Tue, 11 May 2010 14:35:43 -0700
6931180: Migration to recent versions of MS Platform SDK
6951582: Build problems on win64
Summary: Changes to enable building JDK7 with Microsoft Visual Studio 2010
Reviewed-by: ohair, art, ccheung, dcubed
duke@435 | 1 | /* |
duke@435 | 2 | * Copyright 1997-2004 Sun Microsystems, Inc. All Rights Reserved. |
duke@435 | 3 | * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. |
duke@435 | 4 | * |
duke@435 | 5 | * This code is free software; you can redistribute it and/or modify it |
duke@435 | 6 | * under the terms of the GNU General Public License version 2 only, as |
duke@435 | 7 | * published by the Free Software Foundation. |
duke@435 | 8 | * |
duke@435 | 9 | * This code is distributed in the hope that it will be useful, but WITHOUT |
duke@435 | 10 | * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or |
duke@435 | 11 | * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License |
duke@435 | 12 | * version 2 for more details (a copy is included in the LICENSE file that |
duke@435 | 13 | * accompanied this code). |
duke@435 | 14 | * |
duke@435 | 15 | * You should have received a copy of the GNU General Public License version |
duke@435 | 16 | * 2 along with this work; if not, write to the Free Software Foundation, |
duke@435 | 17 | * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. |
duke@435 | 18 | * |
duke@435 | 19 | * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara, |
duke@435 | 20 | * CA 95054 USA or visit www.sun.com if you need additional information or |
duke@435 | 21 | * have any questions. |
duke@435 | 22 | * |
duke@435 | 23 | */ |
duke@435 | 24 | |
duke@435 | 25 | # include "incls/_precompiled.incl" |
duke@435 | 26 | # include "incls/_utf8.cpp.incl" |
duke@435 | 27 | |
duke@435 | 28 | // Assume the utf8 string is in legal form and has been |
duke@435 | 29 | // checked in the class file parser/format checker. |
duke@435 | 30 | char* UTF8::next(const char* str, jchar* value) { |
duke@435 | 31 | unsigned const char *ptr = (const unsigned char *)str; |
duke@435 | 32 | unsigned char ch, ch2, ch3; |
duke@435 | 33 | int length = -1; /* bad length */ |
duke@435 | 34 | jchar result; |
duke@435 | 35 | switch ((ch = ptr[0]) >> 4) { |
duke@435 | 36 | default: |
duke@435 | 37 | result = ch; |
duke@435 | 38 | length = 1; |
duke@435 | 39 | break; |
duke@435 | 40 | |
duke@435 | 41 | case 0x8: case 0x9: case 0xA: case 0xB: case 0xF: |
duke@435 | 42 | /* Shouldn't happen. */ |
duke@435 | 43 | break; |
duke@435 | 44 | |
duke@435 | 45 | case 0xC: case 0xD: |
duke@435 | 46 | /* 110xxxxx 10xxxxxx */ |
duke@435 | 47 | if (((ch2 = ptr[1]) & 0xC0) == 0x80) { |
duke@435 | 48 | unsigned char high_five = ch & 0x1F; |
duke@435 | 49 | unsigned char low_six = ch2 & 0x3F; |
duke@435 | 50 | result = (high_five << 6) + low_six; |
duke@435 | 51 | length = 2; |
duke@435 | 52 | break; |
duke@435 | 53 | } |
duke@435 | 54 | break; |
duke@435 | 55 | |
duke@435 | 56 | case 0xE: |
duke@435 | 57 | /* 1110xxxx 10xxxxxx 10xxxxxx */ |
duke@435 | 58 | if (((ch2 = ptr[1]) & 0xC0) == 0x80) { |
duke@435 | 59 | if (((ch3 = ptr[2]) & 0xC0) == 0x80) { |
duke@435 | 60 | unsigned char high_four = ch & 0x0f; |
duke@435 | 61 | unsigned char mid_six = ch2 & 0x3f; |
duke@435 | 62 | unsigned char low_six = ch3 & 0x3f; |
duke@435 | 63 | result = (((high_four << 6) + mid_six) << 6) + low_six; |
duke@435 | 64 | length = 3; |
duke@435 | 65 | } |
duke@435 | 66 | } |
duke@435 | 67 | break; |
duke@435 | 68 | } /* end of switch */ |
duke@435 | 69 | |
duke@435 | 70 | if (length <= 0) { |
duke@435 | 71 | *value = ptr[0]; /* default bad result; */ |
duke@435 | 72 | return (char*)(ptr + 1); // make progress somehow |
duke@435 | 73 | } |
duke@435 | 74 | |
duke@435 | 75 | *value = result; |
duke@435 | 76 | |
duke@435 | 77 | // The assert is correct but the .class file is wrong |
duke@435 | 78 | // assert(UNICODE::utf8_size(result) == length, "checking reverse computation"); |
duke@435 | 79 | return (char *)(ptr + length); |
duke@435 | 80 | } |
duke@435 | 81 | |
duke@435 | 82 | char* UTF8::next_character(const char* str, jint* value) { |
duke@435 | 83 | unsigned const char *ptr = (const unsigned char *)str; |
duke@435 | 84 | /* See if it's legal supplementary character: |
duke@435 | 85 | 11101101 1010xxxx 10xxxxxx 11101101 1011xxxx 10xxxxxx */ |
duke@435 | 86 | if (is_supplementary_character(ptr)) { |
duke@435 | 87 | *value = get_supplementary_character(ptr); |
duke@435 | 88 | return (char *)(ptr + 6); |
duke@435 | 89 | } |
duke@435 | 90 | jchar result; |
duke@435 | 91 | char* next_ch = next(str, &result); |
duke@435 | 92 | *value = result; |
duke@435 | 93 | return next_ch; |
duke@435 | 94 | } |
duke@435 | 95 | |
duke@435 | 96 | // Count bytes of the form 10xxxxxx and deduct this count |
duke@435 | 97 | // from the total byte count. The utf8 string must be in |
duke@435 | 98 | // legal form which has been verified in the format checker. |
duke@435 | 99 | int UTF8::unicode_length(const char* str, int len) { |
duke@435 | 100 | int num_chars = len; |
duke@435 | 101 | for (int i = 0; i < len; i++) { |
duke@435 | 102 | if ((str[i] & 0xC0) == 0x80) { |
duke@435 | 103 | --num_chars; |
duke@435 | 104 | } |
duke@435 | 105 | } |
duke@435 | 106 | return num_chars; |
duke@435 | 107 | } |
duke@435 | 108 | |
duke@435 | 109 | // Count bytes of the utf8 string except those in form |
duke@435 | 110 | // 10xxxxxx which only appear in multibyte characters. |
duke@435 | 111 | // The utf8 string must be in legal form and has been |
duke@435 | 112 | // verified in the format checker. |
duke@435 | 113 | int UTF8::unicode_length(const char* str) { |
duke@435 | 114 | int num_chars = 0; |
duke@435 | 115 | for (const char* p = str; *p; p++) { |
duke@435 | 116 | if (((*p) & 0xC0) != 0x80) { |
duke@435 | 117 | num_chars++; |
duke@435 | 118 | } |
duke@435 | 119 | } |
duke@435 | 120 | return num_chars; |
duke@435 | 121 | } |
duke@435 | 122 | |
duke@435 | 123 | // Writes a jchar a utf8 and returns the end |
duke@435 | 124 | static u_char* utf8_write(u_char* base, jchar ch) { |
duke@435 | 125 | if ((ch != 0) && (ch <=0x7f)) { |
duke@435 | 126 | base[0] = (u_char) ch; |
duke@435 | 127 | return base + 1; |
duke@435 | 128 | } |
duke@435 | 129 | |
duke@435 | 130 | if (ch <= 0x7FF) { |
duke@435 | 131 | /* 11 bits or less. */ |
duke@435 | 132 | unsigned char high_five = ch >> 6; |
duke@435 | 133 | unsigned char low_six = ch & 0x3F; |
duke@435 | 134 | base[0] = high_five | 0xC0; /* 110xxxxx */ |
duke@435 | 135 | base[1] = low_six | 0x80; /* 10xxxxxx */ |
duke@435 | 136 | return base + 2; |
duke@435 | 137 | } |
duke@435 | 138 | /* possibly full 16 bits. */ |
duke@435 | 139 | char high_four = ch >> 12; |
duke@435 | 140 | char mid_six = (ch >> 6) & 0x3F; |
duke@435 | 141 | char low_six = ch & 0x3f; |
duke@435 | 142 | base[0] = high_four | 0xE0; /* 1110xxxx */ |
duke@435 | 143 | base[1] = mid_six | 0x80; /* 10xxxxxx */ |
duke@435 | 144 | base[2] = low_six | 0x80; /* 10xxxxxx */ |
duke@435 | 145 | return base + 3; |
duke@435 | 146 | } |
duke@435 | 147 | |
duke@435 | 148 | void UTF8::convert_to_unicode(const char* utf8_str, jchar* unicode_str, int unicode_length) { |
duke@435 | 149 | unsigned char ch; |
duke@435 | 150 | const char *ptr = (const char *)utf8_str; |
duke@435 | 151 | int index = 0; |
duke@435 | 152 | |
duke@435 | 153 | /* ASCII case loop optimization */ |
duke@435 | 154 | for (; index < unicode_length; index++) { |
duke@435 | 155 | if((ch = ptr[0]) > 0x7F) { break; } |
duke@435 | 156 | unicode_str[index] = ch; |
duke@435 | 157 | ptr = (const char *)(ptr + 1); |
duke@435 | 158 | } |
duke@435 | 159 | |
duke@435 | 160 | for (; index < unicode_length; index++) { |
duke@435 | 161 | ptr = UTF8::next(ptr, &unicode_str[index]); |
duke@435 | 162 | } |
duke@435 | 163 | } |
duke@435 | 164 | |
duke@435 | 165 | // Returns NULL if 'c' it not found. This only works as long |
duke@435 | 166 | // as 'c' is an ASCII character |
duke@435 | 167 | jbyte* UTF8::strrchr(jbyte* base, int length, jbyte c) { |
duke@435 | 168 | assert(length >= 0, "sanity check"); |
duke@435 | 169 | assert(c >= 0, "does not work for non-ASCII characters"); |
duke@435 | 170 | // Skip backwards in string until 'c' is found or end is reached |
duke@435 | 171 | while(--length >= 0 && base[length] != c); |
duke@435 | 172 | return (length < 0) ? NULL : &base[length]; |
duke@435 | 173 | } |
duke@435 | 174 | |
duke@435 | 175 | bool UTF8::equal(jbyte* base1, int length1, jbyte* base2, int length2) { |
duke@435 | 176 | // Length must be the same |
duke@435 | 177 | if (length1 != length2) return false; |
duke@435 | 178 | for (int i = 0; i < length1; i++) { |
duke@435 | 179 | if (base1[i] != base2[i]) return false; |
duke@435 | 180 | } |
duke@435 | 181 | return true; |
duke@435 | 182 | } |
duke@435 | 183 | |
duke@435 | 184 | bool UTF8::is_supplementary_character(const unsigned char* str) { |
duke@435 | 185 | return ((str[0] & 0xFF) == 0xED) && ((str[1] & 0xF0) == 0xA0) && ((str[2] & 0xC0) == 0x80) |
duke@435 | 186 | && ((str[3] & 0xFF) == 0xED) && ((str[4] & 0xF0) == 0xB0) && ((str[5] & 0xC0) == 0x80); |
duke@435 | 187 | } |
duke@435 | 188 | |
duke@435 | 189 | jint UTF8::get_supplementary_character(const unsigned char* str) { |
duke@435 | 190 | return 0x10000 + ((str[1] & 0x0f) << 16) + ((str[2] & 0x3f) << 10) |
duke@435 | 191 | + ((str[4] & 0x0f) << 6) + (str[5] & 0x3f); |
duke@435 | 192 | } |
duke@435 | 193 | |
duke@435 | 194 | |
duke@435 | 195 | //------------------------------------------------------------------------------------- |
duke@435 | 196 | |
duke@435 | 197 | |
duke@435 | 198 | int UNICODE::utf8_size(jchar c) { |
duke@435 | 199 | if ((0x0001 <= c) && (c <= 0x007F)) return 1; |
duke@435 | 200 | if (c <= 0x07FF) return 2; |
duke@435 | 201 | return 3; |
duke@435 | 202 | } |
duke@435 | 203 | |
duke@435 | 204 | int UNICODE::utf8_length(jchar* base, int length) { |
duke@435 | 205 | int result = 0; |
duke@435 | 206 | for (int index = 0; index < length; index++) { |
duke@435 | 207 | jchar c = base[index]; |
duke@435 | 208 | if ((0x0001 <= c) && (c <= 0x007F)) result += 1; |
duke@435 | 209 | else if (c <= 0x07FF) result += 2; |
duke@435 | 210 | else result += 3; |
duke@435 | 211 | } |
duke@435 | 212 | return result; |
duke@435 | 213 | } |
duke@435 | 214 | |
duke@435 | 215 | char* UNICODE::as_utf8(jchar* base, int length) { |
duke@435 | 216 | int utf8_len = utf8_length(base, length); |
duke@435 | 217 | u_char* result = NEW_RESOURCE_ARRAY(u_char, utf8_len + 1); |
duke@435 | 218 | u_char* p = result; |
duke@435 | 219 | for (int index = 0; index < length; index++) { |
duke@435 | 220 | p = utf8_write(p, base[index]); |
duke@435 | 221 | } |
duke@435 | 222 | *p = '\0'; |
duke@435 | 223 | assert(p == &result[utf8_len], "length prediction must be correct"); |
duke@435 | 224 | return (char*) result; |
duke@435 | 225 | } |
duke@435 | 226 | |
duke@435 | 227 | char* UNICODE::as_utf8(jchar* base, int length, char* buf, int buflen) { |
duke@435 | 228 | u_char* p = (u_char*)buf; |
duke@435 | 229 | u_char* end = (u_char*)buf + buflen; |
duke@435 | 230 | for (int index = 0; index < length; index++) { |
duke@435 | 231 | jchar c = base[index]; |
duke@435 | 232 | if (p + utf8_size(c) >= end) break; // string is truncated |
duke@435 | 233 | p = utf8_write(p, base[index]); |
duke@435 | 234 | } |
duke@435 | 235 | *p = '\0'; |
duke@435 | 236 | return buf; |
duke@435 | 237 | } |
duke@435 | 238 | |
duke@435 | 239 | void UNICODE::convert_to_utf8(const jchar* base, int length, char* utf8_buffer) { |
duke@435 | 240 | for(int index = 0; index < length; index++) { |
duke@435 | 241 | utf8_buffer = (char*)utf8_write((u_char*)utf8_buffer, base[index]); |
duke@435 | 242 | } |
duke@435 | 243 | *utf8_buffer = '\0'; |
duke@435 | 244 | } |