khtml/xml/xml_tokenizer.h

   1 /*
   2  * This file is part of the DOM implementation for KDE.
   3  *
   4  * Copyright (C) 2000 Peter Kelly (pmk@post.com)
   5  *
   6  * This library is free software; you can redistribute it and/or
   7  * modify it under the terms of the GNU Library General Public
   8  * License as published by the Free Software Foundation; either
   9  * version 2 of the License, or (at your option) any later version.
  10  *
  11  * This library is distributed in the hope that it will be useful,
  12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  14  * Library General Public License for more details.
  15  *
  16  * You should have received a copy of the GNU Library General Public License
  17  * along with this library; see the file COPYING.LIB.  If not, write to
  18  * the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
  19  * Boston, MA 02110-1301, USA.
  20  *
  21  */
  22
  23 #ifndef _XML_Tokenizer_h_
  24 #define _XML_Tokenizer_h_
  25
  26 #include <QtXml/qxml.h>
  27 #include <QtCore/QStack>
  28 #include <QtCore/QLinkedList>
  29 #include <QtCore/QMap>
  30 #include <QtCore/QObject>
  31 #include "misc/loader_client.h"
  32 #include "misc/stringit.h"
  33
  34 class KHTMLView;
  35
  36 namespace khtml {
  37     class CachedObject;
  38     class CachedScript;
  39 }
  40
  41 namespace DOM {
  42     class DocumentImpl;
  43     class NodeImpl;
  44     class HTMLScriptElementImpl;
  45     class DocumentImpl;
  46     class HTMLScriptElementImpl;
  47 }
  48
  49 namespace khtml {
  50
  51 class XMLHandler : public QXmlDefaultHandler
  52 {
  53 public:
  54     XMLHandler(DOM::DocumentImpl *_doc, KHTMLView *_view);
  55     virtual ~XMLHandler();
  56
  57     // return the error protocol if parsing failed
  58     QString errorProtocol();
  59
  60     // overloaded handler functions
  61     bool startDocument();
  62     bool startElement(const QString& namespaceURI, const QString& localName, const QString& qName, const QXmlAttributes& atts);
  63     bool endElement(const QString& namespaceURI, const QString& localName, const QString& qName);
  64     bool startCDATA();
  65     bool endCDATA();
  66     bool characters(const QString& ch);
  67     bool comment(const QString & ch);
  68     bool processingInstruction(const QString &target, const QString &data);
  69
  70     // namespace handling, to workaround problem in QXML where some attributes
  71     // do not get the namespace resolved properly
  72     bool startPrefixMapping(const QString& prefix, const QString& uri);
  73     bool endPrefixMapping(const QString& prefix);
  74     void fixUpNSURI(QString& uri, const QString& qname);
  75     QMap<QString, QStack<QString> > namespaceInfo;
  76
  77
  78     // from QXmlDeclHandler
  79     bool attributeDecl(const QString &eName, const QString &aName, const QString &type, const QString &valueDefault, const QString &value);
  80     bool externalEntityDecl(const QString &name, const QString &publicId, const QString &systemId);
  81     bool internalEntityDecl(const QString &name, const QString &value);
  82
  83     // from QXmlDTDHandler
  84     bool notationDecl(const QString &name, const QString &publicId, const QString &systemId);
  85     bool unparsedEntityDecl(const QString &name, const QString &publicId, const QString &systemId, const QString &notationName);
  86
  87     bool enterText();
  88     void exitText();
  89
  90     QString errorString() const;
  91
  92     bool fatalError( const QXmlParseException& exception );
  93
  94     unsigned long errorLine;
  95     unsigned long errorCol;
  96
  97 private:
  98     void pushNode( DOM::NodeImpl *node );
  99     DOM::NodeImpl *popNode();
 100     DOM::NodeImpl *currentNode() const;
 101 private:
 102     QString errorProt;
 103     DOM::DocumentImpl *m_doc;
 104     KHTMLView *m_view;
 105     QStack<DOM::NodeImpl*> m_nodes;
 106     DOM::NodeImpl *m_rootNode;
 107
 108     enum State {
 109         StateInit,
 110         StateDocument,
 111         StateQuote,
 112         StateLine,
 113         StateHeading,
 114         StateP
 115     };
 116     State state;
 117 };
 118
 119 class Tokenizer : public QObject
 120 {
 121     Q_OBJECT
 122 public:
 123     virtual void begin() = 0;
 124     // script output must be prepended, while new data
 125     // received during executing a script must be appended, hence the
 126     // extra bool to be able to distinguish between both cases. document.write()
 127     // always uses false, while khtmlpart uses true
 128     virtual void write( const TokenizerString &str, bool appendData) = 0;
 129     virtual void end() = 0;
 130     virtual void finish() = 0;
 131     virtual void setOnHold(bool /*_onHold*/) {}
 132     virtual bool isWaitingForScripts() const = 0;
 133     virtual bool isExecutingScript() const = 0;
 134     virtual void setNormalYeldDelay() {}
 135     virtual void abort() {}
 136     virtual void setAutoClose(bool b=true) = 0;
 137
 138 Q_SIGNALS:
 139     void finishedParsing();
 140
 141 };
 142
 143 class XMLIncrementalSource : public QXmlInputSource
 144 {
 145 public:
 146     XMLIncrementalSource();
 147     virtual void fetchData();
 148     virtual QChar next();
 149     virtual void setData( const QString& str );
 150     virtual void setData( const QByteArray& data );
 151     virtual QString data() const;
 152
 153     void appendXML( const QString& str );
 154     void setFinished( bool );
 155
 156     inline void setPaused(bool paused = true) { m_paused = paused; }
 157
 158 private:
 159     QString      m_data;
 160     int          m_pos;
 161     const QChar *m_unicode;
 162     bool         m_finished;
 163     bool         m_paused; // if waiting for scripts
 164 };
 165
 166 class XMLTokenizer : public Tokenizer, public khtml::CachedObjectClient
 167 {
 168 public:
 169     XMLTokenizer(DOM::DocumentImpl *, KHTMLView * = 0);
 170     virtual ~XMLTokenizer();
 171     virtual void begin();
 172     virtual void write( const TokenizerString &str, bool );
 173     virtual void end();
 174     virtual void finish();
 175     virtual void setAutoClose(bool b=true) { qWarning("XMLTokenizer::setAutoClose: stub."); (void)b; }
 176
 177     // from CachedObjectClient
 178     void notifyFinished(khtml::CachedObject *finishedObj);
 179
 180     virtual bool isWaitingForScripts() const;
 181     virtual bool isExecutingScript() const { return m_executingScript; }
 182
 183     // execute script in place, if it contains src attribute we stop parsing till it's downloaded
 184     void executeScript(DOM::NodeImpl *n);
 185
 186 protected:
 187     DOM::DocumentImpl *m_doc;
 188     KHTMLView *m_view;
 189
 190     khtml::CachedScript *m_cachedScript;
 191
 192     QString m_bufferedData;
 193
 194     XMLHandler m_handler;
 195     QXmlSimpleReader m_reader;
 196     XMLIncrementalSource m_source;
 197     bool m_noErrors;
 198     bool m_executingScript;
 199     bool m_explicitFinishParsingNeeded;
 200     bool m_insideWrite;
 201 };
 202
 203 } // end namespace
 204
 205 #endif