Encoding API: Use 'windows-1252' as name for 'us-ascii' 'iso-8859-1'

Source/modules/encoding/TextDecoder.cpp

Index: Source/modules/encoding/TextDecoder.cpp
diff --git a/Source/modules/encoding/TextDecoder.cpp b/Source/modules/encoding/TextDecoder.cpp
index 69d37d050a4bf4165db0af3150655c7f24844327..a81f60f2c901f4630d2cf849c5554a05432d8899 100644
--- a/Source/modules/encoding/TextDecoder.cpp
+++ b/Source/modules/encoding/TextDecoder.cpp
@@ -68,7 +68,11 @@ TextDecoder::~TextDecoder()
 
 String TextDecoder::encoding() const
 {
-    return String(m_encoding.name()).lower();
+    String name = String(m_encoding.name()).lower();
+    // The same codec is used, but WTF maintains a different name/identity for these.
+    if (name == "iso-8859-1" || name == "us-ascii")

[jsbell, 2013/08/27 00:21:04]

This could also be done around line 44, during the construction. That would
ensure that even if Blink diverges from having these all use the same codec
class (TextCodecLatin1) they will still all use windows-1252. On the other hand,
aliasing here means we'd actually exercise WTF's aliasing, so I'm leaning
towards keeping it here.

[Ken Russell (switch to Gerrit), 2013/08/27 03:32:54]

I defer to your judgment. If the number of aliases grows large in the future, I
assume an actual data structure will be used.

+        return "windows-1252";
+    return name;
 }
 
 String TextDecoder::decode(ArrayBufferView* input, const Dictionary& options, ExceptionState& es)