Functions are similar to objects in Python  

  • An object has a class name and an address: 

    class Person:
    def __init__(self, name, age):
        self.name = name
        self.age = age

p1 = Person("John", 36)
print(p1)
<__main__.Person object at 0x7f093be7d850>

  • A function has the class function and an address: 

    # This is how to "create" a function "object":
def f1():
    print("Called f1")

print(f1)
<function f1 at 0x7f093c0ba2a0>

DEMO: progs/func-as-obj.py

  Function as a parameter  

  • Functions can have functions as parameters: 

    def f1():
    print("Called f1")

def f2(f):     # f is a function parameter
    f()

f2( f1 )

Called f1

  • Functions in Python are in fact like an object:

    • Functions can be stored in a variable

    • Functions can be passed as a parameter

    • Functions can also be returned by a function !!!

DEMO: progs/func-as-param.py

  Wrapper functions: returning a function  

  • Example of wrapper function: 

    def w(func):
    def wrapper():           # This function is LOCAL to w( )
        print("Started")
        func()
        print("Ended")

    return wrapper           # Returns a function !


def f1():
    print("Hello")

f2 = w(f1)      # Pass f1 to w() and receive a NEW function f2
f2()            # Call the new function f2()

Started
Hello
Ended

DEMO: progs/wrapper.py

  The function name is actually a variable  

  • We can assign a function to a function name: 

    def w(func):
    def wrapper():           # This function is LOCAL to w( )
        print("Started")
        func()
        print("Ended")

    return wrapper           # Returns a function !


def f1():
    print("Hello")

f1()   Hello

# f1 is a VARIABLE and can be assigned....

f1 = w(f1)      # Pass f1 to w() and receive a NEW function
f1()   Started Hello Ended

f1 = w(f1)      # Pass f1 to w() and receive a NEW function
f1()   Started Started Hello Ended Ended

DEMO: progs/wrapper2.py

  Decorators  

  • Decorator notation: 

    def w(func):
    def wrapper():           # This function is LOCAL to w( )
        print("Started")
        func()
        print("Ended")

    return wrapper           # Returns a function !

# Instead of writing:
def f1():
    print("Hello")

f1 = w(f1)      # Pass f1 to w() and receive a NEW function f2

f1()     Started Hello Ended

# We can write: ("decorator notation")
@w
def f2():
    print("Hello")

f2()     Started Hello Ended

DEMO: progs/decorator.py

  Using * args and ** kwargs to create general decorators  

  • A function with a parameter will break the previous example: 

    def w(func):
    def wrapper():           # This function is LOCAL to w( )
        print("Started")
        func()               # *** Function has NO parameter !
        print("Ended")

    return wrapper           # Returns a function !

# Instead of writing:
def f1(s):
    print(s)

f1 = w(f1)      # Pass f1 to w() and receive a NEW function f2

f1("New") 
TypeError: w.wrapper() takes 0 positional arguments but 1 was given
# We can write: ("decorator notation")
@w
def f2(s):      # Function has 1 parameter
    print(s)

f2("New")

DEMO: progs/decorator2.py

  Using * args and ** kwargs to create general decorators  

  • Solution: capture the positional args with *args and the keyword args with **kwargs: 

    def w(func):
    def wrapper(*args, **kwargs):          
        print("Started")
        func(*args, **kwargs) # *** Function has general parameters !
        print("Ended")

    return wrapper           # Returns a function !

# Instead of writing:
def f1(s):
    print(s)

f1 = w(f1)      # Pass f1 to w() and receive a NEW function f2

f1("New")     Started New Ended

# We can write: ("decorator notation")
@w
def f2(s):      # Function has 1 parameter
    print(s)

f2("New")     Started New Ended

DEMO: progs/decorator3.py