This last it has different meaning, we don't want it to be parsed by myitems handler!
+ // handler can handle only 'myitems' (e.g. it's GetSupportedTags returns "MYITEMS")
+ // you can call PushTagHandler(handler, "IT") when you find
+ // and call PopTagHandler() when you find
+ void PushTagHandler(wxHtmlTagHandler *handler, wxString tags);
+
+ // Restores state before last call to PushTagHandler
+ void PopTagHandler();
+
+ wxString* GetSource() {return &m_Source;}
+ void SetSource(const wxString& src);
+
+ // Sets HTML source and remebers current parser's state so that it can
+ // later be restored. This is useful for on-line modifications of
+ // HTML source (for example, handler replaces spaces with
+ // and newlines with
)
+ virtual void SetSourceAndSaveState(const wxString& src);
+ // Restores parser's state from stack or returns FALSE if the stack is
+ // empty
+ virtual bool RestoreState();
+
+protected:
+ // DOM structure
+ void CreateDOMTree();
+ void DestroyDOMTree();
+ void CreateDOMSubTree(wxHtmlTag *cur,
+ int begin_pos, int end_pos,
+ wxHtmlTagsCache *cache);
+
+ // Adds text to the output.
+ // This is called from Parse() and must be overriden in derived classes.
+ // txt is not guaranteed to be only one word. It is largest continuous part of text
+ // (= not broken by tags)
+ // NOTE : using char* because of speed improvements
+ virtual void AddText(const wxChar* txt) = 0;
+
+ // Adds tag and proceeds it. Parse() may (and usually is) called from this method.
+ // This is called from Parse() and may be overriden.
+ // Default behavior is that it looks for proper handler in m_Handlers. The tag is
+ // ignored if no hander is found.
+ // Derived class is *responsible* for filling in m_Handlers table.
+ virtual void AddTag(const wxHtmlTag& tag);
+
+ // Returns entity parser object, used to substitute HTML &entities;
+ wxHtmlEntitiesParser *GetEntitiesParser() const { return m_entitiesParser; }
+
+protected:
+ // DOM tree:
+ wxHtmlTag *m_CurTag;
+ wxHtmlTag *m_Tags;
+ wxHtmlTextPieces *m_TextPieces;
+ size_t m_CurTextPiece;
+
+ wxString m_Source;
+
+ wxHtmlParserState *m_SavedStates;
+
+ // handlers that handle particular tags. The table is accessed by
+ // key = tag's name.
+ // This attribute MUST be filled by derived class otherwise it would
+ // be empty and no tags would be recognized
+ // (see wxHtmlWinParser for details about filling it)
+ // m_HandlersHash is for random access based on knowledge of tag name (BR, P, etc.)
+ // it may (and often does) contain more references to one object
+ // m_HandlersList is list of all handlers and it is guaranteed to contain
+ // only one reference to each handler instance.
+ wxList m_HandlersList;
+ wxHashTable m_HandlersHash;
+
+ // class for opening files (file system)
+ wxFileSystem *m_FS;
+ // handlers stack used by PushTagHandler and PopTagHandler
+ wxList *m_HandlersStack;
+
+ // entity parse
+ wxHtmlEntitiesParser *m_entitiesParser;
};
-
-
-
-//--------------------------------------------------------------------------------
-// wxHtmlTagHandler
-// This class (and derived classes) cooperates with wxHtmlParser.
-// Each recognized tag is passed to handler which is capable
-// of handling it. Each tag is handled in 3 steps:
-// 1. Handler will modifies state of parser
-// (using it's public methods)
-// 2. Parser parses source between starting and ending tag
-// 3. Handler restores original state of the parser
-//--------------------------------------------------------------------------------
-
+// This class (and derived classes) cooperates with wxHtmlParser.
+// Each recognized tag is passed to handler which is capable
+// of handling it. Each tag is handled in 3 steps:
+// 1. Handler will modifies state of parser
+// (using it's public methods)
+// 2. Parser parses source between starting and ending tag
+// 3. Handler restores original state of the parser
class WXDLLEXPORT wxHtmlTagHandler : public wxObject
{
DECLARE_ABSTRACT_CLASS(wxHtmlTagHandler)
- protected:
- wxHtmlParser *m_Parser;
-
- public:
- wxHtmlTagHandler() : wxObject () {m_Parser = NULL;};
-
- virtual void SetParser(wxHtmlParser *parser) {m_Parser = parser;}
- // Sets the parser.
- // NOTE : each _instance_ of handler is guaranteed to be called
- // only by one parser. This means you don't have to care about
- // reentrancy.
-
- virtual wxString GetSupportedTags() = 0;
- // Returns list of supported tags. The list is in uppercase and
- // tags are delimited by ','.
- // Example : "I,B,FONT,P"
- // is capable of handling italic, bold, font and paragraph tags
-
- virtual bool HandleTag(const wxHtmlTag& tag) = 0;
- // This is hadling core method. It does all the Steps 1-3.
- // To process step 2, you can call ParseInner()
- // returned value : TRUE if it called ParseInner(),
- // FALSE etherwise
-
- protected:
- void ParseInner(const wxHtmlTag& tag) {m_Parser -> DoParsing(tag.GetBeginPos(), tag.GetEndPos1());}
- // parses input between beginning and ending tag.
- // m_Parser must be set.
+public:
+ wxHtmlTagHandler() : wxObject () { m_Parser = NULL; }
+
+ // Sets the parser.
+ // NOTE : each _instance_ of handler is guaranteed to be called
+ // only by one parser. This means you don't have to care about
+ // reentrancy.
+ virtual void SetParser(wxHtmlParser *parser)
+ { m_Parser = parser; }
+
+ // Returns list of supported tags. The list is in uppercase and
+ // tags are delimited by ','.
+ // Example : "I,B,FONT,P"
+ // is capable of handling italic, bold, font and paragraph tags
+ virtual wxString GetSupportedTags() = 0;
+
+ // This is hadling core method. It does all the Steps 1-3.
+ // To process step 2, you can call ParseInner()
+ // returned value : TRUE if it called ParseInner(),
+ // FALSE etherwise
+ virtual bool HandleTag(const wxHtmlTag& tag) = 0;
+
+protected:
+ // parses input between beginning and ending tag.
+ // m_Parser must be set.
+ void ParseInner(const wxHtmlTag& tag)
+ { m_Parser->DoParsing(tag.GetBeginPos(), tag.GetEndPos1()); }
+
+ wxHtmlParser *m_Parser;
};
+// This class is used to parse HTML entities in strings. It can handle
+// both named entities and &#xxxx entries where xxxx is Unicode code.
+class WXDLLEXPORT wxHtmlEntitiesParser : public wxObject
+{
+ DECLARE_DYNAMIC_CLASS(wxHtmlEntitiesParser)
+
+public:
+ wxHtmlEntitiesParser();
+ virtual ~wxHtmlEntitiesParser();
+
+ // Sets encoding of output string.
+ // Has no effect if wxUSE_WCHAR_T==0 or wxUSE_UNICODE==1
+ void SetEncoding(wxFontEncoding encoding);
+
+ // Parses entities in input and replaces them with respective characters
+ // (with respect to output encoding)
+ wxString Parse(const wxString& input);
+
+protected:
+ wxChar GetEntityChar(const wxString& entity);
+ wxChar GetCharForCode(unsigned code);
+
+#if wxUSE_WCHAR_T && !wxUSE_UNICODE
+ wxMBConv *m_conv;
+ wxFontEncoding m_encoding;
+#endif
+};
#endif