diff options
Diffstat (limited to 'test/ruby/test_transcode.rb')
| -rw-r--r-- | test/ruby/test_transcode.rb | 70 |
1 files changed, 64 insertions, 6 deletions
diff --git a/test/ruby/test_transcode.rb b/test/ruby/test_transcode.rb index f405877dd5..24ee9b9533 100644 --- a/test/ruby/test_transcode.rb +++ b/test/ruby/test_transcode.rb @@ -126,6 +126,28 @@ class TestTranscode < Test::Unit::TestCase assert_equal("D\xFCrst".force_encoding('iso-8859-2'), "D\xFCrst".encode('iso-8859-2', 'iso-8859-1')) end + def test_encode_xml_multibyte + encodings = %w'UTF-8 UTF-16LE UTF-16BE UTF-32LE UTF-32BE' + encodings.each do |src_enc| + encodings.each do |dst_enc| + escaped = "<>".encode(src_enc).encode(dst_enc, :xml=>:text) + assert_equal("<>", escaped.encode('UTF-8'), "failed encoding #{src_enc} to #{dst_enc} with xml: :text") + + escaped = '<">'.encode(src_enc).encode(dst_enc, :xml=>:attr) + assert_equal('"<">"', escaped.encode('UTF-8'), "failed encoding #{src_enc} to #{dst_enc} with xml: :attr") + + escaped = "<>".encode(src_enc).force_encoding("UTF-8").encode(dst_enc, src_enc, :xml=>:text) + assert_equal("<>", escaped.encode('UTF-8'), "failed encoding #{src_enc} to #{dst_enc} with xml: :text") + + escaped = '<">'.encode(src_enc).force_encoding("UTF-8").encode(dst_enc, src_enc, :xml=>:attr) + assert_equal('"<">"', escaped.encode('UTF-8'), "failed encoding #{src_enc} to #{dst_enc} with xml: :attr") + end + end + # regression test; U+6E7F (湿) uses the same bytes in ISO-2022-JP as "<>" + assert_equal( "<>\u6E7F", "<>\u6E7F".encode("ISO-2022-JP").encode("ISO-2022-JP", :xml=>:text).encode("UTF-8")) + assert_equal("\"<>\u6E7F\"", "<>\u6E7F".encode("ISO-2022-JP").encode("ISO-2022-JP", :xml=>:attr).encode("UTF-8")) + end + def test_ascii_range encodings = [ 'US-ASCII', 'ASCII-8BIT', @@ -469,6 +491,25 @@ class TestTranscode < Test::Unit::TestCase check_both_ways("\u00A0", "\xFF", 'IBM437') # non-breaking space end + def test_IBM720 + assert_raise(Encoding::UndefinedConversionError) { "\x80".encode("utf-8", 'IBM720') } + assert_raise(Encoding::UndefinedConversionError) { "\x8F".encode("utf-8", 'IBM720') } + assert_raise(Encoding::UndefinedConversionError) { "\x90".encode("utf-8", 'IBM720') } + check_both_ways("\u0627", "\x9F", 'IBM720') # ا + check_both_ways("\u0628", "\xA0", 'IBM720') # ب + check_both_ways("\u00BB", "\xAF", 'IBM720') # » + check_both_ways("\u2591", "\xB0", 'IBM720') # ░ + check_both_ways("\u2510", "\xBF", 'IBM720') # ┐ + check_both_ways("\u2514", "\xC0", 'IBM720') # └ + check_both_ways("\u2567", "\xCF", 'IBM720') # ╧ + check_both_ways("\u2568", "\xD0", 'IBM720') # ╨ + check_both_ways("\u2580", "\xDF", 'IBM720') # ▀ + check_both_ways("\u0636", "\xE0", 'IBM720') # ض + check_both_ways("\u064A", "\xEF", 'IBM720') # ي + check_both_ways("\u2261", "\xF0", 'IBM720') # ≡ + check_both_ways("\u00A0", "\xFF", 'IBM720') # non-breaking space + end + def test_IBM775 check_both_ways("\u0106", "\x80", 'IBM775') # Ć check_both_ways("\u00C5", "\x8F", 'IBM775') # Å @@ -2183,12 +2224,20 @@ class TestTranscode < Test::Unit::TestCase assert_equal("U+3042", "\u{3042}".encode("US-ASCII", fallback: fallback.method(:escape))) end - bug8940 = '[ruby-core:57318] [Bug #8940]' - %w[UTF-32 UTF-16].each do |enc| - define_method("test_pseudo_encoding_inspect(#{enc})") do - assert_normal_exit("'aaa'.encode('#{enc}').inspect", bug8940) - assert_equal(4, 'aaa'.encode(enc).length, "should count in #{enc} with BOM") + def test_fallback_aref + fallback = Object.new + def fallback.[](x) + "U+%.4X" % x.unpack("U") end + assert_equal("U+3042", "\u{3042}".encode("US-ASCII", fallback: fallback)) + end + + def test_pseudo_encoding_inspect + s = 'aaa'.encode "UTF-16" + assert_equal '"\xFE\xFF\x00\x61\x00\x61\x00\x61"', s.inspect + + s = 'aaa'.encode "UTF-32" + assert_equal '"\x00\x00\xFE\xFF\x00\x00\x00\x61\x00\x00\x00\x61\x00\x00\x00\x61"', s.inspect end def test_encode_with_invalid_chars @@ -2242,12 +2291,21 @@ class TestTranscode < Test::Unit::TestCase "#{bug} coderange should not have side effects") end - def test_universal_newline + def test_newline_options bug11324 = '[ruby-core:69841] [Bug #11324]' usascii = Encoding::US_ASCII s = "A\nB\r\nC".force_encoding(usascii) assert_equal("A\nB\nC", s.encode(usascii, universal_newline: true), bug11324) assert_equal("A\nB\nC", s.encode(usascii, universal_newline: true, undef: :replace), bug11324) assert_equal("A\nB\nC", s.encode(usascii, universal_newline: true, undef: :replace, replace: ''), bug11324) + assert_equal("A\nB\nC", s.encode(usascii, newline: :universal)) + assert_equal("A\nB\nC", s.encode(usascii, newline: :universal, undef: :replace)) + assert_equal("A\nB\nC", s.encode(usascii, newline: :universal, undef: :replace, replace: '')) + assert_equal("A\rB\r\rC", s.encode(usascii, cr_newline: true)) + assert_equal("A\rB\r\rC", s.encode(usascii, newline: :cr)) + assert_equal("A\r\nB\r\r\nC", s.encode(usascii, crlf_newline: true)) + assert_equal("A\r\nB\r\r\nC", s.encode(usascii, newline: :crlf)) + assert_equal("A\nB\nC", s.encode(usascii, lf_newline: true)) + assert_equal("A\nB\nC", s.encode(usascii, newline: :lf)) end end |