Python does integer division when both operands are integers. The behavior of Python’s division operators have changed from Python 2.x and 3.x (see also http://stackoverflow.com/documentation/python/809/incompatibilities-moving-from-python-2-to-python-3/2797/integer-division ).
a, b, c, d, e = 3, 2, 2.0, -3, 10In Python 2 the result of the ’ / ’ operator depends on the type of the numerator and denominator.
a / b # = 1
a / c # = 1.5
d / b # = -2
b / a # = 0
d / e # = -1Note that because both a and b are ints, the result is an int.
The result is always rounded down (floored).
Because c is a float, the result of a / c is a float.
You can also use the operator module:
import operator # the operator module provides 2-argument arithmetic functions
operator.div(a, b) # = 1
operator.__div__(a, b) # = 1What if you want float division:
Recommended:
from __future__ import division # applies Python 3 style division to the entire module
a / b # = 1.5
a // b # = 1Okay (if you don’t want to apply to the whole module):
a / (b * 1.0) # = 1.5
1.0 * a / b # = 1.5
a / b * 1.0 # = 1.0 (careful with order of operations)
from operator import truediv
truediv(a, b) # = 1.5Not recommended (may raise TypeError, eg if argument is complex):
float(a) / b # = 1.5
a / float(b) # = 1.5The ’ // ’ operator in Python 2 forces floored division regardless of type.
a // b # = 1
a // c # = 1.0In Python 3 the / operator performs ‘true’ division regardless of types. The // operator performs floor division and maintains type.
a / b # = 1.5
e / b # = 5.0
a // b # = 1
a // c # = 1.0
import operator # the operator module provides 2-argument arithmetic functions
operator.truediv(a, b) # = 1.5
operator.floordiv(a, b) # = 1
operator.floordiv(a, c) # = 1.0Possible combinations (builtin types):
int and int (gives an int in Python 2 and a float in Python 3)
int and float (gives a float)
int and complex (gives a complex)
float and float (gives a float)
float and complex (gives a complex)
complex and complex (gives a complex)
See PEP 238 for more information.