backported correct CDATA and <script> handling

git-svn-id: https://svn.wxwidgets.org/svn/wx/wxWidgets/branches/WX_2_4_BRANCH@20065 c3d73ce0-8a6f-49c7-b76d-6d57e0e08775
This commit is contained in:
Václav Slavík
2003-04-07 22:11:53 +00:00
parent 9d308583a4
commit 0da58ad502
3 changed files with 68 additions and 0 deletions

View File

@@ -127,6 +127,8 @@ void wxHtmlParser::CreateDOMTree()
m_CurTextPiece = 0;
}
extern bool wxIsCDATAElement(const wxChar *tag);
void wxHtmlParser::CreateDOMSubTree(wxHtmlTag *cur,
int begin_pos, int end_pos,
wxHtmlTagsCache *cache)
@@ -137,6 +139,15 @@ void wxHtmlParser::CreateDOMSubTree(wxHtmlTag *cur,
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);
@@ -209,6 +220,7 @@ void wxHtmlParser::CreateDOMSubTree(wxHtmlTag *cur,
}
else
i = chd->GetBeginPos();
textBeginning = i;
}

View File

@@ -57,6 +57,12 @@ IMPLEMENT_CLASS(wxHtmlTagsCache,wxObject)
#define CACHE_INCREMENT 64
bool wxIsCDATAElement(const wxChar *tag)
{
return (wxStrcmp(tag, _T("SCRIPT")) == 0) ||
(wxStrcmp(tag, _T("STYLE")) == 0);
}
wxHtmlTagsCache::wxHtmlTagsCache(const wxString& source)
{
const wxChar *src = source.c_str();
@@ -108,6 +114,47 @@ wxHtmlTagsCache::wxHtmlTagsCache(const wxString& source)
else
{
m_Cache[tg].End1 = m_Cache[tg].End2 = -1;
if (wxIsCDATAElement(tagBuffer))
{
// find next matching tag
int tag_len = wxStrlen(tagBuffer);
while (pos < lng)
{
// find the ending tag
while (pos + 1 < lng &&
(src[pos] != '<' || src[pos+1] != '/'))
++pos;
if (src[pos] == '<')
++pos;
// see if it matches
int match_pos = 0;
while (pos < lng && match_pos < tag_len && src[pos] != '>' && src[pos] != '<') {
if (wxToupper(src[pos]) == tagBuffer[match_pos]) {
++match_pos;
}
else if (src[pos] == wxT(' ') || src[pos] == wxT('\n') ||
src[pos] == wxT('\r') || src[pos] == wxT('\t')) {
// need to skip over these
}
else {
match_pos = 0;
}
++pos;
}
// found a match
if (match_pos == tag_len) {
pos = pos - tag_len - 3;
stpos = pos;
break;
}
else {
++pos;
}
}
}
}
}

View File

@@ -224,6 +224,14 @@ TAG_HANDLER_END(BLOCKQUOTE)
// Tag handler for tags that we have to ignore, otherwise non-text data
// would show up as text:
TAG_HANDLER_BEGIN(DoNothing, "SCRIPT")
TAG_HANDLER_PROC(tag)
{
return true;
}
TAG_HANDLER_END(DoNothing)
@@ -236,6 +244,7 @@ TAGS_MODULE_BEGIN(Layout)
TAGS_MODULE_ADD(TITLE)
TAGS_MODULE_ADD(BODY)
TAGS_MODULE_ADD(BLOCKQUOTE)
TAGS_MODULE_ADD(DoNothing)
TAGS_MODULE_END(Layout)