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:2001 23 */ 24 25 /* 26 * BIG5-HKSCS:2001 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:1999 through 116 characters. 31 * 32 * It extends BIG5 (without the rows 0xC6..0xC7) through the ranges 33 * 34 * 0x{88..8D}{40..7E,A1..FE} 757 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 "hkscs2001.h" 49 #include "flushwc.h" 50 51 static int 52 big5hkscs2001_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 if (c == 0x88) { 91 if (n < 2) 92 return RET_TOOFEW(0); 93 { 94 unsigned char c2 = s[1]; 95 if (c2 == 0x62 || c2 == 0x64 || c2 == 0xa3 || c2 == 0xa5) { 96 /* It's a composed character. */ 97 ucs4_t wc1 = ((c2 >> 3) << 2) + 0x009a; /* = 0x00ca or 0x00ea */ 98 ucs4_t wc2 = ((c2 & 6) << 2) + 0x02fc; /* = 0x0304 or 0x030c */ 99 /* We cannot output two Unicode characters at once. So, 100 output the first character and buffer the second one. */ 101 *pwc = wc1; 102 conv->istate = wc2; 103 return 2; 104 } 105 } 106 } 107 return RET_ILSEQ; 108 } 109 } 110 111 #define big5hkscs2001_flushwc normal_flushwc 112 113 static int 114 big5hkscs2001_wctomb (conv_t conv, unsigned char *r, ucs4_t wc, int n) 115 { 116 int count = 0; 117 unsigned char last = conv->ostate; 118 119 if (last) { 120 /* last is = 0x66 or = 0xa7. */ 121 if (wc == 0x0304 || wc == 0x030c) { 122 /* Output the combined character. */ 123 if (n >= 2) { 124 r[0] = 0x88; 125 r[1] = last + ((wc & 24) >> 2) - 4; /* = 0x62 or 0x64 or 0xa3 or 0xa5 */ 126 conv->ostate = 0; 127 return 2; 128 } else 129 return RET_TOOSMALL; 130 } 131 132 /* Output the buffered character. */ 133 if (n < 2) 134 return RET_TOOSMALL; 135 r[0] = 0x88; 136 r[1] = last; 137 r += 2; 138 count = 2; 139 } 140 141 /* Code set 0 (ASCII) */ 142 if (wc < 0x0080) { 143 /* Plain ASCII character. */ 144 if (n > count) { 145 r[0] = (unsigned char) wc; 146 conv->ostate = 0; 147 return count+1; 148 } else 149 return RET_TOOSMALL; 150 } else { 151 unsigned char buf[2]; 152 int ret; 153 154 /* Code set 1 (BIG5 extended) */ 155 ret = big5_wctomb(conv,buf,wc,2); 156 if (ret != RET_ILUNI) { 157 if (ret != 2) abort(); 158 if (!((buf[0] == 0xc6 && buf[1] >= 0xa1) || buf[0] == 0xc7)) { 159 if (n >= count+2) { 160 r[0] = buf[0]; 161 r[1] = buf[1]; 162 conv->ostate = 0; 163 return count+2; 164 } else 165 return RET_TOOSMALL; 166 } 167 } 168 ret = hkscs1999_wctomb(conv,buf,wc,2); 169 if (ret != RET_ILUNI) { 170 if (ret != 2) abort(); 171 if ((wc & ~0x0020) == 0x00ca) { 172 /* A possible first character of a multi-character sequence. We have to 173 buffer it. */ 174 if (!(buf[0] == 0x88 && (buf[1] == 0x66 || buf[1] == 0xa7))) abort(); 175 conv->ostate = buf[1]; /* = 0x66 or = 0xa7 */ 176 return count+0; 177 } 178 if (n >= count+2) { 179 r[0] = buf[0]; 180 r[1] = buf[1]; 181 conv->ostate = 0; 182 return count+2; 183 } else 184 return RET_TOOSMALL; 185 } 186 ret = hkscs2001_wctomb(conv,buf,wc,2); 187 if (ret != RET_ILUNI) { 188 if (ret != 2) abort(); 189 if (n >= count+2) { 190 r[0] = buf[0]; 191 r[1] = buf[1]; 192 conv->ostate = 0; 193 return count+2; 194 } else 195 return RET_TOOSMALL; 196 } 197 return RET_ILUNI; 198 } 199 } 200 201 static int 202 big5hkscs2001_reset (conv_t conv, unsigned char *r, int n) 203 { 204 unsigned char last = conv->ostate; 205 206 if (last) { 207 if (n < 2) 208 return RET_TOOSMALL; 209 r[0] = 0x88; 210 r[1] = last; 211 /* conv->ostate = 0; will be done by the caller */ 212 return 2; 213 } else 214 return 0; 215 } 216