#ifdef __GNUG__
-#pragma implementation
+#pragma implementation "htmlpars.h"
#endif
#include "wx/wxprec.h"
WX_DECLARE_OBJARRAY(wxHtmlTextPiece, wxHtmlTextPieces);
WX_DEFINE_OBJARRAY(wxHtmlTextPieces);
-struct wxHtmlParserState
+class wxHtmlParserState
{
+public:
wxHtmlTag *m_curTag;
wxHtmlTag *m_tags;
wxHtmlTextPieces *m_textPieces;
wxHtmlParser::~wxHtmlParser()
{
- while (RestoreState())
- DestroyDOMTree();
+ while (RestoreState()) {}
+ DestroyDOMTree();
+
delete m_HandlersStack;
m_HandlersHash.Clear();
m_HandlersList.DeleteContents(TRUE);
wxChar c;
int i = begin_pos;
int textBeginning = begin_pos;
-
+
while (i < end_pos)
{
c = m_Source.GetChar(i);
while (i < end_pos)
{
c = m_Source.GetChar(i++);
- if ((c == wxT(' ') || c == wxT('\n') ||
+ if ((c == wxT(' ') || c == wxT('\n') ||
c == wxT('\r') || c == wxT('\t')) && dashes >= 2) {}
else if (c == wxT('>') && dashes >= 2)
{
}
else if (c == wxT('-'))
dashes++;
- else
+ else
dashes = 0;
}
}
-
+
// 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,
+ if (cur)
+ chd = new wxHtmlTag(cur, m_Source,
i, end_pos, cache, m_entitiesParser);
- else
+ else
{
chd = new wxHtmlTag(NULL, m_Source,
i, end_pos, cache, m_entitiesParser);
- if (!m_Tags)
+ if (!m_Tags)
{
- // if this is the first tag to be created make the root
+ // if this is the first tag to be created make the root
// m_Tags point to it:
m_Tags = chd;
}
else
{
- // if there is already a root tag add this tag as
+ // if there is already a root tag add this tag as
// the last sibling:
chd->m_Prev = m_Tags->GetLastSibling();
chd->m_Prev->m_Next = chd;
if (chd->HasEnding())
{
CreateDOMSubTree(chd,
- chd->GetBeginPos(), chd->GetEndPos1(),
+ chd->GetBeginPos(), chd->GetEndPos1(),
cache);
i = chd->GetEndPos2();
}
}
// ... or skip ending tag:
- else
+ else
{
while (i < end_pos && m_Source.GetChar(i) != wxT('>')) i++;
textBeginning = i+1;
m_TextPieces = NULL;
}
-void wxHtmlParser::DoParsing()
+void wxHtmlParser::DoParsing()
{
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)
{
if (end_pos <= begin_pos) return;
-
+
wxHtmlTextPieces& pieces = *m_TextPieces;
size_t piecesCnt = pieces.GetCount();
-
+
while (begin_pos < end_pos)
{
while (m_CurTag && m_CurTag->GetBeginPos() < begin_pos)
m_CurTag = m_CurTag->GetNextTag();
- while (m_CurTextPiece < piecesCnt &&
+ while (m_CurTextPiece < piecesCnt &&
pieces[m_CurTextPiece].m_pos < begin_pos)
m_CurTextPiece++;
- if (m_CurTextPiece < piecesCnt &&
- (!m_CurTag ||
+ if (m_CurTextPiece < piecesCnt &&
+ (!m_CurTag ||
pieces[m_CurTextPiece].m_pos < m_CurTag->GetBeginPos()))
{
// Add text:
- AddText(m_Source.Mid(pieces[m_CurTextPiece].m_pos,
- pieces[m_CurTextPiece].m_lng));
- begin_pos = pieces[m_CurTextPiece].m_pos +
+ 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++;
}
{
if (m_CurTag->HasEnding())
begin_pos = m_CurTag->GetEndPos2();
- else
+ else
begin_pos = m_CurTag->GetBeginPos();
}
wxHtmlTag *t = m_CurTag;
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);
while (tokenizer.HasMoreTokens())
{
- key = tokenizer.NextToken();
+ key = tokenizer.GetNextToken();
m_HandlersHash.Delete(key);
m_HandlersHash.Put(key, handler);
}
m_TextPieces = NULL;
m_CurTextPiece = 0;
m_Source = wxEmptyString;
-
+
SetSource(src);
}
bool wxHtmlParser::RestoreState()
{
if (!m_SavedStates) return FALSE;
-
+
+ DestroyDOMTree();
+
wxHtmlParserState *s = m_SavedStates;
m_SavedStates = s->m_nextState;
-
+
m_CurTag = s->m_curTag;
m_Tags = s->m_tags;
m_TextPieces = s->m_textPieces;
m_CurTextPiece = s->m_curTextPiece;
m_Source = s->m_source;
-
+
delete s;
return TRUE;
}
m_encoding = encoding;
if (m_encoding != wxFONTENCODING_SYSTEM)
m_conv = new wxCSConv(wxFontMapper::GetEncodingName(m_encoding));
+#else
+ (void) encoding;
#endif
}
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')) ||
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);
+ wxLogDebug(wxT("Unrecognized HTML entity: '%s'"), entity.c_str());
+ }
}
}
if (*last != wxT('\0'))
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);
}
wbuf[0] = (wchar_t)code;
wbuf[1] = 0;
wxMBConv *conv = m_conv ? m_conv : &wxConvLocal;
- if (conv->WC2MB(buf, wbuf, 1) == (size_t)-1)
+ if (conv->WC2MB(buf, wbuf, 2) == (size_t)-1)
return '?';
return buf[0];
#else
info = (wxHtmlEntityInfo*) bsearch(entity.c_str(), substitutions,
substitutions_cnt,
sizeof(wxHtmlEntityInfo),
- compar_entity);
+ wxHtmlEntityCompare);
if (info)
code = info->code;
}
if (code == 0)
- return wxT('?');
+ return 0;
else
return GetCharForCode(code);
}
+wxFSFile *wxHtmlParser::OpenURL(wxHtmlURLType WXUNUSED(type),
+ const wxString& url) const
+{
+ return GetFS()->OpenFile(url);
+}
+
#endif
projects / wxWidgets.git / blobdiff