Range with step of type float [duplicate]

Question

The documentation basically says that range must behave exactly as this implementation (for positive step):

def range(start, stop, step):
  x = start
  while True:
    if x >= stop: return
    yield x
    x += step

It also says that its arguments must be integers. Why is that? Isn't that definition also perfectly valid if step is a float?

In my case, I am esp. needing a range function which accepts a float type as its step argument. Is there any in Python or do I need to implement my own?

---

More specific: How would I translate this C code directly to Python in a nice way (i.e. not just doing it via a while-loop manually):

for(float x = 0; x < 10; x += 0.5f) { /* ... */ }

Answer 1

You could use numpy.arange.

EDIT: The docs prefer numpy.linspace. _{Thanks @Droogans for noticing =)}

Answer 2

One explanation might be floating point rounding issues. For example, if you could call

range(0, 0.4, 0.1)

you might expect an output of

[0, 0.1, 0.2, 0.3]

but you in fact get something like

[0, 0.1, 0.2000000001, 0.3000000001]

due to rounding issues. And since range is often used to generate indices of some sort, it's integers only.

Still, if you want a range generator for floats, you can just roll your own.

def xfrange(start, stop, step):
    i = 0
    while start + i * step < stop:
        yield start + i * step
        i += 1

Answer 3

In order to be able to use decimal numbers in a range expression a cool way for doing it is the following: [x * 0.1 for x in range(0, 10)]