iterating in reverse

[Greg Chapman]

On 16 Jan 2003 15:46:32 -0800, jbperez808 at yahoo.com (Jonathan P.) wrote:

>First of all, many thanks to all the people in the list
>who have taken the time to respond to my earlier 
>questions with excellent answers. Now on to the
>problem at hand...
>
>A list's contents are ordered such that:
>
>>>> x=[1,2,3,4,5]
>
>>>> for i in x:
>      print i
>
>will always print x's contents in the
>same sequence.  Is there an elegant and
>efficient way to iterate through x in
>reverse without having to create a reversed
>copy of it (i.e. y=x[:]; y.reverse())?

How about something like this (in 2.3):

def seqiter(seq, *args):
    '''seqiter(seq, [[start], stop, [step]]) -> seq item iterator'''
    if not args:
        xr = xrange(len(seq))
    else:
        slc = slice(*args)
        xr = xrange(*slc.indices(len(seq)))
    for index in xr:
        yield seq[index]

for i in seqiter(x, None, None, -1):
    print i


The above avoids constructing a new list, but it might not be too efficient (I'd
guess that for most lists, it would be better to iterate over the slice, 
e.g. x[::-1]).  Perhaps the list object should grow a new method taking 
([start], stop, [step]) parameters and returning an iterator which produces the
items in the desired order.

---
Greg Chapman