khtml/html/htmlparser.h

   1 /*
   2     This file is part of the KDE libraries
   3
   4     Copyright (C) 1997 Martin Jones (mjones@kde.org)
   5               (C) 1997 Torben Weis (weis@kde.org)
   6               (C) 1998 Waldo Bastian (bastian@kde.org)
   7               (C) 1999 Lars Knoll (knoll@kde.org)
   8               (C) 2003 Apple Computer, Inc.
   9
  10     This library is free software; you can redistribute it and/or
  11     modify it under the terms of the GNU Library General Public
  12     License as published by the Free Software Foundation; either
  13     version 2 of the License, or (at your option) any later version.
  14
  15     This library is distributed in the hope that it will be useful,
  16     but WITHOUT ANY WARRANTY; without even the implied warranty of
  17     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  18     Library General Public License for more details.
  19
  20     You should have received a copy of the GNU Library General Public License
  21     along with this library; see the file COPYING.LIB.  If not, write to
  22     the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
  23     Boston, MA 02110-1301, USA.
  24 */
  25 //----------------------------------------------------------------------------
  26 //
  27 // KDE HTML Widget -- HTML Parser
  28
  29 #ifndef HTMLPARSER_H
  30 #define HTMLPARSER_H
  31
  32 // 0 all
  33 // 1 domtree + rendertree + styleForElement, no layouting
  34 // 2 domtree only
  35 #define SPEED_DEBUG 0
  36
  37 #ifdef SPEED_DEBUG
  38 #include <QDateTime>
  39 #endif
  40
  41
  42 #include "dom/dom_string.h"
  43 #include "xml/dom_nodeimpl.h"
  44 #include "html/html_documentimpl.h"
  45
  46 class KHTMLView;
  47 class HTMLStackElem;
  48
  49 namespace DOM {
  50     class HTMLDocumentImpl;
  51     class DocumentImpl;
  52     class NodeImpl;
  53     class HTMLFormElementImpl;
  54     class HTMLMapElementImpl;
  55     class HTMLHeadElementImpl;
  56     class DocumentFragmentImpl;
  57 }
  58
  59 namespace khtml {
  60
  61 class Token;
  62 class DoctypeToken;
  63
  64 /**
  65  * The parser for html. It receives a stream of tokens from the HTMLTokenizer, and
  66  * builds up the Document structure form it.
  67  */
  68 class KHTMLParser
  69 {
  70 public:
  71     KHTMLParser( KHTMLView *w, DOM::DocumentImpl *i );
  72     KHTMLParser( DOM::DocumentFragmentImpl *frag, DOM::DocumentImpl *doc );
  73     virtual ~KHTMLParser();
  74
  75     /**
  76      * parses one token delivered by the tokenizer
  77      */
  78     void parseToken(Token *_t);
  79
  80     /**
  81      * parses a doctype token delivered by the tokenizer
  82      */
  83     void parseDoctypeToken(DoctypeToken *_t);
  84
  85     /**
  86      * resets the parser
  87      */
  88     void reset();
  89
  90     bool skipMode() const { return (discard_until != 0); }
  91     bool noSpaces() const { return (inSelect || !m_inline  || !inBody); }
  92     bool selectMode() const { return inSelect; }
  93
  94     DOM::HTMLDocumentImpl *doc() const { return static_cast<DOM::HTMLDocumentImpl *>(document); }
  95     DOM::DocumentImpl *docPtr() const { return document; }
  96
  97 protected:
  98
  99     KHTMLView *HTMLWidget;
 100     DOM::DocumentImpl *document;
 101
 102     /*
 103      * generate an element from the token
 104      */
 105     DOM::NodeImpl *getElement(Token *);
 106
 107     void processCloseTag(Token *);
 108
 109     bool insertNode(DOM::NodeImpl *n, bool flat = false);
 110
 111     /*
 112      * The currently active element (the one new elements will be added to)
 113      */
 114     void setCurrent( DOM::NodeImpl* newNode )
 115     {
 116         if ( newNode ) newNode->ref();
 117         if ( current ) current->deref();
 118         current = newNode;
 119     }
 120
 121 private:
 122     DOM::NodeImpl *current;
 123
 124     HTMLStackElem *blockStack;
 125
 126     void pushBlock( int _id, int _level);
 127
 128     void generateImpliedEndTags( int _id );
 129     void popOptionalBlock( int _id );
 130     void popBlock( int _id );
 131     void popOneBlock(bool delBlock = true);
 132     void popInlineBlocks();
 133
 134     void freeBlock( void);
 135
 136     void createHead();
 137
 138     bool isResidualStyleTag(int _id);
 139     bool isAffectedByResidualStyle(int _id);
 140     void handleResidualStyleCloseTagAcrossBlocks(HTMLStackElem* elem);
 141     void reopenResidualStyleTags(HTMLStackElem* elem, DOM::NodeImpl* malformedTableParent);
 142
 143     ushort *forbiddenTag;
 144
 145     /*
 146      * currently active form
 147      */
 148     DOM::HTMLFormElementImpl *form;
 149
 150     /*
 151      * current map
 152      */
 153     DOM::HTMLMapElementImpl *map;
 154
 155     /*
 156      * the head element. Needed for crappy html which defines <base> after </head>
 157      */
 158     DOM::HTMLHeadElementImpl *head;
 159
 160     /*
 161      * a possible <isindex> element in the head. Compatibility hack for
 162      * html from the stone age
 163      */
 164     DOM::NodeImpl *isindex;
 165     DOM::NodeImpl *handleIsindex( Token *t );
 166
 167     /*
 168      * inserts the stupid isIndex element.
 169      */
 170     void startBody();
 171
 172     bool inBody;
 173     bool haveContent;
 174     bool haveBody;
 175     bool haveFrameSet;
 176     bool haveTitle;
 177     bool m_inline;
 178     bool end;
 179     bool inSelect;
 180
 181
 182     /*
 183      * tells the parser to discard all tags, until it reaches the one specified
 184      */
 185     int discard_until;
 186
 187     bool headLoaded;
 188     int inStrayTableContent;
 189
 190 #if SPEED_DEBUG > 0
 191     QTime qt;
 192 #endif
 193 };
 194
 195 } // namespace khtml
 196
 197 #endif // HTMLPARSER_H
 198