1/* Look at first character in UTF-8 string. 2 Copyright (C) 1999-2002, 2006-2007 Free Software Foundation, Inc. 3 Written by Bruno Haible <bruno@clisp.org>, 2001. 4 5 This program is free software: you can redistribute it and/or modify it 6 under the terms of the GNU General Public License as published 7 by the Free Software Foundation; either version 3 of the License, or 8 (at your option) any later version. 9 10 This program is distributed in the hope that it will be useful, 11 but WITHOUT ANY WARRANTY; without even the implied warranty of 12 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU 13 Lesser General Public License for more details. 14 15 You should have received a copy of the GNU General Public License 16 along with this program. If not, see <http://www.gnu.org/licenses/>. */ 17 18#include <config.h> 19 20/* Specification. */ 21#include "unistr.h" 22 23#if !HAVE_INLINE 24 25int 26u8_mbtouc (ucs4_t *puc, const uint8_t *s, size_t n) 27{ 28 uint8_t c = *s; 29 30 if (c < 0x80) 31 { 32 *puc = c; 33 return 1; 34 } 35 else if (c >= 0xc2) 36 { 37 if (c < 0xe0) 38 { 39 if (n >= 2) 40 { 41 if ((s[1] ^ 0x80) < 0x40) 42 { 43 *puc = ((unsigned int) (c & 0x1f) << 6) 44 | (unsigned int) (s[1] ^ 0x80); 45 return 2; 46 } 47 /* invalid multibyte character */ 48 } 49 else 50 { 51 /* incomplete multibyte character */ 52 *puc = 0xfffd; 53 return n; 54 } 55 } 56 else if (c < 0xf0) 57 { 58 if (n >= 3) 59 { 60 if ((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40 61 && (c >= 0xe1 || s[1] >= 0xa0) 62 && (c != 0xed || s[1] < 0xa0)) 63 { 64 *puc = ((unsigned int) (c & 0x0f) << 12) 65 | ((unsigned int) (s[1] ^ 0x80) << 6) 66 | (unsigned int) (s[2] ^ 0x80); 67 return 3; 68 } 69 /* invalid multibyte character */ 70 } 71 else 72 { 73 /* incomplete multibyte character */ 74 *puc = 0xfffd; 75 return n; 76 } 77 } 78 else if (c < 0xf8) 79 { 80 if (n >= 4) 81 { 82 if ((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40 83 && (s[3] ^ 0x80) < 0x40 84 && (c >= 0xf1 || s[1] >= 0x90) 85#if 1 86 && (c < 0xf4 || (c == 0xf4 && s[1] < 0x90)) 87#endif 88 ) 89 { 90 *puc = ((unsigned int) (c & 0x07) << 18) 91 | ((unsigned int) (s[1] ^ 0x80) << 12) 92 | ((unsigned int) (s[2] ^ 0x80) << 6) 93 | (unsigned int) (s[3] ^ 0x80); 94 return 4; 95 } 96 /* invalid multibyte character */ 97 } 98 else 99 { 100 /* incomplete multibyte character */ 101 *puc = 0xfffd; 102 return n; 103 } 104 } 105#if 0 106 else if (c < 0xfc) 107 { 108 if (n >= 5) 109 { 110 if ((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40 111 && (s[3] ^ 0x80) < 0x40 && (s[4] ^ 0x80) < 0x40 112 && (c >= 0xf9 || s[1] >= 0x88)) 113 { 114 *puc = ((unsigned int) (c & 0x03) << 24) 115 | ((unsigned int) (s[1] ^ 0x80) << 18) 116 | ((unsigned int) (s[2] ^ 0x80) << 12) 117 | ((unsigned int) (s[3] ^ 0x80) << 6) 118 | (unsigned int) (s[4] ^ 0x80); 119 return 5; 120 } 121 /* invalid multibyte character */ 122 } 123 else 124 { 125 /* incomplete multibyte character */ 126 *puc = 0xfffd; 127 return n; 128 } 129 } 130 else if (c < 0xfe) 131 { 132 if (n >= 6) 133 { 134 if ((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40 135 && (s[3] ^ 0x80) < 0x40 && (s[4] ^ 0x80) < 0x40 136 && (s[5] ^ 0x80) < 0x40 137 && (c >= 0xfd || s[1] >= 0x84)) 138 { 139 *puc = ((unsigned int) (c & 0x01) << 30) 140 | ((unsigned int) (s[1] ^ 0x80) << 24) 141 | ((unsigned int) (s[2] ^ 0x80) << 18) 142 | ((unsigned int) (s[3] ^ 0x80) << 12) 143 | ((unsigned int) (s[4] ^ 0x80) << 6) 144 | (unsigned int) (s[5] ^ 0x80); 145 return 6; 146 } 147 /* invalid multibyte character */ 148 } 149 else 150 { 151 /* incomplete multibyte character */ 152 *puc = 0xfffd; 153 return n; 154 } 155 } 156#endif 157 } 158 /* invalid multibyte character */ 159 *puc = 0xfffd; 160 return 1; 161} 162 163#endif 164