  strcpy(s, "mango");
  puts(s);
  s[3] = 'P';
  puts(s);
  free(s);
  return 0;
}

>what is unsigned ?? it should be unsigned int.
unsigned and unsigned int are the same thing.

>N.B : is it not necesarry to null terminate inside ?
When you reverse a string you want the present nul to stay where it is, lest you end up with an empty string by swapping the nul character with the first character.