2.3 list reverse() bug?

[Alex Martelli]

Paul Rubin wrote:

> Andrew MacIntyre <andymac at bullseye.apana.org.au> writes:
>> > list(x) is my favourite way to make a shallow copy of list x, but x[:]
>> > and copy.copy(x) are essentially equivalent.
>> 
>> Behaviourally yes, performance wise no.
> 
> Eh?  Last time I looked at the copy module, x[:] is exactly what it did.

copy.copy(x) will have two extra name look-ups at the start (one for
the outer 'copy' name, one for the 'copy' name inside the module), a
function call, and inside that function a type(x) and a dispatch via
a type-to-function dictionary.  list(x) has roughly similar overhead.
Apparently, these miniscule extra bits of overhead appear to be enough
for Andrew to reject the "essentially equivalent" evaluation.


Alex