Python3: on removing map, reduce, filter

[Steven Bethard]

beliavsky at aol.com wrote:
> Steve Bethard wrote:
>> Robert Kern wrote: 
>>>   def square(x):
>>>       return x*x
>>>   map(square, range(1000))
>>> 
>>> versus
>>>   [x*x for x in range(1000)]
>>> 
>>> Hint: function calls are expensive.
>> 
>> $ python -m timeit -s "def square(x): return x*x" "map(square, range(1000))"
>> 1000 loops, best of 3: 693 usec per loop
>> 
>> $ python -m timeit -s "[x*x for x in range(1000)]"
>> 10000000 loops, best of 3: 0.0505 usec per loop
> 
> Functions will often be complicated enought that inlining them is not
> feasible.

True, true.  However, list comprehensions still seem to be comparable in 
speed (at least in Python 2.4):

$ python -m timeit -s "def f(x): return x*x" "[f(x) for x in xrange(1000)]"
1000 loops, best of 3: 686 usec per loop

$ python -m timeit -s "def f(x): return x*x" "map(f, xrange(1000))"
1000 loops, best of 3: 690 usec per loop

Presumably this is because the C code for the byte codes generated by a 
list comprehension isn't too far off of the C code in map.  I looked at 
bltinmodule.c for a bit, but I'm not ambitious enough to try verify this 
hypothesis. ;)

Steve