A script to convert filenames with Japanese characters to romanized versions

romanize.sh

#!/usr/bin/env bash

set -eu -o pipefail
# set -x

REMOVE_ORIGINALS=${REMOVE_ORIGINALS="0"}
ROMANIZED_DIR=${ROMANIZED_DIR="romanized"}

mkdir -p "$ROMANIZED_DIR"


main() {
 count=`list_bad_files | wc -l`
 index=1
 list_bad_files | while IFS= read -r file; do process_file "$file" "$count" $((index++)); done
 overwrite_line "Done"
}

process_file() { 
  local file=$1
  local total=$(printf "%04d" $2)
  local index=$(printf "%04d" $3)
  local romanized_file=`generate_romanized_name "$file"`

  overwrite_line "($index/$total) Processing: $romanized_file"

  copy_file "$file" "$ROMANIZED_DIR/$romanized_file"
  cond_remove_file "$file"
}

copy_file() {
  rsync -a -- "$1" "$2" || cp -n -- "$1" "$2"
}

cond_remove_file() {
  if [ "$REMOVE_ORIGINALS" -eq "1" ] ; then
    rm "$1"
  fi
}

overwrite_line() {
  echo -en "\e[1A";
  tput el;
  echo -e "$1"
}

generate_romanized_name() { echo "$1" | iconv -f utf8 -t eucjp | kakasi -i euc -w | kakasi -i euc -Ha -Ka -Ja -Ea -ka | iconv -f utf8 -t ascii//translit; }

list_bad_files() {
  all_files |
  find_non_ascii_chars |
  remove_incomplete_downloads
}

all_files() { find . -maxdepth 1 -type f; }

find_non_ascii_chars() { perl -ne 'print if ! /^[[:ascii:]]+$/'; }

remove_incomplete_downloads() { 
  grep -v ".!qB$" |
  grep -v ".part$"
}

main