Question

I happen need to use the following shell script

find . -type f -exec sh -c '
mv "$0" "${0%/*}/$(printf "%s\n" "${0##*/}" | sha1sum | cut -d" " -f1)"
' {} \;

But I do not understand how does this script work? For instance, how to analyze

mv "$0" "${0%/*}/$(printf "%s\n" "${0##*/}" | sha1sum | cut -d" " -f1)"
' {} \;

piece by piece? Thanks.

Answer 1

printf "%s\n" "${0##*/}"

${0##*/} takes the path $0 and strips off any leading directory names, leaving only the file name. The printf command adds a newline to the end and then this file name is piped to…

| sha1sum | cut -d" " -f1

This computes the SHA-1 hash of the file name and then uses cut to extract just the hash from sha1sum’s output.

${0%/*}

This is the opposite of ${0##*/}—this one gets the directories from $0 and throws away the file name.

So effectively, what ends up getting run is:

mv "$DIR/$FILENAME" "$DIR/$HASH_OF_FILENAME"

In English, it renames every file it finds to the SHA-1 hash of the original file name.

---

For what it’s worth, it could be simplified a bit and made more readable. I might write the mv command as:

mv "$0" "$(dirname "$0")/$(basename "$0" | sha1sum | awk "{print \$1}")