lib/uniwbrk/wbrktable.c

   1 /* Word break auxiliary table.  -*- coding: utf-8 -*-
   2    Copyright (C) 2009-2024 Free Software Foundation, Inc.
   3    Written by Bruno Haible <bruno@clisp.org>, 2009.
   4
   5    This file is free software.
   6    It is dual-licensed under "the GNU LGPLv3+ or the GNU GPLv2+".
   7    You can redistribute it and/or modify it under either
   8      - the terms of the GNU Lesser General Public License as published
   9        by the Free Software Foundation, either version 3, or (at your
  10        option) any later version, or
  11      - the terms of the GNU General Public License as published by the
  12        Free Software Foundation; either version 2, or (at your option)
  13        any later version, or
  14      - the same dual license "the GNU LGPLv3+ or the GNU GPLv2+".
  15
  16    This file is distributed in the hope that it will be useful,
  17    but WITHOUT ANY WARRANTY; without even the implied warranty of
  18    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  19    Lesser General Public License and the GNU General Public License
  20    for more details.
  21
  22    You should have received a copy of the GNU Lesser General Public
  23    License and of the GNU General Public License along with this
  24    program.  If not, see <https://www.gnu.org/licenses/>.  */
  25
  26 #include <config.h>
  27
  28 /* Specification.  */
  29 #include "wbrktable.h"
  30
  31 const int uniwbrk_prop_index[23] =
  32 {
  33   /* WBP_OTHER */         0,
  34   /* WBP_KATAKANA */      1,
  35   /* WBP_ALETTER */       2,
  36   /* WBP_MIDNUMLET */     3,
  37   /* WBP_MIDLETTER */     4,
  38   /* WBP_MIDNUM */        5,
  39   /* WBP_NUMERIC */       6,
  40   /* WBP_EXTENDNUMLET */  7,
  41   /* WBP_EXTEND */       -1,
  42   /* WBP_FORMAT */       -1,
  43   /* WBP_NEWLINE */      -1,
  44   /* WBP_CR */           -1,
  45   /* WBP_LF */           -1,
  46   /* WBP_RI */           -1,
  47   /* WBP_DQ */            8,
  48   /* WBP_SQ */            9,
  49   /* WBP_HL */           10,
  50   /* WBP_ZWJ */          -1,
  51   /* WBP_EB */           -1,
  52   /* WBP_EM */           -1,
  53   /* WBP_GAZ */          -1,
  54   /* WBP_EBG */          -1,
  55   /* WBP_WSS */          11
  56 };
  57
  58 /* This table contains the following rules (see UAX #29):
  59
  60                            last         current
  61
  62                      (ALetter | HL) × (ALetter | HL)                       (WB5)
  63                      (ALetter | HL) × Numeric                              (WB9)
  64                                  HL × SQ                                   (WB7a)
  65                             Numeric × (ALetter | HL)                       (WB10)
  66                             Numeric × Numeric                              (WB8)
  67                            Katakana × Katakana                             (WB13)
  68 (ALetter | HL | Numeric | Katakana) × ExtendNumLet                         (WB13a)
  69                        ExtendNumLet × ExtendNumLet                         (WB13a)
  70                        ExtendNumLet × (ALetter | HL | Numeric | Katakana)  (WB13b)
  71
  72    Note that the following rules are not handled here but in the loop in u-wordbreaks.h:
  73    - The rules need to look back or look ahead the second character (WB6, WB7, WB7b, WB7c, WB11, WB12)
  74    - The rules with a higher precedence over the "ignore" rule (WB4), such as WB3c
  75  */
  76
  77 const unsigned char uniwbrk_table[12][12] =
  78 {        /* current:        OTHER       MIDNUMLET   NUMERIC     SQ          */
  79          /*                     KATAKANA    MIDLETTER   EXNUMLET    HL      */
  80          /*                         ALETTER     MIDNUM      DQ          WSS */
  81   /* last */
  82   /* WBP_OTHER */        {  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1 },
  83   /* WBP_KATAKANA */     {  1,  0,  1,  1,  1,  1,  1,  0,  1,  1,  1,  1 },
  84   /* WBP_ALETTER */      {  1,  1,  0,  1,  1,  1,  0,  0,  1,  1,  0,  1 },
  85   /* WBP_MIDNUMLET */    {  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1 },
  86   /* WBP_MIDLETTER */    {  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1 },
  87   /* WBP_MIDNUM */       {  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1 },
  88   /* WBP_NUMERIC */      {  1,  1,  0,  1,  1,  1,  0,  0,  1,  1,  0,  1 },
  89   /* WBP_EXTENDNUMLET */ {  1,  0,  0,  1,  1,  1,  0,  0,  1,  1,  0,  1 },
  90   /* WBP_DQ */           {  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1 },
  91   /* WBP_SQ */           {  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1 },
  92   /* WBP_HL */           {  1,  1,  0,  1,  1,  1,  0,  0,  1,  0,  0,  1 },
  93   /* WBP_WSS */          {  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1 }
  94 };