1 /////////////////////////////////////////////////////////////////////////////
3 // Purpose: wxHtmlParser class (generic parser)
4 // Author: Vaclav Slavik
6 // Copyright: (c) 1999 Vaclav Slavik
7 // Licence: wxWindows Licence
8 /////////////////////////////////////////////////////////////////////////////
11 #ifndef _WX_HTMLPARS_H_
12 #define _WX_HTMLPARS_H_
21 #include "wx/html/htmltag.h"
22 #include "wx/filesys.h"
25 class wxHtmlTagHandler
;
27 //--------------------------------------------------------------------------------
29 // This class handles generic parsing of HTML document : it scans
30 // the document and divide it into blocks of tags (where one block
31 // consists of starting and ending tag and of text between these
33 //--------------------------------------------------------------------------------
35 class WXDLLEXPORT wxHtmlParser
: public wxObject
37 DECLARE_ABSTRACT_CLASS(wxHtmlParser
)
40 wxHtmlParser() : wxObject(), m_HandlersHash(wxKEY_STRING
) {m_FS
= NULL
; m_Cache
= NULL
; m_HandlersStack
= NULL
;}
41 virtual ~wxHtmlParser();
43 void SetFS(wxFileSystem
*fs
) {m_FS
= fs
;}
44 // Sets the class which will be used for opening files
45 wxFileSystem
* GetFS() const {return m_FS
;}
47 wxObject
* Parse(const wxString
& source
);
48 // You can simply call this method when you need parsed output.
49 // This method does these things:
50 // 1. call InitParser(source);
51 // 2. call DoParsing();
52 // 3. call GetProduct(); (it's return value is then returned)
53 // 4. call DoneParser();
55 virtual void InitParser(const wxString
& source
);
56 // Sets the source. This must be called before running Parse() method.
57 virtual void DoneParser();
58 // This must be called after Parse().
60 void DoParsing(int begin_pos
, int end_pos
);
61 inline void DoParsing() {DoParsing(0, m_Source
.Length());};
62 // Parses the m_Source from begin_pos to end_pos-1.
63 // (in noparams version it parses whole m_Source)
65 virtual wxObject
* GetProduct() = 0;
66 // Returns product of parsing
67 // Returned value is result of parsing of the part. The type of this result
68 // depends on internal representation in derived parser
69 // (see wxHtmlWinParser for details).
71 virtual void AddTagHandler(wxHtmlTagHandler
*handler
);
72 // adds handler to the list & hash table of handlers.
74 void PushTagHandler(wxHtmlTagHandler
*handler
, wxString tags
);
75 // Forces the handler to handle additional tags (not returned by GetSupportedTags).
76 // The handler should already be in use by this parser.
77 // Example: you want to parse following pseudo-html structure:
79 // <it name="one" value="1">
80 // <it name="two" value="2">
82 // <it> This last it has different meaning, we don't want it to be parsed by myitems handler!
83 // handler can handle only 'myitems' (e.g. it's GetSupportedTags returns "MYITEMS")
84 // you can call PushTagHandler(handler, "IT") when you find <myitems>
85 // and call PopTagHandler() when you find </myitems>
88 // Restores state before last call to PushTagHandler
90 wxString
* GetSource() {return &m_Source
;}
92 virtual wxList
* GetTempData() {return NULL
;}
93 // this method returns list of wxObjects that represents
94 // all data allocated by the parser. These can't be freeded
95 // by destructor because they must be valid as long as
96 // GetProduct's return value is valid - the caller must
97 // explicitly call delete MyParser -> GetTempData() to free
99 // (this method always sets the list to delete its contents)
103 virtual void AddText(const char* txt
) = 0;
104 // Adds text to the output.
105 // This is called from Parse() and must be overriden in derived classes.
106 // txt is not guaranteed to be only one word. It is largest continuous part of text
107 // (= not broken by tags)
108 // NOTE : using char* because of speed improvements
110 virtual void AddTag(const wxHtmlTag
& tag
);
111 // Adds tag and proceeds it. Parse() may (and usually is) called from this method.
112 // This is called from Parse() and may be overriden.
113 // Default behavior is that it looks for proper handler in m_Handlers. The tag is
114 // ignored if no hander is found.
115 // Derived class is *responsible* for filling in m_Handlers table.
120 // source being parsed
121 wxHtmlTagsCache
*m_Cache
;
122 // tags cache, used during parsing.
123 wxHashTable m_HandlersHash
;
124 wxList m_HandlersList
;
125 // handlers that handle particular tags. The table is accessed by
127 // This attribute MUST be filled by derived class otherwise it would
128 // be empty and no tags would be recognized
129 // (see wxHtmlWinParser for details about filling it)
130 // m_HandlersHash is for random access based on knowledge of tag name (BR, P, etc.)
131 // it may (and often does) contain more references to one object
132 // m_HandlersList is list of all handlers and it is guaranteed to contain
133 // only one reference to each handler instance.
135 // class for opening files (file system)
136 wxList
*m_HandlersStack
;
137 // handlers stack used by PushTagHandler and PopTagHandler
146 //--------------------------------------------------------------------------------
148 // This class (and derived classes) cooperates with wxHtmlParser.
149 // Each recognized tag is passed to handler which is capable
150 // of handling it. Each tag is handled in 3 steps:
151 // 1. Handler will modifies state of parser
152 // (using it's public methods)
153 // 2. Parser parses source between starting and ending tag
154 // 3. Handler restores original state of the parser
155 //--------------------------------------------------------------------------------
157 class WXDLLEXPORT wxHtmlTagHandler
: public wxObject
159 DECLARE_ABSTRACT_CLASS(wxHtmlTagHandler
)
162 wxHtmlParser
*m_Parser
;
165 wxHtmlTagHandler() : wxObject () {m_Parser
= NULL
;};
167 virtual void SetParser(wxHtmlParser
*parser
) {m_Parser
= parser
;}
169 // NOTE : each _instance_ of handler is guaranteed to be called
170 // only by one parser. This means you don't have to care about
173 virtual wxString
GetSupportedTags() = 0;
174 // Returns list of supported tags. The list is in uppercase and
175 // tags are delimited by ','.
176 // Example : "I,B,FONT,P"
177 // is capable of handling italic, bold, font and paragraph tags
179 virtual bool HandleTag(const wxHtmlTag
& tag
) = 0;
180 // This is hadling core method. It does all the Steps 1-3.
181 // To process step 2, you can call ParseInner()
182 // returned value : TRUE if it called ParseInner(),
186 void ParseInner(const wxHtmlTag
& tag
) {m_Parser
-> DoParsing(tag
.GetBeginPos(), tag
.GetEndPos1());}
187 // parses input between beginning and ending tag.
188 // m_Parser must be set.
196 #endif // _WX_HTMLPARS_H_