inline void kbe_split(const std::basic_string<T>& s, T c, std::vector< std::basic_string<T> > &v)
{
if(s.size() == 0)
return;
typename std::basic_string< T >::size_type i = 0;
typename std::basic_string< T >::size_type j = s.find(c);
while(j != std::basic_string<T>::npos)
{
std::basic_string<T> buf = s.substr(i, j - i);
if(buf.size() > 0)
v.push_back(buf);
i = ++j;
j = s.find(c, j);
}
if(j == std::basic_string<T>::npos)
{
std::basic_string<T> buf = s.substr(i, s.length() - i);
if(buf.size() > 0)
v.push_back(buf);
}
}
std::basic_string<C>
perl_s (std::basic_string<C> const& src, std::basic_string<C> const& e)
{
typedef std::basic_string<C> string;
typedef typename string::size_type size;
if (e.empty ())
return src;
C delimiter (e[0]);
size first = e.find (delimiter);
size middle = e.find (delimiter, first + 1);
size last = e.find (delimiter, middle + 1);
string pattern (e, first + 1, middle - first - 1);
string format (e, middle + 1, last - middle - 1);
//std::cout << pattern << " " << format << std::endl;
boost::basic_regex<C> expr (pattern);
return regex_merge (
src,
expr,
format,
boost::match_default | boost::format_all );
}
ResultList split(const std::basic_string<CharT> &s,
const std::basic_string<CharT> &delimiter, bool keepEmptyParts = true)
{
ResultList slist;
// If delimiter.empty():
// pos = s.find(delimiter, start);
// pos will be 0.
if (delimiter.empty()) {
slist.push_back(s);
return slist;
}
typename std::basic_string<CharT>::size_type start = 0;
typename std::basic_string<CharT>::size_type pos;
std::basic_string<CharT> part;
if (delimiter.length() == 1) {
CharT ch = delimiter[0];
// Hope that:
// find(std::basic_string<CharT>, CharT ch)
// will be faster than:
// find(std::basic_string<CharT>, std::basic_string<CharT>)
while ((pos = s.find(ch, start)) != s.npos) { // Use strchr/wcschr instead?
part = s.substr(start, pos - start);
if (!part.empty() || keepEmptyParts)
slist.push_back(part);
start = pos + delimiter.length();
}
} else {
while ((pos = s.find(delimiter, start)) != s.npos) { // Use strstr/wcsstr instead?
part = s.substr(start, pos - start);
if (!part.empty() || keepEmptyParts)
slist.push_back(part);
start = pos + delimiter.length();
}
}
if (start != s.length() || keepEmptyParts)
slist.push_back(s.substr(start));
return slist;
}
std::basic_string < Elem, Traits > cell_encode(std::basic_string < Elem, Traits > Str, Elem Sep_, Elem Esc){
if(Str.find(Sep_) < Str.size() || Str.find(Esc) < Str.size()){
for(auto itr = Str.begin(); itr != Str.end(); ++itr){
if(*itr == Esc){
itr = Str.insert(++itr, Esc);
}
}
Str.insert(Str.begin(), Esc);
Str.push_back(Esc);
}
return std::move(Str);
}
std::vector<std::basic_string<T_Char>> split(
const std::basic_string<T_Char>& src,
const std::basic_string<T_Char>& delimit) {
std::vector<std::basic_string<T_Char>> array;
typedef typename std::basic_string<T_Char>::size_type size_type;
typedef typename std::basic_string<T_Char>::const_iterator const_iterator;
std::basic_string<T_Char> tmp;
if (src.empty())
return array;
if (delimit.empty()) {
array.reserve(array.size() + src.size());
for (const_iterator it = src.begin(); it != src.end(); ++it)
array.push_back(std::basic_string<T_Char>(1, *it));
return array;
}
size_type src_pos = 0;
while (src.begin() + src_pos != src.end()) {
size_type fnd_pos = src.find(delimit, src_pos);
if (fnd_pos == std::basic_string<T_Char>::npos) {
array.push_back(std::basic_string<T_Char>(src, src_pos));
break;
}
array.push_back(
std::basic_string<T_Char>(src, src_pos, fnd_pos - src_pos));
src_pos = fnd_pos + delimit.length();
}
return array;
}
void read_cmdline(int argc,
typename Ptree::char_type *argv[],
const std::basic_string<typename Ptree::char_type> &metachars,
Ptree &pt)
{
typedef typename Ptree::char_type Ch;
typedef std::basic_string<Ch> Str;
Ptree local;
// For all arguments
for (int i = 0; i < argc; ++i)
{
Str text = detail::trim<Ch>(argv[i]);
if (!text.empty())
if (metachars.find(text[0]) != Str::npos)
{
if (text.size() == 1)
{
Ptree &child = local.put(text, Str());
Str key;
if (child.size() < 10)
key.push_back(typename Ptree::char_type('0' + child.size()));
child.push_back(std::make_pair(key, Ptree(child.data())));
}
else if (text.size() == 2)
{
Ptree &child = local.put(text.substr(1, 1), Str());
Str key;
if (child.size() < 10)
key.push_back(typename Ptree::char_type('0' + child.size()));
child.push_back(std::make_pair(key, Ptree(child.data())));
}
else
{
Ptree &child = local.put(text.substr(1, 1), detail::trim<Ch>(text.substr(2, Str::npos)));
Str key;
if (child.size() < 10)
key.push_back(typename Ptree::char_type('0' + child.size()));
child.push_back(std::make_pair(key, Ptree(child.data())));
}
}
else
{
Ptree &child = local.put(Str(), detail::trim<Ch>(text));
Str key;
if (child.size() < 10)
key.push_back(typename Ptree::char_type('0' + child.size()));
child.push_back(std::make_pair(key, Ptree(child.data())));
}
}
// Swap local and pt
pt.swap(local);
}
template<typename char_t> std::basic_string<char_t> strip(const std::basic_string<char_t>& what, const std::basic_string<char_t> of_what)
{
std::basic_string<char_t> result;
for ( size_t i=0; i<what.length(); i++ )
{
if ( of_what.find(what[i]) == std::basic_string<char_t>::npos )
{
result.push_back(what[i]);
}
}
return result;
}
//@todo test
template<typename TARGET, typename charT, typename traits> std::list<TARGET>
stringToList( const std::basic_string<charT, traits> &source, charT separator )
{
std::list<TARGET> ret;
for (
size_t next = source.find_first_not_of( separator );
next != std::string::npos;
next = source.find_first_not_of( separator, next )
) {
const size_t start = next;
next = source.find( separator, start );
ret.push_back( boost::lexical_cast<TARGET>( source.substr( start, next - start ) ) );
}
return ret;
}
std::basic_string<CharT> to_str_gcc_workaround(std::basic_string<CharT> str)
{
std::locale loc;
std::numpunct<CharT> const& np = BOOST_USE_FACET(std::numpunct<CharT>, loc);
std::ctype<CharT> const& ct = BOOST_USE_FACET(std::ctype<CharT>, loc);
if(np.grouping().empty())
return str;
CharT prefix[3] = { ct.widen('-'), np.thousands_sep(), CharT() };
if(str.find(prefix) != 0)
return str;
prefix[1] = CharT();
str.replace(0, 2, prefix);
return str;
}
std::vector<std::basic_string<TCHAR>> Split (const std::basic_string<TCHAR> &inString,
const std::basic_string<TCHAR> &separator)
{
std::vector<std::basic_string<TCHAR>> returnVector;
std::basic_string<TCHAR>::size_type start = 0;
std::basic_string<TCHAR>::size_type end = 0;
while ((end=inString.find (separator, start)) != std::string::npos)
{
returnVector.push_back (inString.substr (start, end-start));
start = end+separator.size();
}
returnVector.push_back (inString.substr (start));
return returnVector;
}
template<typename char_t> bool ends_with(const std::basic_string<char_t>& what, const std::basic_string<char_t>& with)
{
return !what.empty() && (what.find(with) == what.length()-with.length());
}
CountT GetCount(const std::basic_string<CharT> & gram) const
{
std::map<std::basic_string<CharT>, CountT>::iterator iter = gram.find(gram);
if (iter == gram.end()) return 0;
else return iter->second;
}
size_t StringReplaceAllT(const std::basic_string<CharType> &find,
const std::basic_string<CharType> &replace,
std::basic_string<CharType> &output)
{
size_t find_length = find.size();
size_t replace_length = replace.size();
size_t offset = 0, endpos;
size_t target = 0, found_pos;
size_t replaced = 0;
CharType *data_ptr;
if (find.empty() || output.empty())
return 0;
/*
* to avoid extra memory reallocating,
* we use two passes to finish the task in the case that replace.size() is greater find.size()
*/
if (find_length < replace_length)
{
/* the first pass, count all available 'find' to be replaced */
for (;;)
{
offset = output.find(find, offset);
if (offset == std::basic_string<CharType>::npos)
break;
replaced++;
offset += find_length;
}
if (replaced == 0)
return 0;
size_t newsize = output.size() + replaced * (replace_length - find_length);
/* we apply for more memory to hold the content to be replaced */
endpos = newsize;
offset = newsize - output.size();
output.resize(newsize);
data_ptr = &output[0];
memmove((void*)(data_ptr + offset),
(void*)data_ptr,
(output.size() - offset) * sizeof(CharType));
}
else
{
endpos = output.size();
offset = 0;
data_ptr = const_cast<CharType *>(&output[0]);
}
/* the second pass, the replacement */
while (offset < endpos)
{
found_pos = output.find(find, offset);
if (found_pos != std::basic_string<CharType>::npos)
{
/* move the content between two targets */
if (target != found_pos)
memmove((void*)(data_ptr + target),
(void*)(data_ptr + offset),
(found_pos - offset) * sizeof(CharType));
target += found_pos - offset;
/* replace */
memcpy(data_ptr + target,
replace.data(),
replace_length * sizeof(CharType));
target += replace_length;
offset = find_length + found_pos;
replaced++;
}
else
{
/* ending work */
if (target != offset)
memcpy((void*)(data_ptr + target),
(void*)(data_ptr + offset),
(endpos - offset) * sizeof(CharType));
break;
}
}
if (replace_length < find_length)
output.resize(output.size() - replaced * (find_length - replace_length));
return replaced;
}
INT ReplaceHtmlEntities(std::basic_string<TCHAR>& htmlText)
{
INT replaces = 0;
size_t startPos = htmlText.find(L"&", 0);
while (startPos != std::wstring::npos)
{
size_t nextEndPos = htmlText.find(L";", startPos + 1);
if (nextEndPos == std::wstring::npos)
{
//=== Last &... wasn't an entity & we ve found EOF
break;
}
BOOL bEntityIsValid = TRUE;
size_t nextStartPos = htmlText.find(L"&", startPos + 1);
if (nextStartPos != std::wstring::npos)
{
if (nextStartPos < nextEndPos)
{
//=== We ve found &..&..; pattern which is illegal
bEntityIsValid = FALSE;
}
}
if (bEntityIsValid)
bEntityIsValid = ((nextEndPos - startPos) <= 8);
if (bEntityIsValid)
{
wchar_t repVal = 0;
LPCTSTR entityStart = &htmlText[startPos+1];
if (*entityStart/*htmlText[startPos+1]*/ == '#')
{
//=== It is a int entity
TCHAR intbf[10];
INT copyChars = (INT) min(nextEndPos - startPos - 2, 9);
_tcsncpy(intbf, &htmlText[startPos+2], copyChars);
intbf[copyChars] = 0;
INT val = _wtoi(intbf);
if (val > 0 && val < 0xFFFE)
repVal = wchar_t(val);
}
else
{
//=== It is a string entity
//=== Check the ISO88591 entities
for (int i = 0; i < sizeof(ISO88591_Entities)/sizeof(LPCTSTR); i++)
{
if (_tcsncmp(ISO88591_Entities[i], entityStart, _tcslen(ISO88591_Entities[i])) == 0)
{
repVal = (wchar_t) (i + 160);
break;
}
}
//=== Check the "other" mapped entities
if (repVal == 0)
{
for (int i = 0; i < sizeof(strentity2wchar_t)/sizeof(HtmlStringEntityMap); i++)
{
if (_tcsncmp(strentity2wchar_t[i].html, entityStart, _tcslen(strentity2wchar_t[i].html)) == 0)
{
repVal = strentity2wchar_t[i].res;
break;
}
}
}
}
if (repVal != 0)
{
htmlText.erase(startPos + 1, nextEndPos - startPos);
htmlText[startPos] = repVal;
replaces++;
}
nextStartPos = htmlText.find(L"&", startPos + 1);
}
startPos = nextStartPos;
}
//for (int i = 0; i < sizeof(html2text)/sizeof(HtmlEntityMap); i++)
// replaces += replace(htmlText, html2text[i].html, html2text[i].text);
return replaces;
}
std::basic_string<char_type> convert ( std::basic_string<char_type> & s )
{
return s.substr( 0, s.find( '\r' ) );
}
