Regular expression with [ or ( in python

I need to extract IP address in the form

prosseek.amer.corp.com [10.0.40.147]

or

prosseek.amer.corp.com (10.0.40.147)

with Python. How c开发者_如何转开发an I get the IP for either case with Python? I started with something like

site = "prosseek.amer.corp.com"
m = re.search("%s.*[\(\[](\d+\.\d+\.\d+\.\d+)" % site, r)

but it doesn't work.

ADDED

m = re.search("%s.+(\(|\[)(\d+\.\d+\.\d+\.\d+)" % site, r)
m.group(2)
m = re.search(r"%s.*[([](\d+\.\d+\.\d+\.\d+)" % site, r)
m.group(1)

seems to work.

You don't need to escape meta-characters (*, (, ), ., ...) in character groups (except ], unless it is the first character in the character group; [][]+ would match a sequence of square brackets.)

Another tip when it comes to Python is to use r'...'-style strings. With them, backslashes has no special meaning. r'\\' would print \\, since backslash has no special meaning:

m = re.search(r"%s.*[([](\d+\.\d+\.\d+\.\d+)" % site, r)

In the above string it doesn't make any difference though, since \d doesn't mean anything in Python, but when it comes to stuff like \r, \\, etc., it makes lives easier.

Use

[([]

The characters inside the outer brackets are taken literally. You do not need to escape them with a backslash.

For example:

import re
site = "prosseek.amer.corp.com "
m = re.search("%s\s*[([](\d+\.\d+\.\d+\.\d+)" % site, 'prosseek.amer.corp.com (10.0.40.147)')

I'd like to suggest a few slight refinements to what you have:

site = "prosseek.amer.corp.com"
m = re.search(r"%s\s+[([](\d+\.\d+\.\d+\.\d+)" % re.escape(site), r)
m.group(2)

The changes are:

• Pass site through re.escape so that it is interpreted literally; otherwise the dots in the domain name can match any character. This is extra important if site comes from user input; you don't want someone to be able to stick a regular expression in there and break your program.
• Use \s+ instead of .+ in between the site name and the IP address, so that it only accepts whitespace.

re.findall("(?:\d{1,3}\.){3}\d{1,3}", site)

How about you just ignore the brackets?

site = "prosseek.amer.corp.com"
m = re.search("%s.*(\d+\.\d+\.\d+\.\d+)" % site, r)

import string    
site='prosseek.amer.corp.com (10.0.40.147)'
''.join([c for c in site if c not in string.ascii_letters+' []()']).strip('.')

For some reason I like this better than regex