Sanitize file/folder names

utf8_to_ascii_rename.sh

#!/bin/bash

# Requires: python3 and unidecode module
# Install with: sudo apt install python3-unidecode OR pip3 install unidecode

dry_run=false
to_lowercase=false
strip_parens=false

print_usage() {
  echo "Usage: $0 [--dry-run] [--lowercase] [--strip-parens] <file-or-directory>"
  echo
  echo "Options:"
  echo "  --dry-run       Show renames without changing anything"
  echo "  --lowercase     Convert names to lowercase"
  echo "  --strip-parens  Remove parentheses and their contents from names"
}

rename_item() {
  local f="$1"
  local dir
  dir=$(dirname "$f")
  local base
  base=$(basename "$f")

  # Safe ASCII transliteration via Python
  local ascii_base
  ascii_base=$(python3 -c "import sys; from unidecode import unidecode; print(unidecode(sys.argv[1]))" "$base")

  # Optional lowercase
  if $to_lowercase; then
    ascii_base=$(echo "$ascii_base" | tr '[:upper:]' '[:lower:]')
  fi

  # Optional parentheses stripping
  if $strip_parens; then
    ascii_base=$(echo "$ascii_base" | sed -E 's/ *\([^)]*\)//g' | sed -E 's/  +/ /g' | sed -E 's/^ *| *$//g')
  fi

  if [[ "$base" != "$ascii_base" ]]; then
    local new_path="$dir/$ascii_base"
    echo "Renaming: $f -> $new_path"
    if ! $dry_run; then
      mv -n "$f" "$new_path"
    fi
  fi
}

rename_to_ascii() {
  local path="$1"
  local items=()

  if [ -d "$path" ]; then
    # Collect paths first
    while IFS= read -r f; do
      items+=("$f")
    done < <(find "$path" -depth)

    # Rename in the order returned by `find -depth`
    for item in "${items[@]}"; do
      rename_item "$item"
    done
  elif [ -f "$path" ]; then
    rename_item "$path"
  else
    echo "Error: '$path' is not a valid file or directory"
    exit 1
  fi
}

# --- Parse flags ---
while [[ "$1" == --* ]]; do
  case "$1" in
    --dry-run) dry_run=true ;;
    --lowercase) to_lowercase=true ;;
    --strip-parens) strip_parens=true ;;
    --help) print_usage; exit 0 ;;
    *) echo "Unknown option: $1"; print_usage; exit 1 ;;
  esac
  shift
done

# --- Path argument required ---
if [ -z "$1" ]; then
  print_usage
  exit 1
fi

rename_to_ascii "$1"