Python - 理解正则表达式

Question

So, I'm taking a list of usernames from a Linux server at school, this top code opens the directory where they are kept and saves it as information

#!/usr/bin/env python
import subprocess, sys

r = subprocess.Popen(['ls','/home/ADILSTU'], stdout=subprocess.PIPE)
information = r.stdout.read()
print information, str(information)

that works just fine and list the users like this... where it list them 1 per line. (there is atleast 100 usernames)

ajax2
jjape3
jaxe32    

my problem is, I want to create a "look-up" for these usernames, this is my code to search for usernames that only start with the letter j (so should only list jaxe32 from this list)

#lookup
import re
p = re.compile(r'j(?!j)\w*')
print p.match(str(information)).group()

but when I run this I get this error, and if I get rid of .group() it then just states "none", but no error. So i'm not sure if the list is getting saved to a string right, or if I'm just missing something obvious. I only want to use regular expression for this, not anything else.

    Traceback (most recent call last):
    File "getInformation.py", line 11, in <module>
    print p.match(str(information)).group()
    AttributeError: 'NoneType' object has no attribute 'group'

Answer 1

From the documentation on re.match :

If zero or more characters at the beginning of string match the regular expression pattern, return a corresponding match object. Return None if the string does not match the pattern;

re.match is only useful if the match starts from the beginning of the string, it does not find all matches in a string.

This leaves you with two main options:

• Split the input file by line and use re.match

• Use multiline matching and re.findall

Option 1 :

r = subprocess.Popen(['ls', '/home/administrator/sotest'], stdout=subprocess.PIPE)
information = r.stdout.read().decode('utf-8').split('\n') # ['ajax2', 'jaxe32', 'jjape3', '']

for user in information:
    s = re.match(r'j(?!j)\w*', user)
    if s:
        print(s.group())

Output:

jaxe32

Option 2 (using (?m)^j(?!j)\\w*$ ):

r = subprocess.Popen(['ls', '/home/administrator/sotest'], stdout=subprocess.PIPE)
information = r.stdout.read().decode('utf-8') # 'ajax2\njaxe32\njjape3\n'

print(re.findall(r'(?m)^j(?!j)\w*$', information))

Output:

['jaxe32']

Answer 2

The problem is that when the match method doesn't match anything, it doesn't return an empty match object, on which you could call the group method, it returns None . Which does not have the group method. Just check for None before you call any methods.

#lookup
import re
p = re.compile(r'j(?!j)\w*')
result = p.match(str(information))
if result:
    print result.group()