Python and Ruby

Chris Rebert clp2 at rebertia.com
Mon Feb 1 01:50:50 CET 2010


On Sun, Jan 31, 2010 at 4:25 PM, Steven D'Aprano
<steven at remove.this.cybersource.com.au> wrote:
> On Sun, 31 Jan 2010 15:40:36 -0800, Chris Rebert wrote:
>> On Sun, Jan 31, 2010 at 2:36 PM, Steven D'Aprano
>> <steve at remove-this-cybersource.com.au> wrote:
>>> On Sun, 31 Jan 2010 04:28:41 -0800, Ed Keith wrote:
>>>> In most functional languages you just name a function to access it and
>>>> you do it ALL the time.
>>>>
>>>> for example, in if you have a function 'f' which takes two parameters
>>>> to call the function and get the result you use:
>>>>
>>>>  f 2 3
>>>>
>>>> If you want the function itself you use:
>>>>
>>>>    f
>>>
>>> How do you call a function of no arguments?
>>
>> It's not really a function in that case, it's just a named constant.
>> (Recall that functions don't/can't have side-effects.)
>
>
>>>> time.time(), random.random()
> (1264983502.7505889, 0.29974255140479633)
>>>> time.time(), random.random()
> (1264983505.9283719, 0.74207867411026329)
>
>
> They don't look terribly constant to me.

Those aren't functions in the pure functional programming sense; which
is unsurprising since Python isn't a [pure] functional language.
They both involve side-effects. time() does I/O to the clock chip to
see what time it is, and random() uses and changes a global seed value
variable (which, in a double-whammy, takes its initial value from
time()).

Pure functions must be referentially transparent
[http://en.wikipedia.org/wiki/Referential_transparency_(computer_science)],
and as you've demonstrated, neither of those Python functions qualify.

Cheers,
Chris
--
http://blog.rebertia.com



More information about the Python-list mailing list