Convert all pdfs to text

convertPdfsToText.rb

#! /usr/bin/env ruby

require 'fileutils.rb'

$home=ENV['HOME']
$documents="#{$home}/Documents"
$already_tried_index="#{$documents}/already_tried_files.txt"

def does_pdf_file_contain_text?(fn)
  text=`pdftotext #{fn.inspect} /dev/stdout`
  text =~ /\w+/
end

$all_pdfs = Dir["#{$documents}/**/*.pdf"]
$already_tried_files = File.read($already_tried_index).split("\n")
$already_tried = $already_tried_files.inject({}) {|h,v| h[v] = 1; h}
$pdfs_without_text = $all_pdfs.select {|fn| !$already_tried.has_key?(fn) && 
    !does_pdf_file_contain_text?(fn) }

if $pdfs_without_text.size > 0
  system "open -a 'Scan to Searchable PDF' #{$pdfs_without_text.map{|n| n.inspect}.join(" ")}"
  $pdfs_without_text.each {|fn|
    puts "Indexing: #{fn}"
    File.open($already_tried_index, "a+") {|idx| idx.puts fn }
  }
else
  puts "No pdfs to index."
end