1 /* 2 * Copyright (C) 1999-2002 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 * SHIFT_JIS 23 */ 24 25 /* 26 Conversion between SJIS codes (s1,s2) and JISX0208 codes (c1,c2): 27 Example. (s1,s2) = 0x8140, (c1,c2) = 0x2121. 28 0x81 <= s1 <= 0x9F || 0xE0 <= s1 <= 0xEA, 29 0x40 <= s2 <= 0x7E || 0x80 <= s2 <= 0xFC, 30 0x21 <= c1 <= 0x74, 0x21 <= c2 <= 0x7E. 31 Invariant: 32 94*2*(s1 < 0xE0 ? s1-0x81 : s1-0xC1) + (s2 < 0x80 ? s2-0x40 : s2-0x41) 33 = 94*(c1-0x21)+(c2-0x21) 34 Conversion (s1,s2) -> (c1,c2): 35 t1 := (s1 < 0xE0 ? s1-0x81 : s1-0xC1) 36 t2 := (s2 < 0x80 ? s2-0x40 : s2-0x41) 37 c1 := 2*t1 + (t2 < 0x5E ? 0 : 1) + 0x21 38 c2 := (t2 < 0x5E ? t2 : t2-0x5E) + 0x21 39 Conversion (c1,c2) -> (s1,s2): 40 t1 := (c1 - 0x21) >> 1 41 t2 := ((c1 - 0x21) & 1) * 0x5E + (c2 - 0x21) 42 s1 := (t1 < 0x1F ? t1+0x81 : t1+0xC1) 43 s2 := (t2 < 0x3F ? t2+0x40 : t2+0x41) 44 */ 45 46 static int 47 sjis_mbtowc (conv_t conv, ucs4_t *pwc, const unsigned char *s, int n) 48 { 49 unsigned char c = *s; 50 if (c < 0x80 || (c >= 0xa1 && c <= 0xdf)) 51 return jisx0201_mbtowc(conv,pwc,s,n); 52 else { 53 unsigned char s1, s2; 54 s1 = c; 55 if ((s1 >= 0x81 && s1 <= 0x9f) || (s1 >= 0xe0 && s1 <= 0xea)) { 56 if (n < 2) 57 return RET_TOOFEW(0); 58 s2 = s[1]; 59 if ((s2 >= 0x40 && s2 <= 0x7e) || (s2 >= 0x80 && s2 <= 0xfc)) { 60 unsigned char t1 = (s1 < 0xe0 ? s1-0x81 : s1-0xc1); 61 unsigned char t2 = (s2 < 0x80 ? s2-0x40 : s2-0x41); 62 unsigned char buf[2]; 63 buf[0] = 2*t1 + (t2 < 0x5e ? 0 : 1) + 0x21; 64 buf[1] = (t2 < 0x5e ? t2 : t2-0x5e) + 0x21; 65 return jisx0208_mbtowc(conv,pwc,buf,2); 66 } 67 } else if (s1 >= 0xf0 && s1 <= 0xf9) { 68 /* User-defined range. See 69 * Ken Lunde's "CJKV Information Processing", table 4-66, p. 206. */ 70 if (n < 2) 71 return RET_TOOFEW(0); 72 s2 = s[1]; 73 if ((s2 >= 0x40 && s2 <= 0x7e) || (s2 >= 0x80 && s2 <= 0xfc)) { 74 *pwc = 0xe000 + 188*(s1 - 0xf0) + (s2 < 0x80 ? s2-0x40 : s2-0x41); 75 return 2; 76 } 77 } 78 return RET_ILSEQ; 79 } 80 } 81 82 static int 83 sjis_wctomb (conv_t conv, unsigned char *r, ucs4_t wc, int n) 84 { 85 unsigned char buf[2]; 86 int ret; 87 88 /* Try JIS X 0201-1976. */ 89 ret = jisx0201_wctomb(conv,buf,wc,1); 90 if (ret != RET_ILUNI) { 91 unsigned char c; 92 if (ret != 1) abort(); 93 c = buf[0]; 94 if (c < 0x80 || (c >= 0xa1 && c <= 0xdf)) { 95 r[0] = c; 96 return 1; 97 } 98 } 99 100 /* Try JIS X 0208-1990. */ 101 ret = jisx0208_wctomb(conv,buf,wc,2); 102 if (ret != RET_ILUNI) { 103 unsigned char c1, c2; 104 if (ret != 2) abort(); 105 if (n < 2) 106 return RET_TOOSMALL; 107 c1 = buf[0]; 108 c2 = buf[1]; 109 if ((c1 >= 0x21 && c1 <= 0x74) && (c2 >= 0x21 && c2 <= 0x7e)) { 110 unsigned char t1 = (c1 - 0x21) >> 1; 111 unsigned char t2 = (((c1 - 0x21) & 1) ? 0x5e : 0) + (c2 - 0x21); 112 r[0] = (t1 < 0x1f ? t1+0x81 : t1+0xc1); 113 r[1] = (t2 < 0x3f ? t2+0x40 : t2+0x41); 114 return 2; 115 } 116 } 117 118 /* User-defined range. See 119 * Ken Lunde's "CJKV Information Processing", table 4-66, p. 206. */ 120 if (wc >= 0xe000 && wc < 0xe758) { 121 unsigned char c1, c2; 122 if (n < 2) 123 return RET_TOOSMALL; 124 c1 = (unsigned int) (wc - 0xe000) / 188; 125 c2 = (unsigned int) (wc - 0xe000) % 188; 126 r[0] = c1+0xf0; 127 r[1] = (c2 < 0x3f ? c2+0x40 : c2+0x41); 128 return 2; 129 } 130 131 return RET_ILUNI; 132 } 133