Use UTS46 API to convert IDN in Unicode to punycode

Source/core/html/forms/EmailInputType.cpp

 /*
  * This file is part of the WebKit project.
  *
  * Copyright (C) 2009 Michelangelo De Simone <micdesim@gmail.com>
  * Copyright (C) 2010 Google Inc. All rights reserved.
  *
  * This library is free software; you can redistribute it and/or
  * modify it under the terms of the GNU Library General Public
  * License as published by the Free Software Foundation; either
  * version 2 of the License, or (at your option) any later version.
@@ skipping 16 matching lines @@
 #include "InputTypeNames.h"
 #include "bindings/v8/ScriptRegexp.h"
 #include "core/html/HTMLInputElement.h"
 #include "core/html/parser/HTMLParserIdioms.h"
 #include "core/page/Chrome.h"
 #include "core/page/ChromeClient.h"
 #include "platform/text/PlatformLocale.h"
 #include "public/platform/Platform.h"
 #include "wtf/PassOwnPtr.h"
 #include "wtf/text/StringBuilder.h"
-#include <unicode/uidna.h>
+#include <unicode/idna.h>
+#include <unicode/unistr.h>
 
 namespace WebCore {
 
 using blink::WebLocalizedString;
 
 // http://www.whatwg.org/specs/web-apps/current-work/multipage/states-of-the-type-attribute.html#valid-e-mail-address
 static const char localPartCharacters[] = "abcdefghijklmnopqrstuvwxyz0123456789!#$%&'*+/=?^_`{|}~.-";
 static const char emailPattern[] =
     "[a-z0-9!#$%&'*+/=?^_`{|}~.-]+" // local part
     "@"
     "[a-z0-9](?:[a-z0-9-]{0,61}[a-z0-9])?" // domain part
     "(?:\\.[a-z0-9]([a-z0-9-]{0,61}[a-z0-9])?)*";
 
 // RFC5321 says the maximum total length of a domain name is 255 octets.
-static const size_t maximumDomainNameLength = 255;
-static const int32_t idnaConversionOption = UIDNA_ALLOW_UNASSIGNED;
+static const int32_t maximumDomainNameLength = 255;
+// Use the same option as in url/url_canon_icu.cc
+static const int32_t idnaConversionOption = UIDNA_CHECK_BIDI;
 
 static String convertEmailAddressToASCII(const String& address)
 {
     if (address.containsOnlyASCII())
         return address;
 
     size_t atPosition = address.find('@');
     if (atPosition == kNotFound)
         return address;
 
-    UErrorCode error = U_ZERO_ERROR;
-    UChar domainNameBuffer[maximumDomainNameLength];
-    int32_t domainNameLength = uidna_IDNToASCII(address.charactersWithNullTermination().data() + atPosition + 1, address.length() - atPosition - 1, domainNameBuffer, WTF_ARRAY_LENGTH(domainNameBuffer), idnaConversionOption, 0, &error);
-    if (error != U_ZERO_ERROR || domainNameLength <= 0)
+    // UnicodeString ctor for copy-on-write does not work reliably (in debug
+    // build.) TODO(jshin): In an unlikely case this is a perf-issue, treat
+    // 8bit and non-8bit strings separately.
+    icu::UnicodeString idnDomainName(address.charactersWithNullTermination().data() + atPosition + 1, address.length() - atPosition - 1);
+    icu::UnicodeString domainName;
+
+    // Leak |idna| at the end.
+    UErrorCode errorCode = U_ZERO_ERROR;
+    static icu::IDNA *idna = icu::IDNA::createUTS46Instance(idnaConversionOption, errorCode);
+    ASSERT(idna);
+    icu::IDNAInfo idnaInfo;
+    idna->nameToASCII(idnDomainName, domainName, idnaInfo, errorCode);
+    if (U_FAILURE(errorCode) || idnaInfo.hasErrors() || domainName.length() > maximumDomainNameLength)
         return address;
 
     StringBuilder builder;
     builder.append(address, 0, atPosition + 1);
-    builder.append(domainNameBuffer, domainNameLength);
+    builder.append(domainName.getBuffer(), domainName.length());
     return builder.toString();
 }
 
 String EmailInputType::convertEmailAddressToUnicode(const String& address) const
 {
     if (!address.containsOnlyASCII())
         return address;
 
     size_t atPosition = address.find('@');
     if (atPosition == kNotFound)
@@ skipping 198 matching lines @@
     builder.reserveCapacity(value.length());
     for (size_t i = 0; i < addresses.size(); ++i) {
         if (i > 0)
             builder.append(",");
         builder.append(convertEmailAddressToUnicode(addresses[i]));
     }
     return builder.toString();
 }
 
 } // namespace WebCore