Python Functions - AskPython

A function is a block of code with a name.
We can call a function by its name.
The code inside a function only runs when it’s called.
A function can accept data from the caller program, it’s called as function parameters.
The function parameters are inside parentheses and separated by a comma. A function can accept any number of arguments.
A function can return data to the caller program. Unlike other popular programming languages, Python functions definition doesn’t specify the return type.
We can’t use reserved keywords as the function name. A function name must follow the Python identifiers definition rules.

How to Define a Function in Python?

We can define a function in Python using def keyword. Let’s look at a couple examples of a function in Python.

def hello():
    print('Hello World')


def add(x, y):
    print(f'arguments are {x} and {y}')
    return x + y

Based on above examples, we can define a function structure as this.

def function_name(arguments):
    # code statements

How to Call a Function in Python?

We can call a function by its name. If the function accept parameters, we have to pass them while calling the function.

hello()
sum = add(10, 5)
print(f'sum is {sum}')

We are calling hello() and add() functions that are defined by us. We are also calling print() function that is one of the built-in function in Python.

Python Function Types

There are two types of functions in Python.

built-in functions: The functions provided by the Python language such as print(), len(), str(), etc.
user-defined functions: The functions defined by us in a Python program.

Can a Function have default parameter value?

Python allows default values for the function parameters. If the caller doesn’t pass the parameter then the default value is used.

def hello(year=2019):
    print(f'Hello World {year}')


hello(2020)  # function parameter is passed
hello()  # function parameter is not passed, so default value will be used

Output:

Hello World 2020
Hello World 2019

Can we have multiple return statements inside a Function?

Yes, a function can have multiple return statements. However, when one of the return statement is reached, the function execution terminates and the value is returned to the caller.

def odd_even_checker(i):
    if i % 2 == 0:
        return 'even'
    else:
        return 'odd'


print(odd_even_checker(20))
print(odd_even_checker(15))

Can a Python Function Return Multiple Values one by one?

Python function can return multiple values one by one. It’s implemented using the yield keyword. It’s useful when you want a function to return a large number of values and process them. We can split the returned values into multiple chunks using yield statement. This type of function is also called a generator function.

def return_odd_ints(i):
    x = 1
    while x <= i:
        yield x
        x += 2


output = return_odd_ints(10)
for out in output:
    print(out)

Output:

Python Function Variable Arguments

Python allows three type of parameters in the function definition.

Formal arguments: the ones we have seen in the examples so far.
Variable Number of non-keyword arguments: for example, def add(*args)
Variable Number of keyword arguments or named arguments: for example, def add(**kwargs)

Some important points about variable arguments in Python are:

The arguments order should be formal arguments, *args, and **kwargs.
It’s not mandatory to use variable parameter names as args and kwargs. However, it’s the best practice to use them for better code readability.
The args type is tuple. So we can pass a tuple to be mapped with *args variable.
The type of kwargs is dict. So we can pass a dictionary to be mapped with **kwargs variable.

Here is a simple example of using variable arguments in a function.

def add(x, y, *args, **kwargs):
    sum = x + y
    for a in args:
        sum += a

    for k, v in kwargs.items():
        sum += v
    return sum


total = add(1, 2, *(3, 4), **{"k1": 5, "k2": 6})
print(total)  # 21

Python Recursive Function

When a function calls itself, it’s called a recursive function. This scenario is called recursion in programming.

You should be very careful when using recursion because there might be a chance that the function never terminates and goes into infinite loop. Here is a simple example to print fibonacci series using recursion.

def fibonacci_numbers_at_index(count):
    if count <= 1:
        return count
    else:
        return fibonacci_numbers_at_index(count - 1) + fibonacci_numbers_at_index(count - 2)


count = 5
i = 1
while i <= count:
    print(fibonacci_numbers_at_index(i))
    i += 1

It’s good to know about recursion but most of the times you don’t need it in programming. You can perform the same thing using for-loop or while-loop.

Data Type of Python Function

Python functions are instances of ‘function’ class. We can check this using type() function.

def foo():
    pass


print(type(foo))

Output: <class ‘function’>

Python Function vs Method

Python function is part of the python script file in which it is defined whereas Methods are defined inside a class definition.
We can call a function directly if it’s in the same module. If the function is defined in a different module, we can import the module and then call the function directly. We need a class or an object of the class to call the methods.
Python function can access all the global variables whereas Python class methods can access global variables as well as class attributes and functions.
Python functions data type is ‘function’ whereas Python methods data type is ‘method’.

Let’s look at a simple example of functions and methods in Python.

class Data:
    def foo(self):
        print('foo method')


def foo():
    print('foo function')


# calling a function
foo()

# calling a method
d = Data()
d.foo()

# checking data types
print(type(foo))
print(type(d.foo))

Output:

foo function
foo method
<class 'function'>
<class 'method'>

Advantages of Python Functions

Code reusability because we can call the same function multiple times
Modular code since we can define different functions for different tasks
Improves maintainability of the code
Abstraction as the caller doesn’t need to know the function implementation

Python Anonymous Function

Anonymous functions doesn’t have a name. We can define anonymous function in Python using lambda keyword.

def square(x):
    return x * x


f_square = lambda x: x * x

print(square(10))  # 100
print(f_square(10))  # 100

Conclusion

Functions are important part of a programming language. Python function are defined using def keyword. We can have variable number of arguments in a Python function. Python also supports anonymous function. They can return a single value or yield a number of values one by one.