// Author: Vaclav Slavik
// RCS-ID: $Id$
// Copyright: (c) 1999 Vaclav Slavik
-// Licence: wxWindows Licence
+// Licence: wxWindows licence
/////////////////////////////////////////////////////////////////////////////
-#ifdef __GNUG__
+#if defined(__GNUG__) && !defined(NO_GCC_PRAGMA)
#pragma implementation "htmlpars.h"
#endif
#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
//-----------------------------------------------------------------------------
{
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();
- m_HandlersList.DeleteContents(TRUE);
- m_HandlersList.Clear();
+ WX_CLEAR_LIST(wxList, m_HandlersList);
delete m_entitiesParser;
}
void wxHtmlParser::InitParser(const wxString& source)
{
SetSource(source);
- m_stopParsing = FALSE;
+ m_stopParsing = false;
}
void wxHtmlParser::DoneParser()
m_CurTextPiece = 0;
}
+extern bool wxIsCDATAElement(const wxChar *tag);
+
void wxHtmlParser::CreateDOMSubTree(wxHtmlTag *cur,
int begin_pos, int end_pos,
wxHtmlTagsCache *cache)
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);
// 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,
}
else
i = chd->GetBeginPos();
+
textBeginning = i;
}
void wxHtmlParser::AddTag(const wxHtmlTag& tag)
{
wxHtmlTagHandler *h;
- bool inner = FALSE;
+ bool inner = false;
h = (wxHtmlTagHandler*) m_HandlersHash.Get(tag.GetName());
if (h)
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())
{
void wxHtmlParser::PopTagHandler()
{
- wxNode *first;
+ wxList::compatibility_iterator first;
- if (m_HandlersStack == NULL ||
- (first = m_HandlersStack->GetFirst()) == NULL)
+ 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());
- m_HandlersStack->DeleteNode(first);
+ delete (wxHashTable*) first->GetData();
+ m_HandlersStack->Erase(first);
}
void wxHtmlParser::SetSourceAndSaveState(const wxString& src)
bool wxHtmlParser::RestoreState()
{
- if (!m_SavedStates) return FALSE;
+ if (!m_SavedStates) return false;
DestroyDOMTree();
m_Source = s->m_source;
delete s;
- return TRUE;
+ return true;
}
//-----------------------------------------------------------------------------
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++)
if (c - last > 0)
output.append(last, c - last);
if (++c == wxT('\0')) break;
-
+
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')) ||
else
{
output.append(ent_s-1, c-ent_s+2);
- wxLogDebug(wxT("Unrecognized HTML entity: '%s'"), entity.c_str());
+ wxLogTrace(wxTRACE_HTML_DEBUG,
+ wxT("Unrecognized HTML entity: '%s'"),
+ entity.c_str());
}
}
}
return GetCharForCode(code);
}
-wxFSFile *wxHtmlParser::OpenURL(wxHtmlURLType WXUNUSED(type),
+wxFSFile *wxHtmlParser::OpenURL(wxHtmlURLType WXUNUSED(type),
const wxString& url) const
{
- return GetFS()->OpenFile(url);
+ return m_FS ? m_FS->OpenFile(url) : NULL;
+
}
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
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;
+ return false;
}
if (tag.HasParam(_T("HTTP-EQUIV")) &&
- tag.GetParam(_T("HTTP-EQUIV")) == _T("Content-Type") &&
+ tag.GetParam(_T("HTTP-EQUIV")).IsSameAs(_T("Content-Type"), false) &&
tag.HasParam(_T("CONTENT")))
{
- wxString content = tag.GetParam(_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;
+ return false;
}
wxString wxHtmlParser::ExtractCharsetInformation(const wxString& markup)
{
wxString charset;
- wxMetaTagParser parser;
- parser.AddTagHandler(new wxMetaTagHandler(&charset));
- parser.Parse(markup);
+ wxMetaTagParser *parser = new wxMetaTagParser();
+ if(parser)
+ {
+ parser->AddTagHandler(new wxMetaTagHandler(&charset));
+ parser->Parse(markup);
+ delete parser;
+ }
return charset;
}
-
#endif