Pass

Very well done Carl!

I like how you dereferensed strlen to write your own! It's a great way to maintain readability (given that your overwrite has the same funtion of course).

Here are some notes on your code:

Cyber-Clara och anmälningslistorna

Your last-length-digit-compare optimisation is expensive. You attempt to avoid calling the (most likely [differs between implementations]) linear strcmp to instead call the linear strlen one more time and assign additional memory. Given that most names differs in length, your optimisation has merit. However, strcmp do also compare the first characters, hence there exist a douplication of logic. Additionally, given that names often differs in characters, strcmp is better than strlen. Therefore I'm sceptical to if your optimisation should count as an optimisation. As proof, David's code also run at 0.5 and he doesn't even call gets over scanf:
David's solution:

int main()
{
   //get the number of names
   unsigned long int size;
   char name[42], surname[21];
   scanf("%lu", &size);
   char names[size][42];

   for (int i = 0; i < size; i++)
   {
      scanf("%s", name);
      strcpy(names[i], name);
      strcat(names[i], " ");
   }

   for (int i = 0; i < size; i++)
   {
      scanf("%s", surname);
      strcat(names[i], surname);
   }

   qsort(names, size, sizeof(names[0]), strcmp);

   unsigned long int output = size;
   for (int i = 1; i < size; i++)
   {

      if (strcmp(names[i - 1], names[i]) == 0)

         output--;
   }

   printf("%lu", output);
}

Keep it up!

Programming is like sex:
One mistake and you have to support it for the rest of your life.

The issue contains feedback and a C code example but names no file, test, entry point, or requested change. No completion condition is provided, so clarify the intended work before starting.