X-Git-Url: https://git.saurik.com/wxWidgets.git/blobdiff_plain/1309ba6c7ec76b4507cf1f2e8656749e3de1befa..c6ea1fdaf924be4ee49a19d12f76e4fcd257bb0a:/include/wx/html/htmlpars.h diff --git a/include/wx/html/htmlpars.h b/include/wx/html/htmlpars.h index 272c44ad5e..e654cb3f0a 100644 --- a/include/wx/html/htmlpars.h +++ b/include/wx/html/htmlpars.h @@ -1,44 +1,59 @@ ///////////////////////////////////////////////////////////////////////////// -// Name: htmlpars.h +// Name: wx/html/htmlpars.h // Purpose: wxHtmlParser class (generic parser) // Author: Vaclav Slavik // RCS-ID: $Id$ // Copyright: (c) 1999 Vaclav Slavik -// Licence: wxWindows Licence +// Licence: wxWindows licence ///////////////////////////////////////////////////////////////////////////// - #ifndef _WX_HTMLPARS_H_ #define _WX_HTMLPARS_H_ -#ifdef __GNUG__ -#pragma interface -#endif - #include "wx/defs.h" #if wxUSE_HTML #include "wx/html/htmltag.h" #include "wx/filesys.h" +#include "wx/hashmap.h" +#include "wx/hashset.h" +#include "wx/vector.h" +#include "wx/fontenc.h" + +class WXDLLIMPEXP_FWD_BASE wxMBConv; +class WXDLLIMPEXP_FWD_HTML wxHtmlParser; +class WXDLLIMPEXP_FWD_HTML wxHtmlTagHandler; +class WXDLLIMPEXP_FWD_HTML wxHtmlEntitiesParser; + +class wxHtmlTextPieces; +class wxHtmlParserState; -class wxHtmlParser; -class wxHtmlTagHandler; +WX_DECLARE_HASH_SET_WITH_DECL(wxHtmlTagHandler*, + wxPointerHash, wxPointerEqual, + wxHtmlTagHandlersSet, + class WXDLLIMPEXP_HTML); +WX_DECLARE_STRING_HASH_MAP_WITH_DECL(wxHtmlTagHandler*, + wxHtmlTagHandlersHash, + class WXDLLIMPEXP_HTML); -//-------------------------------------------------------------------------------- -// wxHtmlParser -// This class handles generic parsing of HTML document : it scans -// the document and divide it into blocks of tags (where one block -// consists of starting and ending tag and of text between these -// 2 tags. -//-------------------------------------------------------------------------------- -class WXDLLEXPORT wxHtmlParser : public wxObject +enum wxHtmlURLType +{ + wxHTML_URL_PAGE, + wxHTML_URL_IMAGE, + wxHTML_URL_OTHER +}; + +// This class handles generic parsing of HTML document : it scans +// the document and divides it into blocks of tags (where one block +// consists of starting and ending tag and of text between these +// 2 tags. +class WXDLLIMPEXP_HTML wxHtmlParser : public wxObject { DECLARE_ABSTRACT_CLASS(wxHtmlParser) public: - wxHtmlParser() : wxObject(), m_HandlersHash(wxKEY_STRING) - { m_FS = NULL; m_Cache = NULL; m_HandlersStack = NULL; } + wxHtmlParser(); virtual ~wxHtmlParser(); // Sets the class which will be used for opening files @@ -46,11 +61,15 @@ public: 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(); (it's return value is then returned) + // 3. call GetProduct(); (its return value is then returned) // 4. call DoneParser(); wxObject* Parse(const wxString& source); @@ -59,10 +78,17 @@ public: // 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(int begin_pos, int end_pos); - inline void DoParsing() {DoParsing(0, m_Source.Length());}; + 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 @@ -73,7 +99,7 @@ public: // 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). + // 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: // @@ -81,39 +107,77 @@ public: // // // 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") + // handler can handle only 'myitems' (e.g. its GetSupportedTags returns "MYITEMS") // you can call PushTagHandler(handler, "IT") when you find // and call PopTagHandler() when you find - void PushTagHandler(wxHtmlTagHandler *handler, wxString tags); + void PushTagHandler(wxHtmlTagHandler *handler, const wxString& tags); // Restores state before last call to PushTagHandler void PopTagHandler(); - wxString* GetSource() {return &m_Source;} + 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,
 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(); + + // 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 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) - // NOTE : using char* because of speed improvements - virtual void AddText(const char* txt) = 0; + // This is called from Parse() and must be overridden 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 + // This is called from Parse() and may be overridden. + // Default behaviour 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: - // source being parsed - wxString m_Source; + // DOM tree: + wxHtmlTag *m_CurTag; + wxHtmlTag *m_Tags; + wxHtmlTextPieces *m_TextPieces; + size_t m_CurTextPiece; - // tags cache, used during parsing. - wxHtmlTagsCache *m_Cache; - wxHashTable m_HandlersHash; + const wxString *m_Source; + + wxHtmlParserState *m_SavedStates; // handlers that handle particular tags. The table is accessed by // key = tag's name. @@ -124,31 +188,33 @@ protected: // 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; + wxHtmlTagHandlersSet m_HandlersSet; + wxHtmlTagHandlersHash m_HandlersHash; + + wxDECLARE_NO_COPY_CLASS(wxHtmlParser); // class for opening files (file system) wxFileSystem *m_FS; // handlers stack used by PushTagHandler and PopTagHandler - wxList *m_HandlersStack; -}; - - + wxVector m_HandlersStack; + // entity parse + wxHtmlEntitiesParser *m_entitiesParser; + // flag indicating that the parser should stop + bool m_stopParsing; +}; -//-------------------------------------------------------------------------------- -// 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 -//-------------------------------------------------------------------------------- -class WXDLLEXPORT wxHtmlTagHandler : public wxObject +// 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 its public methods) +// 2. Parser parses source between starting and ending tag +// 3. Handler restores original state of the parser +class WXDLLIMPEXP_HTML wxHtmlTagHandler : public wxObject { DECLARE_ABSTRACT_CLASS(wxHtmlTagHandler) @@ -159,7 +225,7 @@ public: // 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) + virtual void SetParser(wxHtmlParser *parser) { m_Parser = parser; } // Returns list of supported tags. The list is in uppercase and @@ -170,20 +236,68 @@ public: // 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 + // 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()); } + void ParseInner(const wxHtmlTag& tag) + { m_Parser->DoParsing(tag.GetBeginIter(), tag.GetEndIter1()); } + + // Parses given source as if it was tag's inner code (see + // wxHtmlParser::GetInnerSource). Unlike ParseInner(), this method lets + // you specify the source code to parse. This is useful when you need to + // modify the inner text before parsing. + void ParseInnerSource(const wxString& source); wxHtmlParser *m_Parser; + + wxDECLARE_NO_COPY_CLASS(wxHtmlTagHandler); }; +// 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 WXDLLIMPEXP_HTML wxHtmlEntitiesParser : public wxObject +{ + DECLARE_DYNAMIC_CLASS(wxHtmlEntitiesParser) + +public: + wxHtmlEntitiesParser(); + virtual ~wxHtmlEntitiesParser(); + + // Sets encoding of output string. + // Has no effect if wxUSE_UNICODE==1 +#if wxUSE_UNICODE + void SetEncoding(wxFontEncoding WXUNUSED(encoding)) {} +#else + void SetEncoding(wxFontEncoding encoding); +#endif + + // Parses entities in input and replaces them with respective characters + // (with respect to output encoding) + wxString Parse(const wxString& input) const; + + // Returns character for given entity or 0 if the enity is unknown + wxChar GetEntityChar(const wxString& entity) const; + + // Returns character that represents given Unicode code +#if wxUSE_UNICODE + wxChar GetCharForCode(unsigned code) const { return (wxChar)code; } +#else + wxChar GetCharForCode(unsigned code) const; +#endif + +protected: +#if !wxUSE_UNICODE + wxMBConv *m_conv; + wxFontEncoding m_encoding; +#endif + + wxDECLARE_NO_COPY_CLASS(wxHtmlEntitiesParser); +}; #endif