// 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"
#include "wx/defs.h"
#if wxUSE_HTML && wxUSE_STREAMS
-#ifdef __BORDLANDC__
+#ifdef __BORLANDC__
#pragma hdrstop
#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")
+
//-----------------------------------------------------------------------------
// 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;
}
wxObject* wxHtmlParser::Parse(const wxString& source)
{
- wxObject *result;
-
InitParser(source);
DoParsing();
- result = GetProduct();
+ wxObject *result = GetProduct();
DoneParser();
return result;
}
void wxHtmlParser::InitParser(const wxString& source)
{
SetSource(source);
+ 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);
}
else
i = chd->GetBeginPos();
+
textBeginning = i;
}
wxHtmlTag *t = m_CurTag;
m_CurTag = m_CurTag->GetNextTag();
AddTag(*t);
+ if (m_stopParsing)
+ return;
}
else break;
}
h = (wxHtmlTagHandler*) m_HandlersHash.Get(tag.GetName());
if (h)
+ {
inner = h->HandleTag(tag);
+ if (m_stopParsing)
+ return;
+ }
if (!inner)
{
if (tag.HasEnding())
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;
-
- 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());
- m_HandlersStack->DeleteNode(first);
+ delete (wxHashTable*) first->GetData();
+ m_HandlersStack->Erase(first);
}
void wxHtmlParser::SetSourceAndSaveState(const wxString& src)
void wxHtmlEntitiesParser::SetEncoding(wxFontEncoding encoding)
{
#if wxUSE_WCHAR_T && !wxUSE_UNICODE
- if (encoding == m_encoding) return;
+ if (encoding == m_encoding)
+ return;
+
delete m_conv;
- m_conv = NULL;
+
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;
return wxStrcmp((wxChar*)key, ((wxHtmlEntityInfo*)item)->name);
}
+#if !wxUSE_UNICODE
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;
return (code < 256) ? (wxChar)code : '?';
#endif
}
+#endif
wxChar wxHtmlEntitiesParser::GetEntityChar(const wxString& entity)
{
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