• /
• ~
• ls
• whoami
• gpg
• ?
• blog
• rss

"Grep" in Python

I use Python as a command line tool. I wanted to be able to do something similar to dir() | grep cmd to find out what packages had been loaded and variables created. I got sick of searching the net and finding nothing, so here's a simple script that does effectively the same thing (with slightly more tedious syntax)

#py grep command
#sample command: grep("^x",dir())
#syntax: grep(regexp_string,list_of_strings_to_search)

import re

def grep(string,list):
    expr = re.compile(string)
    for text in list:
        match = expr.search(text)
        if match != None:
            print match.string

# Two alternate solutions, courtesy of Gene H
# These are more elegant, and more detailed descriptions can be found here:
# http://www.faqs.org/docs/diveintopython/apihelper_filter.html
# http://www.faqs.org/docs/diveintopython/regression_filter.html

def grep(string,list):
    expr = re.compile(string)
    return [elem for elem in list if expr.match(elem)]

def grep(string,list):
    expr = re.compile(string)
    return filter(expr.search,list)

# Note a subtle difference between the above two:
# because the first uses expr.match(), it will only return exact matches
# while the latter will return a list containing strings that contain the
# search string in any position
# e.g.:
# list = ['normalize','size','nonzero','zenith']
# grep('ze',list)
# First one returns: ['zenith']
# Second one returns: ['normalize','size','nonzero','zenith']

If you know of a better (i.e. cleverer) way to accomplish this same feat, please let me know! Two of the solutions above were provided by a visitor, and I think they are a better solution to the problem presented here. Personally, I haven't ended up using the 'grep' function as much as I expected; instead I tend to get by on using the Python debugger (pdb) and working on a smaller globals()/locals() list and using who/whos at the ipython command line.

If you are interested in using python on the unix command line in place of grep, I recommend `the second hit on google <http://www.zazzybob.com/bin/grep.py.html>'_

# tags