release/src/router/libiconv/lib/big5hkscs2004.h

   1 /*
   2  * Copyright (C) 1999-2002, 2006 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 Library General Public
   7  * License as published by the Free Software Foundation; either version 2
   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  * Library General Public License for more details.
  14  *
  15  * You should have received a copy of the GNU Library General Public
  16  * License along with the GNU LIBICONV Library; see the file COPYING.LIB.
  17  * If not, write to the Free Software Foundation, Inc., 51 Franklin Street,
  18  * Fifth Floor, Boston, MA 02110-1301, USA.
  19  */
  20
  21 /*
  22  * BIG5-HKSCS:2004
  23  */
  24
  25 /*
  26  * BIG5-HKSCS:2004 can be downloaded from
  27  *   http://www.info.gov.hk/digital21/eng/hkscs/download.html
  28  *   http://www.info.gov.hk/digital21/eng/hkscs/index.html
  29  *
  30  * It extends BIG5-HKSCS:2001 through 123 characters.
  31  *
  32  * It extends BIG5 (without the rows 0xC6..0xC7) through the ranges
  33  *
  34  *   0x{87..8D}{40..7E,A1..FE}      880 characters
  35  *   0x{8E..A0}{40..7E,A1..FE}     2898 characters
  36  *   0x{C6..C8}{40..7E,A1..FE}      359 characters
  37  *   0xF9{D6..FE}                    41 characters
  38  *   0x{FA..FE}{40..7E,A1..FE}      763 characters
  39  *
  40  * Note that some HKSCS characters are not contained in Unicode 3.2
  41  * and are therefore best represented as sequences of Unicode characters:
  42  *   0x8862  U+00CA U+0304  LATIN CAPITAL LETTER E WITH CIRCUMFLEX AND MACRON
  43  *   0x8864  U+00CA U+030C  LATIN CAPITAL LETTER E WITH CIRCUMFLEX AND CARON
  44  *   0x88A3  U+00EA U+0304  LATIN SMALL LETTER E WITH CIRCUMFLEX AND MACRON
  45  *   0x88A5  U+00EA U+030C  LATIN SMALL LETTER E WITH CIRCUMFLEX AND CARON
  46  */
  47
  48 #include "hkscs2004.h"
  49 #include "flushwc.h"
  50
  51 static int
  52 big5hkscs2004_mbtowc (conv_t conv, ucs4_t *pwc, const unsigned char *s, int n)
  53 {
  54   ucs4_t last_wc = conv->istate;
  55   if (last_wc) {
  56     /* Output the buffered character. */
  57     conv->istate = 0;
  58     *pwc = last_wc;
  59     return 0; /* Don't advance the input pointer. */
  60   } else {
  61     unsigned char c = *s;
  62     /* Code set 0 (ASCII) */
  63     if (c < 0x80)
  64       return ascii_mbtowc(conv,pwc,s,n);
  65     /* Code set 1 (BIG5 extended) */
  66     if (c >= 0xa1 && c < 0xff) {
  67       if (n < 2)
  68         return RET_TOOFEW(0);
  69       {
  70         unsigned char c2 = s[1];
  71         if ((c2 >= 0x40 && c2 < 0x7f) || (c2 >= 0xa1 && c2 < 0xff)) {
  72           if (!((c == 0xc6 && c2 >= 0xa1) || c == 0xc7)) {
  73             int ret = big5_mbtowc(conv,pwc,s,2);
  74             if (ret != RET_ILSEQ)
  75               return ret;
  76           }
  77         }
  78       }
  79     }
  80     {
  81       int ret = hkscs1999_mbtowc(conv,pwc,s,n);
  82       if (ret != RET_ILSEQ)
  83         return ret;
  84     }
  85     {
  86       int ret = hkscs2001_mbtowc(conv,pwc,s,n);
  87       if (ret != RET_ILSEQ)
  88         return ret;
  89     }
  90     {
  91       int ret = hkscs2004_mbtowc(conv,pwc,s,n);
  92       if (ret != RET_ILSEQ)
  93         return ret;
  94     }
  95     if (c == 0x88) {
  96       if (n < 2)
  97         return RET_TOOFEW(0);
  98       {
  99         unsigned char c2 = s[1];
 100         if (c2 == 0x62 || c2 == 0x64 || c2 == 0xa3 || c2 == 0xa5) {
 101           /* It's a composed character. */
 102           ucs4_t wc1 = ((c2 >> 3) << 2) + 0x009a; /* = 0x00ca or 0x00ea */
 103           ucs4_t wc2 = ((c2 & 6) << 2) + 0x02fc; /* = 0x0304 or 0x030c */
 104           /* We cannot output two Unicode characters at once. So,
 105              output the first character and buffer the second one. */
 106           *pwc = wc1;
 107           conv->istate = wc2;
 108           return 2;
 109         }
 110       }
 111     }
 112     return RET_ILSEQ;
 113   }
 114 }
 115
 116 #define big5hkscs2004_flushwc normal_flushwc
 117
 118 static int
 119 big5hkscs2004_wctomb (conv_t conv, unsigned char *r, ucs4_t wc, int n)
 120 {
 121   int count = 0;
 122   unsigned char last = conv->ostate;
 123
 124   if (last) {
 125     /* last is = 0x66 or = 0xa7. */
 126     if (wc == 0x0304 || wc == 0x030c) {
 127       /* Output the combined character. */
 128       if (n >= 2) {
 129         r[0] = 0x88;
 130         r[1] = last + ((wc & 24) >> 2) - 4; /* = 0x62 or 0x64 or 0xa3 or 0xa5 */
 131         conv->ostate = 0;
 132         return 2;
 133       } else
 134         return RET_TOOSMALL;
 135     }
 136
 137     /* Output the buffered character. */
 138     if (n < 2)
 139       return RET_TOOSMALL;
 140     r[0] = 0x88;
 141     r[1] = last;
 142     r += 2;
 143     count = 2;
 144   }
 145
 146   /* Code set 0 (ASCII) */
 147   if (wc < 0x0080) {
 148     /* Plain ASCII character. */
 149     if (n > count) {
 150       r[0] = (unsigned char) wc;
 151       conv->ostate = 0;
 152       return count+1;
 153     } else
 154       return RET_TOOSMALL;
 155   } else {
 156     unsigned char buf[2];
 157     int ret;
 158
 159     /* Code set 1 (BIG5 extended) */
 160     ret = big5_wctomb(conv,buf,wc,2);
 161     if (ret != RET_ILUNI) {
 162       if (ret != 2) abort();
 163       if (!((buf[0] == 0xc6 && buf[1] >= 0xa1) || buf[0] == 0xc7)) {
 164         if (n >= count+2) {
 165           r[0] = buf[0];
 166           r[1] = buf[1];
 167           conv->ostate = 0;
 168           return count+2;
 169         } else
 170           return RET_TOOSMALL;
 171       }
 172     }
 173     ret = hkscs1999_wctomb(conv,buf,wc,2);
 174     if (ret != RET_ILUNI) {
 175       if (ret != 2) abort();
 176       if ((wc & ~0x0020) == 0x00ca) {
 177         /* A possible first character of a multi-character sequence. We have to
 178            buffer it. */
 179         if (!(buf[0] == 0x88 && (buf[1] == 0x66 || buf[1] == 0xa7))) abort();
 180         conv->ostate = buf[1]; /* = 0x66 or = 0xa7 */
 181         return count+0;
 182       }
 183       if (n >= count+2) {
 184         r[0] = buf[0];
 185         r[1] = buf[1];
 186         conv->ostate = 0;
 187         return count+2;
 188       } else
 189         return RET_TOOSMALL;
 190     }
 191     ret = hkscs2001_wctomb(conv,buf,wc,2);
 192     if (ret != RET_ILUNI) {
 193       if (ret != 2) abort();
 194       if (n >= count+2) {
 195         r[0] = buf[0];
 196         r[1] = buf[1];
 197         conv->ostate = 0;
 198         return count+2;
 199       } else
 200         return RET_TOOSMALL;
 201     }
 202     ret = hkscs2004_wctomb(conv,buf,wc,2);
 203     if (ret != RET_ILUNI) {
 204       if (ret != 2) abort();
 205       if (n >= count+2) {
 206         r[0] = buf[0];
 207         r[1] = buf[1];
 208         conv->ostate = 0;
 209         return count+2;
 210       } else
 211         return RET_TOOSMALL;
 212     }
 213     return RET_ILUNI;
 214   }
 215 }
 216
 217 static int
 218 big5hkscs2004_reset (conv_t conv, unsigned char *r, int n)
 219 {
 220   unsigned char last = conv->ostate;
 221
 222   if (last) {
 223     if (n < 2)
 224       return RET_TOOSMALL;
 225     r[0] = 0x88;
 226     r[1] = last;
 227     /* conv->ostate = 0; will be done by the caller */
 228     return 2;
 229   } else
 230     return 0;
 231 }