range creates a list in memory with the number of elements defined by the programmer. Since xrange has more performance, I do not know if in all cases, since it does not end up generating a list.
Should I always use xrange above range ? Are there instances where the use of range is more advantageous than the use of xrange ?
In fact, in Python 3 there is no longer the equivalent of the range of Python 2 - the range is now what was the old xrange.
But in the code in Python 2, yes, there is no reason to use the normal range. The range (from Python 3 - xrange) "seems" a function, but it is a class, quite simple to do the same. And the efficiency is so much bigger than a list, that the question that remains is "my God, how did you ever make the range of Python 2 ??": -)
In very little case in which you will call a function that really needs a list, you can convert a range object to a list by doing list(xrange(10)) .
How to create your: everything an object needs is a way to (1) retrieve a given number from an index, (2) need to interact numbers starting at "beginning", and going to the "end" of "step by step", if used in for , for example. (3) Also needs to return its own length.
In Python, this means you have to have a class that implements (1) __getitem__ , (2) __iter__ , (3) __len__ :
class MyRange(object): # em Python 2 é obrigatório herdar de object
def __init__(self, start, stop=None, step=1):
if stop is None:
stop = start
start = 0
self.start = start
self.stop = stop
self.step = step
def __getitem__(self, index):
return self.start + self.step * index
def __len__(self):
return (self.stop - self.start) // self.step
def __iter__(self):
v = self.start
while (v < self.stop) if self.step > 0 else (v > self.stop):
yield v
v += self.step
This class is functionally the same. It does not handle some corner cases, but it also does not check if any of the values is an integer: that is, it works for decimal numbers as well.