1/* 2 * Copyright (C) 1999-2002, 2006, 2010 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:2008 23 */ 24 25/* 26 * BIG5-HKSCS:2008 can be downloaded from 27 * http://www.ogcio.gov.hk/ccli/eng/hkscs/download.html 28 * http://www.ogcio.gov.hk/ccli/eng/hkscs/introduction.html 29 * 30 * It extends BIG5-HKSCS:2004 through 68 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 "hkscs2008.h" 49#include "flushwc.h" 50 51static int 52big5hkscs2008_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 { 96 int ret = hkscs2008_mbtowc(conv,pwc,s,n); 97 if (ret != RET_ILSEQ) 98 return ret; 99 } 100 if (c == 0x88) { 101 if (n < 2) 102 return RET_TOOFEW(0); 103 { 104 unsigned char c2 = s[1]; 105 if (c2 == 0x62 || c2 == 0x64 || c2 == 0xa3 || c2 == 0xa5) { 106 /* It's a composed character. */ 107 ucs4_t wc1 = ((c2 >> 3) << 2) + 0x009a; /* = 0x00ca or 0x00ea */ 108 ucs4_t wc2 = ((c2 & 6) << 2) + 0x02fc; /* = 0x0304 or 0x030c */ 109 /* We cannot output two Unicode characters at once. So, 110 output the first character and buffer the second one. */ 111 *pwc = wc1; 112 conv->istate = wc2; 113 return 2; 114 } 115 } 116 } 117 return RET_ILSEQ; 118 } 119} 120 121#define big5hkscs2008_flushwc normal_flushwc 122 123static int 124big5hkscs2008_wctomb (conv_t conv, unsigned char *r, ucs4_t wc, int n) 125{ 126 int count = 0; 127 unsigned char last = conv->ostate; 128 129 if (last) { 130 /* last is = 0x66 or = 0xa7. */ 131 if (wc == 0x0304 || wc == 0x030c) { 132 /* Output the combined character. */ 133 if (n >= 2) { 134 r[0] = 0x88; 135 r[1] = last + ((wc & 24) >> 2) - 4; /* = 0x62 or 0x64 or 0xa3 or 0xa5 */ 136 conv->ostate = 0; 137 return 2; 138 } else 139 return RET_TOOSMALL; 140 } 141 142 /* Output the buffered character. */ 143 if (n < 2) 144 return RET_TOOSMALL; 145 r[0] = 0x88; 146 r[1] = last; 147 r += 2; 148 count = 2; 149 } 150 151 /* Code set 0 (ASCII) */ 152 if (wc < 0x0080) { 153 /* Plain ASCII character. */ 154 if (n > count) { 155 r[0] = (unsigned char) wc; 156 conv->ostate = 0; 157 return count+1; 158 } else 159 return RET_TOOSMALL; 160 } else { 161 unsigned char buf[2]; 162 int ret; 163 164 /* Code set 1 (BIG5 extended) */ 165 ret = big5_wctomb(conv,buf,wc,2); 166 if (ret != RET_ILUNI) { 167 if (ret != 2) abort(); 168 if (!((buf[0] == 0xc6 && buf[1] >= 0xa1) || buf[0] == 0xc7)) { 169 if (n >= count+2) { 170 r[0] = buf[0]; 171 r[1] = buf[1]; 172 conv->ostate = 0; 173 return count+2; 174 } else 175 return RET_TOOSMALL; 176 } 177 } 178 ret = hkscs1999_wctomb(conv,buf,wc,2); 179 if (ret != RET_ILUNI) { 180 if (ret != 2) abort(); 181 if ((wc & ~0x0020) == 0x00ca) { 182 /* A possible first character of a multi-character sequence. We have to 183 buffer it. */ 184 if (!(buf[0] == 0x88 && (buf[1] == 0x66 || buf[1] == 0xa7))) abort(); 185 conv->ostate = buf[1]; /* = 0x66 or = 0xa7 */ 186 return count+0; 187 } 188 if (n >= count+2) { 189 r[0] = buf[0]; 190 r[1] = buf[1]; 191 conv->ostate = 0; 192 return count+2; 193 } else 194 return RET_TOOSMALL; 195 } 196 ret = hkscs2001_wctomb(conv,buf,wc,2); 197 if (ret != RET_ILUNI) { 198 if (ret != 2) abort(); 199 if (n >= count+2) { 200 r[0] = buf[0]; 201 r[1] = buf[1]; 202 conv->ostate = 0; 203 return count+2; 204 } else 205 return RET_TOOSMALL; 206 } 207 ret = hkscs2004_wctomb(conv,buf,wc,2); 208 if (ret != RET_ILUNI) { 209 if (ret != 2) abort(); 210 if (n >= count+2) { 211 r[0] = buf[0]; 212 r[1] = buf[1]; 213 conv->ostate = 0; 214 return count+2; 215 } else 216 return RET_TOOSMALL; 217 } 218 ret = hkscs2008_wctomb(conv,buf,wc,2); 219 if (ret != RET_ILUNI) { 220 if (ret != 2) abort(); 221 if (n >= count+2) { 222 r[0] = buf[0]; 223 r[1] = buf[1]; 224 conv->ostate = 0; 225 return count+2; 226 } else 227 return RET_TOOSMALL; 228 } 229 return RET_ILUNI; 230 } 231} 232 233static int 234big5hkscs2008_reset (conv_t conv, unsigned char *r, int n) 235{ 236 unsigned char last = conv->ostate; 237 238 if (last) { 239 if (n < 2) 240 return RET_TOOSMALL; 241 r[0] = 0x88; 242 r[1] = last; 243 /* conv->ostate = 0; will be done by the caller */ 244 return 2; 245 } else 246 return 0; 247} 248