On Jun 23, 2019, at 17:01, Steven D'Aprano <steve@pearwood.info> wrote:

...

On Sun, Jun 23, 2019 at 03:22:37PM -0400, James Lu wrote: Make $ a valid identifier and a singleton.

$ is a useful placeholder in [].

Possible function partial syntax:

def foo(x, y): print(x, y)

partialized = foo[$, 10]

I don't think that creating partial functions is important enough to:

1. make partial functions a built-in language feature; 2. using syntax (square brackets); 3. and require a new built-in singleton.

I think that functools.partial is missing two features:

- the ability to close over positional parameters from the right, rather than the left;

- the ability to skip parameters.

The first could be solved with a "rpartial" function, and the second by using ellipsis ... as a placeholder, or a named constant in the functools module.

I’m pretty sure there are multiple improved-partial projects on PyPI. Maybe just picking out one and adding a link to it from the functools docs is sufficient?

On Jun 23, 2019, at 17:01, Steven D'Aprano <steve@pearwood.info> wrote:

...

On Sun, Jun 23, 2019 at 03:22:37PM -0400, James Lu wrote: Make $ a valid identifier and a singleton.

$ is a useful placeholder in [].

Possible function partial syntax:

def foo(x, y): print(x, y)

partialized = foo[$, 10]

I don't think that creating partial functions is important enough to:

1. make partial functions a built-in language feature; 2. using syntax (square brackets); 3. and require a new built-in singleton.

I think that functools.partial is missing two features:

- the ability to close over positional parameters from the right, rather than the left;

- the ability to skip parameters.

The first could be solved with a "rpartial" function, and the second by using ellipsis ... as a placeholder, or a named constant in the functools module.

I’m pretty sure there are multiple improved-partial projects on PyPI. Maybe just picking out one and adding a link to it from the functools docs is sufficient?