J. Duke | 319a3b9 | 2007-12-01 00:00:00 +0000 | [diff] [blame^] | 1 | /* |
| 2 | * Copyright 2004-2005 Sun Microsystems, Inc. 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. Sun designates this |
| 8 | * particular file as subject to the "Classpath" exception as provided |
| 9 | * by Sun in the LICENSE file that accompanied this code. |
| 10 | * |
| 11 | * This code is distributed in the hope that it will be useful, but WITHOUT |
| 12 | * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or |
| 13 | * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License |
| 14 | * version 2 for more details (a copy is included in the LICENSE file that |
| 15 | * accompanied this code). |
| 16 | * |
| 17 | * You should have received a copy of the GNU General Public License version |
| 18 | * 2 along with this work; if not, write to the Free Software Foundation, |
| 19 | * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. |
| 20 | * |
| 21 | * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara, |
| 22 | * CA 95054 USA or visit www.sun.com if you need additional information or |
| 23 | * have any questions. |
| 24 | */ |
| 25 | |
| 26 | /* Misc functions for conversion of Unicode and UTF-8 and platform encoding */ |
| 27 | |
| 28 | #include <stdio.h> |
| 29 | #include <stddef.h> |
| 30 | #include <stdlib.h> |
| 31 | #include <stdarg.h> |
| 32 | #include <string.h> |
| 33 | #include <ctype.h> |
| 34 | |
| 35 | #include "jni.h" |
| 36 | |
| 37 | #include "utf.h" |
| 38 | |
| 39 | /* |
| 40 | * Error handler |
| 41 | */ |
| 42 | void |
| 43 | utfError(char *file, int line, char *message) |
| 44 | { |
| 45 | (void)fprintf(stderr, "UTF ERROR [\"%s\":%d]: %s\n", file, line, message); |
| 46 | abort(); |
| 47 | } |
| 48 | |
| 49 | /* |
| 50 | * Convert UTF-8 to UTF-16 |
| 51 | * Returns length or -1 if output overflows. |
| 52 | */ |
| 53 | int JNICALL |
| 54 | utf8ToUtf16(struct UtfInst *ui, jbyte *utf8, int len, unsigned short *output, int outputMaxLen) |
| 55 | { |
| 56 | int outputLen; |
| 57 | int i; |
| 58 | |
| 59 | UTF_ASSERT(utf8); |
| 60 | UTF_ASSERT(len>=0); |
| 61 | UTF_ASSERT(output); |
| 62 | UTF_ASSERT(outputMaxLen>0); |
| 63 | |
| 64 | i = 0; |
| 65 | outputLen = 0; |
| 66 | while ( i<len ) { |
| 67 | unsigned code, x, y, z; |
| 68 | |
| 69 | if ( outputLen >= outputMaxLen ) { |
| 70 | return -1; |
| 71 | } |
| 72 | x = (unsigned char)utf8[i++]; |
| 73 | code = x; |
| 74 | if ( (x & 0xE0)==0xE0 ) { |
| 75 | y = (unsigned char)utf8[i++]; |
| 76 | z = (unsigned char)utf8[i++]; |
| 77 | code = ((x & 0xF)<<12) + ((y & 0x3F)<<6) + (z & 0x3F); |
| 78 | } else if ( (x & 0xC0)==0xC0 ) { |
| 79 | y = (unsigned char)utf8[i++]; |
| 80 | code = ((x & 0x1F)<<6) + (y & 0x3F); |
| 81 | } |
| 82 | output[outputLen++] = code; |
| 83 | } |
| 84 | return outputLen; |
| 85 | } |
| 86 | |
| 87 | /* |
| 88 | * Convert UTF-16 to UTF-8 Modified |
| 89 | * Returns length or -1 if output overflows. |
| 90 | */ |
| 91 | int JNICALL |
| 92 | utf16ToUtf8m(struct UtfInst *ui, unsigned short *utf16, int len, jbyte *output, int outputMaxLen) |
| 93 | { |
| 94 | int i; |
| 95 | int outputLen; |
| 96 | |
| 97 | UTF_ASSERT(utf16); |
| 98 | UTF_ASSERT(len>=0); |
| 99 | UTF_ASSERT(output); |
| 100 | UTF_ASSERT(outputMaxLen>0); |
| 101 | |
| 102 | outputLen = 0; |
| 103 | for (i = 0; i < len; i++) { |
| 104 | unsigned code; |
| 105 | |
| 106 | code = utf16[i]; |
| 107 | if ( code >= 0x0001 && code <= 0x007F ) { |
| 108 | output[outputLen++] = code; |
| 109 | } else if ( code == 0 || ( code >= 0x0080 && code <= 0x07FF ) ) { |
| 110 | output[outputLen++] = ((code>>6) & 0x1F) | 0xC0; |
| 111 | output[outputLen++] = (code & 0x3F) | 0x80; |
| 112 | } else if ( code >= 0x0800 && code <= 0xFFFF ) { |
| 113 | output[outputLen++] = ((code>>12) & 0x0F) | 0xE0; |
| 114 | output[outputLen++] = ((code>>6) & 0x3F) | 0x80; |
| 115 | output[outputLen++] = (code & 0x3F) | 0x80; |
| 116 | } |
| 117 | if ( outputLen > outputMaxLen ) { |
| 118 | return -1; |
| 119 | } |
| 120 | } |
| 121 | output[outputLen] = 0; |
| 122 | return outputLen; |
| 123 | } |
| 124 | |
| 125 | int JNICALL |
| 126 | utf16ToUtf8s(struct UtfInst *ui, unsigned short *utf16, int len, jbyte *output, int outputMaxLen) |
| 127 | { |
| 128 | return -1; /* FIXUP */ |
| 129 | } |
| 130 | |
| 131 | /* Determine length of this Standard UTF-8 in Modified UTF-8. |
| 132 | * Validation is done of the basic UTF encoding rules, returns |
| 133 | * length (no change) when errors are detected in the UTF encoding. |
| 134 | * |
| 135 | * Note: Accepts Modified UTF-8 also, no verification on the |
| 136 | * correctness of Standard UTF-8 is done. e,g, 0xC080 input is ok. |
| 137 | */ |
| 138 | int JNICALL |
| 139 | utf8sToUtf8mLength(struct UtfInst *ui, jbyte *string, int length) |
| 140 | { |
| 141 | int newLength; |
| 142 | int i; |
| 143 | |
| 144 | newLength = 0; |
| 145 | for ( i = 0 ; i < length ; i++ ) { |
| 146 | unsigned byte; |
| 147 | |
| 148 | byte = (unsigned char)string[i]; |
| 149 | if ( (byte & 0x80) == 0 ) { /* 1byte encoding */ |
| 150 | newLength++; |
| 151 | if ( byte == 0 ) { |
| 152 | newLength++; /* We gain one byte in length on NULL bytes */ |
| 153 | } |
| 154 | } else if ( (byte & 0xE0) == 0xC0 ) { /* 2byte encoding */ |
| 155 | /* Check encoding of following bytes */ |
| 156 | if ( (i+1) >= length || (string[i+1] & 0xC0) != 0x80 ) { |
| 157 | break; /* Error condition */ |
| 158 | } |
| 159 | i++; /* Skip next byte */ |
| 160 | newLength += 2; |
| 161 | } else if ( (byte & 0xF0) == 0xE0 ) { /* 3byte encoding */ |
| 162 | /* Check encoding of following bytes */ |
| 163 | if ( (i+2) >= length || (string[i+1] & 0xC0) != 0x80 |
| 164 | || (string[i+2] & 0xC0) != 0x80 ) { |
| 165 | break; /* Error condition */ |
| 166 | } |
| 167 | i += 2; /* Skip next two bytes */ |
| 168 | newLength += 3; |
| 169 | } else if ( (byte & 0xF8) == 0xF0 ) { /* 4byte encoding */ |
| 170 | /* Check encoding of following bytes */ |
| 171 | if ( (i+3) >= length || (string[i+1] & 0xC0) != 0x80 |
| 172 | || (string[i+2] & 0xC0) != 0x80 |
| 173 | || (string[i+3] & 0xC0) != 0x80 ) { |
| 174 | break; /* Error condition */ |
| 175 | } |
| 176 | i += 3; /* Skip next 3 bytes */ |
| 177 | newLength += 6; /* 4byte encoding turns into 2 3byte ones */ |
| 178 | } else { |
| 179 | break; /* Error condition */ |
| 180 | } |
| 181 | } |
| 182 | if ( i != length ) { |
| 183 | /* Error in finding new length, return old length so no conversion */ |
| 184 | /* FIXUP: ERROR_MESSAGE? */ |
| 185 | return length; |
| 186 | } |
| 187 | return newLength; |
| 188 | } |
| 189 | |
| 190 | /* Convert Standard UTF-8 to Modified UTF-8. |
| 191 | * Assumes the UTF-8 encoding was validated by utf8mLength() above. |
| 192 | * |
| 193 | * Note: Accepts Modified UTF-8 also, no verification on the |
| 194 | * correctness of Standard UTF-8 is done. e,g, 0xC080 input is ok. |
| 195 | */ |
| 196 | void JNICALL |
| 197 | utf8sToUtf8m(struct UtfInst *ui, jbyte *string, int length, jbyte *newString, int newLength) |
| 198 | { |
| 199 | int i; |
| 200 | int j; |
| 201 | |
| 202 | j = 0; |
| 203 | for ( i = 0 ; i < length ; i++ ) { |
| 204 | unsigned byte1; |
| 205 | |
| 206 | byte1 = (unsigned char)string[i]; |
| 207 | |
| 208 | /* NULL bytes and bytes starting with 11110xxx are special */ |
| 209 | if ( (byte1 & 0x80) == 0 ) { /* 1byte encoding */ |
| 210 | if ( byte1 == 0 ) { |
| 211 | /* Bits out: 11000000 10000000 */ |
| 212 | newString[j++] = (jbyte)0xC0; |
| 213 | newString[j++] = (jbyte)0x80; |
| 214 | } else { |
| 215 | /* Single byte */ |
| 216 | newString[j++] = byte1; |
| 217 | } |
| 218 | } else if ( (byte1 & 0xE0) == 0xC0 ) { /* 2byte encoding */ |
| 219 | newString[j++] = byte1; |
| 220 | newString[j++] = string[++i]; |
| 221 | } else if ( (byte1 & 0xF0) == 0xE0 ) { /* 3byte encoding */ |
| 222 | newString[j++] = byte1; |
| 223 | newString[j++] = string[++i]; |
| 224 | newString[j++] = string[++i]; |
| 225 | } else if ( (byte1 & 0xF8) == 0xF0 ) { /* 4byte encoding */ |
| 226 | /* Beginning of 4byte encoding, turn into 2 3byte encodings */ |
| 227 | unsigned byte2, byte3, byte4, u21; |
| 228 | |
| 229 | /* Bits in: 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx */ |
| 230 | byte2 = (unsigned char)string[++i]; |
| 231 | byte3 = (unsigned char)string[++i]; |
| 232 | byte4 = (unsigned char)string[++i]; |
| 233 | /* Reconstruct full 21bit value */ |
| 234 | u21 = (byte1 & 0x07) << 18; |
| 235 | u21 += (byte2 & 0x3F) << 12; |
| 236 | u21 += (byte3 & 0x3F) << 6; |
| 237 | u21 += (byte4 & 0x3F); |
| 238 | /* Bits out: 11101101 1010xxxx 10xxxxxx */ |
| 239 | newString[j++] = (jbyte)0xED; |
| 240 | newString[j++] = (jbyte)(0xA0 + (((u21 >> 16) - 1) & 0x0F)); |
| 241 | newString[j++] = (jbyte)(0x80 + ((u21 >> 10) & 0x3F)); |
| 242 | /* Bits out: 11101101 1011xxxx 10xxxxxx */ |
| 243 | newString[j++] = (jbyte)0xED; |
| 244 | newString[j++] = (jbyte)(0xB0 + ((u21 >> 6) & 0x0F)); |
| 245 | newString[j++] = byte4; |
| 246 | } |
| 247 | } |
| 248 | UTF_ASSERT(i==length); |
| 249 | UTF_ASSERT(j==newLength); |
| 250 | newString[j] = (jbyte)0; |
| 251 | } |
| 252 | |
| 253 | /* Given a Modified UTF-8 string, calculate the Standard UTF-8 length. |
| 254 | * Basic validation of the UTF encoding rules is done, and length is |
| 255 | * returned (no change) when errors are detected. |
| 256 | * |
| 257 | * Note: No validation is made that this is indeed Modified UTF-8 coming in. |
| 258 | * |
| 259 | */ |
| 260 | int JNICALL |
| 261 | utf8mToUtf8sLength(struct UtfInst *ui, jbyte *string, int length) |
| 262 | { |
| 263 | int newLength; |
| 264 | int i; |
| 265 | |
| 266 | newLength = 0; |
| 267 | for ( i = 0 ; i < length ; i++ ) { |
| 268 | unsigned byte1, byte2, byte3, byte4, byte5, byte6; |
| 269 | |
| 270 | byte1 = (unsigned char)string[i]; |
| 271 | if ( (byte1 & 0x80) == 0 ) { /* 1byte encoding */ |
| 272 | newLength++; |
| 273 | } else if ( (byte1 & 0xE0) == 0xC0 ) { /* 2byte encoding */ |
| 274 | /* Check encoding of following bytes */ |
| 275 | if ( (i+1) >= length || (string[i+1] & 0xC0) != 0x80 ) { |
| 276 | break; /* Error condition */ |
| 277 | } |
| 278 | byte2 = (unsigned char)string[++i]; |
| 279 | if ( byte1 != 0xC0 || byte2 != 0x80 ) { |
| 280 | newLength += 2; /* Normal 2byte encoding, not 0xC080 */ |
| 281 | } else { |
| 282 | newLength++; /* We will turn 0xC080 into 0 */ |
| 283 | } |
| 284 | } else if ( (byte1 & 0xF0) == 0xE0 ) { /* 3byte encoding */ |
| 285 | /* Check encoding of following bytes */ |
| 286 | if ( (i+2) >= length || (string[i+1] & 0xC0) != 0x80 |
| 287 | || (string[i+2] & 0xC0) != 0x80 ) { |
| 288 | break; /* Error condition */ |
| 289 | } |
| 290 | byte2 = (unsigned char)string[++i]; |
| 291 | byte3 = (unsigned char)string[++i]; |
| 292 | newLength += 3; |
| 293 | /* Possible process a second 3byte encoding */ |
| 294 | if ( (i+3) < length && byte1 == 0xED && (byte2 & 0xF0) == 0xA0 ) { |
| 295 | /* See if this is a pair of 3byte encodings */ |
| 296 | byte4 = (unsigned char)string[i+1]; |
| 297 | byte5 = (unsigned char)string[i+2]; |
| 298 | byte6 = (unsigned char)string[i+3]; |
| 299 | if ( byte4 == 0xED && (byte5 & 0xF0) == 0xB0 ) { |
| 300 | /* Check encoding of 3rd byte */ |
| 301 | if ( (byte6 & 0xC0) != 0x80 ) { |
| 302 | break; /* Error condition */ |
| 303 | } |
| 304 | newLength++; /* New string will have 4byte encoding */ |
| 305 | i += 3; /* Skip next 3 bytes */ |
| 306 | } |
| 307 | } |
| 308 | } else { |
| 309 | break; /* Error condition */ |
| 310 | } |
| 311 | } |
| 312 | if ( i != length ) { |
| 313 | /* Error in UTF encoding */ |
| 314 | /* FIXUP: ERROR_MESSAGE()? */ |
| 315 | return length; |
| 316 | } |
| 317 | return newLength; |
| 318 | } |
| 319 | |
| 320 | /* Convert a Modified UTF-8 string into a Standard UTF-8 string |
| 321 | * It is assumed that this string has been validated in terms of the |
| 322 | * basic UTF encoding rules by utf8Length() above. |
| 323 | * |
| 324 | * Note: No validation is made that this is indeed Modified UTF-8 coming in. |
| 325 | * |
| 326 | */ |
| 327 | void JNICALL |
| 328 | utf8mToUtf8s(struct UtfInst *ui, jbyte *string, int length, jbyte *newString, int newLength) |
| 329 | { |
| 330 | int i; |
| 331 | int j; |
| 332 | |
| 333 | j = 0; |
| 334 | for ( i = 0 ; i < length ; i++ ) { |
| 335 | unsigned byte1, byte2, byte3, byte4, byte5, byte6; |
| 336 | |
| 337 | byte1 = (unsigned char)string[i]; |
| 338 | if ( (byte1 & 0x80) == 0 ) { /* 1byte encoding */ |
| 339 | /* Single byte */ |
| 340 | newString[j++] = byte1; |
| 341 | } else if ( (byte1 & 0xE0) == 0xC0 ) { /* 2byte encoding */ |
| 342 | byte2 = (unsigned char)string[++i]; |
| 343 | if ( byte1 != 0xC0 || byte2 != 0x80 ) { |
| 344 | newString[j++] = byte1; |
| 345 | newString[j++] = byte2; |
| 346 | } else { |
| 347 | newString[j++] = 0; |
| 348 | } |
| 349 | } else if ( (byte1 & 0xF0) == 0xE0 ) { /* 3byte encoding */ |
| 350 | byte2 = (unsigned char)string[++i]; |
| 351 | byte3 = (unsigned char)string[++i]; |
| 352 | if ( i+3 < length && byte1 == 0xED && (byte2 & 0xF0) == 0xA0 ) { |
| 353 | /* See if this is a pair of 3byte encodings */ |
| 354 | byte4 = (unsigned char)string[i+1]; |
| 355 | byte5 = (unsigned char)string[i+2]; |
| 356 | byte6 = (unsigned char)string[i+3]; |
| 357 | if ( byte4 == 0xED && (byte5 & 0xF0) == 0xB0 ) { |
| 358 | unsigned u21; |
| 359 | |
| 360 | /* Bits in: 11101101 1010xxxx 10xxxxxx */ |
| 361 | /* Bits in: 11101101 1011xxxx 10xxxxxx */ |
| 362 | i += 3; |
| 363 | |
| 364 | /* Reconstruct 21 bit code */ |
| 365 | u21 = ((byte2 & 0x0F) + 1) << 16; |
| 366 | u21 += (byte3 & 0x3F) << 10; |
| 367 | u21 += (byte5 & 0x0F) << 6; |
| 368 | u21 += (byte6 & 0x3F); |
| 369 | |
| 370 | /* Bits out: 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx */ |
| 371 | |
| 372 | /* Convert to 4byte encoding */ |
| 373 | newString[j++] = 0xF0 + ((u21 >> 18) & 0x07); |
| 374 | newString[j++] = 0x80 + ((u21 >> 12) & 0x3F); |
| 375 | newString[j++] = 0x80 + ((u21 >> 6) & 0x3F); |
| 376 | newString[j++] = 0x80 + (u21 & 0x3F); |
| 377 | continue; |
| 378 | } |
| 379 | } |
| 380 | /* Normal 3byte encoding */ |
| 381 | newString[j++] = byte1; |
| 382 | newString[j++] = byte2; |
| 383 | newString[j++] = byte3; |
| 384 | } |
| 385 | } |
| 386 | UTF_ASSERT(i==length); |
| 387 | UTF_ASSERT(j==newLength); |
| 388 | newString[j] = 0; |
| 389 | } |
| 390 | |
| 391 | /* ================================================================= */ |
| 392 | |
| 393 | #if 1 /* Test program */ |
| 394 | |
| 395 | /* |
| 396 | * Convert any byte array into a printable string. |
| 397 | * Returns length or -1 if output overflows. |
| 398 | */ |
| 399 | static int |
| 400 | bytesToPrintable(struct UtfInst *ui, char *bytes, int len, char *output, int outputMaxLen) |
| 401 | { |
| 402 | int outputLen; |
| 403 | int i; |
| 404 | |
| 405 | UTF_ASSERT(bytes); |
| 406 | UTF_ASSERT(len>=0); |
| 407 | UTF_ASSERT(output); |
| 408 | UTF_ASSERT(outputMaxLen>=0); |
| 409 | |
| 410 | outputLen = 0; |
| 411 | for ( i=0; i<len ; i++ ) { |
| 412 | unsigned byte; |
| 413 | |
| 414 | byte = bytes[i]; |
| 415 | if ( outputLen >= outputMaxLen ) { |
| 416 | return -1; |
| 417 | } |
| 418 | if ( byte <= 0x7f && isprint(byte) && !iscntrl(byte) ) { |
| 419 | output[outputLen++] = (char)byte; |
| 420 | } else { |
| 421 | (void)sprintf(output+outputLen,"\\x%02x",byte); |
| 422 | outputLen += 4; |
| 423 | } |
| 424 | } |
| 425 | output[outputLen] = 0; |
| 426 | return outputLen; |
| 427 | } |
| 428 | |
| 429 | static void |
| 430 | test(void) |
| 431 | { |
| 432 | static char *strings[] = { |
| 433 | "characters", |
| 434 | "abcdefghijklmnopqrstuvwxyz", |
| 435 | "0123456789", |
| 436 | "!@#$%^&*()_+=-{}[]:;", |
| 437 | NULL }; |
| 438 | int i; |
| 439 | struct UtfInst *ui; |
| 440 | |
| 441 | ui = utfInitialize(NULL); |
| 442 | |
| 443 | i = 0; |
| 444 | while ( strings[i] != NULL ) { |
| 445 | char *str; |
| 446 | #define MAX 1024 |
| 447 | char buf0[MAX]; |
| 448 | char buf1[MAX]; |
| 449 | char buf2[MAX]; |
| 450 | unsigned short buf3[MAX]; |
| 451 | int len1; |
| 452 | int len2; |
| 453 | int len3; |
| 454 | |
| 455 | str = strings[i]; |
| 456 | |
| 457 | (void)bytesToPrintable(ui, str, (int)strlen(str), buf0, 1024); |
| 458 | |
| 459 | len1 = utf8FromPlatform(ui, str, (int)strlen(str), (jbyte*)buf1, 1024); |
| 460 | |
| 461 | UTF_ASSERT(len1==(int)strlen(str)); |
| 462 | |
| 463 | len3 = utf8ToUtf16(ui, (jbyte*)buf1, len1, (jchar*)buf3, 1024); |
| 464 | |
| 465 | UTF_ASSERT(len3==len1); |
| 466 | |
| 467 | len1 = utf16ToUtf8m(ui, (jchar*)buf3, len3, (jbyte*)buf1, 1024); |
| 468 | |
| 469 | UTF_ASSERT(len1==len3); |
| 470 | UTF_ASSERT(strcmp(str, buf1) == 0); |
| 471 | |
| 472 | len2 = utf8ToPlatform(ui, (jbyte*)buf1, len1, buf2, 1024); |
| 473 | |
| 474 | UTF_ASSERT(len2==len1); |
| 475 | UTF_ASSERT(strcmp(str, buf2) == 0); |
| 476 | |
| 477 | i++; |
| 478 | } |
| 479 | |
| 480 | utfTerminate(ui, NULL); |
| 481 | |
| 482 | } |
| 483 | |
| 484 | int |
| 485 | main(int argc, char **argv) |
| 486 | { |
| 487 | test(); |
| 488 | return 0; |
| 489 | } |
| 490 | |
| 491 | #endif |