Replaces the duplicate files with the hard link to the file which comes first in the sorting

fdedupes.sh

#!/usr/bin/env bash

# set -x  # Uncomment to debug this shell script
set -o errexit \
	-o noclobber \
	-o pipefail

readonly PROGNAME="$(basename "$0")"
readonly VERSION="0.0.4"

declare -a FDUPES_OPTS=( "${FDUPES_OPTS[@]}" )
declare -a LN_OPTS=( "${FDUPES_OPTS[@]}" )
declare -i SHOW_DISK_USAGE="${SHOW_DISK_USAGE:-0}"

err() {
	printf "%s: %b\n" "$PROGNAME" "$*" >&2
}

usage() {
	cat <<-EOF
	This script replaces the duplicate files with hard links.

	Usage:
	  $PROGNAME [options] DIRECTORY...

	OPTIONS:
	  -r      for every directory given follow subdirectories
	          encountered within
	  -R dir  for each directory given after this option follow
	          subdirectories encountered within
	  -s      follow symlinks
	  -H      remap two and more hard links to one inode
	  -n      exclude zero-length files from consideration
	  -A      exclude hidden files from consideration
	  -o BY   select sort order for output and deleting; by file
	          modification time (BY='time'; default), status
	          change time (BY='ctime'), or filename (BY='name')
	  -i      reverse order while sorting
	  -U      show the total size before and after deduplication
	  -V      print $PROGNAME version and exit
	  -h      show this help

	KNOWN ISSUES:
	  * unable to deduplicate files that contain a newline in a filename
	EOF

	exit "${1:-2}"
}

get_total_size() {
	du -csh "$@" | awk '$2 == "total" {print $1}'
}

fdedupes() {
	local dup_file
	local -a dup_files=()

	# shellcheck disable=SC2162
	fdupes --sameline "${FDUPES_OPTS[@]}" -- "$@" |
		while read -d $'\n' -a dup_files; do
			[ -e "${dup_files[0]}" ] || continue
			for dup_file in "${dup_files[@]:1}"; do
				[ -e "$dup_file" ] || continue
				ln -f "${LN_OPTS[@]}" -- "${dup_files[0]}" "$dup_file" || continue
			done
	done
}

while getopts ":AiHnrR:sUo:vVh" opt; do
	case "$opt" in
		A | \
		i | \
		H | \
		n | \
		s | \
		r ) FDUPES_OPTS+=("-$opt") ;;
		R | \
		o ) FDUPES_OPTS+=("-$opt" "$OPTARG") ;;
		U ) SHOW_DISK_USAGE=1 ;;
		v ) LN_OPTS+=("-$opt") ;;
		V ) printf "%s %s\n" "$PROGNAME" "$VERSION"
			exit 0
			;;
		h ) usage 0 ;;
		: ) err "option requires an argument -- '-$OPTARG'"
			usage
			;;
		\?) err "illegal option -- '-$OPTARG'"
			usage
			;;
	esac
done

shift $((OPTIND-1))

# Return an error if positional parameters are not found
if [ -z "$1" ]; then
	err "no directories specified"
	usage
fi

if [ "$SHOW_DISK_USAGE" -eq 1 ]; then
	total_size_before="$(get_total_size "$@")"
	fdedupes "$@"
	total_size_after="$(get_total_size "$@")"

	printf 'Disk usage has been reduced from %s to %s.\n' \
		"$total_size_before" "$total_size_after"
else
	fdedupes "$@"
fi