Compute Checksums for all files in a directory

generate_checksums.sh

#!/bin/bash
#
# Autor: Robin Moser, 2021
# This script lists all files of a specified directory and saves it
# with its computed sha1 checksum in a text file. Useful for comparing,
# finding duplicates, logging, etc.
#
# Usage:
# ./generate_checksums.sh <directory> <slug>
#
# Where <directory> (without a trailing slash) specifies the directory
# which shall be scanned and where <slug> is an arbitrary name for that task,
# used in the list filename and in the log.
#
# Example:
# ./generate_checksums.sh /volume/media media
# ./generate_checksums.sh /volume/documents/private privatedocs

# direcotires which should be excluded
excludes+=(".git")
excludes+=("@eaDir")
excludes+=("excludedir - works with spaces")

# The path, where to store the file lists
path="/volume1/Backups/Synology/Checksums"
log="$path/log.txt"

# CLI arguments
dir="$1"
slug="$2"

if [ -z ${dir+x} ] || [ ! -d "$dir" ]; then
	echo "Directory needs to be defined and needs to exist"
	exit 1
fi

if [ -z ${slug+x} ]; then
	echo "Task name needs to be defined"
	exit 1
fi

# date for the filename
date="$(date +%Y%m%d)"

# date + time for the log entry
datetime="$(date "+%Y-%m-%dT%H:%M:%S")"

# the file, where the checksums with filenames should be stored
listfile="$path/sha1_nas_${slug}_${date}.txt"

# create output path if not existant
mkdir -p "$path"

# reserve the log entry, write start date
# Table HEAD: STARTDATE | ENDDATE | EXITCODE | TASKNAME
echo "$datetime | ------------------- | - | $slug" >> "$log"

# prepare the excludes statement
pattern=(-false)
for exclude in "${excludes[@]}"; do
	pattern+=(-o -name "$exclude");
done

# find all files, save the sha1sum, and write to the output file
find "$dir" -type d \( "${pattern[@]}" \) -prune -false \
	-o -type f -exec sha1sum {} \; > "$listfile"

# update the log file with the end date and the exit code of find
sed -i ".bak" "s/\($datetime\) | --* | -/\1 | $(date "+%Y-%m-%dT%H:%M:%S") | $?/" "$log" && \
	rm "${log}.bak"

exit 0