Shulou(Shulou.com)06/02 Report--

This article mainly introduces "how to understand Python monkey patch". In daily operation, I believe many people have doubts about how to understand Python monkey patch. The editor consulted all kinds of materials and sorted out simple and easy-to-use operation methods. I hope it will be helpful for you to answer the doubts about "how to understand Python monkey patch"! Next, please follow the editor to study!

Topic: talk about your understanding of "Monkey Patch" (monkey patching).

"Monkey patch" is a feature of dynamically typed language. When the code runs, it changes the methods, properties, functions, etc., in the code without modifying the source code to achieve the effect of hot patch. Many system security patches are also implemented through monkey patches, but the actual development should avoid the use of monkey patches, so as not to cause inconsistent code behavior.

When using the gevent library, we will execute gevent.monkey.patch_all () at the beginning of the code. The purpose of this line of code is to replace the socket module in the standard library, so that when we use socket, we can achieve the coordination of the code without modifying any code to improve performance. This is the application of the monkey patch.

In addition, if you want to replace the json in the standard library with the ujson tripartite library, you can also use the monkey patch, as shown below.

Import json, ujson

Json.__name__ = 'ujson'

Json.dumps = ujson.dumps

Json.loads = ujson.loads

The Mock technology in the unit test is also the application of the monkey patch. The unittest.mock module in Python is the module to solve the problem of using the Mock object to replace the dependent object in the unit test.

Topic 32: read the following code to say the result of the run.

Class A:

Def who (self):

Print ('Aids, end='')

Class B (A):

Def who (self):

Super (B, self). Who ()

Print ('Bamboo, end='')

Class C (A):

Def who (self):

Super (C, self). Who ()

Print ('Che, end='')

Class D (B, C):

Def who (self):

Super (D, self). Who ()

Print ('Downs, end='')

Item = D ()

Item.who ()

Comments: two knowledge points have been examined in this question. Knowledge point 1: MRO (method parsing order) in Python. In the absence of multiple inheritance, send a message to the object, and if the object does not have a corresponding method, the order of the search up (parent class) is very clear. If no corresponding method is found going up to the object class (the parent of all classes), an AttributeError exception will be thrown. But when there is multiple inheritance, especially when there is diamond inheritance (diamond inheritance), it is necessary to determine the MRO if you go back up and find out which method you should find. Classes in Python 3 and new classes in Python 2 use the C3 algorithm to determine MRO, which is similar to breadth-first search, while legacy classes (classical classes) in Python 2 use depth-first search to determine MRO. If you don't know MRO, you can use the mro method of the class or the _ _ mro__ property to get the MRO list of the class. Knowledge point 2: the use of the super () function. When using the super function, you can use super (type, object) to specify which object to search up for parent methods with which class as the starting point. So the super (B, self). Who () in the above class B code means to search up the who method of self (class D object) starting with class B, so you will find the who method in class C, because the MRO list of class D objects is D-> B-> C-> A-> object.

ACBD

Topic 33: write a function to evaluate inverse Polish expressions. You can't use Python's built-in functions.

Comments: the inverse Polish expression is also known as the "suffix expression". Compared with the "suffix expression" we usually use, the inverse Polish expression does not need parentheses to determine the priority of the operation. For example, the inverse Polish expression corresponding to 5 * (2 + 3) is 523 + *. The evaluation of the inverse Polish expression needs the help of the stack structure. The scan expression enters the stack when it encounters the Operand, and the two elements of the stack are calculated when the operator is encountered, and the operation result is put into the stack. At the end of the expression scan, there is only one number in the stack, which is the final result of the operation, which can be directly out of the stack.

Import operatorclass Stack:

"" FILO (Stack) ""

Def _ init__ (self):

Self.elems = []

Def push (self, elem):

"" Stack ""

Self.elems.append (elem)

Def pop (self):

"" out of the stack ""

Return self.elems.pop ()

@ property

Def is_empty (self):

"" check whether the stack is empty ""

Return len (self.elems) = = 0def eval_suffix (expr):

"" inverse Polish expression evaluation ""

Operators = {

'+': operator.add

'-': operator.sub

'*': operator.mul

'/': operator.truediv

}

Stack = Stack ()

For item in expr.split ():

If item.isdigit ():

Stack.push (float (item))

Else:

Num2 = stack.pop ()

Num1 = stack.pop ()

Stack.push (operators [item] (num1, num2))

Return stack.pop ()

How to implement string replacement operation in 34:Python?

There are roughly two methods to implement string substitution in Python: the replace method of a string and the sub method of a regular expression.

Method 1: use the replace method of the string.

Message = 'hello, worldview'

Print (message.replace ('he',' O'). Replace ('he','). Replace ('he',' HE'))

Method 2: use the sub method of regular expressions.

Import re

Message = 'hello, worldview'

Pattern = re.compile ('[aeiou]')

Print (pattern.sub ('#', message))

Extension: there is also an interview question in which a series of file names are stored in the list, such as filenames = ['a9.txtnames,' a12.txtnames, 'a8.txtnames,' b2.txtnames, 'b19.txtnames], which are sorted by literal table and numeric size, with a9.txt in front of a12.txt and b2.txt in front of b19.txt. You can think about how to solve this problem.

Topic 35: how to analyze the execution performance of Python code?

Profiling code performance can use the cProfile and pstats modules in the Python standard library, and the run function of cProfile can execute code and collect statistics, create Stats objects, and print simple profiling reports. Stats is a class in the pstats module, which is a statistical object. Of course, you can also use the tripartite tools line_profiler and memory_profiler to analyze the time and memory consumed by each line of code, both of which output the profiling structure in a very friendly way. If you use PyCharm, you can use the "Profile" menu item of the "Run" menu to analyze the performance of the code, and PyCharm can display the results of the performance analysis in the form of Statistics or Call Graph.

The following is an example of parsing code performance using cProfile.

Example.py

Import cProfiledef is_prime (num):

For factor in range (2, int (num * * 0.5) + 1):

If num% factor = = 0:

Return False

Return Trueclass PrimeIter:

Def _ _ init__ (self, total):

Self.counter = 0

Self.current = 1

Self.total = total

Def _ iter__ (self):

Return self

Def _ next__ (self):

If self.counter < self.total:

Self.current + = 1

While not is_prime (self.current):

Self.current + = 1

Self.counter + = 1

Return self.current

Raise StopIteration () cProfile.run ('list (PrimeIter (10000))')

If you use the line_profiler tripartite tool, you can directly analyze the performance of each line of code of the is_prime function, and you need to add a profiler decorator to the is_prime function, as shown below.

@ profiler

Def is_prime (num):

For factor in range (2, int (num * * 0.5) + 1):

If num% factor = = 0:

Return False

Return True

Install line_profiler.

Pip install line_profiler

Use line_profiler.

Kernprof-lv example.py

The result of the run is as follows.

Line # Hits Time Per Hit% Time Line Contents

=

1 @ profile

2 def is_prime (num):

3 86624 48420.0 0.6 50.5 for factor in range (2, int (num * * 0.5) + 1):

4 85624 44000.0 0.5 45.9 if num% factor = 0:

5 6918 3080.0 0.4 3.2 return False

6 1000 430.0 0.4 return True, this is the end of the study on "how to understand Python Monkey Patch". I hope I can solve your doubts. The collocation of theory and practice can better help you learn, go and try it! If you want to continue to learn more related knowledge, please continue to follow the website, the editor will continue to work hard to bring you more practical articles!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.