awkmd5sum

How to use `md5sum` to process `$1` of `awk`?

test file as example:

cat <<'EOF'> test
/boot/efi/EFI/debian/grubx64.efi root root 700 1625230722
/boot/efi/EFI/debian/mmx64.efi root root 700 1625230722
/boot/efi/EFI/debian/shimx64.efi root root 700 1625230722
/boot/grub/fonts/unicode.pf2 root root 644 1625230723
EOF

I plan to add one more column to test file,the value of new added column depends on the first column(md5sum --binary($1)),such as md5sum --binary /boot/efi/EFI/debian/grubx64.efi.

My script is awk '{ cmd = "md5sum --binary" close(cmd) $(NF+1) = (cmd($1)); print }' test > test_new but got error as below:

awk: line 1: syntax error at or near =

I am new to awk,what's the problem?

Solution

  • This awk should work:

    awk '{cmd = "md5sum --binary \047" $1 "\047"; 
if ((cmd | getline out) > 0) print $0, out; close(cmd)}' file.test

    However for task like this a bash loop should do job better as we are calling an external program using just the first column:

    while read -r col1 rest; do
   echo "$col1 $rest" "$(md5sum --binary "$col1")"
done < file.test