PEP 309 (Partial Function Application) Idea

[Ronald Mai]

In my opinion, Ellipsis might be in the middle, not only in leftmost or
rightmost, so a placeholder approach can be much more flexible and
convenient.

Here is a reference implementation:

_ = lambda x: x.pop(0)

def partial(func, *args, **keywords):
def newfunc(*fargs, **fkeywords):
newkeywords = keywords.copy()
newkeywords.upd ate(fkeywords)
newargs = (lambda seq: tuple([(a == _ and a(seq)) or a for
a in args] + seq))(list(farg s))
return func(*newargs, **newkeywords)
newfunc.func = func
newfunc.args = args
newfunc.keyword s = keywords
return newfunc

Here is example of use:
[color=blue][color=green][color=darkred]
>>> def capture(*args):[/color][/color][/color]
return args
[color=blue][color=green][color=darkred]
>>> partial(capture )()[/color][/color][/color]
()[color=blue][color=green][color=darkred]
>>> partial(capture , _)(1)[/color][/color][/color]
(1,)[color=blue][color=green][color=darkred]
>>> partial(capture , _, 2)(1)[/color][/color][/color]
(1, 2)[color=blue][color=green][color=darkred]
>>> partial(capture , 1)(2)[/color][/color][/color]
(1, 2)[color=blue][color=green][color=darkred]
>>> partial(capture , 1, _)(2)[/color][/color][/color]
(1, 2)[color=blue][color=green][color=darkred]
>>> partial(capture , 1, _)()[/color][/color][/color]
IndexError: pop from empty list[color=blue][color=green][color=darkred]
>>> partial(capture , 1, _, _)(2, 3)[/color][/color][/color]
(1, 2, 3)

Chris Perkins :
[color=blue]
> Random idea of the day: How about having syntax support for
> currying/partial function application, like this:
>
> func(..., a, b)
> func(a, ..., b)
> func(a, b, ...)
>
> That is:
> 1) Make an Ellipsis literal legal syntax in an argument list.
> 2) Have the compiler recognize the Ellipsis literal and transform the
> function call into a curried/parially applied function.
>
> So the compiler would essentially do something like this:
>
> func(a, ...) ==> curry(func, a)
> func(..., a) ==> rightcurry(func , a)
> func(a, ..., b) ==> rightcurry(curr y(func,a), b)
>
> I haven't though this through much, and I'm sure there are issues, but
> I do like the way it looks. The "..." really stands out as saying
> "something is omitted here".
>
>
> Chris Perkins[/color]

[Giovanni Bajo]

Re: PEP 309 (Partial Function Application) Idea

Ronald Mai wrote:
[color=blue]
> Here is a reference implementation:
>
> _ = lambda x: x.pop(0)
>
> def partial(func, *args, **keywords):
> def newfunc(*fargs, **fkeywords):
> newkeywords = keywords.copy()
> newkeywords.upd ate(fkeywords)
> newargs = (lambda seq: tuple([(a == _ and a(seq)) or a for
> a in args] + seq))(list(farg s))
> return func(*newargs, **newkeywords)
> newfunc.func = func
> newfunc.args = args
> newfunc.keyword s = keywords
> return newfunc
>
> Here is example of use:
>[color=green][color=darkred]
>>>> def capture(*args):[/color][/color]
> return args
>[color=green][color=darkred]
>>>> partial(capture )()[/color][/color]
> ()[color=green][color=darkred]
>>>> partial(capture , _)(1)[/color][/color]
> (1,)[color=green][color=darkred]
>>>> partial(capture , _, 2)(1)[/color][/color]
> (1, 2)[color=green][color=darkred]
>>>> partial(capture , 1)(2)[/color][/color]
> (1, 2)[color=green][color=darkred]
>>>> partial(capture , 1, _)(2)[/color][/color]
> (1, 2)[color=green][color=darkred]
>>>> partial(capture , 1, _)()[/color][/color]
> IndexError: pop from empty list[color=green][color=darkred]
>>>> partial(capture , 1, _, _)(2, 3)[/color][/color]
> (1, 2, 3)[/color]

Other implementations I have seen (boost::bind comes to mind) use ordered
placeholders such as _1, _2, _3, etc to provide more flexibility in adaptation:
[color=blue][color=green][color=darkred]
>>> partial(capture , "a", _1, _2)("b", "c")[/color][/color][/color]
("a", "b", "c")[color=blue][color=green][color=darkred]
>>> partial(capture , "a", _2, _1)("b", "c")[/color][/color][/color]
("a", "c", "b")

I don't see mention of this in the PEP, but it's a nice feature to have IMO.
--
Giovanni Bajo

[bonono@gmail.com]

Re: PEP 309 (Partial Function Application) Idea


Giovanni Bajo wrote:[color=blue]
> Ronald Mai wrote:
>[color=green]
> > Here is a reference implementation:
> >
> > _ = lambda x: x.pop(0)
> >
> > def partial(func, *args, **keywords):
> > def newfunc(*fargs, **fkeywords):
> > newkeywords = keywords.copy()
> > newkeywords.upd ate(fkeywords)
> > newargs = (lambda seq: tuple([(a == _ and a(seq)) or a for
> > a in args] + seq))(list(farg s))
> > return func(*newargs, **newkeywords)
> > newfunc.func = func
> > newfunc.args = args
> > newfunc.keyword s = keywords
> > return newfunc
> >
> > Here is example of use:
> >[color=darkred]
> >>>> def capture(*args):[/color]
> > return args
> >[color=darkred]
> >>>> partial(capture )()[/color]
> > ()[color=darkred]
> >>>> partial(capture , _)(1)[/color]
> > (1,)[color=darkred]
> >>>> partial(capture , _, 2)(1)[/color]
> > (1, 2)[color=darkred]
> >>>> partial(capture , 1)(2)[/color]
> > (1, 2)[color=darkred]
> >>>> partial(capture , 1, _)(2)[/color]
> > (1, 2)[color=darkred]
> >>>> partial(capture , 1, _)()[/color]
> > IndexError: pop from empty list[color=darkred]
> >>>> partial(capture , 1, _, _)(2, 3)[/color]
> > (1, 2, 3)[/color]
>
> Other implementations I have seen (boost::bind comes to mind) use ordered
> placeholders such as _1, _2, _3, etc to provide more flexibility in adaptation:
>[color=green][color=darkred]
> >>> partial(capture , "a", _1, _2)("b", "c")[/color][/color]
> ("a", "b", "c")[color=green][color=darkred]
> >>> partial(capture , "a", _2, _1)("b", "c")[/color][/color]
> ("a", "c", "b")
>
> I don't see mention of this in the PEP, but it's a nice feature to have IMO.
> --
> Giovanni Bajo[/color]
Since python has named parameter(and I assume this PEP would support it
as well), is it really that useful to have these place holder things ?
As when the parameter list gets long, named param should be easier to
read.

The only case I find it useful is for binary ops where I would like to
either bind the left hand side or the right hand side but that can be
handled easily with a "flip" function as in haskell.

[Giovanni Bajo]

Re: PEP 309 (Partial Function Application) Idea

bonono@gmail.co m wrote:
[color=blue]
> Since python has named parameter(and I assume this PEP would support
> it as well), is it really that useful to have these place holder things
> ?[/color]

Probably not so much, you're right.
[color=blue]
> As when the parameter list gets long, named param should be easier to
> read.
>
> The only case I find it useful is for binary ops where I would like to
> either bind the left hand side or the right hand side but that can be
> handled easily with a "flip" function as in haskell.[/color]

I'm not sure I'd like a specialized flip function, compared to the numbered
placeholders solution.
--
Giovanni Bajo

[Ronald Mai]

Re: PEP 309 (Partial Function Application) Idea

for a new function, named parameter might be a better choise, but for
existing function, placeholder would be better, since you don't need to
modify the exiting code.

[Ronald Mai]

Re: PEP 309 (Partial Function Application) Idea

sorry, should be "existing"

[Ronald Mai]

Re: PEP 309 (Partial Function Application) Idea

for a new function, named parameter might be a better choise, but for
existing function, placeholder would be better, since you don't need to
modify the existing code.