How to filter the lines of two files and save the difference into a third file

Question

radiumc 0 Newbie Poster

12 Years Ago

Hello,

I am very new to python. I have two files containing multiple lines of files name, and I need to filter out if file name present on file1 than remove it from file2 and save the file2 difference into a third file.

i.e.

file1:
/users/ux454500/radpres.tar
/paci/ucb/ux453039/source/amr.12.20.2002.tar~
/paci/ucb/ux453039/source/amr.1.25.2003.htar.idx
/paci/ucb/ux453039/source/amr.1.18.2003.htar.idx

file2:
DIRECTORY /paci/ucb/ux453039/.
FILE /paci/ucb/ux453039/.ktb_ux453039 05/31/2006 17:52:04 09/01/2007 14:25:33
FILE /paci/ucb/ux453039/source/amr.12.20.2002.tar~ 03/10/2005 20:56:50 09/02/2007 10:35:41
FILE /paci/ucb/ux453039/source/amr.1.25.2003.htar.idx 02/23/2007 14:20:15 08/27/2007 14:53:48
FILE /paci/ucb/ux453039/bhsf1.0000.out 02/23/2007 14:20:13 08/27/2007 14:53:48
FILE /users/ux454500/AIX.mpCC.DEBUG.ex 02/23/2007 14:20:13 02/28/2007 14:47:55
DIRECTORY /paci/ucb/ux453039/runs/bondi

file3:
DIRECTORY /paci/ucb/ux453039/.
FILE /paci/ucb/ux453039/.ktb_ux453039 05/31/2006 17:52:04 09/01/2007 14:25:33
FILE /paci/ucb/ux453039/bhsf1.0000.out 02/23/2007 14:20:13 08/27/2007 14:53:48
FILE /users/ux454500/AIX.mpCC.DEBUG.ex 02/23/2007 14:20:13 02/28/2007 14:47:55
DIRECTORY /paci/ucb/ux453039/runs/bondi

Kindly help.

With my best regards,

python

4 Contributors
8 Replies
144 Views
1 Day Discussion Span
Latest Post 12 Years Ago Latest Post by radiumc

TrustyTony 888 pyMod

12 Years Ago

We are glad to check your code, either push the (CODE) button right before doing paste or select the code you pasted and push the (CODE) button after.

Edited 11 Years Ago by mike_2000_17 because: Fixed formatting

JoshuaBurleson 23 Posting Whiz

12 Years Ago

file1:
/users/ux454500/radpres.tar
/paci/ucb/ux453039/source/amr.12.20.2002.tar~
/paci/ucb/ux453039/source/amr.1.25.2003.htar.idx
/paci/ucb/ux453039/source/amr.1.18.2003.htar.idx

file2:
DIRECTORY /paci/ucb/ux453039/.
FILE /paci/ucb/ux453039/.ktb_ux453039 05/31/2006 17:52:04 09/01/2007 14:25:33
FILE /paci/ucb/ux453039/source/amr.12.20.2002.tar~ 03/10/2005 20:56:50 09/02/2007 10:35:41
FILE /paci/ucb/ux453039/source/amr.1.25.2003.htar.idx 02/23/2007 14:20:15 08/27/2007 14:53:48
FILE /paci/ucb/ux453039/bhsf1.0000.out 02/23/2007 14:20:13 08/27/2007 14:53:48
FILE /users/ux454500/AIX.mpCC.DEBUG.ex 02/23/2007 14:20:13 02/28/2007 14:47:55
DIRECTORY /paci/ucb/ux453039/runs/bondi

file3:
DIRECTORY /paci/ucb/ux453039/.
FILE /paci/ucb/ux453039/.ktb_ux453039 05/31/2006 17:52:04 09/01/2007 14:25:33
FILE /paci/ucb/ux453039/bhsf1.0000.out 02/23/2007 14:20:13 08/27/2007 14:53:48
FILE /users/ux454500/AIX.mpCC.DEBUG.ex 02/23/2007 14:20:13 02/28/2007 14:47:55
DIRECTORY /paci/ucb/ux453039/runs/bondi

and your effort on the code? Please read the forum rules, or for a simplified version; pytony's signature.

Reply to this topic

Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.

radiumc 0 Newbie Poster · Answer 1 · 2011-10-14T03:17:30+00:00

file1:
/users/ux454500/radpres.tar
/paci/ucb/ux453039/source/amr.12.20.2002.tar~
/paci/ucb/ux453039/source/amr.1.25.2003.htar.idx
/paci/ucb/ux453039/source/amr.1.18.2003.htar.idx

file2:
DIRECTORY /paci/ucb/ux453039/.
FILE /paci/ucb/ux453039/.ktb_ux453039 05/31/2006 17:52:04 09/01/2007 14:25:33
FILE /paci/ucb/ux453039/source/amr.12.20.2002.tar~ 03/10/2005 20:56:50 09/02/2007 10:35:41
FILE /paci/ucb/ux453039/source/amr.1.25.2003.htar.idx 02/23/2007 14:20:15 08/27/2007 14:53:48
FILE /paci/ucb/ux453039/bhsf1.0000.out 02/23/2007 14:20:13 08/27/2007 14:53:48
FILE /users/ux454500/AIX.mpCC.DEBUG.ex 02/23/2007 14:20:13 02/28/2007 14:47:55
DIRECTORY /paci/ucb/ux453039/runs/bondi

file3:
DIRECTORY /paci/ucb/ux453039/.
FILE /paci/ucb/ux453039/.ktb_ux453039 05/31/2006 17:52:04 09/01/2007 14:25:33
FILE /paci/ucb/ux453039/bhsf1.0000.out 02/23/2007 14:20:13 08/27/2007 14:53:48
FILE /users/ux454500/AIX.mpCC.DEBUG.ex 02/23/2007 14:20:13 02/28/2007 14:47:55
DIRECTORY /paci/ucb/ux453039/runs/bondi

radiumc 0 Newbie Poster · Answer 2 · 2011-10-14T07:06:11+00:00

I am sorry. This is my first time login and post the question. Attached is the code that I copy from someone had posted before and try to modify and use it.

Thank you very much for your help.

#!/usr/local/bin/python -u
# Filename: file_comp.py

file1=['/Users/file_comp/1.md']
file2=['/Users/file_comp/2.out']
file3=['/Users/file_comp/3.comp']


def key(line):
    return tuple(line.strip().split()[0:2])
 
def make_key_set(file_path):
    return set(key(line) for line in open(file_path))
 
 
def filtered_lines(file_path1, file_path2):
    key_set = make_key_set(file_path2)
    return (line for line in open(file_path1) if key(line) in key_set)
 
if __name__ == "__main__":
    file3 = open("file3", "w")
    for line in filtered_lines("file1", "file2"):
        file3.write(line)
    file3.close()

snippsat 661 Master Poster · Answer 3 · 2011-10-14T19:05:59+00:00

import os

with open('file_1.txt') as file_1, open('file_2.txt') as file_2:
    f1 = [i.strip() for i in file_1]
    f2 = [i.strip() for i in file_2]
    comp =  [f for f in f2 if f not in f1]
    for f in comp:
        #print f
        #print os.path.basename(f)
        with open('new.txt', 'a') as f_out:
            f_out.write(f + '\n')

So in file_1 1.txt .
In file file_2 1.txt 2.txt 3.txt .
Saved to new.txt 2.txt 3.txt .

This will not work for you because your filename is strange. .ktb_ux453039 . amr.12.20.2002.tar~ In a normal case os.path.basename() would extract filename.

>>> import os
>>> os.path.basename('/root/desktop/python.py')
'python.py'
>>> os.path.basename('FILE /paci/ucb/ux453039/.ktb_ux453039 05/31/2006 17:52:04 09/01/2007 14:25:33')
'2007 14:25:33'

If your filename look like this i think write a regex is the only option,to extract those strange looking filename.

TrustyTony 888 pyMod Team Colleague Featured Poster · Answer 4 · 2011-10-14T21:05:44+00:00

import os

with open('file_1.txt') as file_1, open('file_2.txt') as file_2:
    f1 = [i.strip() for i in file_1]
    f2 = [i.strip() for i in file_2]
    comp =  [f for f in f2 if f not in f1]
    for f in comp:
        #print f
        #print os.path.basename(f)
        with open('new.txt', 'a') as f_out:
            f_out.write(f + '\n')
So in file_1 1.txt .
In file file_2 1.txt 2.txt 3.txt .
Saved to new.txt 2.txt 3.txt .
This will not work for you because your filename is strange. .ktb_ux453039 . amr.12.20.2002.tar~ In a normal case os.path.basename() would extract filename.
>>> import os
>>> os.path.basename('/root/desktop/python.py')
'python.py'
>>> os.path.basename('FILE /paci/ucb/ux453039/.ktb_ux453039 05/31/2006 17:52:04 09/01/2007 14:25:33')
'2007 14:25:33'
If your filename look like this i think write a regex is the only option,to extract those strange looking filename.

# Who need regexp when we have Python ;) (... mostly)
import os

name = 'FILE /paci/ucb/ux453039/.ktb_ux453039 05/31/2006 17:52:04 09/01/2007 14:25:33'
name = name.split()[1]
print name
print os.path.basename(name)

radiumc 0 Newbie Poster · Answer 5 · 2011-10-15T02:56:19+00:00

radiumc 0 Newbie Poster

12 Years Ago

Excellent.
Thank you very much gentlemen.

radiumc 0 Newbie Poster · Answer 6 · 2011-10-15T05:55:11+00:00

Ok, here is my final code for your information. Again, thank you for your time and help.

#!/usr/local/bin/python -u
# Filename: file_comp.py
 
file_path1='/Users/file_comp/1.md'
file_path2='/Users/file_comp/2.out'
file_path3='/Users/file_comp/3.comp'
file_path3='/Users/file_comp/4.missing'

import os

def key_file2(line):
    t = tuple(line.strip().split()[1:2])
    #print "key_file2: ", t
    return t

def key_file1(line):
    t = tuple(line.strip().split()[0:1])
    # print "key_file1: ", t
    return t

def make_key_set(file_path):
    return set(key_file1(line) for line in open(file_path))

def filtered_lines(file_path1, file_path2):
    key_set = make_key_set(file_path1)

    file3 = open(file_path3, "w")
    for line in open(file_path2):
        if line.split()[0:1][0] == 'DIRECTORY':
            file3.write(line)
        elif key_file2(line) not in key_set:
            file3.write(line)
        else:
            key_set.remove(key_file2(line))

    file3.close()

    file4 = open(file_path4, "w")
    for key in key_set:
        file4.write(key[0] + "\n")

if __name__ == "__main__":
    filtered_lines(file_path1, file_path2)