encoding_test.rb

# encoding: utf-8
require 'minitest/spec'
require 'minitest/autorun'

puts RUBY_DESCRIPTION
describe "Encoding conversions" do

  def serialize(string)
    begin
      data = string.encode('utf-8')
    rescue EncodingError
      data = string.dup
      data.force_encoding 'utf-8'

      raise unless data.valid_encoding?
    end

    data.force_encoding('binary')
  end

  def deserialize(data)
    data.force_encoding('utf-8')
  end

  before do
    @source = "europäischen"
  end

  it "losslessly serializes and deserializes utf-8 data" do
    deserialize(serialize(@source)).must_equal @source
  end

  it "serializes and deserializes iso-8859-1 data as utf-8" do
    deserialize(serialize(@source.encode('iso-8859-1'))).must_equal @source
  end

  it "serializes and deserializes utf-8 data in binary form" do
    binary = @source.dup.force_encoding('binary')
    deserialize(serialize(binary)).must_equal @source
  end

  it "raises an exception when the data cannot be converted to utf-8" do
    binary = "\xFF".force_encoding('binary')
    lambda { deserialize(serialize(binary)) }.must_raise Encoding::UndefinedConversionError
  end

end

__END__
$ ruby encoding_test.rb
ruby 1.9.3p125 (2012-02-16 revision 34643) [x86_64-darwin11.3.0]
Run options: --seed 44410

# Running tests:

....

Finished tests in 0.000923s, 4333.6945 tests/s, 4333.6945 assertions/s.

4 tests, 4 assertions, 0 failures, 0 errors, 0 skips

$ jruby --1.9 encoding_test.rb
jruby 1.7.0.dev (ruby-1.9.3-p139) (2012-04-10 96f32d2) (Java HotSpot(TM) 64-Bit Server VM 1.6.0_29) [darwin-x86_64-java]
Run options: --seed 28683

# Running tests:

.F.F

Finished tests in 0.116000s, 34.4828 tests/s, 34.4828 assertions/s.

  1) Failure:
test_0004_raises_an_exception_when_the_data_cannot_be_converted_to_utf_8(Encoding conversions) [encoding_test.rb:44]:
Encoding::UndefinedConversionError expected but nothing was raised.

  2) Failure:
test_0003_serializes_and_deserializes_utf_8_data_in_binary_form(Encoding conversions) [encoding_test.rb:39]:
Expected: "europäischen"
  Actual: "europäischen"

4 tests, 4 assertions, 2 failures, 0 errors, 0 skips