lib/utf32.h

   1 /*
   2  * Copyright (C) 1999-2001, 2008, 2011, 2016, 2024 Free Software Foundation, Inc.
   3  * This file is part of the GNU LIBICONV Library.
   4  *
   5  * The GNU LIBICONV Library is free software; you can redistribute it
   6  * and/or modify it under the terms of the GNU Lesser General Public
   7  * License as published by the Free Software Foundation; either version 2.1
   8  * of the License, or (at your option) any later version.
   9  *
  10  * The GNU LIBICONV Library is distributed in the hope that it will be
  11  * useful, but WITHOUT ANY WARRANTY; without even the implied warranty of
  12  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  13  * Lesser General Public License for more details.
  14  *
  15  * You should have received a copy of the GNU Lesser General Public
  16  * License along with the GNU LIBICONV Library; see the file COPYING.LIB.
  17  * If not, see <https://www.gnu.org/licenses/>.
  18  */
  19
  20 /*
  21  * UTF-32
  22  */
  23
  24 /* Specification: Unicode 3.1 Standard Annex #19 */
  25
  26 /* Here we accept FFFE0000/0000FEFF marks as endianness indicators
  27    everywhere in the stream, not just at the beginning. (This is contrary
  28    to what #19 D36c specifies, but it allows concatenation of byte
  29    sequences to work flawlessly, while disagreeing with #19 behaviour
  30    only for strings containing U+FEFF characters, which is quite rare.)
  31    The default is big-endian. */
  32 /* The state is 0 if big-endian, 1 if little-endian. */
  33 static int
  34 utf32_mbtowc (conv_t conv, ucs4_t *pwc, const unsigned char *s, size_t n)
  35 {
  36   state_t state = conv->istate;
  37   int count = 0;
  38   for (; n >= 4 && count <= RET_COUNT_MAX && count <= INT_MAX-4;) {
  39     ucs4_t wc = (state
  40                   ?    (ucs4_t) s[0]
  41                     + ((ucs4_t) s[1] << 8)
  42                     + ((ucs4_t) s[2] << 16)
  43                     + ((ucs4_t) s[3] << 24)
  44                   :   ((ucs4_t) s[0] << 24)
  45                     + ((ucs4_t) s[1] << 16)
  46                     + ((ucs4_t) s[2] << 8)
  47                     +  (ucs4_t) s[3]);
  48     if (wc == 0x0000feff) {
  49     } else if (wc == 0xfffe0000u) {
  50       state ^= 1;
  51     } else {
  52       if (wc < 0x110000 && !(wc >= 0xd800 && wc < 0xe000)) {
  53         *pwc = wc;
  54         conv->istate = state;
  55         return count+4;
  56       } else {
  57         conv->istate = state;
  58         return RET_SHIFT_ILSEQ(count);
  59       }
  60     }
  61     s += 4; n -= 4; count += 4;
  62   }
  63   conv->istate = state;
  64   return RET_TOOFEW(count);
  65 }
  66
  67 /* We output UTF-32 in big-endian order, with byte-order mark. */
  68 /* The state is 0 at the beginning, 1 after the BOM has been written. */
  69 static int
  70 utf32_wctomb (conv_t conv, unsigned char *r, ucs4_t wc, size_t n)
  71 {
  72   if (wc < 0x110000 && !(wc >= 0xd800 && wc < 0xe000)) {
  73     int count = 0;
  74     if (!conv->ostate) {
  75       if (n >= 4) {
  76         r[0] = 0x00;
  77         r[1] = 0x00;
  78         r[2] = 0xFE;
  79         r[3] = 0xFF;
  80         r += 4; n -= 4; count += 4;
  81       } else
  82         return RET_TOOSMALL;
  83     }
  84     if (wc < 0x110000) {
  85       if (n >= 4) {
  86         r[0] = 0;
  87         r[1] = (unsigned char) (wc >> 16);
  88         r[2] = (unsigned char) (wc >> 8);
  89         r[3] = (unsigned char) wc;
  90         conv->ostate = 1;
  91         return count+4;
  92       } else
  93         return RET_TOOSMALL;
  94     }
  95   }
  96   return RET_ILUNI;
  97 }