// // Copyright (c) 2009-2011 Artyom Beilis (Tonkikh) // // Distributed under the Boost Software License, Version 1.0. (See // accompanying file LICENSE_1_0.txt or copy at // http://www.boost.org/LICENSE_1_0.txt) // #ifndef BOOST_LOCALE_BOUNDARY_SEGMENT_HPP_INCLUDED #define BOOST_LOCALE_BOUNDARY_SEGMENT_HPP_INCLUDED #include #ifdef BOOST_MSVC # pragma warning(push) # pragma warning(disable : 4275 4251 4231 4660) #endif #include #include #include #include namespace boost { namespace locale { namespace boundary { /// \cond INTERNAL namespace details { template int compare_text(LeftIterator l_begin,LeftIterator l_end,RightIterator r_begin,RightIterator r_end) { typedef LeftIterator left_iterator; typedef RightIterator right_iterator; typedef typename std::iterator_traits::value_type char_type; typedef std::char_traits traits; while(l_begin!=l_end && r_begin!=r_end) { char_type lchar = *l_begin++; char_type rchar = *r_begin++; if(traits::eq(lchar,rchar)) continue; if(traits::lt(lchar,rchar)) return -1; else return 1; } if(l_begin==l_end && r_begin==r_end) return 0; if(l_begin==l_end) return -1; else return 1; } template int compare_text(Left const &l,Right const &r) { return compare_text(l.begin(),l.end(),r.begin(),r.end()); } template int compare_string(Left const &l,Char const *begin) { Char const *end = begin; while(*end!=0) end++; return compare_text(l.begin(),l.end(),begin,end); } template int compare_string(Char const *begin,Right const &r) { Char const *end = begin; while(*end!=0) end++; return compare_text(begin,end,r.begin(),r.end()); } } /// \endcond /// /// \addtogroup boundary /// @{ /// /// \brief a segment object that represents a pair of two iterators that define the range where /// this segment exits and a rule that defines it. /// /// This type of object is dereferenced by the iterators of segment_index. Using a rule() member function /// you can get a specific rule this segment was selected with. For example, when you use /// word boundary analysis, you can check if the specific word contains Kana letters by checking (rule() & \ref word_kana)!=0 /// For a sentence analysis you can check if the sentence is selected because a sentence terminator is found (\ref sentence_term) or /// there is a line break (\ref sentence_sep). /// /// This object can be automatically converted to std::basic_string with the same type of character. It is also /// valid range that has begin() and end() member functions returning iterators on the location of the segment. /// /// \see /// /// - \ref segment_index /// - \ref boundary_point /// - \ref boundary_point_index /// template class segment : public std::pair { public: /// /// The type of the underlying character /// typedef typename std::iterator_traits::value_type char_type; /// /// The type of the string it is converted to /// typedef std::basic_string string_type; /// /// The value that iterators return - the character itself /// typedef char_type value_type; /// /// The iterator that allows to iterate the range /// typedef IteratorType iterator; /// /// The iterator that allows to iterate the range /// typedef IteratorType const_iterator; /// /// The type that represent a difference between two iterators /// typedef typename std::iterator_traits::difference_type difference_type; /// /// Default constructor /// segment() {} /// /// Create a segment using two iterators and a rule that represents this point /// segment(iterator b,iterator e,rule_type r) : std::pair(b,e), rule_(r) { } /// /// Set the start of the range /// void begin(iterator const &v) { this->first = v; } /// /// Set the end of the range /// void end(iterator const &v) { this->second = v; } /// /// Get the start of the range /// IteratorType begin() const { return this->first; } /// /// Set the end of the range /// IteratorType end() const { return this->second; } /// /// Convert the range to a string automatically /// template operator std::basic_string ()const { return std::basic_string(this->first, this->second); } /// /// Create a string from the range explicitly /// string_type str() const { return string_type(begin(),end()); } /// /// Get the length of the text chunk /// size_t length() const { return std::distance(begin(),end()); } /// /// Check if the segment is empty /// bool empty() const { return begin() == end(); } /// /// Get the rule that is used for selection of this segment. /// rule_type rule() const { return rule_; } /// /// Set a rule that is used for segment selection /// void rule(rule_type r) { rule_ = r; } // make sure we override std::pair's operator== /// Compare two segments bool operator==(segment const &other) { return details::compare_text(*this,other) == 0; } /// Compare two segments bool operator!=(segment const &other) { return details::compare_text(*this,other) != 0; } private: rule_type rule_; }; /// Compare two segments template bool operator==(segment const &l,segment const &r) { return details::compare_text(l,r) == 0; } /// Compare two segments template bool operator!=(segment const &l,segment const &r) { return details::compare_text(l,r) != 0; } /// Compare two segments template bool operator<(segment const &l,segment const &r) { return details::compare_text(l,r) < 0; } /// Compare two segments template bool operator<=(segment const &l,segment const &r) { return details::compare_text(l,r) <= 0; } /// Compare two segments template bool operator>(segment const &l,segment const &r) { return details::compare_text(l,r) > 0; } /// Compare two segments template bool operator>=(segment const &l,segment const &r) { return details::compare_text(l,r) >= 0; } /// Compare string and segment template bool operator==(std::basic_string const &l,segment const &r) { return details::compare_text(l,r) == 0; } /// Compare string and segment template bool operator!=(std::basic_string const &l,segment const &r) { return details::compare_text(l,r) != 0; } /// Compare string and segment template bool operator<(std::basic_string const &l,segment const &r) { return details::compare_text(l,r) < 0; } /// Compare string and segment template bool operator<=(std::basic_string const &l,segment const &r) { return details::compare_text(l,r) <= 0; } /// Compare string and segment template bool operator>(std::basic_string const &l,segment const &r) { return details::compare_text(l,r) > 0; } /// Compare string and segment template bool operator>=(std::basic_string const &l,segment const &r) { return details::compare_text(l,r) >= 0; } /// Compare string and segment template bool operator==(segment const &l,std::basic_string const &r) { return details::compare_text(l,r) == 0; } /// Compare string and segment template bool operator!=(segment const &l,std::basic_string const &r) { return details::compare_text(l,r) != 0; } /// Compare string and segment template bool operator<(segment const &l,std::basic_string const &r) { return details::compare_text(l,r) < 0; } /// Compare string and segment template bool operator<=(segment const &l,std::basic_string const &r) { return details::compare_text(l,r) <= 0; } /// Compare string and segment template bool operator>(segment const &l,std::basic_string const &r) { return details::compare_text(l,r) > 0; } /// Compare string and segment template bool operator>=(segment const &l,std::basic_string const &r) { return details::compare_text(l,r) >= 0; } /// Compare C string and segment template bool operator==(CharType const *l,segment const &r) { return details::compare_string(l,r) == 0; } /// Compare C string and segment template bool operator!=(CharType const *l,segment const &r) { return details::compare_string(l,r) != 0; } /// Compare C string and segment template bool operator<(CharType const *l,segment const &r) { return details::compare_string(l,r) < 0; } /// Compare C string and segment template bool operator<=(CharType const *l,segment const &r) { return details::compare_string(l,r) <= 0; } /// Compare C string and segment template bool operator>(CharType const *l,segment const &r) { return details::compare_string(l,r) > 0; } /// Compare C string and segment template bool operator>=(CharType const *l,segment const &r) { return details::compare_string(l,r) >= 0; } /// Compare C string and segment template bool operator==(segment const &l,CharType const *r) { return details::compare_string(l,r) == 0; } /// Compare C string and segment template bool operator!=(segment const &l,CharType const *r) { return details::compare_string(l,r) != 0; } /// Compare C string and segment template bool operator<(segment const &l,CharType const *r) { return details::compare_string(l,r) < 0; } /// Compare C string and segment template bool operator<=(segment const &l,CharType const *r) { return details::compare_string(l,r) <= 0; } /// Compare C string and segment template bool operator>(segment const &l,CharType const *r) { return details::compare_string(l,r) > 0; } /// Compare C string and segment template bool operator>=(segment const &l,CharType const *r) { return details::compare_string(l,r) >= 0; } typedef segment ssegment; ///< convenience typedef typedef segment wssegment; ///< convenience typedef #ifdef BOOST_HAS_CHAR16_T typedef segment u16ssegment;///< convenience typedef #endif #ifdef BOOST_HAS_CHAR32_T typedef segment u32ssegment;///< convenience typedef #endif typedef segment csegment; ///< convenience typedef typedef segment wcsegment; ///< convenience typedef #ifdef BOOST_HAS_CHAR16_T typedef segment u16csegment; ///< convenience typedef #endif #ifdef BOOST_HAS_CHAR32_T typedef segment u32csegment; ///< convenience typedef #endif /// /// Write the segment to the stream character by character /// template std::basic_ostream &operator<<( std::basic_ostream &out, segment const &tok) { for(Iterator p=tok.begin(),e=tok.end();p!=e;++p) out << *p; return out; } /// @} } // boundary } // locale } // boost #ifdef BOOST_MSVC #pragma warning(pop) #endif #endif // vim: tabstop=4 expandtab shiftwidth=4 softtabstop=4