Ruby script to find "strange" characters on Excel XLSX documents.

find.rb

# Installation: requires loofah gem
#
# Usage: ruby find.rb name_of_the.xlsx

require 'loofah'
require 'fileutils'

TMP_DIR        = 'tmp'
ASCII_TRESHOLD = 255
FILE_EXT       = '.xml'

Dir.mkdir TMP_DIR unless File.directory? TMP_DIR

content = ''
file    = ARGV[0]

FileUtils.cp(file, TMP_DIR)

`unzip #{TMP_DIR}/#{file} -d #{TMP_DIR}`

Dir.glob("#{TMP_DIR}/**/**").each do |file|
  content += Loofah.fragment(File.read(file)).scrub!(:strip) if file.split(//).last(FILE_EXT.length).join == FILE_EXT
end

FileUtils.rm_rf TMP_DIR

content.split('').sort.each { |c| puts "'#{c}' -> #{c.ord}" if c.ord > ASCII_TRESHOLD }