utf8.cpp revision 1472:c18cbe5936b8
1/* 2 * Copyright (c) 1997, 2004, Oracle and/or its affiliates. All rights reserved. 3 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. 4 * 5 * This code is free software; you can redistribute it and/or modify it 6 * under the terms of the GNU General Public License version 2 only, as 7 * published by the Free Software Foundation. 8 * 9 * This code is distributed in the hope that it will be useful, but WITHOUT 10 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or 11 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License 12 * version 2 for more details (a copy is included in the LICENSE file that 13 * accompanied this code). 14 * 15 * You should have received a copy of the GNU General Public License version 16 * 2 along with this work; if not, write to the Free Software Foundation, 17 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. 18 * 19 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA 20 * or visit www.oracle.com if you need additional information or have any 21 * questions. 22 * 23 */ 24 25# include "incls/_precompiled.incl" 26# include "incls/_utf8.cpp.incl" 27 28// Assume the utf8 string is in legal form and has been 29// checked in the class file parser/format checker. 30char* UTF8::next(const char* str, jchar* value) { 31 unsigned const char *ptr = (const unsigned char *)str; 32 unsigned char ch, ch2, ch3; 33 int length = -1; /* bad length */ 34 jchar result; 35 switch ((ch = ptr[0]) >> 4) { 36 default: 37 result = ch; 38 length = 1; 39 break; 40 41 case 0x8: case 0x9: case 0xA: case 0xB: case 0xF: 42 /* Shouldn't happen. */ 43 break; 44 45 case 0xC: case 0xD: 46 /* 110xxxxx 10xxxxxx */ 47 if (((ch2 = ptr[1]) & 0xC0) == 0x80) { 48 unsigned char high_five = ch & 0x1F; 49 unsigned char low_six = ch2 & 0x3F; 50 result = (high_five << 6) + low_six; 51 length = 2; 52 break; 53 } 54 break; 55 56 case 0xE: 57 /* 1110xxxx 10xxxxxx 10xxxxxx */ 58 if (((ch2 = ptr[1]) & 0xC0) == 0x80) { 59 if (((ch3 = ptr[2]) & 0xC0) == 0x80) { 60 unsigned char high_four = ch & 0x0f; 61 unsigned char mid_six = ch2 & 0x3f; 62 unsigned char low_six = ch3 & 0x3f; 63 result = (((high_four << 6) + mid_six) << 6) + low_six; 64 length = 3; 65 } 66 } 67 break; 68 } /* end of switch */ 69 70 if (length <= 0) { 71 *value = ptr[0]; /* default bad result; */ 72 return (char*)(ptr + 1); // make progress somehow 73 } 74 75 *value = result; 76 77 // The assert is correct but the .class file is wrong 78 // assert(UNICODE::utf8_size(result) == length, "checking reverse computation"); 79 return (char *)(ptr + length); 80} 81 82char* UTF8::next_character(const char* str, jint* value) { 83 unsigned const char *ptr = (const unsigned char *)str; 84 /* See if it's legal supplementary character: 85 11101101 1010xxxx 10xxxxxx 11101101 1011xxxx 10xxxxxx */ 86 if (is_supplementary_character(ptr)) { 87 *value = get_supplementary_character(ptr); 88 return (char *)(ptr + 6); 89 } 90 jchar result; 91 char* next_ch = next(str, &result); 92 *value = result; 93 return next_ch; 94} 95 96// Count bytes of the form 10xxxxxx and deduct this count 97// from the total byte count. The utf8 string must be in 98// legal form which has been verified in the format checker. 99int UTF8::unicode_length(const char* str, int len) { 100 int num_chars = len; 101 for (int i = 0; i < len; i++) { 102 if ((str[i] & 0xC0) == 0x80) { 103 --num_chars; 104 } 105 } 106 return num_chars; 107} 108 109// Count bytes of the utf8 string except those in form 110// 10xxxxxx which only appear in multibyte characters. 111// The utf8 string must be in legal form and has been 112// verified in the format checker. 113int UTF8::unicode_length(const char* str) { 114 int num_chars = 0; 115 for (const char* p = str; *p; p++) { 116 if (((*p) & 0xC0) != 0x80) { 117 num_chars++; 118 } 119 } 120 return num_chars; 121} 122 123// Writes a jchar a utf8 and returns the end 124static u_char* utf8_write(u_char* base, jchar ch) { 125 if ((ch != 0) && (ch <=0x7f)) { 126 base[0] = (u_char) ch; 127 return base + 1; 128 } 129 130 if (ch <= 0x7FF) { 131 /* 11 bits or less. */ 132 unsigned char high_five = ch >> 6; 133 unsigned char low_six = ch & 0x3F; 134 base[0] = high_five | 0xC0; /* 110xxxxx */ 135 base[1] = low_six | 0x80; /* 10xxxxxx */ 136 return base + 2; 137 } 138 /* possibly full 16 bits. */ 139 char high_four = ch >> 12; 140 char mid_six = (ch >> 6) & 0x3F; 141 char low_six = ch & 0x3f; 142 base[0] = high_four | 0xE0; /* 1110xxxx */ 143 base[1] = mid_six | 0x80; /* 10xxxxxx */ 144 base[2] = low_six | 0x80; /* 10xxxxxx */ 145 return base + 3; 146} 147 148void UTF8::convert_to_unicode(const char* utf8_str, jchar* unicode_str, int unicode_length) { 149 unsigned char ch; 150 const char *ptr = (const char *)utf8_str; 151 int index = 0; 152 153 /* ASCII case loop optimization */ 154 for (; index < unicode_length; index++) { 155 if((ch = ptr[0]) > 0x7F) { break; } 156 unicode_str[index] = ch; 157 ptr = (const char *)(ptr + 1); 158 } 159 160 for (; index < unicode_length; index++) { 161 ptr = UTF8::next(ptr, &unicode_str[index]); 162 } 163} 164 165// Returns NULL if 'c' it not found. This only works as long 166// as 'c' is an ASCII character 167jbyte* UTF8::strrchr(jbyte* base, int length, jbyte c) { 168 assert(length >= 0, "sanity check"); 169 assert(c >= 0, "does not work for non-ASCII characters"); 170 // Skip backwards in string until 'c' is found or end is reached 171 while(--length >= 0 && base[length] != c); 172 return (length < 0) ? NULL : &base[length]; 173} 174 175bool UTF8::equal(jbyte* base1, int length1, jbyte* base2, int length2) { 176 // Length must be the same 177 if (length1 != length2) return false; 178 for (int i = 0; i < length1; i++) { 179 if (base1[i] != base2[i]) return false; 180 } 181 return true; 182} 183 184bool UTF8::is_supplementary_character(const unsigned char* str) { 185 return ((str[0] & 0xFF) == 0xED) && ((str[1] & 0xF0) == 0xA0) && ((str[2] & 0xC0) == 0x80) 186 && ((str[3] & 0xFF) == 0xED) && ((str[4] & 0xF0) == 0xB0) && ((str[5] & 0xC0) == 0x80); 187} 188 189jint UTF8::get_supplementary_character(const unsigned char* str) { 190 return 0x10000 + ((str[1] & 0x0f) << 16) + ((str[2] & 0x3f) << 10) 191 + ((str[4] & 0x0f) << 6) + (str[5] & 0x3f); 192} 193 194 195//------------------------------------------------------------------------------------- 196 197 198int UNICODE::utf8_size(jchar c) { 199 if ((0x0001 <= c) && (c <= 0x007F)) return 1; 200 if (c <= 0x07FF) return 2; 201 return 3; 202} 203 204int UNICODE::utf8_length(jchar* base, int length) { 205 int result = 0; 206 for (int index = 0; index < length; index++) { 207 jchar c = base[index]; 208 if ((0x0001 <= c) && (c <= 0x007F)) result += 1; 209 else if (c <= 0x07FF) result += 2; 210 else result += 3; 211 } 212 return result; 213} 214 215char* UNICODE::as_utf8(jchar* base, int length) { 216 int utf8_len = utf8_length(base, length); 217 u_char* result = NEW_RESOURCE_ARRAY(u_char, utf8_len + 1); 218 u_char* p = result; 219 for (int index = 0; index < length; index++) { 220 p = utf8_write(p, base[index]); 221 } 222 *p = '\0'; 223 assert(p == &result[utf8_len], "length prediction must be correct"); 224 return (char*) result; 225} 226 227char* UNICODE::as_utf8(jchar* base, int length, char* buf, int buflen) { 228 u_char* p = (u_char*)buf; 229 u_char* end = (u_char*)buf + buflen; 230 for (int index = 0; index < length; index++) { 231 jchar c = base[index]; 232 if (p + utf8_size(c) >= end) break; // string is truncated 233 p = utf8_write(p, base[index]); 234 } 235 *p = '\0'; 236 return buf; 237} 238 239void UNICODE::convert_to_utf8(const jchar* base, int length, char* utf8_buffer) { 240 for(int index = 0; index < length; index++) { 241 utf8_buffer = (char*)utf8_write((u_char*)utf8_buffer, base[index]); 242 } 243 *utf8_buffer = '\0'; 244} 245