replace from control chars to control pictures

replace-control-chars-to-control-pictures.rb

#!/usr/bin/env ruby

replace_chars = {}
(0...0x20).each do |n|
  replace_chars[n.chr] = (0x2400 + n).chr("utf-8")
end
#replace_chars[" "] = "\u{2423}"

while xml = ARGF.gets(">")
  print xml.gsub(/[\x00-\x1F&&[^\t\r\n]]/, replace_chars)
end

replace-control-chars.rb

#!/usr/bin/env ruby
require 'nokogiri'

puts RUBY_DESCRIPTION
puts "Nokogiri::VERSION = #{Nokogiri::VERSION}"
puts "Nokogiri::LIBXML_VERSION = #{Nokogiri::LIBXML_VERSION}"

class MyDocument < Nokogiri::XML::SAX::Document
  DEBUG_PRINT = Hash.new($VERBOSE)
  #DEBUG_PRINT[:fields_data] = true

  def xmldecl(version, encoding, standalone)
    p [:xmldecl, version, encoding, standalone] if DEBUG_PRINT[:xmldecl]
  end
  def start_document
    p [:start_document] if DEBUG_PRINT[:start_document]
    @text_stack = []
  end
  def end_document
    p [:end_document] if DEBUG_PRINT[:end_document]
  end
  def start_element_namespace(name, attrs = [], prefix = nil, uri = nil, ns = [])
    p [:start_element_namespace, name, attrs, prefix, uri, ns] if DEBUG_PRINT[:start_element_namespace]
    @text_stack.push ""
  end
  def end_element_namespace(name, prefix = nil, uri = nil)
    p [:end_element_namespace, name, prefix, uri] if DEBUG_PRINT[:end_element_namespace]
    texts = @text_stack.pop
    p [:end_element_namespace_texts, name, texts] if DEBUG_PRINT[:end_element_namespace_texts]
  end
  def characters(string)
    p [:characters, string] if DEBUG_PRINT[:characters]
    @text_stack.each do |text|
      text << string
    end
  end
  def comment(string)
    p [:comment, string] if DEBUG_PRINT[:comment]
  end
  def warning(string)
    p [:warning, string] if DEBUG_PRINT[:warning]
  end
  def error(string)
    p [:error, string] if DEBUG_PRINT[:error]
  end
  def cdata_block(string)
    p [:cdata_block, string] if DEBUG_PRINT[:cdata_block]
    @text_stack.each do |text|
      text << string
    end
  end
  def processing_instruction name, content
    p [:processing_instruction, name, content] if DEBUG_PRINT[:processing_instruction]
  end
end

parser = Nokogiri::XML::SAX::Parser.new(MyDocument.new)
replace_chars = {}
xml = "<x>"
(0..0x20).each do |n|
  replace_chars[n.chr] = [0x2400 + n].pack('U')
  xml << n.chr
end
xml << " </x>"
xml = xml.gsub(/[\x00-\x1F&&[^\t\r\n]]/, replace_chars)
p xml
parser.parse(xml)
xml = xml.gsub(/[\x00-\x20]/, replace_chars)
p xml
replace_chars[" "] = "\u{2423}"
xml = xml.gsub(/[\x00-\x20]/, replace_chars)
p xml

result.txt

% ruby -v replace-control-chars.rb
ruby 2.0.0p247 (2013-06-27 revision 41674) [x86_64-darwin11.4.2]
ruby 2.0.0p247 (2013-06-27 revision 41674) [x86_64-darwin11.4.2]
Nokogiri::VERSION = 1.6.0
Nokogiri::LIBXML_VERSION = 2.8.0
"<x>␀␁␂␃␄␅␆␇␈\t\n␋␌\r␎␏␐␑␒␓␔␕␖␗␘␙␚␛␜␝␞␟  </x>"
[:start_document]
[:start_element_namespace, "x", [], nil, nil, []]
[:characters, "␀␁␂␃␄␅␆␇␈\t\n␋␌\n␎␏␐␑␒␓␔␕␖␗␘␙␚␛␜␝␞␟  "]
[:end_element_namespace, "x", nil, nil]
[:end_element_namespace_texts, "x", "␀␁␂␃␄␅␆␇␈\t\n␋␌\n␎␏␐␑␒␓␔␕␖␗␘␙␚␛␜␝␞␟  "]
[:end_document]
"<x>␀␁␂␃␄␅␆␇␈␉␊␋␌␍␎␏␐␑␒␓␔␕␖␗␘␙␚␛␜␝␞␟␠␠</x>"
"<x>␀␁␂␃␄␅␆␇␈␉␊␋␌␍␎␏␐␑␒␓␔␕␖␗␘␙␚␛␜␝␞␟␠␠</x>"