Chromium Code Reviews| OLD | NEW |
|---|---|
| 1 /* | 1 /* |
| 2 * Copyright (C) 2013 Google, Inc. All Rights Reserved. | 2 * Copyright (C) 2013 Google, Inc. All Rights Reserved. |
| 3 * | 3 * |
| 4 * Redistribution and use in source and binary forms, with or without | 4 * Redistribution and use in source and binary forms, with or without |
| 5 * modification, are permitted provided that the following conditions | 5 * modification, are permitted provided that the following conditions |
| 6 * are met: | 6 * are met: |
| 7 * 1. Redistributions of source code must retain the above copyright | 7 * 1. Redistributions of source code must retain the above copyright |
| 8 * notice, this list of conditions and the following disclaimer. | 8 * notice, this list of conditions and the following disclaimer. |
| 9 * 2. Redistributions in binary form must reproduce the above copyright | 9 * 2. Redistributions in binary form must reproduce the above copyright |
| 10 * notice, this list of conditions and the following disclaimer in the | 10 * notice, this list of conditions and the following disclaimer in the |
| 11 * documentation and/or other materials provided with the distribution. | 11 * documentation and/or other materials provided with the distribution. |
| 12 * | 12 * |
| 13 * THIS SOFTWARE IS PROVIDED BY GOOGLE INC. ``AS IS'' AND ANY | 13 * THIS SOFTWARE IS PROVIDED BY GOOGLE INC. ``AS IS'' AND ANY |
| 14 * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE | 14 * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE |
| 15 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR | 15 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR |
| 16 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GOOGLE INC. OR | 16 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GOOGLE INC. OR |
| 17 * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, | 17 * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, |
| 18 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, | 18 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, |
| 19 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR | 19 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR |
| 20 * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY | 20 * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY |
| 21 * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT | 21 * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT |
| 22 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE | 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. | 23 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. |
| 24 */ | 24 */ |
| 25 | 25 |
| 26 #ifndef BackgroundHTMLParser_h | 26 #ifndef BackgroundHTMLParser_h |
| 27 #define BackgroundHTMLParser_h | 27 #define BackgroundHTMLParser_h |
| 28 | 28 |
| 29 #include <memory> | |
| 29 #include "core/dom/DocumentEncodingData.h" | 30 #include "core/dom/DocumentEncodingData.h" |
| 30 #include "core/html/parser/BackgroundHTMLInputStream.h" | 31 #include "core/html/parser/BackgroundHTMLInputStream.h" |
| 31 #include "core/html/parser/CompactHTMLToken.h" | 32 #include "core/html/parser/CompactHTMLToken.h" |
| 32 #include "core/html/parser/HTMLParserOptions.h" | 33 #include "core/html/parser/HTMLParserOptions.h" |
| 33 #include "core/html/parser/HTMLPreloadScanner.h" | 34 #include "core/html/parser/HTMLPreloadScanner.h" |
| 34 #include "core/html/parser/HTMLSourceTracker.h" | 35 #include "core/html/parser/HTMLSourceTracker.h" |
| 35 #include "core/html/parser/HTMLTreeBuilderSimulator.h" | 36 #include "core/html/parser/HTMLTreeBuilderSimulator.h" |
| 36 #include "core/html/parser/TextResourceDecoder.h" | 37 #include "core/html/parser/TextResourceDecoder.h" |
| 37 #include "core/html/parser/TokenizedChunkQueue.h" | 38 #include "core/html/parser/TokenizedChunkQueue.h" |
| 38 #include "core/html/parser/XSSAuditorDelegate.h" | 39 #include "core/html/parser/XSSAuditorDelegate.h" |
| 40 #include "platform/heap/GarbageCollected.h" | |
| 39 #include "wtf/WeakPtr.h" | 41 #include "wtf/WeakPtr.h" |
| 40 #include <memory> | |
| 41 | 42 |
| 42 namespace blink { | 43 namespace blink { |
| 43 | 44 |
| 44 class HTMLDocumentParser; | 45 class HTMLDocumentParser; |
| 45 class XSSAuditor; | 46 class XSSAuditor; |
| 46 class WebTaskRunner; | 47 class WebTaskRunner; |
| 47 | 48 |
| 48 class BackgroundHTMLParser { | 49 class BackgroundHTMLParser |
| 49 USING_FAST_MALLOC(BackgroundHTMLParser); | 50 : public GarbageCollectedFinalized<BackgroundHTMLParser> { |
|
Yoav Weiss
2017/02/09 10:39:00
I don't think this needs to be Finalized, as it ha
Charlie Harrison
2017/02/13 22:55:32
It does need to be finalized, because a bunch of f
Yoav Weiss
2017/02/14 06:45:47
ok
| |
| 50 WTF_MAKE_NONCOPYABLE(BackgroundHTMLParser); | 51 WTF_MAKE_NONCOPYABLE(BackgroundHTMLParser); |
| 51 | 52 |
| 52 public: | 53 public: |
| 53 struct Configuration { | 54 struct Configuration { |
| 54 USING_FAST_MALLOC(Configuration); | 55 USING_FAST_MALLOC(Configuration); |
| 55 | 56 |
| 56 public: | 57 public: |
| 57 Configuration(); | 58 Configuration(); |
| 58 HTMLParserOptions options; | 59 HTMLParserOptions options; |
| 59 WeakPtr<HTMLDocumentParser> parser; | |
| 60 std::unique_ptr<XSSAuditor> xssAuditor; | 60 std::unique_ptr<XSSAuditor> xssAuditor; |
| 61 std::unique_ptr<TextResourceDecoder> decoder; | 61 std::unique_ptr<TextResourceDecoder> decoder; |
| 62 RefPtr<TokenizedChunkQueue> tokenizedChunkQueue; | 62 RefPtr<TokenizedChunkQueue> tokenizedChunkQueue; |
| 63 // outstandingTokenLimit must be greater than or equal to | 63 // outstandingTokenLimit must be greater than or equal to |
| 64 // pendingTokenLimit | 64 // pendingTokenLimit |
| 65 size_t outstandingTokenLimit; | 65 size_t outstandingTokenLimit; |
| 66 size_t pendingTokenLimit; | 66 size_t pendingTokenLimit; |
| 67 bool shouldCoalesceChunks; | 67 bool shouldCoalesceChunks; |
| 68 }; | 68 }; |
| 69 | 69 |
| 70 // The returned BackgroundHTMLParser should only be used on the parser | 70 static BackgroundHTMLParser* create(HTMLDocumentParser*, |
| 71 // thread: it must first be initialized by calling init(), and free by | 71 Document&, |
| 72 // calling stop(). | 72 std::unique_ptr<Configuration>); |
| 73 static WeakPtr<BackgroundHTMLParser> create(std::unique_ptr<Configuration>, | 73 ~BackgroundHTMLParser(); |
| 74 RefPtr<WebTaskRunner>); | 74 |
| 75 void init(const KURL& documentURL, | 75 DECLARE_TRACE(); |
| 76 std::unique_ptr<CachedDocumentParameters>, | |
| 77 const MediaValuesCached::MediaValuesCachedData&); | |
| 78 | 76 |
| 79 struct Checkpoint { | 77 struct Checkpoint { |
| 80 USING_FAST_MALLOC(Checkpoint); | 78 USING_FAST_MALLOC(Checkpoint); |
| 81 | 79 |
| 82 public: | 80 public: |
| 83 WeakPtr<HTMLDocumentParser> parser; | 81 WeakPtr<HTMLDocumentParser> parser; |
| 84 std::unique_ptr<HTMLToken> token; | 82 std::unique_ptr<HTMLToken> token; |
| 85 std::unique_ptr<HTMLTokenizer> tokenizer; | 83 std::unique_ptr<HTMLTokenizer> tokenizer; |
| 86 HTMLTreeBuilderSimulator::State treeBuilderState; | 84 HTMLTreeBuilderSimulator::State treeBuilderState; |
| 87 HTMLInputCheckpoint inputCheckpoint; | 85 HTMLInputCheckpoint inputCheckpoint; |
| 88 TokenPreloadScannerCheckpoint preloadScannerCheckpoint; | 86 TokenPreloadScannerCheckpoint preloadScannerCheckpoint; |
| 89 String unparsedInput; | 87 String unparsedInput; |
| 90 }; | 88 }; |
| 91 | 89 |
| 92 void appendRawBytesFromMainThread(std::unique_ptr<Vector<char>>, | 90 void appendRawBytesFromMainThread(std::unique_ptr<Vector<char>>, |
| 93 double bytesReceivedTime); | 91 double bytesReceivedTime); |
| 94 void setDecoder(std::unique_ptr<TextResourceDecoder>); | 92 void setDecoder(std::unique_ptr<TextResourceDecoder>); |
| 95 void flush(); | 93 void flush(); |
| 96 void resumeFrom(std::unique_ptr<Checkpoint>); | 94 void resumeFrom(std::unique_ptr<Checkpoint>); |
| 97 void startedChunkWithCheckpoint(HTMLInputCheckpoint); | 95 void startedChunkWithCheckpoint(HTMLInputCheckpoint); |
| 98 void finish(); | 96 void finish(); |
| 99 void stop(); | |
| 100 | 97 |
| 101 void forcePlaintextForTextDocument(); | 98 void forcePlaintextForTextDocument(); |
| 102 | |
| 103 private: | 99 private: |
| 104 BackgroundHTMLParser(std::unique_ptr<Configuration>, RefPtr<WebTaskRunner>); | 100 BackgroundHTMLParser(HTMLDocumentParser*, |
| 105 ~BackgroundHTMLParser(); | 101 std::unique_ptr<Configuration>, |
| 102 RefPtr<WebTaskRunner>, | |
| 103 std::unique_ptr<TokenPreloadScanner>); | |
| 106 | 104 |
| 107 void appendDecodedBytes(const String&); | 105 void appendDecodedBytes(const String&); |
| 108 void markEndOfFile(); | 106 void markEndOfFile(); |
| 109 void pumpTokenizer(); | 107 void pumpTokenizer(); |
| 110 | 108 |
| 109 void onResumeFrom(std::unique_ptr<Checkpoint>); | |
| 110 void onFinish(); | |
| 111 | |
| 111 // Returns whether or not the HTMLDocumentParser should be notified of | 112 // Returns whether or not the HTMLDocumentParser should be notified of |
| 112 // pending chunks. | 113 // pending chunks. |
| 113 bool queueChunkForMainThread(); | 114 bool queueChunkForMainThread(); |
| 114 void notifyMainThreadOfNewChunks(); | 115 void notifyMainThreadOfNewChunks(); |
| 115 void updateDocument(const String& decodedData); | 116 void updateDocument(const String& decodedData); |
| 116 | 117 |
| 117 template <typename FunctionType, typename... Ps> | |
| 118 void runOnMainThread(FunctionType, Ps&&...); | |
| 119 | |
| 120 WeakPtrFactory<BackgroundHTMLParser> m_weakFactory; | |
| 121 BackgroundHTMLInputStream m_input; | 118 BackgroundHTMLInputStream m_input; |
| 122 HTMLSourceTracker m_sourceTracker; | 119 HTMLSourceTracker m_sourceTracker; |
| 123 std::unique_ptr<HTMLToken> m_token; | 120 std::unique_ptr<HTMLToken> m_token; |
| 124 std::unique_ptr<HTMLTokenizer> m_tokenizer; | 121 std::unique_ptr<HTMLTokenizer> m_tokenizer; |
| 125 HTMLTreeBuilderSimulator m_treeBuilderSimulator; | 122 HTMLTreeBuilderSimulator m_treeBuilderSimulator; |
| 126 HTMLParserOptions m_options; | 123 HTMLParserOptions m_options; |
| 127 const size_t m_outstandingTokenLimit; | 124 const size_t m_outstandingTokenLimit; |
| 128 WeakPtr<HTMLDocumentParser> m_parser; | 125 Member<HTMLDocumentParser> m_parser; |
| 129 | 126 |
| 130 std::unique_ptr<CompactHTMLTokenStream> m_pendingTokens; | 127 std::unique_ptr<CompactHTMLTokenStream> m_pendingTokens; |
| 131 const size_t m_pendingTokenLimit; | 128 const size_t m_pendingTokenLimit; |
| 132 PreloadRequestStream m_pendingPreloads; | 129 PreloadRequestStream m_pendingPreloads; |
| 133 // Indices into |m_pendingTokens|. | 130 // Indices into |m_pendingTokens|. |
| 134 Vector<int> m_likelyDocumentWriteScriptIndices; | 131 Vector<int> m_likelyDocumentWriteScriptIndices; |
| 135 ViewportDescriptionWrapper m_viewportDescription; | 132 ViewportDescriptionWrapper m_viewportDescription; |
| 136 XSSInfoStream m_pendingXSSInfos; | 133 XSSInfoStream m_pendingXSSInfos; |
| 137 | 134 |
| 138 std::unique_ptr<XSSAuditor> m_xssAuditor; | 135 std::unique_ptr<XSSAuditor> m_xssAuditor; |
| 139 std::unique_ptr<TokenPreloadScanner> m_preloadScanner; | 136 std::unique_ptr<TokenPreloadScanner> m_preloadScanner; |
| 140 std::unique_ptr<TextResourceDecoder> m_decoder; | 137 std::unique_ptr<TextResourceDecoder> m_decoder; |
| 141 DocumentEncodingData m_lastSeenEncodingData; | 138 DocumentEncodingData m_lastSeenEncodingData; |
| 142 RefPtr<WebTaskRunner> m_loadingTaskRunner; | 139 RefPtr<WebTaskRunner> m_loadingTaskRunner; |
| 143 RefPtr<TokenizedChunkQueue> m_tokenizedChunkQueue; | 140 RefPtr<TokenizedChunkQueue> m_tokenizedChunkQueue; |
|
Yoav Weiss
2017/02/09 10:39:00
Seems like we could have moved TokenizedChunkQueue
Charlie Harrison
2017/02/13 22:55:32
We'll probably want this to be a raw pointer, and
kouhei (in TOK)
2017/02/13 23:09:14
Optional (assuming TokenizedChunkQueue to also go
Charlie Harrison
2017/02/13 23:19:31
Done.
| |
| 144 | 141 |
| 145 // Index into |m_pendingTokens| of the last <meta> csp token found. Will be | 142 // Index into |m_pendingTokens| of the last <meta> csp token found. Will be |
| 146 // |TokenizedChunk::noPendingToken| if none have been found. | 143 // |TokenizedChunk::noPendingToken| if none have been found. |
| 147 int m_pendingCSPMetaTokenIndex; | 144 int m_pendingCSPMetaTokenIndex; |
| 148 | 145 |
| 149 bool m_startingScript; | 146 bool m_startingScript; |
| 150 double m_lastBytesReceivedTime; | 147 double m_lastBytesReceivedTime; |
| 151 bool m_shouldCoalesceChunks; | 148 bool m_shouldCoalesceChunks; |
| 152 }; | 149 }; |
| 153 | 150 |
| 154 } // namespace blink | 151 } // namespace blink |
| 155 | 152 |
| 156 #endif | 153 #endif |
| OLD | NEW |