Strange behaviour with reversed()

[Duncan Booth]

Andreas Kraemer <akraemer at sbcglobal.net> wrote:

>> The only other behaviours I would regard as intuitive for iteration over
>> a mutating sequence would be to throw an exception either for mutating
>> the sequence while the iterator exists or for using the iterator after a
>> mutation.
> 
> Maybe it would have been slightly more intuitive if reversed() had
> been implemented like this,
> 
> def Reversed(seq):
>   for i in xrange(len(seq)-1,-1,-1):
>     yield seq[i]
> 
> so that the length of the sequence is determined when the iteration
> starts, not when the iterator is created?

Perhaps, but either way it comes down to "don't modify the sequence while 
iterating".