Storing components of a complex number

I'm writing a program in which a user enters a complex number (a + bi). I need to store the real numbers a and b in floating points *x1 and *i1 respectively but I'm having a hard time figuring out how to approach it.

It would be easy to ask for each part individually, but the program I have to write doesn't allow for that. The user enters the entire number.

Can anyone give me any suggestions on how to look at it?

You'll need to "parse" the string "by hand". You can perhaps use strtok() to split the string, or write your own code that walks through the string and splits it into it's respective components.

--
Mats

Okay, using strtok(a, b), the command will search through a to find a character that doesn't match what is in b and then turns the character after that into a null character.

so if a = 1.000+2i, if I use strtok(a, "+") it will stop at + and change 2 into null, leaving the i.

Is there a way with strtok to search for a token that is an arbitrary number to that the + can be changed into null instead of the number following the +?

strtol() and strtod() will respectively parse ints and floats, AND set a pointer to the point where they managed to parse to (in other words, where your + might be).

strtod() can be used to parse floats. And according to all of the sites I've found explaining the function, it is of the format:

strtod(const char *nptr, char **endptr, base)

so assuming nptr is going to be the variable containing the complex number, it will read the value until it gets to the + and put it in endptr?

Sorry if this is a stupid question, I'm just not understanding what I'm reading on the pages I've found and the examples are cryptic.

This is what I currently have. I've been messing with declarations and things just to see if it makes any difference, so I know the type identifiers are probably all wrong.

Code:

void sub_sum(float *x1, float *i1, float *x2, float *i2)
{
        const char num1, num2;
        float a1, a2, b1, b2;
        char endptr1, junk, endptr2;

        a1 = 0;
        a2 = 0;
        b1 = 0;
        b2 = 0;

        printf("Enter first complex number: ");
        scanf("%c", &num1);

        printf("Enter second complex number: ");
        scanf("%c", &num2);

        a1 = strtod(num1, &endptr1);
        b1 = strtod(endptr1, &junk);
        a2 = strtod(num2, &endptr2);
        b2 = strtod(endptr2, &junk);
        x1 = &a1;
        i1 = &b1;
        x2 = &a2;
        i2 = &b2;

        printf("%.6f %.6f \n", *x1, *i1);
        printf("%.6f %.6f \n", *x2, *i2);

}

Why use strtok() for this?

Code:

float realpart, imagpart;
if(sscanf(input, "%f+%fi", &realpart, &imagpart) != 2)
{
    /* Parse failed */
}