1// std::ctype implementation details, GNU version -*- C++ -*- 2 3// Copyright (C) 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010 4// Free Software Foundation, Inc. 5// 6// This file is part of the GNU ISO C++ Library. This library is free 7// software; you can redistribute it and/or modify it under the 8// terms of the GNU General Public License as published by the 9// Free Software Foundation; either version 3, or (at your option) 10// any later version. 11 12// This library is distributed in the hope that it will be useful, 13// but WITHOUT ANY WARRANTY; without even the implied warranty of 14// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 15// GNU General Public License for more details. 16 17// Under Section 7 of GPL version 3, you are granted additional 18// permissions described in the GCC Runtime Library Exception, version 19// 3.1, as published by the Free Software Foundation. 20 21// You should have received a copy of the GNU General Public License and 22// a copy of the GCC Runtime Library Exception along with this program; 23// see the files COPYING3 and COPYING.RUNTIME respectively. If not, see 24// <http://www.gnu.org/licenses/>. 25 26// 27// ISO C++ 14882: 22.2.1.1.2 ctype virtual functions. 28// 29 30// Written by Benjamin Kosnik <bkoz@redhat.com> 31 32#include <locale> 33#include <cstdio> 34#include <bits/c++locale_internal.h> 35 36_GLIBCXX_BEGIN_NAMESPACE(std) 37 38 // NB: The other ctype<char> specializations are in src/locale.cc and 39 // various /config/os/* files. 40 ctype_byname<char>::ctype_byname(const char* __s, size_t __refs) 41 : ctype<char>(0, false, __refs) 42 { 43 if (std::strcmp(__s, "C") != 0 && std::strcmp(__s, "POSIX") != 0) 44 { 45 this->_S_destroy_c_locale(this->_M_c_locale_ctype); 46 this->_S_create_c_locale(this->_M_c_locale_ctype, __s); 47 this->_M_toupper = this->_M_c_locale_ctype->__ctype_toupper; 48 this->_M_tolower = this->_M_c_locale_ctype->__ctype_tolower; 49 this->_M_table = this->_M_c_locale_ctype->__ctype_b; 50 } 51 } 52 53 ctype_byname<char>::~ctype_byname() 54 { } 55 56#ifdef _GLIBCXX_USE_WCHAR_T 57 ctype<wchar_t>::__wmask_type 58 ctype<wchar_t>::_M_convert_to_wmask(const mask __m) const throw() 59 { 60 __wmask_type __ret; 61 switch (__m) 62 { 63 case space: 64 __ret = __wctype_l("space", _M_c_locale_ctype); 65 break; 66 case print: 67 __ret = __wctype_l("print", _M_c_locale_ctype); 68 break; 69 case cntrl: 70 __ret = __wctype_l("cntrl", _M_c_locale_ctype); 71 break; 72 case upper: 73 __ret = __wctype_l("upper", _M_c_locale_ctype); 74 break; 75 case lower: 76 __ret = __wctype_l("lower", _M_c_locale_ctype); 77 break; 78 case alpha: 79 __ret = __wctype_l("alpha", _M_c_locale_ctype); 80 break; 81 case digit: 82 __ret = __wctype_l("digit", _M_c_locale_ctype); 83 break; 84 case punct: 85 __ret = __wctype_l("punct", _M_c_locale_ctype); 86 break; 87 case xdigit: 88 __ret = __wctype_l("xdigit", _M_c_locale_ctype); 89 break; 90 case alnum: 91 __ret = __wctype_l("alnum", _M_c_locale_ctype); 92 break; 93 case graph: 94 __ret = __wctype_l("graph", _M_c_locale_ctype); 95 break; 96 default: 97 __ret = __wmask_type(); 98 } 99 return __ret; 100 } 101 102 wchar_t 103 ctype<wchar_t>::do_toupper(wchar_t __c) const 104 { return __towupper_l(__c, _M_c_locale_ctype); } 105 106 const wchar_t* 107 ctype<wchar_t>::do_toupper(wchar_t* __lo, const wchar_t* __hi) const 108 { 109 while (__lo < __hi) 110 { 111 *__lo = __towupper_l(*__lo, _M_c_locale_ctype); 112 ++__lo; 113 } 114 return __hi; 115 } 116 117 wchar_t 118 ctype<wchar_t>::do_tolower(wchar_t __c) const 119 { return __towlower_l(__c, _M_c_locale_ctype); } 120 121 const wchar_t* 122 ctype<wchar_t>::do_tolower(wchar_t* __lo, const wchar_t* __hi) const 123 { 124 while (__lo < __hi) 125 { 126 *__lo = __towlower_l(*__lo, _M_c_locale_ctype); 127 ++__lo; 128 } 129 return __hi; 130 } 131 132 bool 133 ctype<wchar_t>:: 134 do_is(mask __m, wchar_t __c) const 135 { 136 // The case of __m == ctype_base::space is particularly important, 137 // due to its use in many istream functions. Therefore we deal with 138 // it first, exploiting the knowledge that on GNU systems _M_bit[5] 139 // is the mask corresponding to ctype_base::space. NB: an encoding 140 // change would not affect correctness! 141 bool __ret = false; 142 if (__m == _M_bit[5]) 143 __ret = __iswctype_l(__c, _M_wmask[5], _M_c_locale_ctype); 144 else 145 { 146 // Highest bitmask in ctype_base == 10, but extra in "C" 147 // library for blank. 148 const size_t __bitmasksize = 11; 149 for (size_t __bitcur = 0; __bitcur <= __bitmasksize; ++__bitcur) 150 if (__m & _M_bit[__bitcur]) 151 { 152 if (__iswctype_l(__c, _M_wmask[__bitcur], _M_c_locale_ctype)) 153 { 154 __ret = true; 155 break; 156 } 157 else if (__m == _M_bit[__bitcur]) 158 break; 159 } 160 } 161 return __ret; 162 } 163 164 const wchar_t* 165 ctype<wchar_t>:: 166 do_is(const wchar_t* __lo, const wchar_t* __hi, mask* __vec) const 167 { 168 for (; __lo < __hi; ++__vec, ++__lo) 169 { 170 // Highest bitmask in ctype_base == 10, but extra in "C" 171 // library for blank. 172 const size_t __bitmasksize = 11; 173 mask __m = 0; 174 for (size_t __bitcur = 0; __bitcur <= __bitmasksize; ++__bitcur) 175 if (__iswctype_l(*__lo, _M_wmask[__bitcur], _M_c_locale_ctype)) 176 __m |= _M_bit[__bitcur]; 177 *__vec = __m; 178 } 179 return __hi; 180 } 181 182 const wchar_t* 183 ctype<wchar_t>:: 184 do_scan_is(mask __m, const wchar_t* __lo, const wchar_t* __hi) const 185 { 186 while (__lo < __hi && !this->do_is(__m, *__lo)) 187 ++__lo; 188 return __lo; 189 } 190 191 const wchar_t* 192 ctype<wchar_t>:: 193 do_scan_not(mask __m, const char_type* __lo, const char_type* __hi) const 194 { 195 while (__lo < __hi && this->do_is(__m, *__lo) != 0) 196 ++__lo; 197 return __lo; 198 } 199 200 wchar_t 201 ctype<wchar_t>:: 202 do_widen(char __c) const 203 { return _M_widen[static_cast<unsigned char>(__c)]; } 204 205 const char* 206 ctype<wchar_t>:: 207 do_widen(const char* __lo, const char* __hi, wchar_t* __dest) const 208 { 209 while (__lo < __hi) 210 { 211 *__dest = _M_widen[static_cast<unsigned char>(*__lo)]; 212 ++__lo; 213 ++__dest; 214 } 215 return __hi; 216 } 217 218 char 219 ctype<wchar_t>:: 220 do_narrow(wchar_t __wc, char __dfault) const 221 { 222 if (__wc >= 0 && __wc < 128 && _M_narrow_ok) 223 return _M_narrow[__wc]; 224#if __GLIBC__ > 2 || (__GLIBC__ == 2 && __GLIBC_MINOR__ > 2) 225 __c_locale __old = __uselocale(_M_c_locale_ctype); 226#endif 227 const int __c = wctob(__wc); 228#if __GLIBC__ > 2 || (__GLIBC__ == 2 && __GLIBC_MINOR__ > 2) 229 __uselocale(__old); 230#endif 231 return (__c == EOF ? __dfault : static_cast<char>(__c)); 232 } 233 234 const wchar_t* 235 ctype<wchar_t>:: 236 do_narrow(const wchar_t* __lo, const wchar_t* __hi, char __dfault, 237 char* __dest) const 238 { 239#if __GLIBC__ > 2 || (__GLIBC__ == 2 && __GLIBC_MINOR__ > 2) 240 __c_locale __old = __uselocale(_M_c_locale_ctype); 241#endif 242 if (_M_narrow_ok) 243 while (__lo < __hi) 244 { 245 if (*__lo >= 0 && *__lo < 128) 246 *__dest = _M_narrow[*__lo]; 247 else 248 { 249 const int __c = wctob(*__lo); 250 *__dest = (__c == EOF ? __dfault : static_cast<char>(__c)); 251 } 252 ++__lo; 253 ++__dest; 254 } 255 else 256 while (__lo < __hi) 257 { 258 const int __c = wctob(*__lo); 259 *__dest = (__c == EOF ? __dfault : static_cast<char>(__c)); 260 ++__lo; 261 ++__dest; 262 } 263#if __GLIBC__ > 2 || (__GLIBC__ == 2 && __GLIBC_MINOR__ > 2) 264 __uselocale(__old); 265#endif 266 return __hi; 267 } 268 269 void 270 ctype<wchar_t>::_M_initialize_ctype() throw() 271 { 272#if __GLIBC__ > 2 || (__GLIBC__ == 2 && __GLIBC_MINOR__ > 2) 273 __c_locale __old = __uselocale(_M_c_locale_ctype); 274#endif 275 wint_t __i; 276 for (__i = 0; __i < 128; ++__i) 277 { 278 const int __c = wctob(__i); 279 if (__c == EOF) 280 break; 281 else 282 _M_narrow[__i] = static_cast<char>(__c); 283 } 284 if (__i == 128) 285 _M_narrow_ok = true; 286 else 287 _M_narrow_ok = false; 288 for (size_t __j = 0; 289 __j < sizeof(_M_widen) / sizeof(wint_t); ++__j) 290 _M_widen[__j] = btowc(__j); 291 292 for (size_t __k = 0; __k <= 11; ++__k) 293 { 294 _M_bit[__k] = static_cast<mask>(_ISbit(__k)); 295 _M_wmask[__k] = _M_convert_to_wmask(_M_bit[__k]); 296 } 297#if __GLIBC__ > 2 || (__GLIBC__ == 2 && __GLIBC_MINOR__ > 2) 298 __uselocale(__old); 299#endif 300 } 301#endif // _GLIBCXX_USE_WCHAR_T 302 303_GLIBCXX_END_NAMESPACE 304