converting characters that blow up our app

char_converter.rb

# config/initializers/char_converter.rb
require 'uri'

module Support
  class CharConverter

    def initialize(app)
      @app = app
    end

    def call(env)
      @app.call(self.class.sanitize_env(env))
    end

    def self.sanitize_env(env)
      ["HTTP_REFERER", "PATH_INFO", "QUERY_STRING", "REQUEST_PATH", "REQUEST_URI"].each do |key|
        next unless value = env[key]
        fixed = sanitize_string(URI.decode(value))
        env[key] = URI.encode(fixed) if fixed
      end
      ["HTTP_COOKIE"].each do |key|
        next unless value = env[key]
        fixed = sanitize_string(value)
        env[key] = fixed if fixed
      end
      env
    end

    def self.sanitize_string(string)
      return if !string.is_a?(String) || string == ''

      # Try it as UTF-8 directly
      cleaned = string.dup.force_encoding(Encoding::UTF_8)
      # Some of it might be old Windows code page
      cleaned.encode(Encoding::UTF_8, Encoding::Windows_1250) unless cleaned.valid_encoding?
    rescue EncodingError
      # Force it to UTF-8, throwing out invalid bits
      cleaned.encode(Encoding::UTF_8, invalid: :replace, undef: :replace)
    end

  end
end

Thanks! I came across this problem today and modified it to work with Ruby 1.9.3:

https://gist.github.com/3639014

In some cases, for example with such url /companies/123-%d0%9d%d0%be%d0%b2%d0%be%d0%b5-%d0%b6%d0%b5%d0%bb%d1%82%d0%be%d0%b5-%d1%82%d0%b0%d0%ba%0d-0.1%81%d0%b8 you code can't help.

Solution is to use

cleaned.encode!('UTF-16', 'UTF-8', :invalid => :replace, :replace => '') cleaned.encode!('UTF-8', 'UTF-16')

Link: http://stackoverflow.com/questions/2982677/ruby-1-9-invalid-byte-sequence-in-utf-8