Reputation: 271
I have three files G_P_map.txt, G_S_map.txt and S_P_map.txt. I have to combine these three files using awk. The example contents are the following -
(G_P_map.txt contains)
test21g|A-CZ|1mos
test21g|A-CZ|2mos
...
(G_S_map.txt contains)
nwtestn5|A-CZ
nwtestn6|A-CZ
...
(S_P_map.txt contains)
3mos|nwtestn5
4mos|nwtestn6
Expected Output :
1mos, 3mos
2mos, 4mos
Here is the code which I tried. I was able to combine the first two, but I couldn't do along with the third one.
awk -F"|" 'NR==FNR {file1[$1]=$1; next} {$2=file[$1]; print}' G_S_map.txt S_P_map.txt
Any ideas/help is much appreciated. Thanks in advance!
Upvotes: 2
Views: 902
Reputation: 360105
GNU AWK (gawk
) 4 has BEGINFILE
and ENDFILE
which would be perfect for this. However, the gawk
manual includes a function that will provide this functionality for most versions of AWK.
#!/usr/bin/awk
BEGIN {
FS = "|"
}
function beginfile(ignoreme) {
files++
}
function endfile(ignoreme) {
# endfile() would be defined here if we were using it
}
FILENAME != _oldfilename \
{
if (_oldfilename != "")
endfile(_oldfilename)
_oldfilename = FILENAME
beginfile(FILENAME)
}
END { endfile(FILENAME) }
files == 1 { # save all the key, value pairs from file 1
file1[$2] = $3
next
}
files == 2 { # save all the key, value pairs from file 2
file2[$1] = $2
next
}
files == 3 { # perform the lookup and output
print file1[file2[$2]], $1
}
# Place the regular END block here, if needed. It would be in addition to the one above (there can be more than one)
Call the script like this:
./scriptname G_P_map.txt G_S_map.txt S_P_map.txt
Upvotes: 2