Question

How do you make an algorithm to validate if a floating point value is a floating point value in C?

Answer 1

So far I've tried using isalpha(<variable name>) but it does not seem to work. Does this work for floating point values?

Answer 2

try the math library, particularly the ceilf() and floorf() functions, or maybe even the modf() function

Answer 3

http://ideone.com/q6FJu

Answer 4

Is it possible to accomplish this using only the libraries string.h and stdio.h?

Answer 5

So you are trying to test string representation of numbers?

Answer 6

I think it is possible.

Answer 7

let me try :-P I will only include <string.h> and <stdio.h>

Answer 8

Well, I'm asking the user to input a floating point value, and check if the input is indeed a floating point.

Answer 9

prompting the user? then it is very possible.

Answer 10

Using scanf, more specifically.

Answer 11

http://ideone.com/DpEMx

The idea is to get input as a string and parse it to test if it is a float.

Answer 12

I still think that using the math.h functions to test the float value itself is far better than using the string and stdio functions the way I just did :-P

Answer 13

This is what I've done so far, but it doesn't seem to be working well. Basically, I'm using strchr to see if the string AmtInput contains the characters ".0123456789", and assigning variables for later use. Where am I going wrong?

int i, FloatPoint;
char AmtInput[99];

int main() {

scanf("%s", AmtInput);
getchar();

// Check if user input is a floating point value
for (i = 0; i <= strlen(AmtInput); i++) {
if (strchr(".0123456789", AmtInput[i]) != 0) {
printf("Not floating pt\n");
// User input IS a floating point value
FloatPoint = 1;
i = strlen(AmtInput) + 1;
}
}
}

Answer 14

Well it may be easier, but the problem is as of now, I don't know much about functions outside string.h and stdio.h. I may just end up stabbing in the dark if I use other libraries.

Answer 15

You won't be stabbing in the dark if you use anything out of the C standard library, though :-D Go ahead, try out the nice math library so that you can test it in one line instead of a complicated loop.

Answer 16

try posting your code on a site like http://ideone.com with input to see how it works.

Right now I think your code's logic is a bit messed up. in particular, the if block contains two different things, one saying the input is not a float and another saying your input is a float.

What happens is the first character (which is always a digit or a decimal point whether it is a whole number or a fractional number) is passed to strchr, returning a non-null pointer and therefore executing the code in the ifblock, saying that your number is "Not floating pt"

Answer 17

Yes. You can do it only using <string.h> and <stdio.h> header file. Try strtok() library function. I think it could help .

Answer 18

Just think that, Floating point number can be divide into token.( By split ".") :D , I think you got my idea.

Answer 19

or just do this

#include <math.h>

int is_float(double num)
{
return ceil(num) != floor(num)
}

Answer 20

I forgot to say that I don't need to check if it's a floating point per se; I just need the algorithm to recognize if it contains characters aside from . 0 1 2 3 4 5 6 7 8 9.

Here's my take:

AmtInput[99]
FloatSymbols[11] = {".0123456789"}
// Check if user input contains characters aside from ".0123456789"
for (i = 0; i <= strlen(AmtInput); i++) {
for (j = 0; j < 11; j++) {
if (FloatSymbols[j] == AmtInput[i]) {
FCount++;
}
}
}

// Check for validity of input
if(FCount != strlen(AmtInput)) {
MsgInvalidInput();
}

There's more to my source code that's not relevant to the question, so I'm leaving them out. For my simple purposes, this works.