/test/ruby/test_transcode.rb

http://github.com/ruby/ruby · Ruby · 2268 lines · 2145 code · 89 blank · 34 comment · 0 complexity · c15189a37de3cb325d5f3b99c587bec2 MD5 · raw file

Large files are truncated click here to view the full file

    

  1. # encoding: ASCII-8BIT   # make sure this runs in binary mode
    2. 

  2. # frozen_string_literal: false
    3. 

  3. # some of the comments are in UTF-8
    4. 

  4. 

  5. require 'test/unit'
    6. 

  6. 

  7. class TestTranscode < Test::Unit::TestCase
    8. 

  8.   def test_errors
    9. 

  9.     assert_raise(Encoding::ConverterNotFoundError) { 'abc'.encode('foo', 'bar') }
    10. 

  10.     assert_raise(Encoding::ConverterNotFoundError) { 'abc'.encode!('foo', 'bar') }
    11. 

  11.     assert_raise(Encoding::ConverterNotFoundError) { 'abc'.force_encoding('utf-8').encode('foo') }
    12. 

  12.     assert_raise(Encoding::ConverterNotFoundError) { 'abc'.force_encoding('utf-8').encode!('foo') }
    13. 

  13.     assert_raise(Encoding::UndefinedConversionError) { "\x80".encode('utf-8','ASCII-8BIT') }
    14. 

  14.     assert_raise(Encoding::InvalidByteSequenceError) { "\x80".encode('utf-8','US-ASCII') }
    15. 

  15.     assert_raise(Encoding::UndefinedConversionError) { "\xA5".encode('utf-8','iso-8859-3') }
    16. 

  16.     assert_raise(FrozenError) { 'hello'.freeze.encode!('iso-8859-1') }
    17. 

  17.     assert_raise(FrozenError) { '\u3053\u3093\u306b\u3061\u306f'.freeze.encode!('iso-8859-1') } # こんにちは
    18. 

  18.   end
    19. 

  19. 

  20.   def test_arguments
    21. 

  21.     assert_equal('abc', 'abc'.force_encoding('utf-8').encode('iso-8859-1'))
    22. 

  22.     # check that encoding is kept when no conversion is done
    23. 

  23.     assert_equal('abc'.force_encoding('Shift_JIS'), 'abc'.force_encoding('Shift_JIS').encode('Shift_JIS'))
    24. 

  24.     assert_equal('abc'.force_encoding('Shift_JIS'), 'abc'.force_encoding('Shift_JIS').encode!('Shift_JIS'))
    25. 

  25.     # assert that encoding is correctly set
    26. 

  26.     assert_equal("D\u00FCrst".encoding, "D\xFCrst".force_encoding('iso-8859-1').encode('utf-8').encoding)
    27. 

  27.     # check that Encoding can be used as parameter
    28. 

  28.     assert_equal("D\u00FCrst", "D\xFCrst".encode('utf-8', Encoding.find('ISO-8859-1')))
    29. 

  29.     assert_equal("D\u00FCrst", "D\xFCrst".encode(Encoding.find('utf-8'), 'ISO-8859-1'))
    30. 

  30.     assert_equal("D\u00FCrst", "D\xFCrst".encode(Encoding.find('utf-8'), Encoding.find('ISO-8859-1')))
    31. 

  31.   end
    32. 

  32. 

  33.   def test_noargument
    34. 

  34.     EnvUtil.with_default_internal(nil) do
    35. 

  35.       assert_equal("\u3042".encode, "\u3042")
    36. 

  36.       assert_equal("\xE3\x81\x82\x81".force_encoding("utf-8").encode,
    37. 

  37.                    "\xE3\x81\x82\x81".force_encoding("utf-8"))
    38. 

  38.     end
    39. 

  39.     EnvUtil.with_default_internal('EUC-JP') do
    40. 

  40.       assert_equal("\u3042".encode, "\xA4\xA2".force_encoding('EUC-JP'))
    41. 

  41.       assert_equal("\xE3\x81\x82\x81".force_encoding("utf-8").encode,
    42. 

  42.                    "\xA4\xA2?".force_encoding('EUC-JP'))
    43. 

  43.     end
    44. 

  44.   end
    45. 

  45. 

  46.   def test_length
    47. 

  47.     assert_equal("\u20AC"*20, ("\xA4"*20).encode('utf-8', 'iso-8859-15'))
    48. 

  48.     assert_equal("\u20AC"*20, ("\xA4"*20).encode!('utf-8', 'iso-8859-15'))
    49. 

  49.     assert_equal("\u20AC"*2000, ("\xA4"*2000).encode('utf-8', 'iso-8859-15'))
    50. 

  50.     assert_equal("\u20AC"*2000, ("\xA4"*2000).encode!('utf-8', 'iso-8859-15'))
    51. 

  51.     assert_equal("\u20AC"*200000, ("\xA4"*200000).encode('utf-8', 'iso-8859-15'))
    52. 

  52.     assert_equal("\u20AC"*200000, ("\xA4"*200000).encode!('utf-8', 'iso-8859-15'))
    53. 

  53.   end
    54. 

  54. 

  55.   def check_both_ways(utf8, raw, encoding)
    56. 

  56.     assert_equal(utf8.force_encoding('utf-8'), raw.encode('utf-8', encoding),utf8.dump+raw.dump)
    57. 

  57.     assert_equal(raw.force_encoding(encoding), utf8.encode(encoding, 'utf-8'))
    58. 

  58.   end
    59. 

  59. 

  60.   def check_both_ways2(str1, enc1, str2, enc2)
    61. 

  61.     assert_equal(str1.force_encoding(enc1), str2.encode(enc1, enc2))
    62. 

  62.     assert_equal(str2.force_encoding(enc2), str1.encode(enc2, enc1))
    63. 

  63.   end
    64. 

  64. 

  65.   def test_encoding_of_ascii_originating_from_binary
    66. 

  66.     binary_string = [0x82, 0x74, 0x68, 0x69, 0x73, 0x20, 0x69, 0x73, 0x20,
    67. 

  67.                      0x61, 0x20, 0x76, 0x65, 0x72, 0x79, 0x20, 0x6c, 0x6f,
    68. 

  68.                      0x6e, 0x67, 0x20, 0x73, 0x74, 0x72, 0x69, 0x6e, 0x67]
    69. 

  69.     class << binary_string
    70. 

  70.       # create a copy on write substring that contains
    71. 

  71.       # just the ascii characters (i.e. this is...), in JRuby
    72. 

  72.       # the underlying string have the same buffer backing
    73. 

  73.       # it up, but the offset of the string will be 1 instead
    74. 

  74.       # of 0.
    75. 

  75.       def make_cow_substring
    76. 

  76.         pack('C27').slice(1, 26)
    77. 

  77.       end
    78. 

  78.     end
    79. 

  79. 

  80.     ascii_string  = binary_string.make_cow_substring
    81. 

  81.     assert_equal("this is a very long string", ascii_string)
    82. 

  82.     assert_equal(Encoding::ASCII_8BIT, ascii_string.encoding)
    83. 

  83.     utf8_string = nil
    84. 

  84.     assert_nothing_raised("JRUBY-6764") do
    85. 

  85.       utf8_string = ascii_string.encode(Encoding::UTF_8)
    86. 

  86.     end
    87. 

  87.     assert_equal("this is a very long string", utf8_string)
    88. 

  88.     assert_equal(Encoding::UTF_8, utf8_string.encoding)
    89. 

  89.   end
    90. 

  90. 

  91.   def test_encodings
    92. 

  92.     check_both_ways("\u307E\u3064\u3082\u3068 \u3086\u304D\u3072\u308D",
    93. 

  93.         "\x82\xdc\x82\xc2\x82\xe0\x82\xc6 \x82\xe4\x82\xab\x82\xd0\x82\xeb", 'shift_jis') # まつもと ゆきひろ
    94. 

  94.     check_both_ways("\u307E\u3064\u3082\u3068 \u3086\u304D\u3072\u308D",
    95. 

  95.         "\xa4\xde\xa4\xc4\xa4\xe2\xa4\xc8 \xa4\xe6\xa4\xad\xa4\xd2\xa4\xed", 'euc-jp')
    96. 

  96.     check_both_ways("\u307E\u3064\u3082\u3068 \u3086\u304D\u3072\u308D",
    97. 

  97.         "\xa4\xde\xa4\xc4\xa4\xe2\xa4\xc8 \xa4\xe6\xa4\xad\xa4\xd2\xa4\xed", 'euc-jis-2004')
    98. 

  98.     check_both_ways("\u677E\u672C\u884C\u5F18", "\x8f\xbc\x96\x7b\x8d\x73\x8d\x4f", 'shift_jis') # 松本行弘
    99. 

  99.     check_both_ways("\u677E\u672C\u884C\u5F18", "\xbe\xbe\xcb\xdc\xb9\xd4\xb9\xb0", 'euc-jp')
    100. 

  100.     check_both_ways("\u677E\u672C\u884C\u5F18", "\xbe\xbe\xcb\xdc\xb9\xd4\xb9\xb0", 'euc-jis-2004')
    101. 

  101.     check_both_ways("D\u00FCrst", "D\xFCrst", 'iso-8859-1') # Dürst
    102. 

  102.     check_both_ways("D\u00FCrst", "D\xFCrst", 'iso-8859-2')
    103. 

  103.     check_both_ways("D\u00FCrst", "D\xFCrst", 'iso-8859-3')
    104. 

  104.     check_both_ways("D\u00FCrst", "D\xFCrst", 'iso-8859-4')
    105. 

  105.     check_both_ways("D\u00FCrst", "D\xFCrst", 'iso-8859-9')
    106. 

  106.     check_both_ways("D\u00FCrst", "D\xFCrst", 'iso-8859-10')
    107. 

  107.     check_both_ways("D\u00FCrst", "D\xFCrst", 'iso-8859-13')
    108. 

  108.     check_both_ways("D\u00FCrst", "D\xFCrst", 'iso-8859-14')
    109. 

  109.     check_both_ways("D\u00FCrst", "D\xFCrst", 'iso-8859-15')
    110. 

  110.     check_both_ways("r\u00E9sum\u00E9", "r\xE9sum\xE9", 'iso-8859-1') # résumé
    111. 

  111.     check_both_ways("\u0065\u006C\u0151\u00ED\u0072\u00E1\u0073", "el\xF5\xEDr\xE1s", 'iso-8859-2') # előírás
    112. 

  112.     check_both_ways("\u043F\u0435\u0440\u0435\u0432\u043E\u0434",
    113. 

  113.          "\xDF\xD5\xE0\xD5\xD2\xDE\xD4", 'iso-8859-5') # перевод
    114. 

  114.     check_both_ways("\u0643\u062A\u0628", "\xE3\xCA\xC8", 'iso-8859-6') # كتب
    115. 

  115.     check_both_ways("\u65E5\u8A18", "\x93\xFA\x8BL", 'shift_jis') # 日記
    116. 

  116.     check_both_ways("\u65E5\u8A18", "\xC6\xFC\xB5\xAD", 'euc-jp')
    117. 

  117.     check_both_ways("\u65E5\u8A18", "\xC6\xFC\xB5\xAD", 'euc-jis-2004')
    118. 

  118.     check_both_ways("\uC560\uC778\uAD6C\uD568\u0020\u6734\uC9C0\uC778",
    119. 

  119.          "\xBE\xD6\xC0\xCE\xB1\xB8\xC7\xD4\x20\xDA\xD3\xC1\xF6\xC0\xCE", 'euc-kr') # 애인구함 朴지인
    120. 

  120.     check_both_ways("\uC544\uD58F\uD58F\u0020\uB620\uBC29\uD6BD\uB2D8\u0020\uC0AC\uB791\uD716",
    121. 

  121.          "\xBE\xC6\xC1\x64\xC1\x64\x20\x8C\x63\xB9\xE6\xC4\x4F\xB4\xD4\x20\xBB\xE7\xB6\xFB\xC5\x42", 'cp949') # 아햏햏 똠방횽님 사랑휖
    122. 

  122.     assert_equal(Encoding::ISO_8859_1, "D\xFCrst".force_encoding('iso-8859-2').encode('iso-8859-1', 'iso-8859-1').encoding)
    123. 

  123.   end
    124. 

  124. 

  125.   def test_twostep
    126. 

  126.     assert_equal("D\xFCrst".force_encoding('iso-8859-2'), "D\xFCrst".encode('iso-8859-2', 'iso-8859-1'))
    127. 

  127.   end
    128. 

  128. 

  129.   def test_ascii_range
    130. 

  130.     encodings = [
    131. 

  131.       'US-ASCII', 'ASCII-8BIT',
    132. 

  132.       'ISO-8859-1', 'ISO-8859-2', 'ISO-8859-3',
    133. 

  133.       'ISO-8859-4', 'ISO-8859-5', 'ISO-8859-6',
    134. 

  134.       'ISO-8859-7', 'ISO-8859-8', 'ISO-8859-9',
    135. 

  135.       'ISO-8859-10', 'ISO-8859-11', 'ISO-8859-13',
    136. 

  136.       'ISO-8859-14', 'ISO-8859-15',
    137. 

  137.       'EUC-JP', 'SHIFT_JIS', 'EUC-KR'
    138. 

  138.     ]
    139. 

  139.     all_ascii = (0..127).to_a.pack 'C*'
    140. 

  140.     encodings.each do |enc|
    141. 

  141.       test_start = all_ascii
    142. 

  142.       assert_equal(test_start, test_start.encode('UTF-8',enc).encode(enc).force_encoding('ASCII-8BIT'))
    143. 

  143.     end
    144. 

  144.   end
    145. 

  145. 

  146.   def test_all_bytes
    147. 

  147.     encodings_8859 = [
    148. 

  148.       'ISO-8859-1', 'ISO-8859-2',
    149. 

  149.       #'ISO-8859-3', # not all bytes used
    150. 

  150.       'ISO-8859-4', 'ISO-8859-5',
    151. 

  151.       #'ISO-8859-6', # not all bytes used
    152. 

  152.       #'ISO-8859-7', # not all bytes used
    153. 

  153.       #'ISO-8859-8', # not all bytes used
    154. 

  154.       'ISO-8859-9', 'ISO-8859-10',
    155. 

  155.       #'ISO-8859-11', # not all bytes used
    156. 

  156.       #'ISO-8859-12', # not available
    157. 

  157.       'ISO-8859-13','ISO-8859-14','ISO-8859-15',
    158. 

  158.       #'ISO-8859-16', # not available
    159. 

  159.     ]
    160. 

  160.     all_bytes = (0..255).to_a.pack 'C*'
    161. 

  161.     encodings_8859.each do |enc|
    162. 

  162.       test_start = all_bytes
    163. 

  163.       assert_equal(test_start, test_start.encode('UTF-8',enc).encode(enc).force_encoding('ASCII-8BIT'))
    164. 

  164.     end
    165. 

  165.   end
    166. 

  166. 

  167.   def test_windows_874
    168. 

  168.     check_both_ways("\u20AC", "\x80", 'windows-874') # €
    169. 

  169.     assert_raise(Encoding::UndefinedConversionError) { "\x81".encode("utf-8", 'windows-874') }
    170. 

  170.     assert_raise(Encoding::UndefinedConversionError) { "\x84".encode("utf-8", 'windows-874') }
    171. 

  171.     check_both_ways("\u2026", "\x85", 'windows-874') # …
    172. 

  172.     assert_raise(Encoding::UndefinedConversionError) { "\x86".encode("utf-8", 'windows-874') }
    173. 

  173.     assert_raise(Encoding::UndefinedConversionError) { "\x8F".encode("utf-8", 'windows-874') }
    174. 

  174.     assert_raise(Encoding::UndefinedConversionError) { "\x90".encode("utf-8", 'windows-874') }
    175. 

  175.     check_both_ways("\u2018", "\x91", 'windows-874') # ‘
    176. 

  176.     check_both_ways("\u2014", "\x97", 'windows-874') # —
    177. 

  177.     assert_raise(Encoding::UndefinedConversionError) { "\x98".encode("utf-8", 'windows-874') }
    178. 

  178.     assert_raise(Encoding::UndefinedConversionError) { "\x9F".encode("utf-8", 'windows-874') }
    179. 

  179.     check_both_ways("\u00A0", "\xA0", 'windows-874') # non-breaking space
    180. 

  180.     check_both_ways("\u0E0F", "\xAF", 'windows-874') # ฏ
    181. 

  181.     check_both_ways("\u0E10", "\xB0", 'windows-874') # ฐ
    182. 

  182.     check_both_ways("\u0E1F", "\xBF", 'windows-874') # ฟ
    183. 

  183.     check_both_ways("\u0E20", "\xC0", 'windows-874') # ภ
    184. 

  184.     check_both_ways("\u0E2F", "\xCF", 'windows-874') # ฯ
    185. 

  185.     check_both_ways("\u0E30", "\xD0", 'windows-874') # ะ
    186. 

  186.     check_both_ways("\u0E3A", "\xDA", 'windows-874') # ฺ
    187. 

  187.     assert_raise(Encoding::UndefinedConversionError) { "\xDB".encode("utf-8", 'windows-874') }
    188. 

  188.     assert_raise(Encoding::UndefinedConversionError) { "\xDE".encode("utf-8", 'windows-874') }
    189. 

  189.     check_both_ways("\u0E3F", "\xDF", 'windows-874') # ฿
    190. 

  190.     check_both_ways("\u0E40", "\xE0", 'windows-874') # เ
    191. 

  191.     check_both_ways("\u0E4F", "\xEF", 'windows-874') # ๏
    192. 

  192.     check_both_ways("\u0E50", "\xF0", 'windows-874') # ๐
    193. 

  193.     check_both_ways("\u0E5B", "\xFB", 'windows-874') # ๛
    194. 

  194.     assert_raise(Encoding::UndefinedConversionError) { "\xFC".encode("utf-8", 'windows-874') }
    195. 

  195.     assert_raise(Encoding::UndefinedConversionError) { "\xFF".encode("utf-8", 'windows-874') }
    196. 

  196.   end
    197. 

  197. 

  198.   def test_windows_1250
    199. 

  199.     check_both_ways("\u20AC", "\x80", 'windows-1250') # €
    200. 

  200.     assert_raise(Encoding::UndefinedConversionError) { "\x81".encode("utf-8", 'windows-1250') }
    201. 

  201.     check_both_ways("\u201A", "\x82", 'windows-1250') # ‚
    202. 

  202.     assert_raise(Encoding::UndefinedConversionError) { "\x83".encode("utf-8", 'windows-1250') }
    203. 

  203.     check_both_ways("\u201E", "\x84", 'windows-1250') # „
    204. 

  204.     check_both_ways("\u2021", "\x87", 'windows-1250') # ‡
    205. 

  205.     assert_raise(Encoding::UndefinedConversionError) { "\x88".encode("utf-8", 'windows-1250') }
    206. 

  206.     check_both_ways("\u2030", "\x89", 'windows-1250') # ‰
    207. 

  207.     check_both_ways("\u0179", "\x8F", 'windows-1250') # Ź
    208. 

  208.     assert_raise(Encoding::UndefinedConversionError) { "\x90".encode("utf-8", 'windows-1250') }
    209. 

  209.     check_both_ways("\u2018", "\x91", 'windows-1250') # ‘
    210. 

  210.     check_both_ways("\u2014", "\x97", 'windows-1250') # —
    211. 

  211.     assert_raise(Encoding::UndefinedConversionError) { "\x98".encode("utf-8", 'windows-1250') }
    212. 

  212.     check_both_ways("\u2122", "\x99", 'windows-1250') # ™
    213. 

  213.     check_both_ways("\u00A0", "\xA0", 'windows-1250') # non-breaking space
    214. 

  214.     check_both_ways("\u017B", "\xAF", 'windows-1250') # Ż
    215. 

  215.     check_both_ways("\u00B0", "\xB0", 'windows-1250') # °
    216. 

  216.     check_both_ways("\u017C", "\xBF", 'windows-1250') # ż
    217. 

  217.     check_both_ways("\u0154", "\xC0", 'windows-1250') # Ŕ
    218. 

  218.     check_both_ways("\u010E", "\xCF", 'windows-1250') # Ď
    219. 

  219.     check_both_ways("\u0110", "\xD0", 'windows-1250') # Đ
    220. 

  220.     check_both_ways("\u00DF", "\xDF", 'windows-1250') # ß
    221. 

  221.     check_both_ways("\u0155", "\xE0", 'windows-1250') # ŕ
    222. 

  222.     check_both_ways("\u010F", "\xEF", 'windows-1250') # ď
    223. 

  223.     check_both_ways("\u0111", "\xF0", 'windows-1250') # đ
    224. 

  224.     check_both_ways("\u02D9", "\xFF", 'windows-1250') # ˙
    225. 

  225.   end
    226. 

  226. 

  227.   def test_windows_1251
    228. 

  228.     check_both_ways("\u0402", "\x80", 'windows-1251') # Ђ
    229. 

  229.     check_both_ways("\u20AC", "\x88", 'windows-1251') # €
    230. 

  230.     check_both_ways("\u040F", "\x8F", 'windows-1251') # Џ
    231. 

  231.     check_both_ways("\u0452", "\x90", 'windows-1251') # ђ
    232. 

  232.     assert_raise(Encoding::UndefinedConversionError) { "\x98".encode("utf-8", 'windows-1251') }
    233. 

  233.     check_both_ways("\u045F", "\x9F", 'windows-1251') # џ
    234. 

  234.     check_both_ways("\u00A0", "\xA0", 'windows-1251') # non-breaking space
    235. 

  235.     check_both_ways("\u0407", "\xAF", 'windows-1251') # Ї
    236. 

  236.     check_both_ways("\u00B0", "\xB0", 'windows-1251') # °
    237. 

  237.     check_both_ways("\u0457", "\xBF", 'windows-1251') # ї
    238. 

  238.     check_both_ways("\u0410", "\xC0", 'windows-1251') # А
    239. 

  239.     check_both_ways("\u041F", "\xCF", 'windows-1251') # П
    240. 

  240.     check_both_ways("\u0420", "\xD0", 'windows-1251') # Р
    241. 

  241.     check_both_ways("\u042F", "\xDF", 'windows-1251') # Я
    242. 

  242.     check_both_ways("\u0430", "\xE0", 'windows-1251') # а
    243. 

  243.     check_both_ways("\u043F", "\xEF", 'windows-1251') # п
    244. 

  244.     check_both_ways("\u0440", "\xF0", 'windows-1251') # р
    245. 

  245.     check_both_ways("\u044F", "\xFF", 'windows-1251') # я
    246. 

  246.   end
    247. 

  247. 

  248.   def test_windows_1252
    249. 

  249.     check_both_ways("\u20AC", "\x80", 'windows-1252') # €
    250. 

  250.     assert_raise(Encoding::UndefinedConversionError) { "\x81".encode("utf-8", 'windows-1252') }
    251. 

  251.     check_both_ways("\u201A", "\x82", 'windows-1252') # ‚
    252. 

  252.     check_both_ways("\u0152", "\x8C", 'windows-1252') # >Œ
    253. 

  253.     assert_raise(Encoding::UndefinedConversionError) { "\x8D".encode("utf-8", 'windows-1252') }
    254. 

  254.     check_both_ways("\u017D", "\x8E", 'windows-1252') # Ž
    255. 

  255.     assert_raise(Encoding::UndefinedConversionError) { "\x8F".encode("utf-8", 'windows-1252') }
    256. 

  256.     assert_raise(Encoding::UndefinedConversionError) { "\x90".encode("utf-8", 'windows-1252') }
    257. 

  257.     check_both_ways("\u2018", "\x91", 'windows-1252') #‘
    258. 

  258.     check_both_ways("\u0153", "\x9C", 'windows-1252') # œ
    259. 

  259.     assert_raise(Encoding::UndefinedConversionError) { "\x9D".encode("utf-8", 'windows-1252') }
    260. 

  260.     check_both_ways("\u017E", "\x9E", 'windows-1252') # ž
    261. 

  261.     check_both_ways("\u00A0", "\xA0", 'windows-1252') # non-breaking space
    262. 

  262.     check_both_ways("\u00AF", "\xAF", 'windows-1252') # ¯
    263. 

  263.     check_both_ways("\u00B0", "\xB0", 'windows-1252') # °
    264. 

  264.     check_both_ways("\u00BF", "\xBF", 'windows-1252') # ¿
    265. 

  265.     check_both_ways("\u00C0", "\xC0", 'windows-1252') # À
    266. 

  266.     check_both_ways("\u00CF", "\xCF", 'windows-1252') # Ï
    267. 

  267.     check_both_ways("\u00D0", "\xD0", 'windows-1252') # Ð
    268. 

  268.     check_both_ways("\u00DF", "\xDF", 'windows-1252') # ß
    269. 

  269.     check_both_ways("\u00E0", "\xE0", 'windows-1252') # à
    270. 

  270.     check_both_ways("\u00EF", "\xEF", 'windows-1252') # ï
    271. 

  271.     check_both_ways("\u00F0", "\xF0", 'windows-1252') # ð
    272. 

  272.     check_both_ways("\u00FF", "\xFF", 'windows-1252') # ÿ
    273. 

  273.   end
    274. 

  274. 

  275.   def test_windows_1253
    276. 

  276.     check_both_ways("\u20AC", "\x80", 'windows-1253') # €
    277. 

  277.     assert_raise(Encoding::UndefinedConversionError) { "\x81".encode("utf-8", 'windows-1253') }
    278. 

  278.     check_both_ways("\u201A", "\x82", 'windows-1253') # ‚
    279. 

  279.     check_both_ways("\u2021", "\x87", 'windows-1253') # ‡
    280. 

  280.     assert_raise(Encoding::UndefinedConversionError) { "\x88".encode("utf-8", 'windows-1253') }
    281. 

  281.     check_both_ways("\u2030", "\x89", 'windows-1253') # ‰
    282. 

  282.     assert_raise(Encoding::UndefinedConversionError) { "\x8A".encode("utf-8", 'windows-1253') }
    283. 

  283.     check_both_ways("\u2039", "\x8B", 'windows-1253') # ‹
    284. 

  284.     assert_raise(Encoding::UndefinedConversionError) { "\x8C".encode("utf-8", 'windows-1253') }
    285. 

  285.     assert_raise(Encoding::UndefinedConversionError) { "\x8F".encode("utf-8", 'windows-1253') }
    286. 

  286.     assert_raise(Encoding::UndefinedConversionError) { "\x90".encode("utf-8", 'windows-1253') }
    287. 

  287.     check_both_ways("\u2018", "\x91", 'windows-1253') # ‘
    288. 

  288.     check_both_ways("\u2014", "\x97", 'windows-1253') # —
    289. 

  289.     assert_raise(Encoding::UndefinedConversionError) { "\x98".encode("utf-8", 'windows-1253') }
    290. 

  290.     check_both_ways("\u2122", "\x99", 'windows-1253') # ™
    291. 

  291.     assert_raise(Encoding::UndefinedConversionError) { "\x9A".encode("utf-8", 'windows-1253') }
    292. 

  292.     check_both_ways("\u203A", "\x9B", 'windows-1253') # ›
    293. 

  293.     assert_raise(Encoding::UndefinedConversionError) { "\x9C".encode("utf-8", 'windows-1253') }
    294. 

  294.     assert_raise(Encoding::UndefinedConversionError) { "\x9F".encode("utf-8", 'windows-1253') }
    295. 

  295.     check_both_ways("\u00A0", "\xA0", 'windows-1253') # non-breaking space
    296. 

  296.     check_both_ways("\u2015", "\xAF", 'windows-1253') # ―
    297. 

  297.     check_both_ways("\u00B0", "\xB0", 'windows-1253') # °
    298. 

  298.     check_both_ways("\u038F", "\xBF", 'windows-1253') # Ώ
    299. 

  299.     check_both_ways("\u0390", "\xC0", 'windows-1253') # ΐ
    300. 

  300.     check_both_ways("\u039F", "\xCF", 'windows-1253') # Ο
    301. 

  301.     check_both_ways("\u03A0", "\xD0", 'windows-1253') # Π
    302. 

  302.     check_both_ways("\u03A1", "\xD1", 'windows-1253') # Ρ
    303. 

  303.     assert_raise(Encoding::UndefinedConversionError) { "\xD2".encode("utf-8", 'windows-1253') }
    304. 

  304.     check_both_ways("\u03A3", "\xD3", 'windows-1253') # Σ
    305. 

  305.     check_both_ways("\u03AF", "\xDF", 'windows-1253') # ί
    306. 

  306.     check_both_ways("\u03B0", "\xE0", 'windows-1253') # ΰ
    307. 

  307.     check_both_ways("\u03BF", "\xEF", 'windows-1253') # ο
    308. 

  308.     check_both_ways("\u03C0", "\xF0", 'windows-1253') # π
    309. 

  309.     check_both_ways("\u03CE", "\xFE", 'windows-1253') # ώ
    310. 

  310.     assert_raise(Encoding::UndefinedConversionError) { "\xFF".encode("utf-8", 'windows-1253') }
    311. 

  311.   end
    312. 

  312. 

  313.   def test_windows_1254
    314. 

  314.     check_both_ways("\u20AC", "\x80", 'windows-1254') # €
    315. 

  315.     assert_raise(Encoding::UndefinedConversionError) { "\x81".encode("utf-8", 'windows-1254') }
    316. 

  316.     check_both_ways("\u201A", "\x82", 'windows-1254') # ‚
    317. 

  317.     check_both_ways("\u0152", "\x8C", 'windows-1254') # Œ
    318. 

  318.     assert_raise(Encoding::UndefinedConversionError) { "\x8D".encode("utf-8", 'windows-1254') }
    319. 

  319.     assert_raise(Encoding::UndefinedConversionError) { "\x8F".encode("utf-8", 'windows-1254') }
    320. 

  320.     assert_raise(Encoding::UndefinedConversionError) { "\x90".encode("utf-8", 'windows-1254') }
    321. 

  321.     check_both_ways("\u2018", "\x91", 'windows-1254') # ‘
    322. 

  322.     check_both_ways("\u0153", "\x9C", 'windows-1254') # œ
    323. 

  323.     assert_raise(Encoding::UndefinedConversionError) { "\x9D".encode("utf-8", 'windows-1254') }
    324. 

  324.     assert_raise(Encoding::UndefinedConversionError) { "\x9E".encode("utf-8", 'windows-1254') }
    325. 

  325.     check_both_ways("\u0178", "\x9F", 'windows-1254') # Ÿ
    326. 

  326.     check_both_ways("\u00A0", "\xA0", 'windows-1254') # non-breaking space
    327. 

  327.     check_both_ways("\u00AF", "\xAF", 'windows-1254') # ¯
    328. 

  328.     check_both_ways("\u00B0", "\xB0", 'windows-1254') # °
    329. 

  329.     check_both_ways("\u00BF", "\xBF", 'windows-1254') # ¿
    330. 

  330.     check_both_ways("\u00C0", "\xC0", 'windows-1254') # À
    331. 

  331.     check_both_ways("\u00CF", "\xCF", 'windows-1254') # Ï
    332. 

  332.     check_both_ways("\u011E", "\xD0", 'windows-1254') # Ğ
    333. 

  333.     check_both_ways("\u00DF", "\xDF", 'windows-1254') # ß
    334. 

  334.     check_both_ways("\u00E0", "\xE0", 'windows-1254') # à
    335. 

  335.     check_both_ways("\u00EF", "\xEF", 'windows-1254') # ï
    336. 

  336.     check_both_ways("\u011F", "\xF0", 'windows-1254') # ğ
    337. 

  337.     check_both_ways("\u00FF", "\xFF", 'windows-1254') # ÿ
    338. 

  338.   end
    339. 

  339. 

  340.   def test_windows_1255
    341. 

  341.     check_both_ways("\u20AC", "\x80", 'windows-1255') # €
    342. 

  342.     assert_raise(Encoding::UndefinedConversionError) { "\x81".encode("utf-8", 'windows-1255') }
    343. 

  343.     check_both_ways("\u201A", "\x82", 'windows-1255') # ‚
    344. 

  344.     check_both_ways("\u2030", "\x89", 'windows-1255') # ‰
    345. 

  345.     assert_raise(Encoding::UndefinedConversionError) { "\x8A".encode("utf-8", 'windows-1255') }
    346. 

  346.     check_both_ways("\u2039", "\x8B", 'windows-1255') # ‹
    347. 

  347.     assert_raise(Encoding::UndefinedConversionError) { "\x8C".encode("utf-8", 'windows-1255') }
    348. 

  348.     assert_raise(Encoding::UndefinedConversionError) { "\x8F".encode("utf-8", 'windows-1255') }
    349. 

  349.     assert_raise(Encoding::UndefinedConversionError) { "\x90".encode("utf-8", 'windows-1255') }
    350. 

  350.     check_both_ways("\u2018", "\x91", 'windows-1255') # ‘
    351. 

  351.     check_both_ways("\u2122", "\x99", 'windows-1255') # ™
    352. 

  352.     assert_raise(Encoding::UndefinedConversionError) { "\x9A".encode("utf-8", 'windows-1255') }
    353. 

  353.     check_both_ways("\u203A", "\x9B", 'windows-1255') # ›
    354. 

  354.     assert_raise(Encoding::UndefinedConversionError) { "\x9C".encode("utf-8", 'windows-1255') }
    355. 

  355.     assert_raise(Encoding::UndefinedConversionError) { "\x9F".encode("utf-8", 'windows-1255') }
    356. 

  356.     check_both_ways("\u00A0", "\xA0", 'windows-1255') # non-breaking space
    357. 

  357.     check_both_ways("\u00A1", "\xA1", 'windows-1255') # ¡
    358. 

  358.     check_both_ways("\u00D7", "\xAA", 'windows-1255') # ×
    359. 

  359.     check_both_ways("\u00AF", "\xAF", 'windows-1255') # ¯
    360. 

  360.     check_both_ways("\u00B0", "\xB0", 'windows-1255') # °
    361. 

  361.     check_both_ways("\u00B8", "\xB8", 'windows-1255') # ¸
    362. 

  362.     check_both_ways("\u00F7", "\xBA", 'windows-1255') # ÷
    363. 

  363.     check_both_ways("\u00BF", "\xBF", 'windows-1255') # ¿
    364. 

  364.     check_both_ways("\u05B0", "\xC0", 'windows-1255') # ְ
    365. 

  365.     check_both_ways("\u05B9", "\xC9", 'windows-1255') # ֹ
    366. 

  366.     check_both_ways("\u05BA", "\xCA", 'windows-1255') # ֺ
    367. 

  367.     check_both_ways("\u05BB", "\xCB", 'windows-1255') # ֻ
    368. 

  368.     check_both_ways("\u05BF", "\xCF", 'windows-1255') # ֿ
    369. 

  369.     check_both_ways("\u05C0", "\xD0", 'windows-1255') # ׀
    370. 

  370.     check_both_ways("\u05F3", "\xD7", 'windows-1255') # ׳
    371. 

  371.     check_both_ways("\u05F4", "\xD8", 'windows-1255') # ״
    372. 

  372.     assert_raise(Encoding::UndefinedConversionError) { "\xD9".encode("utf-8", 'windows-1255') }
    373. 

  373.     assert_raise(Encoding::UndefinedConversionError) { "\xDF".encode("utf-8", 'windows-1255') }
    374. 

  374.     check_both_ways("\u05D0", "\xE0", 'windows-1255') # א
    375. 

  375.     check_both_ways("\u05DF", "\xEF", 'windows-1255') # ן
    376. 

  376.     check_both_ways("\u05E0", "\xF0", 'windows-1255') # נ
    377. 

  377.     check_both_ways("\u05EA", "\xFA", 'windows-1255') # ת
    378. 

  378.     assert_raise(Encoding::UndefinedConversionError) { "\xFB".encode("utf-8", 'windows-1255') }
    379. 

  379.     assert_raise(Encoding::UndefinedConversionError) { "\xFC".encode("utf-8", 'windows-1255') }
    380. 

  380.     check_both_ways("\u200E", "\xFD", 'windows-1255') # left-to-right mark
    381. 

  381.     check_both_ways("\u200F", "\xFE", 'windows-1255') # right-to-left mark
    382. 

  382.     assert_raise(Encoding::UndefinedConversionError) { "\xFF".encode("utf-8", 'windows-1255') }
    383. 

  383.   end
    384. 

  384. 

  385.   def test_windows_1256
    386. 

  386.     check_both_ways("\u20AC", "\x80", 'windows-1256') # €
    387. 

  387.     check_both_ways("\u0679", "\x8A", 'windows-1256') # ٹ
    388. 

  388.     check_both_ways("\u0688", "\x8F", 'windows-1256') # ڈ
    389. 

  389.     check_both_ways("\u06AF", "\x90", 'windows-1256') # گ
    390. 

  390.     check_both_ways("\u06A9", "\x98", 'windows-1256') # ک
    391. 

  391.     check_both_ways("\u0691", "\x9A", 'windows-1256') # ڑ
    392. 

  392.     check_both_ways("\u06BA", "\x9F", 'windows-1256') # ں
    393. 

  393.     check_both_ways("\u00A0", "\xA0", 'windows-1256') # non-breaking space
    394. 

  394.     check_both_ways("\u06BE", "\xAA", 'windows-1256') # ھ
    395. 

  395.     check_both_ways("\u00AF", "\xAF", 'windows-1256') # ¯
    396. 

  396.     check_both_ways("\u00B0", "\xB0", 'windows-1256') # °
    397. 

  397.     check_both_ways("\u061F", "\xBF", 'windows-1256') # ؟
    398. 

  398.     check_both_ways("\u06C1", "\xC0", 'windows-1256') # ہ
    399. 

  399.     check_both_ways("\u062F", "\xCF", 'windows-1256') # د
    400. 

  400.     check_both_ways("\u0630", "\xD0", 'windows-1256') # ذ
    401. 

  401.     check_both_ways("\u0643", "\xDF", 'windows-1256') # ك
    402. 

  402.     check_both_ways("\u00E0", "\xE0", 'windows-1256') # à
    403. 

  403.     check_both_ways("\u00EF", "\xEF", 'windows-1256') # ï
    404. 

  404.     check_both_ways("\u064B", "\xF0", 'windows-1256') #  ًً
    405. 

  405.     check_both_ways("\u06D2", "\xFF", 'windows-1256') # ے
    406. 

  406.   end
    407. 

  407. 

  408.   def test_windows_1257
    409. 

  409.     check_both_ways("\u20AC", "\x80", 'windows-1257') # €
    410. 

  410.     assert_raise(Encoding::UndefinedConversionError) { "\x81".encode("utf-8", 'windows-1257') }
    411. 

  411.     check_both_ways("\u201A", "\x82", 'windows-1257') # ‚
    412. 

  412.     assert_raise(Encoding::UndefinedConversionError) { "\x83".encode("utf-8", 'windows-1257') }
    413. 

  413.     check_both_ways("\u201E", "\x84", 'windows-1257') # „
    414. 

  414.     check_both_ways("\u2021", "\x87", 'windows-1257') # ‡
    415. 

  415.     assert_raise(Encoding::UndefinedConversionError) { "\x88".encode("utf-8", 'windows-1257') }
    416. 

  416.     check_both_ways("\u2030", "\x89", 'windows-1257') # ‰
    417. 

  417.     assert_raise(Encoding::UndefinedConversionError) { "\x8A".encode("utf-8", 'windows-1257') }
    418. 

  418.     check_both_ways("\u2039", "\x8B", 'windows-1257') # ‹
    419. 

  419.     assert_raise(Encoding::UndefinedConversionError) { "\x8C".encode("utf-8", 'windows-1257') }
    420. 

  420.     check_both_ways("\u00A8", "\x8D", 'windows-1257') # ¨
    421. 

  421.     check_both_ways("\u02C7", "\x8E", 'windows-1257') # ˇ
    422. 

  422.     check_both_ways("\u00B8", "\x8F", 'windows-1257') # ¸
    423. 

  423.     assert_raise(Encoding::UndefinedConversionError) { "\x90".encode("utf-8", 'windows-1257') }
    424. 

  424.     check_both_ways("\u2018", "\x91", 'windows-1257') # ‘
    425. 

  425.     check_both_ways("\u2014", "\x97", 'windows-1257') # —
    426. 

  426.     assert_raise(Encoding::UndefinedConversionError) { "\x98".encode("utf-8", 'windows-1257') }
    427. 

  427.     check_both_ways("\u2122", "\x99", 'windows-1257') # ™
    428. 

  428.     assert_raise(Encoding::UndefinedConversionError) { "\x9A".encode("utf-8", 'windows-1257') }
    429. 

  429.     check_both_ways("\u203A", "\x9B", 'windows-1257') # ›
    430. 

  430.     assert_raise(Encoding::UndefinedConversionError) { "\x9C".encode("utf-8", 'windows-1257') }
    431. 

  431.     check_both_ways("\u00AF", "\x9D", 'windows-1257') # ¯
    432. 

  432.     check_both_ways("\u02DB", "\x9E", 'windows-1257') # ˛
    433. 

  433.     assert_raise(Encoding::UndefinedConversionError) { "\x9F".encode("utf-8", 'windows-1257') }
    434. 

  434.     check_both_ways("\u00A0", "\xA0", 'windows-1257') # non-breaking space
    435. 

  435.     assert_raise(Encoding::UndefinedConversionError) { "\xA1".encode("utf-8", 'windows-1257') }
    436. 

  436.     check_both_ways("\u00A2", "\xA2", 'windows-1257') # ¢
    437. 

  437.     check_both_ways("\u00A4", "\xA4", 'windows-1257') # ¤
    438. 

  438.     assert_raise(Encoding::UndefinedConversionError) { "\xA5".encode("utf-8", 'windows-1257') }
    439. 

  439.     check_both_ways("\u00A6", "\xA6", 'windows-1257') # ¦
    440. 

  440.     check_both_ways("\u00C6", "\xAF", 'windows-1257') # Æ
    441. 

  441.     check_both_ways("\u00B0", "\xB0", 'windows-1257') # °
    442. 

  442.     check_both_ways("\u00E6", "\xBF", 'windows-1257') # æ
    443. 

  443.     check_both_ways("\u0104", "\xC0", 'windows-1257') # Ą
    444. 

  444.     check_both_ways("\u013B", "\xCF", 'windows-1257') # Ļ
    445. 

  445.     check_both_ways("\u0160", "\xD0", 'windows-1257') # Š
    446. 

  446.     check_both_ways("\u00DF", "\xDF", 'windows-1257') # ß
    447. 

  447.     check_both_ways("\u0105", "\xE0", 'windows-1257') # ą
    448. 

  448.     check_both_ways("\u013C", "\xEF", 'windows-1257') # ļ
    449. 

  449.     check_both_ways("\u0161", "\xF0", 'windows-1257') # š
    450. 

  450.     check_both_ways("\u02D9", "\xFF", 'windows-1257') # ˙
    451. 

  451.   end
    452. 

  452. 

  453.   def test_IBM437
    454. 

  454.     check_both_ways("\u00C7", "\x80", 'IBM437') # Ç
    455. 

  455.     check_both_ways("\u00C5", "\x8F", 'IBM437') # Å
    456. 

  456.     check_both_ways("\u00C9", "\x90", 'IBM437') # É
    457. 

  457.     check_both_ways("\u0192", "\x9F", 'IBM437') # ƒ
    458. 

  458.     check_both_ways("\u00E1", "\xA0", 'IBM437') # á
    459. 

  459.     check_both_ways("\u00BB", "\xAF", 'IBM437') # »
    460. 

  460.     check_both_ways("\u2591", "\xB0", 'IBM437') # ░
    461. 

  461.     check_both_ways("\u2510", "\xBF", 'IBM437') # ┐
    462. 

  462.     check_both_ways("\u2514", "\xC0", 'IBM437') # └
    463. 

  463.     check_both_ways("\u2567", "\xCF", 'IBM437') # ╧
    464. 

  464.     check_both_ways("\u2568", "\xD0", 'IBM437') # ╨
    465. 

  465.     check_both_ways("\u2580", "\xDF", 'IBM437') # ▀
    466. 

  466.     check_both_ways("\u03B1", "\xE0", 'IBM437') # α
    467. 

  467.     check_both_ways("\u2229", "\xEF", 'IBM437') # ∩
    468. 

  468.     check_both_ways("\u2261", "\xF0", 'IBM437') # ≡
    469. 

  469.     check_both_ways("\u00A0", "\xFF", 'IBM437') # non-breaking space
    470. 

  470.   end
    471. 

  471. 

  472.   def test_IBM775
    473. 

  473.     check_both_ways("\u0106", "\x80", 'IBM775') # Ć
    474. 

  474.     check_both_ways("\u00C5", "\x8F", 'IBM775') # Å
    475. 

  475.     check_both_ways("\u00C9", "\x90", 'IBM775') # É
    476. 

  476.     check_both_ways("\u00A4", "\x9F", 'IBM775') # ¤
    477. 

  477.     check_both_ways("\u0100", "\xA0", 'IBM775') # Ā
    478. 

  478.     check_both_ways("\u00BB", "\xAF", 'IBM775') # »
    479. 

  479.     check_both_ways("\u2591", "\xB0", 'IBM775') # ░
    480. 

  480.     check_both_ways("\u2510", "\xBF", 'IBM775') # ┐
    481. 

  481.     check_both_ways("\u2514", "\xC0", 'IBM775') # └
    482. 

  482.     check_both_ways("\u017D", "\xCF", 'IBM775') # Ž
    483. 

  483.     check_both_ways("\u0105", "\xD0", 'IBM775') # ą
    484. 

  484.     check_both_ways("\u2580", "\xDF", 'IBM775') # ▀
    485. 

  485.     check_both_ways("\u00D3", "\xE0", 'IBM775') # Ó
    486. 

  486.     check_both_ways("\u2019", "\xEF", 'IBM775') # ’
    487. 

  487.     check_both_ways("\u00AD", "\xF0", 'IBM775') # soft hyphen
    488. 

  488.     check_both_ways("\u00A0", "\xFF", 'IBM775') # non-breaking space
    489. 

  489.   end
    490. 

  490. 

  491.   def test_IBM852
    492. 

  492.     check_both_ways("\u00C7", "\x80", 'IBM852') # Ç
    493. 

  493.     check_both_ways("\u0106", "\x8F", 'IBM852') # Ć
    494. 

  494.     check_both_ways("\u00C9", "\x90", 'IBM852') # É
    495. 

  495.     check_both_ways("\u010D", "\x9F", 'IBM852') # č
    496. 

  496.     check_both_ways("\u00E1", "\xA0", 'IBM852') # á
    497. 

  497.     check_both_ways("\u00BB", "\xAF", 'IBM852') # »
    498. 

  498.     check_both_ways("\u2591", "\xB0", 'IBM852') # ░
    499. 

  499.     check_both_ways("\u2510", "\xBF", 'IBM852') # ┐
    500. 

  500.     check_both_ways("\u2514", "\xC0", 'IBM852') # └
    501. 

  501.     check_both_ways("\u00A4", "\xCF", 'IBM852') # ¤
    502. 

  502.     check_both_ways("\u0111", "\xD0", 'IBM852') # đ
    503. 

  503.     check_both_ways("\u2580", "\xDF", 'IBM852') # ▀
    504. 

  504.     check_both_ways("\u00D3", "\xE0", 'IBM852') # Ó
    505. 

  505.     check_both_ways("\u00B4", "\xEF", 'IBM852') # ´
    506. 

  506.     check_both_ways("\u00AD", "\xF0", 'IBM852') # soft hyphen
    507. 

  507.     check_both_ways("\u00A0", "\xFF", 'IBM852') # non-breaking space
    508. 

  508.   end
    509. 

  509. 

  510.   def test_IBM855
    511. 

  511.     check_both_ways("\u0452", "\x80", 'IBM855') # ђ
    512. 

  512.     check_both_ways("\u0408", "\x8F", 'IBM855') # Ј
    513. 

  513.     check_both_ways("\u0459", "\x90", 'IBM855') # љ
    514. 

  514.     check_both_ways("\u042A", "\x9F", 'IBM855') # Ъ
    515. 

  515.     check_both_ways("\u0430", "\xA0", 'IBM855') # а
    516. 

  516.     check_both_ways("\u00BB", "\xAF", 'IBM855') # »
    517. 

  517.     check_both_ways("\u2591", "\xB0", 'IBM855') # ░
    518. 

  518.     check_both_ways("\u2510", "\xBF", 'IBM855') # ┐
    519. 

  519.     check_both_ways("\u2514", "\xC0", 'IBM855') # └
    520. 

  520.     check_both_ways("\u00A4", "\xCF", 'IBM855') # ¤
    521. 

  521.     check_both_ways("\u043B", "\xD0", 'IBM855') # л
    522. 

  522.     check_both_ways("\u2580", "\xDF", 'IBM855') # ▀
    523. 

  523.     check_both_ways("\u042F", "\xE0", 'IBM855') # Я
    524. 

  524.     check_both_ways("\u2116", "\xEF", 'IBM855') # №
    525. 

  525.     check_both_ways("\u00AD", "\xF0", 'IBM855') # soft hyphen
    526. 

  526.     check_both_ways("\u00A0", "\xFF", 'IBM855') # non-breaking space
    527. 

  527.   end
    528. 

  528. 

  529.   def test_IBM857
    530. 

  530.     check_both_ways("\u00C7", "\x80", 'IBM857') # Ç
    531. 

  531.     check_both_ways("\u00C5", "\x8F", 'IBM857') # Å
    532. 

  532.     check_both_ways("\u00C9", "\x90", 'IBM857') # É
    533. 

  533.     check_both_ways("\u015F", "\x9F", 'IBM857') # ş
    534. 

  534.     check_both_ways("\u00E1", "\xA0", 'IBM857') # á
    535. 

  535.     check_both_ways("\u00BB", "\xAF", 'IBM857') # »
    536. 

  536.     check_both_ways("\u2591", "\xB0", 'IBM857') # ░
    537. 

  537.     check_both_ways("\u2510", "\xBF", 'IBM857') # ┐
    538. 

  538.     check_both_ways("\u2514", "\xC0", 'IBM857') # └
    539. 

  539.     check_both_ways("\u00A4", "\xCF", 'IBM857') # ¤
    540. 

  540.     check_both_ways("\u00BA", "\xD0", 'IBM857') # º
    541. 

  541.     check_both_ways("\u00C8", "\xD4", 'IBM857') # È
    542. 

  542.     assert_raise(Encoding::UndefinedConversionError) { "\xD5".encode("utf-8", 'IBM857') }
    543. 

  543.     check_both_ways("\u00CD", "\xD6", 'IBM857') # Í
    544. 

  544.     check_both_ways("\u2580", "\xDF", 'IBM857') # ▀
    545. 

  545.     check_both_ways("\u00D3", "\xE0", 'IBM857') # Ó
    546. 

  546.     check_both_ways("\u00B5", "\xE6", 'IBM857') # µ
    547. 

  547.     assert_raise(Encoding::UndefinedConversionError) { "\xE7".encode("utf-8", 'IBM857') }
    548. 

  548.     check_both_ways("\u00D7", "\xE8", 'IBM857') # ×
    549. 

  549.     check_both_ways("\u00B4", "\xEF", 'IBM857') # ´
    550. 

  550.     check_both_ways("\u00AD", "\xF0", 'IBM857') # soft hyphen
    551. 

  551.     check_both_ways("\u00B1", "\xF1", 'IBM857') # ±
    552. 

  552.     assert_raise(Encoding::UndefinedConversionError) { "\xF2".encode("utf-8", 'IBM857') }
    553. 

  553.     check_both_ways("\u00BE", "\xF3", 'IBM857') # ¾
    554. 

  554.     check_both_ways("\u00A0", "\xFF", 'IBM857') # non-breaking space
    555. 

  555.   end
    556. 

  556. 

  557.   def test_IBM860
    558. 

  558.     check_both_ways("\u00C7", "\x80", 'IBM860') # Ç
    559. 

  559.     check_both_ways("\u00C2", "\x8F", 'IBM860') # Â
    560. 

  560.     check_both_ways("\u00C9", "\x90", 'IBM860') # É
    561. 

  561.     check_both_ways("\u00D3", "\x9F", 'IBM860') # Ó
    562. 

  562.     check_both_ways("\u00E1", "\xA0", 'IBM860') # á
    563. 

  563.     check_both_ways("\u00BB", "\xAF", 'IBM860') # »
    564. 

  564.     check_both_ways("\u2591", "\xB0", 'IBM860') # ░
    565. 

  565.     check_both_ways("\u2510", "\xBF", 'IBM860') # ┐
    566. 

  566.     check_both_ways("\u2514", "\xC0", 'IBM860') # └
    567. 

  567.     check_both_ways("\u2567", "\xCF", 'IBM860') # ╧
    568. 

  568.     check_both_ways("\u2568", "\xD0", 'IBM860') # ╨
    569. 

  569.     check_both_ways("\u2580", "\xDF", 'IBM860') # ▀
    570. 

  570.     check_both_ways("\u03B1", "\xE0", 'IBM860') # α
    571. 

  571.     check_both_ways("\u2229", "\xEF", 'IBM860') # ∩
    572. 

  572.     check_both_ways("\u2261", "\xF0", 'IBM860') # ≡
    573. 

  573.     check_both_ways("\u00A0", "\xFF", 'IBM860') # non-breaking space
    574. 

  574.   end
    575. 

  575. 

  576.   def test_IBM861
    577. 

  577.     check_both_ways("\u00C7", "\x80", 'IBM861') # Ç
    578. 

  578.     check_both_ways("\u00C5", "\x8F", 'IBM861') # Å
    579. 

  579.     check_both_ways("\u00C9", "\x90", 'IBM861') # É
    580. 

  580.     check_both_ways("\u0192", "\x9F", 'IBM861') # ƒ
    581. 

  581.     check_both_ways("\u00E1", "\xA0", 'IBM861') # á
    582. 

  582.     check_both_ways("\u00BB", "\xAF", 'IBM861') # »
    583. 

  583.     check_both_ways("\u2591", "\xB0", 'IBM861') # ░
    584. 

  584.     check_both_ways("\u2510", "\xBF", 'IBM861') # ┐
    585. 

  585.     check_both_ways("\u2514", "\xC0", 'IBM861') # └
    586. 

  586.     check_both_ways("\u2567", "\xCF", 'IBM861') # ╧
    587. 

  587.     check_both_ways("\u2568", "\xD0", 'IBM861') # ╨
    588. 

  588.     check_both_ways("\u2580", "\xDF", 'IBM861') # ▀
    589. 

  589.     check_both_ways("\u03B1", "\xE0", 'IBM861') # α
    590. 

  590.     check_both_ways("\u2229", "\xEF", 'IBM861') # ∩
    591. 

  591.     check_both_ways("\u2261", "\xF0", 'IBM861') # ≡
    592. 

  592.     check_both_ways("\u00A0", "\xFF", 'IBM861') # non-breaking space
    593. 

  593.   end
    594. 

  594. 

  595.   def test_IBM862
    596. 

  596.     check_both_ways("\u05D0", "\x80", 'IBM862') # א
    597. 

  597.     check_both_ways("\u05DF", "\x8F", 'IBM862') # ן
    598. 

  598.     check_both_ways("\u05E0", "\x90", 'IBM862') # נ
    599. 

  599.     check_both_ways("\u0192", "\x9F", 'IBM862') # ƒ
    600. 

  600.     check_both_ways("\u00E1", "\xA0", 'IBM862') # á
    601. 

  601.     check_both_ways("\u00BB", "\xAF", 'IBM862') # »
    602. 

  602.     check_both_ways("\u2591", "\xB0", 'IBM862') # ░
    603. 

  603.     check_both_ways("\u2510", "\xBF", 'IBM862') # ┐
    604. 

  604.     check_both_ways("\u2514", "\xC0", 'IBM862') # └
    605. 

  605.     check_both_ways("\u2567", "\xCF", 'IBM862') # ╧
    606. 

  606.     check_both_ways("\u2568", "\xD0", 'IBM862') # ╨
    607. 

  607.     check_both_ways("\u2580", "\xDF", 'IBM862') # ▀
    608. 

  608.     check_both_ways("\u03B1", "\xE0", 'IBM862') # α
    609. 

  609.     check_both_ways("\u2229", "\xEF", 'IBM862') # ∩
    610. 

  610.     check_both_ways("\u2261", "\xF0", 'IBM862') # ≡
    611. 

  611.     check_both_ways("\u00A0", "\xFF", 'IBM862') # non-breaking space
    612. 

  612.   end
    613. 

  613. 

  614.   def test_IBM863
    615. 

  615.     check_both_ways("\u00C7", "\x80", 'IBM863') # Ç
    616. 

  616.     check_both_ways("\u00A7", "\x8F", 'IBM863') # §
    617. 

  617.     check_both_ways("\u00C9", "\x90", 'IBM863') # É
    618. 

  618.     check_both_ways("\u0192", "\x9F", 'IBM863') # ƒ
    619. 

  619.     check_both_ways("\u00A6", "\xA0", 'IBM863') # ¦
    620. 

  620.     check_both_ways("\u00BB", "\xAF", 'IBM863') # »
    621. 

  621.     check_both_ways("\u2591", "\xB0", 'IBM863') # ░
    622. 

  622.     check_both_ways("\u2510", "\xBF", 'IBM863') # ┐
    623. 

  623.     check_both_ways("\u2514", "\xC0", 'IBM863') # └
    624. 

  624.     check_both_ways("\u2567", "\xCF", 'IBM863') # ╧
    625. 

  625.     check_both_ways("\u2568", "\xD0", 'IBM863') # ╨
    626. 

  626.     check_both_ways("\u2580", "\xDF", 'IBM863') # ▀
    627. 

  627.     check_both_ways("\u03B1", "\xE0", 'IBM863') # α
    628. 

  628.     check_both_ways("\u2229", "\xEF", 'IBM863') # ∩
    629. 

  629.     check_both_ways("\u2261", "\xF0", 'IBM863') # ≡
    630. 

  630.     check_both_ways("\u00A0", "\xFF", 'IBM863') # non-breaking space
    631. 

  631.   end
    632. 

  632. 

  633.   def test_IBM865
    634. 

  634.     check_both_ways("\u00C7", "\x80", 'IBM865') # Ç
    635. 

  635.     check_both_ways("\u00C5", "\x8F", 'IBM865') # Å
    636. 

  636.     check_both_ways("\u00C9", "\x90", 'IBM865') # É
    637. 

  637.     check_both_ways("\u0192", "\x9F", 'IBM865') # ƒ
    638. 

  638.     check_both_ways("\u00E1", "\xA0", 'IBM865') # á
    639. 

  639.     check_both_ways("\u00A4", "\xAF", 'IBM865') # ¤
    640. 

  640.     check_both_ways("\u2591", "\xB0", 'IBM865') # ░
    641. 

  641.     check_both_ways("\u2510", "\xBF", 'IBM865') # ┐
    642. 

  642.     check_both_ways("\u2514", "\xC0", 'IBM865') # └
    643. 

  643.     check_both_ways("\u2567", "\xCF", 'IBM865') # ╧
    644. 

  644.     check_both_ways("\u2568", "\xD0", 'IBM865') # ╨
    645. 

  645.     check_both_ways("\u2580", "\xDF", 'IBM865') # ▀
    646. 

  646.     check_both_ways("\u03B1", "\xE0", 'IBM865') # α
    647. 

  647.     check_both_ways("\u2229", "\xEF", 'IBM865') # ∩
    648. 

  648.     check_both_ways("\u2261", "\xF0", 'IBM865') # ≡
    649. 

  649.     check_both_ways("\u00A0", "\xFF", 'IBM865') # non-breaking space
    650. 

  650.   end
    651. 

  651. 

  652.   def test_IBM866
    653. 

  653.     check_both_ways("\u0410", "\x80", 'IBM866') # А
    654. 

  654.     check_both_ways("\u041F", "\x8F", 'IBM866') # П
    655. 

  655.     check_both_ways("\u0420", "\x90", 'IBM866') # Р
    656. 

  656.     check_both_ways("\u042F", "\x9F", 'IBM866') # Я
    657. 

  657.     check_both_ways("\u0430", "\xA0", 'IBM866') # а
    658. 

  658.     check_both_ways("\u043F", "\xAF", 'IBM866') # п
    659. 

  659.     check_both_ways("\u2591", "\xB0", 'IBM866') # ░
    660. 

  660.     check_both_ways("\u2510", "\xBF", 'IBM866') # ┐
    661. 

  661.     check_both_ways("\u2514", "\xC0", 'IBM866') # └
    662. 

  662.     check_both_ways("\u2567", "\xCF", 'IBM866') # ╧
    663. 

  663.     check_both_ways("\u2568", "\xD0", 'IBM866') # ╨
    664. 

  664.     check_both_ways("\u2580", "\xDF", 'IBM866') # ▀
    665. 

  665.     check_both_ways("\u0440", "\xE0", 'IBM866') # р
    666. 

  666.     check_both_ways("\u044F", "\xEF", 'IBM866') # я
    667. 

  667.     check_both_ways("\u0401", "\xF0", 'IBM866') # Ё
    668. 

  668.     check_both_ways("\u00A0", "\xFF", 'IBM866') # non-breaking space
    669. 

  669.   end
    670. 

  670. 

  671.   def test_IBM869
    672. 

  672.     assert_raise(Encoding::UndefinedConversionError) { "\x80".encode("utf-8", 'IBM869') }
    673. 

  673.     assert_raise(Encoding::UndefinedConversionError) { "\x85".encode("utf-8", 'IBM869') }
    674. 

  674.     check_both_ways("\u0386", "\x86", 'IBM869') # Ά
    675. 

  675.     assert_raise(Encoding::UndefinedConversionError) { "\x87".encode("utf-8", 'IBM869') }
    676. 

  676.     check_both_ways("\u00B7", "\x88", 'IBM869') # ·
    677. 

  677.     check_both_ways("\u0389", "\x8F", 'IBM869') # Ή
    678. 

  678.     check_both_ways("\u038A", "\x90", 'IBM869') # Ί
    679. 

  679.     check_both_ways("\u038C", "\x92", 'IBM869') # Ό
    680. 

  680.     assert_raise(Encoding::UndefinedConversionError) { "\x93".encode("utf-8", 'IBM869') }
    681. 

  681.     assert_raise(Encoding::UndefinedConversionError) { "\x94".encode("utf-8", 'IBM869') }
    682. 

  682.     check_both_ways("\u038E", "\x95", 'IBM869') # Ύ
    683. 

  683.     check_both_ways("\u03AF", "\x9F", 'IBM869') # ί
    684. 

  684.     check_both_ways("\u03CA", "\xA0", 'IBM869') # ϊ
    685. 

  685.     check_both_ways("\u00BB", "\xAF", 'IBM869') # »
    686. 

  686.     check_both_ways("\u2591", "\xB0", 'IBM869') # ░
    687. 

  687.     check_both_ways("\u2510", "\xBF", 'IBM869') # ┐
    688. 

  688.     check_both_ways("\u2514", "\xC0", 'IBM869') # └
    689. 

  689.     check_both_ways("\u03A3", "\xCF", 'IBM869') # Σ
    690. 

  690.     check_both_ways("\u03A4", "\xD0", 'IBM869') # Τ
    691. 

  691.     check_both_ways("\u2580", "\xDF", 'IBM869') # ▀
    692. 

  692.     check_both_ways("\u03B6", "\xE0", 'IBM869') # ζ
    693. 

  693.     check_both_ways("\u0384", "\xEF", 'IBM869') # ΄
    694. 

  694.     check_both_ways("\u00AD", "\xF0", 'IBM869') # soft hyphen
    695. 

  695.     check_both_ways("\u00A0", "\xFF", 'IBM869') # non-breaking space
    696. 

  696.   end
    697. 

  697. 

  698.   def test_macCroatian
    699. 

  699.     check_both_ways("\u00C4", "\x80", 'macCroatian') # Ä
    700. 

  700.     check_both_ways("\u00E8", "\x8F", 'macCroatian') # è
    701. 

  701.     check_both_ways("\u00EA", "\x90", 'macCroatian') # ê
    702. 

  702.     check_both_ways("\u00FC", "\x9F", 'macCroatian') # ü
    703. 

  703.     check_both_ways("\u2020", "\xA0", 'macCroatian') # †
    704. 

  704.     check_both_ways("\u00D8", "\xAF", 'macCroatian') # Ø
    705. 

  705.     check_both_ways("\u221E", "\xB0", 'macCroatian') # ∞
    706. 

  706.     check_both_ways("\u00F8", "\xBF", 'macCroatian') # ø
    707. 

  707.     check_both_ways("\u00BF", "\xC0", 'macCroatian') # ¿
    708. 

  708.     check_both_ways("\u0153", "\xCF", 'macCroatian') # œ
    709. 

  709.     check_both_ways("\u0110", "\xD0", 'macCroatian') # Đ
    710. 

  710.     check_both_ways("\u00A9", "\xD9", 'macCroatian') # ©
    711. 

  711.     check_both_ways("\u2044", "\xDA", 'macCroatian') # ⁄
    712. 

  712.     check_both_ways("\u203A", "\xDD", 'macCroatian') # ›
    713. 

  713.     check_both_ways("\u00C6", "\xDE", 'macCroatian') # Æ
    714. 

  714.     check_both_ways("\u00BB", "\xDF", 'macCroatian') # »
    715. 

  715.     check_both_ways("\u2013", "\xE0", 'macCroatian') # –
    716. 

  716.     check_both_ways("\u00B7", "\xE1", 'macCroatian') # ·
    717. 

  717.     check_both_ways("\u00C2", "\xE5", 'macCroatian') # Â
    718. 

  718.     check_both_ways("\u0107", "\xE6", 'macCroatian') # ć
    719. 

  719.     check_both_ways("\u00C1", "\xE7", 'macCroatian') # Á
    720. 

  720.     check_both_ways("\u010D", "\xE8", 'macCroatian') # č
    721. 

  721.     check_both_ways("\u00C8", "\xE9", 'macCroatian') # È
    722. 

  722.     check_both_ways("\u00D4", "\xEF", 'macCroatian') # Ô
    723. 

  723.     check_both_ways("\u0111", "\xF0", 'macCroatian') # đ
    724. 

  724.     check_both_ways("\u00D2", "\xF1", 'macCroatian') # Ò
    725. 

  725.     check_both_ways("\u00AF", "\xF8", 'macCroatian') # ¯
    726. 

  726.     check_both_ways("\u03C0", "\xF9", 'macCroatian') # π
    727. 

  727.     check_both_ways("\u00CB", "\xFA", 'macCroatian') # Ë
    728. 

  728.     check_both_ways("\u00CA", "\xFD", 'macCroatian') # Ê
    729. 

  729.     check_both_ways("\u00E6", "\xFE", 'macCroatian') # æ
    730. 

  730.     check_both_ways("\u02C7", "\xFF", 'macCroatian') # ˇ
    731. 

  731.   end
    732. 

  732. 

  733.   def test_macCyrillic
    734. 

  734.     check_both_ways("\u0410", "\x80", 'macCyrillic') # А
    735. 

  735.     check_both_ways("\u041F", "\x8F", 'macCyrillic') # П
    736. 

  736.     check_both_ways("\u0420", "\x90", 'macCyrillic') # Р
    737. 

  737.     check_both_ways("\u042F", "\x9F", 'macCyrillic') # Я
    738. 

  738.     check_both_ways("\u2020", "\xA0", 'macCyrillic') # †
    739. 

  739.     check_both_ways("\u0453", "\xAF", 'macCyrillic') # ѓ
    740. 

  740.     check_both_ways("\u221E", "\xB0", 'macCyrillic') # ∞
    741. 

  741.     check_both_ways("\u045A", "\xBF", 'macCyrillic') # њ
    742. 

  742.     check_both_ways("\u0458", "\xC0", 'macCyrillic') # ј
    743. 

  743.     check_both_ways("\u0455", "\xCF", 'macCyrillic') # ѕ
    744. 

  744.     check_both_ways("\u2013", "\xD0", 'macCyrillic') # –
    745. 

  745.     check_both_ways("\u044F", "\xDF", 'macCyrillic') # я
    746. 

  746.     check_both_ways("\u0430", "\xE0", 'macCyrillic') # а
    747. 

  747.     check_both_ways("\u043F", "\xEF", 'macCyrillic') # п
    748. 

  748.     check_both_ways("\u0440", "\xF0", 'macCyrillic') # р
    749. 

  749.     check_both_ways("\u00A4", "\xFF", 'macCyrillic') # ¤
    750. 

  750.   end
    751. 

  751. 

  752.   def test_macGreek
    753. 

  753.     check_both_ways("\u00C4", "\x80", 'macGreek') # Ä
    754. 

  754.     check_both_ways("\u00E8", "\x8F", 'macGreek') # è
    755. 

  755.     check_both_ways("\u00EA", "\x90", 'macGreek') # ê
    756. 

  756.     check_both_ways("\u00FC", "\x9F", 'macGreek') # ü
    757. 

  757.     check_both_ways("\u2020", "\xA0", 'macGreek') # †
    758. 

  758.     check_both_ways("\u0393", "\xA1", 'macGreek') # Γ
    759. 

  759.     check_both_ways("\u0387", "\xAF", 'macGreek') # ·
    760. 

  760.     check_both_ways("\u0391", "\xB0", 'macGreek') # Α
    761. 

  761.     check_both_ways("\u03A9", "\xBF", 'macGreek') # Ω
    762. 

  762.     check_both_ways("\u03AC", "\xC0", 'macGreek') # ά
    763. 

  763.     check_both_ways("\u0153", "\xCF", 'macGreek') # œ
    764. 

  764.     check_both_ways("\u2013", "\xD0", 'macGreek') # –
    765. 

  765.     check_both_ways("\u038F", "\xDF", 'macGreek') # Ώ
    766. 

  766.     check_both_ways("\u03CD", "\xE0", 'macGreek') # ύ
    767. 

  767.     check_both_ways("\u03BF", "\xEF", 'macGreek') # ο
    768. 

  768.     check_both_ways("\u03C0", "\xF0", 'macGreek') # π
    769. 

  769.     check_both_ways("\u03B0", "\xFE", 'macGreek') # ΰ
    770. 

  770.     assert_raise(Encoding::UndefinedConversionError) { "\xFF".encode("utf-8", 'macGreek') }
    771. 

  771.   end
    772. 

  772. 

  773.   def test_macIceland
    774. 

  774.     check_both_ways("\u00C4", "\x80", 'macIceland') # Ä
    775. 

  775.     check_both_ways("\u00E8", "\x8F", 'macIceland') # è
    776. 

  776.     check_both_ways("\u00EA", "\x90", 'macIceland') # ê
    777. 

  777.     check_both_ways("\u00FC", "\x9F", 'macIceland') # ü
    778. 

  778.     check_both_ways("\u00DD", "\xA0", 'macIceland') # Ý
    779. 

  779.     check_both_ways("\u00D8", "\xAF", 'macIceland') # Ø
    780. 

  780.     check_both_ways("\u221E", "\xB0", 'macIceland') # ∞
    781. 

  781.     check_both_ways("\u00F8", "\xBF", 'macIceland') # ø
    782. 

  782.     check_both_ways("\u00BF", "\xC0", 'macIceland') # ¿
    783. 

  783.     check_both_ways("\u0153", "\xCF", 'macIceland') # œ
    784. 

  784.     check_both_ways("\u2013", "\xD0", 'macIceland') # –
    785. 

  785.     check_both_ways("\u00FE", "\xDF", 'macIceland') # þ
    786. 

  786.     check_both_ways("\u00FD", "\xE0", 'macIceland') # ý
    787. 

  787.     check_both_ways("\u00D4", "\xEF", 'macIceland') # Ô
    788. 

  788.     #check_both_ways("\uF8FF", "\xF0", 'macIceland') # Apple logo
    789. 

  789.     check_both_ways("\u02C7", "\xFF", 'macIceland') # ˇ
    790. 

  790.   end
    791. 

  791. 

  792.   def test_macRoman
    793. 

  793.     check_both_ways("\u00C4", "\x80", 'macRoman') # Ä
    794. 

  794.     check_both_ways("\u00E8", "\x8F", 'macRoman') # è
    795. 

  795.     check_both_ways("\u00EA", "\x90", 'macRoman') # ê
    796. 

  796.     check_both_ways("\u00FC", "\x9F", 'macRoman') # ü
    797. 

  797.     check_both_ways("\u2020", "\xA0", 'macRoman') # †
    798. 

  798.     #check_both_ways("\u00DB", "\xAF", 'macRoman') # Ø
    799. 

  799.     check_both_ways("\u221E", "\xB0", 'macRoman') # ∞
    800. 

  800.     check_both_ways("\u00F8", "\xBF", 'macRoman') # ø
    801. 

  801.     check_both_ways("\u00BF", "\xC0", 'macRoman') # ¿
    802. 

  802.     check_both_ways("\u0153", "\xCF", 'macRoman') # œ
    803. 

  803.     check_both_ways("\u2013", "\xD0", 'macRoman') # –
    804. 

  804.     check_both_ways("\u00A4", "\xDB", 'macRoman') # ¤
    805. 

  805.     check_both_ways("\uFB02", "\xDF", 'macRoman') # fl
    806. 

  806.     check_both_ways("\u2021", "\xE0", 'macRoman') # ‡
    807. 

  807.     check_both_ways("\u00D4", "\xEF", 'macRoman') # Ô
    808. 

  808.     #check_both_ways("\uF8FF", "\xF0", 'macRoman') # Apple logo
    809. 

  809.     check_both_ways("\u02C7", "\xFF", 'macRoman') # ˇ
    810. 

  810.   end
    811. 

  811. 

  812.   def test_macRomania
    813. 

  813.     check_both_ways("\u00C4", "\x80", 'macRomania') # Ä
    814. 

  814.     check_both_ways("\u00E8", "\x8F", 'macRomania') # è
    815. 

  815.     check_both_ways("\u00EA", "\x90", 'macRomania') # ê
    816. 

  816.     check_both_ways("\u00FC", "\x9F", 'macRomania') # ü
    817. 

  817.     check_both_ways("\u2020", "\xA0", 'macRomania') # †
    818. 

  818.     check_both_ways("\u015E", "\xAF", 'macRomania') # Ş
    819. 

  819.     check_both_ways("\u221E", "\xB0", 'macRomania') # ∞
    820. 

  820.     check_both_ways("\u015F", "\xBF", 'macRomania') # ş
    821. 

  821.     check_both_ways("\u00BF", "\xC0", 'macRomania') # ¿
    822. 

  822.     check_both_ways("\u0153", "\xCF", 'macRomania') # œ
    823. 

  823.     check_both_ways("\u2013", "\xD0", 'macRomania') # –
    824. 

  824.     check_both_ways("\u00A4", "\xDB", 'macRomania') # €
    825. 

  825.     check_both_ways("\u0163", "\xDF", 'macRomania') # ţ
    826. 

  826.     check_both_ways("\u2021", "\xE0", 'macRomania') # ‡
    827. 

  827.     check_both_ways("\u00D4", "\xEF", 'macRomania') # Ô
    828. 

  828.     #check_both_ways("\uF8FF", "\xF0", 'macRomania') # Apple logo
    829. 

  829.     check_both_ways("\u02C7", "\xFF", 'macRomania') # ˇ
    830. 

  830.   end
    831. 

  831. 

  832.   def test_macTurkish
    833. 

  833.     check_both_ways("\u00C4", "\x80", 'macTurkish') # Ä
    834. 

  834.     check_both_ways("\u00E8", "\x8F", 'macTurkish') # è
    835. 

  835.     check_both_ways("\u00EA", "\x90", 'macTurkish') # ê
    836. 

  836.     check_both_ways("\u00FC", "\x9F", 'macTurkish') # ü
    837. 

  837.     check_both_ways("\u2020", "\xA0", 'macTurkish') # †
    838. 

  838.     check_both_ways("\u00D8", "\xAF", 'macTurkish') # Ø
    839. 

  839.     check_both_ways("\u221E", "\xB0", 'macTurkish') # ∞
    840. 

  840.     check_both_ways("\u00F8", "\xBF", 'macTurkish') # ø
    841. 

  841.     check_both_ways("\u00BF", "\xC0", 'macTurkish') # ¿
    842. 

  842.     check_both_ways("\u0153", "\xCF", 'macTurkish') # œ
    843. 

  843.     check_both_ways("\u2013", "\xD0", 'macTurkish') # –
    844. 

  844.     check_both_ways("\u015F", "\xDF", 'macTurkish') # ş
    845. 

  845.     check_both_ways("\u2021", "\xE0", 'macTurkish') # ‡
    846. 

  846.     check_both_ways("\u00D4", "\xEF", 'macTurkish') # Ô
    847. 

  847.     #check_both_ways("\uF8FF", "\xF0", 'macTurkish') # Apple logo
    848. 

  848.     check_both_ways("\u00D9", "\xF4", 'macTurkish') # Ù
    849. 

  849.     assert_raise(Encoding::UndefinedConversionError) { "\xF5".encode("utf-8", 'macTurkish') }
    850. 

  850.     check_both_ways("\u02C6", "\xF6", 'macTurkish') # ˆ
    851. 

  851.     check_both_ways("\u02C7", "\xFF", 'macTurkish') # ˇ
    852. 

  852.   end
    853. 

  853. 

  854.   def test_macUkraine
    855. 

  855.     check_both_ways("\u0410", "\x80", 'macUkraine') # А
    856. 

  856.     check_both_ways("\u041F", "\x8F", 'macUkraine') # П
    857. 

  857.     check_both_ways("\u0420", "\x90", 'macUkraine') # Р
    858. 

  858.     check_both_ways("\u042F", "\x9F", 'macUkraine') # Я
    859. 

  859.     check_both_ways("\u2020", "\xA0", 'macUkraine') # †
    860. 

  860.     check_both_ways("\u0453", "\xAF", 'macUkraine') # ѓ
    861. 

  861.     check_both_ways("\u221E", "\xB0", 'macUkraine') # ∞
    862. 

  862.     check_both_ways("\u045A", "\xBF", 'macUkraine') # њ
    863. 

  863.     check_both_ways("\u0458", "\xC0", 'macUkraine') # ј
    864. 

  864.     check_both_ways("\u0455", "\xCF", 'macUkraine') # ѕ
    865. 

  865.     check_both_ways("\u2013", "\xD0", 'macUkraine') # –
    866. 

  866.     check_both_ways("\u044F", "\xDF", 'macUkraine') # я
    867. 

  867.     check_both_ways("\u0430", "\xE0", 'macUkraine') # а
    868. 

  868.     check_both_ways("\u043F", "\xEF", 'macUkraine') # п
    869. 

  869.     check_both_ways("\u0440", "\xF0", 'macUkraine') # р
    870. 

  870.     check_both_ways("\u00A4", "\xFF", 'macUkraine') # ¤
    871. 

  871.   end
    872. 

  872. 

  873.   def test_koi8_u
    874. 

  874.     check_both_ways("\u2500", "\x80", 'KOI8-U') # ─
    875. 

  875.     check_both_ways("\u2590", "\x8F", 'KOI8-U') # ▐
    876. 

  876.     check_both_ways("\u2591", "\x90", 'KOI8-U') # ░
    877. 

  877.     check_both_ways("\u00F7", "\x9F", 'KOI8-U') # ÷
    878. 

  878.     check_both_ways("\u2550", "\xA0", 'KOI8-U') # ═
    879. 

  879.     check_both_ways("\u0454", "\xA4", 'KOI8-U') # є
    880. 

  880.     check_both_ways("\u0456", "\xA6", 'KOI8-U') # і
    881. 

  881.     check_both_ways("\u0457", "\xA7", 'KOI8-U') # ї
    882. 

  882.     check_both_ways("\u0491", "\xAD", 'KOI8-U') # ґ
    883. 

  883.     check_both_ways("\u255E", "\xAF", 'KOI8-U') # ╞
    884. 

  884.     check_both_ways("\u255F", "\xB0", 'KOI8-U') # ╟
    885. 

  885.     check_both_ways("\u0404", "\xB4", 'KOI8-U') # Є
    886. 

  886.     check_both_ways("\u0406", "\xB6", 'KOI8-U') # І
    887. 

  887.     check_both_ways("\u0407", "\xB7", 'KOI8-U') # Ї
    888. 

  888.     check_both_ways("\u0490", "\xBD", 'KOI8-U') # Ґ
    889. 

  889.     check_both_ways("\u00A9", "\xBF", 'KOI8-U') # ©
    890. 

  890.     check_both_ways("\u044E", "\xC0", 'KOI8-U') # ю
    891. 

  891.     check_both_ways("\u043E", "\xCF", 'KOI8-U') # о
    892. 

  892.     check_both_ways("\u043F", "\xD0", 'KOI8-U') # п
    893. 

  893.     check_both_ways("\u044A", "\xDF", 'KOI8-U') # ъ
    894. 

  894.     check_both_ways("\u042E", "\xE0", 'KOI8-U') # Ю
    895. 

  895.     check_both_ways("\u041E", "\xEF", 'KOI8-U') # О
    896. 

  896.     check_both_ways("\u041F", "\xF0", 'KOI8-U') # П
    897. 

  897.     check_both_ways("\u042A", "\xFF", 'KOI8-U') # Ъ
    898. 

  898.   end
    899. 

  899. 

  900.   def test_koi8_r
    901. 

  901.     check_both_ways("\u2500", "\x80", 'KOI8-R') # ─
    902. 

  902.     check_both_ways("\u2590", "\x8F", 'KOI8-R') # ▐
    903. 

  903.     check_both_ways("\u2591", "\x90", 'KOI8-R') # ░
    904. 

  904.     check_both_ways("\u00F7", "\x9F", 'KOI8-R') # ÷
    905. 

  905.     check_both_ways("\u2550", "\xA0", 'KOI8-R') # ═
    906. 

  906.     check_both_ways("\u255E", "\xAF", 'KOI8-R') # ╞
    907. 

  907.     check_both_ways("\u255F", "\xB0", 'KOI8-R') # ╟
    908. 

  908.     check_both_ways("\u00A9", "\xBF", 'KOI8-R') # ©
    909. 

  909.     check_both_ways("\u044E", "\xC0", 'KOI8-R') # ю
    910. 

  910.     check_both_ways("\u043E", "\xCF", 'KOI8-R') # о
    911. 

  911.     check_both_ways("\u043F", "\xD0", 'KOI8-R') # п
    912. 

  912.     check_both_ways("\u044A", "\xDF", 'KOI8-R') # ъ
    913. 

  913.     check_both_ways("\u042E", "\xE0", 'KOI8-R') # Ю
    914. 

  914.     check_both_ways("\u041E", "\xEF", 'KOI8-R') # О
    915. 

  915.     check_both_ways("\u041F", "\xF0", 'KOI8-R') # П
    916. 

  916.     check_both_ways("\u042A", "\xFF", 'KOI8-R') # Ъ
    917. 

  917.   end
    918. 

  918. 

  919.   def test_TIS_620
    920. 

  920.     assert_raise(Encoding::UndefinedConversionError) { "\x80".encode("utf-8", 'TIS-620') }
    921. 

  921.     assert_raise(Encoding::UndefinedConversionError) { "\x8F".encode("utf-8", 'TIS-620') }
    922. 

  922.     assert_raise(Encoding::UndefinedConversionError) { "\x90".encode("utf-8", 'TIS-620') }
    923. 

  923.     assert_raise(Encoding::Undef…
    924.