How to remove underscore character with awk

You can use this snippet:

$ awk '{ gsub("_", "", $NF); print "Country: " $NF }' report.txt
Country: PLUTO
Country: PINEAPPLE
Country: ORANGE
Country: RICE

Note that gsub() will perform the modification in place, so it will store the result of the substitution back to $NF, in your case.

If you're using GNU awk, you can use gensub() instead, which is slightly simpler:

$ gawk '{ print "Country: " gensub("_", "", "g", $NF) }' report.txt
Country: PLUTO
Country: PINEAPPLE
Country: ORANGE
Country: RICE

See GNU awk documentation for gsub() and gensub() for more details.

try

awk -F_ '{ print "Country: " $(NF-1) }' infile

You could try sed instead.

sed -r 's/[^_]*_([^_]*)_.*/Country: \1/' infile

• [^_]*_ matches everything until a first _ seen.
• ([^_]*)_ matches everything after above match untill next _ seen and .* matches everything after that, but only keep (...) part as a captured group.
• \1 is the back-reference to the ([^_]*) captured group.

Using sed instead:

$ sed -E 's/^This is (an? )?/Country: /; s/\<_//; s/_\>//' file
Country: PLUTO
Country: PINEAPPLE
Country: ORANGE
Country: RICE

This applies three substitutions:

1. Replaces the text This is optionally followed by either a or an with Country:.
2. Removes _ at the start of a word.
3. Removes _ at the end of a word.

The last two substitutions allows for data on the form

This is a _big_blue_ball_

which would be transformed into

Country: big_blue_ball

and not

Country: big blue ball

An awk alternative that just ignores the first part of each line and trims the first and last characters off of the last whitespace-delimited field:

awk '{ printf("Country: %s\n", substr($NF, 2, length($NF)-2)) }'

Done by using python

#!/usr/bin/python
import re
l=[]
k=open('file.txt','r')
for i in k:
        l.append(i)
m=re.compile(r'_.*')
for h in l:
        out=re.search(m,h)
        print "Country:",out.group().split('_')[-2]

output

Country: PLUTO
Country: PINEAPPLE
Country: ORANGE
Country: RICE