implemented GetLabel() (part of patch 1679337)

[wxWidgets.git] / src / html / htmlpars.cpp

diff --git a/src/html/htmlpars.cpp b/src/html/htmlpars.cpp
index 57a7406d032463b5e827531475c00ff58fc023db..7efb69af345f19d0b43b0fec336b5933a00a9a1a 100644 (file)
--- a/src/html/htmlpars.cpp
+++ b/src/html/htmlpars.cpp
@@ -1,29 +1,25 @@
 /////////////////////////////////////////////////////////////////////////////
 /////////////////////////////////////////////////////////////////////////////

-// Name:        htmlpars.cpp
+// Name:        src/html/htmlpars.cpp

 // Purpose:     wxHtmlParser class (generic parser)
 // Author:      Vaclav Slavik
 // RCS-ID:      $Id$
 // Copyright:   (c) 1999 Vaclav Slavik
 // Purpose:     wxHtmlParser class (generic parser)
 // Author:      Vaclav Slavik
 // RCS-ID:      $Id$
 // Copyright:   (c) 1999 Vaclav Slavik

-// Licence:     wxWindows Licence
+// Licence:     wxWindows licence

 /////////////////////////////////////////////////////////////////////////////
 
 /////////////////////////////////////////////////////////////////////////////
 

+#include "wx/wxprec.h"

 
 

-#ifdef __GNUG__
-#pragma implementation
+#ifdef __BORLANDC__
+    #pragma hdrstop

 #endif
 
 #endif
 

-#include "wx/wxprec.h"
-
-#include "wx/defs.h"

 #if wxUSE_HTML && wxUSE_STREAMS
 
 #if wxUSE_HTML && wxUSE_STREAMS
 

-#ifdef __BORDLANDC__
-#pragma hdrstop
-#endif
-

 #ifndef WXPRECOMP
 #ifndef WXPRECOMP

+    #include "wx/dynarray.h"

     #include "wx/log.h"
     #include "wx/intl.h"
     #include "wx/log.h"
     #include "wx/intl.h"

+    #include "wx/app.h"

 #endif
 
 #include "wx/tokenzr.h"
 #endif
 
 #include "wx/tokenzr.h"


@@ -32,9 +28,17 @@
 #include "wx/fontmap.h"
 #include "wx/html/htmldefs.h"
 #include "wx/html/htmlpars.h"
 #include "wx/fontmap.h"
 #include "wx/html/htmldefs.h"
 #include "wx/html/htmlpars.h"

-#include "wx/dynarray.h"

 #include "wx/arrimpl.cpp"
 
 #include "wx/arrimpl.cpp"
 

+#ifdef __WXWINCE__
+    #include "wx/msw/wince/missing.h"       // for bsearch()
+#endif
+
+// DLL options compatibility check:
+WX_CHECK_BUILD_OPTIONS("wxHTML")
+
+const wxChar *wxTRACE_HTML_DEBUG = _T("htmldebug");
+

 //-----------------------------------------------------------------------------
 // wxHtmlParser helpers
 //-----------------------------------------------------------------------------
 //-----------------------------------------------------------------------------
 // wxHtmlParser helpers
 //-----------------------------------------------------------------------------


@@ -47,7 +51,7 @@ public:
 };
 
 WX_DECLARE_OBJARRAY(wxHtmlTextPiece, wxHtmlTextPieces);
 };
 
 WX_DECLARE_OBJARRAY(wxHtmlTextPiece, wxHtmlTextPieces);

-WX_DEFINE_OBJARRAY(wxHtmlTextPieces);
+WX_DEFINE_OBJARRAY(wxHtmlTextPieces)

 
 class wxHtmlParserState
 {
 
 class wxHtmlParserState
 {


@@ -80,22 +84,28 @@ wxHtmlParser::wxHtmlParser()
 
 wxHtmlParser::~wxHtmlParser()
 {
 
 wxHtmlParser::~wxHtmlParser()
 {

-    while (RestoreState())
-        DestroyDOMTree();
+    while (RestoreState()) {}
+    DestroyDOMTree();
+
+    if (m_HandlersStack)
+    {
+        wxList& tmp = *m_HandlersStack;
+        wxList::iterator it, en;
+        for( it = tmp.begin(), en = tmp.end(); it != en; ++it )
+            delete (wxHashTable*)*it;
+        tmp.clear();
+    }

     delete m_HandlersStack;
     m_HandlersHash.Clear();
     delete m_HandlersStack;
     m_HandlersHash.Clear();

-    m_HandlersList.DeleteContents(TRUE);
-    m_HandlersList.Clear();
+    WX_CLEAR_LIST(wxList, m_HandlersList);

     delete m_entitiesParser;
 }
 
 wxObject* wxHtmlParser::Parse(const wxString& source)
 {
     delete m_entitiesParser;
 }
 
 wxObject* wxHtmlParser::Parse(const wxString& source)
 {

-    wxObject *result;
-

     InitParser(source);
     DoParsing();
     InitParser(source);
     DoParsing();

-    result = GetProduct();
+    wxObject *result = GetProduct();

     DoneParser();
     return result;
 }
     DoneParser();
     return result;
 }


@@ -103,6 +113,7 @@ wxObject* wxHtmlParser::Parse(const wxString& source)
 void wxHtmlParser::InitParser(const wxString& source)
 {
     SetSource(source);
 void wxHtmlParser::InitParser(const wxString& source)
 {
     SetSource(source);

+    m_stopParsing = false;

 }
 
 void wxHtmlParser::DoneParser()
 }
 
 void wxHtmlParser::DoneParser()


@@ -123,10 +134,12 @@ void wxHtmlParser::CreateDOMTree()
 {
     wxHtmlTagsCache cache(m_Source);
     m_TextPieces = new wxHtmlTextPieces;
 {
     wxHtmlTagsCache cache(m_Source);
     m_TextPieces = new wxHtmlTextPieces;

-    CreateDOMSubTree(NULL, 0, m_Source.Length(), &cache);
+    CreateDOMSubTree(NULL, 0, m_Source.length(), &cache);

     m_CurTextPiece = 0;
 }
 
     m_CurTextPiece = 0;
 }
 

+extern bool wxIsCDATAElement(const wxChar *tag);
+

 void wxHtmlParser::CreateDOMSubTree(wxHtmlTag *cur,
                                     int begin_pos, int end_pos,
                                     wxHtmlTagsCache *cache)
 void wxHtmlParser::CreateDOMSubTree(wxHtmlTag *cur,
                                     int begin_pos, int end_pos,
                                     wxHtmlTagsCache *cache)


@@ -137,6 +150,15 @@ void wxHtmlParser::CreateDOMSubTree(wxHtmlTag *cur,
     int i = begin_pos;
     int textBeginning = begin_pos;
 
     int i = begin_pos;
     int textBeginning = begin_pos;
 

+    // If the tag contains CDATA text, we include the text between beginning
+    // and ending tag verbosely. Setting i=end_pos will skip to the very
+    // end of this function where text piece is added, bypassing any child
+    // tags parsing (CDATA element can't have child elements by definition):
+    if (cur != NULL && wxIsCDATAElement(cur->GetName().c_str()))
+    {
+        i = end_pos;
+    }
+

     while (i < end_pos)
     {
         c = m_Source.GetChar(i);
     while (i < end_pos)
     {
         c = m_Source.GetChar(i);


@@ -176,7 +198,7 @@ void wxHtmlParser::CreateDOMSubTree(wxHtmlTag *cur,
 
             // add another tag to the tree:
             else if (i < end_pos-1 && m_Source.GetChar(i+1) != wxT('/'))
 
             // add another tag to the tree:
             else if (i < end_pos-1 && m_Source.GetChar(i+1) != wxT('/'))

-               {
+            {

                 wxHtmlTag *chd;
                 if (cur)
                     chd = new wxHtmlTag(cur, m_Source,
                 wxHtmlTag *chd;
                 if (cur)
                     chd = new wxHtmlTag(cur, m_Source,


@@ -209,6 +231,7 @@ void wxHtmlParser::CreateDOMSubTree(wxHtmlTag *cur,
                 }
                 else
                     i = chd->GetBeginPos();
                 }
                 else
                     i = chd->GetBeginPos();

+

                 textBeginning = i;
             }
 
                 textBeginning = i;
             }
 


@@ -248,7 +271,7 @@ void wxHtmlParser::DoParsing()
 {
     m_CurTag = m_Tags;
     m_CurTextPiece = 0;
 {
     m_CurTag = m_Tags;
     m_CurTextPiece = 0;

-    DoParsing(0, m_Source.Length());
+    DoParsing(0, m_Source.length());

 }
 
 void wxHtmlParser::DoParsing(int begin_pos, int end_pos)
 }
 
 void wxHtmlParser::DoParsing(int begin_pos, int end_pos)


@@ -280,17 +303,15 @@ void wxHtmlParser::DoParsing(int begin_pos, int end_pos)
         }
         else if (m_CurTag)
         {
         }
         else if (m_CurTag)
         {

-            // Add tag:
-            if (m_CurTag)
-            {
-                if (m_CurTag->HasEnding())
-                    begin_pos = m_CurTag->GetEndPos2();
-                else
-                    begin_pos = m_CurTag->GetBeginPos();
-            }
+            if (m_CurTag->HasEnding())
+                begin_pos = m_CurTag->GetEndPos2();
+            else
+                begin_pos = m_CurTag->GetBeginPos();

             wxHtmlTag *t = m_CurTag;
             m_CurTag = m_CurTag->GetNextTag();
             AddTag(*t);
             wxHtmlTag *t = m_CurTag;
             m_CurTag = m_CurTag->GetNextTag();
             AddTag(*t);

+            if (m_stopParsing)
+                return;

         }
         else break;
     }
         }
         else break;
     }


@@ -299,11 +320,15 @@ void wxHtmlParser::DoParsing(int begin_pos, int end_pos)
 void wxHtmlParser::AddTag(const wxHtmlTag& tag)
 {
     wxHtmlTagHandler *h;
 void wxHtmlParser::AddTag(const wxHtmlTag& tag)
 {
     wxHtmlTagHandler *h;

-    bool inner = FALSE;
+    bool inner = false;

 
     h = (wxHtmlTagHandler*) m_HandlersHash.Get(tag.GetName());
     if (h)
 
     h = (wxHtmlTagHandler*) m_HandlersHash.Get(tag.GetName());
     if (h)

+    {

         inner = h->HandleTag(tag);
         inner = h->HandleTag(tag);

+        if (m_stopParsing)
+            return;
+    }

     if (!inner)
     {
         if (tag.HasEnding())
     if (!inner)
     {
         if (tag.HasEnding())


@@ -317,7 +342,7 @@ void wxHtmlParser::AddTagHandler(wxHtmlTagHandler *handler)
     wxStringTokenizer tokenizer(s, wxT(", "));
 
     while (tokenizer.HasMoreTokens())
     wxStringTokenizer tokenizer(s, wxT(", "));
 
     while (tokenizer.HasMoreTokens())

-        m_HandlersHash.Put(tokenizer.NextToken(), handler);
+        m_HandlersHash.Put(tokenizer.GetNextToken(), handler);

 
     if (m_HandlersList.IndexOf(handler) == wxNOT_FOUND)
         m_HandlersList.Append(handler);
 
     if (m_HandlersList.IndexOf(handler) == wxNOT_FOUND)
         m_HandlersList.Append(handler);


@@ -325,7 +350,7 @@ void wxHtmlParser::AddTagHandler(wxHtmlTagHandler *handler)
     handler->SetParser(this);
 }
 
     handler->SetParser(this);
 }
 

-void wxHtmlParser::PushTagHandler(wxHtmlTagHandler *handler, wxString tags)
+void wxHtmlParser::PushTagHandler(wxHtmlTagHandler *handler, const wxString& tags)

 {
     wxStringTokenizer tokenizer(tags, wxT(", "));
     wxString key;
 {
     wxStringTokenizer tokenizer(tags, wxT(", "));
     wxString key;


@@ -333,14 +358,13 @@ void wxHtmlParser::PushTagHandler(wxHtmlTagHandler *handler, wxString tags)
     if (m_HandlersStack == NULL)
     {
         m_HandlersStack = new wxList;
     if (m_HandlersStack == NULL)
     {
         m_HandlersStack = new wxList;

-        m_HandlersStack->DeleteContents(TRUE);

     }
 
     }
 

-    m_HandlersStack->Insert(new wxHashTable(m_HandlersHash));
+    m_HandlersStack->Insert((wxObject*)new wxHashTable(m_HandlersHash));

 
     while (tokenizer.HasMoreTokens())
     {
 
     while (tokenizer.HasMoreTokens())
     {

-        key = tokenizer.NextToken();
+        key = tokenizer.GetNextToken();

         m_HandlersHash.Delete(key);
         m_HandlersHash.Put(key, handler);
     }
         m_HandlersHash.Delete(key);
         m_HandlersHash.Put(key, handler);
     }


@@ -348,16 +372,22 @@ void wxHtmlParser::PushTagHandler(wxHtmlTagHandler *handler, wxString tags)
 
 void wxHtmlParser::PopTagHandler()
 {
 
 void wxHtmlParser::PopTagHandler()
 {

-    wxNode *first;
-
-    if (m_HandlersStack == NULL ||
-        (first = m_HandlersStack->GetFirst()) == NULL)
+    wxList::compatibility_iterator first;
+
+    if ( !m_HandlersStack ||
+#if wxUSE_STL
+         !(first = m_HandlersStack->GetFirst())
+#else // !wxUSE_STL
+         ((first = m_HandlersStack->GetFirst()) == NULL)
+#endif // wxUSE_STL/!wxUSE_STL
+        )

     {
         wxLogWarning(_("Warning: attempt to remove HTML tag handler from empty stack."));
         return;
     }
     m_HandlersHash = *((wxHashTable*) first->GetData());
     {
         wxLogWarning(_("Warning: attempt to remove HTML tag handler from empty stack."));
         return;
     }
     m_HandlersHash = *((wxHashTable*) first->GetData());

-    m_HandlersStack->DeleteNode(first);
+    delete (wxHashTable*) first->GetData();
+    m_HandlersStack->Erase(first);

 }
 
 void wxHtmlParser::SetSourceAndSaveState(const wxString& src)
 }
 
 void wxHtmlParser::SetSourceAndSaveState(const wxString& src)


@@ -384,7 +414,9 @@ void wxHtmlParser::SetSourceAndSaveState(const wxString& src)
 
 bool wxHtmlParser::RestoreState()
 {
 
 bool wxHtmlParser::RestoreState()
 {

-    if (!m_SavedStates) return FALSE;
+    if (!m_SavedStates) return false;
+
+    DestroyDOMTree();

 
     wxHtmlParserState *s = m_SavedStates;
     m_SavedStates = s->m_nextState;
 
     wxHtmlParserState *s = m_SavedStates;
     m_SavedStates = s->m_nextState;


@@ -396,7 +428,13 @@ bool wxHtmlParser::RestoreState()
     m_Source = s->m_source;
 
     delete s;
     m_Source = s->m_source;
 
     delete s;

-    return TRUE;
+    return true;
+}
+
+wxString wxHtmlParser::GetInnerSource(const wxHtmlTag& tag)
+{
+    return GetSource()->Mid(tag.GetBeginPos(),
+                            tag.GetEndPos1() - tag.GetBeginPos());

 }
 
 //-----------------------------------------------------------------------------
 }
 
 //-----------------------------------------------------------------------------


@@ -405,6 +443,15 @@ bool wxHtmlParser::RestoreState()
 
 IMPLEMENT_ABSTRACT_CLASS(wxHtmlTagHandler,wxObject)
 
 
 IMPLEMENT_ABSTRACT_CLASS(wxHtmlTagHandler,wxObject)
 

+void wxHtmlTagHandler::ParseInnerSource(const wxString& source)
+{
+    // It is safe to temporarily change the source being parsed,
+    // provided we restore the state back after parsing
+    m_Parser->SetSourceAndSaveState(source);
+    m_Parser->DoParsing();
+    m_Parser->RestoreState();
+}
+

 
 //-----------------------------------------------------------------------------
 // wxHtmlEntitiesParser
 
 //-----------------------------------------------------------------------------
 // wxHtmlEntitiesParser


@@ -429,11 +476,15 @@ wxHtmlEntitiesParser::~wxHtmlEntitiesParser()
 void wxHtmlEntitiesParser::SetEncoding(wxFontEncoding encoding)
 {
 #if wxUSE_WCHAR_T && !wxUSE_UNICODE
 void wxHtmlEntitiesParser::SetEncoding(wxFontEncoding encoding)
 {
 #if wxUSE_WCHAR_T && !wxUSE_UNICODE

-    if (encoding == m_encoding) return;
+    if (encoding == m_encoding)
+        return;
+

     delete m_conv;
     delete m_conv;

-    m_conv = NULL;
+

     m_encoding = encoding;
     m_encoding = encoding;

-    if (m_encoding != wxFONTENCODING_SYSTEM)
+    if (m_encoding == wxFONTENCODING_SYSTEM)
+        m_conv = NULL;
+    else

         m_conv = new wxCSConv(wxFontMapper::GetEncodingName(m_encoding));
 #else
     (void) encoding;
         m_conv = new wxCSConv(wxFontMapper::GetEncodingName(m_encoding));
 #else
     (void) encoding;


@@ -446,15 +497,21 @@ wxString wxHtmlEntitiesParser::Parse(const wxString& input)
     const wxChar *in_str = input.c_str();
     wxString output;
 
     const wxChar *in_str = input.c_str();
     wxString output;
 

+    output.reserve(input.length());
+

     for (c = in_str, last = in_str; *c != wxT('\0'); c++)
     {
         if (*c == wxT('&'))
         {
             if (c - last > 0)
                 output.append(last, c - last);
     for (c = in_str, last = in_str; *c != wxT('\0'); c++)
     {
         if (*c == wxT('&'))
         {
             if (c - last > 0)
                 output.append(last, c - last);

-            if (++c == wxT('\0')) break;
+            if ( *++c == wxT('\0') )
+                break;
+

             wxString entity;
             const wxChar *ent_s = c;
             wxString entity;
             const wxChar *ent_s = c;

+            wxChar entity_char;
+

             for (; (*c >= wxT('a') && *c <= wxT('z')) ||
                    (*c >= wxT('A') && *c <= wxT('Z')) ||
                    (*c >= wxT('0') && *c <= wxT('9')) ||
             for (; (*c >= wxT('a') && *c <= wxT('z')) ||
                    (*c >= wxT('A') && *c <= wxT('Z')) ||
                    (*c >= wxT('0') && *c <= wxT('9')) ||


@@ -462,7 +519,16 @@ wxString wxHtmlEntitiesParser::Parse(const wxString& input)
             entity.append(ent_s, c - ent_s);
             if (*c != wxT(';')) c--;
             last = c+1;
             entity.append(ent_s, c - ent_s);
             if (*c != wxT(';')) c--;
             last = c+1;

-            output << GetEntityChar(entity);
+            entity_char = GetEntityChar(entity);
+            if (entity_char)
+                output << entity_char;
+            else
+            {
+                output.append(ent_s-1, c-ent_s+2);
+                wxLogTrace(wxTRACE_HTML_DEBUG,
+                           wxT("Unrecognized HTML entity: '%s'"),
+                           entity.c_str());
+            }

         }
     }
     if (*last != wxT('\0'))
         }
     }
     if (*last != wxT('\0'))


@@ -476,16 +542,15 @@ struct wxHtmlEntityInfo
     unsigned code;
 };
 
     unsigned code;
 };
 

-static int LINKAGEMODE compar_entity(const void *key, const void *item)
+extern "C" int LINKAGEMODE wxHtmlEntityCompare(const void *key, const void *item)

 {
     return wxStrcmp((wxChar*)key, ((wxHtmlEntityInfo*)item)->name);
 }
 
 {
     return wxStrcmp((wxChar*)key, ((wxHtmlEntityInfo*)item)->name);
 }
 

+#if !wxUSE_UNICODE

 wxChar wxHtmlEntitiesParser::GetCharForCode(unsigned code)
 {
 wxChar wxHtmlEntitiesParser::GetCharForCode(unsigned code)
 {

-#if wxUSE_UNICODE
-    return (wxChar)code;
-#elif wxUSE_WCHAR_T
+#if wxUSE_WCHAR_T

     char buf[2];
     wchar_t wbuf[2];
     wbuf[0] = (wchar_t)code;
     char buf[2];
     wchar_t wbuf[2];
     wbuf[0] = (wchar_t)code;


@@ -498,6 +563,7 @@ wxChar wxHtmlEntitiesParser::GetCharForCode(unsigned code)
     return (code < 256) ? (wxChar)code : '?';
 #endif
 }
     return (code < 256) ? (wxChar)code : '?';
 #endif
 }

+#endif

 
 wxChar wxHtmlEntitiesParser::GetEntityChar(const wxString& entity)
 {
 
 wxChar wxHtmlEntitiesParser::GetEntityChar(const wxString& entity)
 {


@@ -782,19 +848,107 @@ wxChar wxHtmlEntitiesParser::GetEntityChar(const wxString& entity)
             while (substitutions[substitutions_cnt].code != 0)
                 substitutions_cnt++;
 
             while (substitutions[substitutions_cnt].code != 0)
                 substitutions_cnt++;
 

-        wxHtmlEntityInfo *info;
+        wxHtmlEntityInfo *info = NULL;
+#ifdef __WXWINCE__
+        // bsearch crashes under WinCE for some reason
+        size_t i;
+        for (i = 0; i < substitutions_cnt; i++)
+        {
+            if (entity == substitutions[i].name)
+            {
+                info = & substitutions[i];
+                break;
+            }
+        }
+#else

         info = (wxHtmlEntityInfo*) bsearch(entity.c_str(), substitutions,
                                            substitutions_cnt,
                                            sizeof(wxHtmlEntityInfo),
         info = (wxHtmlEntityInfo*) bsearch(entity.c_str(), substitutions,
                                            substitutions_cnt,
                                            sizeof(wxHtmlEntityInfo),

-                                           compar_entity);
+                                           wxHtmlEntityCompare);
+#endif

         if (info)
             code = info->code;
     }
 
     if (code == 0)
         if (info)
             code = info->code;
     }
 
     if (code == 0)

-        return wxT('?');
+        return 0;

     else
         return GetCharForCode(code);
 }
 
     else
         return GetCharForCode(code);
 }
 

+wxFSFile *wxHtmlParser::OpenURL(wxHtmlURLType WXUNUSED(type),
+                                const wxString& url) const
+{
+    return m_FS ? m_FS->OpenFile(url) : NULL;
+
+}
+
+
+//-----------------------------------------------------------------------------
+// wxHtmlParser::ExtractCharsetInformation
+//-----------------------------------------------------------------------------
+
+class wxMetaTagParser : public wxHtmlParser
+{
+public:
+    wxMetaTagParser() { }
+
+    wxObject* GetProduct() { return NULL; }
+
+protected:
+    virtual void AddText(const wxChar* WXUNUSED(txt)) {}
+
+    DECLARE_NO_COPY_CLASS(wxMetaTagParser)
+};
+
+class wxMetaTagHandler : public wxHtmlTagHandler
+{
+public:
+    wxMetaTagHandler(wxString *retval) : wxHtmlTagHandler(), m_retval(retval) {}
+    wxString GetSupportedTags() { return wxT("META,BODY"); }
+    bool HandleTag(const wxHtmlTag& tag);
+
+private:
+    wxString *m_retval;
+
+    DECLARE_NO_COPY_CLASS(wxMetaTagHandler)
+};
+
+bool wxMetaTagHandler::HandleTag(const wxHtmlTag& tag)
+{
+    if (tag.GetName() == _T("BODY"))
+    {
+        m_Parser->StopParsing();
+        return false;
+    }
+
+    if (tag.HasParam(_T("HTTP-EQUIV")) &&
+        tag.GetParam(_T("HTTP-EQUIV")).IsSameAs(_T("Content-Type"), false) &&
+        tag.HasParam(_T("CONTENT")))
+    {
+        wxString content = tag.GetParam(_T("CONTENT")).Lower();
+        if (content.Left(19) == _T("text/html; charset="))
+        {
+            *m_retval = content.Mid(19);
+            m_Parser->StopParsing();
+        }
+    }
+    return false;
+}
+
+
+/*static*/
+wxString wxHtmlParser::ExtractCharsetInformation(const wxString& markup)
+{
+    wxString charset;
+    wxMetaTagParser *parser = new wxMetaTagParser();
+    if(parser)
+    {
+        parser->AddTagHandler(new wxMetaTagHandler(&charset));
+        parser->Parse(markup);
+        delete parser;
+    }
+    return charset;
+}
+

 #endif
 #endif