hashdeep orchestrator for data integrity checks in home folder

rehash.sh

#!/bin/sh -v

cd ~
hash_file="Documents/hashdeep-checksums.txt"

mv "$hash_file" "$hash_file.bak"

time hashdeep -c md5 -of -r -l Music Downloads Documents Pictures Phone Dropbox repo oneplus9-home | tee "$hash_file"

verify-hashes-backup.sh

#!/bin/sh

backuproot="/media/tim/backup/backintime/any/tim/1/last_snapshot/backup/home/tim"

hash_file="$backuproot/Documents/hashdeep-checksums.txt"

tmp_dir=$(mktemp -d -t ci-XXXXXXXXXX)
output_file="$tmp_dir/hashdeep-checksums-verification.txt" 

cd "$backuproot"

time hashdeep -k  "$hash_file" -rle -of Music Downloads Documents Pictures Phone Dropbox repo oneplus9-home -avv | tee "$output_file" | grep "Known file not used"

xed "$output_file" &

echo '"Known file not used" means there was no match for this hash, i.e. deleted file'
echo '"No match" means new/modified file'

verify-hashes-previous.sh

#!/bin/sh

cd ~
hash_file="Documents/hashdeep-checksums.txt.bak"
output_file="Documents/hashdeep-checksums-verification.txt" 

mv "$output_file" "$output_file.bak"

time hashdeep -k  "$hash_file" -rle -of Music Downloads Documents Pictures Phone Dropbox repo oneplus9-home -avv | tee "$output_file" | grep "Known file not used"

xed "$output_file" &

echo '"Known file not used" means there was no match for this hash, i.e. deleted file'
echo '"No match" means new/modified file'

verify-hashes.sh

#!/bin/sh

cd ..
hash_file="Documents/hashdeep-checksums.txt"
output_file="/tmp/hashdeep-checksums-verification.txt" 
missing_file="/tmp/hashdeep-checksums-verification-missing.html" 

mv "$output_file" "$output_file.bak"

time hashdeep -k  "$hash_file" -rle -of Music Downloads Documents Pictures Phone Dropbox repo oneplus9-home -avv | tee "$output_file"

# https://github.com/timabell/paths2html
grep "Known file not used" "$output_file" | sed 's/:.*//'| sort | paths2html | tee "$missing_file" && xdg-open "$missing_file"

# Odd terminology from md5deep verify:
# - "Known file not used" means there was no match for this hash, i.e. deleted file
# - "No match" means new/modified file