When working with money values in Django, the first thought is to use FloatField to represent currency in the model. However, FloatField uses the float type internally which comes with some precision issues.

The problem

Take a look at this piece of code. It's just a simple operation with float numbers in Python.

>>> .1 + .1 + .1 == .3
False
>>> .1 + .1 + .1
0.30000000000000004

Normally, you think that the addition will make sense but because of some issues with the float approximation, the equation is not equal to 3.
You can fix these issues using rounding but if you are dealing with money values and precision matters, it might be time to use decimals.

The solution

Basically, use decimals instead of floats when precision matters. Let's rewrite the previous example but with Decimal.

>>> from decimal import Decimal
>>> Decimal('.1') + Decimal('.1') + Decimal('.1') == Decimal('.3')
True
>>> Decimal('.1') + Decimal('.1') + Decimal('.1')
Decimal('0.3')

Notice that here we are initializing the decimals values with string values. You can use floats but as we said earlier, floats have their approximation issues.

Then when working with Decimal in Django with the DecimalField, it's always a good habit to precise the decimal_places attribute when defining the field.

class Product(models.Model):
    title = models.CharField(max_length=64)
    description = models.TextField()

    price = models.DecimalField(max_digits=6, decimal_places=2)

You can learn more about DecimalField here.

Article posted using bloggu.io. Try it for free.