Objective-C: how to get 123 from 0.123 or generically xyz... from 0.xyz...?

Question

As title states, if I have a float, I need to get the fraction part as an integer, how do I do it?

I was thinking:

get index(position) of decimal point
then I can know how many digits after decimal point
get those digits as substring
convert it to an integer

is there any better/smarter way?

update:

I forgot to mention, the float has format like: X.YZ so there are at most two digits after decimal point.

What is your exact requirement? If you're using the float type, even a number like 12 might have an exact value of 12.000001. If the number is captured from the user, you should probably be using a char* or NSString instead. — Rajesh J Advani, Jul 31 '12 at 06:37

score 9 · Answer 1 · edited May 23 '17 at 12:09

9

You use the modf function:

double integral;
double fractional = modf(some_double, &integral);

refer this fractional part of NSDecimalnumber

edited May 23 '17 at 12:09

Community

1
1

answered Jul 31 '12 at 06:43

Hector

3,909
2
30
46

1

Bad copy paste, I think you've meant `modff` - `modf` operates on doubles. – matm Jul 31 '12 at 09:37

score 5 · Accepted Answer · answered Jul 31 '12 at 07:04

5

 float temp=12.123;
NSString *str=[NSString stringWithFormat:@"%f",temp];
NSArray *arr=[str componentsSeparatedByString:@"."];
int tempInt=[[arr lastObject] intValue];

answered Jul 31 '12 at 07:04

Nikunj

987
11
25

I picked your answer but I need to make some change in my case, as the 'temp' is a returned value from some method, it may be 12.12300001, but I only need 123 or even 12, so I changed %f to %.2f and removed the last 0 if there is any. – hzxu Jul 31 '12 at 07:17
hmm need change as per your above requirement – Nikunj Jul 31 '12 at 07:40

Inder Kumar Rathore · Answer 3 · 2012-07-31T12:19:50.540

4

NSString *str = [NSString stringWithFormat:@"%f", value];
NSArray *arr = [str componentsSeparatedByString:@"."];
int num = [[arr objectAtIndex:1] intValue];

edited Jul 31 '12 at 12:19

answered Jul 31 '12 at 06:54

Inder Kumar Rathore

39,458
17
135
184

Small mistake: pointer on `int`? `intValue` returns `(int)`, not `(int *)`. – matm Jul 31 '12 at 09:36

score 2 · Answer 4 · answered Jul 31 '12 at 06:42

2

Let take a example: x = 129.567;

convert x into integer and put in y. y = int(x); so y = 129;

now subtract y from x. so z = x-y; z = .567

z = z*1000; so z = 567

I think thats what you are looking for.

answered Jul 31 '12 at 06:42

Rahul Bhansali

135
1
1
8

@Hector for 129.5678, after getting .5678 you can find the length of your fraction and multiply 10 that many times. What say??? – Rahul Bhansali Jul 31 '12 at 06:55
But I don't know how many digits after ., so if it is 1.1, then I will get 1000 instead of 1 – hzxu Jul 31 '12 at 07:07
@HZXU you have to find the length of fraction and multiply 10 that no. of times. – Rahul Bhansali Jul 31 '12 at 07:10
@HZXU As you are saying there at most 2 digit after decimal point. So you can check X.YZ whether z is 0 or non zero. if z is 0 multiply by 10 else multiply by 100. – Rahul Bhansali Jul 31 '12 at 07:18

score 1 · Answer 5 · answered Jul 31 '12 at 06:42

1

How about taking the float and multiply by 1000 and convert the result to an integer?

answered Jul 31 '12 at 06:42

Tommy Hui

1,306
6
9

1

this isn't an answer.. it should be a comment – Inder Kumar Rathore Jul 31 '12 at 06:48

score 0 · Answer 6 · answered Jul 31 '12 at 06:38

0

Take a look at NSDecimalNumber and NSDecimal.

answered Jul 31 '12 at 06:38

Julien

3,427
20
22

score 0 · Answer 7 · answered Jul 31 '12 at 06:48

0

Float never represents a number with accuracy. The best way is to create a string from "double" number and then do the string manipulation to get the digits after decimal point.

answered Jul 31 '12 at 06:48

Apurv

17,116
8
51
67

score 0 · Answer 8 · edited May 23 '17 at 12:09

Here's my solution:

float value = 12.345f;
float integral;
float divisor = 1000.0f;
float fractional = modff(value, &integral);  // breaks a float into fractional and integral parts

int xyz = (int)fmodf(fractional*divisor, divisor);  // modulo (cannot just use %)

NSLog(@"XYZ: %d", xyz); // logs "XYZ: 345"

By changing the value of divisor you can get the precision you want, e.g. only XY (100.0f) or X (10.0f).

The solution is not the most performant, but useful. I think it could be optimized using bitwise operations.

Notes on using "f":

Objective-C: how to get 123 from 0.123 or generically xyz... from 0.xyz...?

8 Answers8