libstdc++-v3/testsuite/22_locale/codecvt/in/wchar_t/9.cc

   1 // { dg-require-namedlocale "en_US.UTF-8" }
   2
   3 // 2003-02-06  Petur Runolfsson  <peturr02@ru.is>
   4
   5 // Copyright (C) 2003-2021 Free Software Foundation, Inc.
   6 //
   7 // This file is part of the GNU ISO C++ Library.  This library is free
   8 // software; you can redistribute it and/or modify it under the
   9 // terms of the GNU General Public License as published by the
  10 // Free Software Foundation; either version 3, or (at your option)
  11 // any later version.
  12
  13 // This library is distributed in the hope that it will be useful,
  14 // but WITHOUT ANY WARRANTY; without even the implied warranty of
  15 // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  16 // GNU General Public License for more details.
  17
  18 // You should have received a copy of the GNU General Public License along
  19 // with this library; see the file COPYING3.  If not see
  20 // <http://www.gnu.org/licenses/>.
  21
  22 // 22.2.1.5 - Template class codecvt [lib.locale.codecvt]
  23
  24 #include <locale>
  25 #include <cstring>
  26 #include <testsuite_hooks.h>
  27
  28 // Need to explicitly set the state(mbstate_t) to zero.
  29 // How to do this is not specified by the ISO C99 standard, so we
  30 // might need to add some operators to make the intuiative case
  31 // work:
  32 //   w_codecvt::state_type state00;
  33 //   state00 = 0;
  34 // or, can use this explicit "C" initialization:
  35 //   w_codecvt::state_type state01 = {0, 0};
  36 // .. except Ulrich says: Use memset. Always use memset. Feel the force...
  37 void
  38 zero_state(std::mbstate_t& state)
  39 { std::memset(&state, 0, sizeof(std::mbstate_t)); }
  40
  41 // Required instantiation
  42 // codecvt<wchar_t, char, mbstate_t>
  43 //
  44 // Test one source character at a time for UTF-8
  45 void test09()
  46 {
  47   using namespace std;
  48   typedef codecvt<wchar_t, char, mbstate_t>     w_codecvt;
  49   typedef codecvt_base::result                  result;
  50   typedef wchar_t                               int_type;
  51   typedef char                                  ext_type;
  52   typedef char_traits<wchar_t>                  int_traits;
  53
  54   const ext_type*       e_lit =
  55     "\x1\x2\x3\x4\x5\x6\x7\x8\x9\xa\xb\xc\xd\xe\xf\x10\x11\x12\x13"
  56     "\x14\x15\x16\x17\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f\x20!\"#$%&"
  57     "'()*+,-./0123456789:;<=>?@}~\x7f\xc2\x80\xc2\x81\xc2\x82\xc2"
  58     "\x83\xc2\x84\xc2\x85\xc2\x86\xc2\x87\xc2\x88\xc2\x89\xc2\x8a"
  59     "\xc2\x8b\xc2\x8c\xc2\x8d\xc2\x8e\xc2\x8f\xc2\x90\xc2\x91\xc2"
  60     "\x92\xc2\x93\xc2\x94\xc2\x95\xc2\x96\xc2\x97\xc2\x98\xc2\x99"
  61     "\xc2\x9a\xc2\x9b\xc2\x9c\xc3\xba\xc3\xbb\xc3\xbc\xc3\xbd\xc3"
  62     "\xbe\xc3\xbf\xc4\x80\xc4\x81\xc4\x82\xc4\x83\xc4\x84\xc4\x85"
  63     "\xc4\x86\xc4\x87\xc4\x88\xc4\x89\xc4\x8a\xc4\x8b\xc4\x8c\xc4"
  64     "\x8d\xc4\x8e\xc4\x8f\xc4\x90\xc4\x91\xc4\x92\xc4\x93\xc4\x94"
  65     "\xc4\x95\xc4\x96\xc4\x97\xc4\x98\xc4\x99\xdf\xb8\xdf\xb9\xdf"
  66     "\xba\xdf\xbb\xdf\xbc\xdf\xbd\xdf\xbe\xdf\xbf\xe0\xa0\x80\xe0"
  67     "\xa0\x81\xe0\xa0\x82\xe0\xa0\x83\xe0\xa0\x84\xe0\xa0\x85\xe0"
  68     "\xa0\x86\xe0\xa0\x87\xe0\xa0\x88\xe0\xa0\x89\xe0\xa0\x8a\xe0"
  69     "\xa0\x8b\xe0\xa0\x8c\xe0\xa0\x8d\xe0\xa0\x8e\xe0\xa0\x8f\xe0"
  70     "\xa0\x90\xe0\xa0\x91\xe0\xa0\x92\xe0\xa0\x93\xe0\xa0\x94\xe0"
  71     "\xa0\x95\xe0\xa0\x96\xe0\xa0\x97\x1\x2\x4\x8\x10\x20@\xc2\x80"
  72     "\xc4\x80\xc8\x80\xd0\x80\xe0\xa0\x80\xe1\x80\x80\xe2\x80\x80"
  73     "\xe4\x80\x80\xe8\x80\x80\xf0\x90\x80\x80\xf0\xa0\x80\x80\xf1"
  74     "\x80\x80\x80\xf2\x80\x80\x80\xf4\x80\x80\x80";
  75
  76   const ext_type*       efrom = e_lit;
  77   const ext_type*       efrom_next;
  78   const int_type        i_lit[] = {
  79     0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, 0x8, 0x9, 0xa, 0xb, 0xc,
  80     0xd, 0xe, 0xf, 0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17,
  81     0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f, 0x20, L'!',
  82     L'"', L'#', L'$', L'%', L'&', L'\'', L'(', L')', L'*', L'+',
  83     L',', L'-', L'.', L'/', L'0', L'1', L'2', L'3', L'4', L'5',
  84     L'6', L'7', L'8', L'9', L':', L';', L'<', L'=', L'>', L'?',
  85     L'@', L'}', L'~', 0x7f, 0x80, 0x81, 0x82, 0x83, 0x84, 0x85,
  86     0x86, 0x87, 0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e, 0x8f,
  87     0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97, 0x98, 0x99,
  88     0x9a, 0x9b, 0x9c, 0xfa, 0xfb, 0xfc, 0xfd, 0xfe, 0xff, 0x100,
  89     0x101, 0x102, 0x103, 0x104, 0x105, 0x106, 0x107, 0x108, 0x109,
  90     0x10a, 0x10b, 0x10c, 0x10d, 0x10e, 0x10f, 0x110, 0x111, 0x112,
  91     0x113, 0x114, 0x115, 0x116, 0x117, 0x118, 0x119, 0x7f8, 0x7f9,
  92     0x7fa, 0x7fb, 0x7fc, 0x7fd, 0x7fe, 0x7ff, 0x800, 0x801, 0x802,
  93     0x803, 0x804, 0x805, 0x806, 0x807, 0x808, 0x809, 0x80a, 0x80b,
  94     0x80c, 0x80d, 0x80e, 0x80f, 0x810, 0x811, 0x812, 0x813, 0x814,
  95     0x815, 0x816, 0x817, 0x1, 0x2, 0x4, 0x8, 0x10, 0x20, L'@',
  96     0x80, 0x100, 0x200, 0x400, 0x800, 0x1000, 0x2000, 0x4000, 0x8000,
  97     0x10000, 0x20000, 0x40000, 0x80000, 0x100000, 0x0
  98   };
  99
 100   int                   esize = strlen(e_lit);
 101   int                   isize = wcslen(i_lit);
 102   int_type*             i_arr = new int_type[esize + 1];
 103   int_type*             i_ref = new int_type[esize + 1];
 104   wmemset(i_arr, 0xdeadbeef, esize + 1);
 105   wmemset(i_ref, 0xdeadbeef, esize + 1);
 106   int_type*             ito = i_arr;
 107   int_type*             ito_next;
 108
 109   locale loc = locale("en_US.UTF-8");
 110   locale::global(loc);
 111   const w_codecvt*      cvt = &use_facet<w_codecvt>(loc);
 112
 113   // in
 114   w_codecvt::state_type state01;
 115   zero_state(state01);
 116   result r1 = codecvt_base::error;
 117
 118   for (int i = 0; i <= esize; ++i)
 119     {
 120       r1 = cvt->in(state01, efrom, e_lit + i, efrom_next,
 121                    ito, i_arr + esize, ito_next);
 122
 123       // It it not clear if partial should ever be returned here
 124       // (see DR 382).
 125       VERIFY( r1 == codecvt_base::ok || r1 == codecvt_base::partial );
 126       VERIFY( efrom_next >= efrom );
 127       VERIFY( efrom_next <= e_lit + i );
 128       VERIFY( ito_next >= ito );
 129       VERIFY( ito_next <= i_arr + isize );
 130       VERIFY( !int_traits::compare(i_arr, i_lit, ito_next - i_arr) );
 131       VERIFY( !int_traits::compare(ito_next, i_ref,
 132                                    i_arr + esize + 1 - ito_next) );
 133
 134       efrom = efrom_next;
 135       ito = ito_next;
 136     }
 137
 138   VERIFY( r1 == codecvt_base::ok );
 139   VERIFY( efrom_next == e_lit + esize );
 140   VERIFY( ito_next == i_arr + isize );
 141
 142   delete [] i_arr;
 143   delete [] i_ref;
 144 }
 145
 146 int main ()
 147 {
 148   test09();
 149   return 0;
 150 }