find deleted files

Sometimes you want to compare an archive backup drive and a target folder and see what items got  deleted in that folder. Useful if you have a flaky backup system like dropbox.

#!/usr/bin/env bash
# find_deleted.sh — list non-image files that exist in SOURCE but not in TARGET
# usage: ./find_deleted.sh [--rsync-check] SOURCE_DIR TARGET_DIR
set -euo pipefail
usage() {
  echo "Usage: $0 [--rsync-check] SOURCE_DIR TARGET_DIR" >&2
  exit 1
}
RSYNC_CHECK=0
if [ "${1:-}" = "--rsync-check" ]; then
  RSYNC_CHECK=1
  shift
fi
[ $# -eq 2 ] || usage
SRC="${1%/}"
DST="${2%/}"
[ -d "$SRC" ] || { echo "Source not found: $SRC" >&2; exit 2; }
[ -d "$DST" ] || { echo "Target not found: $DST" >&2; exit 3; }
# Returns 0 if the path looks like an image, 1 otherwise.
is_image_path() {
  # lower-case path (Bash-only ${var,,})
  local p="${1,,}"
  case "$p" in
    *.jpg|*.jpeg|*.png|*.gif|*.bmp|*.tif|*.tiff|*.webp|*.heic|*.heif|*.svg|\
    *.cr2|*.nef|*.arw|*.dng|*.rw2|*.orf|*.raf)
      return 0 ;;
    *)  return 1 ;;
  esac
}
missing_count=0
image_skipped=0
scanned=0
echo "Scanning (non-images only). Source: $SRC  Target: $DST" >&2
# Use a pipeline instead of process substitution for /bin/sh safety (still Bash script).
find "$SRC" -type f -print0 | while IFS= read -r -d '' f; do
  scanned=$((scanned + 1))
  rel="${f#"$SRC"/}"
  if is_image_path "$rel"; then
    image_skipped=$((image_skipped + 1))
    continue
  fi
  tgt="$DST/$rel"
  if [ ! -e "$tgt" ]; then
    echo "$rel"
    missing_count=$((missing_count + 1))
  fi
  # Lightweight progress feedback every 500 files
  if [ $((scanned % 500)) -eq 0 ]; then
    echo "…scanned $scanned (missing so far: $missing_count, images skipped: $image_skipped)" >&2
  fi
done
echo "Done. Scanned: $scanned; Missing (non-images): $missing_count; Images skipped: $image_skipped" >&2
# Optional cross-check using rsync (dry-run, excludes images).
if [ "$RSYNC_CHECK" -eq 1 ]; then
  echo >&2
  echo "rsync dry-run cross-check (non-images). Differences rsync would copy from SRC->DST:" >&2
  rsync -ani \
    --exclude='*.jpg'  --exclude='*.jpeg' --exclude='*.png'  --exclude='*.gif'  \
    --exclude='*.bmp'  --exclude='*.tif'  --exclude='*.tiff' --exclude='*.webp' \
    --exclude='*.heic' --exclude='*.heif' --exclude='*.svg'  --exclude='*.cr2'  \
    --exclude='*.nef'  --exclude='*.arw'  --exclude='*.dng'  --exclude='*.rw2'  \
    --exclude='*.orf'  --exclude='*.raf' \
    --delete --ignore-existing \
    "$SRC/" "$DST/" | sed 's/^/> /' >&2
  echo "Note: rsync output is informational only (dry run). No changes made." >&2
fi

Popular posts from this blog

Automatically Fix Song Metadata and Filenames on Linux with Beets

 🎵 Automatically Fix Song Metadata and Filenames on Linux with Beets Tired of sorting through unknown MP3s with cryptic filenames like Track01.mp3? Sick of missing album info, incorrect genres, and zero cover art? Here's a clean way to batch-fix your music collection with open source tools — no manual editing required. ✅ What You’ll Use beets — the core tool that organizes, renames, and tags your music by matching against the MusicBrainz database. libchromaprint-tools — provides fpcalc, which generates audio fingerprints to identify unlabeled tracks. ffmpeg (optional) — used by beets for transcoding or embedded cover art processing. 🔧 Installation sudo apt update sudo apt install beets libchromaprint-tools ffmpeg 🚀 One Command to Organize It All beet import /path/to/your/music/ Beets will: Fingerprint each file Query MusicBrainz for metadata Rename files (e.g., Artist - Title.mp3) Write correct ID3/metadata tag Optionally move files into organized folders 📁 Example Output Your ...
Read more

throttle traffic on apache

sudo apt install apache2-utils sudo apt install libapache2-mod-evasive vi /etc/apache2/mods-enabled/evasive.conf service apache restart edit it and set values as per below or whatever else you like, the time quantities are seconds     DOSSiteInterval     1     DOSBlockingPeriod   10     #DOSLogDir           "/var/lock/mod_evasive"
Read more

script to show all files with disk usage sizes

Normally to get the disk usage of a folder you use get properties on the rightclick or similar. However there are hidden files starting with . in their name that you can get info on, only in the terminal, using the  du -sh  command. If you have hundreds of them (which you do!) this becomes tedious. The following script gives you a directory listing with the files all indicated in file sizes and highlights on files of megabytes or gigabytes in size.  #!/bin/bash # Set Internal Field Separator to handle spaces in filenames IFS=$'\n' # Loop through all files and directories for i in $(/bin/ls -A); do        size=$(du -sh "$i" | awk '{print $1}')           # Apply colors: bold for MB, bright bold for GB      case "$size" in          *M) color="\e[1m" ;;         # Bold for Megabytes          *G) color="\e[1;97m" ;;      # Br...
Read more