Boost.Locale
boundary.cpp

Example of using segment_index

//
// Copyright (c) 2009-2011 Artyom Beilis (Tonkikh)
//
// Distributed under the Boost Software License, Version 1.0.
// https://www.boost.org/LICENSE_1_0.txt
#include <boost/locale.hpp>
#include <ctime>
#include <iostream>
int main()
{
using namespace boost::locale;
generator gen;
// Make system default locale global
std::locale loc = gen("");
// We need the boundary facet, currently only available via ICU
if(!std::has_facet<boundary::boundary_indexing<char>>(loc)) {
std::cout << "boundary detection not implemented in this environment\n";
return 0;
}
std::locale::global(loc);
std::cout.imbue(loc);
std::string text = "Hello World! あにま! Linux2.6 and Windows7 is word and number. שָלוֹם עוֹלָם!";
std::cout << text << std::endl;
boundary::ssegment_index index(boundary::word, text.begin(), text.end());
for(boundary::ssegment_index::iterator p = index.begin(), e = index.end(); p != e; ++p) {
std::cout << "Part [" << *p << "] has ";
if(p->rule() & boundary::word_number)
std::cout << "number(s) ";
if(p->rule() & boundary::word_letter)
std::cout << "letter(s) ";
if(p->rule() & boundary::word_kana)
std::cout << "kana character(s) ";
if(p->rule() & boundary::word_ideo)
std::cout << "ideographic character(s) ";
if(p->rule() & boundary::word_none)
std::cout << "no word characters";
std::cout << std::endl;
}
index.map(boundary::character, text.begin(), text.end());
for(const boundary::ssegment& p : index)
std::cout << "|" << p;
std::cout << "|\n\n";
index.map(boundary::line, text.begin(), text.end());
for(const boundary::ssegment& p : index)
std::cout << "|" << p;
std::cout << "|\n\n";
index.map(boundary::sentence, text.begin(), text.end());
for(const boundary::ssegment& p : index)
std::cout << "|" << p;
std::cout << "|\n\n";
}
// boostinspect:noascii
This facet generates an index for boundary analysis of a given text.
Definition: facets.hpp:59
This class holds an index of segments in the text range and allows to iterate over them.
Definition: index.hpp:475
a segment object that represents a pair of two iterators that define the range where this segment exi...
Definition: segment.hpp:90
the major class used for locale generation
Definition: generator.hpp:101
void map(boundary_type type, base_iterator begin, base_iterator end, const std::locale &loc=std::locale())
Definition: index.hpp:556
iterator end() const
Definition: index.hpp:578
iterator begin() const
Definition: index.hpp:568
unspecified_iterator_type iterator
Definition: index.hpp:493
This is the main namespace that encloses all localization classes.
Definition: boundary_point.hpp:13