lib/c99.h

   1 /*
   2  * Copyright (C) 1999-2002, 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  * C99
  22  * This is ASCII with \uXXXX and \UXXXXXXXX escape sequences, denoting Unicode
  23  * characters. See ISO/IEC 9899:1999, section 6.4.3.
  24  * The treatment of control characters in the range U+0080..U+009F is not
  25  * specified; we pass them through unmodified.
  26  */
  27
  28 static int
  29 c99_mbtowc (conv_t conv, ucs4_t *pwc, const unsigned char *s, size_t n)
  30 {
  31   unsigned char c;
  32   ucs4_t wc;
  33   int i;
  34
  35   c = s[0];
  36   if (c < 0xa0) {
  37     if (c != '\\') {
  38       *pwc = c;
  39       return 1;
  40     }
  41     if (n < 2)
  42       return RET_TOOFEW(0);
  43     c = s[1];
  44     if (c == 'u') {
  45       wc = 0;
  46       for (i = 2; i < 6; i++) {
  47         if (n <= i)
  48           return RET_TOOFEW(0);
  49         c = s[i];
  50         if (c >= '0' && c <= '9')
  51           c -= '0';
  52         else if (c >= 'A' && c <= 'Z')
  53           c -= 'A'-10;
  54         else if (c >= 'a' && c <= 'z')
  55           c -= 'a'-10;
  56         else
  57           goto simply_backslash;
  58         wc |= (ucs4_t) c << (4 * (5-i));
  59       }
  60       if ((wc >= 0x00a0 && !(wc >= 0xd800 && wc < 0xe000))
  61           || wc == 0x0024 || wc == 0x0040 || wc == 0x0060) {
  62         *pwc = wc;
  63         return 6;
  64       }
  65     } else if (c == 'U') {
  66       wc = 0;
  67       for (i = 2; i < 10; i++) {
  68         if (n <= i)
  69           return RET_TOOFEW(0);
  70         c = s[i];
  71         if (c >= '0' && c <= '9')
  72           c -= '0';
  73         else if (c >= 'A' && c <= 'Z')
  74           c -= 'A'-10;
  75         else if (c >= 'a' && c <= 'z')
  76           c -= 'a'-10;
  77         else
  78           goto simply_backslash;
  79         wc |= (ucs4_t) c << (4 * (9-i));
  80       }
  81       if ((wc >= 0x00a0 && !(wc >= 0xd800 && wc < 0xe000))
  82           || wc == 0x0024 || wc == 0x0040 || wc == 0x0060) {
  83         *pwc = wc;
  84         return 10;
  85       }
  86     } else
  87       goto simply_backslash;
  88   }
  89   return RET_ILSEQ;
  90 simply_backslash:
  91   *pwc = '\\';
  92   return 1;
  93 }
  94
  95 static int
  96 c99_wctomb (conv_t conv, unsigned char *r, ucs4_t wc, size_t n)
  97 {
  98   if (wc < 0xa0) {
  99     *r = wc;
 100     return 1;
 101   } else {
 102     int result;
 103     unsigned char u;
 104     if (wc < 0x10000) {
 105       result = 6;
 106       u = 'u';
 107     } else {
 108       result = 10;
 109       u = 'U';
 110     }
 111     if (n >= result) {
 112       int count;
 113       r[0] = '\\';
 114       r[1] = u;
 115       r += 2;
 116       for (count = result-3; count >= 0; count--) {
 117         unsigned int i = (wc >> (4*count)) & 0x0f;
 118         *r++ = (i < 10 ? '0'+i : 'a'-10+i);
 119       }
 120       return result;
 121     } else
 122       return RET_TOOSMALL;
 123   }
 124 }