diff options
| -rw-r--r-- | activesupport/CHANGELOG.md | 6 | ||||
| -rw-r--r-- | activesupport/lib/active_support/json/encoding.rb | 8 | ||||
| -rw-r--r-- | activesupport/test/json/encoding_test.rb | 19 |
3 files changed, 23 insertions, 10 deletions
diff --git a/activesupport/CHANGELOG.md b/activesupport/CHANGELOG.md index 82db1bb6f9..5e7121181d 100644 --- a/activesupport/CHANGELOG.md +++ b/activesupport/CHANGELOG.md @@ -1,4 +1,10 @@ ## Rails 4.0.0 (unreleased) ## +* Remove surrogate unicode character encoding from ActiveSupport::JSON.encode + The encoding scheme was broken for unicode characters outside the basic multilingual plane; + since json is assumed to be UTF-8, and we already force the encoding to UTF-8 simply pass through + the un-encoded characters. + + *Brett Carter* * Deprecate `Time.time_with_date_fallback`, `Time.utc_time` and `Time.local_time`. These methods were added to handle the limited range of Ruby's native Time diff --git a/activesupport/lib/active_support/json/encoding.rb b/activesupport/lib/active_support/json/encoding.rb index 7a5c351ca8..832d1ce6d5 100644 --- a/activesupport/lib/active_support/json/encoding.rb +++ b/activesupport/lib/active_support/json/encoding.rb @@ -129,13 +129,7 @@ module ActiveSupport def escape(string) string = string.encode(::Encoding::UTF_8, :undef => :replace).force_encoding(::Encoding::BINARY) - json = string. - gsub(escape_regex) { |s| ESCAPED_CHARS[s] }. - gsub(/([\xC0-\xDF][\x80-\xBF]| - [\xE0-\xEF][\x80-\xBF]{2}| - [\xF0-\xF7][\x80-\xBF]{3})+/nx) { |s| - s.unpack("U*").pack("n*").unpack("H*")[0].gsub(/.{4}/n, '\\\\u\&') - } + json = string.gsub(escape_regex) { |s| ESCAPED_CHARS[s] } json = %("#{json}") json.force_encoding(::Encoding::UTF_8) json diff --git a/activesupport/test/json/encoding_test.rb b/activesupport/test/json/encoding_test.rb index 5bb2a45c87..b6e2cd4529 100644 --- a/activesupport/test/json/encoding_test.rb +++ b/activesupport/test/json/encoding_test.rb @@ -112,21 +112,34 @@ class TestJSONEncoding < ActiveSupport::TestCase def test_utf8_string_encoded_properly result = ActiveSupport::JSON.encode('€2.99') - assert_equal '"\\u20ac2.99"', result + assert_equal '"€2.99"', result assert_equal(Encoding::UTF_8, result.encoding) result = ActiveSupport::JSON.encode('✎☺') - assert_equal '"\\u270e\\u263a"', result + assert_equal '"✎☺"', result assert_equal(Encoding::UTF_8, result.encoding) end def test_non_utf8_string_transcodes s = '二'.encode('Shift_JIS') result = ActiveSupport::JSON.encode(s) - assert_equal '"\\u4e8c"', result + assert_equal '"二"', result assert_equal Encoding::UTF_8, result.encoding end + def test_wide_utf8_chars + w = '𠜎' + result = ActiveSupport::JSON.encode(w) + assert_equal '"𠜎"', result + end + + def test_wide_utf8_roundtrip + hash = { string: "𐒑" } + json = ActiveSupport::JSON.encode(hash) + decoded_hash = ActiveSupport::JSON.decode(json) + assert_equal "𐒑", decoded_hash['string'] + end + def test_exception_raised_when_encoding_circular_reference_in_array a = [1] a << a |