void splittext(const char *text, u32t width, TStrings &lst, u32t flags) {
lst.Clear();
if (!text || width<8) return;
const char *ps = text, *pps;
char softcr = flags&SplitText_HelpMode?'&':0;
int ln = 0;
spstr linehead;
lst.Add(spstr());
do {
while (*ps && (strchr(AllowSet,*ps) || *ps==softcr)) {
register char cr=*ps++;
if (cr!=' ') {
if (cr==softcr) {
pps = ps;
if (*pps) pps++;
while (*pps==' ' || *pps=='\t') pps++;
/* get 1 char after eol and all spaces after it and use result
as a header for all next lines until real cr */
linehead = spstr(ps, pps-ps).expandtabs(4);
ps = pps;
} else {
lst.Add(spstr());
if (cr=='\r' && *ps=='\n') ps++;
linehead.clear();
if (lst[ln].lastchar()==' ') lst[ln].dellast();
ln++;
}
}
}
if (*ps==0) break;
pps = strpbrk(ps, softcr ? CarrySet AllowSet "&" : CarrySet AllowSet);
int carry = pps&&strchr(CarrySet,*pps)?1:0;
spstr curr(ps,pps?pps-ps+carry:0), sum;
sum = lst[ln]+curr;
sum.expandtabs(4);
if ((flags&SplitText_NoAnsi?strlen:str_length)(sum())<width-2)
lst[ln]=sum;
else {
lst.Add(linehead+curr);
if (lst[ln].lastchar()==' ') lst[ln].dellast();
ln++;
}
if (!carry||pps&&pps[1]==' ') lst[ln]+=' ';
ps=pps;
if (carry) ps++;
} while (pps);
}
std::string HtmlStrDecode(const std::string& htmlStr, const char* encTo, bool decodeChar){
if(htmlStr.empty() || encTo == NULL)
return "";
std::string destStr = htmlStr;
std::string tmpStr = destStr;
boost::smatch what;
// dec unicode convert 年
boost::regex unicodeDecReg("&#\\d{1,6};");
std::string::const_iterator start = tmpStr.begin();
std::string::const_iterator end = tmpStr.end();
while ( boost::regex_search(start, end, what, unicodeDecReg) )
{
// split unicode number
std::string unicode(what[0].str(), 2, what[0].str().size() - 3);
// convert unicode to char
std::string dest = ConvChar(atoi(unicode.c_str()), encTo, "UTF-16");
// find next unicode number
start = what[0].second;
// replace html unicode to char
boost::replace_all(destStr, what[0].str(), dest);
}
// standard unicode convert \u5e74
boost::regex unicodeReg("\\\\u\\w{1,4}");
tmpStr = destStr; start = tmpStr.begin(); end = tmpStr.end();
while ( boost::regex_search(start, end, what, unicodeReg) )
{
std::string str(what[0].str(), 2, what[0].str().size() - 2);
uint32_t unicode = hex_to_decimal(str.c_str(), str.size());
std::string dest = ConvChar(unicode, encTo, "UTF-16");
boost::replace_all(destStr, what[0].str(), dest);
start = what[0].second;
}
// Hex unicode convert 年
boost::regex unicodeHexReg("&#x\\w{1,4}");
tmpStr = destStr; start = tmpStr.begin(); end = tmpStr.end();
while ( boost::regex_search(start, end, what, unicodeHexReg) )
{
// split unicode number
std::string str(what[0].str(), 3, what[0].str().size() - 3);
uint32_t unicode = hex_to_decimal(str.c_str(), str.size());
std::string dest = ConvChar(unicode, encTo, "UTF-16");
boost::replace_all(destStr, what[0].str(), dest);
start = what[0].second;
}
// special char convert
boost::regex spCharReg("&[A-Za-z0-9]{2,8};");
tmpStr = destStr; start = tmpStr.begin(); end = tmpStr.end();
while ( boost::regex_search(start, end, what, spCharReg) )
{
std::string spstr(what[0].str(), 1, what[0].str().size() - 2);
std::map<std::string, uint32_t>::const_iterator it = HtmlSpMap.find(spstr);
if(HtmlSpMap.end() != it){
std::string dest = ConvChar(it->second, encTo, "UTF-16");
boost::replace_all(destStr, what[0].str(), dest);
}
start = what[0].second;
}
return destStr;
}
