icu.hpp

来自「support vector clustering for vc++」· HPP 代码 · 共 1,018 行 · 第 1/3 页

#endif
inline bool u32regex_search(const UnicodeString& s, 
                        match_results<const UChar*>& m, 
                        const u32regex& e, 
                        match_flag_type flags = match_default)
{
   return re_detail::do_regex_search(s.getBuffer(), s.getBuffer() + s.length(), m, e, flags, s.getBuffer(), static_cast<mpl::int_<sizeof(wchar_t)> const*>(0));
}
template <class BidiIterator>
inline bool u32regex_search(BidiIterator first, BidiIterator last, 
                 const u32regex& e, 
                 match_flag_type flags = match_default)
{
   match_results<BidiIterator> m;
   return re_detail::do_regex_search(first, last, m, e, flags, first, static_cast<mpl::int_<sizeof(*first)> const*>(0));
}
inline bool u32regex_search(const UChar* p, 
                 const u32regex& e, 
                 match_flag_type flags = match_default)
{
   match_results<const UChar*> m;
   return re_detail::do_regex_search(p, p+u_strlen(p), m, e, flags, p, static_cast<mpl::int_<2> const*>(0));
}
#if !defined(U_WCHAR_IS_UTF16) && !defined(BOOST_NO_WREGEX)
inline bool u32regex_search(const wchar_t* p, 
                 const u32regex& e, 
                 match_flag_type flags = match_default)
{
   match_results<const wchar_t*> m;
   return re_detail::do_regex_search(p, p+std::wcslen(p), m, e, flags, p, static_cast<mpl::int_<sizeof(wchar_t)> const*>(0));
}
#endif
inline bool u32regex_search(const char* p, 
                 const u32regex& e, 
                 match_flag_type flags = match_default)
{
   match_results<const char*> m;
   return re_detail::do_regex_search(p, p+std::strlen(p), m, e, flags, p, static_cast<mpl::int_<1> const*>(0));
}
inline bool u32regex_search(const unsigned char* p, 
                 const u32regex& e, 
                 match_flag_type flags = match_default)
{
   match_results<const unsigned char*> m;
   return re_detail::do_regex_search(p, p+std::strlen((const char*)p), m, e, flags, p, static_cast<mpl::int_<1> const*>(0));
}
inline bool u32regex_search(const std::string& s, 
                        const u32regex& e, 
                        match_flag_type flags = match_default)
{
   match_results<std::string::const_iterator> m;
   return re_detail::do_regex_search(s.begin(), s.end(), m, e, flags, s.begin(), static_cast<mpl::int_<1> const*>(0));
}
#ifndef BOOST_NO_STD_WSTRING
inline bool u32regex_search(const std::wstring& s, 
                        const u32regex& e, 
                        match_flag_type flags = match_default)
{
   match_results<std::wstring::const_iterator> m;
   return re_detail::do_regex_search(s.begin(), s.end(), m, e, flags, s.begin(), static_cast<mpl::int_<sizeof(wchar_t)> const*>(0));
}
#endif
inline bool u32regex_search(const UnicodeString& s, 
                        const u32regex& e, 
                        match_flag_type flags = match_default)
{
   match_results<const UChar*> m;
   return re_detail::do_regex_search(s.getBuffer(), s.getBuffer() + s.length(), m, e, flags, s.getBuffer(), static_cast<mpl::int_<sizeof(wchar_t)> const*>(0));
}

//
// overloads for regex_replace with utf-8 and utf-16 data types:
//
namespace re_detail{
template <class I>
inline std::pair< boost::u8_to_u32_iterator<I>, boost::u8_to_u32_iterator<I> >
   make_utf32_seq(I i, I j, mpl::int_<1> const*)
{
   return std::pair< boost::u8_to_u32_iterator<I>, boost::u8_to_u32_iterator<I> >(boost::u8_to_u32_iterator<I>(i), boost::u8_to_u32_iterator<I>(j));
}
template <class I>
inline std::pair< boost::u16_to_u32_iterator<I>, boost::u16_to_u32_iterator<I> >
   make_utf32_seq(I i, I j, mpl::int_<2> const*)
{
   return std::pair< boost::u16_to_u32_iterator<I>, boost::u16_to_u32_iterator<I> >(boost::u16_to_u32_iterator<I>(i), boost::u16_to_u32_iterator<I>(j));
}
template <class I>
inline std::pair< I, I >
   make_utf32_seq(I i, I j, mpl::int_<4> const*)
{
   return std::pair< I, I >(i, j);
}
template <class charT>
inline std::pair< boost::u8_to_u32_iterator<const charT*>, boost::u8_to_u32_iterator<const charT*> >
   make_utf32_seq(const charT* p, mpl::int_<1> const*)
{
   return std::pair< boost::u8_to_u32_iterator<const charT*>, boost::u8_to_u32_iterator<const charT*> >(boost::u8_to_u32_iterator<const charT*>(p), boost::u8_to_u32_iterator<const charT*>(p+std::strlen((const char*)p)));
}
template <class charT>
inline std::pair< boost::u16_to_u32_iterator<const charT*>, boost::u16_to_u32_iterator<const charT*> >
   make_utf32_seq(const charT* p, mpl::int_<2> const*)
{
   return std::pair< boost::u16_to_u32_iterator<const charT*>, boost::u16_to_u32_iterator<const charT*> >(boost::u16_to_u32_iterator<const charT*>(p), boost::u16_to_u32_iterator<const charT*>(p+u_strlen((const UChar*)p)));
}
template <class charT>
inline std::pair< const charT*, const charT* >
   make_utf32_seq(const charT* p, mpl::int_<4> const*)
{
   return std::pair< const charT*, const charT* >(p, p+icu_regex_traits::length((UChar32 const*)p));
}
template <class OutputIterator>
inline OutputIterator make_utf32_out(OutputIterator o, mpl::int_<4> const*)
{
   return o;
}
template <class OutputIterator>
inline utf16_output_iterator<OutputIterator> make_utf32_out(OutputIterator o, mpl::int_<2> const*)
{
   return o;
}
template <class OutputIterator>
inline utf8_output_iterator<OutputIterator> make_utf32_out(OutputIterator o, mpl::int_<1> const*)
{
   return o;
}

template <class OutputIterator, class I1, class I2>
OutputIterator do_regex_replace(OutputIterator out,
                                 std::pair<I1, I1> const& in,
                                 const u32regex& e, 
                                 const std::pair<I2, I2>& fmt, 
                                 match_flag_type flags
                                 )
{
   // unfortunately we have to copy the format string in order to pass in onward:
   std::vector<UChar32> f;
#ifndef BOOST_NO_TEMPLATED_ITERATOR_CONSTRUCTORS
   f.assign(fmt.first, fmt.second);
#else
   f.clear();
   I2 pos = fmt.first;
   while(pos != fmt.second)
      f.push_back(*pos++);
#endif
   
   regex_iterator<I1, UChar32, icu_regex_traits> i(in.first, in.second, e, flags);
   regex_iterator<I1, UChar32, icu_regex_traits> j;
   if(i == j)
   {
      if(!(flags & regex_constants::format_no_copy))
         out = re_detail::copy(in.first, in.second, out);
   }
   else
   {
      I1 last_m = in.first;
      while(i != j)
      {
         if(!(flags & regex_constants::format_no_copy))
            out = re_detail::copy(i->prefix().first, i->prefix().second, out); 
         if(f.size())
            out = ::boost::re_detail::regex_format_imp(out, *i, &*f.begin(), &*f.begin() + f.size(), flags, e.get_traits());
         else
            out = ::boost::re_detail::regex_format_imp(out, *i, static_cast<UChar32 const*>(0), static_cast<UChar32 const*>(0), flags, e.get_traits());
         last_m = (*i)[0].second;
         if(flags & regex_constants::format_first_only)
            break;
         ++i;
      }
      if(!(flags & regex_constants::format_no_copy))
         out = re_detail::copy(last_m, in.second, out);
   }
   return out;
}
template <class BaseIterator>
inline const BaseIterator& extract_output_base(const BaseIterator& b)
{
   return b;
}
template <class BaseIterator>
inline BaseIterator extract_output_base(const utf8_output_iterator<BaseIterator>& b)
{
   return b.base();
}
template <class BaseIterator>
inline BaseIterator extract_output_base(const utf16_output_iterator<BaseIterator>& b)
{
   return b.base();
}
}  // re_detail

template <class OutputIterator, class BidirectionalIterator, class charT>
inline OutputIterator u32regex_replace(OutputIterator out,
                         BidirectionalIterator first,
                         BidirectionalIterator last,
                         const u32regex& e, 
                         const charT* fmt, 
                         match_flag_type flags = match_default)
{
   return re_detail::extract_output_base
#if BOOST_WORKAROUND(BOOST_MSVC, <= 1300)
   <OutputIterator>
#endif
    (
      re_detail::do_regex_replace(
         re_detail::make_utf32_out(out, static_cast<mpl::int_<sizeof(*first)> const*>(0)),
         re_detail::make_utf32_seq(first, last, static_cast<mpl::int_<sizeof(*first)> const*>(0)),
         e,
         re_detail::make_utf32_seq(fmt, static_cast<mpl::int_<sizeof(*fmt)> const*>(0)),
         flags)
      );
}

template <class OutputIterator, class Iterator, class charT>
inline OutputIterator u32regex_replace(OutputIterator out,
                         Iterator first,
                         Iterator last,
                         const u32regex& e, 
                         const std::basic_string<charT>& fmt,
                         match_flag_type flags = match_default)
{
   return re_detail::extract_output_base
#if BOOST_WORKAROUND(BOOST_MSVC, <= 1300)
   <OutputIterator>
#endif
    (
      re_detail::do_regex_replace(
         re_detail::make_utf32_out(out, static_cast<mpl::int_<sizeof(*first)> const*>(0)),
         re_detail::make_utf32_seq(first, last, static_cast<mpl::int_<sizeof(*first)> const*>(0)),
         e,
         re_detail::make_utf32_seq(fmt.begin(), fmt.end(), static_cast<mpl::int_<sizeof(charT)> const*>(0)),
         flags)
      );
}

template <class OutputIterator, class Iterator>
inline OutputIterator u32regex_replace(OutputIterator out,
                         Iterator first,
                         Iterator last,
                         const u32regex& e, 
                         const UnicodeString& fmt,
                         match_flag_type flags = match_default)
{
   return re_detail::extract_output_base
#if BOOST_WORKAROUND(BOOST_MSVC, <= 1300)
   <OutputIterator>
#endif
   (
      re_detail::do_regex_replace(
         re_detail::make_utf32_out(out, static_cast<mpl::int_<sizeof(*first)> const*>(0)),
         re_detail::make_utf32_seq(first, last, static_cast<mpl::int_<sizeof(*first)> const*>(0)),
         e,
         re_detail::make_utf32_seq(fmt.getBuffer(), fmt.getBuffer() + fmt.length(), static_cast<mpl::int_<2> const*>(0)),
         flags)
      );
}

template <class charT>
std::basic_string<charT> u32regex_replace(const std::basic_string<charT>& s,
                         const u32regex& e, 
                         const charT* fmt,
                         match_flag_type flags = match_default)
{
   std::basic_string<charT> result;
   re_detail::string_out_iterator<std::basic_string<charT> > i(result);
   u32regex_replace(i, s.begin(), s.end(), e, fmt, flags);
   return result;
}

template <class charT>
std::basic_string<charT> u32regex_replace(const std::basic_string<charT>& s,
                         const u32regex& e, 
                         const std::basic_string<charT>& fmt,
                         match_flag_type flags = match_default)
{
   std::basic_string<charT> result;
   re_detail::string_out_iterator<std::basic_string<charT> > i(result);
   u32regex_replace(i, s.begin(), s.end(), e, fmt.c_str(), flags);
   return result;
}

namespace re_detail{

class unicode_string_out_iterator
{
   UnicodeString* out;
public:
   unicode_string_out_iterator(UnicodeString& s) : out(&s) {}
   unicode_string_out_iterator& operator++() { return *this; }
   unicode_string_out_iterator& operator++(int) { return *this; }
   unicode_string_out_iterator& operator*() { return *this; }
   unicode_string_out_iterator& operator=(UChar v) 
   { 
      *out += v; 
      return *this; 
   }
   typedef std::ptrdiff_t difference_type;
   typedef UChar value_type;
   typedef value_type* pointer;
   typedef value_type& reference;
   typedef std::output_iterator_tag iterator_category;
};

}

inline UnicodeString u32regex_replace(const UnicodeString& s,
                         const u32regex& e, 
                         const UChar* fmt,
                         match_flag_type flags = match_default)
{
   UnicodeString result;
   re_detail::unicode_string_out_iterator i(result);
   u32regex_replace(i, s.getBuffer(), s.getBuffer()+s.length(), e, fmt, flags);
   return result;
}

inline UnicodeString u32regex_replace(const UnicodeString& s,
                         const u32regex& e, 
                         const UnicodeString& fmt,
                         match_flag_type flags = match_default)
{
   UnicodeString result;
   re_detail::unicode_string_out_iterator i(result);
   re_detail::do_regex_replace(
         re_detail::make_utf32_out(i, static_cast<mpl::int_<2> const*>(0)),
         re_detail::make_utf32_seq(s.getBuffer(), s.getBuffer()+s.length(), static_cast<mpl::int_<2> const*>(0)),
         e,
         re_detail::make_utf32_seq(fmt.getBuffer(), fmt.getBuffer() + fmt.length(), static_cast<mpl::int_<2> const*>(0)),
         flags);
   return result;
}

} // namespace boost.

#include <boost/regex/v4/u32regex_iterator.hpp>
#include <boost/regex/v4/u32regex_token_iterator.hpp>

#endif