| 1 | // |
| 2 | // Copyright (c) 2009-2011 Artyom Beilis (Tonkikh) |
| 3 | // |
| 4 | // Distributed under the Boost Software License, Version 1.0. |
| 5 | // https://www.boost.org/LICENSE_1_0.txt |
| 6 | |
| 7 | // |
| 8 | // ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! |
| 9 | // |
| 10 | // BIG FAT WARNING FOR Microsoft Visual Studio Users |
| 11 | // |
| 12 | // YOU NEED TO CONVERT THIS SOURCE FILE ENCODING TO UTF-8 WITH BOM ENCODING. |
| 13 | // |
| 14 | // Unfortunately MSVC understands that the source code is encoded as |
| 15 | // UTF-8 only if you add useless BOM in the beginning. |
| 16 | // |
| 17 | // So, before you compile "wide" examples with MSVC, please convert them to text |
| 18 | // files with BOM. There are two very simple ways to do it: |
| 19 | // |
| 20 | // 1. Open file with Notepad and save it from there. It would convert |
| 21 | // it to file with BOM. |
| 22 | // 2. In Visual Studio go File->Advances Save Options... and select |
| 23 | // Unicode (UTF-8 with signature) Codepage 65001 |
| 24 | // |
| 25 | // Note: once converted to UTF-8 with BOM, this source code would not |
| 26 | // compile with other compilers, because no-one uses BOM with UTF-8 today |
| 27 | // because it is absolutely meaningless in context of UTF-8. |
| 28 | // |
| 29 | // ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! |
| 30 | // |
| 31 | #include <boost/locale.hpp> |
| 32 | #include <ctime> |
| 33 | #include <iostream> |
| 34 | |
| 35 | int main() |
| 36 | { |
| 37 | using namespace boost::locale; |
| 38 | |
| 39 | // Create system default locale |
| 40 | generator gen; |
| 41 | std::locale loc = gen("" ); |
| 42 | // We need the boundary facet, currently only available via ICU |
| 43 | if(!std::has_facet<boundary::boundary_indexing<wchar_t>>(loc: loc)) { |
| 44 | std::cout << "boundary detection not implemented in this environment\n" ; |
| 45 | return 0; |
| 46 | } |
| 47 | std::locale::global(loc: loc); |
| 48 | std::wcout.imbue(loc: loc); |
| 49 | |
| 50 | // This is needed to prevent the C stdio library from |
| 51 | // converting strings to narrow on some platforms |
| 52 | std::ios_base::sync_with_stdio(sync: false); |
| 53 | |
| 54 | std::wstring text = L"Hello World! あにま! Linux2.6 and Windows7 is word and number. שָלוֹם עוֹלָם!" ; |
| 55 | |
| 56 | std::wcout << text << std::endl; |
| 57 | |
| 58 | boundary::wssegment_index index(boundary::word, text.begin(), text.end()); |
| 59 | |
| 60 | for(boundary::wssegment_index::iterator p = index.begin(), e = index.end(); p != e; ++p) { |
| 61 | std::wcout << L"Part [" << *p << L"] has " ; |
| 62 | if(p->rule() & boundary::word_number) |
| 63 | std::wcout << L"number(s) " ; |
| 64 | if(p->rule() & boundary::word_letter) |
| 65 | std::wcout << L"letter(s) " ; |
| 66 | if(p->rule() & boundary::word_kana) |
| 67 | std::wcout << L"kana character(s) " ; |
| 68 | if(p->rule() & boundary::word_ideo) |
| 69 | std::wcout << L"ideographic character(s) " ; |
| 70 | if(p->rule() & boundary::word_none) |
| 71 | std::wcout << L"no word characters" ; |
| 72 | std::wcout << std::endl; |
| 73 | } |
| 74 | |
| 75 | index.map(type: boundary::character, begin: text.begin(), end: text.end()); |
| 76 | |
| 77 | for(const boundary::wssegment& p : index) |
| 78 | std::wcout << L"|" << p; |
| 79 | std::wcout << L"|\n\n" ; |
| 80 | |
| 81 | index.map(type: boundary::line, begin: text.begin(), end: text.end()); |
| 82 | |
| 83 | for(const boundary::wssegment& p : index) |
| 84 | std::wcout << L"|" << p; |
| 85 | std::wcout << L"|\n\n" ; |
| 86 | |
| 87 | index.map(type: boundary::sentence, begin: text.begin(), end: text.end()); |
| 88 | |
| 89 | for(const boundary::wssegment& p : index) |
| 90 | std::wcout << L"|" << p; |
| 91 | std::wcout << "|\n\n" ; |
| 92 | } |
| 93 | |
| 94 | // boostinspect:noascii |
| 95 | |