Tue, 12 Feb 2013 12:19:28 -0500
8007950: Undo hs_file permission change
Summary: Reverse hs_err file permission back to 0666, as early push was premature
Reviewed-by: dsamersoff, dcubed, acorn
duke@435 | 1 | /* |
minqi@4267 | 2 | * Copyright (c) 1997, 2012, Oracle and/or its affiliates. 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 | * |
trims@1907 | 19 | * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA |
trims@1907 | 20 | * or visit www.oracle.com if you need additional information or have any |
trims@1907 | 21 | * questions. |
duke@435 | 22 | * |
duke@435 | 23 | */ |
duke@435 | 24 | |
stefank@2314 | 25 | #include "precompiled.hpp" |
stefank@2314 | 26 | #include "utilities/utf8.hpp" |
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; |
minqi@4267 | 150 | const char *ptr = 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 | |
minqi@4267 | 165 | // returns the quoted ascii length of a 0-terminated utf8 string |
minqi@4267 | 166 | int UTF8::quoted_ascii_length(const char* utf8_str, int utf8_length) { |
minqi@4267 | 167 | const char *ptr = utf8_str; |
minqi@4267 | 168 | const char* end = ptr + utf8_length; |
minqi@4267 | 169 | int result = 0; |
minqi@4267 | 170 | while (ptr < end) { |
minqi@4267 | 171 | jchar c; |
minqi@4267 | 172 | ptr = UTF8::next(ptr, &c); |
minqi@4267 | 173 | if (c >= 32 && c < 127) { |
minqi@4267 | 174 | result++; |
minqi@4267 | 175 | } else { |
minqi@4267 | 176 | result += 6; |
minqi@4267 | 177 | } |
minqi@4267 | 178 | } |
minqi@4267 | 179 | return result; |
minqi@4267 | 180 | } |
minqi@4267 | 181 | |
minqi@4267 | 182 | // converts a utf8 string to quoted ascii |
minqi@4267 | 183 | void UTF8::as_quoted_ascii(const char* utf8_str, char* buf, int buflen) { |
minqi@4267 | 184 | const char *ptr = utf8_str; |
minqi@4267 | 185 | char* p = buf; |
minqi@4267 | 186 | char* end = buf + buflen; |
minqi@4267 | 187 | while (*ptr != '\0') { |
minqi@4267 | 188 | jchar c; |
minqi@4267 | 189 | ptr = UTF8::next(ptr, &c); |
minqi@4267 | 190 | if (c >= 32 && c < 127) { |
minqi@4267 | 191 | if (p + 1 >= end) break; // string is truncated |
minqi@4267 | 192 | *p++ = (char)c; |
minqi@4267 | 193 | } else { |
minqi@4267 | 194 | if (p + 6 >= end) break; // string is truncated |
minqi@4267 | 195 | sprintf(p, "\\u%04x", c); |
minqi@4267 | 196 | p += 6; |
minqi@4267 | 197 | } |
minqi@4267 | 198 | } |
minqi@4267 | 199 | *p = '\0'; |
minqi@4267 | 200 | } |
minqi@4267 | 201 | |
minqi@4267 | 202 | |
minqi@4267 | 203 | const char* UTF8::from_quoted_ascii(const char* quoted_ascii_str) { |
minqi@4267 | 204 | const char *ptr = quoted_ascii_str; |
minqi@4267 | 205 | char* result = NULL; |
minqi@4267 | 206 | while (*ptr != '\0') { |
minqi@4267 | 207 | char c = *ptr; |
minqi@4267 | 208 | if (c < 32 || c >= 127) break; |
minqi@4267 | 209 | } |
minqi@4267 | 210 | if (*ptr == '\0') { |
minqi@4267 | 211 | // nothing to do so return original string |
minqi@4267 | 212 | return quoted_ascii_str; |
minqi@4267 | 213 | } |
minqi@4267 | 214 | // everything up to this point was ok. |
minqi@4267 | 215 | int length = ptr - quoted_ascii_str; |
minqi@4267 | 216 | char* buffer = NULL; |
minqi@4267 | 217 | for (int round = 0; round < 2; round++) { |
minqi@4267 | 218 | while (*ptr != '\0') { |
minqi@4267 | 219 | if (*ptr != '\\') { |
minqi@4267 | 220 | if (buffer != NULL) { |
minqi@4267 | 221 | buffer[length] = *ptr; |
minqi@4267 | 222 | } |
minqi@4267 | 223 | length++; |
minqi@4267 | 224 | } else { |
minqi@4267 | 225 | switch (ptr[1]) { |
minqi@4267 | 226 | case 'u': { |
minqi@4267 | 227 | ptr += 2; |
minqi@4267 | 228 | jchar value=0; |
minqi@4267 | 229 | for (int i=0; i<4; i++) { |
minqi@4267 | 230 | char c = *ptr++; |
minqi@4267 | 231 | switch (c) { |
minqi@4267 | 232 | case '0': case '1': case '2': case '3': case '4': |
minqi@4267 | 233 | case '5': case '6': case '7': case '8': case '9': |
minqi@4267 | 234 | value = (value << 4) + c - '0'; |
minqi@4267 | 235 | break; |
minqi@4267 | 236 | case 'a': case 'b': case 'c': |
minqi@4267 | 237 | case 'd': case 'e': case 'f': |
minqi@4267 | 238 | value = (value << 4) + 10 + c - 'a'; |
minqi@4267 | 239 | break; |
minqi@4267 | 240 | case 'A': case 'B': case 'C': |
minqi@4267 | 241 | case 'D': case 'E': case 'F': |
minqi@4267 | 242 | value = (value << 4) + 10 + c - 'A'; |
minqi@4267 | 243 | break; |
minqi@4267 | 244 | default: |
minqi@4267 | 245 | ShouldNotReachHere(); |
minqi@4267 | 246 | } |
minqi@4267 | 247 | } |
minqi@4267 | 248 | if (buffer == NULL) { |
minqi@4267 | 249 | char utf8_buffer[4]; |
minqi@4267 | 250 | char* next = (char*)utf8_write((u_char*)utf8_buffer, value); |
minqi@4267 | 251 | length += next - utf8_buffer; |
minqi@4267 | 252 | } else { |
minqi@4267 | 253 | char* next = (char*)utf8_write((u_char*)&buffer[length], value); |
minqi@4267 | 254 | length += next - &buffer[length]; |
minqi@4267 | 255 | } |
minqi@4267 | 256 | break; |
minqi@4267 | 257 | } |
minqi@4267 | 258 | case 't': if (buffer != NULL) buffer[length] = '\t'; ptr += 2; length++; break; |
minqi@4267 | 259 | case 'n': if (buffer != NULL) buffer[length] = '\n'; ptr += 2; length++; break; |
minqi@4267 | 260 | case 'r': if (buffer != NULL) buffer[length] = '\r'; ptr += 2; length++; break; |
minqi@4267 | 261 | case 'f': if (buffer != NULL) buffer[length] = '\f'; ptr += 2; length++; break; |
minqi@4267 | 262 | default: |
minqi@4267 | 263 | ShouldNotReachHere(); |
minqi@4267 | 264 | } |
minqi@4267 | 265 | } |
minqi@4267 | 266 | } |
minqi@4267 | 267 | if (round == 0) { |
minqi@4267 | 268 | buffer = NEW_RESOURCE_ARRAY(char, length + 1); |
minqi@4267 | 269 | ptr = quoted_ascii_str; |
minqi@4267 | 270 | } else { |
minqi@4267 | 271 | buffer[length] = '\0'; |
minqi@4267 | 272 | } |
minqi@4267 | 273 | } |
minqi@4267 | 274 | return buffer; |
minqi@4267 | 275 | } |
minqi@4267 | 276 | |
minqi@4267 | 277 | |
duke@435 | 278 | // Returns NULL if 'c' it not found. This only works as long |
duke@435 | 279 | // as 'c' is an ASCII character |
coleenp@2497 | 280 | const jbyte* UTF8::strrchr(const jbyte* base, int length, jbyte c) { |
duke@435 | 281 | assert(length >= 0, "sanity check"); |
duke@435 | 282 | assert(c >= 0, "does not work for non-ASCII characters"); |
duke@435 | 283 | // Skip backwards in string until 'c' is found or end is reached |
duke@435 | 284 | while(--length >= 0 && base[length] != c); |
duke@435 | 285 | return (length < 0) ? NULL : &base[length]; |
duke@435 | 286 | } |
duke@435 | 287 | |
coleenp@2497 | 288 | bool UTF8::equal(const jbyte* base1, int length1, const jbyte* base2, int length2) { |
duke@435 | 289 | // Length must be the same |
duke@435 | 290 | if (length1 != length2) return false; |
duke@435 | 291 | for (int i = 0; i < length1; i++) { |
duke@435 | 292 | if (base1[i] != base2[i]) return false; |
duke@435 | 293 | } |
duke@435 | 294 | return true; |
duke@435 | 295 | } |
duke@435 | 296 | |
duke@435 | 297 | bool UTF8::is_supplementary_character(const unsigned char* str) { |
duke@435 | 298 | return ((str[0] & 0xFF) == 0xED) && ((str[1] & 0xF0) == 0xA0) && ((str[2] & 0xC0) == 0x80) |
duke@435 | 299 | && ((str[3] & 0xFF) == 0xED) && ((str[4] & 0xF0) == 0xB0) && ((str[5] & 0xC0) == 0x80); |
duke@435 | 300 | } |
duke@435 | 301 | |
duke@435 | 302 | jint UTF8::get_supplementary_character(const unsigned char* str) { |
duke@435 | 303 | return 0x10000 + ((str[1] & 0x0f) << 16) + ((str[2] & 0x3f) << 10) |
duke@435 | 304 | + ((str[4] & 0x0f) << 6) + (str[5] & 0x3f); |
duke@435 | 305 | } |
duke@435 | 306 | |
duke@435 | 307 | |
duke@435 | 308 | //------------------------------------------------------------------------------------- |
duke@435 | 309 | |
duke@435 | 310 | |
duke@435 | 311 | int UNICODE::utf8_size(jchar c) { |
duke@435 | 312 | if ((0x0001 <= c) && (c <= 0x007F)) return 1; |
duke@435 | 313 | if (c <= 0x07FF) return 2; |
duke@435 | 314 | return 3; |
duke@435 | 315 | } |
duke@435 | 316 | |
duke@435 | 317 | int UNICODE::utf8_length(jchar* base, int length) { |
duke@435 | 318 | int result = 0; |
duke@435 | 319 | for (int index = 0; index < length; index++) { |
duke@435 | 320 | jchar c = base[index]; |
duke@435 | 321 | if ((0x0001 <= c) && (c <= 0x007F)) result += 1; |
duke@435 | 322 | else if (c <= 0x07FF) result += 2; |
duke@435 | 323 | else result += 3; |
duke@435 | 324 | } |
duke@435 | 325 | return result; |
duke@435 | 326 | } |
duke@435 | 327 | |
duke@435 | 328 | char* UNICODE::as_utf8(jchar* base, int length) { |
duke@435 | 329 | int utf8_len = utf8_length(base, length); |
duke@435 | 330 | u_char* result = NEW_RESOURCE_ARRAY(u_char, utf8_len + 1); |
duke@435 | 331 | u_char* p = result; |
duke@435 | 332 | for (int index = 0; index < length; index++) { |
duke@435 | 333 | p = utf8_write(p, base[index]); |
duke@435 | 334 | } |
duke@435 | 335 | *p = '\0'; |
duke@435 | 336 | assert(p == &result[utf8_len], "length prediction must be correct"); |
duke@435 | 337 | return (char*) result; |
duke@435 | 338 | } |
duke@435 | 339 | |
duke@435 | 340 | char* UNICODE::as_utf8(jchar* base, int length, char* buf, int buflen) { |
duke@435 | 341 | u_char* p = (u_char*)buf; |
duke@435 | 342 | u_char* end = (u_char*)buf + buflen; |
duke@435 | 343 | for (int index = 0; index < length; index++) { |
duke@435 | 344 | jchar c = base[index]; |
duke@435 | 345 | if (p + utf8_size(c) >= end) break; // string is truncated |
duke@435 | 346 | p = utf8_write(p, base[index]); |
duke@435 | 347 | } |
duke@435 | 348 | *p = '\0'; |
duke@435 | 349 | return buf; |
duke@435 | 350 | } |
duke@435 | 351 | |
duke@435 | 352 | void UNICODE::convert_to_utf8(const jchar* base, int length, char* utf8_buffer) { |
duke@435 | 353 | for(int index = 0; index < length; index++) { |
duke@435 | 354 | utf8_buffer = (char*)utf8_write((u_char*)utf8_buffer, base[index]); |
duke@435 | 355 | } |
duke@435 | 356 | *utf8_buffer = '\0'; |
duke@435 | 357 | } |
minqi@4267 | 358 | |
minqi@4267 | 359 | // returns the quoted ascii length of a unicode string |
minqi@4267 | 360 | int UNICODE::quoted_ascii_length(jchar* base, int length) { |
minqi@4267 | 361 | int result = 0; |
minqi@4267 | 362 | for (int i = 0; i < length; i++) { |
minqi@4267 | 363 | jchar c = base[i]; |
minqi@4267 | 364 | if (c >= 32 && c < 127) { |
minqi@4267 | 365 | result++; |
minqi@4267 | 366 | } else { |
minqi@4267 | 367 | result += 6; |
minqi@4267 | 368 | } |
minqi@4267 | 369 | } |
minqi@4267 | 370 | return result; |
minqi@4267 | 371 | } |
minqi@4267 | 372 | |
minqi@4267 | 373 | // converts a utf8 string to quoted ascii |
minqi@4267 | 374 | void UNICODE::as_quoted_ascii(const jchar* base, int length, char* buf, int buflen) { |
minqi@4267 | 375 | char* p = buf; |
minqi@4267 | 376 | char* end = buf + buflen; |
minqi@4267 | 377 | for (int index = 0; index < length; index++) { |
minqi@4267 | 378 | jchar c = base[index]; |
minqi@4267 | 379 | if (c >= 32 && c < 127) { |
minqi@4267 | 380 | if (p + 1 >= end) break; // string is truncated |
minqi@4267 | 381 | *p++ = (char)c; |
minqi@4267 | 382 | } else { |
minqi@4267 | 383 | if (p + 6 >= end) break; // string is truncated |
minqi@4267 | 384 | sprintf(p, "\\u%04x", c); |
minqi@4267 | 385 | p += 6; |
minqi@4267 | 386 | } |
minqi@4267 | 387 | } |
minqi@4267 | 388 | *p = '\0'; |
minqi@4267 | 389 | } |