WebCore/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 #if ENABLE(WML)
  76         bool isWMLDocument() const;
  77 #endif
  78
  79         // from CachedResourceClient
  80         virtual void notifyFinished(CachedResource* finishedObj);
  81
  82
  83         void handleError(ErrorType type, const char* m, int lineNumber, int columnNumber);
  84
  85         virtual bool wellFormed() const { return !m_sawError; }
  86
  87         int lineNumber() const;
  88         int columnNumber() const;
  89
  90 #if USE(QXMLSTREAM)
  91 private:
  92         void parse();
  93         void startDocument();
  94         void parseStartElement();
  95         void parseEndElement();
  96         void parseCharacters();
  97         void parseProcessingInstruction();
  98         void parseCdata();
  99         void parseComment();
 100         void endDocument();
 101         void parseDtd();
 102         bool hasError() const;
 103 #else
 104 public:
 105         // callbacks from parser SAX
 106         void error(ErrorType, const char* message, va_list args) WTF_ATTRIBUTE_PRINTF(3, 0);
 107         void startElementNs(const xmlChar* xmlLocalName, const xmlChar* xmlPrefix, const xmlChar* xmlURI, int nb_namespaces,
 108                             const xmlChar** namespaces, int nb_attributes, int nb_defaulted, const xmlChar** libxmlAttributes);
 109         void endElementNs();
 110         void characters(const xmlChar* s, int len);
 111         void processingInstruction(const xmlChar* target, const xmlChar* data);
 112         void cdataBlock(const xmlChar* s, int len);
 113         void comment(const xmlChar* s);
 114         void startDocument(const xmlChar* version, const xmlChar* encoding, int standalone);
 115         void internalSubset(const xmlChar* name, const xmlChar* externalID, const xmlChar* systemID);
 116         void endDocument();
 117 #endif
 118     private:
 119         friend bool parseXMLDocumentFragment(const String& chunk, DocumentFragment* fragment, Element* parent);
 120
 121         void initializeParserContext(const char* chunk = 0);
 122         void setCurrentNode(Node*);
 123
 124         void insertErrorMessageBlock();
 125
 126         bool enterText();
 127         void exitText();
 128
 129         void doWrite(const String&);
 130         void doEnd();
 131
 132         Document* m_doc;
 133         FrameView* m_view;
 134
 135         String m_originalSourceForTransform;
 136
 137 #if USE(QXMLSTREAM)
 138         QXmlStreamReader m_stream;
 139         bool m_wroteText;
 140 #else
 141         xmlParserCtxtPtr m_context;
 142         OwnPtr<PendingCallbacks> m_pendingCallbacks;
 143         Vector<xmlChar> m_bufferedText;
 144 #endif
 145         Node* m_currentNode;
 146         bool m_currentNodeIsReferenced;
 147
 148         bool m_sawError;
 149         bool m_sawXSLTransform;
 150         bool m_sawFirstElement;
 151         bool m_isXHTMLDocument;
 152
 153         bool m_parserPaused;
 154         bool m_requestingScript;
 155         bool m_finishCalled;
 156
 157         int m_errorCount;
 158         int m_lastErrorLine;
 159         int m_lastErrorColumn;
 160         String m_errorMessages;
 161
 162         CachedResourceHandle<CachedScript> m_pendingScript;
 163         RefPtr<Element> m_scriptElement;
 164         int m_scriptStartLine;
 165
 166         bool m_parsingFragment;
 167         String m_defaultNamespaceURI;
 168
 169         typedef HashMap<String, String> PrefixForNamespaceMap;
 170         PrefixForNamespaceMap m_prefixToNamespaceMap;
 171         SegmentedString m_pendingSrc;
 172     };
 173
 174 #if ENABLE(XSLT)
 175 void* xmlDocPtrForString(DocLoader*, const String& source, const String& url);
 176 #endif
 177
 178 HashMap<String, String> parseAttributes(const String&, bool& attrsOK);
 179 bool parseXMLDocumentFragment(const String&, DocumentFragment*, Element* parent = 0);
 180
 181 } // namespace WebCore
 182
 183 #endif // XMLTokenizer_h