iconvdata/iso-ir-165.h

   1 /* Tables for conversion to and from ISO-IR-165.
   2    converting from UCS using gaps.
   3    Copyright (C) 2000-2019 Free Software Foundation, Inc.
   4    This file is part of the GNU C Library.
   5    Contributed by Ulrich Drepper <drepper@cygnus.com>, 2000.
   6
   7    The GNU C Library is free software; you can redistribute it and/or
   8    modify it under the terms of the GNU Lesser General Public
   9    License as published by the Free Software Foundation; either
  10    version 2.1 of the License, or (at your option) any later version.
  11
  12    The GNU C 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 GNU
  15    Lesser General Public License for more details.
  16
  17    You should have received a copy of the GNU Lesser General Public
  18    License along with the GNU C Library; if not, see
  19    <https://www.gnu.org/licenses/>.  */
  20
  21 #ifndef _ISO_IR_165_H
  22 #define _ISO_IR_165_H   1
  23
  24 #include <gconv.h>
  25 #include <stdint.h>
  26
  27 struct gap
  28 {
  29   uint16_t start;
  30   uint16_t end;
  31   int32_t idx;
  32 };
  33
  34 /* Table for ISO-IR-165 (CCITT Chinese) to UCS4 conversion.  */
  35 #define ISOIR165_FROMSIZE       0x2284
  36 extern const uint16_t __isoir165_to_tab[ISOIR165_FROMSIZE];
  37
  38
  39 /* XXX If we at some point need an offset value to decode the byte
  40    sequences another parameter can be added.  */
  41 static inline uint32_t
  42 __attribute ((always_inline))
  43 isoir165_to_ucs4 (const unsigned char **s, size_t avail)
  44 {
  45   unsigned char ch = *(*s);
  46   unsigned char ch2;
  47   uint32_t res;
  48
  49   if (ch <= 0x20 || ch >= 0x7f)
  50     return __UNKNOWN_10646_CHAR;
  51
  52   if (avail < 2)
  53     return 0;
  54
  55   ch2 = (*s)[1];
  56   if (ch2 <= 0x20 || ch2 >= 0x7f)
  57     return __UNKNOWN_10646_CHAR;
  58
  59   res = __isoir165_to_tab[(ch - 0x21) * 94 + (ch2 - 0x21)];
  60   if (res == 0)
  61     return __UNKNOWN_10646_CHAR;
  62
  63   *s += 2;
  64   return res;
  65 }
  66
  67
  68 /* Tables for ISO-IR-165 (CCITT Chinese) from UCS4 conversion.  */
  69 extern const struct gap __isoir165_from_idx[];
  70 extern const char __isoir165_from_tab[];
  71
  72 static inline size_t
  73 __attribute ((always_inline))
  74 ucs4_to_isoir165 (uint32_t wch, unsigned char *s, size_t avail)
  75 {
  76   unsigned int ch = (unsigned int) wch;
  77   const char *cp;
  78   const struct gap *rp = __isoir165_from_idx;
  79
  80   if (ch > 0xffe5)
  81     /* This is an illegal character.  */
  82     return __UNKNOWN_10646_CHAR;
  83
  84   while (ch > rp->end)
  85     ++rp;
  86   if (ch < rp->start)
  87     /* This is an illegal character.  */
  88     return __UNKNOWN_10646_CHAR;
  89
  90   /* The two bytes following the index given in this record give the
  91      encoding in ISO-IR-165.  Unless the bytes are zero.  */
  92   cp = &__isoir165_from_tab[(ch + rp->idx) * 2];
  93   if (*cp == '\0')
  94     /* This is an illegal character.  */
  95     return __UNKNOWN_10646_CHAR;
  96
  97   if (avail < 2)
  98     return 0;
  99
 100   s[0] = cp[0];
 101   s[1] = cp[1];
 102
 103   return 2;
 104 }
 105
 106 #endif  /* iso-ir-165.h */