I need help decifering this short script. There's no big thing I'm trying to accomplish, and I didn't write it, but I need to understand what its doing through each step. I'm not really a script writer, so I thought you all might be able to help me.

for D in $(cat /etc/passwd | awk -F : '{print $6}' | sort -u) ; do
ls -d "$D" >/dev/null 2>&1 || echo "$D" 1>&2 ;
done

My general understanding is that the output of the passwd file is being compared against another file and the results sorted to show only unique results. A directory listing is being run against the results of the $D variable and the output is being discarded.

I'm sure I've misinterpreted something, or just plain missed something all together. I am trying to get a better understanding of what's happening.

It generates a list of home directories that do not exist.

EDIT:

In more detail the only purpose of the ls command is to test whether each home directory exists; if it does not then ls returns a non-zero exit status and the following logical OR (||) test is passed and the name of the missing directory is printed on stderr (which has file descriptor 2).

Could be done a little more elegantly and robustly.

If you try the script with
for D in $(cat /etc/passwd | awk -F : '{print $6}' | sort -u) ; do
echo $D
done

you will see that the for loop is giving the /etc/passwd to AWK. AWK prints column number 6 which has the Home Directory included.

now look at the ls command. This will try to list the entries in column number 6 as directories and throw the ouptput away. if this fails the echo command is in charge and it will output the name .

The conclusion is that the script shows non existing home directories

Regard Norbert

EDIT: remove typos

The ls symdrome strikes back I'd rather use the -d builtin. You can save yourself one pipe as well.

Thanks for the feedback. That helps a lot.