lib/unictype/categ_name.c

   1 /* Categories of Unicode characters.
   2    Copyright (C) 2002, 2006-2007, 2011-2021 Free Software Foundation, Inc.
   3    Written by Bruno Haible <bruno@clisp.org>, 2002.
   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 "unictype.h"
  22
  23 static const char u_category_name[30][3] =
  24 {
  25   "Lu", "Ll", "Lt", "Lm", "Lo", "Mn", "Mc", "Me", "Nd", "Nl",
  26   "No", "Pc", "Pd", "Ps", "Pe", "Pi", "Pf", "Po", "Sm", "Sc",
  27   "Sk", "So", "Zs", "Zl", "Zp", "Cc", "Cf", "Cs", "Co", "Cn"
  28 };
  29
  30 const char *
  31 uc_general_category_name (uc_general_category_t category)
  32 {
  33   uint32_t bitmask = category.bitmask;
  34   /* bitmask should consist of a single bit.  */
  35   if (bitmask != 0)
  36     {
  37       if ((bitmask & (bitmask - 1)) == 0)
  38         {
  39           int bit;
  40           /* Take log2 using a variant of Robert Harley's method.
  41              Found by Bruno Haible 1996.  */
  42           uint32_t n = bitmask;
  43           static const char ord2_tab[64] =
  44             {
  45               -1,  0,  1, 12,  2,  6, -1, 13,  3, -1,  7, -1, -1, -1, -1, 14,
  46               10,  4, -1, -1,  8, -1, -1, 25, -1, -1, -1, -1, -1, 21, 27, 15,
  47               31, 11,  5, -1, -1, -1, -1, -1,  9, -1, -1, 24, -1, -1, 20, 26,
  48               30, -1, -1, -1, -1, 23, -1, 19, 29, -1, 22, 18, 28, 17, 16, -1
  49             };
  50           n += n << 4;
  51           n += n << 6;
  52           n = (n << 16) - n;
  53           bit = ord2_tab[n >> 26];
  54
  55           if (bit < sizeof (u_category_name) / sizeof (u_category_name[0]))
  56             return u_category_name[bit];
  57         }
  58       else
  59         {
  60           if (bitmask == UC_CATEGORY_MASK_L)
  61             return "L";
  62           if (bitmask == UC_CATEGORY_MASK_LC)
  63             return "LC";
  64           if (bitmask == UC_CATEGORY_MASK_M)
  65             return "M";
  66           if (bitmask == UC_CATEGORY_MASK_N)
  67             return "N";
  68           if (bitmask == UC_CATEGORY_MASK_P)
  69             return "P";
  70           if (bitmask == UC_CATEGORY_MASK_S)
  71             return "S";
  72           if (bitmask == UC_CATEGORY_MASK_Z)
  73             return "Z";
  74           if (bitmask == UC_CATEGORY_MASK_C)
  75             return "C";
  76         }
  77     }
  78   return NULL;
  79 }