Vrftjkry

So I'm working on a small project whereas I want to take mock data and separate them into structs. But I was thinking of the issue of people with multiple first and/or last names.

I want to write first names like you would do (like "Michael") and last names in all capital letters (like "JAMESON").

But what if I'm reading a name like Michael Daniel VAN DOORNE, etc. I don't know how I'd be able to separate "Michael Daniel" as first name, and "VAN DOORNE" as the last name. I tried to separate by stopping at the first capital letter, but I am of course capitalizing the first letter in someone's first names as well.

Example:

I want to read Michael Daniel VAN DOORNE, and separate it into "Michael Daniel" as firstname, and "VAN DOORNE" as the surname.

sscanf(buffer, "%s %s", firstName, lastName);

That wouldnt work naturally. But i am kinda stuck on coming up with a solution for mock names with multiple first and last names.

As you seem to be in total control of the data, I rather recommend a different approach:

A specific separator character in between forename(s) and surname(s). Then you don't rely on case sensitivity any more, especially the single character name issue appearing in another answer isn't an issue any more.

Separator character should be one that won't ever appear in any name, such as a tab (in contrast to space) character, #, '|', ... Even comma or semicolon should be fine, though the period might appear in abbreviated names and thus should not be used.

So knowing if it is part of a first name or last name is a bit of a challenge, but from the sound of it, you are in control of the data, so you can either lowercase the first name and capitalize the last or use some other method.

Breaking up the string, this is relatively easy by using strtok.

Making some assumptions that you are reading names line by line and stuffing them into buffer.

Use strtok to break buffer into "names".

char *token

token = strtok(buffer, " "); //note the second parameter is what you want to parse the array by

while(token != NULL)

{

    if(isupper(token[0]))

        //store that token into your struct (first or last name) allow for multiple

    else

        //store into the other 



    token = strtok(NULL, " "); //keep looping on string for names

}

strspn and strcspn could be used to iterate through the string and locate the first word that is all uppercase.

#include <stdio.h>

#include <string.h>



int main( void) {

    char line = "Michael Daniel VAN DOORNE";

    char upper = "ABCDEFGHIJKLMNOPQURSTUVWXYZ";

    int last = 0;

    int start = 0;

    int check = 0;

    do {

        last = start + check;

        check += start + 1;

        start = strspn ( line + check, upper);//span of characters that are uppercase

        if ( ' ' != line[start + check]) {//is the next character a space

            start = strcspn ( line + check, " ");//find the space

            if ( line[start + check]) {//not the terminatins zero

                start++;

            }

        }

    } while ( line[start + check] && ' ' != line[start + check]);



    printf ( "%sn", &line[last]);



    return 0;

}

Assuming last names are always written in upper case, start reading the string from the end and see when you have your last lower case.

int i=strlen(buffer)-1;

while(!islower(buffer[i]) && i>0)

    i--;

strncpy(firstName,buffer,i+1);

strcpy(lastName,&buffer[i+2]);

Here's another solution.Read until there are two capitals after each other or a capital and a space. Then use pointer arithmetic to fill first name and lastname.

char name = "Michael Daniel VAN DOORNE";



char *p = name;



char firstname[100] = { 0 };

char lastname[100] = { 0 };



while (*p)

{

    if (isupper(p[0]) && (isupper(p[1]) || p[1] == ' '))

    {

        strcpy(lastname, p);

        strncpy(firstname, name, p - name - 1);

        break;

    }

    p++;

}

If you're working with ASCII, here is a charset-specific trick that will help you:

#define TWOUPPER(c0, c1) (!((c0) & 32) && !((c1) & 32))

This will work even on single character last names since the null character will fail the 5th bit check, and single character middle names will not be taken as the last name since the following space will not succeed the test.

Works with the following test cases for me by comparing every two characters in the string and stopping on a match:

    char test1[100] = "Otto VON BISMARK",

    test2[100] = "Johannes Diderik VAN DER WAALS",

    test3[100] = "Vincent VAN GOGH",

    test4[100] = "Govind A B C D P"; // Only the "P" is counted as the last name here