1999-07-07 18:04:58 -04:00
|
|
|
/////////////////////////////////////////////////////////////////////////////
|
2011-03-19 20:14:35 -04:00
|
|
|
// Name: wx/html/htmlpars.h
|
1999-07-07 18:04:58 -04:00
|
|
|
// Purpose: wxHtmlParser class (generic parser)
|
|
|
|
// Author: Vaclav Slavik
|
1999-10-02 14:19:46 -04:00
|
|
|
// RCS-ID: $Id$
|
1999-07-07 18:04:58 -04:00
|
|
|
// Copyright: (c) 1999 Vaclav Slavik
|
2004-05-23 16:53:33 -04:00
|
|
|
// Licence: wxWindows licence
|
1999-07-07 18:04:58 -04:00
|
|
|
/////////////////////////////////////////////////////////////////////////////
|
|
|
|
|
1999-10-02 14:19:46 -04:00
|
|
|
#ifndef _WX_HTMLPARS_H_
|
|
|
|
#define _WX_HTMLPARS_H_
|
1999-07-07 18:04:58 -04:00
|
|
|
|
|
|
|
#include "wx/defs.h"
|
|
|
|
#if wxUSE_HTML
|
|
|
|
|
1999-10-02 14:19:46 -04:00
|
|
|
#include "wx/html/htmltag.h"
|
|
|
|
#include "wx/filesys.h"
|
2007-10-05 18:56:55 -04:00
|
|
|
#include "wx/hashmap.h"
|
|
|
|
#include "wx/hashset.h"
|
|
|
|
#include "wx/vector.h"
|
2001-08-06 18:38:53 -04:00
|
|
|
#include "wx/fontenc.h"
|
1999-07-07 18:04:58 -04:00
|
|
|
|
2007-07-09 06:09:52 -04:00
|
|
|
class WXDLLIMPEXP_FWD_BASE wxMBConv;
|
|
|
|
class WXDLLIMPEXP_FWD_HTML wxHtmlParser;
|
|
|
|
class WXDLLIMPEXP_FWD_HTML wxHtmlTagHandler;
|
|
|
|
class WXDLLIMPEXP_FWD_HTML wxHtmlEntitiesParser;
|
2001-07-01 11:09:35 -04:00
|
|
|
|
2001-07-19 18:30:22 -04:00
|
|
|
class wxHtmlTextPieces;
|
|
|
|
class wxHtmlParserState;
|
|
|
|
|
2007-10-05 18:56:55 -04:00
|
|
|
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);
|
|
|
|
|
2002-01-27 14:03:10 -05:00
|
|
|
|
|
|
|
enum wxHtmlURLType
|
|
|
|
{
|
|
|
|
wxHTML_URL_PAGE,
|
|
|
|
wxHTML_URL_IMAGE,
|
|
|
|
wxHTML_URL_OTHER
|
|
|
|
};
|
|
|
|
|
2001-07-01 11:09:35 -04:00
|
|
|
// This class handles generic parsing of HTML document : it scans
|
2006-07-02 17:40:38 -04:00
|
|
|
// the document and divides it into blocks of tags (where one block
|
2001-07-01 11:09:35 -04:00
|
|
|
// consists of starting and ending tag and of text between these
|
|
|
|
// 2 tags.
|
2003-07-04 09:04:05 -04:00
|
|
|
class WXDLLIMPEXP_HTML wxHtmlParser : public wxObject
|
1999-07-07 18:04:58 -04:00
|
|
|
{
|
|
|
|
DECLARE_ABSTRACT_CLASS(wxHtmlParser)
|
|
|
|
|
2001-04-24 18:05:19 -04:00
|
|
|
public:
|
2001-07-01 11:09:35 -04:00
|
|
|
wxHtmlParser();
|
2001-04-24 18:05:19 -04:00
|
|
|
virtual ~wxHtmlParser();
|
|
|
|
|
|
|
|
// Sets the class which will be used for opening files
|
|
|
|
void SetFS(wxFileSystem *fs) { m_FS = fs; }
|
|
|
|
|
|
|
|
wxFileSystem* GetFS() const { return m_FS; }
|
|
|
|
|
2002-01-27 14:03:10 -05:00
|
|
|
// Opens file if the parser is allowed to open given URL (may be forbidden
|
2004-09-27 15:06:39 -04:00
|
|
|
// for security reasons)
|
2002-01-27 14:03:10 -05:00
|
|
|
virtual wxFSFile *OpenURL(wxHtmlURLType type, const wxString& url) const;
|
2002-01-19 18:23:15 -05:00
|
|
|
|
2001-04-24 18:05:19 -04:00
|
|
|
// You can simply call this method when you need parsed output.
|
|
|
|
// This method does these things:
|
|
|
|
// 1. call InitParser(source);
|
|
|
|
// 2. call DoParsing();
|
2006-07-02 17:40:38 -04:00
|
|
|
// 3. call GetProduct(); (its return value is then returned)
|
2001-04-24 18:05:19 -04:00
|
|
|
// 4. call DoneParser();
|
|
|
|
wxObject* Parse(const wxString& source);
|
|
|
|
|
|
|
|
// Sets the source. This must be called before running Parse() method.
|
|
|
|
virtual void InitParser(const wxString& source);
|
|
|
|
// This must be called after Parse().
|
|
|
|
virtual void DoneParser();
|
2004-09-27 15:06:39 -04:00
|
|
|
|
2002-12-04 09:11:26 -05:00
|
|
|
// May be called during parsing to immediately return from Parse().
|
2004-09-27 15:06:39 -04:00
|
|
|
virtual void StopParsing() { m_stopParsing = true; }
|
2001-04-24 18:05:19 -04:00
|
|
|
|
|
|
|
// Parses the m_Source from begin_pos to end_pos-1.
|
|
|
|
// (in noparams version it parses whole m_Source)
|
2007-08-26 06:04:11 -04:00
|
|
|
void DoParsing(const wxString::const_iterator& begin_pos,
|
|
|
|
const wxString::const_iterator& end_pos);
|
2001-07-19 18:30:22 -04:00
|
|
|
void DoParsing();
|
|
|
|
|
|
|
|
// Returns pointer to the tag at parser's current position
|
|
|
|
wxHtmlTag *GetCurrentTag() const { return m_CurTag; }
|
2001-04-24 18:05:19 -04:00
|
|
|
|
|
|
|
// Returns product of parsing
|
|
|
|
// Returned value is result of parsing of the part. The type of this result
|
|
|
|
// depends on internal representation in derived parser
|
|
|
|
// (see wxHtmlWinParser for details).
|
|
|
|
virtual wxObject* GetProduct() = 0;
|
|
|
|
|
|
|
|
// adds handler to the list & hash table of handlers.
|
|
|
|
virtual void AddTagHandler(wxHtmlTagHandler *handler);
|
|
|
|
|
2004-09-27 15:06:39 -04:00
|
|
|
// Forces the handler to handle additional tags (not returned by GetSupportedTags).
|
2001-04-24 18:05:19 -04:00
|
|
|
// The handler should already be in use by this parser.
|
|
|
|
// Example: you want to parse following pseudo-html structure:
|
|
|
|
// <myitems>
|
|
|
|
// <it name="one" value="1">
|
|
|
|
// <it name="two" value="2">
|
|
|
|
// </myitems>
|
|
|
|
// <it> This last it has different meaning, we don't want it to be parsed by myitems handler!
|
2006-07-02 17:40:38 -04:00
|
|
|
// handler can handle only 'myitems' (e.g. its GetSupportedTags returns "MYITEMS")
|
2001-04-24 18:05:19 -04:00
|
|
|
// you can call PushTagHandler(handler, "IT") when you find <myitems>
|
|
|
|
// and call PopTagHandler() when you find </myitems>
|
2005-09-24 17:43:15 -04:00
|
|
|
void PushTagHandler(wxHtmlTagHandler *handler, const wxString& tags);
|
2001-04-24 18:05:19 -04:00
|
|
|
|
|
|
|
// Restores state before last call to PushTagHandler
|
|
|
|
void PopTagHandler();
|
|
|
|
|
2007-08-26 06:04:11 -04:00
|
|
|
const wxString* GetSource() {return m_Source;}
|
2001-04-24 18:05:19 -04:00
|
|
|
void SetSource(const wxString& src);
|
2004-09-27 15:06:39 -04:00
|
|
|
|
2006-07-02 17:40:38 -04:00
|
|
|
// Sets HTML source and remembers current parser's state so that it can
|
2004-09-27 15:06:39 -04:00
|
|
|
// later be restored. This is useful for on-line modifications of
|
2001-07-19 18:30:22 -04:00
|
|
|
// HTML source (for example, <pre> handler replaces spaces with
|
|
|
|
// and newlines with <br>)
|
|
|
|
virtual void SetSourceAndSaveState(const wxString& src);
|
2004-09-27 15:06:39 -04:00
|
|
|
// Restores parser's state from stack or returns false if the stack is
|
2001-07-19 18:30:22 -04:00
|
|
|
// empty
|
|
|
|
virtual bool RestoreState();
|
2004-09-27 15:06:39 -04:00
|
|
|
|
2006-04-03 15:34:21 -04:00
|
|
|
// Returns HTML source inside the element (i.e. between the starting
|
|
|
|
// and ending tag)
|
|
|
|
wxString GetInnerSource(const wxHtmlTag& tag);
|
|
|
|
|
2002-12-04 09:11:26 -05:00
|
|
|
// 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);
|
2004-09-27 15:06:39 -04:00
|
|
|
|
2004-06-30 06:22:23 -04:00
|
|
|
// Returns entity parser object, used to substitute HTML &entities;
|
|
|
|
wxHtmlEntitiesParser *GetEntitiesParser() const { return m_entitiesParser; }
|
2001-04-24 18:05:19 -04:00
|
|
|
|
2007-04-08 17:02:03 -04:00
|
|
|
// 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);
|
|
|
|
|
2001-04-24 18:05:19 -04:00
|
|
|
protected:
|
2001-07-19 18:30:22 -04:00
|
|
|
// DOM structure
|
|
|
|
void CreateDOMTree();
|
|
|
|
void DestroyDOMTree();
|
|
|
|
void CreateDOMSubTree(wxHtmlTag *cur,
|
2007-08-26 06:04:11 -04:00
|
|
|
const wxString::const_iterator& begin_pos,
|
|
|
|
const wxString::const_iterator& end_pos,
|
2001-07-19 18:30:22 -04:00
|
|
|
wxHtmlTagsCache *cache);
|
|
|
|
|
2001-04-24 18:05:19 -04:00
|
|
|
// Adds text to the output.
|
2011-03-22 10:17:38 -04:00
|
|
|
// This is called from Parse() and must be overridden in derived classes.
|
2007-06-11 08:11:15 -04:00
|
|
|
// 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;
|
2001-04-24 18:05:19 -04:00
|
|
|
|
|
|
|
// Adds tag and proceeds it. Parse() may (and usually is) called from this method.
|
2011-03-22 10:17:38 -04:00
|
|
|
// This is called from Parse() and may be overridden.
|
|
|
|
// Default behaviour is that it looks for proper handler in m_Handlers. The tag is
|
2001-04-24 18:05:19 -04:00
|
|
|
// ignored if no hander is found.
|
|
|
|
// Derived class is *responsible* for filling in m_Handlers table.
|
|
|
|
virtual void AddTag(const wxHtmlTag& tag);
|
|
|
|
|
|
|
|
protected:
|
2001-07-19 18:30:22 -04:00
|
|
|
// DOM tree:
|
|
|
|
wxHtmlTag *m_CurTag;
|
|
|
|
wxHtmlTag *m_Tags;
|
|
|
|
wxHtmlTextPieces *m_TextPieces;
|
|
|
|
size_t m_CurTextPiece;
|
2001-04-24 18:05:19 -04:00
|
|
|
|
2007-08-26 06:04:11 -04:00
|
|
|
const wxString *m_Source;
|
2004-09-27 15:06:39 -04:00
|
|
|
|
2001-07-19 18:30:22 -04:00
|
|
|
wxHtmlParserState *m_SavedStates;
|
2004-09-27 15:06:39 -04:00
|
|
|
|
2001-04-24 18:05:19 -04:00
|
|
|
// 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
|
|
|
|
// be empty and no tags would be recognized
|
|
|
|
// (see wxHtmlWinParser for details about filling it)
|
|
|
|
// m_HandlersHash is for random access based on knowledge of tag name (BR, P, etc.)
|
|
|
|
// 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.
|
2007-10-05 18:56:55 -04:00
|
|
|
wxHtmlTagHandlersSet m_HandlersSet;
|
|
|
|
wxHtmlTagHandlersHash m_HandlersHash;
|
2001-04-24 18:05:19 -04:00
|
|
|
|
2009-02-08 06:45:59 -05:00
|
|
|
wxDECLARE_NO_COPY_CLASS(wxHtmlParser);
|
2003-01-02 18:38:11 -05:00
|
|
|
|
2001-04-24 18:05:19 -04:00
|
|
|
// class for opening files (file system)
|
|
|
|
wxFileSystem *m_FS;
|
|
|
|
// handlers stack used by PushTagHandler and PopTagHandler
|
2007-10-05 18:56:55 -04:00
|
|
|
wxVector<wxHtmlTagHandlersHash*> m_HandlersStack;
|
2004-09-27 15:06:39 -04:00
|
|
|
|
2001-07-01 11:09:35 -04:00
|
|
|
// entity parse
|
|
|
|
wxHtmlEntitiesParser *m_entitiesParser;
|
2004-09-27 15:06:39 -04:00
|
|
|
|
2002-12-04 09:11:26 -05:00
|
|
|
// flag indicating that the parser should stop
|
|
|
|
bool m_stopParsing;
|
1999-07-07 18:04:58 -04:00
|
|
|
};
|
|
|
|
|
|
|
|
|
|
|
|
|
2001-07-01 11:09:35 -04:00
|
|
|
// 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
|
2006-07-02 17:40:38 -04:00
|
|
|
// (using its public methods)
|
2001-07-01 11:09:35 -04:00
|
|
|
// 2. Parser parses source between starting and ending tag
|
|
|
|
// 3. Handler restores original state of the parser
|
2003-07-04 09:04:05 -04:00
|
|
|
class WXDLLIMPEXP_HTML wxHtmlTagHandler : public wxObject
|
1999-07-07 18:04:58 -04:00
|
|
|
{
|
|
|
|
DECLARE_ABSTRACT_CLASS(wxHtmlTagHandler)
|
|
|
|
|
2001-04-24 18:05:19 -04:00
|
|
|
public:
|
|
|
|
wxHtmlTagHandler() : wxObject () { m_Parser = NULL; }
|
|
|
|
|
|
|
|
// Sets the parser.
|
|
|
|
// NOTE : each _instance_ of handler is guaranteed to be called
|
|
|
|
// only by one parser. This means you don't have to care about
|
|
|
|
// reentrancy.
|
2004-09-27 15:06:39 -04:00
|
|
|
virtual void SetParser(wxHtmlParser *parser)
|
2001-04-24 18:05:19 -04:00
|
|
|
{ m_Parser = parser; }
|
|
|
|
|
|
|
|
// Returns list of supported tags. The list is in uppercase and
|
|
|
|
// tags are delimited by ','.
|
|
|
|
// Example : "I,B,FONT,P"
|
|
|
|
// is capable of handling italic, bold, font and paragraph tags
|
|
|
|
virtual wxString GetSupportedTags() = 0;
|
|
|
|
|
|
|
|
// This is hadling core method. It does all the Steps 1-3.
|
|
|
|
// To process step 2, you can call ParseInner()
|
2004-09-27 15:06:39 -04:00
|
|
|
// returned value : true if it called ParseInner(),
|
|
|
|
// false etherwise
|
2001-04-24 18:05:19 -04:00
|
|
|
virtual bool HandleTag(const wxHtmlTag& tag) = 0;
|
|
|
|
|
|
|
|
protected:
|
|
|
|
// parses input between beginning and ending tag.
|
|
|
|
// m_Parser must be set.
|
2004-09-27 15:06:39 -04:00
|
|
|
void ParseInner(const wxHtmlTag& tag)
|
2007-08-26 06:04:11 -04:00
|
|
|
{ m_Parser->DoParsing(tag.GetBeginIter(), tag.GetEndIter1()); }
|
2001-04-24 18:05:19 -04:00
|
|
|
|
2006-04-03 15:34:21 -04:00
|
|
|
// 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);
|
|
|
|
|
2001-04-24 18:05:19 -04:00
|
|
|
wxHtmlParser *m_Parser;
|
2003-01-02 18:38:11 -05:00
|
|
|
|
2009-02-08 06:45:59 -05:00
|
|
|
wxDECLARE_NO_COPY_CLASS(wxHtmlTagHandler);
|
1999-07-07 18:04:58 -04:00
|
|
|
};
|
|
|
|
|
|
|
|
|
2001-07-01 11:09:35 -04:00
|
|
|
// This class is used to parse HTML entities in strings. It can handle
|
|
|
|
// both named entities and &#xxxx entries where xxxx is Unicode code.
|
2003-07-04 09:04:05 -04:00
|
|
|
class WXDLLIMPEXP_HTML wxHtmlEntitiesParser : public wxObject
|
2001-07-01 11:09:35 -04:00
|
|
|
{
|
|
|
|
DECLARE_DYNAMIC_CLASS(wxHtmlEntitiesParser)
|
|
|
|
|
|
|
|
public:
|
|
|
|
wxHtmlEntitiesParser();
|
|
|
|
virtual ~wxHtmlEntitiesParser();
|
2004-09-27 15:06:39 -04:00
|
|
|
|
2001-07-01 11:09:35 -04:00
|
|
|
// Sets encoding of output string.
|
2010-04-16 06:43:18 -04:00
|
|
|
// Has no effect if wxUSE_UNICODE==1
|
|
|
|
#if wxUSE_UNICODE
|
|
|
|
void SetEncoding(wxFontEncoding WXUNUSED(encoding)) {}
|
|
|
|
#else
|
2001-07-01 11:09:35 -04:00
|
|
|
void SetEncoding(wxFontEncoding encoding);
|
2010-04-16 06:43:18 -04:00
|
|
|
#endif
|
2004-09-27 15:06:39 -04:00
|
|
|
|
2001-07-01 11:09:35 -04:00
|
|
|
// Parses entities in input and replaces them with respective characters
|
|
|
|
// (with respect to output encoding)
|
2007-08-20 16:08:40 -04:00
|
|
|
wxString Parse(const wxString& input) const;
|
2001-08-08 19:18:39 -04:00
|
|
|
|
2001-10-05 18:05:45 -04:00
|
|
|
// Returns character for given entity or 0 if the enity is unknown
|
2007-08-20 16:08:40 -04:00
|
|
|
wxChar GetEntityChar(const wxString& entity) const;
|
2001-08-08 19:18:39 -04:00
|
|
|
|
|
|
|
// Returns character that represents given Unicode code
|
2002-12-04 09:11:26 -05:00
|
|
|
#if wxUSE_UNICODE
|
2007-08-20 16:08:40 -04:00
|
|
|
wxChar GetCharForCode(unsigned code) const { return (wxChar)code; }
|
2002-12-04 09:11:26 -05:00
|
|
|
#else
|
2007-08-20 16:08:40 -04:00
|
|
|
wxChar GetCharForCode(unsigned code) const;
|
2002-12-04 09:11:26 -05:00
|
|
|
#endif
|
2001-07-01 11:09:35 -04:00
|
|
|
|
2001-08-08 19:18:39 -04:00
|
|
|
protected:
|
2010-04-16 06:43:18 -04:00
|
|
|
#if !wxUSE_UNICODE
|
2001-07-01 11:09:35 -04:00
|
|
|
wxMBConv *m_conv;
|
|
|
|
wxFontEncoding m_encoding;
|
|
|
|
#endif
|
2003-01-02 18:38:11 -05:00
|
|
|
|
2009-02-08 06:45:59 -05:00
|
|
|
wxDECLARE_NO_COPY_CLASS(wxHtmlEntitiesParser);
|
2001-07-01 11:09:35 -04:00
|
|
|
};
|
1999-07-07 18:04:58 -04:00
|
|
|
|
|
|
|
|
|
|
|
#endif
|
1999-10-02 14:19:46 -04:00
|
|
|
|
|
|
|
#endif // _WX_HTMLPARS_H_
|