Question

I made this range function so I could have something other than an integer step, and it works, but I am wondering why the floats are not truncated.

def drange(start, step):
    values = []
    r = start
    while r >= 0:
        values.append(r)
        r += step
    return values


print drange(2, -0.2)

Upon debugging I find that instead of this printing

[2, 1.8, 1.6, 1.4, 1.2, 1.0, 0.8, 0.6, 0.4, 0.2, 0]

it instead prints

[2, 1.8, 1.6, 1.4000000000000001, 1.2000000000000002, 1.0000000000000002, 0.8000
000000000003, 0.6000000000000003, 0.4000000000000003, 0.2000000000000003, 2.7755
575615628914e-16]

Lol, no wonder my module isn’t working. Why does this happen and how might I fix it?

Answer 1

This is correct behavior, since one cannot express 0.2 = 1/5 in base 2, just like there is no way to express 1/3 in base 10.

Use decimal instead if you want to calculate in base 10.

Additionally, you should really use a generator, as in

def drange(start, step):
    r = start
    while r >= 0:
        yield r
        r += step
print list(drange(2, -0.2))

That allows users of drange to iterate over the values without memory being allocated for the whole list.