X-Git-Url: https://git.saurik.com/wxWidgets.git/blobdiff_plain/9749497156909657d17ef972857caaca4bd09da2..ab9d0a8ca3d306898d92960ecfaefb8c1191b18a:/include/wx/html/htmlpars.h

diff --git a/include/wx/html/htmlpars.h b/include/wx/html/htmlpars.h
index 6a93d07ddb..d0392418b2 100644
--- a/include/wx/html/htmlpars.h
+++ b/include/wx/html/htmlpars.h
@@ -4,14 +4,14 @@
 // 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__
+#if defined(__GNUG__) && !defined(NO_GCC_PRAGMA)
 #pragma interface "htmlpars.h"
 #endif
 
@@ -20,25 +20,35 @@
 
 #include "wx/html/htmltag.h"
 #include "wx/filesys.h"
+#include "wx/hash.h"
+#include "wx/fontenc.h"
 
-class wxHtmlParser;
-class wxHtmlTagHandler;
+class WXDLLIMPEXP_BASE wxMBConv;
+class WXDLLIMPEXP_HTML wxHtmlParser;
+class WXDLLIMPEXP_HTML wxHtmlTagHandler;
+class WXDLLIMPEXP_HTML wxHtmlEntitiesParser;
 
-//--------------------------------------------------------------------------------
-// wxHtmlParser
-//                  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 wxHtmlTextPieces;
+class wxHtmlParserState;
 
-class WXDLLEXPORT wxHtmlParser : public wxObject
+
+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 WXDLLIMPEXP_HTML wxHtmlParser : public wxObject
 {
     DECLARE_ABSTRACT_CLASS(wxHtmlParser)
 
 public:
-    wxHtmlParser() : wxObject(), m_HandlersHash(wxKEY_STRING) 
-        { m_FS = NULL; m_Cache = NULL; m_HandlersStack = NULL; }
+    wxHtmlParser();
     virtual ~wxHtmlParser();
 
     // Sets the class which will be used for opening files
@@ -46,6 +56,10 @@ public:
 
     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);
@@ -59,10 +73,16 @@ public:
     // 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);
-    inline void DoParsing() {DoParsing(0, m_Source.Length());};
+    void DoParsing();
+
+    // Returns pointer to the tag at parser's current position
+    wxHtmlTag *GetCurrentTag() const { return m_CurTag; }
 
     // Returns product of parsing
     // Returned value is result of parsing of the part. The type of this result
@@ -73,7 +93,7 @@ public:
     // 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>
@@ -92,13 +112,37 @@ public:
     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
+    // HTML source (for example, <pre> handler replaces spaces with &nbsp;
+    // and newlines with <br>)
+    virtual void SetSourceAndSaveState(const wxString& src);
+    // Restores parser's state from stack or returns false if the stack is
+    // empty
+    virtual bool RestoreState();
+
+    // 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,
+                          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 char* txt) = 0;
+    virtual void AddText(const wxChar* 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.
@@ -108,12 +152,15 @@ protected:
     virtual void AddTag(const wxHtmlTag& tag);
 
 protected:
-    // source being parsed
+    // DOM tree:
+    wxHtmlTag *m_CurTag;
+    wxHtmlTag *m_Tags;
+    wxHtmlTextPieces *m_TextPieces;
+    size_t m_CurTextPiece;
+
     wxString m_Source;
 
-    // tags cache, used during parsing.
-    wxHtmlTagsCache *m_Cache;
-    wxHashTable m_HandlersHash;
+    wxHtmlParserState *m_SavedStates;
 
     // handlers that handle particular tags. The table is accessed by
     // key = tag's name.
@@ -125,30 +172,32 @@ protected:
     // 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;
+
+    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;
+};
 
 
-//--------------------------------------------------------------------------------
-// wxHtmlTagHandler
-//                  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
-//                    (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
+// 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
+//    (using it's public methods)
+// 2. Parser parses source between starting and ending tag
+// 3. Handler restores original state of the parser
+class WXDLLIMPEXP_HTML wxHtmlTagHandler : public wxObject
 {
     DECLARE_ABSTRACT_CLASS(wxHtmlTagHandler)
 
@@ -159,7 +208,7 @@ public:
     // 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
@@ -170,20 +219,58 @@ public:
 
     // 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()); }
 
     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 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 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)
+};
 
 
 #endif