third_party/WebKit/Source/core/html/parser/HTMLDocumentParser.h

   1 /*
   2  * Copyright (C) 2010 Google, Inc. All Rights Reserved.
   3  *
   4  * Redistribution and use in source and binary forms, with or without
   5  * modification, are permitted provided that the following conditions
   6  * are met:
   7  * 1. Redistributions of source code must retain the above copyright
   8  *    notice, this list of conditions and the following disclaimer.
   9  * 2. Redistributions in binary form must reproduce the above copyright
  10  *    notice, this list of conditions and the following disclaimer in the
  11  *    documentation and/or other materials provided with the distribution.
  12  *
  13  * THIS SOFTWARE IS PROVIDED BY APPLE INC. ``AS IS'' AND ANY
  14  * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  15  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
  16  * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL APPLE INC. OR
  17  * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
  18  * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
  19  * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
  20  * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
  21  * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
  22  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
  23  * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  24  */
  25
  26 #ifndef HTMLDocumentParser_h
  27 #define HTMLDocumentParser_h
  28
  29 #include "core/dom/ParserContentPolicy.h"
  30 #include "core/dom/ScriptableDocumentParser.h"
  31 #include "core/fetch/ResourceClient.h"
  32 #include "core/frame/UseCounter.h"
  33 #include "core/html/parser/BackgroundHTMLInputStream.h"
  34 #include "core/html/parser/CompactHTMLToken.h"
  35 #include "core/html/parser/HTMLInputStream.h"
  36 #include "core/html/parser/HTMLParserOptions.h"
  37 #include "core/html/parser/HTMLPreloadScanner.h"
  38 #include "core/html/parser/HTMLScriptRunnerHost.h"
  39 #include "core/html/parser/HTMLSourceTracker.h"
  40 #include "core/html/parser/HTMLToken.h"
  41 #include "core/html/parser/HTMLTokenizer.h"
  42 #include "core/html/parser/HTMLTreeBuilderSimulator.h"
  43 #include "core/html/parser/ParserSynchronizationPolicy.h"
  44 #include "core/html/parser/TextResourceDecoder.h"
  45 #include "core/html/parser/XSSAuditor.h"
  46 #include "core/html/parser/XSSAuditorDelegate.h"
  47 #include "platform/text/SegmentedString.h"
  48 #include "wtf/Deque.h"
  49 #include "wtf/OwnPtr.h"
  50 #include "wtf/WeakPtr.h"
  51 #include "wtf/text/TextPosition.h"
  52
  53 namespace blink {
  54
  55 class BackgroundHTMLParser;
  56 class CompactHTMLToken;
  57 class Document;
  58 class DocumentEncodingData;
  59 class DocumentFragment;
  60 class Element;
  61 class HTMLDocument;
  62 class HTMLParserScheduler;
  63 class HTMLScriptRunner;
  64 class HTMLTreeBuilder;
  65 class HTMLResourcePreloader;
  66
  67 class PumpSession;
  68
  69 class HTMLDocumentParser :  public ScriptableDocumentParser, private HTMLScriptRunnerHost {
  70     WTF_MAKE_FAST_ALLOCATED_WILL_BE_REMOVED(HTMLDocumentParser);
  71     WILL_BE_USING_GARBAGE_COLLECTED_MIXIN(HTMLDocumentParser);
  72 public:
  73     static PassRefPtrWillBeRawPtr<HTMLDocumentParser> create(HTMLDocument& document, bool reportErrors, ParserSynchronizationPolicy backgroundParsingPolicy)
  74     {
  75         return adoptRefWillBeNoop(new HTMLDocumentParser(document, reportErrors, backgroundParsingPolicy));
  76     }
  77     ~HTMLDocumentParser() override;
  78     DECLARE_VIRTUAL_TRACE();
  79
  80     // Exposed for HTMLParserScheduler
  81     void resumeParsingAfterYield();
  82
  83     static void parseDocumentFragment(const String&, DocumentFragment*, Element* contextElement, ParserContentPolicy = AllowScriptingContent);
  84
  85     HTMLTokenizer* tokenizer() const { return m_tokenizer.get(); }
  86
  87     TextPosition textPosition() const final;
  88     bool isParsingAtLineNumber() const final;
  89     OrdinalNumber lineNumber() const final;
  90
  91     void suspendScheduledTasks() final;
  92     void resumeScheduledTasks() final;
  93
  94     struct ParsedChunk {
  95         WTF_MAKE_FAST_ALLOCATED(ParsedChunk);
  96     public:
  97         OwnPtr<CompactHTMLTokenStream> tokens;
  98         PreloadRequestStream preloads;
  99         XSSInfoStream xssInfos;
 100         HTMLTokenizer::State tokenizerState;
 101         HTMLTreeBuilderSimulator::State treeBuilderState;
 102         HTMLInputCheckpoint inputCheckpoint;
 103         TokenPreloadScannerCheckpoint preloadScannerCheckpoint;
 104         bool startingScript;
 105     };
 106     void didReceiveParsedChunkFromBackgroundParser(PassOwnPtr<ParsedChunk>);
 107     void didReceiveEncodingDataFromBackgroundParser(const DocumentEncodingData&);
 108
 109     void appendBytes(const char* bytes, size_t length) override;
 110     void flush() final;
 111     void setDecoder(PassOwnPtr<TextResourceDecoder>) final;
 112
 113     UseCounter* useCounter() { return UseCounter::getFrom(contextForParsingSession()); }
 114
 115 protected:
 116     void insert(const SegmentedString&) final;
 117     void append(const String&) override;
 118     void finish() final;
 119
 120     HTMLDocumentParser(HTMLDocument&, bool reportErrors, ParserSynchronizationPolicy);
 121     HTMLDocumentParser(DocumentFragment*, Element* contextElement, ParserContentPolicy);
 122
 123     HTMLTreeBuilder* treeBuilder() const { return m_treeBuilder.get(); }
 124
 125     void forcePlaintextForTextDocument();
 126
 127 private:
 128     static PassRefPtrWillBeRawPtr<HTMLDocumentParser> create(DocumentFragment* fragment, Element* contextElement, ParserContentPolicy parserContentPolicy)
 129     {
 130         return adoptRefWillBeNoop(new HTMLDocumentParser(fragment, contextElement, parserContentPolicy));
 131     }
 132
 133     // DocumentParser
 134     void detach() final;
 135     bool hasInsertionPoint() final;
 136     bool processingData() const final;
 137     void prepareToStopParsing() final;
 138     void stopParsing() final;
 139     bool isWaitingForScripts() const final;
 140     bool isExecutingScript() const final;
 141     void executeScriptsWaitingForResources() final;
 142
 143     // HTMLScriptRunnerHost
 144     void notifyScriptLoaded(Resource*) final;
 145     HTMLInputStream& inputStream() final { return m_input; }
 146     bool hasPreloadScanner() const final { return m_preloadScanner.get() && !shouldUseThreading(); }
 147     void appendCurrentInputStreamToPreloadScannerAndScan() final;
 148
 149     void startBackgroundParser();
 150     void stopBackgroundParser();
 151     void validateSpeculations(PassOwnPtr<ParsedChunk> lastChunk);
 152     void discardSpeculationsAndResumeFrom(PassOwnPtr<ParsedChunk> lastChunk, PassOwnPtr<HTMLToken>, PassOwnPtr<HTMLTokenizer>);
 153     size_t processParsedChunkFromBackgroundParser(PassOwnPtr<ParsedChunk>);
 154     void pumpPendingSpeculations();
 155
 156     Document* contextForParsingSession();
 157
 158     bool canTakeNextToken();
 159     void pumpTokenizer();
 160     void pumpTokenizerIfPossible();
 161     void constructTreeFromHTMLToken();
 162     void constructTreeFromCompactHTMLToken(const CompactHTMLToken&);
 163
 164     void runScriptsForPausedTreeBuilder();
 165     void resumeParsingAfterScriptExecution();
 166
 167     void attemptToEnd();
 168     void endIfDelayed();
 169     void attemptToRunDeferredScriptsAndEnd();
 170     void end();
 171
 172     bool shouldUseThreading() const { return m_shouldUseThreading; }
 173
 174     bool isParsingFragment() const;
 175     bool isScheduledForResume() const;
 176     bool inPumpSession() const { return m_pumpSessionNestingLevel > 0; }
 177     bool shouldDelayEnd() const { return inPumpSession() || isWaitingForScripts() || isScheduledForResume() || isExecutingScript(); }
 178
 179     HTMLToken& token() { return *m_token; }
 180
 181     HTMLParserOptions m_options;
 182     HTMLInputStream m_input;
 183
 184     OwnPtr<HTMLToken> m_token;
 185     OwnPtr<HTMLTokenizer> m_tokenizer;
 186     OwnPtrWillBeMember<HTMLScriptRunner> m_scriptRunner;
 187     OwnPtrWillBeMember<HTMLTreeBuilder> m_treeBuilder;
 188     OwnPtr<HTMLPreloadScanner> m_preloadScanner;
 189     OwnPtr<HTMLPreloadScanner> m_insertionPreloadScanner;
 190     OwnPtr<HTMLParserScheduler> m_parserScheduler;
 191     HTMLSourceTracker m_sourceTracker;
 192     TextPosition m_textPosition;
 193     XSSAuditor m_xssAuditor;
 194     XSSAuditorDelegate m_xssAuditorDelegate;
 195
 196     // FIXME: m_lastChunkBeforeScript, m_tokenizer, m_token, and m_input should be combined into a single state object
 197     // so they can be set and cleared together and passed between threads together.
 198     OwnPtr<ParsedChunk> m_lastChunkBeforeScript;
 199     Deque<OwnPtr<ParsedChunk>> m_speculations;
 200     WeakPtrFactory<HTMLDocumentParser> m_weakFactory;
 201     WeakPtr<BackgroundHTMLParser> m_backgroundParser;
 202     OwnPtrWillBeMember<HTMLResourcePreloader> m_preloader;
 203     PreloadRequestStream m_queuedPreloads;
 204
 205     bool m_shouldUseThreading;
 206     bool m_endWasDelayed;
 207     bool m_haveBackgroundParser;
 208     bool m_tasksWereSuspended;
 209     unsigned m_pumpSessionNestingLevel;
 210     unsigned m_pumpSpeculationsSessionNestingLevel;
 211     bool m_isParsingAtLineNumber;
 212 };
 213
 214 }
 215
 216 #endif