include/wx/html/htmlparser.h

   1 /////////////////////////////////////////////////////////////////////////////
   2 // Name:        htmlparser.h
   3 // Purpose:     wxHtmlParser class (generic parser)
   4 // Author:      Vaclav Slavik
   5 // Copyright:   (c) 1999 Vaclav Slavik
   6 // Licence:     wxWindows Licence
   7 /////////////////////////////////////////////////////////////////////////////
   8
   9
  10 #ifndef __HTMLPARSER_H__
  11 #define __HTMLPARSER_H__
  12
  13 #ifdef __GNUG__
  14 #pragma interface
  15 #endif
  16
  17 #include "wx/defs.h"
  18 #if wxUSE_HTML
  19
  20 #include <wx/html/htmltag.h>
  21 #include <wx/filesys.h>
  22
  23 class wxHtmlParser;
  24 class wxHtmlTagHandler;
  25
  26 //--------------------------------------------------------------------------------
  27 // wxHtmlParser
  28 //                  This class handles generic parsing of HTML document : it scans
  29 //                  the document and divide it into blocks of tags (where one block
  30 //                  consists of starting and ending tag and of text between these
  31 //                  2 tags.
  32 //--------------------------------------------------------------------------------
  33
  34 class WXDLLEXPORT wxHtmlParser : public wxObject
  35 {
  36     DECLARE_ABSTRACT_CLASS(wxHtmlParser)
  37
  38     protected:
  39         wxString m_Source;
  40                 // source being parsed
  41         wxHtmlTagsCache *m_Cache;
  42                 // tags cache, used during parsing.
  43         wxHashTable m_HandlersHash;
  44         wxList m_HandlersList;
  45                 // handlers that handle particular tags. The table is accessed by
  46                 // key = tag's name.
  47                 // This attribute MUST be filled by derived class otherwise it would
  48                 // be empty and no tags would be recognized
  49                 // (see wxHtmlWinParser for details about filling it)
  50                 // m_HandlersHash is for random access based on knowledge of tag name (BR, P, etc.)
  51                 //      it may (and often does) contain more references to one object
  52                 // m_HandlersList is list of all handlers and it is guaranteed to contain
  53                 //      only one reference to each handler instance.
  54         wxFileSystem *m_FS;
  55                 // class for opening files (file system)
  56
  57     public:
  58         wxHtmlParser() : wxObject(), m_HandlersHash(wxKEY_STRING) {m_FS = NULL; m_Cache = NULL;}
  59         virtual ~wxHtmlParser();
  60
  61         void SetFS(wxFileSystem *fs) {m_FS = fs;}
  62                 // Sets the class which will be used for opening files
  63         wxFileSystem* GetFS() const {return m_FS;}
  64
  65         wxObject* Parse(const wxString& source);
  66                 // You can simply call this method when you need parsed output.
  67                 // This method does these things:
  68                 // 1. call InitParser(source);
  69                 // 2. call DoParsing();
  70                 // 3. call GetProduct(); (it's return value is then returned)
  71                 // 4. call DoneParser();
  72
  73         virtual void InitParser(const wxString& source);
  74                 // Sets the source. This must be called before running Parse() method.
  75         virtual void DoneParser();
  76                 // This must be called after Parse().
  77
  78         void DoParsing(int begin_pos, int end_pos);
  79         inline void DoParsing() {DoParsing(0, m_Source.Length());};
  80                 // Parses the m_Source from begin_pos to end_pos-1.
  81                 // (in noparams version it parses whole m_Source)
  82
  83         virtual wxObject* GetProduct() = 0;
  84                 // Returns product of parsing
  85                 // Returned value is result of parsing of the part. The type of this result
  86                 // depends on internal representation in derived parser
  87                 // (see wxHtmlWinParser for details).
  88
  89         virtual void AddTagHandler(wxHtmlTagHandler *handler);
  90                 // adds handler to the list & hash table of handlers.
  91
  92         wxString* GetSource() {return &m_Source;}
  93
  94         virtual wxList* GetTempData() {return NULL;}
  95                 // this method returns list of wxObjects that represents
  96                 // all data allocated by the parser. These can't be freeded
  97                 // by destructor because they must be valid as long as
  98                 // GetProduct's return value is valid - the caller must
  99                 // explicitly call delete MyParser -> GetTempData() to free
 100                 // the memory
 101                 // (this method always sets the list to delete its contents)
 102
 103     protected:
 104
 105         virtual void AddText(const char* txt) = 0;
 106                 // Adds text to the output.
 107                 // This is called from Parse() and must be overriden in derived classes.
 108                 // txt is not guaranteed to be only one word. It is largest continuous part of text
 109                 // (= not broken by tags)
 110                 // NOTE : using char* because of speed improvements
 111
 112         virtual void AddTag(const wxHtmlTag& tag);
 113                 // Adds tag and proceeds it. Parse() may (and usually is) called from this method.
 114                 // This is called from Parse() and may be overriden.
 115                 // Default behavior is that it looks for proper handler in m_Handlers. The tag is
 116                 // ignored if no hander is found.
 117                 // Derived class is *responsible* for filling in m_Handlers table.
 118 };
 119
 120
 121
 122
 123
 124
 125 //--------------------------------------------------------------------------------
 126 // wxHtmlTagHandler
 127 //                  This class (and derived classes) cooperates with wxHtmlParser.
 128 //                  Each recognized tag is passed to handler which is capable
 129 //                  of handling it. Each tag is handled in 3 steps:
 130 //                  1. Handler will modifies state of parser
 131 //                    (using it's public methods)
 132 //                  2. Parser parses source between starting and ending tag
 133 //                  3. Handler restores original state of the parser
 134 //--------------------------------------------------------------------------------
 135
 136 class WXDLLEXPORT wxHtmlTagHandler : public wxObject
 137 {
 138     DECLARE_ABSTRACT_CLASS(wxHtmlTagHandler)
 139
 140     protected:
 141         wxHtmlParser *m_Parser;
 142
 143     public:
 144         wxHtmlTagHandler() : wxObject () {m_Parser = NULL;};
 145
 146         virtual void SetParser(wxHtmlParser *parser) {m_Parser = parser;}
 147                 // Sets the parser.
 148                 // NOTE : each _instance_ of handler is guaranteed to be called
 149                 // only by one parser. This means you don't have to care about
 150                 // reentrancy.
 151
 152         virtual wxString GetSupportedTags() = 0;
 153                 // Returns list of supported tags. The list is in uppercase and
 154                 // tags are delimited by ','.
 155                 // Example : "I,B,FONT,P"
 156                 //   is capable of handling italic, bold, font and paragraph tags
 157
 158         virtual bool HandleTag(const wxHtmlTag& tag) = 0;
 159                 // This is hadling core method. It does all the Steps 1-3.
 160                 // To process step 2, you can call ParseInner()
 161                 // returned value : TRUE if it called ParseInner(),
 162                 //                  FALSE etherwise
 163
 164     protected:
 165         void ParseInner(const wxHtmlTag& tag) {m_Parser -> DoParsing(tag.GetBeginPos(), tag.GetEndPos1());}
 166                 // parses input between beginning and ending tag.
 167                 // m_Parser must be set.
 168 };
 169
 170
 171
 172
 173
 174 #endif // __HTMLPARSER_H__
 175
 176 #endif