[SOLVED] grep pattern grep -Ff

phpshell · 11-24-2013, 12:26 PM

file1

Code:

hello  333
hi  333
foo  111
ok  333
egypt 11
nas  333
nop 444

file2

Code:

hi
foo
nassw
local

i am using to find matching but i need to show the remain words from file2

grep -Ff file2 file1

output should be

Code:

nassw
local

any idea

colucix · 11-24-2013, 12:42 PM

I don't think it's possible to show the not matched patterns using grep. What about an awk solution like this?

Code:

awk 'FNR == NR{p[$1]++} FNR < NR{if ($1 in p) delete p[$1]} END{for (i in p) print i}' file2 file1

druuna · 11-24-2013, 12:45 PM

colucix was faster with the awk solution, here's a less elegant one:

Code:

comm -13 <(cut -f1 -d" " file1 | sort) <(sort file2)
local
nassw

sycamorex · 11-24-2013, 12:50 PM

See if that works for you.

Code:

#!/bin/bash

while read p; do
        if ! grep -q $p file1; then
               echo $p
        fi
done < file2

Edit: Ok, some people are just muuuch quicker than others

ntubski · 11-24-2013, 12:52 PM

Code:

cut -f1 -d' ' file1 | grep -Fwvf - file2

grail · 11-24-2013, 10:21 PM

As we know it is field 1 in File1, I would switch the order for the awk to be a little shorter:

Code:

awk 'FNR==NR{a[$1]}!($1 in a)' File1 File2

phpshell · 11-24-2013, 11:31 PM

many thanks to you all