diff options
author | usa <usa@b2dd03c8-39d4-4d8f-98ff-823fe69b080e> | 2019-08-26 16:44:03 +0000 |
---|---|---|
committer | usa <usa@b2dd03c8-39d4-4d8f-98ff-823fe69b080e> | 2019-08-26 16:44:03 +0000 |
commit | f8efd7f9e852adda9fddb5682360bde263352e02 (patch) | |
tree | f40db02b1bec803fd8e4ae0c44394fadacfc65d3 | |
parent | ef626cd09c448df9193c8d5f40143317d0185fd9 (diff) |
merge revision(s) 5b1bf8dd2d08ae7371ecf025967376bb794ed651: [Backport #16099]
UTF LE is fixed at least the first 2 bytes
* io.c (io_strip_bom): if the first 2 bytes are 0xFF0xFE, it
should be a little-endian UTF, 16 or 32. [Bug #16099]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_5@67772 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-rw-r--r-- | io.c | 7 | ||||
-rw-r--r-- | test/ruby/test_file.rb | 2 | ||||
-rw-r--r-- | test/ruby/test_io_m17n.rb | 10 | ||||
-rw-r--r-- | version.h | 2 |
4 files changed, 9 insertions, 12 deletions
@@ -6025,12 +6025,9 @@ io_strip_bom(VALUE io) return ENCINDEX_UTF_32LE; } rb_io_ungetbyte(io, b4); - rb_io_ungetbyte(io, b3); - } - else { - rb_io_ungetbyte(io, b3); - return ENCINDEX_UTF_16LE; } + rb_io_ungetbyte(io, b3); + return ENCINDEX_UTF_16LE; } rb_io_ungetbyte(io, b2); break; diff --git a/test/ruby/test_file.rb b/test/ruby/test_file.rb index 9a795ee02f..10bfbd9ae0 100644 --- a/test/ruby/test_file.rb +++ b/test/ruby/test_file.rb @@ -87,7 +87,7 @@ class TestFile < Test::Unit::TestCase end def test_bom_32le - assert_bom(["\xFF\xFE\0", "\0"], __method__) + assert_bom(["\xFF", "\xFE\0\0"], __method__) end def test_truncate_wbuf diff --git a/test/ruby/test_io_m17n.rb b/test/ruby/test_io_m17n.rb index 8ba34845fa..9ff5307fc3 100644 --- a/test/ruby/test_io_m17n.rb +++ b/test/ruby/test_io_m17n.rb @@ -2082,14 +2082,14 @@ EOT def test_strip_bom with_tmpdir { - text = "\uFEFFa" - stripped = "a" + text = "\uFEFF\u0100a" + stripped = "\u0100a" %w/UTF-8 UTF-16BE UTF-16LE UTF-32BE UTF-32LE/.each do |name| path = '%s-bom.txt' % name content = text.encode(name) generate_file(path, content) result = File.read(path, mode: 'rb:BOM|UTF-8') - assert_equal(content[1].force_encoding("ascii-8bit"), + assert_equal(content[1..-1].force_encoding("ascii-8bit"), result.force_encoding("ascii-8bit")) result = File.read(path, mode: 'rb:BOM|UTF-8:UTF-8') assert_equal(Encoding::UTF_8, result.encoding) @@ -2099,10 +2099,10 @@ EOT bug3407 = '[ruby-core:30641]' path = 'UTF-8-bom.txt' result = File.read(path, encoding: 'BOM|UTF-8') - assert_equal("a", result.force_encoding("ascii-8bit"), bug3407) + assert_equal(stripped.b, result.force_encoding("ascii-8bit"), bug3407) bug8323 = '[ruby-core:54563] [Bug #8323]' - expected = "a\xff".force_encoding("utf-8") + expected = (stripped.b + "\xff").force_encoding("utf-8") open(path, 'ab') {|f| f.write("\xff")} result = File.read(path, encoding: 'BOM|UTF-8') assert_not_predicate(result, :valid_encoding?, bug8323) @@ -1,6 +1,6 @@ #define RUBY_VERSION "2.5.6" #define RUBY_RELEASE_DATE "2019-08-27" -#define RUBY_PATCHLEVEL 190 +#define RUBY_PATCHLEVEL 191 #define RUBY_RELEASE_YEAR 2019 #define RUBY_RELEASE_MONTH 8 |