diff options
Diffstat (limited to 'test')
| -rw-r--r-- | test/ruby/test_transcode.rb | 22 |
1 files changed, 22 insertions, 0 deletions
diff --git a/test/ruby/test_transcode.rb b/test/ruby/test_transcode.rb index 04c8248697..c8b0034e06 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', |