"Integer Float" Exact Conversion to Float-CodePudding

I am working with a library that uses int for representing money amounts. For example, the following is what this looks like:

599 --> $5.99
500 --> $5.00
5 --> $0.05
0 --> $0.00
-599 --> $-5.99

I wrote the following function which does an exact float conversion without causing floating point errors, but what I am wondering is if I am approaching this the wrong way or if there is a much cleaner solution.

def money_int_to_float(value: int) -> float:
    neg: str = '-' if value < 0 else ''
    value = abs(value)

    cents: int = value % 100
    remaining: int = value - cents
    dollars: int = int(str(remaining).removesuffix('00'))
    return float(f'{neg}{dollars}.{str(cents).zfill(2)}')

My reasoning for wanting an exact float conversion is that when dealing with money you can never be too careful with floats.

CodePudding user response：

divmod sounds like something you should know about:

def money_int_to_str(value: int) -> str:
    sign = '-' if value < 0 else ''
    dollars, cents = divmod(abs(value), 100)
    return f'${sign}{dollars}.{cents:02}'

the reason for staying away from floats would be that, e.g., money_int_to_float(123456789123456789) won't give you 89 cents from any implementation – there isn't enough precision in a float to store it.

CodePudding user response：

Thanks to @Samwise's comment, I have found a clean solution to this question. Using the Decimal class we can rewrite the function as follows:

from decimal import Decimal

def money_int_to_float(value: int) -> float:
    return float(Decimal(value).quantize(Decimal('0.00')).shift(-2))