]> git.saurik.com Git - wxWidgets.git/blobdiff - src/html/htmlpars.cpp
Compilo.
[wxWidgets.git] / src / html / htmlpars.cpp
index 64054fc2d2f246bdf211e123bd3d607688c28cc0..7cee2a8e061775d50413d7ef1c9830562e680bd6 100644 (file)
@@ -4,12 +4,12 @@
 // Author:      Vaclav Slavik
 // RCS-ID:      $Id$
 // Copyright:   (c) 1999 Vaclav Slavik
 // Author:      Vaclav Slavik
 // RCS-ID:      $Id$
 // Copyright:   (c) 1999 Vaclav Slavik
-// Licence:     wxWindows Licence
+// Licence:     wxWindows licence
 /////////////////////////////////////////////////////////////////////////////
 
 
 /////////////////////////////////////////////////////////////////////////////
 
 
-#ifdef __GNUG__
-#pragma implementation
+#if defined(__GNUG__) && !defined(NO_GCC_PRAGMA)
+#pragma implementation "htmlpars.h"
 #endif
 
 #include "wx/wxprec.h"
 #endif
 
 #include "wx/wxprec.h"
@@ -17,7 +17,7 @@
 #include "wx/defs.h"
 #if wxUSE_HTML && wxUSE_STREAMS
 
 #include "wx/defs.h"
 #if wxUSE_HTML && wxUSE_STREAMS
 
-#ifdef __BORDLANDC__
+#ifdef __BORLANDC__
 #pragma hdrstop
 #endif
 
 #pragma hdrstop
 #endif
 
 #include "wx/dynarray.h"
 #include "wx/arrimpl.cpp"
 
 #include "wx/dynarray.h"
 #include "wx/arrimpl.cpp"
 
+#ifdef __WXWINCE__
+    #include "wx/msw/wince/missing.h"       // for bsearch()
+#endif
+
+// DLL options compatibility check:
+#include "wx/app.h"
+WX_CHECK_BUILD_OPTIONS("wxHTML")
+
+const wxChar *wxTRACE_HTML_DEBUG = _T("htmldebug");
+
 //-----------------------------------------------------------------------------
 // wxHtmlParser helpers
 //-----------------------------------------------------------------------------
 //-----------------------------------------------------------------------------
 // wxHtmlParser helpers
 //-----------------------------------------------------------------------------
@@ -80,22 +90,28 @@ 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 +119,7 @@ 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()
@@ -127,6 +144,8 @@ void wxHtmlParser::CreateDOMTree()
     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 +156,15 @@ 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);
@@ -209,6 +237,7 @@ void wxHtmlParser::CreateDOMSubTree(wxHtmlTag *cur,
                 }
                 else
                     i = chd->GetBeginPos();
                 }
                 else
                     i = chd->GetBeginPos();
+                
                 textBeginning = i;
             }
 
                 textBeginning = i;
             }
 
@@ -271,8 +300,9 @@ void wxHtmlParser::DoParsing(int begin_pos, int end_pos)
              pieces[m_CurTextPiece].m_pos < m_CurTag->GetBeginPos()))
         {
             // Add text:
              pieces[m_CurTextPiece].m_pos < m_CurTag->GetBeginPos()))
         {
             // Add text:
-            AddText(m_Source.Mid(pieces[m_CurTextPiece].m_pos,
-                                 pieces[m_CurTextPiece].m_lng));
+            AddText(GetEntitiesParser()->Parse(
+                       m_Source.Mid(pieces[m_CurTextPiece].m_pos,
+                                    pieces[m_CurTextPiece].m_lng)));
             begin_pos = pieces[m_CurTextPiece].m_pos +
                         pieces[m_CurTextPiece].m_lng;
             m_CurTextPiece++;
             begin_pos = pieces[m_CurTextPiece].m_pos +
                         pieces[m_CurTextPiece].m_lng;
             m_CurTextPiece++;
@@ -290,6 +320,8 @@ void wxHtmlParser::DoParsing(int begin_pos, int end_pos)
             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;
     }
@@ -302,7 +334,11 @@ void wxHtmlParser::AddTag(const wxHtmlTag& tag)
 
     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())
@@ -316,7 +352,7 @@ 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);
@@ -332,14 +368,13 @@ 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);
     }
@@ -347,16 +382,22 @@ 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)
@@ -385,6 +426,8 @@ 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;
 
@@ -428,11 +471,15 @@ 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;
@@ -444,6 +491,8 @@ wxString wxHtmlEntitiesParser::Parse(const wxString& input)
     const wxChar *c, *last;
     const wxChar *in_str = input.c_str();
     wxString output;
     const wxChar *c, *last;
     const wxChar *in_str = input.c_str();
     wxString output;
+    
+    output.reserve(input.length());
 
     for (c = in_str, last = in_str; *c != wxT('\0'); c++)
     {
 
     for (c = in_str, last = in_str; *c != wxT('\0'); c++)
     {
@@ -452,8 +501,11 @@ wxString wxHtmlEntitiesParser::Parse(const wxString& input)
             if (c - last > 0)
                 output.append(last, c - last);
             if (++c == wxT('\0')) break;
             if (c - last > 0)
                 output.append(last, c - last);
             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')) ||
@@ -461,7 +513,16 @@ 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'))
@@ -475,16 +536,15 @@ 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;
@@ -497,6 +557,7 @@ 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)
 {
@@ -785,15 +846,86 @@ wxChar wxHtmlEntitiesParser::GetEntityChar(const wxString& entity)
         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);
         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;
+    parser.AddTagHandler(new wxMetaTagHandler(&charset));
+    parser.Parse(markup);
+    return charset;
+}
+
 #endif
 #endif