Two practical cold tips on Python

I. I remember when I first started my work, the boss gave us the C ++ Basic Course and told us that the string literal volume can be wrapped (the following code). It felt like a dream.

 
 

  
  
#include <stdio.h>  
  
  
 
  
  
int main(int argc, char** argv)  
  
  
{  
  
  
        char* w = "hello" 
  
  
                " " 
  
  
                "world." 
  
  
                ;  
  
  
        printf("%s", w);  
  
  
        return 0;  
  
  
}  
 
 

Output:

 
 

  
  
hello world. 
 
 

Later, after writing Python for a long time, I realized that Python is actually okay:

 
 

  
  
>>> t = ('hello' 
  
  
... ' ' 
  
  
... 'world')  
  
  
>>> t  
  
  
'hello world' 
 
 

This feature is very useful and allows you to elegantly classify long code into several lines. I remember that when I used to spell SQL language and write log entries, I always felt tangled for the length of the code line exceeding 78 (see our programming specification: http://blog.csdn.net/lanphaday/article/details/6601123), and now there is no pressure.

Ii. When writing the test code of absolute32 (see: http://blog.csdn.net/lanphaday/article/details/6762023), to make the test code compatible with Python2.x/3.x, introduce the ugly code:

 
 

  
  
If sys. version <'3 ':
  
  
Exec ("chinese = unicode ('rayong hao', 'utf-8 ')")
  
  
Else:
  
  
Exec ("chinese = 'rayong hao '")
 
 

This is because in Python2.x

 
 

  
  
Chinese = 'rayong hao'
 
 

The encoding is not unicode, but the prefix u of the string literal is canceled in Python3.x, so

 
 

  
  
Chinese = u'rayong hao'
 
 

And directly syntax errors. At that time, I had to write exec code for compilation based on different versions. Later I learned that unicode_literals was introduced in Python2.6, which can easily write 2.x/ 3.x compatible code:

 
 

  
  
>>> X = 'China'
  
  
>>> X
  
  
'\ Xe4 \ xb8 \ xad \ xe5 \ x9b \ xbd'
  
  
>>> From _ future _ import unicode_literals
  
  
>>> Y = 'China'
  
  
>>> Y
  
  
U' \ u4e2d \ u56fd'
 
 

In this way, my ugly code can also be beautified!

Original article: http://blog.csdn.net/lanphaday/article/details/6793567