lib/uninorm/decomposition.c

   1 /* Decomposition of Unicode characters.
   2    Copyright (C) 2009-2019 Free Software Foundation, Inc.
   3    Written by Bruno Haible <bruno@clisp.org>, 2009.
   4
   5    This program is free software: you can redistribute it and/or modify it
   6    under the terms of the GNU Lesser General Public License as published
   7    by the Free Software Foundation; either version 3 of the License, or
   8    (at your option) any later version.
   9
  10    This program is distributed in the hope that it will be useful,
  11    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 License
  16    along with this program.  If not, see <https://www.gnu.org/licenses/>.  */
  17
  18 #include <config.h>
  19
  20 /* Specification.  */
  21 #include "uninorm.h"
  22
  23 #include "uninorm/decomposition-table.h"
  24
  25 int
  26 uc_decomposition (ucs4_t uc, int *decomp_tag, ucs4_t *decomposition)
  27 {
  28   if (uc >= 0xAC00 && uc < 0xD7A4)
  29     {
  30       /* Hangul syllable.  See Unicode standard, chapter 3, section
  31          "Hangul Syllable Decomposition",  See also the clarification at
  32          <http://www.unicode.org/versions/Unicode5.1.0/>, section
  33          "Clarification of Hangul Jamo Handling".  */
  34       unsigned int t;
  35
  36       uc -= 0xAC00;
  37       t = uc % 28;
  38
  39       *decomp_tag = UC_DECOMP_CANONICAL;
  40       if (t == 0)
  41         {
  42           unsigned int v, l;
  43
  44           uc = uc / 28;
  45           v = uc % 21;
  46           l = uc / 21;
  47
  48           decomposition[0] = 0x1100 + l;
  49           decomposition[1] = 0x1161 + v;
  50           return 2;
  51         }
  52       else
  53         {
  54 #if 1 /* Return the pairwise decomposition, not the full decomposition.  */
  55           decomposition[0] = 0xAC00 + uc - t; /* = 0xAC00 + (l * 21 + v) * 28; */
  56           decomposition[1] = 0x11A7 + t;
  57           return 2;
  58 #else
  59           unsigned int v, l;
  60
  61           uc = uc / 28;
  62           v = uc % 21;
  63           l = uc / 21;
  64
  65           decomposition[0] = 0x1100 + l;
  66           decomposition[1] = 0x1161 + v;
  67           decomposition[2] = 0x11A7 + t;
  68           return 3;
  69 #endif
  70         }
  71     }
  72   else if (uc < 0x110000)
  73     {
  74       unsigned short entry = decomp_index (uc);
  75       if (entry != (unsigned short)(-1))
  76         {
  77           const unsigned char *p;
  78           unsigned int element;
  79           unsigned int length;
  80
  81           p = &gl_uninorm_decomp_chars_table[3 * (entry & 0x7FFF)];
  82           element = (p[0] << 16) | (p[1] << 8) | p[2];
  83           /* The first element has 5 bits for the decomposition type.  */
  84           *decomp_tag = (element >> 18) & 0x1f;
  85           length = 1;
  86           for (;;)
  87             {
  88               /* Every element has an 18 bits wide Unicode code point.  */
  89               *decomposition = element & 0x3ffff;
  90               /* Bit 23 tells whether there are more elements,  */
  91               if ((element & (1 << 23)) == 0)
  92                 break;
  93               p += 3;
  94               element = (p[0] << 16) | (p[1] << 8) | p[2];
  95               decomposition++;
  96               length++;
  97             }
  98           return length;
  99         }
 100     }
 101   return -1;
 102 }