Slicing [N::-1]

[Gary Herron]

Mensanator wrote:
> On Mar 5, 12:28 pm, Steven D'Aprano <st... at REMOVE-THIS-
> cybersource.com.au> wrote:
>   
>> On Fri, 05 Mar 2010 18:12:05 +0000, Arnaud Delobelle wrote:
>>     
>>>>>> l = range(10)
>>>>>> l
>>>>>>             
>>> [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
>>>       
>>>>>> l[7::-1]
>>>>>>             
>>> [7, 6, 5, 4, 3, 2, 1, 0]
>>>       
>>>>>> [l[i] for i in range(7, -1, -1)]
>>>>>>             
>>> [7, 6, 5, 4, 3, 2, 1, 0]
>>>       
>> Where does the first -1 come from? Slices are supposed to have default
>> values of 0 and len(seq):
>>     
>
> The only way to get a 0 from a reverse range() is to have a bound of
> -1.
>   

Not quite.  An empty second bound goes all the way to the zero index:

 >>> range(9)[2::-1]
[2, 1, 0]

Gary Herron




>   
>>>>> l[7::1]
>>>>>           
>> [7, 8, 9]
>>     
>>>>> [l[i] for i in range(7, len(l), 1)]
>>>>>           
>> [7, 8, 9]
>>     
>>>>> [l[i] for i in range(7, len(l), -1)]
>>>>>           
>> []
>>
>> I don't believe the actual behaviour is documented anywhere.
>>     
>
> Well, it's implied. If the stopping bound in a reverse range()
> is greater than the starting bound, you get an empty return.
>
>   
>> --
>> Steven
>>     
>
>