A list generator is an expression used to generate a list in a specific syntax form . yes Python Provides a concise form of generating lists , Can quickly generate a new list.
1、 Usage method
Common grammar :[exp for iter_var in iterable]
Syntax format with filter function : [exp for iter_var in iterable if_exp]
Syntax format with filter function : [exp for iter_var in iterable if_exp]
2、 Basic application
It can transform tedious methods 1 Simplify to method 2
Method 1
import string
import random
codes = []
for count in range (100): # loop 100 Time , Every time it is generated arbitrarily 4 A string
code = "".join(random.sample(string.ascii_letters,4))
codes.append(code)
print(codes)
# result
['MIFp', 'zJct', 'KpNB', 'afcS', 'zCnv', 'nzgH', 'IlhR', 'UHJk', ....
Method 2
codes = [ "".join(random.sample(string.ascii_letters,4)) for i in range(100)]
print(codes)
# result
['mnLM', 'LVKe', 'ntlM', 'BjYl', 'KIAt', 'gRBl', 'QiqH', ...
3、 Exercises
find 1 To 100 It can be 3 Divisible number
Method 1
nums = []
for num in range(1,100):
if num % 3 == 0:
nums.append(num)
print(nums)
Method 2
Using the formula generator
nums = [num for num in range(1,100) if num % 3 == 0]
print(nums)
The set generation formula is as follows
result = {
i**2 for i in range(10)}
print(result)
# result
{
0, 1, 64, 4, 36, 9, 16, 49, 81, 25}
The dictionary generation formula is as follows
result={
i:i**2 for i in range(10)}
print(result)
# result
{
0: 0, 1: 1, 2: 4, 3: 9, 4: 16, 5: 25, 6: 36, 7: 49, 8: 64, 9: 81}
stay python The mechanism of loop and compute in is called generator 
Use scenarios 
1、 Generator method
Method 1： The first way to implement ( Rewrite the generative expression into a generator ）
nums = (i**2 for i in range(1000))
print(nums)
# result
<generator object <genexpr> at 0x0000018FFF234900>
The above is not convenient to view , You can view the values in the following way
nums = (i**2 for i in range(1000))
for num in nums:
print(num)
# result
0
1
4
9
16
25
36
...
Method 2：yield keyword
return： Function encountered return Just go back to , The following functions will not execute .
yield: encounter yield Then stop executing the code , Call again next When the method is used , Since the last stop yiled Continue to the next yield
def login():
print('step1') # step
yield 1 #output
print('step')
yield 2
print('step 3')
yield 3
# If there is in the function yield keyword , The return value of the function is a generator
g = login()
print(next(g))
print(next(g))
import time
start_time=time.time() # from 1970.1.1 The number of seconds to now
time.sleep(2) # Sleep for two seconds
end_time = time.time() # from 1970.1.1 The number of seconds to now
print(end_time-start_time)
# result
2.00744891166687
1、 Functions are nested inside functions
2、 The return value of an external function is a reference to an internal function
3、 Internal functions can use variables of external functions
import time
def timeit(name):
def wrapper():
print('wrapper'+name)
print('timeit')
return wrapper
in_fun = timeit(name = 'westos') in_fun At this time, it is equivalent to wrapper function
in_fun() It is equivalent to calling internal and external functions at the same time
1、 Basic template
import time
from functools import wraps
def Name of decorator (f):
@wraps(f) Keep the help documentation for the decorator function
def wrapper(*args,**kwargs):
# Here you can add things before executing the function
result=f(*args,**kargs)
# What you do after executing the function
return result Return the function to the decorated function
return wrapper()
2、 give an example
# Decorator ： Functions used to decorate functions
# Without changing the source code , Tools to add additional functionality
# Using closures to implement decorators
import time
def timeit(f):
def wrapper(x,y):
start = time.time()
result= f(x,y) # Execution is add(x,y)
end = time.time()
print(' The function runs for ：%.4f'%(end-start))
return result
return wrapper
@timeit # Decorative sugar , The time here will be add Function to timeit(add())
def add(x,y):
return x+y
result = add(1,3)
print(result)
3、 add to wraps, Help documents of functions can be described separately
# Decorator ： Functions used to decorate functions
# Without changing the source code , Tools to add additional functionality
# Using closures to implement decorators
import time
from functools import wraps
def timeit(f):
"""" The decorator of the timer """
@wraps(f) # Keep the attribute information and help documents of the decorated function
def wrapper(*args,**kwargs):
start = time.time()
result= f() # Execution is login()
end = time.time()
print(' The function runs for ：%.4f'%(end-start))
return result
return wrapper
@timeit # Decorative sugar , The time here will be add Function to timeit(add())
def login():
""""login Help document for """
print('login')
print(help(timeit))
import time
from functools import wraps
def timeit(args='seconds'):
def desc(f):
"""" The decorator of the timer """
@wraps(f) Keep the attribute information and help documents of the decorated function
def wrapper(*args,**kwargs):
start = time.time()
result= f() Execution is login()
end = time.time()
if args =='seconds':
print(f" function {
f.__name__} The running time is {
end-start} second ")
elif args == 'mintues':
print(f" function {
f.__name__} The running time is {
(end - start)/60} second ")
return result
return wrapper
return desc
@timeit(args='mintues') To achieve timeit() It will @ perform desc function , namely desc=desc(login)
def login():
""""login Help document for """
print('login')
login()
Note that the decorator is called from bottom to top , Call the second decorator first is_permisson, Then call the first decorator is_login, But the end result is from top to bottom , Output the result of a decorator , Then output the result of the second decorator , Finally, the result of executing the function .
from functools import wraps
def is_login(f):
@wraps(f)
def wrapper1(*args,**kwargs):
print('is_login, Is the user logged in ')
result=f(*args,**kwargs)
return result
return wrapper1
def is_permisson(f):
@wraps(f)
def wrapper2(*args,**kwargs):
print('is_permission, Whether the user has permission ')
result=f(*args,**kwargs)
return result
return wrapper2
@is_login
@is_permisson Note that the decorator is called from bottom to top , First call is_permisson, Call again is_login
def show_hosts():
print(' Display all virtual machines ')
show_hosts()
The specific calling sequence is as follows
1、is_permisson The module calls first
First show_host=is_permisson(show_hosts)
Get the results show_host=wrapper2
2、is_login The module is called again
show_hosts = is_login(wrapper2) ( Because the last decorator made show_hosts by wrapper2, So the function passed in here is wrapper2)
Get the results show_hosts=wrapper1
3、 The above two steps are finished , Carry out the third step show_hosts function .
Now? show_hosts Function time wrapper1, So execute show_hosts The function actually executes wrapper1 function , Output is ：
’is_login, Is the user logged in ‘
And then execute f(*args,**kwargs), Follow the steps 2 It is found that the function passed in is wrapper2, So execute wrapper2 function , Output is ：
‘is_permission, Whether the user has permission ’
Finally, execute f(*args,**kwargs), Follow the steps 1 It is found that the function passed in is show_hosts, So execute show_hosts Self function , Output is ：
‘ Display all virtual machines ’
Processing in parallel at the same time
# Calculation x Of 2 Power
result = map(lambda x:x**2,[1,2,3,4])
print(list(result))
# result
[1, 4, 9, 16]
result=map(lambda x,y:x+y,[1,2,3],[4,5,6])
print(list(result))
# result
[5, 7, 9]
Take the result of the former as the input of the latter
# reduce function
# Calculation ((2**3)**3)=reduce result
from functools import reduce
result = reduce(lambda x, y: x**y, [2, 3, 3])
print(result)
practice ·： Calculation 1×2×...100
from functools import reduce
result = reduce(lambda x,y: x * y,list(range(1,100))) Be sure to convert to a list
print(result)
Filter function
Filter out even numbers
result=filter(lambda x:x % 2==0,[1,2,4,5])
print(list(result))
# result
# filter function
result=filter(lambda x:x % 2==0,[1,2,4,5])
print(list(result)) Output as list
Sort numbers by size , From small to large
result = sorted([1,4,321,2])
print(result)
result = sorted([1,4,321,2],reverse=True)
print(result)
# result
[321, 4, 2, 1]
According to whether the number is 0 Sort
The rule in the function is to set non - 01 Is endowed with 1, And then according to 0 ,1 Sort , Return the real value after sorting
result = sorted([1,4,0,2],key=lambda x:0 if x==0 else 1)
print(result)
[0, 1, 4, 2]
