Arithmetic NSNumber

Question

I want to do some simple arithmetic in NSNumbers and save the type. Is it possible?

For instance:

- (NSNumber *)add:(NSNumber *)firstNumber to:(NSNumber *)secondNumber; 

Is the definition of my method, and firstNumber and secondNumber are integers, then I would like to return an integer, which is their integer. Similarly, if both are doubles, then to return the result as a double.

It looks like I can get the type (other than the boolean) using [NSNumber objCType] found in this question: get the type NSNumber , but I can't seem to extract these types and do the calculations without a lot of code to extract the values, do the calculation and return the result for each possible type.

Is there a short and concise way to do this?

Answer 1

If you want to do arithmetic, the best option would be to use NSDecimalNumber .

NSDecimalNumber have methods for performing arithmetic operations, such as:

 – decimalNumberByAdding: – decimalNumberBySubtracting: – decimalNumberByMultiplyingBy: – decimalNumberByDividingBy: – decimalNumberByRaisingToPower: – decimalNumberByMultiplyingByPowerOf10: – decimalNumberByAdding:withBehavior: – decimalNumberBySubtracting:withBehavior: – decimalNumberByMultiplyingBy:withBehavior: – decimalNumberByDividingBy:withBehavior: – decimalNumberByRaisingToPower:withBehavior: – decimalNumberByMultiplyingByPowerOf10:withBehavior: 

And since NSDecimalNumber extends NSNumber, it also has all the NSNumber methods, so I think you can use it in your case without any problems.

Answer 2

For almost all applications, it will be convenient to convert to double and back using -doubleValue and –initWithDouble: This will allow you to use the standard characters C (+, -, ...) and the functions ( exp() , sin() ). The only way you would run into a problem is to use maximum precision for 64-bit integer values.

If you want to stick with the operations of the Objective-C class, you can use NSDecimalNumber .

See also: How to add two NSNumber objects?

Answer 3

How about calculating the value of an expression as double (with all inputs as double values) and then checking if the result is integer? Then you can use NSNumber numberWithInt: or NSNumber numberWithDouble: to return the result.

When you check if the result value is integer, be sure to consider the rounding error (for example, when 1 is expressed as 0.99999999, etc.).

EDIT . Just noticed in docs for NSNumber this phrase:

Note that numeric objects do not necessarily preserve the type that they are created with.

I think this means that you cannot reliably do what you are trying to do.