// 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 "htmlpars.h"
-#endif
-
#include "wx/defs.h"
#if wxUSE_HTML
#include "wx/hash.h"
#include "wx/fontenc.h"
-class WXDLLEXPORT wxMBConv;
-class WXDLLEXPORT wxHtmlParser;
-class WXDLLEXPORT wxHtmlTagHandler;
-class WXDLLEXPORT wxHtmlEntitiesParser;
+class WXDLLIMPEXP_BASE wxMBConv;
+class WXDLLIMPEXP_HTML wxHtmlParser;
+class WXDLLIMPEXP_HTML wxHtmlTagHandler;
+class WXDLLIMPEXP_HTML wxHtmlEntitiesParser;
class wxHtmlTextPieces;
class wxHtmlParserState;
+
+enum wxHtmlURLType
+{
+ wxHTML_URL_PAGE,
+ wxHTML_URL_IMAGE,
+ wxHTML_URL_OTHER
+};
+
// 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
+class WXDLLIMPEXP_HTML wxHtmlParser : public wxObject
{
DECLARE_ABSTRACT_CLASS(wxHtmlParser)
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);
// 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);
// 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:
// <myitems>
// handler can handle only 'myitems' (e.g. it's GetSupportedTags returns "MYITEMS")
// you can call PushTagHandler(handler, "IT") when you find <myitems>
// and call PopTagHandler() when you find </myitems>
- 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;}
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
+ // 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
+ // 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; }
+
protected:
// DOM structure
void CreateDOMTree();
void DestroyDOMTree();
void CreateDOMSubTree(wxHtmlTag *cur,
- int begin_pos, int end_pos,
+ int begin_pos, int end_pos,
wxHtmlTagsCache *cache);
// Adds text to the output.
// 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:
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
wxList m_HandlersList;
wxHashTable m_HandlersHash;
+ DECLARE_NO_COPY_CLASS(wxHtmlParser)
+
// class for opening files (file system)
wxFileSystem *m_FS;
// handlers stack used by PushTagHandler and PopTagHandler
wxList *m_HandlersStack;
-
+
// entity parse
wxHtmlEntitiesParser *m_entitiesParser;
+
+ // flag indicating that the parser should stop
+ bool m_stopParsing;
};
// (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
+class WXDLLIMPEXP_HTML wxHtmlTagHandler : public wxObject
{
DECLARE_ABSTRACT_CLASS(wxHtmlTagHandler)
// 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
// 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)
+ void ParseInner(const wxHtmlTag& tag)
{ m_Parser->DoParsing(tag.GetBeginPos(), tag.GetEndPos1()); }
+ // 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;
+
+ DECLARE_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 WXDLLEXPORT wxHtmlEntitiesParser : public wxObject
+class WXDLLIMPEXP_HTML 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);
- // Returns character for given entity
+ // Returns character for given entity or 0 if the enity is unknown
wxChar GetEntityChar(const wxString& entity);
// Returns character that represents given Unicode code
+#if wxUSE_UNICODE
+ wxChar GetCharForCode(unsigned code) { return (wxChar)code; }
+#else
wxChar GetCharForCode(unsigned code);
+#endif
protected:
#if wxUSE_WCHAR_T && !wxUSE_UNICODE
wxMBConv *m_conv;
wxFontEncoding m_encoding;
#endif
+
+ DECLARE_NO_COPY_CLASS(wxHtmlEntitiesParser)
};
projects / wxWidgets.git / blobdiff