1 /* 2 * string.c : an XML string utilities module 3 * 4 * This module provides various utility functions for manipulating 5 * the xmlChar* type. All functions named xmlStr* have been moved here 6 * from the parser.c file (their original home). 7 * 8 * See Copyright for the status of this software. 9 * 10 * UTF8 string routines from: 11 * William Brack <wbrack (at) mmm.com.hk> 12 * 13 * daniel (at) veillard.com 14 */ 15 16 #define IN_LIBXML 17 #include "libxml.h" 18 19 #include <stdlib.h> 20 #include <string.h> 21 #include <libxml/xmlmemory.h> 22 #include <libxml/parserInternals.h> 23 #include <libxml/xmlstring.h> 24 25 /************************************************************************ 26 * * 27 * Commodity functions to handle xmlChars * 28 * * 29 ************************************************************************/ 30 31 /** 32 * xmlStrndup: 33 * @cur: the input xmlChar * 34 * @len: the len of @cur 35 * 36 * a strndup for array of xmlChar's 37 * 38 * Returns a new xmlChar * or NULL 39 */ 40 xmlChar * 41 xmlStrndup(const xmlChar *cur, int len) { 42 xmlChar *ret; 43 44 if ((cur == NULL) || (len < 0)) return(NULL); 45 ret = (xmlChar *) xmlMallocAtomic((len + 1) * sizeof(xmlChar)); 46 if (ret == NULL) { 47 xmlErrMemory(NULL, NULL); 48 return(NULL); 49 } 50 memcpy(ret, cur, len * sizeof(xmlChar)); 51 ret[len] = 0; 52 return(ret); 53 } 54 55 /** 56 * xmlStrdup: 57 * @cur: the input xmlChar * 58 * 59 * a strdup for array of xmlChar's. Since they are supposed to be 60 * encoded in UTF-8 or an encoding with 8bit based chars, we assume 61 * a termination mark of '0'. 62 * 63 * Returns a new xmlChar * or NULL 64 */ 65 xmlChar * 66 xmlStrdup(const xmlChar *cur) { 67 const xmlChar *p = cur; 68 69 if (cur == NULL) return(NULL); 70 while (*p != 0) p++; /* non input consuming */ 71 return(xmlStrndup(cur, p - cur)); 72 } 73 74 /** 75 * xmlCharStrndup: 76 * @cur: the input char * 77 * @len: the len of @cur 78 * 79 * a strndup for char's to xmlChar's 80 * 81 * Returns a new xmlChar * or NULL 82 */ 83 84 xmlChar * 85 xmlCharStrndup(const char *cur, int len) { 86 int i; 87 xmlChar *ret; 88 89 if ((cur == NULL) || (len < 0)) return(NULL); 90 ret = (xmlChar *) xmlMallocAtomic((len + 1) * sizeof(xmlChar)); 91 if (ret == NULL) { 92 xmlErrMemory(NULL, NULL); 93 return(NULL); 94 } 95 for (i = 0;i < len;i++) { 96 ret[i] = (xmlChar) cur[i]; 97 if (ret[i] == 0) return(ret); 98 } 99 ret[len] = 0; 100 return(ret); 101 } 102 103 /** 104 * xmlCharStrdup: 105 * @cur: the input char * 106 * 107 * a strdup for char's to xmlChar's 108 * 109 * Returns a new xmlChar * or NULL 110 */ 111 112 xmlChar * 113 xmlCharStrdup(const char *cur) { 114 const char *p = cur; 115 116 if (cur == NULL) return(NULL); 117 while (*p != '\0') p++; /* non input consuming */ 118 return(xmlCharStrndup(cur, p - cur)); 119 } 120 121 /** 122 * xmlStrcmp: 123 * @str1: the first xmlChar * 124 * @str2: the second xmlChar * 125 * 126 * a strcmp for xmlChar's 127 * 128 * Returns the integer result of the comparison 129 */ 130 131 int 132 xmlStrcmp(const xmlChar *str1, const xmlChar *str2) { 133 register int tmp; 134 135 if (str1 == str2) return(0); 136 if (str1 == NULL) return(-1); 137 if (str2 == NULL) return(1); 138 do { 139 tmp = *str1++ - *str2; 140 if (tmp != 0) return(tmp); 141 } while (*str2++ != 0); 142 return 0; 143 } 144 145 /** 146 * xmlStrEqual: 147 * @str1: the first xmlChar * 148 * @str2: the second xmlChar * 149 * 150 * Check if both strings are equal of have same content. 151 * Should be a bit more readable and faster than xmlStrcmp() 152 * 153 * Returns 1 if they are equal, 0 if they are different 154 */ 155 156 int 157 xmlStrEqual(const xmlChar *str1, const xmlChar *str2) { 158 if (str1 == str2) return(1); 159 if (str1 == NULL) return(0); 160 if (str2 == NULL) return(0); 161 do { 162 if (*str1++ != *str2) return(0); 163 } while (*str2++); 164 return(1); 165 } 166 167 /** 168 * xmlStrQEqual: 169 * @pref: the prefix of the QName 170 * @name: the localname of the QName 171 * @str: the second xmlChar * 172 * 173 * Check if a QName is Equal to a given string 174 * 175 * Returns 1 if they are equal, 0 if they are different 176 */ 177 178 int 179 xmlStrQEqual(const xmlChar *pref, const xmlChar *name, const xmlChar *str) { 180 if (pref == NULL) return(xmlStrEqual(name, str)); 181 if (name == NULL) return(0); 182 if (str == NULL) return(0); 183 184 do { 185 if (*pref++ != *str) return(0); 186 } while ((*str++) && (*pref)); 187 if (*str++ != ':') return(0); 188 do { 189 if (*name++ != *str) return(0); 190 } while (*str++); 191 return(1); 192 } 193 194 /** 195 * xmlStrncmp: 196 * @str1: the first xmlChar * 197 * @str2: the second xmlChar * 198 * @len: the max comparison length 199 * 200 * a strncmp for xmlChar's 201 * 202 * Returns the integer result of the comparison 203 */ 204 205 int 206 xmlStrncmp(const xmlChar *str1, const xmlChar *str2, int len) { 207 register int tmp; 208 209 if (len <= 0) return(0); 210 if (str1 == str2) return(0); 211 if (str1 == NULL) return(-1); 212 if (str2 == NULL) return(1); 213 #ifdef __GNUC__ 214 tmp = strncmp((const char *)str1, (const char *)str2, len); 215 return tmp; 216 #else 217 do { 218 tmp = *str1++ - *str2; 219 if (tmp != 0 || --len == 0) return(tmp); 220 } while (*str2++ != 0); 221 return 0; 222 #endif 223 } 224 225 static const xmlChar casemap[256] = { 226 0x00,0x01,0x02,0x03,0x04,0x05,0x06,0x07, 227 0x08,0x09,0x0A,0x0B,0x0C,0x0D,0x0E,0x0F, 228 0x10,0x11,0x12,0x13,0x14,0x15,0x16,0x17, 229 0x18,0x19,0x1A,0x1B,0x1C,0x1D,0x1E,0x1F, 230 0x20,0x21,0x22,0x23,0x24,0x25,0x26,0x27, 231 0x28,0x29,0x2A,0x2B,0x2C,0x2D,0x2E,0x2F, 232 0x30,0x31,0x32,0x33,0x34,0x35,0x36,0x37, 233 0x38,0x39,0x3A,0x3B,0x3C,0x3D,0x3E,0x3F, 234 0x40,0x61,0x62,0x63,0x64,0x65,0x66,0x67, 235 0x68,0x69,0x6A,0x6B,0x6C,0x6D,0x6E,0x6F, 236 0x70,0x71,0x72,0x73,0x74,0x75,0x76,0x77, 237 0x78,0x79,0x7A,0x7B,0x5C,0x5D,0x5E,0x5F, 238 0x60,0x61,0x62,0x63,0x64,0x65,0x66,0x67, 239 0x68,0x69,0x6A,0x6B,0x6C,0x6D,0x6E,0x6F, 240 0x70,0x71,0x72,0x73,0x74,0x75,0x76,0x77, 241 0x78,0x79,0x7A,0x7B,0x7C,0x7D,0x7E,0x7F, 242 0x80,0x81,0x82,0x83,0x84,0x85,0x86,0x87, 243 0x88,0x89,0x8A,0x8B,0x8C,0x8D,0x8E,0x8F, 244 0x90,0x91,0x92,0x93,0x94,0x95,0x96,0x97, 245 0x98,0x99,0x9A,0x9B,0x9C,0x9D,0x9E,0x9F, 246 0xA0,0xA1,0xA2,0xA3,0xA4,0xA5,0xA6,0xA7, 247 0xA8,0xA9,0xAA,0xAB,0xAC,0xAD,0xAE,0xAF, 248 0xB0,0xB1,0xB2,0xB3,0xB4,0xB5,0xB6,0xB7, 249 0xB8,0xB9,0xBA,0xBB,0xBC,0xBD,0xBE,0xBF, 250 0xC0,0xC1,0xC2,0xC3,0xC4,0xC5,0xC6,0xC7, 251 0xC8,0xC9,0xCA,0xCB,0xCC,0xCD,0xCE,0xCF, 252 0xD0,0xD1,0xD2,0xD3,0xD4,0xD5,0xD6,0xD7, 253 0xD8,0xD9,0xDA,0xDB,0xDC,0xDD,0xDE,0xDF, 254 0xE0,0xE1,0xE2,0xE3,0xE4,0xE5,0xE6,0xE7, 255 0xE8,0xE9,0xEA,0xEB,0xEC,0xED,0xEE,0xEF, 256 0xF0,0xF1,0xF2,0xF3,0xF4,0xF5,0xF6,0xF7, 257 0xF8,0xF9,0xFA,0xFB,0xFC,0xFD,0xFE,0xFF 258 }; 259 260 /** 261 * xmlStrcasecmp: 262 * @str1: the first xmlChar * 263 * @str2: the second xmlChar * 264 * 265 * a strcasecmp for xmlChar's 266 * 267 * Returns the integer result of the comparison 268 */ 269 270 int 271 xmlStrcasecmp(const xmlChar *str1, const xmlChar *str2) { 272 register int tmp; 273 274 if (str1 == str2) return(0); 275 if (str1 == NULL) return(-1); 276 if (str2 == NULL) return(1); 277 do { 278 tmp = casemap[*str1++] - casemap[*str2]; 279 if (tmp != 0) return(tmp); 280 } while (*str2++ != 0); 281 return 0; 282 } 283 284 /** 285 * xmlStrncasecmp: 286 * @str1: the first xmlChar * 287 * @str2: the second xmlChar * 288 * @len: the max comparison length 289 * 290 * a strncasecmp for xmlChar's 291 * 292 * Returns the integer result of the comparison 293 */ 294 295 int 296 xmlStrncasecmp(const xmlChar *str1, const xmlChar *str2, int len) { 297 register int tmp; 298 299 if (len <= 0) return(0); 300 if (str1 == str2) return(0); 301 if (str1 == NULL) return(-1); 302 if (str2 == NULL) return(1); 303 do { 304 tmp = casemap[*str1++] - casemap[*str2]; 305 if (tmp != 0 || --len == 0) return(tmp); 306 } while (*str2++ != 0); 307 return 0; 308 } 309 310 /** 311 * xmlStrchr: 312 * @str: the xmlChar * array 313 * @val: the xmlChar to search 314 * 315 * a strchr for xmlChar's 316 * 317 * Returns the xmlChar * for the first occurrence or NULL. 318 */ 319 320 const xmlChar * 321 xmlStrchr(const xmlChar *str, xmlChar val) { 322 if (str == NULL) return(NULL); 323 while (*str != 0) { /* non input consuming */ 324 if (*str == val) return((xmlChar *) str); 325 str++; 326 } 327 return(NULL); 328 } 329 330 /** 331 * xmlStrstr: 332 * @str: the xmlChar * array (haystack) 333 * @val: the xmlChar to search (needle) 334 * 335 * a strstr for xmlChar's 336 * 337 * Returns the xmlChar * for the first occurrence or NULL. 338 */ 339 340 const xmlChar * 341 xmlStrstr(const xmlChar *str, const xmlChar *val) { 342 int n; 343 344 if (str == NULL) return(NULL); 345 if (val == NULL) return(NULL); 346 n = xmlStrlen(val); 347 348 if (n == 0) return(str); 349 while (*str != 0) { /* non input consuming */ 350 if (*str == *val) { 351 if (!xmlStrncmp(str, val, n)) return((const xmlChar *) str); 352 } 353 str++; 354 } 355 return(NULL); 356 } 357 358 /** 359 * xmlStrcasestr: 360 * @str: the xmlChar * array (haystack) 361 * @val: the xmlChar to search (needle) 362 * 363 * a case-ignoring strstr for xmlChar's 364 * 365 * Returns the xmlChar * for the first occurrence or NULL. 366 */ 367 368 const xmlChar * 369 xmlStrcasestr(const xmlChar *str, const xmlChar *val) { 370 int n; 371 372 if (str == NULL) return(NULL); 373 if (val == NULL) return(NULL); 374 n = xmlStrlen(val); 375 376 if (n == 0) return(str); 377 while (*str != 0) { /* non input consuming */ 378 if (casemap[*str] == casemap[*val]) 379 if (!xmlStrncasecmp(str, val, n)) return(str); 380 str++; 381 } 382 return(NULL); 383 } 384 385 /** 386 * xmlStrsub: 387 * @str: the xmlChar * array (haystack) 388 * @start: the index of the first char (zero based) 389 * @len: the length of the substring 390 * 391 * Extract a substring of a given string 392 * 393 * Returns the xmlChar * for the first occurrence or NULL. 394 */ 395 396 xmlChar * 397 xmlStrsub(const xmlChar *str, int start, int len) { 398 int i; 399 400 if (str == NULL) return(NULL); 401 if (start < 0) return(NULL); 402 if (len < 0) return(NULL); 403 404 for (i = 0;i < start;i++) { 405 if (*str == 0) return(NULL); 406 str++; 407 } 408 if (*str == 0) return(NULL); 409 return(xmlStrndup(str, len)); 410 } 411 412 /** 413 * xmlStrlen: 414 * @str: the xmlChar * array 415 * 416 * length of a xmlChar's string 417 * 418 * Returns the number of xmlChar contained in the ARRAY. 419 */ 420 421 int 422 xmlStrlen(const xmlChar *str) { 423 int len = 0; 424 425 if (str == NULL) return(0); 426 while (*str != 0) { /* non input consuming */ 427 str++; 428 len++; 429 } 430 return(len); 431 } 432 433 /** 434 * xmlStrncat: 435 * @cur: the original xmlChar * array 436 * @add: the xmlChar * array added 437 * @len: the length of @add 438 * 439 * a strncat for array of xmlChar's, it will extend @cur with the len 440 * first bytes of @add. Note that if @len < 0 then this is an API error 441 * and NULL will be returned. 442 * 443 * Returns a new xmlChar *, the original @cur is reallocated and should 444 * not be freed. 445 */ 446 447 xmlChar * 448 xmlStrncat(xmlChar *cur, const xmlChar *add, int len) { 449 int size; 450 xmlChar *ret; 451 452 if ((add == NULL) || (len == 0)) 453 return(cur); 454 if (len < 0) 455 return(NULL); 456 if (cur == NULL) 457 return(xmlStrndup(add, len)); 458 459 size = xmlStrlen(cur); 460 if (size < 0) 461 return(NULL); 462 ret = (xmlChar *) xmlRealloc(cur, (size + len + 1) * sizeof(xmlChar)); 463 if (ret == NULL) { 464 xmlErrMemory(NULL, NULL); 465 return(cur); 466 } 467 memcpy(&ret[size], add, len * sizeof(xmlChar)); 468 ret[size + len] = 0; 469 return(ret); 470 } 471 472 /** 473 * xmlStrncatNew: 474 * @str1: first xmlChar string 475 * @str2: second xmlChar string 476 * @len: the len of @str2 or < 0 477 * 478 * same as xmlStrncat, but creates a new string. The original 479 * two strings are not freed. If @len is < 0 then the length 480 * will be calculated automatically. 481 * 482 * Returns a new xmlChar * or NULL 483 */ 484 xmlChar * 485 xmlStrncatNew(const xmlChar *str1, const xmlChar *str2, int len) { 486 int size; 487 xmlChar *ret; 488 489 if (len < 0) { 490 len = xmlStrlen(str2); 491 if (len < 0) 492 return(NULL); 493 } 494 if ((str2 == NULL) || (len == 0)) 495 return(xmlStrdup(str1)); 496 if (str1 == NULL) 497 return(xmlStrndup(str2, len)); 498 499 size = xmlStrlen(str1); 500 if (size < 0) 501 return(NULL); 502 ret = (xmlChar *) xmlMalloc((size + len + 1) * sizeof(xmlChar)); 503 if (ret == NULL) { 504 xmlErrMemory(NULL, NULL); 505 return(xmlStrndup(str1, size)); 506 } 507 memcpy(ret, str1, size * sizeof(xmlChar)); 508 memcpy(&ret[size], str2, len * sizeof(xmlChar)); 509 ret[size + len] = 0; 510 return(ret); 511 } 512 513 /** 514 * xmlStrcat: 515 * @cur: the original xmlChar * array 516 * @add: the xmlChar * array added 517 * 518 * a strcat for array of xmlChar's. Since they are supposed to be 519 * encoded in UTF-8 or an encoding with 8bit based chars, we assume 520 * a termination mark of '0'. 521 * 522 * Returns a new xmlChar * containing the concatenated string. The original 523 * @cur is reallocated and should not be freed. 524 */ 525 xmlChar * 526 xmlStrcat(xmlChar *cur, const xmlChar *add) { 527 const xmlChar *p = add; 528 529 if (add == NULL) return(cur); 530 if (cur == NULL) 531 return(xmlStrdup(add)); 532 533 while (*p != 0) p++; /* non input consuming */ 534 return(xmlStrncat(cur, add, p - add)); 535 } 536 537 /** 538 * xmlStrPrintf: 539 * @buf: the result buffer. 540 * @len: the result buffer length. 541 * @msg: the message with printf formatting. 542 * @...: extra parameters for the message. 543 * 544 * Formats @msg and places result into @buf. 545 * 546 * Returns the number of characters written to @buf or -1 if an error occurs. 547 */ 548 int XMLCDECL 549 xmlStrPrintf(xmlChar *buf, int len, const char *msg, ...) { 550 va_list args; 551 int ret; 552 553 if((buf == NULL) || (msg == NULL)) { 554 return(-1); 555 } 556 557 va_start(args, msg); 558 ret = vsnprintf((char *) buf, len, (const char *) msg, args); 559 va_end(args); 560 buf[len - 1] = 0; /* be safe ! */ 561 562 return(ret); 563 } 564 565 /** 566 * xmlStrVPrintf: 567 * @buf: the result buffer. 568 * @len: the result buffer length. 569 * @msg: the message with printf formatting. 570 * @ap: extra parameters for the message. 571 * 572 * Formats @msg and places result into @buf. 573 * 574 * Returns the number of characters written to @buf or -1 if an error occurs. 575 */ 576 int 577 xmlStrVPrintf(xmlChar *buf, int len, const char *msg, va_list ap) { 578 int ret; 579 580 if((buf == NULL) || (msg == NULL)) { 581 return(-1); 582 } 583 584 ret = vsnprintf((char *) buf, len, (const char *) msg, ap); 585 buf[len - 1] = 0; /* be safe ! */ 586 587 return(ret); 588 } 589 590 /************************************************************************ 591 * * 592 * Generic UTF8 handling routines * 593 * * 594 * From rfc2044: encoding of the Unicode values on UTF-8: * 595 * * 596 * UCS-4 range (hex.) UTF-8 octet sequence (binary) * 597 * 0000 0000-0000 007F 0xxxxxxx * 598 * 0000 0080-0000 07FF 110xxxxx 10xxxxxx * 599 * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx * 600 * * 601 * I hope we won't use values > 0xFFFF anytime soon ! * 602 * * 603 ************************************************************************/ 604 605 606 /** 607 * xmlUTF8Size: 608 * @utf: pointer to the UTF8 character 609 * 610 * calculates the internal size of a UTF8 character 611 * 612 * returns the numbers of bytes in the character, -1 on format error 613 */ 614 int 615 xmlUTF8Size(const xmlChar *utf) { 616 xmlChar mask; 617 int len; 618 619 if (utf == NULL) 620 return -1; 621 if (*utf < 0x80) 622 return 1; 623 /* check valid UTF8 character */ 624 if (!(*utf & 0x40)) 625 return -1; 626 /* determine number of bytes in char */ 627 len = 2; 628 for (mask=0x20; mask != 0; mask>>=1) { 629 if (!(*utf & mask)) 630 return len; 631 len++; 632 } 633 return -1; 634 } 635 636 /** 637 * xmlUTF8Charcmp: 638 * @utf1: pointer to first UTF8 char 639 * @utf2: pointer to second UTF8 char 640 * 641 * compares the two UCS4 values 642 * 643 * returns result of the compare as with xmlStrncmp 644 */ 645 int 646 xmlUTF8Charcmp(const xmlChar *utf1, const xmlChar *utf2) { 647 648 if (utf1 == NULL ) { 649 if (utf2 == NULL) 650 return 0; 651 return -1; 652 } 653 return xmlStrncmp(utf1, utf2, xmlUTF8Size(utf1)); 654 } 655 656 /** 657 * xmlUTF8Strlen: 658 * @utf: a sequence of UTF-8 encoded bytes 659 * 660 * compute the length of an UTF8 string, it doesn't do a full UTF8 661 * checking of the content of the string. 662 * 663 * Returns the number of characters in the string or -1 in case of error 664 */ 665 int 666 xmlUTF8Strlen(const xmlChar *utf) { 667 int ret = 0; 668 669 if (utf == NULL) 670 return(-1); 671 672 while (*utf != 0) { 673 if (utf[0] & 0x80) { 674 if ((utf[1] & 0xc0) != 0x80) 675 return(-1); 676 if ((utf[0] & 0xe0) == 0xe0) { 677 if ((utf[2] & 0xc0) != 0x80) 678 return(-1); 679 if ((utf[0] & 0xf0) == 0xf0) { 680 if ((utf[0] & 0xf8) != 0xf0 || (utf[3] & 0xc0) != 0x80) 681 return(-1); 682 utf += 4; 683 } else { 684 utf += 3; 685 } 686 } else { 687 utf += 2; 688 } 689 } else { 690 utf++; 691 } 692 ret++; 693 } 694 return(ret); 695 } 696 697 /** 698 * xmlGetUTF8Char: 699 * @utf: a sequence of UTF-8 encoded bytes 700 * @len: a pointer to the minimum number of bytes present in 701 * the sequence. This is used to assure the next character 702 * is completely contained within the sequence. 703 * 704 * Read the first UTF8 character from @utf 705 * 706 * Returns the char value or -1 in case of error, and sets *len to 707 * the actual number of bytes consumed (0 in case of error) 708 */ 709 int 710 xmlGetUTF8Char(const unsigned char *utf, int *len) { 711 unsigned int c; 712 713 if (utf == NULL) 714 goto error; 715 if (len == NULL) 716 goto error; 717 if (*len < 1) 718 goto error; 719 720 c = utf[0]; 721 if (c & 0x80) { 722 if (*len < 2) 723 goto error; 724 if ((utf[1] & 0xc0) != 0x80) 725 goto error; 726 if ((c & 0xe0) == 0xe0) { 727 if (*len < 3) 728 goto error; 729 if ((utf[2] & 0xc0) != 0x80) 730 goto error; 731 if ((c & 0xf0) == 0xf0) { 732 if (*len < 4) 733 goto error; 734 if ((c & 0xf8) != 0xf0 || (utf[3] & 0xc0) != 0x80) 735 goto error; 736 *len = 4; 737 /* 4-byte code */ 738 c = (utf[0] & 0x7) << 18; 739 c |= (utf[1] & 0x3f) << 12; 740 c |= (utf[2] & 0x3f) << 6; 741 c |= utf[3] & 0x3f; 742 } else { 743 /* 3-byte code */ 744 *len = 3; 745 c = (utf[0] & 0xf) << 12; 746 c |= (utf[1] & 0x3f) << 6; 747 c |= utf[2] & 0x3f; 748 } 749 } else { 750 /* 2-byte code */ 751 *len = 2; 752 c = (utf[0] & 0x1f) << 6; 753 c |= utf[1] & 0x3f; 754 } 755 } else { 756 /* 1-byte code */ 757 *len = 1; 758 } 759 return(c); 760 761 error: 762 if (len != NULL) 763 *len = 0; 764 return(-1); 765 } 766 767 /** 768 * xmlCheckUTF8: 769 * @utf: Pointer to putative UTF-8 encoded string. 770 * 771 * Checks @utf for being valid UTF-8. @utf is assumed to be 772 * null-terminated. This function is not super-strict, as it will 773 * allow longer UTF-8 sequences than necessary. Note that Java is 774 * capable of producing these sequences if provoked. Also note, this 775 * routine checks for the 4-byte maximum size, but does not check for 776 * 0x10ffff maximum value. 777 * 778 * Return value: true if @utf is valid. 779 **/ 780 int 781 xmlCheckUTF8(const unsigned char *utf) 782 { 783 int ix; 784 unsigned char c; 785 786 if (utf == NULL) 787 return(0); 788 /* 789 * utf is a string of 1, 2, 3 or 4 bytes. The valid strings 790 * are as follows (in "bit format"): 791 * 0xxxxxxx valid 1-byte 792 * 110xxxxx 10xxxxxx valid 2-byte 793 * 1110xxxx 10xxxxxx 10xxxxxx valid 3-byte 794 * 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx valid 4-byte 795 */ 796 for (ix = 0; (c = utf[ix]);) { /* string is 0-terminated */ 797 if ((c & 0x80) == 0x00) { /* 1-byte code, starts with 10 */ 798 ix++; 799 } else if ((c & 0xe0) == 0xc0) {/* 2-byte code, starts with 110 */ 800 if ((utf[ix+1] & 0xc0 ) != 0x80) 801 return 0; 802 ix += 2; 803 } else if ((c & 0xf0) == 0xe0) {/* 3-byte code, starts with 1110 */ 804 if (((utf[ix+1] & 0xc0) != 0x80) || 805 ((utf[ix+2] & 0xc0) != 0x80)) 806 return 0; 807 ix += 3; 808 } else if ((c & 0xf8) == 0xf0) {/* 4-byte code, starts with 11110 */ 809 if (((utf[ix+1] & 0xc0) != 0x80) || 810 ((utf[ix+2] & 0xc0) != 0x80) || 811 ((utf[ix+3] & 0xc0) != 0x80)) 812 return 0; 813 ix += 4; 814 } else /* unknown encoding */ 815 return 0; 816 } 817 return(1); 818 } 819 820 /** 821 * xmlUTF8Strsize: 822 * @utf: a sequence of UTF-8 encoded bytes 823 * @len: the number of characters in the array 824 * 825 * storage size of an UTF8 string 826 * the behaviour is not guaranteed if the input string is not UTF-8 827 * 828 * Returns the storage size of 829 * the first 'len' characters of ARRAY 830 */ 831 832 int 833 xmlUTF8Strsize(const xmlChar *utf, int len) { 834 const xmlChar *ptr=utf; 835 xmlChar ch; 836 837 if (utf == NULL) 838 return(0); 839 840 if (len <= 0) 841 return(0); 842 843 while ( len-- > 0) { 844 if ( !*ptr ) 845 break; 846 if ( (ch = *ptr++) & 0x80) 847 while ((ch<<=1) & 0x80 ) { 848 if (*ptr == 0) break; 849 ptr++; 850 } 851 } 852 return (ptr - utf); 853 } 854 855 856 /** 857 * xmlUTF8Strndup: 858 * @utf: the input UTF8 * 859 * @len: the len of @utf (in chars) 860 * 861 * a strndup for array of UTF8's 862 * 863 * Returns a new UTF8 * or NULL 864 */ 865 xmlChar * 866 xmlUTF8Strndup(const xmlChar *utf, int len) { 867 xmlChar *ret; 868 int i; 869 870 if ((utf == NULL) || (len < 0)) return(NULL); 871 i = xmlUTF8Strsize(utf, len); 872 ret = (xmlChar *) xmlMallocAtomic((i + 1) * sizeof(xmlChar)); 873 if (ret == NULL) { 874 xmlGenericError(xmlGenericErrorContext, 875 "malloc of %ld byte failed\n", 876 (len + 1) * (long)sizeof(xmlChar)); 877 return(NULL); 878 } 879 memcpy(ret, utf, i * sizeof(xmlChar)); 880 ret[i] = 0; 881 return(ret); 882 } 883 884 /** 885 * xmlUTF8Strpos: 886 * @utf: the input UTF8 * 887 * @pos: the position of the desired UTF8 char (in chars) 888 * 889 * a function to provide the equivalent of fetching a 890 * character from a string array 891 * 892 * Returns a pointer to the UTF8 character or NULL 893 */ 894 const xmlChar * 895 xmlUTF8Strpos(const xmlChar *utf, int pos) { 896 xmlChar ch; 897 898 if (utf == NULL) return(NULL); 899 if (pos < 0) 900 return(NULL); 901 while (pos--) { 902 if ((ch=*utf++) == 0) return(NULL); 903 if ( ch & 0x80 ) { 904 /* if not simple ascii, verify proper format */ 905 if ( (ch & 0xc0) != 0xc0 ) 906 return(NULL); 907 /* then skip over remaining bytes for this char */ 908 while ( (ch <<= 1) & 0x80 ) 909 if ( (*utf++ & 0xc0) != 0x80 ) 910 return(NULL); 911 } 912 } 913 return((xmlChar *)utf); 914 } 915 916 /** 917 * xmlUTF8Strloc: 918 * @utf: the input UTF8 * 919 * @utfchar: the UTF8 character to be found 920 * 921 * a function to provide the relative location of a UTF8 char 922 * 923 * Returns the relative character position of the desired char 924 * or -1 if not found 925 */ 926 int 927 xmlUTF8Strloc(const xmlChar *utf, const xmlChar *utfchar) { 928 int i, size; 929 xmlChar ch; 930 931 if (utf==NULL || utfchar==NULL) return -1; 932 size = xmlUTF8Strsize(utfchar, 1); 933 for(i=0; (ch=*utf) != 0; i++) { 934 if (xmlStrncmp(utf, utfchar, size)==0) 935 return(i); 936 utf++; 937 if ( ch & 0x80 ) { 938 /* if not simple ascii, verify proper format */ 939 if ( (ch & 0xc0) != 0xc0 ) 940 return(-1); 941 /* then skip over remaining bytes for this char */ 942 while ( (ch <<= 1) & 0x80 ) 943 if ( (*utf++ & 0xc0) != 0x80 ) 944 return(-1); 945 } 946 } 947 948 return(-1); 949 } 950 /** 951 * xmlUTF8Strsub: 952 * @utf: a sequence of UTF-8 encoded bytes 953 * @start: relative pos of first char 954 * @len: total number to copy 955 * 956 * Create a substring from a given UTF-8 string 957 * Note: positions are given in units of UTF-8 chars 958 * 959 * Returns a pointer to a newly created string 960 * or NULL if any problem 961 */ 962 963 xmlChar * 964 xmlUTF8Strsub(const xmlChar *utf, int start, int len) { 965 int i; 966 xmlChar ch; 967 968 if (utf == NULL) return(NULL); 969 if (start < 0) return(NULL); 970 if (len < 0) return(NULL); 971 972 /* 973 * Skip over any leading chars 974 */ 975 for (i = 0;i < start;i++) { 976 if ((ch=*utf++) == 0) return(NULL); 977 if ( ch & 0x80 ) { 978 /* if not simple ascii, verify proper format */ 979 if ( (ch & 0xc0) != 0xc0 ) 980 return(NULL); 981 /* then skip over remaining bytes for this char */ 982 while ( (ch <<= 1) & 0x80 ) 983 if ( (*utf++ & 0xc0) != 0x80 ) 984 return(NULL); 985 } 986 } 987 988 return(xmlUTF8Strndup(utf, len)); 989 } 990 991 /** 992 * xmlEscapeFormatString: 993 * @msg: a pointer to the string in which to escape '%' characters. 994 * Must be a heap-allocated buffer created by libxml2 that may be 995 * returned, or that may be freed and replaced. 996 * 997 * Replaces the string pointed to by 'msg' with an escaped string. 998 * Returns the same string with all '%' characters escaped. 999 */ 1000 xmlChar * 1001 xmlEscapeFormatString(xmlChar **msg) 1002 { 1003 xmlChar *msgPtr = NULL; 1004 xmlChar *result = NULL; 1005 xmlChar *resultPtr = NULL; 1006 size_t count = 0; 1007 size_t msgLen = 0; 1008 size_t resultLen = 0; 1009 1010 if (!msg || !*msg) 1011 return(NULL); 1012 1013 for (msgPtr = *msg; *msgPtr != '\0'; ++msgPtr) { 1014 ++msgLen; 1015 if (*msgPtr == '%') 1016 ++count; 1017 } 1018 1019 if (count == 0) 1020 return(*msg); 1021 1022 resultLen = msgLen + count + 1; 1023 result = (xmlChar *) xmlMallocAtomic(resultLen * sizeof(xmlChar)); 1024 if (result == NULL) { 1025 /* Clear *msg to prevent format string vulnerabilities in 1026 out-of-memory situations. */ 1027 xmlFree(*msg); 1028 *msg = NULL; 1029 xmlErrMemory(NULL, NULL); 1030 return(NULL); 1031 } 1032 1033 for (msgPtr = *msg, resultPtr = result; *msgPtr != '\0'; ++msgPtr, ++resultPtr) { 1034 *resultPtr = *msgPtr; 1035 if (*msgPtr == '%') 1036 *(++resultPtr) = '%'; 1037 } 1038 result[resultLen - 1] = '\0'; 1039 1040 xmlFree(*msg); 1041 *msg = result; 1042 1043 return *msg; 1044 } 1045 1046 #define bottom_xmlstring 1047 #include "elfgcchack.h" 1048
