+public:
+ wxHtmlParser();
+ virtual ~wxHtmlParser();
+
+ // Sets the class which will be used for opening files
+ void SetFS(wxFileSystem *fs) { m_FS = fs; }
+
+ wxFileSystem* GetFS() const { return m_FS; }
+
+ // Opens file if the parser is allowed to open given URL (may be forbidden
+ // for security reasons)
+ virtual wxFSFile *OpenURL(wxHtmlURLType type, const wxString& url) const;
+
+ // You can simply call this method when you need parsed output.
+ // This method does these things:
+ // 1. call InitParser(source);
+ // 2. call DoParsing();
+ // 3. call GetProduct(); (its return value is then returned)
+ // 4. call DoneParser();
+ wxObject* Parse(const wxString& source);
+
+ // Sets the source. This must be called before running Parse() method.
+ virtual void InitParser(const wxString& source);
+ // This must be called after Parse().
+ virtual void DoneParser();
+
+ // May be called during parsing to immediately return from Parse().
+ virtual void StopParsing() { m_stopParsing = true; }
+
+ // Parses the m_Source from begin_pos to end_pos-1.
+ // (in noparams version it parses whole m_Source)
+ void DoParsing(const wxString::const_iterator& begin_pos,
+ const wxString::const_iterator& end_pos);
+ void DoParsing();
+
+ // Returns pointer to the tag at parser's current position
+ wxHtmlTag *GetCurrentTag() const { return m_CurTag; }
+
+ // Returns product of parsing
+ // Returned value is result of parsing of the part. The type of this result
+ // depends on internal representation in derived parser
+ // (see wxHtmlWinParser for details).
+ virtual wxObject* GetProduct() = 0;
+
+ // adds handler to the list & hash table of handlers.
+ virtual void AddTagHandler(wxHtmlTagHandler *handler);
+
+ // Forces the handler to handle additional tags (not returned by GetSupportedTags).
+ // The handler should already be in use by this parser.
+ // Example: you want to parse following pseudo-html structure:
+ // <myitems>
+ // <it name="one" value="1">
+ // <it name="two" value="2">
+ // </myitems>
+ // <it> This last it has different meaning, we don't want it to be parsed by myitems handler!
+ // handler can handle only 'myitems' (e.g. its GetSupportedTags returns "MYITEMS")
+ // you can call PushTagHandler(handler, "IT") when you find <myitems>
+ // and call PopTagHandler() when you find </myitems>
+ void PushTagHandler(wxHtmlTagHandler *handler, const wxString& tags);
+
+ // Restores state before last call to PushTagHandler
+ void PopTagHandler();
+
+ const wxString* GetSource() {return m_Source;}
+ void SetSource(const wxString& src);
+
+ // Sets HTML source and remembers current parser's state so that it can
+ // later be restored. This is useful for on-line modifications of
+ // HTML source (for example, <pre> handler replaces spaces with
+ // and newlines with <br>)
+ virtual void SetSourceAndSaveState(const wxString& src);
+ // Restores parser's state from stack or returns false if the stack is
+ // empty
+ virtual bool RestoreState();
+
+ // Returns HTML source inside the element (i.e. between the starting
+ // and ending tag)
+ wxString GetInnerSource(const wxHtmlTag& tag);
+
+ // Parses HTML string 'markup' and extracts charset info from <meta> tag
+ // if present. Returns empty string if the tag is missing.
+ // For wxHTML's internal use.
+ static wxString ExtractCharsetInformation(const wxString& markup);
+
+ // Returns entity parser object, used to substitute HTML &entities;
+ wxHtmlEntitiesParser *GetEntitiesParser() const { return m_entitiesParser; }
+
+ // Returns true if the tag starting at the given position is a comment tag
+ //
+ // p should point to '<' character and is modified to point to the closing
+ // '>' of the end comment tag if this is indeed a comment
+ static bool
+ SkipCommentTag(wxString::const_iterator& p, wxString::const_iterator end);
+
+protected:
+ // DOM structure
+ void CreateDOMTree();
+ void DestroyDOMTree();
+ void CreateDOMSubTree(wxHtmlTag *cur,
+ const wxString::const_iterator& begin_pos,
+ const wxString::const_iterator& 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)
+ virtual void AddText(const wxString& 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);
+
+protected:
+ // DOM tree:
+ wxHtmlTag *m_CurTag;
+ wxHtmlTag *m_Tags;
+ wxHtmlTextPieces *m_TextPieces;
+ size_t m_CurTextPiece;
+
+ const 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.
+ wxHtmlTagHandlersSet m_HandlersSet;
+ wxHtmlTagHandlersHash m_HandlersHash;
+
+ DECLARE_NO_COPY_CLASS(wxHtmlParser)
+
+ // class for opening files (file system)
+ wxFileSystem *m_FS;
+ // handlers stack used by PushTagHandler and PopTagHandler
+ wxVector<wxHtmlTagHandlersHash*> m_HandlersStack;
+
+ // entity parse
+ wxHtmlEntitiesParser *m_entitiesParser;
+
+ // flag indicating that the parser should stop
+ bool m_stopParsing;
+};
projects / wxWidgets.git / blobdiff