1
/////////////////////////////////////////////////////////////////////////////
3
// Purpose: wxHtmlParser class (generic parser)
4
// Author: Vaclav Slavik
5
// RCS-ID: $Id: htmlpars.h,v 1.22 2004/09/27 19:06:39 ABX Exp $
6
// Copyright: (c) 1999 Vaclav Slavik
7
// Licence: wxWindows licence
8
/////////////////////////////////////////////////////////////////////////////
11
#ifndef _WX_HTMLPARS_H_
12
#define _WX_HTMLPARS_H_
14
#if defined(__GNUG__) && !defined(NO_GCC_PRAGMA)
15
#pragma interface "htmlpars.h"
21
#include "wx/html/htmltag.h"
22
#include "wx/filesys.h"
24
#include "wx/fontenc.h"
26
class WXDLLIMPEXP_BASE wxMBConv;
27
class WXDLLIMPEXP_HTML wxHtmlParser;
28
class WXDLLIMPEXP_HTML wxHtmlTagHandler;
29
class WXDLLIMPEXP_HTML wxHtmlEntitiesParser;
31
class wxHtmlTextPieces;
32
class wxHtmlParserState;
42
// This class handles generic parsing of HTML document : it scans
43
// the document and divide it into blocks of tags (where one block
44
// consists of starting and ending tag and of text between these
46
class WXDLLIMPEXP_HTML wxHtmlParser : public wxObject
48
DECLARE_ABSTRACT_CLASS(wxHtmlParser)
52
virtual ~wxHtmlParser();
54
// Sets the class which will be used for opening files
55
void SetFS(wxFileSystem *fs) { m_FS = fs; }
57
wxFileSystem* GetFS() const { return m_FS; }
59
// Opens file if the parser is allowed to open given URL (may be forbidden
60
// for security reasons)
61
virtual wxFSFile *OpenURL(wxHtmlURLType type, const wxString& url) const;
63
// You can simply call this method when you need parsed output.
64
// This method does these things:
65
// 1. call InitParser(source);
66
// 2. call DoParsing();
67
// 3. call GetProduct(); (it's return value is then returned)
68
// 4. call DoneParser();
69
wxObject* Parse(const wxString& source);
71
// Sets the source. This must be called before running Parse() method.
72
virtual void InitParser(const wxString& source);
73
// This must be called after Parse().
74
virtual void DoneParser();
76
// May be called during parsing to immediately return from Parse().
77
virtual void StopParsing() { m_stopParsing = true; }
79
// Parses the m_Source from begin_pos to end_pos-1.
80
// (in noparams version it parses whole m_Source)
81
void DoParsing(int begin_pos, int end_pos);
84
// Returns pointer to the tag at parser's current position
85
wxHtmlTag *GetCurrentTag() const { return m_CurTag; }
87
// Returns product of parsing
88
// Returned value is result of parsing of the part. The type of this result
89
// depends on internal representation in derived parser
90
// (see wxHtmlWinParser for details).
91
virtual wxObject* GetProduct() = 0;
93
// adds handler to the list & hash table of handlers.
94
virtual void AddTagHandler(wxHtmlTagHandler *handler);
96
// Forces the handler to handle additional tags (not returned by GetSupportedTags).
97
// The handler should already be in use by this parser.
98
// Example: you want to parse following pseudo-html structure:
100
// <it name="one" value="1">
101
// <it name="two" value="2">
103
// <it> This last it has different meaning, we don't want it to be parsed by myitems handler!
104
// handler can handle only 'myitems' (e.g. it's GetSupportedTags returns "MYITEMS")
105
// you can call PushTagHandler(handler, "IT") when you find <myitems>
106
// and call PopTagHandler() when you find </myitems>
107
void PushTagHandler(wxHtmlTagHandler *handler, wxString tags);
109
// Restores state before last call to PushTagHandler
110
void PopTagHandler();
112
wxString* GetSource() {return &m_Source;}
113
void SetSource(const wxString& src);
115
// Sets HTML source and remebers current parser's state so that it can
116
// later be restored. This is useful for on-line modifications of
117
// HTML source (for example, <pre> handler replaces spaces with
118
// and newlines with <br>)
119
virtual void SetSourceAndSaveState(const wxString& src);
120
// Restores parser's state from stack or returns false if the stack is
122
virtual bool RestoreState();
124
// Parses HTML string 'markup' and extracts charset info from <meta> tag
125
// if present. Returns empty string if the tag is missing.
126
// For wxHTML's internal use.
127
static wxString ExtractCharsetInformation(const wxString& markup);
129
// Returns entity parser object, used to substitute HTML &entities;
130
wxHtmlEntitiesParser *GetEntitiesParser() const { return m_entitiesParser; }
134
void CreateDOMTree();
135
void DestroyDOMTree();
136
void CreateDOMSubTree(wxHtmlTag *cur,
137
int begin_pos, int end_pos,
138
wxHtmlTagsCache *cache);
140
// Adds text to the output.
141
// This is called from Parse() and must be overriden in derived classes.
142
// txt is not guaranteed to be only one word. It is largest continuous part of text
143
// (= not broken by tags)
144
// NOTE : using char* because of speed improvements
145
virtual void AddText(const wxChar* txt) = 0;
147
// Adds tag and proceeds it. Parse() may (and usually is) called from this method.
148
// This is called from Parse() and may be overriden.
149
// Default behavior is that it looks for proper handler in m_Handlers. The tag is
150
// ignored if no hander is found.
151
// Derived class is *responsible* for filling in m_Handlers table.
152
virtual void AddTag(const wxHtmlTag& tag);
158
wxHtmlTextPieces *m_TextPieces;
159
size_t m_CurTextPiece;
163
wxHtmlParserState *m_SavedStates;
165
// handlers that handle particular tags. The table is accessed by
167
// This attribute MUST be filled by derived class otherwise it would
168
// be empty and no tags would be recognized
169
// (see wxHtmlWinParser for details about filling it)
170
// m_HandlersHash is for random access based on knowledge of tag name (BR, P, etc.)
171
// it may (and often does) contain more references to one object
172
// m_HandlersList is list of all handlers and it is guaranteed to contain
173
// only one reference to each handler instance.
174
wxList m_HandlersList;
175
wxHashTable m_HandlersHash;
177
DECLARE_NO_COPY_CLASS(wxHtmlParser)
179
// class for opening files (file system)
181
// handlers stack used by PushTagHandler and PopTagHandler
182
wxList *m_HandlersStack;
185
wxHtmlEntitiesParser *m_entitiesParser;
187
// flag indicating that the parser should stop
193
// This class (and derived classes) cooperates with wxHtmlParser.
194
// Each recognized tag is passed to handler which is capable
195
// of handling it. Each tag is handled in 3 steps:
196
// 1. Handler will modifies state of parser
197
// (using it's public methods)
198
// 2. Parser parses source between starting and ending tag
199
// 3. Handler restores original state of the parser
200
class WXDLLIMPEXP_HTML wxHtmlTagHandler : public wxObject
202
DECLARE_ABSTRACT_CLASS(wxHtmlTagHandler)
205
wxHtmlTagHandler() : wxObject () { m_Parser = NULL; }
208
// NOTE : each _instance_ of handler is guaranteed to be called
209
// only by one parser. This means you don't have to care about
211
virtual void SetParser(wxHtmlParser *parser)
212
{ m_Parser = parser; }
214
// Returns list of supported tags. The list is in uppercase and
215
// tags are delimited by ','.
216
// Example : "I,B,FONT,P"
217
// is capable of handling italic, bold, font and paragraph tags
218
virtual wxString GetSupportedTags() = 0;
220
// This is hadling core method. It does all the Steps 1-3.
221
// To process step 2, you can call ParseInner()
222
// returned value : true if it called ParseInner(),
224
virtual bool HandleTag(const wxHtmlTag& tag) = 0;
227
// parses input between beginning and ending tag.
228
// m_Parser must be set.
229
void ParseInner(const wxHtmlTag& tag)
230
{ m_Parser->DoParsing(tag.GetBeginPos(), tag.GetEndPos1()); }
232
wxHtmlParser *m_Parser;
234
DECLARE_NO_COPY_CLASS(wxHtmlTagHandler)
238
// This class is used to parse HTML entities in strings. It can handle
239
// both named entities and &#xxxx entries where xxxx is Unicode code.
240
class WXDLLIMPEXP_HTML wxHtmlEntitiesParser : public wxObject
242
DECLARE_DYNAMIC_CLASS(wxHtmlEntitiesParser)
245
wxHtmlEntitiesParser();
246
virtual ~wxHtmlEntitiesParser();
248
// Sets encoding of output string.
249
// Has no effect if wxUSE_WCHAR_T==0 or wxUSE_UNICODE==1
250
void SetEncoding(wxFontEncoding encoding);
252
// Parses entities in input and replaces them with respective characters
253
// (with respect to output encoding)
254
wxString Parse(const wxString& input);
256
// Returns character for given entity or 0 if the enity is unknown
257
wxChar GetEntityChar(const wxString& entity);
259
// Returns character that represents given Unicode code
261
wxChar GetCharForCode(unsigned code) { return (wxChar)code; }
263
wxChar GetCharForCode(unsigned code);
267
#if wxUSE_WCHAR_T && !wxUSE_UNICODE
269
wxFontEncoding m_encoding;
272
DECLARE_NO_COPY_CLASS(wxHtmlEntitiesParser)
278
#endif // _WX_HTMLPARS_H_