Extract ISBN from PDF files using pdftotext utility

scan_isbn.rb

#!/usr/bin/env ruby
# frozen_string_literal: true

# Scan PDF files for ISBN Number using pdftotext.
#
# This ruby script will scan pdf files and extract ISBN
# Number and append it to the filename (`%filename%_[ISBN].pdf`)
#
# Run this script in folder
#

require 'fileutils'
require 'logger'
require 'tmpdir'

require 'lisbn'

logger = Logger.new($stdout)
logger.level = Logger::DEBUG

list = Dir.entries('.', encoding: 'UTF-8').select do |e|
  File.extname(e).downcase == '.pdf' && !e.match?(/_\[\d+\]\.pdf/)
end

Dir.mktmpdir('isbn') do |tmp|
  list.each do |pdf_file|
    page_count = `pdfinfo "#{pdf_file}"`.scan(/Pages:[^\r\n\d]+(\d+)/).flatten.first.to_i

    text_f = File.join(tmp, "#{pdf_file}_f.txt")
    text_l = File.join(tmp, "#{pdf_file}_l.txt")

    _ = `pdftotext -l 20 -enc "UTF-8" "#{pdf_file}" "#{text_f}"`
    _ = `pdftotext -f #{page_count - 20} -l #{page_count} -enc "UTF-8" "#{pdf_file}" "#{text_l}"`

    isbn = (File.read(text_f, encoding: 'UTF-8') + File.read(text_l, encoding: 'UTF-8'))
           .scan(/(?:ISBN)?(?:-)?(?:10|13)?(?:[:-]+|10|13)?([\d\-X]{10,19})/)
           .flatten

    next if isbn.empty?

    isbn = isbn.flatten.map { |e| e.gsub(/[^\dX]+/, '') }
    isbn_check = isbn.select { |e| Lisbn.new(e).valid? }
    if !isbn_check.empty?
      new_name = pdf_file.gsub(/.pdf/i, "_[#{Lisbn.new(isbn_check.first).isbn13}].pdf")
      logger.info(format('ISBN Found -> "%s"', pdf_file))
      logger.info(format('RENAME_FILE "%s" -> "%s"', pdf_file, new_name))
      FileUtils.mv(pdf_file, new_name)
    else
      logger.error(format('ISBN_INVALID [%s] -> %s', isbn, pdf_file))
    end
  end
  puts 'Done with the task. Enter to exit ...'
  gets
end