/////////////////////////////////////////////////////////////////////////////
-// Name: htmlpars.cpp
+// Name: src/html/htmlpars.cpp
// Purpose: wxHtmlParser class (generic parser)
// Author: Vaclav Slavik
// RCS-ID: $Id$
// Copyright: (c) 1999 Vaclav Slavik
-// Licence: wxWindows Licence
+// Licence: wxWindows licence
/////////////////////////////////////////////////////////////////////////////
-
-#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 __BORLANDC__
-#pragma hdrstop
+ #pragma hdrstop
#endif
-#ifndef WXPRECOMP
+#if wxUSE_HTML && wxUSE_STREAMS
+
+#ifndef WX_PRECOMP
+ #include "wx/dynarray.h"
#include "wx/log.h"
#include "wx/intl.h"
+ #include "wx/app.h"
+ #include "wx/wxcrtvararg.h"
#endif
#include "wx/tokenzr.h"
#include "wx/fontmap.h"
#include "wx/html/htmldefs.h"
#include "wx/html/htmlpars.h"
-#include "wx/dynarray.h"
#include "wx/arrimpl.cpp"
#ifdef __WXWINCE__
#endif
// DLL options compatibility check:
-#include "wx/app.h"
WX_CHECK_BUILD_OPTIONS("wxHTML")
+const wxChar *wxTRACE_HTML_DEBUG = _T("htmldebug");
+
//-----------------------------------------------------------------------------
// wxHtmlParser helpers
//-----------------------------------------------------------------------------
};
WX_DECLARE_OBJARRAY(wxHtmlTextPiece, wxHtmlTextPieces);
-WX_DEFINE_OBJARRAY(wxHtmlTextPieces);
+WX_DEFINE_OBJARRAY(wxHtmlTextPieces)
class wxHtmlParserState
{
void wxHtmlParser::InitParser(const wxString& source)
{
SetSource(source);
- m_stopParsing = FALSE;
+ m_stopParsing = false;
}
void wxHtmlParser::DoneParser()
{
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;
}
wxHtmlTextPiece(textBeginning, i - textBeginning));
// if it is a comment, skip it:
- if (i < end_pos-6 && m_Source.GetChar(i+1) == wxT('!') &&
- m_Source.GetChar(i+2) == wxT('-') &&
- m_Source.GetChar(i+3) == wxT('-'))
+ wxString::const_iterator iter = m_Source.begin() + i;
+ if ( SkipCommentTag(iter, m_Source.end()) )
{
- // Comments begin with "<!--" and end with "--[ \t\r\n]*>"
- // according to HTML 4.0
- int dashes = 0;
- i += 4;
- while (i < end_pos)
- {
- c = m_Source.GetChar(i++);
- if ((c == wxT(' ') || c == wxT('\n') ||
- c == wxT('\r') || c == wxT('\t')) && dashes >= 2) {}
- else if (c == wxT('>') && dashes >= 2)
- {
- textBeginning = i;
- break;
- }
- else if (c == wxT('-'))
- dashes++;
- else
- dashes = 0;
- }
+ textBeginning =
+ i = iter - m_Source.begin() + 1; // skip closing '>' too
}
// 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;
}
{
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)
}
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);
void wxHtmlParser::AddTag(const wxHtmlTag& tag)
{
wxHtmlTagHandler *h;
- bool inner = FALSE;
+ bool inner = false;
h = (wxHtmlTagHandler*) m_HandlersHash.Get(tag.GetName());
if (h)
handler->SetParser(this);
}
-void wxHtmlParser::PushTagHandler(wxHtmlTagHandler *handler, wxString tags)
+void wxHtmlParser::PushTagHandler(wxHtmlTagHandler *handler, const wxString& tags)
{
wxStringTokenizer tokenizer(tags, wxT(", "));
wxString key;
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;
+}
+
+wxString wxHtmlParser::GetInnerSource(const wxHtmlTag& tag)
+{
+ return GetSource()->Mid(tag.GetBeginPos(),
+ tag.GetEndPos1() - tag.GetBeginPos());
}
//-----------------------------------------------------------------------------
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
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;
-
+ 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());
}
}
}
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),
wxHtmlEntityCompare);
+#endif
if (info)
code = info->code;
}
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;
+
}
if (tag.GetName() == _T("BODY"))
{
m_Parser->StopParsing();
- return FALSE;
+ return false;
}
if (tag.HasParam(_T("HTTP-EQUIV")) &&
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
+/* static */
+bool
+wxHtmlParser::SkipCommentTag(wxString::const_iterator& start,
+ wxString::const_iterator end)
+{
+ wxASSERT_MSG( *start == '<', _T("should be called on the tag start") );
+
+ wxString::const_iterator p = start;
+
+ // comments begin with "<!--" in HTML 4.0
+ if ( end - p < 3 || *++p != '!' || *++p != '-' || *++p != '-' )
+ {
+ // not a comment at all
+ return false;
+ }
+
+ // skip the start of the comment tag in any case, if we don't find the
+ // closing tag we should ignore broken markup
+ start = p;
+
+ // comments end with "--[ \t\r\n]*>", i.e. white space is allowed between
+ // comment delimiter and the closing tag character (section 3.2.4 of
+ // http://www.w3.org/TR/html401/)
+ int dashes = 0;
+ while ( ++p < end )
+ {
+ const wxChar c = *p;
+
+ if ( (c == wxT(' ') || c == wxT('\n') ||
+ c == wxT('\r') || c == wxT('\t')) && dashes >= 2 )
+ {
+ // ignore white space before potential tag end
+ continue;
+ }
+
+ if ( c == wxT('>') && dashes >= 2 )
+ {
+ // found end of comment
+ start = p;
+ break;
+ }
+
+ if ( c == wxT('-') )
+ dashes++;
+ else
+ dashes = 0;
+ }
+
+ return true;
+}
+
+#endif // wxUSE_HTML
projects / wxWidgets.git / blobdiff