/////////////////////////////////////////////////////////////////////////////
-// Name: htmlpars.h
+// Name: wx/html/htmlpars.h
// Purpose: wxHtmlParser class (generic parser)
// Author: Vaclav Slavik
// RCS-ID: $Id$
#include "wx/html/htmltag.h"
#include "wx/filesys.h"
-#include "wx/hash.h"
+#include "wx/hashmap.h"
+#include "wx/hashset.h"
+#include "wx/vector.h"
#include "wx/fontenc.h"
-class WXDLLIMPEXP_BASE wxMBConv;
-class WXDLLIMPEXP_HTML wxHtmlParser;
-class WXDLLIMPEXP_HTML wxHtmlTagHandler;
-class WXDLLIMPEXP_HTML wxHtmlEntitiesParser;
+class WXDLLIMPEXP_FWD_BASE wxMBConv;
+class WXDLLIMPEXP_FWD_HTML wxHtmlParser;
+class WXDLLIMPEXP_FWD_HTML wxHtmlTagHandler;
+class WXDLLIMPEXP_FWD_HTML wxHtmlEntitiesParser;
class wxHtmlTextPieces;
class wxHtmlParserState;
+WX_DECLARE_HASH_SET_WITH_DECL_PTR(wxHtmlTagHandler*,
+ wxPointerHash, wxPointerEqual,
+ wxHtmlTagHandlersSet,
+ class WXDLLIMPEXP_HTML);
+WX_DECLARE_STRING_HASH_MAP_WITH_DECL(wxHtmlTagHandler*,
+ wxHtmlTagHandlersHash,
+ class WXDLLIMPEXP_HTML);
+
enum wxHtmlURLType
{
// 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);
+ 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
// 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
void CreateDOMTree();
void DestroyDOMTree();
void CreateDOMSubTree(wxHtmlTag *cur,
- int begin_pos, int end_pos,
+ 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 wxChar* 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);
wxHtmlTextPieces *m_TextPieces;
size_t m_CurTextPiece;
- wxString m_Source;
+ const wxString *m_Source;
wxHtmlParserState *m_SavedStates;
// 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;
+ wxHtmlTagHandlersSet m_HandlersSet;
+ wxHtmlTagHandlersHash m_HandlersHash;
- DECLARE_NO_COPY_CLASS(wxHtmlParser)
+ 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<wxHtmlTagHandlersHash*> m_HandlersStack;
// entity parse
wxHtmlEntitiesParser *m_entitiesParser;
// parses input between beginning and ending tag.
// m_Parser must be set.
void ParseInner(const wxHtmlTag& tag)
- { m_Parser->DoParsing(tag.GetBeginPos(), tag.GetEndPos1()); }
+ { 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
wxHtmlParser *m_Parser;
- DECLARE_NO_COPY_CLASS(wxHtmlTagHandler)
+ wxDECLARE_NO_COPY_CLASS(wxHtmlTagHandler);
};
virtual ~wxHtmlEntitiesParser();
// Sets encoding of output string.
- // Has no effect if wxUSE_WCHAR_T==0 or wxUSE_UNICODE==1
+ // 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);
+ wxString Parse(const wxString& input) const;
// Returns character for given entity or 0 if the enity is unknown
- wxChar GetEntityChar(const wxString& entity);
+ wxChar GetEntityChar(const wxString& entity) const;
// Returns character that represents given Unicode code
#if wxUSE_UNICODE
- wxChar GetCharForCode(unsigned code) { return (wxChar)code; }
+ wxChar GetCharForCode(unsigned code) const { return (wxChar)code; }
#else
- wxChar GetCharForCode(unsigned code);
+ wxChar GetCharForCode(unsigned code) const;
#endif
protected:
-#if wxUSE_WCHAR_T && !wxUSE_UNICODE
+#if !wxUSE_UNICODE
wxMBConv *m_conv;
wxFontEncoding m_encoding;
#endif
- DECLARE_NO_COPY_CLASS(wxHtmlEntitiesParser)
+ wxDECLARE_NO_COPY_CLASS(wxHtmlEntitiesParser);
};