Source: http://www.liaoxuefeng.com/

♥ Function

• In python, name of a function could be assigned to a variable, for example:

>>> a = abs;

>>> a(-12)

12

• function definition:

def funtion_name(input_variable):

    function body

    return variables   # usually *return* signifies the end of the function,

                       # without which function will return *None*.  

# Define a null function

def nop()

    pass   # *pass* could be used as a placeholder, to ensure a smooth running

def

• import a self-defined function:

from name_of_function_definition_file (without .py) import function_name

• two functions:
  
  1 isinstance: check type of parameter;
  
  2 raise TypeError('...'): return TypeError with '...' as additional information.
• multiple returned values are actually one tuple;

Parameters

• Default parameters
  
  Note that defalut parameters must be unchangable objects.

# Set 'Westeros' and 'dead' as default parameters for 'loaction' and 'status' respectively

def game_of_thorn(name, gender, location = 'Westeros', status = 'dead'):

    print('name: ', name)

    print('gender: ', gender)

    print('location: ', location)

    print('status: ', status)

>>> game_of_thorn('Eddard Satrk', 'M')

name: Eddard Stark

gender: M

location: Westeros

status: dead

>>> game_of_thorn('Benjen Stark', 'M', 'the Wall')

name: Roose Bolton

gender: M

location: the North

status: dead

# You could ignore the order of input parameters as long as you assign

# the inputs directly to certain parameters

>>> game_of_thorn('Cersei Lannister', 'F', status = 'alive')

name: Cersei Lannister

gender: F

location: Westeros

status: alive

• Variable parameters

# An example

def calc(*numbers):

    sum = 0

    for n in numbers:   # variable parameters are treated as a tuple

        sum = sum + n * n

    return sum

>>> calc(1, 2)

5

>>> nums = [1, 2, 3]

>>> calc(*nums)   # list or tuple inputs could be sent in as variables parameters as well

14

• Keyword parameter
  
  Variable parameters are treated as a dict inside the function.

def person(name, age, **kw):   # kw as keyword parameter

    print('name:', name, 'age:', age, 'other:', kw)

>>> person('Michael', 30)

name: Michael age: 30 other: {}

>>> person('Adam', 45, gender='M', job='Engineer')

name: Adam age: 45 other: {'gender': 'M', 'job': 'Engineer'}

>>> extra = {'city': 'Beijing', 'job': 'Engineer'}

>>> person('Jack', 24, **extra)

name: Jack age: 24 other: {'city': 'Beijing', 'job': 'Engineer'}

• Named keyword parameter

def person(name, age, *, city, job):   # Only those using 'city' and 'job' as

                                       # keys will be accepted as keyword parameters

    print(name, age, city, job)

>>> person('Jack', 24, city='Beijing', job='Engineer')

Jack 24 Beijing Engineer

# if there is alrealy a variable parameters in the function, then '*,'

# is not needed for named keyword parameters

def person(name, age, *args, city, job):

    print(name, age, args, city, job)

Note

1 for named keyword parameters, parameter name is always needed when using the function;

2 named keyword parameter could be assigned a default value;

3 all the parameters could be used in one time, while the input order should be: regular parameter, default parameter, variable parameter (*argus), named keyword parameter and keyword parameter(**kw).

# Any fuction could be called using func(*args, **kw)

def f1(a, b, c=0, *args, **kw):

    print('a =', a, 'b =', b, 'c =', c, 'args =', args, 'kw =', kw)

def f2(a, b, c=0, *, d, **kw):

    print('a =', a, 'b =', b, 'c =', c, 'd =', d, 'kw =', kw)

>>> args = (1, 2, 3, 4)

>>> kw = {'d': 99, 'x': '#'}

>>> f1(*args, **kw)

a = 1 b = 2 c = 3 args = (4,) kw = {'d': 99, 'x': '#'}

>>> args = (1, 2, 3)

>>> kw = {'d': 88, 'x': '#'}

>>> f2(*args, **kw)

a = 1 b = 2 c = 3 d = 88 kw = {'x': '#'}

Recursive function

Recursive function is the one call itself. But a stack flow may appear with multiple recursion. One way to avoid the situation is : tail recursion.

A recursive function is tail recursive if the final result of the recursive call is the final result of the function itself (https://wiki.haskell.org/Tail_recursion). Tail recursion occupies constant RAM, thus could effectively avoid stack overflow during function calling.