avdosev / bindpy

The bindpy package allows for partial application of arguments to a function, making it easy to create specialized versions of the function with some arguments pre-filled.

bindpy

The bindpy package allows for partial application of arguments to a function, making it easy to create specialized versions of the function with some arguments pre-filled. It is a better version of the Python's standard partial function from the functools package inspired by C++'s std::bind.

Install

Expect gracefully crafted support for any version of Python 3+, but confidently tested in version 3.7 and higher.

pip install bindpy

Usage

from bindpy import *

bind function

The show function takes three arguments, a1, a2 and a3, and returns a string composed of their values separated by spaces. The show_10 function is a partially applied version of show, with a2 bound to _1, a1 bound to _2 and a3 bound to 10.

Bind support placeholders : _1, _2, ... _10. The placeholders allow you to partially apply a function and leave certain arguments to be filled in later. This allows you to reuse the partially applied function and pass different values for the placeholder argument.

Overall, bind and placeholders make it easier to create reusable and composable functions by allowing you to fix certain arguments and create new functions that take fewer arguments.

def show(a1, a2, a3):
    return " ".join(map(str, [a1, a2, a3]))
    
show_10 = bind(show, _2, _1, a3=10)

print(show_10(20, 30)) # output: 30 20 10

Convenient to use with functional style.

If you find lambda expressions unappealing, you can use bind for a more convenient and aesthetically pleasing experience with functional programming.

def add(a, b, c):
  return a + b * c
  
numbers = [1, 2, 3, 4]
print(list(map(bind(add, _1, 10, 2), numbers))) # output 21 22 23 24
# same code with lambda
print(list(map(lambda x: add(x, 10, 2), numbers)))

import os # used for example

files = ['a.txt', 'b.json']
my_join = bind(os.path.join, '.', 'data')
print(list(map(my_join, files))) # output ['./data/a.txt', './data/b.json']

bind_front function

bind_front pre-specifies function arguments like functools.partial. It takes a function and values, returns new function with values bound to front. When called with remaining args, values passed to bind_front are automatically inserted in front.

def add(a, b, c=0):
    return a + b + c


add_10 = bind_front(add, 10)
result = add_10(20, c=30)
print(result)  # 60

add_20_30 = bind_front(add, 20, 30)
result = add_20_30() # call add(20, 30)
print(result)  # 50

bind_back function

bind_back also pre-specifies function arguments but with values bound to end of arg list after all others. It takes a function and key-value pairs, returns new function with values bound to end. When called with remaining args, values passed to bind_back are automatically inserted at end.

add_30 = bind_back(add, c=30)
result = add_30(10, 20)
print(result)  # 60

add_40 = bind_back(add, 40)
result = add_40(10, 20) # call add(10, 20, 40)
print(result)  # 70

add_10 = bind_back(add, 10)
result = add_10(12)  # call add(10, 12), c=0 by  default
print(result)  # 22

Sequential binding

You can combine bind_front and bind_back to create a function that has arguments pre-specified at both the front and end of the argument list. For example, the code:

def func(p1, p2, p3, p4, p5):
    return " ".join(map(str, [p1, p2, p3, p4, p5]))
    
b_func = bind_front(bind_back(func, 4, 5), 1, 2)
print(b_func(3)) # 1 2 3 4 5

can be replaced with:

b_func_v2 = bind(func, 1, 2, _1, 4, 5) # using placeholder *_1*
print(b_func_v2(3)) # 1 2 3 4 5

---

We hope this information helps you effectively use the bind functions in your project. If you have any questions or feedback, please reach out to us. Happy coding!

---

Acknowledgements

We would like to express our sincere gratitude to all the individuals who have made this project a reality. Their contributions, guidance, and support have been invaluable. Thank you to everyone who has played a part in bringing this project to life.

• Daniil Dudkin
• ChatGPT