| 1 | // |
|---|
| 2 | // Copyright (c) 2009-2011 Artyom Beilis (Tonkikh) |
|---|
| 3 | // Copyright (c) 2023 Alexander Grund |
|---|
| 4 | // |
|---|
| 5 | // Distributed under the Boost Software License, Version 1.0. |
|---|
| 6 | // https://www.boost.org/LICENSE_1_0.txt |
|---|
| 7 | |
|---|
| 8 | // |
|---|
| 9 | // ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! |
|---|
| 10 | // |
|---|
| 11 | // BIG FAT WARNING FOR Microsoft Visual Studio Users |
|---|
| 12 | // |
|---|
| 13 | // YOU NEED TO CONVERT THIS SOURCE FILE ENCODING TO UTF-8 WITH BOM ENCODING. |
|---|
| 14 | // |
|---|
| 15 | // Unfortunately MSVC understands that the source code is encoded as |
|---|
| 16 | // UTF-8 only if you add useless BOM in the beginning. |
|---|
| 17 | // |
|---|
| 18 | // So, before you compile "wide" examples with MSVC, please convert them to text |
|---|
| 19 | // files with BOM. There are two very simple ways to do it: |
|---|
| 20 | // |
|---|
| 21 | // 1. Open file with Notepad and save it from there. It would convert |
|---|
| 22 | // it to file with BOM. |
|---|
| 23 | // 2. In Visual Studio go File->Advances Save Options... and select |
|---|
| 24 | // Unicode (UTF-8 with signature) Codepage 65001 |
|---|
| 25 | // |
|---|
| 26 | // Note: once converted to UTF-8 with BOM, this source code would not |
|---|
| 27 | // compile with other compilers, because no-one uses BOM with UTF-8 today |
|---|
| 28 | // because it is absolutely meaningless in context of UTF-8. |
|---|
| 29 | // |
|---|
| 30 | // ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! |
|---|
| 31 | // |
|---|
| 32 | #include <boost/locale.hpp> |
|---|
| 33 | #include <boost/algorithm/string/case_conv.hpp> |
|---|
| 34 | #include <iostream> |
|---|
| 35 | |
|---|
| 36 | int main() |
|---|
| 37 | { |
|---|
| 38 | if(boost::locale::localization_backend_manager::global().get_all_backends().at(n: 0) != "icu" ) |
|---|
| 39 | std::wcout << L"Need ICU support for this example!\nConversion below will likely be wrong!\n" ; |
|---|
| 40 | |
|---|
| 41 | // Create system default locale |
|---|
| 42 | boost::locale::generator gen; |
|---|
| 43 | std::locale loc = gen("" ); |
|---|
| 44 | std::locale::global(loc: loc); |
|---|
| 45 | std::wcout.imbue(loc: loc); |
|---|
| 46 | |
|---|
| 47 | // This is needed to prevent the C stdio library from |
|---|
| 48 | // converting strings to narrow on some platforms |
|---|
| 49 | std::ios_base::sync_with_stdio(sync: false); |
|---|
| 50 | |
|---|
| 51 | std::wcout << L"Correct case conversion can't be done by simple, character by character conversion\n" ; |
|---|
| 52 | std::wcout << L"because case conversion is context sensitive and not a 1-to-1 conversion.\n" ; |
|---|
| 53 | std::wcout << L"For example:\n" ; |
|---|
| 54 | const std::wstring gruessen(L"grüßen" ); |
|---|
| 55 | std::wcout << L" German " << gruessen << " would be incorrectly converted to " << boost::to_upper_copy(Input: gruessen); |
|---|
| 56 | std::wcout << ", while Boost.Locale converts it to " << boost::locale::to_upper(str: gruessen) << std::endl |
|---|
| 57 | << L" where ß is replaced with SS.\n" ; |
|---|
| 58 | const std::wstring greek(L"ὈΔΥΣΣΕΎΣ" ); |
|---|
| 59 | std::wcout << L" Greek " << greek << " would be incorrectly converted to " << boost::to_lower_copy(Input: greek); |
|---|
| 60 | std::wcout << ", while Boost.Locale correctly converts it to " << boost::locale::to_lower(str: greek) << std::endl |
|---|
| 61 | << L" where Σ is converted to σ or to ς, according to position in the word.\n" ; |
|---|
| 62 | std::wcout |
|---|
| 63 | << L"Such type of conversion just can't be done using std::toupper/boost::to_upper* that work on character " |
|---|
| 64 | L"by character base.\n" |
|---|
| 65 | L"Also std::toupper is not fully applicable when working with variable character length like UTF-8 or UTF-16\n" |
|---|
| 66 | L"limiting the correct behavior to BMP or ASCII only\n" ; |
|---|
| 67 | } |
|---|
| 68 | |
|---|
| 69 | // boostinspect:noascii |
|---|
| 70 | |
|---|
