From c910388587220e962682b0b9187e79b8f1641c17 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafael=20Mendon=C3=A7a=20Fran=C3=A7a?=
 <rafaelmfranca@gmail.com>
Date: Wed, 8 May 2013 21:03:37 -0300
Subject: Revert "Remove unicode character encoding from
 ActiveSupport::JSON.encode"

This reverts commit 815a9431ab61376a7e8e1bdff21f87bc557992f8.

Conflicts:
	activesupport/test/json/encoding_test.rb

Reason: This was causing a regression where the resulting string is always
returning UTF-8. Also it changes the behavior of this method on a stable release.

Fixes #9498.
---
 activesupport/CHANGELOG.md                        | 14 ++++++-------
 activesupport/lib/active_support/json/encoding.rb |  8 +++++++-
 activesupport/test/json/encoding_test.rb          | 24 +++++++++--------------
 3 files changed, 23 insertions(+), 23 deletions(-)

(limited to 'activesupport')

diff --git a/activesupport/CHANGELOG.md b/activesupport/CHANGELOG.md
index 649766cc9b..2cbdfb015b 100644
--- a/activesupport/CHANGELOG.md
+++ b/activesupport/CHANGELOG.md
@@ -1,5 +1,12 @@
 ## unreleased ##
 
+*   Revert the changes on unicode character encoding from `ActiveSupport::JSON.encode`.
+    This was causing a regression where the resulting string is always returning UTF-8.
+    Also it changes the behavior of this method on a stable release.
+    Fixes #9498.
+
+    *Rafael Mendonça França*
+
 *   Fix `ActiveSupport::TimeZone.parse` when time is at a local DST jump.
     Fixes #9678.
 
@@ -13,13 +20,6 @@
 
     *Dan Kubb*
 
-*   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*
-
 *   Fix mocha v0.13.0 compatibility. *James Mead*
 
 *   `#as_json` isolates options when encoding a hash. [Backport #8185]
diff --git a/activesupport/lib/active_support/json/encoding.rb b/activesupport/lib/active_support/json/encoding.rb
index a50e6524c6..bd2f909ca9 100644
--- a/activesupport/lib/active_support/json/encoding.rb
+++ b/activesupport/lib/active_support/json/encoding.rb
@@ -122,7 +122,13 @@ module ActiveSupport
           if string.respond_to?(:force_encoding)
             string = string.encode(::Encoding::UTF_8, :undef => :replace).force_encoding(::Encoding::BINARY)
           end
-          json = string.gsub(escape_regex) { |s| ESCAPED_CHARS[s] }
+          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 = %("#{json}")
           json.force_encoding(::Encoding::UTF_8) if json.respond_to?(:force_encoding)
           json
diff --git a/activesupport/test/json/encoding_test.rb b/activesupport/test/json/encoding_test.rb
index e4e13c3f25..67b648e6db 100644
--- a/activesupport/test/json/encoding_test.rb
+++ b/activesupport/test/json/encoding_test.rb
@@ -100,11 +100,11 @@ class TestJSONEncoding < Test::Unit::TestCase
   def test_utf8_string_encoded_properly_when_kcode_is_utf8
     with_kcode 'UTF8' do
       result = ActiveSupport::JSON.encode('€2.99')
-      assert_equal '"€2.99"', result
+      assert_equal '"\\u20ac2.99"', result
       assert_equal(Encoding::UTF_8, result.encoding) if result.respond_to?(:encoding)
 
       result = ActiveSupport::JSON.encode('✎☺')
-      assert_equal '"✎☺"', result
+      assert_equal '"\\u270e\\u263a"', result
       assert_equal(Encoding::UTF_8, result.encoding) if result.respond_to?(:encoding)
     end
   end
@@ -113,22 +113,16 @@ class TestJSONEncoding < Test::Unit::TestCase
     def test_non_utf8_string_transcodes
       s = '二'.encode('Shift_JIS')
       result = ActiveSupport::JSON.encode(s)
-      assert_equal '"二"', result
+      assert_equal '"\\u4e8c"', result
       assert_equal Encoding::UTF_8, result.encoding
     end
-  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']
+    def test_utf8_hash_key
+      w = { '𠜎' => 'a' }
+      result = ActiveSupport::JSON.encode(w)
+      assert_equal '{"\\u070e":"a"}', result
+      assert_equal Encoding::US_ASCII, result.encoding
+    end
   end
 
   def test_exception_raised_when_encoding_circular_reference_in_array
-- 
cgit v1.2.3