Moved text decoding to the parser thread

Source/core/html/parser/BackgroundHTMLParser.cpp

Index: Source/core/html/parser/BackgroundHTMLParser.cpp
diff --git a/Source/core/html/parser/BackgroundHTMLParser.cpp b/Source/core/html/parser/BackgroundHTMLParser.cpp
index b04ebab5b4ac66afde6a7b60a64db55a496fab2b..2d81ba2dac9fd64a2125b0b8298bdf3a31af9ff5 100644
--- a/Source/core/html/parser/BackgroundHTMLParser.cpp
+++ b/Source/core/html/parser/BackgroundHTMLParser.cpp
@@ -26,10 +26,13 @@
 #include "config.h"
 #include "core/html/parser/BackgroundHTMLParser.h"
 
+#include "core/fetch/TextResourceDecoder.h"

[abarth-chromium, 2013/11/18 06:57:10]

Should this class move to the core/html/parser directory?

[oystein (OOO til 10th of July), 2013/11/18 22:18:42]

It does live in an awkward location right now, but it handles more than html
too. Then again core/html/parser/ does have other non-HTML things like the CSS
preload scanner, so maybe that's OK?

 #include "core/html/parser/HTMLDocumentParser.h"
 #include "core/html/parser/HTMLParserThread.h"
+#include "core/html/parser/HTMLToken.h"
 #include "core/html/parser/HTMLTokenizer.h"
 #include "core/html/parser/XSSAuditor.h"
+#include "platform/SharedBuffer.h"
 #include "wtf/MainThread.h"
 #include "wtf/text/TextPosition.h"
 
@@ -77,7 +80,13 @@ static void checkThatXSSInfosAreSafeToSendToAnotherThread(const XSSInfoStream& i
 
 #endif
 
-BackgroundHTMLParser::BackgroundHTMLParser(PassRefPtr<WeakReference<BackgroundHTMLParser> > reference, PassOwnPtr<Configuration> config)
+void BackgroundHTMLParser::create(PassRefPtr<WeakReference<BackgroundHTMLParser> > reference, PassOwnPtr<Configuration> config, PassOwnPtr<TextResourceDecoder> decoder)
+{
+    new BackgroundHTMLParser(reference, config, decoder);

[eseidel, 2013/11/18 20:07:40]

This reads very strange.

[oystein (OOO til 10th of July), 2013/11/18 22:18:42]

I agree (I just moved this from the header), I don't see any big benefits to
using the WeakPtrFactory thing here. Should I just yank it out?

+    // Caller must free by calling stop().

[eseidel, 2013/11/18 20:07:40]

If this is deleted via "stop" maybe it should be created via start()?

[oystein (OOO til 10th of July), 2013/11/18 22:18:42]

Done.

+}
+
+BackgroundHTMLParser::BackgroundHTMLParser(PassRefPtr<WeakReference<BackgroundHTMLParser> > reference, PassOwnPtr<Configuration> config, PassOwnPtr<TextResourceDecoder> decoder)
     : m_weakFactory(reference, this)
     , m_token(adoptPtr(new HTMLToken))
     , m_tokenizer(HTMLTokenizer::create(config->options))
@@ -87,6 +96,11 @@ BackgroundHTMLParser::BackgroundHTMLParser(PassRefPtr<WeakReference<BackgroundHT
     , m_pendingTokens(adoptPtr(new CompactHTMLTokenStream))
     , m_xssAuditor(config->xssAuditor.release())
     , m_preloadScanner(config->preloadScanner.release())
+    , m_decoder(decoder)
+{
+}
+
+BackgroundHTMLParser::~BackgroundHTMLParser()
 {
 }
 
@@ -97,6 +111,37 @@ void BackgroundHTMLParser::append(const String& input)
     pumpTokenizer();
 }
 
+void BackgroundHTMLParser::appendBytes(PassRefPtr<SharedBuffer> data)
+{
+    RefPtr<SharedBuffer> buffer = data;
+    String decoded = m_decoder->decode(buffer->data(), buffer->size());
+    updateDocument(decoded);
+}
+
+void BackgroundHTMLParser::flush()
+{
+    String remainingData = m_decoder->flush();

[eseidel, 2013/11/18 20:07:40]

I'm not sure the local buys you much.

[oystein (OOO til 10th of July), 2013/11/18 22:18:42]

error: non-const lvalue reference to type 'WTF::String' cannot bind to a
temporary of type 'WTF::String'

+    updateDocument(remainingData);
+}
+
+void BackgroundHTMLParser::updateDocument(String& decodedData)
+{
+    DocumentEncodingData encodingData;
+    encodingData.encoding = m_decoder->encoding();

[eseidel, 2013/11/18 20:07:40]

This feels like DocumentEndodingData should be something the Decoder knows how
to return for you?  Or maybe DocumentEncodingData has a constructor which knows
how to fill itself from a Decoder?  This just doesn't feel like it belongs here.

[oystein (OOO til 10th of July), 2013/11/18 22:18:42]

Done; Added TextResourceDecoder::getEncodingData(). 

+    encodingData.wasDetectedHeuristically = m_decoder->encodingWasDetectedHeuristically();
+    encodingData.sawDecodingError = m_decoder->sawError();
+
+    if (encodingData != m_lastSeenEncodingData) {
+        m_lastSeenEncodingData = encodingData;
+        callOnMainThread(bind(&HTMLDocumentParser::didReceiveEncodingDataFromBackgroundParser, m_parser, encodingData));
+    }
+
+    if (decodedData.isEmpty())
+        return;
+
+    append(decodedData);
+}
+
 void BackgroundHTMLParser::resumeFrom(PassOwnPtr<Checkpoint> checkpoint)
 {
     m_parser = checkpoint->parser;