Define a variable to distinguish system_icu from bundled_icu in Blink

LayoutTests/fast/encoding/char-decoding.html

Index: LayoutTests/fast/encoding/char-decoding.html
diff --git a/LayoutTests/fast/encoding/char-decoding.html b/LayoutTests/fast/encoding/char-decoding.html
index 5ac47ad1f94fdef4295bb8af7587ca13740d00df..38bb6599bc6049e1c180426d9db3d6e7e4a073c3 100644
--- a/LayoutTests/fast/encoding/char-decoding.html
+++ b/LayoutTests/fast/encoding/char-decoding.html
@@ -12,19 +12,29 @@ testDecode('UTF-8', '%E2%88%9A', 'U+221A');
 
 // <http://bugs.webkit.org/show_bug.cgi?id=17014> EUC-CN code A3A0 is mapped to U+E5E5 instead of U+3000

[jsbell, 2015/06/02 00:06:32]

Can we remove this comment? Looks like it passes now?

[jungshik at Google, 2015/06/02 00:20:34]

Yes, it passes because the patch for the above bug manually added back the
mapping in the code. 

I think the comment is as to why this particular mapping is tested. 
Pending the resolution of GBK vs GB18030, I have not yet updated GBK/GB18030
tables to match the encoding spec more or less. When I do, I can remove the code
to map A3A0 to U+3000 (the patch for the above webkit bug). 

 testDecode('gb2312', '%A3%A0', 'U+3000');
+testDecode('gb_2312', '%A3%A0', 'U+3000');
 testDecode('gb_2312-80', '%A3%A0', 'U+3000');
+testDecode('csgb2312', '%A3%A0', 'U+3000');
+testDecode('iso-ir-58', '%A3%A0', 'U+3000');
+testDecode('csiso58gb231280', '%A3%A0', 'U+3000');
 testDecode('chinese', '%A3%A0', 'U+3000');
 testDecode('gbk', '%A3%A0', 'U+3000');
+testDecode('x-gbk', '%A3%A0', 'U+3000');
 testDecode('gb18030', '%A3%A0', 'U+3000');
 testDecode('EUC-CN', '%A3%A0', 'U+3000');
 
 // Test Shift_JIS aliases.
 testDecode('Shift_JIS', '%82%d0', 'U+3072');
 testDecode('shift-jis', '%82%d0', 'U+3072');
+testDecode('csshiftjis', '%82%d0', 'U+3072');
+testDecode('sjis', '%82%d0', 'U+3072');
+testDecode('x-sjis', '%82%d0', 'U+3072');
+testDecode('ms_kanji', '%82%d0', 'U+3072');
+testDecode('windows-31j', '%82%d0', 'U+3072');
 
 // Test that all Korean encodings of EUC-KR family are treated as windows-949.
 var korean = {
-    encodings: ['korean', 'EUC-KR', 'windows-949', 'x-windows-949', 'x-uhc',
+    encodings: ['korean', 'EUC-KR', 'windows-949', 'cseuckr', 'csksc56011987',
                 'iso-ir-149', 'KS_C_5601-1987', 'KS_C_5601-1989',
                 'KSC5601', 'KSC_5601'],
     encoded: ['%A2%E6', '%A1%A4', '%A1%A9', '%A1%AA', '%A1%AD', '%A2%A6',