lib/utf8.h

   1 /*
   2  * Copyright (C) 1999-2001, 2004, 2016 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-8
  22  */
  23
  24 /* Specification: RFC 3629 */
  25
  26 static int
  27 utf8_mbtowc (conv_t conv, ucs4_t *pwc, const unsigned char *s, size_t n)
  28 {
  29   unsigned char c = s[0];
  30
  31   if (c < 0x80) {
  32     *pwc = c;
  33     return 1;
  34   } else if (c < 0xc2) {
  35     return RET_ILSEQ;
  36   } else if (c < 0xe0) {
  37     if (n < 2)
  38       return RET_TOOFEW(0);
  39     if (!((s[1] ^ 0x80) < 0x40))
  40       return RET_ILSEQ;
  41     *pwc = ((ucs4_t) (c & 0x1f) << 6)
  42            | (ucs4_t) (s[1] ^ 0x80);
  43     return 2;
  44   } else if (c < 0xf0) {
  45     if (n < 3)
  46       return RET_TOOFEW(0);
  47     if (!((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
  48           && (c >= 0xe1 || s[1] >= 0xa0)
  49           && (c != 0xed || s[1] < 0xa0)))
  50       return RET_ILSEQ;
  51     *pwc = ((ucs4_t) (c & 0x0f) << 12)
  52            | ((ucs4_t) (s[1] ^ 0x80) << 6)
  53            | (ucs4_t) (s[2] ^ 0x80);
  54     return 3;
  55   } else if (c < 0xf8 && sizeof(ucs4_t)*8 >= 32) {
  56     if (n < 4)
  57       return RET_TOOFEW(0);
  58     if (!((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
  59           && (s[3] ^ 0x80) < 0x40
  60           && (c >= 0xf1 || s[1] >= 0x90)
  61           && (c < 0xf4 || (c == 0xf4 && s[1] < 0x90))))
  62       return RET_ILSEQ;
  63     *pwc = ((ucs4_t) (c & 0x07) << 18)
  64            | ((ucs4_t) (s[1] ^ 0x80) << 12)
  65            | ((ucs4_t) (s[2] ^ 0x80) << 6)
  66            | (ucs4_t) (s[3] ^ 0x80);
  67     return 4;
  68   } else
  69     return RET_ILSEQ;
  70 }
  71
  72 static int
  73 utf8_wctomb (conv_t conv, unsigned char *r, ucs4_t wc, size_t n) /* n == 0 is acceptable */
  74 {
  75   int count;
  76   if (wc < 0x80)
  77     count = 1;
  78   else if (wc < 0x800)
  79     count = 2;
  80   else if (wc < 0x10000) {
  81     if (wc < 0xd800 || wc >= 0xe000)
  82       count = 3;
  83     else
  84       return RET_ILUNI;
  85   } else if (wc < 0x110000)
  86     count = 4;
  87   else
  88     return RET_ILUNI;
  89   if (n < count)
  90     return RET_TOOSMALL;
  91   switch (count) { /* note: code falls through cases! */
  92     case 4: r[3] = 0x80 | (wc & 0x3f); wc = wc >> 6; wc |= 0x10000;
  93     case 3: r[2] = 0x80 | (wc & 0x3f); wc = wc >> 6; wc |= 0x800;
  94     case 2: r[1] = 0x80 | (wc & 0x3f); wc = wc >> 6; wc |= 0xc0;
  95     case 1: r[0] = wc;
  96   }
  97   return count;
  98 }