Canonical name change: windows-949 => EUC-KR

Source/wtf/text/TextCodecICU.cpp

 /*
  * Copyright (C) 2004, 2006, 2007, 2008, 2011 Apple Inc. All rights reserved.
  * Copyright (C) 2006 Alexey Proskuryakov <ap@nypop.com>
  *
  * Redistribution and use in source and binary forms, with or without
  * modification, are permitted provided that the following conditions
  * are met:
  * 1. Redistributions of source code must retain the above copyright
  *    notice, this list of conditions and the following disclaimer.
  * 2. Redistributions in binary form must reproduce the above copyright
@@ skipping 66 matching lines @@
         const char* standardName = ucnv_getStandardName(name, "MIME", &error);
         if (!U_SUCCESS(error) || !standardName) {
             error = U_ZERO_ERROR;
             // Try IANA to pick up 'windows-12xx' and other names
             // which are not preferred MIME names but are widely used.
             standardName = ucnv_getStandardName(name, "IANA", &error);
             if (!U_SUCCESS(error) || !standardName)
                 continue;
         }
 
+        // A number of these aliases are handled in Chrome's copy of ICU, but
+        // Chromium can be compiled with the system ICU.
+
         // 1. Treat GB2312 encoding as GBK (its more modern superset), to match other browsers.
         // 2. On the Web, GB2312 is encoded as EUC-CN or HZ, while ICU provides a native encoding
         //    for encoding GB_2312-80 and several others. So, we need to override this behavior, too.
-        if (strcmp(standardName, "GB2312") == 0 || strcmp(standardName, "GB_2312-80") == 0)
+        if (!strcmp(standardName, "GB2312") || !strcmp(standardName, "GB_2312-80"))
             standardName = "GBK";
-        // Similarly, EUC-KR encodings all map to an extended version.
-        else if (strcmp(standardName, "KSC_5601") == 0 || strcmp(standardName, "EUC-KR") == 0 || strcmp(standardName, "cp1363") == 0)
-            standardName = "windows-949";
+        // Similarly, EUC-KR encodings all map to an extended version, but
+        // per HTML5, the canonical name still should be EUC-KR.
+        else if (!strcmp(standardName, "EUC-KR") || !strcmp(standardName, "KSC_5601") || !strcmp(standardName, "cp1363"))
+            standardName = "EUC-KR";
         // And so on.
-        else if (strcasecmp(standardName, "iso-8859-9") == 0) // This name is returned in different case by ICU 3.2 and 3.6.
+        else if (!strcasecmp(standardName, "iso-8859-9")) // This name is returned in different case by ICU 3.2 and 3.6.
             standardName = "windows-1254";
-        else if (strcmp(standardName, "TIS-620") == 0)
+        else if (!strcmp(standardName, "TIS-620"))
             standardName = "windows-874";
 
         registrar(standardName, standardName);
 
         uint16_t numAliases = ucnv_countAliases(name, &error);
         ASSERT(U_SUCCESS(error));
         if (U_SUCCESS(error))
             for (uint16_t j = 0; j < numAliases; ++j) {
                 error = U_ZERO_ERROR;
                 const char* alias = ucnv_getAlias(name, j, &error);
@@ skipping 31 matching lines @@
     registrar("iso8859-11", "windows-874");
     registrar("dos-874", "windows-874");
     registrar("wingreek", "windows-1253");
     registrar("winhebrew", "windows-1255");
     registrar("winlatin2", "windows-1250");
     registrar("winturkish", "windows-1254");
     registrar("winvietnamese", "windows-1258");
     registrar("x-cp1250", "windows-1250");
     registrar("x-cp1251", "windows-1251");
     registrar("x-euc", "EUC-JP");
-    registrar("x-windows-949", "windows-949");
-    registrar("KSC5601", "KSC_5601");
-    registrar("x-uhc", "windows-949");
+    registrar("x-windows-949", "EUC-KR");
+    registrar("KSC5601", "EUC-KR");
+    registrar("x-uhc", "EUC-KR");
     registrar("shift-jis", "Shift_JIS");
 
     // These aliases are present in modern versions of ICU, but use different codecs, and have no standard names.
     // They are not present in ICU 3.2.
     registrar("dos-720", "cp864");
     registrar("jis7", "ISO-2022-JP");
 
     // Alternative spelling of ISO encoding names.
     registrar("ISO8859-1", "ISO-8859-1");
     registrar("ISO8859-2", "ISO-8859-2");
@@ skipping 163 matching lines @@
         do {
             decodeToBuffer(buffer, bufferLimit, source, sourceLimit, offsets, true, err);
         } while (source < sourceLimit);
         sawError = true;
     }
 
     String resultString = result.toString();
 
     // <http://bugs.webkit.org/show_bug.cgi?id=17014>
     // Simplified Chinese pages use the code A3A0 to mean "full-width space", but ICU decodes it as U+E5E5.
-    if (strcmp(m_encoding.name(), "GBK") == 0 || strcasecmp(m_encoding.name(), "gb18030") == 0)
+    if (!strcmp(m_encoding.name(), "GBK") || !strcasecmp(m_encoding.name(), "gb18030"))
         resultString.replace(0xE5E5, ideographicSpace);
 
     return resultString;
 }
 
 // We need to apply these fallbacks ourselves as they are not currently supported by ICU and
 // they were provided by the old TEC encoding path. Needed to fix <rdar://problem/4708689>.
 static UChar fallbackForGBK(UChar32 character)
 {
     switch (character) {
@@ skipping 168 matching lines @@
 {
     return encodeCommon(characters, length, handling);
 }
 
 CString TextCodecICU::encode(const LChar* characters, size_t length, UnencodableHandling handling)
 {
     return encodeCommon(characters, length, handling);
 }
 
 } // namespace WTF