Diff
Modified: trunk/Source/_javascript_Core/ChangeLog (134172 => 134173)
--- trunk/Source/_javascript_Core/ChangeLog 2012-11-11 13:12:54 UTC (rev 134172)
+++ trunk/Source/_javascript_Core/ChangeLog 2012-11-11 14:17:46 UTC (rev 134173)
@@ -1,3 +1,15 @@
+2012-11-11 Kenichi Ishibashi <[email protected]>
+
+ WTFString::utf8() should have a mode of conversion to use replacement character
+ https://bugs.webkit.org/show_bug.cgi?id=101678
+
+ Reviewed by Alexey Proskuryakov.
+
+ Follow the change on String::utf8()
+
+ * runtime/JSGlobalObjectFunctions.cpp:
+ (JSC::encode): Pass String::StrictConversion instead of true to String::utf8().
+
2012-11-10 Filip Pizlo <[email protected]>
DFG should optimize out the NaN check on loads from double arrays if the array prototype chain is having a great time
Modified: trunk/Source/_javascript_Core/runtime/JSGlobalObjectFunctions.cpp (134172 => 134173)
--- trunk/Source/_javascript_Core/runtime/JSGlobalObjectFunctions.cpp 2012-11-11 13:12:54 UTC (rev 134172)
+++ trunk/Source/_javascript_Core/runtime/JSGlobalObjectFunctions.cpp 2012-11-11 14:17:46 UTC (rev 134173)
@@ -52,7 +52,7 @@
static JSValue encode(ExecState* exec, const char* doNotEscape)
{
- CString cstr = exec->argument(0).toString(exec)->value(exec).utf8(true);
+ CString cstr = exec->argument(0).toString(exec)->value(exec).utf8(String::StrictConversion);
if (!cstr.data())
return throwError(exec, createURIError(exec, ASCIILiteral("String contained an illegal UTF-16 sequence.")));
Modified: trunk/Source/WTF/ChangeLog (134172 => 134173)
--- trunk/Source/WTF/ChangeLog 2012-11-11 13:12:54 UTC (rev 134172)
+++ trunk/Source/WTF/ChangeLog 2012-11-11 14:17:46 UTC (rev 134173)
@@ -1,3 +1,24 @@
+2012-11-11 Kenichi Ishibashi <[email protected]>
+
+ WTFString::utf8() should have a mode of conversion to use replacement character
+ https://bugs.webkit.org/show_bug.cgi?id=101678
+
+ Reviewed by Alexander Pavlov.
+
+ Introduce conversion mode to String::utf8().
+ There are three conversion modes; lenient mode, strict mode, and
+ "replacing unpaired surrogates with the replacement character" (replacement) mode.
+ Lenient mode converts unpaired surrogates. Strict mode fails when there is an unpaired
+ surrogates and returns CString(). Replacement mode replaces unpaired surrogates with
+ the replacement character(U+FFFD). Replacement mode implements the algorithm defined at
+ http://dev.w3.org/2006/webapi/WebIDL/#dfn-obtain-unicode. WebSocket::send() requires
+ this algorithm to encode a string to utf-8.
+
+ * wtf/text/WTFString.cpp:
+ (WTF::String::utf8): Changed to take ConversionMode as the argument.
+ * wtf/text/WTFString.h:
+ (String):
+
2012-11-09 Alexei Filippov <[email protected]>
Web Inspector: Fix heap snapshots counted several times by NMI
Modified: trunk/Source/WTF/wtf/text/WTFString.cpp (134172 => 134173)
--- trunk/Source/WTF/wtf/text/WTFString.cpp 2012-11-11 13:12:54 UTC (rev 134172)
+++ trunk/Source/WTF/wtf/text/WTFString.cpp 2012-11-11 14:17:46 UTC (rev 134173)
@@ -32,6 +32,7 @@
#include <wtf/StringExtras.h>
#include <wtf/Vector.h>
#include <wtf/dtoa.h>
+#include <wtf/unicode/CharacterNames.h>
#include <wtf/unicode/UTF8.h>
#include <wtf/unicode/Unicode.h>
@@ -753,7 +754,7 @@
*buffer++ = static_cast<char>((ch & 0x3F) | 0x80);
}
-CString String::utf8(bool strict) const
+CString String::utf8(ConversionMode mode) const
{
unsigned length = this->length();
@@ -784,26 +785,48 @@
} else {
const UChar* characters = this->characters16();
- ConversionResult result = convertUTF16ToUTF8(&characters, characters + length, &buffer, buffer + bufferVector.size(), strict);
- ASSERT(result != targetExhausted); // (length * 3) should be sufficient for any conversion
+ if (mode == StrictConversionReplacingUnpairedSurrogatesWithFFFD) {
+ const UChar* charactersEnd = characters + length;
+ char* bufferEnd = buffer + bufferVector.size();
+ while (characters < charactersEnd) {
+ // Use strict conversion to detect unpaired surrogates.
+ ConversionResult result = convertUTF16ToUTF8(&characters, charactersEnd, &buffer, bufferEnd, true);
+ ASSERT(result != targetExhausted);
+ // Conversion fails when there is an unpaired surrogate.
+ // Put replacement character (U+FFFD) instead of the unpaired surrogate.
+ if (result != conversionOK) {
+ ASSERT((0xD800 <= *characters && *characters <= 0xDFFF));
+ // There should be room left, since one UChar hasn't been converted.
+ ASSERT((buffer + 3) <= bufferEnd);
+ putUTF8Triple(buffer, replacementCharacter);
+ ++characters;
+ }
+ }
+ } else {
+ bool strict = mode == StrictConversion;
+ ConversionResult result = convertUTF16ToUTF8(&characters, characters + length, &buffer, buffer + bufferVector.size(), strict);
+ ASSERT(result != targetExhausted); // (length * 3) should be sufficient for any conversion
- // Only produced from strict conversion.
- if (result == sourceIllegal)
- return CString();
+ // Only produced from strict conversion.
+ if (result == sourceIllegal) {
+ ASSERT(strict);
+ return CString();
+ }
- // Check for an unconverted high surrogate.
- if (result == sourceExhausted) {
- if (strict)
- return CString();
- // This should be one unpaired high surrogate. Treat it the same
- // was as an unpaired high surrogate would have been handled in
- // the middle of a string with non-strict conversion - which is
- // to say, simply encode it to UTF-8.
- ASSERT((characters + 1) == (this->characters() + length));
- ASSERT((*characters >= 0xD800) && (*characters <= 0xDBFF));
- // There should be room left, since one UChar hasn't been converted.
- ASSERT((buffer + 3) <= (buffer + bufferVector.size()));
- putUTF8Triple(buffer, *characters);
+ // Check for an unconverted high surrogate.
+ if (result == sourceExhausted) {
+ if (strict)
+ return CString();
+ // This should be one unpaired high surrogate. Treat it the same
+ // was as an unpaired high surrogate would have been handled in
+ // the middle of a string with non-strict conversion - which is
+ // to say, simply encode it to UTF-8.
+ ASSERT((characters + 1) == (this->characters() + length));
+ ASSERT((*characters >= 0xD800) && (*characters <= 0xDBFF));
+ // There should be room left, since one UChar hasn't been converted.
+ ASSERT((buffer + 3) <= (buffer + bufferVector.size()));
+ putUTF8Triple(buffer, *characters);
+ }
}
}
Modified: trunk/Source/WTF/wtf/text/WTFString.h (134172 => 134173)
--- trunk/Source/WTF/wtf/text/WTFString.h 2012-11-11 13:12:54 UTC (rev 134172)
+++ trunk/Source/WTF/wtf/text/WTFString.h 2012-11-11 14:17:46 UTC (rev 134173)
@@ -211,8 +211,15 @@
WTF_EXPORT_STRING_API CString ascii() const;
WTF_EXPORT_STRING_API CString latin1() const;
- WTF_EXPORT_STRING_API CString utf8(bool strict = false) const;
+ typedef enum {
+ LenientConversion,
+ StrictConversion,
+ StrictConversionReplacingUnpairedSurrogatesWithFFFD,
+ } ConversionMode;
+
+ WTF_EXPORT_STRING_API CString utf8(ConversionMode = LenientConversion) const;
+
UChar operator[](unsigned index) const
{
if (!m_impl || index >= m_impl->length())
Modified: trunk/Source/WebCore/ChangeLog (134172 => 134173)
--- trunk/Source/WebCore/ChangeLog 2012-11-11 13:12:54 UTC (rev 134172)
+++ trunk/Source/WebCore/ChangeLog 2012-11-11 14:17:46 UTC (rev 134173)
@@ -1,3 +1,27 @@
+2012-11-11 Kenichi Ishibashi <[email protected]>
+
+ WTFString::utf8() should have a mode of conversion to use replacement character
+ https://bugs.webkit.org/show_bug.cgi?id=101678
+
+ Reviewed by Alexey Proskuryakov.
+
+ Follow the change on String::utf8()
+
+ No new tests. No changes in behavior.
+
+ * Modules/websockets/WebSocket.cpp:
+ (WebCore::WebSocket::close): Pass String::StrictConversion instead of true to String::utf8().
+ * Modules/websockets/WebSocketChannel.cpp:
+ (WebCore::WebSocketChannel::send): Ditto.
+ * html/MediaFragmentURIParser.cpp:
+ (WebCore::MediaFragmentURIParser::parseFragments): Ditto.
+ * platform/graphics/blackberry/MediaPlayerPrivateBlackBerry.cpp:
+ (WebCore::MediaPlayerPrivate::notifyChallengeResult): Ditto.
+ * platform/network/blackberry/rss/RSSFilterStream.cpp:
+ (WebCore::RSSFilterStream::convertContentToHtml): Ditto.
+ * platform/network/blackberry/rss/RSSGenerator.cpp:
+ (WebCore::RSSGenerator::generateHtml): Ditto.
+
2012-11-10 Simon Fraser <[email protected]>
Coalesce main thread scroll position updates
Modified: trunk/Source/WebCore/Modules/websockets/WebSocket.cpp (134172 => 134173)
--- trunk/Source/WebCore/Modules/websockets/WebSocket.cpp 2012-11-11 13:12:54 UTC (rev 134172)
+++ trunk/Source/WebCore/Modules/websockets/WebSocket.cpp 2012-11-11 14:17:46 UTC (rev 134173)
@@ -341,7 +341,7 @@
ec = INVALID_ACCESS_ERR;
return;
}
- CString utf8 = reason.utf8(true);
+ CString utf8 = reason.utf8(String::StrictConversion);
if (utf8.length() > maxReasonSizeInBytes) {
scriptExecutionContext()->addConsoleMessage(JSMessageSource, LogMessageType, ErrorMessageLevel, "WebSocket close message is too long.");
ec = SYNTAX_ERR;
Modified: trunk/Source/WebCore/Modules/websockets/WebSocketChannel.cpp (134172 => 134173)
--- trunk/Source/WebCore/Modules/websockets/WebSocketChannel.cpp 2012-11-11 13:12:54 UTC (rev 134172)
+++ trunk/Source/WebCore/Modules/websockets/WebSocketChannel.cpp 2012-11-11 14:17:46 UTC (rev 134173)
@@ -137,7 +137,7 @@
ThreadableWebSocketChannel::SendResult WebSocketChannel::send(const String& message)
{
LOG(Network, "WebSocketChannel %p send %s", this, message.utf8().data());
- CString utf8 = message.utf8(true);
+ CString utf8 = message.utf8(String::StrictConversion);
if (utf8.isNull() && message.length())
return InvalidMessage;
enqueueTextFrame(utf8);
Modified: trunk/Source/WebCore/html/MediaFragmentURIParser.cpp (134172 => 134173)
--- trunk/Source/WebCore/html/MediaFragmentURIParser.cpp 2012-11-11 13:12:54 UTC (rev 134172)
+++ trunk/Source/WebCore/html/MediaFragmentURIParser.cpp 2012-11-11 14:17:46 UTC (rev 134173)
@@ -142,11 +142,11 @@
// name or value are not valid UTF-8 strings, then remove the name-value pair from the list.
bool validUTF8 = true;
if (!name.isEmpty()) {
- name = name.utf8(true).data();
+ name = name.utf8(String::StrictConversion).data();
validUTF8 = !name.isEmpty();
}
if (validUTF8 && !value.isEmpty()) {
- value = value.utf8(true).data();
+ value = value.utf8(String::StrictConversion).data();
validUTF8 = !value.isEmpty();
}
Modified: trunk/Source/WebCore/platform/graphics/blackberry/MediaPlayerPrivateBlackBerry.cpp (134172 => 134173)
--- trunk/Source/WebCore/platform/graphics/blackberry/MediaPlayerPrivateBlackBerry.cpp 2012-11-11 13:12:54 UTC (rev 134172)
+++ trunk/Source/WebCore/platform/graphics/blackberry/MediaPlayerPrivateBlackBerry.cpp 2012-11-11 14:17:46 UTC (rev 134173)
@@ -749,9 +749,9 @@
if (result != AuthenticationChallengeSuccess || !url.isValid())
return;
- m_platformPlayer->reloadWithCredential(credential.user().utf8(true).data(),
- credential.password().utf8(true).data(),
- static_cast<MMRAuthChallenge::CredentialPersistence>(credential.persistence()));
+ m_platformPlayer->reloadWithCredential(credential.user().utf8(String::StrictConversion).data(),
+ credential.password().utf8(String::StrictConversion).data(),
+ static_cast<MMRAuthChallenge::CredentialPersistence>(credential.persistence()));
}
void MediaPlayerPrivate::onAuthenticationAccepted(const MMRAuthChallenge& authChallenge) const
Modified: trunk/Source/WebCore/platform/network/blackberry/rss/RSSFilterStream.cpp (134172 => 134173)
--- trunk/Source/WebCore/platform/network/blackberry/rss/RSSFilterStream.cpp 2012-11-11 13:12:54 UTC (rev 134172)
+++ trunk/Source/WebCore/platform/network/blackberry/rss/RSSFilterStream.cpp 2012-11-11 14:17:46 UTC (rev 134173)
@@ -527,7 +527,7 @@
OwnPtr<RSSGenerator> generator = adoptPtr(new RSSGenerator());
String html = generator->generateHtml(parser->m_root);
- result = html.utf8(true).data();
+ result = html.utf8(String::StrictConversion).data();
return true;
}
Modified: trunk/Source/WebCore/platform/network/blackberry/rss/RSSGenerator.cpp (134172 => 134173)
--- trunk/Source/WebCore/platform/network/blackberry/rss/RSSGenerator.cpp 2012-11-11 13:12:54 UTC (rev 134172)
+++ trunk/Source/WebCore/platform/network/blackberry/rss/RSSGenerator.cpp 2012-11-11 14:17:46 UTC (rev 134173)
@@ -83,10 +83,10 @@
builder.append(articleName);
builder.appendLiteral("\" class=\"article\">\n<a href=""
if (!item->m_link.isEmpty())
- builder.append(item->m_link.utf8(true).data());
+ builder.append(item->m_link.utf8(String::StrictConversion).data());
builder.appendLiteral("\"><b>");
if (!item->m_title.isEmpty())
- builder.append(item->m_title.utf8(true).data());
+ builder.append(item->m_title.utf8(String::StrictConversion).data());
else
builder.append(s_defaultEntryTitle);
builder.appendLiteral("</b></a>\n<br />");
@@ -94,13 +94,13 @@
if (!item->m_author.isEmpty()) {
builder.append(i18n("By"));
builder.appendLiteral(" <b>");
- builder.append(item->m_author.utf8(true).data());
+ builder.append(item->m_author.utf8(String::StrictConversion).data());
builder.appendLiteral("</b> ");
} else {
if (!feed->m_author.isEmpty()) {
builder.append(i18n("By"));
builder.appendLiteral(" <b>");
- builder.append(feed->m_author.utf8(true).data());
+ builder.append(feed->m_author.utf8(String::StrictConversion).data());
builder.appendLiteral("</b> ");
}
}
@@ -113,7 +113,7 @@
for (unsigned i = 0; i < item->m_categories.size() ; ++i) {
builder.appendLiteral("<b>");
- builder.append(item->m_categories[i].utf8(true).data());
+ builder.append(item->m_categories[i].utf8(String::StrictConversion).data());
builder.appendLiteral("</b>");
if (i < item->m_categories.size() - 1)
@@ -123,11 +123,11 @@
builder.appendLiteral("<br />");
if (!item->m_pubDate.isEmpty())
- builder.append(item->m_pubDate.utf8(true).data());
+ builder.append(item->m_pubDate.utf8(String::StrictConversion).data());
builder.appendLiteral("<br />");
if (!item->m_description.isEmpty())
- builder.append(item->m_description.utf8(true).data());
+ builder.append(item->m_description.utf8(String::StrictConversion).data());
builder.appendLiteral("<br />");
if (item->m_enclosure) {
Modified: trunk/Source/WebKit2/ChangeLog (134172 => 134173)
--- trunk/Source/WebKit2/ChangeLog 2012-11-11 13:12:54 UTC (rev 134172)
+++ trunk/Source/WebKit2/ChangeLog 2012-11-11 14:17:46 UTC (rev 134173)
@@ -1,3 +1,15 @@
+2012-11-11 Kenichi Ishibashi <[email protected]>
+
+ WTFString::utf8() should have a mode of conversion to use replacement character
+ https://bugs.webkit.org/show_bug.cgi?id=101678
+
+ Reviewed by Alexey Proskuryakov.
+
+ Update the symbol for String::utf8().
+
+ * win/WebKit2.def:
+ * win/WebKit2CFLite.def:
+
2012-11-10 Zeno Albisser <[email protected]>
[Qt][WK2] Use QLibraryInfo to search for executables.
Modified: trunk/Source/WebKit2/win/WebKit2.def (134172 => 134173)
--- trunk/Source/WebKit2/win/WebKit2.def 2012-11-11 13:12:54 UTC (rev 134172)
+++ trunk/Source/WebKit2/win/WebKit2.def 2012-11-11 14:17:46 UTC (rev 134173)
@@ -267,7 +267,7 @@
?treeScope@Node@WebCore@@QBEPAVTreeScope@2@XZ
?updateLayoutIgnorePendingStylesheets@Document@WebCore@@QAEXXZ
?userPreferredLanguages@WebCore@@YA?AV?$Vector@VString@WTF@@$0A@@WTF@@XZ
- ?utf8@String@WTF@@QBE?AVCString@2@_N@Z
+ ?utf8@String@WTF@@QBE?AVCString@2@W4ConversionMode@12@@Z
?view@Document@WebCore@@QBEPAVFrameView@2@XZ
??1ContextDestructionObserver@WebCore@@MAE@XZ
?contextDestroyed@ContextDestructionObserver@WebCore@@UAEXXZ
Modified: trunk/Source/WebKit2/win/WebKit2CFLite.def (134172 => 134173)
--- trunk/Source/WebKit2/win/WebKit2CFLite.def 2012-11-11 13:12:54 UTC (rev 134172)
+++ trunk/Source/WebKit2/win/WebKit2CFLite.def 2012-11-11 14:17:46 UTC (rev 134173)
@@ -260,7 +260,7 @@
?treeScope@Node@WebCore@@QBEPAVTreeScope@2@XZ
?updateLayoutIgnorePendingStylesheets@Document@WebCore@@QAEXXZ
?userPreferredLanguages@WebCore@@YA?AV?$Vector@VString@WTF@@$0A@@WTF@@XZ
- ?utf8@String@WTF@@QBE?AVCString@2@_N@Z
+ ?utf8@String@WTF@@QBE?AVCString@2@W4ConversionMode@12@@Z
?view@Document@WebCore@@QBEPAVFrameView@2@XZ
??1ContextDestructionObserver@WebCore@@MAE@XZ
?contextDestroyed@ContextDestructionObserver@WebCore@@UAEXXZ
