This is your standard vanilla range functions, converted to work as an iterator: UPDATE: xrange already does that. So feel free to ignore this.
Download
Copy to clipboard1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 | def irange(start, stop=None, step=None):
'''Return a range - as iterator.'''
if stop == None:
start, stop = 0, start
if stop > start:
_comp = lambda x, y: x <= y
step = step or 1
else:
_comp = lambda x, y: x >= y
step = step or -1
if _comp(step, 0): # Safeguard: Wrong way => empty range
raise StopIteration
while _comp(start, stop):
yield start
start += step
|
One thing that has long puzzled me about python is its awkward for i in range(...) construct. Not the look of it, mind you - syntactically it is, IMHO, most elegant. But it may take lots of memory, if the range is very big.
The thing that seems to be missing from the itertools package is an "irange" function that does the same as range, only with iterators. So, here you have it.
Note: It even does one thing "better" than the original range, namely automagically choosing a negative step if stop < start and no step is specified. If you dislike this, you can move the "step = step or 1" out of the if and remove the "step = step or -1". Happy iterating!
UPDATE: I should have looked into the function reference...
