Problem with structure with a float variable

Hi,

I created a structure which has 2 char arrays and 1 float variable. When trying to input data into the float variable the program crashes. On the output screen I can see the following error.

Code:

---

scanf : floating point formats not linked

Abnormal program termination.

---

Following is the code:

Code:

---


#include<stdio.h>

#define NAME_LIMIT 100
#define TEAM_LIMIT 20
#define NO_OF_PLAYERS 10

struct cricket
{
        char name[NAME_LIMIT];
        char team[TEAM_LIMIT];
        float bat_avg;
};
typedef struct cricket cricket;

void getS(char [], int);

void main()
{
        cricket player[NO_OF_PLAYERS];
        int i;
        clrscr();
        printf("Please enter details of %d players:-",NO_OF_PLAYERS);
        for(i=0;i<NO_OF_PLAYERS;i++)
        {
                printf("\n\nPlayer %d:-",i+1);

                printf("\nPlayer Name: ");
                getS(player[i].name,NAME_LIMIT);

                printf("\nTeam: ");
                getS(player[i].team,TEAM_LIMIT);

                printf("\nBatting Average:");
                scanf("%f",&player[i].bat_avg);
        }
}

void getS(char str[], int max)
{
        int i=-1;
        do
        {
                str[++i] = getche();
        }
        while(i<max-1 && str[i] != '\n' && str[i] != '\r');
        str[++i] = '\0';
}

---

I am using Turbo C compiler. I cannot understand what I am doing wrong. Please help me.

Thank you,
-Sreejit

Aside from void main being wrong (see the FAQ, and never mind Borland's autogenerated template) and the fact that your compiler is probably 13 years old (if it's TC 3.0, as I suspect), I think that you are missing some compiler options that make using scanf with floats possible. The error message hints at that.

Best solution would be to use a proper compiler.

Code:

---

void getS(char str[], int max)
{
        int i=-1;
        do
        {
                str[++i] = getche();
        }
        while(i<max-1 && str[i] != '\n' && str[i] != '\r');
        str[++i] = '\0';
}

---

As CornedBee said that u are using a very old compiler where u to seriously think about change your compiler. well, u could use DEV-C++ with the good IDE for your use and which is not that difficult to use that. Dev-c++ connects to one of the port of gcc compiler which basically mean u are compiling as per the gcc standards.

And in the above code you ar using a non standard function which is getche change to getchar which is standard function.

ssharish2005

I remember that I read somewhere that the borland compilers don't recognize that the floatingpoint support has to be linked to your program if the only place a float is used is in a scanf or printf statement.
A workaround is to initialize a dummy float somewhere in your program.
e.g.

Code:

---

int main() {
    float dummy = 1.0f;
    ....
}

---

Kurt

Thank you so much for the replies. I am using the compiler because I have to appear for a lab test next month and we will be asked to code on Turbo C.

Kurt, your solution of declaring and initializing a dummy variable did not help. However, I used the dummy variable to get the input. The following code works fine now:

Code:

---

scanf("%f",&dummy);
player[i].bat_avg = dummy;

---

But I still do not understand why does scanf have problems with a float variable inside a structure but not with a normal float variable.

Thanks anyway for the suggestions. I will certainly try and get a better compiler once I am through with my lab test.

-Sreejit

Quote:

---

Aside from void main being wrong (see the FAQ, and never mind Borland's autogenerated template) and the fact that your compiler is probably 13 years old (if it's TC 3.0, as I suspect), I think that you are missing some compiler options that make using scanf with floats possible. The error message hints at that.

Best solution would be to use a proper compiler.

---

CornedBee, Thank you for information on void main. I am a beginner and did not know that we are not supposed to use void with main. http://faq.cprogramming.com/cgi-bin/...&id=1043284376 helped.

Thanks a lot.

-Sreejit

In addition to what Zuk said, I thought you had to do some dummy calculation on the float to actually get the float support linked in (not just simply declaring a float variable), by calling a floating-point function, such as sin:

Code:

---

#include <math.h>

...

sin(dummy);

---

http://www.google.co.uk/search?q=flo...ats+not+linked

Thanks a lot Salem and hk_mp5kpdw. Salem, I tried google but my search text was ... well... stupid I guess, after looking at your search results. Thank you for the help.

Searching for pointed results is an acquired skill.
http://cboard.cprogramming.com/showp...71&postcount=4