The "@ P" and "@ PP" Syntax

Related discussion: #6

Motivation

For partial functions, currently we have syntax F(func, *args, **kwargs) or FF(func, *args, **kwargs). The syntax have some drawbacks:

1. It's hard to tell what the function is at the first sight, especially when func is a lambda expression.
2. It's hard to tell what the parameters are, since the function and parameters are enclosed by the same parenthesis.

This PR propose an alternative syntax, placing the function to the very front of the expression, and separating function and the parameters using some operators. The syntax is described in the following section.

Syntax Description

The new syntax has four variants:

1. func @ P(*args, **kwargs) is equivalent to F(func, *args, **kwargs);
2. func @ P is equivalent to F(func);
3. func @ PP(*args, **kwargs) is equivalent to FF(func, *args, **kwargs);
4. func @ PP is equivalent to FF(func).

Examples

# Sum up all odd numbers in [0, 9]
range(10) | filter @ P(lambda x: x % 2) | sum @ P == 25
# equiv. to
range(10) | F(filter, lambda x: x % 2) | F(sum)

# Spread parameters to a summing function
[1, 2, 3] | (lambda x, y, z: x + y + z) @ PP == 6
# equiv. to
[1, 2, 3] | FF(lambda x, y, z: x + y + z)