The Android Open Source Project | 8b23a6c | 2009-03-03 19:30:32 -0800 | [diff] [blame] | 1 | /* Copyright (C) 2007-2008 The Android Open Source Project |
| 2 | ** |
| 3 | ** This software is licensed under the terms of the GNU General Public |
| 4 | ** License version 2, as published by the Free Software Foundation, and |
| 5 | ** may be copied, distributed, and modified under those terms. |
| 6 | ** |
| 7 | ** This program is distributed in the hope that it will be useful, |
| 8 | ** but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 9 | ** MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 10 | ** GNU General Public License for more details. |
| 11 | */ |
| 12 | #include "gsm.h" |
| 13 | #include <stdlib.h> |
| 14 | #include <string.h> |
| 15 | |
| 16 | /** UTILITIES |
| 17 | **/ |
| 18 | byte_t |
| 19 | gsm_int_to_bcdi( int value ) |
| 20 | { |
| 21 | return (byte_t)((value / 10) | ((value % 10) << 4)); |
| 22 | } |
| 23 | |
| 24 | int |
| 25 | gsm_int_from_bcdi( byte_t val ) |
| 26 | { |
| 27 | int ret = 0; |
| 28 | |
| 29 | if ((val & 0xf0) <= 0x90) |
| 30 | ret = (val >> 4); |
| 31 | |
| 32 | if ((val & 0x0f) <= 0x90) |
| 33 | ret |= (val % 0xf)*10; |
| 34 | |
| 35 | return ret; |
| 36 | } |
| 37 | |
| 38 | #if 0 |
| 39 | static int |
| 40 | gsm_bcdi_to_ascii( cbytes_t bcd, int bcdlen, bytes_t dst ) |
| 41 | { |
| 42 | static byte_t bcdichars[14] = "0123456789*#,N"; |
| 43 | |
| 44 | int result = 0; |
| 45 | int shift = 0; |
| 46 | |
| 47 | while (bcdlen > 0) { |
| 48 | int c = (bcd[0] >> shift) & 0xf; |
| 49 | |
| 50 | if (c == 0xf && bcdlen == 1) |
| 51 | break; |
| 52 | |
| 53 | if (c < 14) { |
| 54 | if (dst) dst[result] = bcdichars[c]; |
| 55 | result += 1; |
| 56 | } |
| 57 | bcdlen --; |
| 58 | shift += 4; |
| 59 | if (shift == 8) { |
| 60 | bcd++; |
| 61 | shift = 0; |
| 62 | } |
| 63 | } |
| 64 | return result; |
| 65 | } |
| 66 | #endif |
| 67 | |
| 68 | #if 0 |
| 69 | static int |
| 70 | gsm_bcdi_from_ascii( cbytes_t ascii, int asciilen, bytes_t dst ) |
| 71 | { |
| 72 | cbytes_t end = ascii + asciilen; |
| 73 | int result = 0; |
| 74 | int phase = 0x01; |
| 75 | |
| 76 | while (ascii < end) { |
| 77 | int c = *ascii++; |
| 78 | |
| 79 | if (c == '*') |
| 80 | c = 11; |
| 81 | else if (c == '#') |
| 82 | c = 12; |
| 83 | else if (c == ',') |
| 84 | c = 13; |
| 85 | else if (c == 'N') |
| 86 | c = 14; |
| 87 | else { |
| 88 | c -= '0'; |
| 89 | if ((unsigned)c >= 10) |
| 90 | break; |
| 91 | } |
| 92 | phase = (phase << 4) | c; |
| 93 | if (phase & 0x100) { |
| 94 | if (dst) dst[result] = (byte_t) phase; |
| 95 | result += 1; |
| 96 | phase = 0x01; |
| 97 | } |
| 98 | } |
| 99 | if (phase != 0x01) { |
| 100 | if (dst) dst[result] = (byte_t)( phase | 0xf0 ); |
| 101 | result += 1; |
| 102 | } |
| 103 | return result; |
| 104 | } |
| 105 | #endif |
| 106 | |
| 107 | int |
| 108 | gsm_hexchar_to_int( char c ) |
| 109 | { |
| 110 | if ((unsigned)(c - '0') < 10) |
| 111 | return c - '0'; |
| 112 | if ((unsigned)(c - 'a') < 6) |
| 113 | return 10 + (c - 'a'); |
| 114 | if ((unsigned)(c - 'A') < 6) |
| 115 | return 10 + (c - 'A'); |
| 116 | return -1; |
| 117 | } |
| 118 | |
| 119 | int |
| 120 | gsm_hexchar_to_int0( char c ) |
| 121 | { |
| 122 | int ret = gsm_hexchar_to_int(c); |
| 123 | |
| 124 | return (ret < 0) ? 0 : ret; |
| 125 | } |
| 126 | |
| 127 | int |
| 128 | gsm_hex2_to_byte( const char* hex ) |
| 129 | { |
| 130 | int hi = gsm_hexchar_to_int(hex[0]); |
| 131 | int lo = gsm_hexchar_to_int(hex[1]); |
| 132 | |
| 133 | if (hi < 0 || lo < 0) |
| 134 | return -1; |
| 135 | |
| 136 | return ( (hi << 4) | lo ); |
| 137 | } |
| 138 | |
| 139 | int |
| 140 | gsm_hex4_to_short( const char* hex ) |
| 141 | { |
| 142 | int hi = gsm_hex2_to_byte(hex); |
| 143 | int lo = gsm_hex2_to_byte(hex+2); |
| 144 | |
| 145 | if (hi < 0 || lo < 0) |
| 146 | return -1; |
| 147 | |
| 148 | return ((hi << 8) | lo); |
| 149 | } |
| 150 | |
| 151 | int |
| 152 | gsm_hex2_to_byte0( const char* hex ) |
| 153 | { |
| 154 | int hi = gsm_hexchar_to_int0(hex[0]); |
| 155 | int lo = gsm_hexchar_to_int0(hex[1]); |
| 156 | |
| 157 | return (byte_t)( (hi << 4) | lo ); |
| 158 | } |
| 159 | |
| 160 | void |
| 161 | gsm_hex_from_byte( char* hex, int val ) |
| 162 | { |
| 163 | static const char hexdigits[] = "0123456789abcdef"; |
| 164 | |
| 165 | hex[0] = hexdigits[(val >> 4) & 15]; |
| 166 | hex[1] = hexdigits[val & 15]; |
| 167 | } |
| 168 | |
| 169 | void |
| 170 | gsm_hex_from_short( char* hex, int val ) |
| 171 | { |
| 172 | gsm_hex_from_byte( hex, (val >> 8) ); |
| 173 | gsm_hex_from_byte( hex+2, val ); |
| 174 | } |
| 175 | |
| 176 | |
| 177 | |
| 178 | /** HEX |
| 179 | **/ |
| 180 | void |
| 181 | gsm_hex_to_bytes0( cbytes_t hex, int hexlen, bytes_t dst ) |
| 182 | { |
| 183 | int nn; |
| 184 | |
| 185 | for (nn = 0; nn < hexlen/2; nn++ ) { |
| 186 | dst[nn] = (byte_t) gsm_hex2_to_byte0( (const char*)hex+2*nn ); |
| 187 | } |
| 188 | if (hexlen & 1) { |
| 189 | dst[nn] = gsm_hexchar_to_int0( hex[2*nn] ) << 4; |
| 190 | } |
| 191 | } |
| 192 | |
| 193 | int |
| 194 | gsm_hex_to_bytes( cbytes_t hex, int hexlen, bytes_t dst ) |
| 195 | { |
| 196 | int nn; |
| 197 | |
| 198 | if (hexlen & 1) /* must be even */ |
| 199 | return -1; |
| 200 | |
| 201 | for (nn = 0; nn < hexlen/2; nn++ ) { |
| 202 | int c = gsm_hex2_to_byte( (const char*)hex+2*nn ); |
| 203 | if (c < 0) return -1; |
| 204 | dst[nn] = (byte_t) c; |
| 205 | } |
| 206 | return hexlen/2; |
| 207 | } |
| 208 | |
| 209 | void |
| 210 | gsm_hex_from_bytes( char* hex, cbytes_t src, int srclen ) |
| 211 | { |
| 212 | int nn; |
| 213 | |
| 214 | for (nn = 0; nn < srclen; nn++) { |
| 215 | gsm_hex_from_byte( hex + 2*nn, src[nn] ); |
| 216 | } |
| 217 | } |
| 218 | |
| 219 | /** ROPES |
| 220 | **/ |
| 221 | |
| 222 | void |
| 223 | gsm_rope_init( GsmRope rope ) |
| 224 | { |
| 225 | rope->data = NULL; |
| 226 | rope->pos = 0; |
| 227 | rope->max = 0; |
| 228 | rope->error = 0; |
| 229 | } |
| 230 | |
| 231 | void |
| 232 | gsm_rope_init_alloc( GsmRope rope, int count ) |
| 233 | { |
| 234 | rope->data = rope->data0; |
| 235 | rope->pos = 0; |
| 236 | rope->max = sizeof(rope->data0); |
| 237 | rope->error = 0; |
| 238 | |
| 239 | if (count > 0) { |
| 240 | rope->data = calloc( count, 1 ); |
| 241 | rope->max = count; |
| 242 | |
| 243 | if (rope->data == NULL) { |
| 244 | rope->error = 1; |
| 245 | rope->max = 0; |
| 246 | } |
| 247 | } |
| 248 | } |
| 249 | |
| 250 | int |
| 251 | gsm_rope_done( GsmRope rope ) |
| 252 | { |
| 253 | int result = rope->error; |
| 254 | |
| 255 | if (rope->data && rope->data != rope->data0) |
| 256 | free(rope->data); |
| 257 | |
| 258 | rope->data = NULL; |
| 259 | rope->pos = 0; |
| 260 | rope->max = 0; |
| 261 | rope->error = 0; |
| 262 | |
| 263 | return result; |
| 264 | } |
| 265 | |
| 266 | |
| 267 | bytes_t |
| 268 | gsm_rope_done_acquire( GsmRope rope, int *psize ) |
| 269 | { |
| 270 | bytes_t result = rope->data; |
| 271 | |
| 272 | *psize = rope->pos; |
| 273 | if (result == rope->data0) { |
| 274 | result = malloc( rope->pos ); |
| 275 | if (result != NULL) |
| 276 | memcpy( result, rope->data, rope->pos ); |
| 277 | } |
| 278 | return result; |
| 279 | } |
| 280 | |
| 281 | |
| 282 | int |
| 283 | gsm_rope_ensure( GsmRope rope, int new_count ) |
| 284 | { |
| 285 | if (rope->data != NULL) { |
| 286 | int old_max = rope->max; |
| 287 | bytes_t old_data = rope->data == rope->data0 ? NULL : rope->data; |
| 288 | int new_max = old_max; |
| 289 | bytes_t new_data; |
| 290 | |
| 291 | while (new_max < new_count) { |
| 292 | new_max += (new_max >> 1) + 4; |
| 293 | } |
| 294 | new_data = realloc( old_data, new_max ); |
| 295 | if (new_data == NULL) { |
| 296 | rope->error = 1; |
| 297 | return -1; |
| 298 | } |
| 299 | rope->data = new_data; |
| 300 | rope->max = new_max; |
| 301 | } else { |
| 302 | rope->max = new_count; |
| 303 | } |
| 304 | return 0; |
| 305 | } |
| 306 | |
| 307 | static int |
| 308 | gsm_rope_can_grow( GsmRope rope, int count ) |
| 309 | { |
| 310 | if (!rope->data || rope->error) |
| 311 | return 0; |
| 312 | |
| 313 | if (rope->pos + count > rope->max) |
| 314 | { |
| 315 | if (rope->data == NULL) |
| 316 | rope->max = rope->pos + count; |
| 317 | |
| 318 | else if (rope->error || |
| 319 | gsm_rope_ensure( rope, rope->pos + count ) < 0) |
| 320 | return 0; |
| 321 | } |
| 322 | return 1; |
| 323 | } |
| 324 | |
| 325 | void |
| 326 | gsm_rope_add_c( GsmRope rope, char c ) |
| 327 | { |
| 328 | if (gsm_rope_can_grow(rope, 1)) { |
| 329 | rope->data[ rope->pos ] = (byte_t) c; |
| 330 | } |
| 331 | rope->pos += 1; |
| 332 | } |
| 333 | |
| 334 | void |
| 335 | gsm_rope_add( GsmRope rope, const void* buf, int buflen ) |
| 336 | { |
| 337 | if (gsm_rope_can_grow(rope, buflen)) { |
| 338 | memcpy( rope->data + rope->pos, (const char*)buf, buflen ); |
| 339 | } |
| 340 | rope->pos += buflen; |
| 341 | } |
| 342 | |
| 343 | void* |
| 344 | gsm_rope_reserve( GsmRope rope, int count ) |
| 345 | { |
| 346 | void* result = NULL; |
| 347 | |
| 348 | if (gsm_rope_can_grow(rope, count)) |
| 349 | { |
| 350 | if (rope->data != NULL) |
| 351 | result = rope->data + rope->pos; |
| 352 | } |
| 353 | rope->pos += count; |
| 354 | |
| 355 | return result; |
| 356 | } |
| 357 | |
| 358 | /* skip a given number of Unicode characters in a utf-8 byte string */ |
| 359 | cbytes_t |
| 360 | utf8_skip( cbytes_t utf8, |
| 361 | cbytes_t utf8end, |
| 362 | int count) |
| 363 | { |
| 364 | cbytes_t p = utf8; |
| 365 | cbytes_t end = utf8end; |
| 366 | |
| 367 | for ( ; count > 0; count-- ) { |
| 368 | int c; |
| 369 | |
| 370 | if (p >= end) |
| 371 | break; |
| 372 | |
| 373 | c = *p++; |
| 374 | if (c > 128) { |
| 375 | while (p < end && (p[0] & 0xc0) == 0x80) |
| 376 | p++; |
| 377 | } |
| 378 | } |
| 379 | return p; |
| 380 | } |
| 381 | |
| 382 | |
| 383 | static __inline__ int |
| 384 | utf8_next( cbytes_t *pp, cbytes_t end ) |
| 385 | { |
| 386 | cbytes_t p = *pp; |
| 387 | int result = -1; |
| 388 | |
| 389 | if (p < end) { |
| 390 | int c= *p++; |
| 391 | if (c >= 128) { |
| 392 | if ((c & 0xe0) == 0xc0) |
| 393 | c &= 0x1f; |
| 394 | else if ((c & 0xf0) == 0xe0) |
| 395 | c &= 0x0f; |
| 396 | else |
| 397 | c &= 0x07; |
| 398 | |
| 399 | while (p < end && (p[0] & 0xc0) == 0x80) { |
| 400 | c = (c << 6) | (p[0] & 0x3f); |
| 401 | p ++; |
| 402 | } |
| 403 | } |
| 404 | result = c; |
| 405 | *pp = p; |
| 406 | } |
| 407 | return result; |
| 408 | } |
| 409 | |
| 410 | |
| 411 | __inline__ int |
| 412 | utf8_write( bytes_t utf8, int offset, int v ) |
| 413 | { |
| 414 | int result; |
| 415 | |
| 416 | if (v < 128) { |
| 417 | result = 1; |
| 418 | if (utf8) |
| 419 | utf8[offset] = (byte_t) v; |
| 420 | } else if (v < 0x800) { |
| 421 | result = 2; |
| 422 | if (utf8) { |
| 423 | utf8[offset+0] = (byte_t)( 0xc0 | (v >> 6) ); |
| 424 | utf8[offset+1] = (byte_t)( 0x80 | (v & 0x3f) ); |
| 425 | } |
| 426 | } else if (v < 0x10000) { |
| 427 | result = 3; |
| 428 | if (utf8) { |
| 429 | utf8[offset+0] = (byte_t)( 0xe0 | (v >> 12) ); |
| 430 | utf8[offset+1] = (byte_t)( 0x80 | ((v >> 6) & 0x3f) ); |
| 431 | utf8[offset+2] = (byte_t)( 0x80 | (v & 0x3f) ); |
| 432 | } |
| 433 | } else { |
| 434 | result = 4; |
| 435 | if (utf8) { |
| 436 | utf8[offset+0] = (byte_t)( 0xf0 | ((v >> 18) & 0x7) ); |
| 437 | utf8[offset+1] = (byte_t)( 0x80 | ((v >> 12) & 0x3f) ); |
| 438 | utf8[offset+2] = (byte_t)( 0x80 | ((v >> 6) & 0x3f) ); |
| 439 | utf8[offset+3] = (byte_t)( 0x80 | (v & 0x3f) ); |
| 440 | } |
| 441 | } |
| 442 | return result; |
| 443 | } |
| 444 | |
| 445 | static __inline__ int |
| 446 | ucs2_write( bytes_t ucs2, int offset, int v ) |
| 447 | { |
| 448 | if (ucs2) { |
| 449 | ucs2[offset+0] = (byte_t) (v >> 8); |
| 450 | ucs2[offset+1] = (byte_t) (v); |
| 451 | } |
| 452 | return 2; |
| 453 | } |
| 454 | |
| 455 | int |
| 456 | utf8_check( cbytes_t p, int utf8len ) |
| 457 | { |
| 458 | cbytes_t end = p + utf8len; |
| 459 | int result = 0; |
| 460 | |
| 461 | if (p) { |
| 462 | while (p < end) { |
| 463 | int c = *p++; |
| 464 | if (c >= 128) { |
| 465 | int len; |
| 466 | if ((c & 0xe0) == 0xc0) { |
| 467 | len = 1; |
| 468 | } |
| 469 | else if ((c & 0xf0) == 0xe0) { |
| 470 | len = 2; |
| 471 | } |
| 472 | else if ((c & 0xf8) == 0xf0) { |
| 473 | len = 3; |
| 474 | } |
| 475 | else |
| 476 | goto Exit; /* malformed utf-8 */ |
| 477 | |
| 478 | if (p+len > end) /* string too short */ |
| 479 | goto Exit; |
| 480 | |
| 481 | for ( ; len > 0; len--, p++ ) { |
| 482 | if ((p[0] & 0xc0) != 0x80) |
| 483 | goto Exit; |
| 484 | } |
| 485 | } |
| 486 | } |
| 487 | result = 1; |
| 488 | } |
| 489 | Exit: |
| 490 | return result; |
| 491 | } |
| 492 | |
| 493 | /** UCS2 to UTF8 |
| 494 | **/ |
| 495 | |
| 496 | /* convert a UCS2 string into a UTF8 byte string, assumes 'buf' is correctly sized */ |
| 497 | int |
| 498 | ucs2_to_utf8( cbytes_t ucs2, |
| 499 | int ucs2len, |
| 500 | bytes_t buf ) |
| 501 | { |
| 502 | int nn; |
| 503 | int result = 0; |
| 504 | |
| 505 | for (nn = 0; nn < ucs2len; ucs2 += 2, nn++) { |
| 506 | int c= (ucs2[0] << 8) | ucs2[1]; |
| 507 | result += utf8_write(buf, result, c); |
| 508 | } |
| 509 | return result; |
| 510 | } |
| 511 | |
| 512 | /* count the number of UCS2 chars contained in a utf8 byte string */ |
| 513 | int |
| 514 | utf8_to_ucs2( cbytes_t utf8, |
| 515 | int utf8len, |
| 516 | bytes_t ucs2 ) |
| 517 | { |
| 518 | cbytes_t p = utf8; |
| 519 | cbytes_t end = p + utf8len; |
| 520 | int result = 0; |
| 521 | |
| 522 | while (p < end) { |
| 523 | int c = utf8_next(&p, end); |
| 524 | |
| 525 | if (c < 0) |
| 526 | break; |
| 527 | |
| 528 | result += ucs2_write(ucs2, result, c); |
| 529 | } |
| 530 | return result/2; |
| 531 | } |
| 532 | |
| 533 | |
| 534 | |
| 535 | /** GSM ALPHABET |
| 536 | **/ |
| 537 | |
| 538 | #define GSM_7BITS_ESCAPE 0x1b |
| 539 | #define GSM_7BITS_UNKNOWN 0 |
| 540 | |
| 541 | static const unsigned short gsm7bits_to_unicode[128] = { |
| 542 | '@', 0xa3, '$', 0xa5, 0xe8, 0xe9, 0xf9, 0xec, 0xf2, 0xc7, '\n', 0xd8, 0xf8, '\r', 0xc5, 0xe5, |
| 543 | 0x394, '_',0x3a6,0x393,0x39b,0x3a9,0x3a0,0x3a8,0x3a3,0x398,0x39e, 0, 0xc6, 0xe6, 0xdf, 0xc9, |
| 544 | ' ', '!', '"', '#', 0xa4, '%', '&', '\'', '(', ')', '*', '+', ',', '-', '.', '/', |
| 545 | '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', ':', ';', '<', '=', '>', '?', |
| 546 | 0xa1, 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', |
| 547 | 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z', 0xc4, 0xd6,0x147, 0xdc, 0xa7, |
| 548 | 0xbf, 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', |
| 549 | 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z', 0xe4, 0xf6, 0xf1, 0xfc, 0xe0, |
| 550 | }; |
| 551 | |
| 552 | static const unsigned short gsm7bits_extend_to_unicode[128] = { |
| 553 | 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,'\f', 0, 0, 0, 0, 0, |
| 554 | 0, 0, 0, 0, '^', 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, |
| 555 | 0, 0, 0, 0, 0, 0, 0, 0, '{', '}', 0, 0, 0, 0, 0,'\\', |
| 556 | 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, '[', '~', ']', 0, |
| 557 | '|', 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, |
| 558 | 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, |
| 559 | 0, 0, 0, 0, 0,0x20ac, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, |
| 560 | 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, |
| 561 | }; |
| 562 | |
| 563 | |
| 564 | static int |
| 565 | unichar_to_gsm7( int unicode ) |
| 566 | { |
| 567 | int nn; |
| 568 | for (nn = 0; nn < 128; nn++) { |
| 569 | if (gsm7bits_to_unicode[nn] == unicode) { |
| 570 | return nn; |
| 571 | } |
| 572 | } |
| 573 | return -1; |
| 574 | } |
| 575 | |
| 576 | static int |
| 577 | unichar_to_gsm7_extend( int unichar ) |
| 578 | { |
| 579 | int nn; |
| 580 | for (nn = 0; nn < 128; nn++) { |
| 581 | if (gsm7bits_extend_to_unicode[nn] == unichar) { |
| 582 | return nn; |
| 583 | } |
| 584 | } |
| 585 | return -1; |
| 586 | } |
| 587 | |
| 588 | |
| 589 | /* return the number of septets needed to encode a unicode charcode */ |
| 590 | static int |
| 591 | unichar_to_gsm7_count( int unicode ) |
| 592 | { |
| 593 | int nn; |
| 594 | |
| 595 | nn = unichar_to_gsm7(unicode); |
| 596 | if (nn >= 0) |
| 597 | return 1; |
| 598 | |
| 599 | nn = unichar_to_gsm7_extend(unicode); |
| 600 | if (nn >= 0) |
| 601 | return 2; |
| 602 | |
| 603 | return 0; |
| 604 | } |
| 605 | |
| 606 | |
| 607 | cbytes_t |
| 608 | utf8_skip_gsm7( cbytes_t utf8, cbytes_t utf8end, int gsm7len ) |
| 609 | { |
| 610 | cbytes_t p = utf8; |
| 611 | cbytes_t end = utf8end; |
| 612 | |
| 613 | while (gsm7len >0) { |
| 614 | cbytes_t q = p; |
| 615 | int c = utf8_next( &q, end ); |
| 616 | int len; |
| 617 | |
| 618 | if (c < 0) |
| 619 | break; |
| 620 | |
| 621 | len = unichar_to_gsm7_count( c ); |
| 622 | if (len == 0) /* unknown chars are replaced by spaces */ |
| 623 | len = 1; |
| 624 | |
| 625 | if (len > gsm7len) |
| 626 | break; |
| 627 | |
| 628 | gsm7len -= len; |
| 629 | p = q; |
| 630 | } |
| 631 | return p; |
| 632 | } |
| 633 | |
| 634 | |
| 635 | int |
| 636 | utf8_check_gsm7( cbytes_t utf8, |
| 637 | int utf8len ) |
| 638 | { |
| 639 | cbytes_t utf8end = utf8 + utf8len; |
| 640 | |
| 641 | while (utf8 < utf8end) { |
| 642 | int c = utf8_next( &utf8, utf8end ); |
| 643 | if (unichar_to_gsm7_count(c) == 0) |
| 644 | return 0; |
| 645 | } |
| 646 | return 1; |
| 647 | } |
| 648 | |
| 649 | |
| 650 | int |
| 651 | utf8_from_gsm7( cbytes_t src, |
| 652 | int septet_offset, |
| 653 | int septet_count, |
| 654 | bytes_t utf8 ) |
| 655 | { |
| 656 | int shift = (septet_offset & 7); |
| 657 | int escaped = 0; |
| 658 | int result = 0; |
| 659 | |
| 660 | src += (septet_offset >> 3); |
| 661 | for ( ; septet_count > 0; septet_count-- ) |
| 662 | { |
| 663 | int c = (src[0] >> shift) & 0x7f; |
| 664 | int v; |
| 665 | |
| 666 | if (shift > 1) { |
| 667 | c = ((src[1] << (8-shift)) | c) & 0x7f; |
| 668 | } |
| 669 | |
| 670 | if (escaped) { |
| 671 | v = gsm7bits_extend_to_unicode[c]; |
| 672 | } else if (c == GSM_7BITS_ESCAPE) { |
| 673 | escaped = 1; |
| 674 | goto NextSeptet; |
| 675 | } else { |
| 676 | v = gsm7bits_to_unicode[c]; |
| 677 | } |
| 678 | |
| 679 | result += utf8_write( utf8, result, v ); |
| 680 | |
| 681 | NextSeptet: |
| 682 | shift += 7; |
| 683 | if (shift >= 8) { |
| 684 | shift -= 8; |
| 685 | src += 1; |
| 686 | } |
| 687 | } |
| 688 | return result; |
| 689 | } |
| 690 | |
| 691 | |
| 692 | int |
| 693 | utf8_from_gsm8( cbytes_t src, int count, bytes_t utf8 ) |
| 694 | { |
| 695 | int result = 0; |
| 696 | int escaped = 0; |
| 697 | |
| 698 | |
| 699 | for ( ; count > 0; count-- ) |
| 700 | { |
| 701 | int c = *src++; |
| 702 | |
| 703 | if (c == 0xff) |
| 704 | break; |
| 705 | |
| 706 | if (c == GSM_7BITS_ESCAPE) { |
| 707 | if (escaped) { /* two escape characters => one space */ |
| 708 | c = 0x20; |
| 709 | escaped = 0; |
| 710 | } else { |
| 711 | escaped = 1; |
| 712 | continue; |
| 713 | } |
| 714 | } |
| 715 | else |
| 716 | { |
| 717 | if (c >= 0x80) { |
| 718 | c = 0x20; |
| 719 | escaped = 0; |
| 720 | } else if (escaped) { |
| 721 | c = gsm7bits_extend_to_unicode[c]; |
| 722 | } else |
| 723 | c = gsm7bits_to_unicode[c]; |
| 724 | } |
| 725 | |
| 726 | result += utf8_write( utf8, result, c ); |
| 727 | } |
| 728 | return result; |
| 729 | } |
| 730 | |
| 731 | /* convert a GSM 7-bit message into a unicode character array |
| 732 | * the 'dst' array must contain at least 160 chars. the function |
| 733 | * returns the number of characters decoded |
| 734 | * |
| 735 | * assumes the 'dst' array has at least septet_count items, returns the |
| 736 | * number of unichars really written |
| 737 | */ |
| 738 | int |
| 739 | ucs2_from_gsm7( bytes_t ucs2, |
| 740 | cbytes_t src, |
| 741 | int septet_offset, |
| 742 | int septet_count ) |
| 743 | { |
| 744 | const unsigned char* p = src + (septet_offset >> 3); |
| 745 | int shift = (septet_offset & 7); |
| 746 | int escaped = 0; |
| 747 | int result = 0; |
| 748 | |
| 749 | for ( ; septet_count > 0; septet_count-- ) |
| 750 | { |
| 751 | unsigned val = (p[0] >> shift) & 0x7f; |
| 752 | |
| 753 | if (shift > 1) |
| 754 | val = (val | (p[1] << (8-shift))) & 0x7f; |
| 755 | |
| 756 | if (escaped) { |
| 757 | int c = gsm7bits_to_unicode[val]; |
| 758 | |
| 759 | result += ucs2_write(ucs2, result, c); |
| 760 | escaped = 0; |
| 761 | } |
| 762 | else if (val == GSM_7BITS_ESCAPE) { |
| 763 | escaped = 1; |
| 764 | } |
| 765 | else { |
| 766 | val = gsm7bits_extend_to_unicode[val]; |
| 767 | if (val == 0) |
| 768 | val = 0x20; |
| 769 | |
| 770 | result += ucs2_write( ucs2, result, val ); |
| 771 | } |
| 772 | } |
| 773 | return result/2; |
| 774 | } |
| 775 | |
| 776 | |
| 777 | /* count the number of septets required to write a utf8 string */ |
| 778 | static int |
| 779 | utf8_to_gsm7_count( cbytes_t utf8, int utf8len ) |
| 780 | { |
| 781 | cbytes_t utf8end = utf8 + utf8len; |
| 782 | int result = 0; |
| 783 | |
| 784 | while ( utf8 < utf8end ) { |
| 785 | int len; |
| 786 | int c = utf8_next( &utf8, utf8end ); |
| 787 | |
| 788 | if (c < 0) |
| 789 | break; |
| 790 | |
| 791 | len = unichar_to_gsm7_count(c); |
| 792 | if (len == 0) /* replace non-representables with space */ |
| 793 | len = 1; |
| 794 | |
| 795 | result += len; |
| 796 | } |
| 797 | return result; |
| 798 | } |
| 799 | |
| 800 | typedef struct { |
| 801 | bytes_t dst; |
| 802 | unsigned pad; |
| 803 | int bits; |
| 804 | int offset; |
| 805 | } BWriterRec, *BWriter; |
| 806 | |
| 807 | static void |
| 808 | bwriter_init( BWriter writer, bytes_t dst, int start ) |
| 809 | { |
| 810 | int shift = start & 7; |
| 811 | |
| 812 | writer->dst = dst + (start >> 3); |
| 813 | writer->pad = 0; |
| 814 | writer->bits = shift; |
| 815 | writer->offset = start; |
| 816 | |
| 817 | if (shift > 0) { |
| 818 | writer->pad = writer->dst[0] & ~(0xFF << shift); |
| 819 | } |
| 820 | } |
| 821 | |
| 822 | static void |
| 823 | bwriter_add7( BWriter writer, unsigned value ) |
| 824 | { |
| 825 | writer->pad |= (unsigned)(value << writer->bits); |
| 826 | writer->bits += 7; |
| 827 | if (writer->bits >= 8) { |
| 828 | writer->dst[0] = (byte_t)writer->pad; |
| 829 | writer->bits -= 8; |
| 830 | writer->pad >>= 8; |
| 831 | writer->dst += 1; |
| 832 | } |
| 833 | writer->offset += 7; |
| 834 | } |
| 835 | |
| 836 | static int |
| 837 | bwriter_done( BWriter writer ) |
| 838 | { |
| 839 | if (writer->bits > 0) { |
| 840 | writer->dst[0] = (byte_t)writer->pad; |
| 841 | writer->pad = 0; |
| 842 | writer->bits = 0; |
| 843 | writer->dst += 1; |
| 844 | } |
| 845 | return writer->offset; |
| 846 | } |
| 847 | |
| 848 | /* convert a utf8 string to a gsm7 byte string - return the number of septets written */ |
| 849 | int |
| 850 | utf8_to_gsm7( cbytes_t utf8, int utf8len, bytes_t dst, int offset ) |
| 851 | { |
| 852 | const unsigned char* utf8end = utf8 + utf8len; |
| 853 | BWriterRec writer[1]; |
| 854 | |
| 855 | if (dst == NULL) |
| 856 | return utf8_to_gsm7_count(utf8, utf8len); |
| 857 | |
| 858 | bwriter_init( writer, dst, offset ); |
| 859 | while ( utf8 < utf8end ) { |
| 860 | int c = utf8_next( &utf8, utf8end ); |
| 861 | int nn; |
| 862 | |
| 863 | if (c < 0) |
| 864 | break; |
| 865 | |
| 866 | nn = unichar_to_gsm7(c); |
| 867 | if (nn >= 0) { |
| 868 | bwriter_add7( writer, nn ); |
| 869 | continue; |
| 870 | } |
| 871 | |
| 872 | nn = unichar_to_gsm7_extend(c); |
| 873 | if (nn >= 0) { |
| 874 | bwriter_add7( writer, GSM_7BITS_ESCAPE ); |
| 875 | bwriter_add7( writer, nn ); |
| 876 | continue; |
| 877 | } |
| 878 | |
| 879 | /* unknown => replaced by space */ |
| 880 | bwriter_add7( writer, 0x20 ); |
| 881 | } |
| 882 | return bwriter_done( writer ); |
| 883 | } |
| 884 | |
| 885 | |
| 886 | int |
| 887 | utf8_to_gsm8( cbytes_t utf8, int utf8len, bytes_t dst ) |
| 888 | { |
| 889 | const unsigned char* utf8end = utf8 + utf8len; |
| 890 | int result = 0; |
| 891 | |
| 892 | while ( utf8 < utf8end ) { |
| 893 | int c = utf8_next( &utf8, utf8end ); |
| 894 | int nn; |
| 895 | |
| 896 | if (c < 0) |
| 897 | break; |
| 898 | |
| 899 | nn = unichar_to_gsm7(c); |
| 900 | if (nn >= 0) { |
| 901 | if (dst) |
| 902 | dst[result] = (byte_t)nn; |
| 903 | result += 1; |
| 904 | continue; |
| 905 | } |
| 906 | |
| 907 | nn = unichar_to_gsm7_extend(c); |
| 908 | if (nn >= 0) { |
| 909 | if (dst) { |
| 910 | dst[result+0] = (byte_t) GSM_7BITS_ESCAPE; |
| 911 | dst[result+1] = (byte_t) nn; |
| 912 | } |
| 913 | result += 2; |
| 914 | continue; |
| 915 | } |
| 916 | |
| 917 | /* unknown => space */ |
| 918 | if (dst) |
| 919 | dst[result] = 0x20; |
| 920 | result += 1; |
| 921 | } |
| 922 | return result; |
| 923 | } |
| 924 | |
| 925 | |
| 926 | int |
| 927 | ucs2_to_gsm7( cbytes_t ucs2, int ucs2len, bytes_t dst, int offset ) |
| 928 | { |
| 929 | const unsigned char* ucs2end = ucs2 + ucs2len*2; |
| 930 | BWriterRec writer[1]; |
| 931 | |
| 932 | bwriter_init( writer, dst, offset ); |
| 933 | while ( ucs2 < ucs2end ) { |
| 934 | int c = *ucs2++; |
| 935 | int nn; |
| 936 | |
| 937 | for (nn = 0; nn < 128; nn++) { |
| 938 | if ( gsm7bits_to_unicode[nn] == c ) { |
| 939 | bwriter_add7( writer, nn ); |
| 940 | goto NextUnicode; |
| 941 | } |
| 942 | } |
| 943 | for (nn = 0; nn < 128; nn++) { |
| 944 | if ( gsm7bits_extend_to_unicode[nn] == c ) { |
| 945 | bwriter_add7( writer, GSM_7BITS_ESCAPE ); |
| 946 | bwriter_add7( writer, nn ); |
| 947 | goto NextUnicode; |
| 948 | } |
| 949 | } |
| 950 | |
| 951 | /* unknown */ |
| 952 | bwriter_add7( writer, 0x20 ); |
| 953 | |
| 954 | NextUnicode: |
| 955 | ; |
| 956 | } |
| 957 | return bwriter_done( writer ); |
| 958 | } |
| 959 | |
| 960 | |
| 961 | int |
| 962 | ucs2_to_gsm8( cbytes_t ucs2, int ucs2len, bytes_t dst ) |
| 963 | { |
| 964 | const unsigned char* ucs2end = ucs2 + ucs2len*2; |
| 965 | bytes_t dst0 = dst; |
| 966 | |
| 967 | while ( ucs2 < ucs2end ) { |
| 968 | int c = *ucs2++; |
| 969 | int nn; |
| 970 | |
| 971 | for (nn = 0; nn < 128; nn++) { |
| 972 | if ( gsm7bits_to_unicode[nn] == c ) { |
| 973 | *dst++ = (byte_t)nn; |
| 974 | goto NextUnicode; |
| 975 | } |
| 976 | } |
| 977 | for (nn = 0; nn < 128; nn++) { |
| 978 | if ( gsm7bits_extend_to_unicode[nn] == c ) { |
| 979 | dst[0] = (byte_t) GSM_7BITS_ESCAPE; |
| 980 | dst[1] = (byte_t) nn; |
| 981 | dst += 2; |
| 982 | goto NextUnicode; |
| 983 | } |
| 984 | } |
| 985 | |
| 986 | /* unknown */ |
| 987 | *dst++ = 0x20; |
| 988 | |
| 989 | NextUnicode: |
| 990 | ; |
| 991 | } |
| 992 | return (dst - dst0); |
| 993 | } |
| 994 | |
| 995 | int |
| 996 | gsm_bcdnum_to_ascii( cbytes_t bcd, int count, bytes_t dst ) |
| 997 | { |
| 998 | int result = 0; |
| 999 | int shift = 0; |
| 1000 | |
| 1001 | while (count > 0) { |
| 1002 | int c = (bcd[0] >> shift) & 0xf; |
| 1003 | |
| 1004 | if (c == 15 && count == 1) /* ignore trailing 0xf */ |
| 1005 | break; |
| 1006 | |
| 1007 | if (c >= 14) |
| 1008 | c = 0; |
| 1009 | |
| 1010 | if (dst) dst[result] = "0123456789*#,N"[c]; |
| 1011 | result += 1; |
| 1012 | |
| 1013 | shift += 4; |
| 1014 | if (shift == 8) { |
| 1015 | shift = 0; |
| 1016 | bcd += 1; |
| 1017 | } |
| 1018 | } |
| 1019 | return result; |
| 1020 | } |
| 1021 | |
| 1022 | |
| 1023 | int |
| 1024 | gsm_bcdnum_from_ascii( cbytes_t ascii, int asciilen, bytes_t dst ) |
| 1025 | { |
| 1026 | cbytes_t end = ascii + asciilen; |
| 1027 | int result = 0; |
| 1028 | int phase = 0x01; |
| 1029 | |
| 1030 | while (ascii < end) { |
| 1031 | int c = *ascii++; |
| 1032 | |
| 1033 | if (c == '*') |
| 1034 | c = 10; |
| 1035 | else if (c == '#') |
| 1036 | c = 11; |
| 1037 | else if (c == ',') |
| 1038 | c = 12; |
| 1039 | else if (c == 'N') |
| 1040 | c = 13; |
| 1041 | else { |
| 1042 | c -= '0'; |
| 1043 | if ((unsigned)c >= 10U) |
| 1044 | return -1; |
| 1045 | } |
| 1046 | phase = (phase << 4) | c; |
| 1047 | result += 1; |
| 1048 | if (phase & 0x100) { |
| 1049 | if (dst) dst[result/2] = (byte_t) phase; |
| 1050 | phase = 0x01; |
| 1051 | } |
| 1052 | } |
| 1053 | |
| 1054 | if (result & 1) { |
| 1055 | if (dst) dst[result/2] = (byte_t)(phase | 0xf0); |
| 1056 | } |
| 1057 | return result; |
| 1058 | } |
| 1059 | |
| 1060 | /** ADN: Abbreviated Dialing Number |
| 1061 | **/ |
| 1062 | |
| 1063 | #define ADN_FOOTER_SIZE 14 |
| 1064 | #define ADN_OFFSET_NUMBER_LENGTH 0 |
| 1065 | #define ADN_OFFSET_TON_NPI 1 |
| 1066 | #define ADN_OFFSET_NUMBER_START 2 |
| 1067 | #define ADN_OFFSET_NUMBER_END 11 |
| 1068 | #define ADN_OFFSET_CAPABILITY_ID 12 |
| 1069 | #define ADN_OFFSET_EXTENSION_ID 13 |
| 1070 | |
| 1071 | /* see 10.5.1 of 3GPP 51.011 */ |
| 1072 | static int |
| 1073 | sim_adn_alpha_to_utf8( cbytes_t alpha, cbytes_t end, bytes_t dst ) |
| 1074 | { |
| 1075 | int result = 0; |
| 1076 | |
| 1077 | /* ignore trailing 0xff */ |
| 1078 | while (alpha < end && end[-1] == 0xff) |
| 1079 | end--; |
| 1080 | |
| 1081 | if (alpha >= end) |
| 1082 | return 0; |
| 1083 | |
| 1084 | if (alpha[0] == 0x80) { /* UCS/2 source encoding */ |
| 1085 | alpha += 1; |
| 1086 | result = ucs2_to_utf8( alpha, (end-alpha)/2, dst ); |
| 1087 | } |
| 1088 | else |
| 1089 | { |
| 1090 | int is_ucs2 = 0; |
| 1091 | int len = 0, base = 0; |
| 1092 | |
| 1093 | if (alpha+3 <= end && alpha[0] == 0x81) { |
| 1094 | is_ucs2 = 1; |
| 1095 | len = alpha[1]; |
| 1096 | base = alpha[2] << 7; |
| 1097 | alpha += 3; |
| 1098 | if (len > end-alpha) |
| 1099 | len = end-alpha; |
| 1100 | } else if (alpha+4 <= end && alpha[0] == 0x82) { |
| 1101 | is_ucs2 = 1; |
| 1102 | len = alpha[1]; |
| 1103 | base = (alpha[2] << 8) | alpha[3]; |
| 1104 | alpha += 4; |
| 1105 | if (len > end-alpha) |
| 1106 | len = end-alpha; |
| 1107 | } |
| 1108 | |
| 1109 | if (is_ucs2) { |
| 1110 | end = alpha + len; |
| 1111 | while (alpha < end) { |
| 1112 | int c = alpha[0]; |
| 1113 | if (c >= 0x80) { |
| 1114 | result += utf8_write(dst, result, base + (c & 0x7f)); |
| 1115 | alpha += 1; |
| 1116 | } else { |
| 1117 | /* GSM character set */ |
| 1118 | int count; |
| 1119 | for (count = 0; alpha+count < end && alpha[count] < 128; count++) |
| 1120 | ; |
| 1121 | result += utf8_from_gsm8(alpha, count, (dst ? dst+result : NULL)); |
| 1122 | alpha += count; |
| 1123 | } |
| 1124 | } |
| 1125 | } |
| 1126 | else { |
| 1127 | result = utf8_from_gsm8(alpha, end-alpha, dst); |
| 1128 | } |
| 1129 | } |
| 1130 | return result; |
| 1131 | } |
| 1132 | |
| 1133 | #if 0 |
| 1134 | static int |
| 1135 | sim_adn_alpha_from_utf8( cbytes_t utf8, int utf8len, bytes_t dst ) |
| 1136 | { |
| 1137 | int result = 0; |
| 1138 | |
| 1139 | if (utf8_check_gsm7(utf8, utf8len)) { |
| 1140 | /* GSM 7-bit compatible, encode directly as 8-bit string */ |
| 1141 | result = utf8_to_gsm8(utf8, utf8len, dst); |
| 1142 | } else { |
| 1143 | /* otherwise, simply try UCS-2 encoding, nothing more serious at the moment */ |
| 1144 | if (dst) { |
| 1145 | dst[0] = 0x80; |
| 1146 | } |
| 1147 | result = 1 + utf8_to_ucs2(utf8, utf8len, dst ? (dst+1) : NULL)*2; |
| 1148 | } |
| 1149 | return result; |
| 1150 | } |
| 1151 | #endif |
| 1152 | |
| 1153 | int |
| 1154 | sim_adn_record_from_bytes( SimAdnRecord rec, cbytes_t data, int len ) |
| 1155 | { |
| 1156 | cbytes_t end = data + len; |
| 1157 | cbytes_t footer = end - ADN_FOOTER_SIZE; |
| 1158 | int num_len; |
| 1159 | |
| 1160 | rec->adn.alpha[0] = 0; |
| 1161 | rec->adn.number[0] = 0; |
| 1162 | rec->ext_record = 0xff; |
| 1163 | |
| 1164 | if (len < ADN_FOOTER_SIZE) |
| 1165 | return -1; |
| 1166 | |
| 1167 | /* alpha is optional */ |
| 1168 | if (len > ADN_FOOTER_SIZE) { |
| 1169 | cbytes_t dataend = data + len - ADN_FOOTER_SIZE; |
| 1170 | int count = sim_adn_alpha_to_utf8(data, dataend, NULL); |
| 1171 | |
| 1172 | if (count > sizeof(rec->adn.alpha)-1) /* too long */ |
| 1173 | return -1; |
| 1174 | |
| 1175 | sim_adn_alpha_to_utf8(data, dataend, rec->adn.alpha); |
| 1176 | rec->adn.alpha[count] = 0; |
| 1177 | } |
| 1178 | |
| 1179 | num_len = footer[ADN_OFFSET_NUMBER_LENGTH]; |
| 1180 | if (num_len > 11) |
| 1181 | return -1; |
| 1182 | |
| 1183 | /* decode TON and number to ASCII, NOTE: this is lossy !! */ |
| 1184 | { |
| 1185 | int ton = footer[ADN_OFFSET_TON_NPI]; |
| 1186 | bytes_t number = (bytes_t) rec->adn.number; |
| 1187 | int len = sizeof(rec->adn.number)-1; |
| 1188 | int count; |
| 1189 | |
| 1190 | if (ton != 0x81 && ton != 0x91) |
| 1191 | return -1; |
| 1192 | |
| 1193 | if (ton == 0x91) { |
| 1194 | *number++ = '+'; |
| 1195 | len -= 1; |
| 1196 | } |
| 1197 | |
| 1198 | count = gsm_bcdnum_to_ascii( footer + ADN_OFFSET_NUMBER_START, |
| 1199 | num_len*2, number ); |
| 1200 | number[count] = 0; |
| 1201 | } |
| 1202 | return 0; |
| 1203 | } |
| 1204 | |
| 1205 | int |
| 1206 | sim_adn_record_to_bytes( SimAdnRecord rec, bytes_t data, int datalen ) |
| 1207 | { |
| 1208 | bytes_t end = data + datalen; |
| 1209 | bytes_t footer = end - ADN_FOOTER_SIZE; |
David 'Digit' Turner | a2c14f9 | 2014-02-04 01:02:30 +0100 | [diff] [blame] | 1210 | //int ton = 0x81; |
The Android Open Source Project | 8b23a6c | 2009-03-03 19:30:32 -0800 | [diff] [blame] | 1211 | cbytes_t number = (cbytes_t) rec->adn.number; |
| 1212 | |
| 1213 | if (number[0] == '+') { |
David 'Digit' Turner | a2c14f9 | 2014-02-04 01:02:30 +0100 | [diff] [blame] | 1214 | //ton = 0x91; |
The Android Open Source Project | 8b23a6c | 2009-03-03 19:30:32 -0800 | [diff] [blame] | 1215 | number += 1; |
| 1216 | } |
| 1217 | footer[0] = (strlen((const char*)number)+1)/2 + 1; |
| 1218 | /* XXXX: TODO */ |
| 1219 | return 0; |
| 1220 | } |