Dump non-ascii benchmark

no-ascii.rb

# encoding: utf-8

require 'benchmark'
require 'iconv'


class String
  def remove_nonascii(replacement='')
    n=self.split("")
    self.slice!(0..self.size)
    n.each { |b|
      if b[0].to_i< 33 || b[0].to_i>127 then
        self.concat(replacement)
      else
        self.concat(b)
      end
    }
    self.to_s
  end

  def remove_non_ascii(replacement="") 
    self.gsub(/[\u0080-\u00ff]/,replacement)
  end

  def iconv
    Iconv.conv('US-ASCII//IGNORE', 'UTF-8', self)
  end

  def with_encode
    # See String#encode
    encoding_options = {
      :invalid           => :replace,  # Replace invalid byte sequences
      :undef             => :replace,  # Replace anything not defined in ASCII
      :replace           => '',        # Replace above with this
      :universal_newline => true       # Always break lines with \n
    }
    self.encode 'ASCII', encoding_options
  end

  def with_chars
    self.chars.select(&:ascii_only?).join.strip.gsub(/[(\s)]+/, " ").gsub("&nbsp;", "").strip
  end

  def regex
    self.gsub(/\P{ASCII}/, '')
  end
end

utf8 = '☼☹☼✿☺☻☹☃⌇♒♒⌨☝♡“¥¥ß©®@÷π≠ascii' * 3 
# puts utf8.force_encoding( Encoding.find('ASCII') )
puts utf8.with_encode
puts utf8.with_chars
puts utf8.iconv
puts utf8.regex
# puts utf8.encode Encoding.find('ASCII'), undef: :replace


n = 10_000
Benchmark.bm do |x|
  x.report('iconv') { n.times { utf8.iconv } }

  x.report('encod') { n.times { utf8.with_encode } }

  x.report('chars') { n.times { utf8.with_chars } }

  # x.report('force') { n.times { utf8.force_encoding('US-ASCII') } }
  x.report('regex') { n.times { utf8.regex } }
end