Moved text decoding to the parser thread

Source/core/html/parser/BackgroundHTMLParser.cpp

 /*
  * Copyright (C) 2013 Google, Inc. All Rights Reserved.
  *
  * 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
  *    notice, this list of conditions and the following disclaimer in the
  *    documentation and/or other materials provided with the distribution.
  *
  * THIS SOFTWARE IS PROVIDED BY GOOGLE INC. ``AS IS'' AND ANY
  * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
  * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL GOOGLE INC. OR
  * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
  * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
  * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
  * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
  * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
  * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  */
 
 #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"
 
 namespace WebCore {
 
 // On a network with high latency and high bandwidth, using a device
 // with a fast CPU, we could end up speculatively tokenizing
 // the whole document, well ahead of when the main-thread actually needs it.
 // This is a waste of memory (and potentially time if the speculation fails).
 // So we limit our outstanding speculations arbitrarily to 10.
@@ skipping 27 matching lines @@
 }
 
 static void checkThatXSSInfosAreSafeToSendToAnotherThread(const XSSInfoStream& infos)
 {
     for (size_t i = 0; i < infos.size(); ++i)
         ASSERT(infos[i]->isSafeToSendToAnotherThread());
 }
 
 #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))
     , m_treeBuilderSimulator(config->options)
     , m_options(config->options)
     , m_parser(config->parser)
     , m_pendingTokens(adoptPtr(new CompactHTMLTokenStream))
     , m_xssAuditor(config->xssAuditor.release())
     , m_preloadScanner(config->preloadScanner.release())
+    , m_decoder(decoder)
 {
 }
 
+BackgroundHTMLParser::~BackgroundHTMLParser()
+{
+}
+
 void BackgroundHTMLParser::append(const String& input)
 {
     ASSERT(!m_input.current().isClosed());
     m_input.append(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;
     m_token = checkpoint->token.release();
     m_tokenizer = checkpoint->tokenizer.release();
     m_treeBuilderSimulator.setState(checkpoint->treeBuilderState);
     m_input.rewindTo(checkpoint->inputCheckpoint, checkpoint->unparsedInput);
     m_preloadScanner->rewindTo(checkpoint->preloadScannerCheckpoint);
     pumpTokenizer();
 }
@@ skipping 91 matching lines @@
     chunk->tokenizerState = m_tokenizer->state();
     chunk->treeBuilderState = m_treeBuilderSimulator.state();
     chunk->inputCheckpoint = m_input.createCheckpoint();
     chunk->preloadScannerCheckpoint = m_preloadScanner->createCheckpoint();
     callOnMainThread(bind(&HTMLDocumentParser::didReceiveParsedChunkFromBackgroundParser, m_parser, chunk.release()));
 
     m_pendingTokens = adoptPtr(new CompactHTMLTokenStream);
 }
 
 }