WebCore/dom/XMLTokenizer.h

   1 /*
   2  * Copyright (C) 2000 Peter Kelly (pmk@post.com)
   3  * Copyright (C) 2005, 2006, 2007 Apple Inc. All rights reserved.
   4  * Copyright (C) 2007 Samuel Weinig (sam@webkit.org)
   5  * Copyright (C) 2008 Nokia Corporation and/or its subsidiary(-ies)
   6  *
   7  * This library is free software; you can redistribute it and/or
   8  * modify it under the terms of the GNU Library General Public
   9  * License as published by the Free Software Foundation; either
  10  * version 2 of the License, or (at your option) any later version.
  11  *
  12  * This library is distributed in the hope that it will be useful,
  13  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  14  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  15  * Library General Public License for more details.
  16  *
  17  * You should have received a copy of the GNU Library General Public License
  18  * along with this library; see the file COPYING.LIB.  If not, write to
  19  * the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
  20  * Boston, MA 02110-1301, USA.
  21  *
  22  */
  23
  24 #ifndef XMLTokenizer_h
  25 #define XMLTokenizer_h
  26
  27 #include "CachedResourceClient.h"
  28 #include "CachedResourceHandle.h"
  29 #include "SegmentedString.h"
  30 #include "StringHash.h"
  31 #include "Tokenizer.h"
  32 #include <wtf/HashMap.h>
  33 #include <wtf/OwnPtr.h>
  34
  35 #if USE(QXMLSTREAM)
  36 #include <QtXml/qxmlstream.h>
  37 #else
  38 #include <libxml/tree.h>
  39 #include <libxml/xmlstring.h>
  40 #endif
  41
  42 namespace WebCore {
  43
  44     class Node;
  45     class CachedScript;
  46     class DocLoader;
  47     class DocumentFragment;
  48     class Document;
  49     class Element;
  50     class FrameView;
  51     class PendingCallbacks;
  52     class ScriptElement;
  53
  54     class XMLTokenizer : public Tokenizer, public CachedResourceClient {
  55     public:
  56         XMLTokenizer(Document*, FrameView* = 0);
  57         XMLTokenizer(DocumentFragment*, Element*);
  58         ~XMLTokenizer();
  59
  60         enum ErrorType { warning, nonFatal, fatal };
  61
  62         // from Tokenizer
  63         virtual bool write(const SegmentedString&, bool appendData);
  64         virtual void finish();
  65         virtual bool isWaitingForScripts() const;
  66         virtual void stopParsing();
  67
  68         void end();
  69
  70         void pauseParsing();
  71         void resumeParsing();
  72
  73         void setIsXHTMLDocument(bool isXHTML) { m_isXHTMLDocument = isXHTML; }
  74         bool isXHTMLDocument() const { return m_isXHTMLDocument; }
  75
  76         // from CachedResourceClient
  77         virtual void notifyFinished(CachedResource* finishedObj);
  78
  79
  80         void handleError(ErrorType type, const char* m, int lineNumber, int columnNumber);
  81
  82         virtual bool wellFormed() const { return !m_sawError; }
  83
  84         int lineNumber() const;
  85         int columnNumber() const;
  86
  87 #if USE(QXMLSTREAM)
  88 private:
  89         void parse();
  90         void startDocument();
  91         void parseStartElement();
  92         void parseEndElement();
  93         void parseCharacters();
  94         void parseProcessingInstruction();
  95         void parseCdata();
  96         void parseComment();
  97         void endDocument();
  98         void parseDtd();
  99         bool hasError() const;
 100 #else
 101 public:
 102         // callbacks from parser SAX
 103         void error(ErrorType, const char* message, va_list args) WTF_ATTRIBUTE_PRINTF(3, 0);
 104         void startElementNs(const xmlChar* xmlLocalName, const xmlChar* xmlPrefix, const xmlChar* xmlURI, int nb_namespaces,
 105                             const xmlChar** namespaces, int nb_attributes, int nb_defaulted, const xmlChar** libxmlAttributes);
 106         void endElementNs();
 107         void characters(const xmlChar* s, int len);
 108         void processingInstruction(const xmlChar* target, const xmlChar* data);
 109         void cdataBlock(const xmlChar* s, int len);
 110         void comment(const xmlChar* s);
 111         void startDocument(const xmlChar* version, const xmlChar* encoding, int standalone);
 112         void internalSubset(const xmlChar* name, const xmlChar* externalID, const xmlChar* systemID);
 113         void endDocument();
 114 #endif
 115     private:
 116         friend bool parseXMLDocumentFragment(const String& chunk, DocumentFragment* fragment, Element* parent);
 117
 118         static void eventuallyMarkAsParserCreated(Element* element);
 119         void initializeParserContext(const char* chunk = 0);
 120         void setCurrentNode(Node*);
 121
 122         void insertErrorMessageBlock();
 123
 124         bool enterText();
 125         void exitText();
 126
 127         void doWrite(const String&);
 128         void doEnd();
 129
 130         Document* m_doc;
 131         FrameView* m_view;
 132
 133         String m_originalSourceForTransform;
 134
 135 #if USE(QXMLSTREAM)
 136         QXmlStreamReader m_stream;
 137         bool m_wroteText;
 138 #else
 139         xmlParserCtxtPtr m_context;
 140         OwnPtr<PendingCallbacks> m_pendingCallbacks;
 141         Vector<xmlChar> m_bufferedText;
 142 #endif
 143         Node* m_currentNode;
 144         bool m_currentNodeIsReferenced;
 145
 146         bool m_sawError;
 147         bool m_sawXSLTransform;
 148         bool m_sawFirstElement;
 149         bool m_isXHTMLDocument;
 150
 151         bool m_parserPaused;
 152         bool m_requestingScript;
 153         bool m_finishCalled;
 154
 155         int m_errorCount;
 156         int m_lastErrorLine;
 157         int m_lastErrorColumn;
 158         String m_errorMessages;
 159
 160         CachedResourceHandle<CachedScript> m_pendingScript;
 161         RefPtr<Element> m_scriptElement;
 162         int m_scriptStartLine;
 163
 164         bool m_parsingFragment;
 165         String m_defaultNamespaceURI;
 166
 167         typedef HashMap<String, String> PrefixForNamespaceMap;
 168         PrefixForNamespaceMap m_prefixToNamespaceMap;
 169         SegmentedString m_pendingSrc;
 170     };
 171
 172 #if ENABLE(XSLT)
 173 void* xmlDocPtrForString(DocLoader*, const String& source, const String& url);
 174 void setLoaderForLibXMLCallbacks(DocLoader*);
 175 #endif
 176
 177 HashMap<String, String> parseAttributes(const String&, bool& attrsOK);
 178 bool parseXMLDocumentFragment(const String&, DocumentFragment*, Element* parent = 0);
 179
 180 bool isScriptElement(Element*);
 181 ScriptElement* castToScriptElement(Element*);
 182
 183 } // namespace WebCore
 184
 185 #endif // XMLTokenizer_h