Can someone explain the implementation of function as objects? It's quite confusing and I'm having a hard time understanding it.

Question

Function as Objects

Answer 1

def thing():
print("thing")

thing()

#def thing() is the function, and won't be executed until you put thing() as a object in your code

Answer 2

Like in mathematics, you use function quiet same way... giving data, you expect a corresponding result: f(x)=ax+b give the value of the corresponding y value for (x,y) dot on the line described by the expression.

Basically, it's what you do to in programming. With Python, equivalent function is defined by:

def get_y(x):
var a = 4
var b = 2
return a*x+b

... and it's used:

var point_y = get_y(10);

In programming, contrarely to matematics, you need often some block of code to be reuse many times, but not necessarly wich return a value... Sometimes we made a distinction between function who's provide return value or not by distinct name: the second one are called "procedure"... and also functions or procedures belonging to a class ( OOP ) are called indifferently "method". But usually they all are as well called "function" :P

Answer 3

>>> help("def") # just the interesting bits
Function definitions
********************
A function definition defines a...function object...[and] is an executable statement [. it] binds the function name...to a function object (a wrapper around the executable code for the function)...**Programmer's note:** Functions are first-class objects...
-----
Takeaway: Write "def demo():" and Python cleverly wraps demo's body in a class (function type):

>>> def demo(): print("You called?");
>>> print(type(demo))
<class 'function'>

It has an attributes directory.
>>> print(dir(demo))
['__annotations__', '__call__', '__class__', '__closure__', '__code__', '__defaults__', '__delattr__', '__dict__', '__dir__', '__doc__', '__eq__', '__format__', '__ge__', '__get__', '__getattribute__', '__globals__', '__gt__', '__hash__', '__init__', '__kwdefaults__', '__le__', '__lt__', '__module__', '__name__', '__ne__', '__new__', '__qualname__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__']

Look at all the things we can do with our function object.

>>> demo.__call__()
You called?

Look deeper? demo.__code__'s also a class (code type).
>>> cheat = demo.__code__
>>> type(cheat)
<class 'code'>

>>> dir(cheat)
[ ... more attributes like above ... 'co_code', 'co_consts', 'co_filename' ... ]

>>> print([cheat.co_filename, cheat.co_argcount, cheat.co_consts, cheat.co_names, cheat.co_nlocals, cheat.co_varnames])
['<stdin>', 0, (None, 'You called?'), ('print',), 0, ()]

'stdin' - yep, I'm in interactive>>> mode. Zero arguments, string constants tuple, names tuple (print function), no local variables, empty variable name tuple.

As a quick tour of Python objects that happen to relate to functions...it's not code...but I can if it's helpful.

Answer 4

Also...you might like a GUI to inspect Python objects:
https://github.com/titusjan/objbrowser

Request 'desktop site' on mobile, then scroll down to "User interface" for a sample. Install with pip.

Answer 5

Since you're passing the "add" object, you can mentally replace "func" with "add" (also: x=a and y=b...I didn't do that):

def do_twice(add, x, y):
return add(add(x,y), add(x,y))

Evaluate from innermost to outermost:
return add(x+y, x+y)

becomes:
return (x+y) + (x+y) # effectively 2*(a+b)

The reason it's done this way (I think) is so you could substitute other functions without much trouble (subtract, concatenate, something weirder..). I would expect this self-similarity in a fractal.

Answer 6

When python interpreter see reserved name-def, he doesn't run the code under def, just adding function name to namespace

Answer 7

I'm actually looking for an analogy that can help me understand with an example of a wrong & working code or may be a detailed explanation. I really appreciate your quick response and please pardon my ignorance.

Answer 8

Thank you all for the help👍! Can someone explain this code to me bit-by-bit?

def add(x, y):
return x + y

def do_twice(func, x, y):
return func(func(x, y), func(x, y))

a = 5
b = 10

print(do_twice(add, a, b))

Basically, line 3 & 4 of the code. Thank you in advance!

Answer 9

@Kirk: Spot on! Thanks a lot for the explanation 👍