1 #include "buffer.h" 2 3 #include <stdlib.h> 4 #include <string.h> 5 6 #include <stdio.h> 7 #include <assert.h> 8 #include <ctype.h> 9 10 11 static const char hex_chars[] = "0123456789abcdef"; 12 13 14 /** 15 * init the buffer 16 * 17 */ 18 19 buffer* buffer_init(void) { 20 buffer *b; 21 22 b = malloc(sizeof(*b)); 23 assert(b); 24 25 b->ptr = NULL; 26 b->size = 0; 27 b->used = 0; 28 29 return b; 30 } 31 32 buffer *buffer_init_buffer(buffer *src) { 33 buffer *b = buffer_init(); 34 buffer_copy_string_buffer(b, src); 35 return b; 36 } 37 38 /** 39 * free the buffer 40 * 41 */ 42 43 void buffer_free(buffer *b) { 44 if (!b) return; 45 46 free(b->ptr); 47 free(b); 48 } 49 50 void buffer_reset(buffer *b) { 51 if (!b) return; 52 53 /* limit don't reuse buffer larger than ... bytes */ 54 if (b->size > BUFFER_MAX_REUSE_SIZE) { 55 free(b->ptr); 56 b->ptr = NULL; 57 b->size = 0; 58 } else if (b->size) { 59 b->ptr[0] = '\0'; 60 } 61 62 b->used = 0; 63 } 64 65 66 /** 67 * 68 * allocate (if neccessary) enough space for 'size' bytes and 69 * set the 'used' counter to 0 70 * 71 */ 72 73 #define BUFFER_PIECE_SIZE 64 74 75 int buffer_prepare_copy(buffer *b, size_t size) { 76 if (!b) return -1; 77 78 if ((0 == b->size) || 79 (size > b->size)) { 80 if (b->size) free(b->ptr); 81 82 b->size = size; 83 84 /* always allocate a multiply of BUFFER_PIECE_SIZE */ 85 b->size += BUFFER_PIECE_SIZE - (b->size % BUFFER_PIECE_SIZE); 86 87 b->ptr = malloc(b->size); 88 assert(b->ptr); 89 } 90 b->used = 0; 91 return 0; 92 } 93 94 /** 95 * 96 * increase the internal buffer (if neccessary) to append another 'size' byte 97 * ->used isn't changed 98 * 99 */ 100 101 int buffer_prepare_append(buffer *b, size_t size) { 102 if (!b) return -1; 103 104 if (0 == b->size) { 105 b->size = size; 106 107 /* always allocate a multiply of BUFFER_PIECE_SIZE */ 108 b->size += BUFFER_PIECE_SIZE - (b->size % BUFFER_PIECE_SIZE); 109 110 b->ptr = malloc(b->size); 111 b->used = 0; 112 assert(b->ptr); 113 } else if (b->used + size > b->size) { 114 b->size += size; 115 116 /* always allocate a multiply of BUFFER_PIECE_SIZE */ 117 b->size += BUFFER_PIECE_SIZE - (b->size % BUFFER_PIECE_SIZE); 118 119 b->ptr = realloc(b->ptr, b->size); 120 assert(b->ptr); 121 } 122 return 0; 123 } 124 125 int buffer_copy_string(buffer *b, const char *s) { 126 size_t s_len; 127 128 if (!s || !b) return -1; 129 130 s_len = strlen(s) + 1; 131 buffer_prepare_copy(b, s_len); 132 133 memcpy(b->ptr, s, s_len); 134 b->used = s_len; 135 136 return 0; 137 } 138 139 int buffer_copy_string_len(buffer *b, const char *s, size_t s_len) { 140 if (!s || !b) return -1; 141 #if 0 142 /* removed optimization as we have to keep the empty string 143 * in some cases for the config handling 144 * 145 * url.access-deny = ( "" ) 146 */ 147 if (s_len == 0) return 0; 148 #endif 149 buffer_prepare_copy(b, s_len + 1); 150 151 memcpy(b->ptr, s, s_len); 152 b->ptr[s_len] = '\0'; 153 b->used = s_len + 1; 154 155 return 0; 156 } 157 158 int buffer_copy_string_buffer(buffer *b, const buffer *src) { 159 if (!src) return -1; 160 161 if (src->used == 0) { 162 buffer_reset(b); 163 return 0; 164 } 165 return buffer_copy_string_len(b, src->ptr, src->used - 1); 166 } 167 168 int buffer_append_string(buffer *b, const char *s) { 169 size_t s_len; 170 171 if (!s || !b) return -1; 172 173 s_len = strlen(s); 174 buffer_prepare_append(b, s_len + 1); 175 if (b->used == 0) 176 b->used++; 177 178 memcpy(b->ptr + b->used - 1, s, s_len + 1); 179 b->used += s_len; 180 181 return 0; 182 } 183 184 int buffer_append_string_rfill(buffer *b, const char *s, size_t maxlen) { 185 size_t s_len; 186 187 if (!s || !b) return -1; 188 189 s_len = strlen(s); 190 if (s_len > maxlen) s_len = maxlen; 191 buffer_prepare_append(b, maxlen + 1); 192 if (b->used == 0) 193 b->used++; 194 195 memcpy(b->ptr + b->used - 1, s, s_len); 196 if (maxlen > s_len) { 197 memset(b->ptr + b->used - 1 + s_len, ' ', maxlen - s_len); 198 } 199 200 b->used += maxlen; 201 b->ptr[b->used - 1] = '\0'; 202 return 0; 203 } 204 205 /** 206 * append a string to the end of the buffer 207 * 208 * the resulting buffer is terminated with a '\0' 209 * s is treated as a un-terminated string (a \0 is handled a normal character) 210 * 211 * @param b a buffer 212 * @param s the string 213 * @param s_len size of the string (without the terminating \0) 214 */ 215 216 int buffer_append_string_len(buffer *b, const char *s, size_t s_len) { 217 if (!s || !b) return -1; 218 if (s_len == 0) return 0; 219 220 buffer_prepare_append(b, s_len + 1); 221 if (b->used == 0) 222 b->used++; 223 224 memcpy(b->ptr + b->used - 1, s, s_len); 225 b->used += s_len; 226 b->ptr[b->used - 1] = '\0'; 227 228 return 0; 229 } 230 231 int buffer_append_string_buffer(buffer *b, const buffer *src) { 232 if (!src) return -1; 233 if (src->used == 0) return 0; 234 235 return buffer_append_string_len(b, src->ptr, src->used - 1); 236 } 237 238 int buffer_append_memory(buffer *b, const char *s, size_t s_len) { 239 if (!s || !b) return -1; 240 if (s_len == 0) return 0; 241 242 buffer_prepare_append(b, s_len); 243 memcpy(b->ptr + b->used, s, s_len); 244 b->used += s_len; 245 246 return 0; 247 } 248 249 int buffer_copy_memory(buffer *b, const char *s, size_t s_len) { 250 if (!s || !b) return -1; 251 252 b->used = 0; 253 254 return buffer_append_memory(b, s, s_len); 255 } 256 257 int buffer_append_long_hex(buffer *b, unsigned long value) { 258 char *buf; 259 int shift = 0; 260 unsigned long copy = value; 261 262 while (copy) { 263 copy >>= 4; 264 shift++; 265 } 266 if (shift == 0) 267 shift++; 268 if (shift & 0x01) 269 shift++; 270 271 buffer_prepare_append(b, shift + 1); 272 if (b->used == 0) 273 b->used++; 274 buf = b->ptr + (b->used - 1); 275 b->used += shift; 276 277 shift <<= 2; 278 while (shift > 0) { 279 shift -= 4; 280 *(buf++) = hex_chars[(value >> shift) & 0x0F]; 281 } 282 *buf = '\0'; 283 284 return 0; 285 } 286 287 int LI_ltostr(char *buf, long val) { 288 char swap; 289 char *end; 290 int len = 1; 291 292 if (val < 0) { 293 len++; 294 *(buf++) = '-'; 295 val = -val; 296 } 297 298 end = buf; 299 while (val > 9) { 300 *(end++) = '0' + (val % 10); 301 val = val / 10; 302 } 303 *(end) = '0' + val; 304 *(end + 1) = '\0'; 305 len += end - buf; 306 307 while (buf < end) { 308 swap = *end; 309 *end = *buf; 310 *buf = swap; 311 312 buf++; 313 end--; 314 } 315 316 return len; 317 } 318 319 int buffer_append_long(buffer *b, long val) { 320 if (!b) return -1; 321 322 buffer_prepare_append(b, 32); 323 if (b->used == 0) 324 b->used++; 325 326 b->used += LI_ltostr(b->ptr + (b->used - 1), val); 327 return 0; 328 } 329 330 int buffer_copy_long(buffer *b, long val) { 331 if (!b) return -1; 332 333 b->used = 0; 334 return buffer_append_long(b, val); 335 } 336 337 #if !defined(SIZEOF_LONG) || (SIZEOF_LONG != SIZEOF_OFF_T) 338 int buffer_append_off_t(buffer *b, off_t val) { 339 char swap; 340 char *end; 341 char *start; 342 int len = 1; 343 344 if (!b) return -1; 345 346 buffer_prepare_append(b, 32); 347 if (b->used == 0) 348 b->used++; 349 350 start = b->ptr + (b->used - 1); 351 if (val < 0) { 352 len++; 353 *(start++) = '-'; 354 val = -val; 355 } 356 357 end = start; 358 while (val > 9) { 359 *(end++) = '0' + (val % 10); 360 val = val / 10; 361 } 362 *(end) = '0' + val; 363 *(end + 1) = '\0'; 364 len += end - start; 365 366 while (start < end) { 367 swap = *end; 368 *end = *start; 369 *start = swap; 370 371 start++; 372 end--; 373 } 374 375 b->used += len; 376 return 0; 377 } 378 379 int buffer_copy_off_t(buffer *b, off_t val) { 380 if (!b) return -1; 381 382 b->used = 0; 383 return buffer_append_off_t(b, val); 384 } 385 #endif /* !defined(SIZEOF_LONG) || (SIZEOF_LONG != SIZEOF_OFF_T) */ 386 387 char int2hex(char c) { 388 return hex_chars[(c & 0x0F)]; 389 } 390 391 /* converts hex char (0-9, A-Z, a-z) to decimal. 392 * returns 0xFF on invalid input. 393 */ 394 char hex2int(unsigned char hex) { 395 hex = hex - '0'; 396 if (hex > 9) { 397 hex = (hex + '0' - 1) | 0x20; 398 hex = hex - 'a' + 11; 399 } 400 if (hex > 15) 401 hex = 0xFF; 402 403 return hex; 404 } 405 406 407 /** 408 * init the buffer 409 * 410 */ 411 412 buffer_array* buffer_array_init(void) { 413 buffer_array *b; 414 415 b = malloc(sizeof(*b)); 416 417 assert(b); 418 b->ptr = NULL; 419 b->size = 0; 420 b->used = 0; 421 422 return b; 423 } 424 425 void buffer_array_reset(buffer_array *b) { 426 size_t i; 427 428 if (!b) return; 429 430 /* if they are too large, reduce them */ 431 for (i = 0; i < b->used; i++) { 432 buffer_reset(b->ptr[i]); 433 } 434 435 b->used = 0; 436 } 437 438 439 /** 440 * free the buffer_array 441 * 442 */ 443 444 void buffer_array_free(buffer_array *b) { 445 size_t i; 446 if (!b) return; 447 448 for (i = 0; i < b->size; i++) { 449 if (b->ptr[i]) buffer_free(b->ptr[i]); 450 } 451 free(b->ptr); 452 free(b); 453 } 454 455 buffer *buffer_array_append_get_buffer(buffer_array *b) { 456 size_t i; 457 458 if (b->size == 0) { 459 b->size = 16; 460 b->ptr = malloc(sizeof(*b->ptr) * b->size); 461 assert(b->ptr); 462 for (i = 0; i < b->size; i++) { 463 b->ptr[i] = NULL; 464 } 465 } else if (b->size == b->used) { 466 b->size += 16; 467 b->ptr = realloc(b->ptr, sizeof(*b->ptr) * b->size); 468 assert(b->ptr); 469 for (i = b->used; i < b->size; i++) { 470 b->ptr[i] = NULL; 471 } 472 } 473 474 if (b->ptr[b->used] == NULL) { 475 b->ptr[b->used] = buffer_init(); 476 } 477 478 b->ptr[b->used]->used = 0; 479 480 return b->ptr[b->used++]; 481 } 482 483 484 char * buffer_search_string_len(buffer *b, const char *needle, size_t len) { 485 size_t i; 486 if (len == 0) return NULL; 487 if (needle == NULL) return NULL; 488 489 if (b->used < len) return NULL; 490 491 for(i = 0; i < b->used - len; i++) { 492 if (0 == memcmp(b->ptr + i, needle, len)) { 493 return b->ptr + i; 494 } 495 } 496 497 return NULL; 498 } 499 500 buffer *buffer_init_string(const char *str) { 501 buffer *b = buffer_init(); 502 503 buffer_copy_string(b, str); 504 505 return b; 506 } 507 508 int buffer_is_empty(buffer *b) { 509 if (!b) return 1; 510 return (b->used == 0); 511 } 512 513 /** 514 * check if two buffer contain the same data 515 * 516 * HISTORY: this function was pretty much optimized, but didn't handled 517 * alignment properly. 518 */ 519 520 int buffer_is_equal(buffer *a, buffer *b) { 521 if (a->used != b->used) return 0; 522 if (a->used == 0) return 1; 523 524 return (0 == strcmp(a->ptr, b->ptr)); 525 } 526 527 int buffer_is_equal_string(buffer *a, const char *s, size_t b_len) { 528 buffer b; 529 530 b.ptr = (char *)s; 531 b.used = b_len + 1; 532 533 return buffer_is_equal(a, &b); 534 } 535 536 /* simple-assumption: 537 * 538 * most parts are equal and doing a case conversion needs time 539 * 540 */ 541 int buffer_caseless_compare(const char *a, size_t a_len, const char *b, size_t b_len) { 542 size_t ndx = 0, max_ndx; 543 size_t *al, *bl; 544 size_t mask = sizeof(*al) - 1; 545 546 al = (size_t *)a; 547 bl = (size_t *)b; 548 549 /* is the alignment correct ? */ 550 if ( ((size_t)al & mask) == 0 && 551 ((size_t)bl & mask) == 0 ) { 552 553 max_ndx = ((a_len < b_len) ? a_len : b_len) & ~mask; 554 555 for (; ndx < max_ndx; ndx += sizeof(*al)) { 556 if (*al != *bl) break; 557 al++; bl++; 558 559 } 560 561 } 562 563 a = (char *)al; 564 b = (char *)bl; 565 566 max_ndx = ((a_len < b_len) ? a_len : b_len); 567 568 for (; ndx < max_ndx; ndx++) { 569 char a1 = *a++, b1 = *b++; 570 571 if (a1 != b1) { 572 if ((a1 >= 'A' && a1 <= 'Z') && (b1 >= 'a' && b1 <= 'z')) 573 a1 |= 32; 574 else if ((a1 >= 'a' && a1 <= 'z') && (b1 >= 'A' && b1 <= 'Z')) 575 b1 |= 32; 576 if ((a1 - b1) != 0) return (a1 - b1); 577 578 } 579 } 580 581 /* all chars are the same, and the length match too 582 * 583 * they are the same */ 584 if (a_len == b_len) return 0; 585 586 /* if a is shorter then b, then b is larger */ 587 return (a_len - b_len); 588 } 589 590 591 /** 592 * check if the rightmost bytes of the string are equal. 593 * 594 * 595 */ 596 597 int buffer_is_equal_right_len(buffer *b1, buffer *b2, size_t len) { 598 /* no, len -> equal */ 599 if (len == 0) return 1; 600 601 /* len > 0, but empty buffers -> not equal */ 602 if (b1->used == 0 || b2->used == 0) return 0; 603 604 /* buffers too small -> not equal */ 605 if (b1->used - 1 < len || b1->used - 1 < len) return 0; 606 607 if (0 == strncmp(b1->ptr + b1->used - 1 - len, 608 b2->ptr + b2->used - 1 - len, len)) { 609 return 1; 610 } 611 612 return 0; 613 } 614 615 int buffer_copy_string_hex(buffer *b, const char *in, size_t in_len) { 616 size_t i; 617 618 /* BO protection */ 619 if (in_len * 2 < in_len) return -1; 620 621 buffer_prepare_copy(b, in_len * 2 + 1); 622 623 for (i = 0; i < in_len; i++) { 624 b->ptr[b->used++] = hex_chars[(in[i] >> 4) & 0x0F]; 625 b->ptr[b->used++] = hex_chars[in[i] & 0x0F]; 626 } 627 b->ptr[b->used++] = '\0'; 628 629 return 0; 630 } 631 632 /* everything except: ! ( ) * - . 0-9 A-Z _ a-z */ 633 const char encoded_chars_rel_uri_part[] = { 634 /* 635 0 1 2 3 4 5 6 7 8 9 A B C D E F 636 */ 637 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 00 - 0F control chars */ 638 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 10 - 1F */ 639 1, 0, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 0, 0, 1, /* 20 - 2F space " # $ % & ' + , / */ 640 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, /* 30 - 3F : ; < = > ? */ 641 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 40 - 4F @ */ 642 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 0, /* 50 - 5F [ \ ] ^ */ 643 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 60 - 6F ` */ 644 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, /* 70 - 7F { | } ~ DEL */ 645 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 80 - 8F */ 646 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 90 - 9F */ 647 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* A0 - AF */ 648 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* B0 - BF */ 649 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* C0 - CF */ 650 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* D0 - DF */ 651 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* E0 - EF */ 652 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* F0 - FF */ 653 }; 654 655 /* everything except: ! ( ) * - . / 0-9 A-Z _ a-z */ 656 const char encoded_chars_rel_uri[] = { 657 /* 658 0 1 2 3 4 5 6 7 8 9 A B C D E F 659 */ 660 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 00 - 0F control chars */ 661 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 10 - 1F */ 662 1, 0, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 0, 0, 0, /* 20 - 2F space " # $ % & ' + , */ 663 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, /* 30 - 3F : ; < = > ? */ 664 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 40 - 4F @ */ 665 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 0, /* 50 - 5F [ \ ] ^ */ 666 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 60 - 6F ` */ 667 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, /* 70 - 7F { | } ~ DEL */ 668 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 80 - 8F */ 669 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 90 - 9F */ 670 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* A0 - AF */ 671 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* B0 - BF */ 672 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* C0 - CF */ 673 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* D0 - DF */ 674 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* E0 - EF */ 675 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* F0 - FF */ 676 }; 677 678 const char encoded_chars_html[] = { 679 /* 680 0 1 2 3 4 5 6 7 8 9 A B C D E F 681 */ 682 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 00 - 0F control chars */ 683 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 10 - 1F */ 684 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 20 - 2F & */ 685 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, /* 30 - 3F < > */ 686 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 40 - 4F */ 687 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 50 - 5F */ 688 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 60 - 6F */ 689 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, /* 70 - 7F DEL */ 690 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 80 - 8F */ 691 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 90 - 9F */ 692 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* A0 - AF */ 693 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* B0 - BF */ 694 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* C0 - CF */ 695 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* D0 - DF */ 696 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* E0 - EF */ 697 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* F0 - FF */ 698 }; 699 700 const char encoded_chars_minimal_xml[] = { 701 /* 702 0 1 2 3 4 5 6 7 8 9 A B C D E F 703 */ 704 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 00 - 0F control chars */ 705 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 10 - 1F */ 706 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 20 - 2F & */ 707 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, /* 30 - 3F < > */ 708 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 40 - 4F */ 709 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 50 - 5F */ 710 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 60 - 6F */ 711 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, /* 70 - 7F DEL */ 712 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 80 - 8F */ 713 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 90 - 9F */ 714 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* A0 - AF */ 715 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* B0 - BF */ 716 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* C0 - CF */ 717 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* D0 - DF */ 718 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* E0 - EF */ 719 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* F0 - FF */ 720 }; 721 722 const char encoded_chars_hex[] = { 723 /* 724 0 1 2 3 4 5 6 7 8 9 A B C D E F 725 */ 726 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 00 - 0F control chars */ 727 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 10 - 1F */ 728 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 20 - 2F */ 729 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 30 - 3F */ 730 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 40 - 4F */ 731 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 50 - 5F */ 732 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 60 - 6F */ 733 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 70 - 7F */ 734 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 80 - 8F */ 735 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 90 - 9F */ 736 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* A0 - AF */ 737 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* B0 - BF */ 738 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* C0 - CF */ 739 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* D0 - DF */ 740 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* E0 - EF */ 741 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* F0 - FF */ 742 }; 743 744 const char encoded_chars_http_header[] = { 745 /* 746 0 1 2 3 4 5 6 7 8 9 A B C D E F 747 */ 748 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, /* 00 - 0F */ 749 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 10 - 1F */ 750 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 20 - 2F */ 751 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 30 - 3F */ 752 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 40 - 4F */ 753 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 50 - 5F */ 754 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 60 - 6F */ 755 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 70 - 7F */ 756 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 80 - 8F */ 757 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 90 - 9F */ 758 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* A0 - AF */ 759 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* B0 - BF */ 760 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* C0 - CF */ 761 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* D0 - DF */ 762 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* E0 - EF */ 763 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* F0 - FF */ 764 }; 765 766 767 768 int buffer_append_string_encoded(buffer *b, const char *s, size_t s_len, buffer_encoding_t encoding) { 769 unsigned char *ds, *d; 770 size_t d_len, ndx; 771 const char *map = NULL; 772 773 if (!s || !b) return -1; 774 775 if (b->ptr[b->used - 1] != '\0') { 776 SEGFAULT(); 777 } 778 779 if (s_len == 0) return 0; 780 781 switch(encoding) { 782 case ENCODING_REL_URI: 783 map = encoded_chars_rel_uri; 784 break; 785 case ENCODING_REL_URI_PART: 786 map = encoded_chars_rel_uri_part; 787 break; 788 case ENCODING_HTML: 789 map = encoded_chars_html; 790 break; 791 case ENCODING_MINIMAL_XML: 792 map = encoded_chars_minimal_xml; 793 break; 794 case ENCODING_HEX: 795 map = encoded_chars_hex; 796 break; 797 case ENCODING_HTTP_HEADER: 798 map = encoded_chars_http_header; 799 break; 800 case ENCODING_UNSET: 801 break; 802 } 803 804 assert(map != NULL); 805 806 /* count to-be-encoded-characters */ 807 for (ds = (unsigned char *)s, d_len = 0, ndx = 0; ndx < s_len; ds++, ndx++) { 808 if (map[*ds]) { 809 switch(encoding) { 810 case ENCODING_REL_URI: 811 case ENCODING_REL_URI_PART: 812 d_len += 3; 813 break; 814 case ENCODING_HTML: 815 case ENCODING_MINIMAL_XML: 816 d_len += 6; 817 break; 818 case ENCODING_HTTP_HEADER: 819 case ENCODING_HEX: 820 d_len += 2; 821 break; 822 case ENCODING_UNSET: 823 break; 824 } 825 } else { 826 d_len ++; 827 } 828 } 829 830 buffer_prepare_append(b, d_len); 831 832 for (ds = (unsigned char *)s, d = (unsigned char *)b->ptr + b->used - 1, d_len = 0, ndx = 0; ndx < s_len; ds++, ndx++) { 833 if (map[*ds]) { 834 switch(encoding) { 835 case ENCODING_REL_URI: 836 case ENCODING_REL_URI_PART: 837 d[d_len++] = '%'; 838 d[d_len++] = hex_chars[((*ds) >> 4) & 0x0F]; 839 d[d_len++] = hex_chars[(*ds) & 0x0F]; 840 break; 841 case ENCODING_HTML: 842 case ENCODING_MINIMAL_XML: 843 d[d_len++] = '&'; 844 d[d_len++] = '#'; 845 d[d_len++] = 'x'; 846 d[d_len++] = hex_chars[((*ds) >> 4) & 0x0F]; 847 d[d_len++] = hex_chars[(*ds) & 0x0F]; 848 d[d_len++] = ';'; 849 break; 850 case ENCODING_HEX: 851 d[d_len++] = hex_chars[((*ds) >> 4) & 0x0F]; 852 d[d_len++] = hex_chars[(*ds) & 0x0F]; 853 break; 854 case ENCODING_HTTP_HEADER: 855 d[d_len++] = *ds; 856 d[d_len++] = '\t'; 857 break; 858 case ENCODING_UNSET: 859 break; 860 } 861 } else { 862 d[d_len++] = *ds; 863 } 864 } 865 866 /* terminate buffer and calculate new length */ 867 b->ptr[b->used + d_len - 1] = '\0'; 868 869 b->used += d_len; 870 871 return 0; 872 } 873 874 875 /* decodes url-special-chars inplace. 876 * replaces non-printable characters with '_' 877 */ 878 879 static int buffer_urldecode_internal(buffer *url, int is_query) { 880 unsigned char high, low; 881 const char *src; 882 char *dst; 883 884 if (!url || !url->ptr) return -1; 885 886 src = (const char*) url->ptr; 887 dst = (char*) url->ptr; 888 889 while ((*src) != '\0') { 890 if (is_query && *src == '+') { 891 *dst = ' '; 892 } else if (*src == '%') { 893 *dst = '%'; 894 895 high = hex2int(*(src + 1)); 896 if (high != 0xFF) { 897 low = hex2int(*(src + 2)); 898 if (low != 0xFF) { 899 high = (high << 4) | low; 900 901 /* map control-characters out */ 902 if (high < 32 || high == 127) high = '_'; 903 904 *dst = high; 905 src += 2; 906 } 907 } 908 } else { 909 *dst = *src; 910 } 911 912 dst++; 913 src++; 914 } 915 916 *dst = '\0'; 917 url->used = (dst - url->ptr) + 1; 918 919 return 0; 920 } 921 922 int buffer_urldecode_path(buffer *url) { 923 return buffer_urldecode_internal(url, 0); 924 } 925 926 int buffer_urldecode_query(buffer *url) { 927 return buffer_urldecode_internal(url, 1); 928 } 929 930 /* Remove "/../", "//", "/./" parts from path. 931 * 932 * /blah/.. gets / 933 * /blah/../foo gets /foo 934 * /abc/./xyz gets /abc/xyz 935 * /abc//xyz gets /abc/xyz 936 * 937 * NOTE: src and dest can point to the same buffer, in which case, 938 * the operation is performed in-place. 939 */ 940 941 int buffer_path_simplify(buffer *dest, buffer *src) 942 { 943 int toklen; 944 char c, pre1; 945 char *start, *slash, *walk, *out; 946 unsigned short pre; 947 948 if (src == NULL || src->ptr == NULL || dest == NULL) 949 return -1; 950 951 if (src == dest) 952 buffer_prepare_append(dest, 1); 953 else 954 buffer_prepare_copy(dest, src->used + 1); 955 956 walk = src->ptr; 957 start = dest->ptr; 958 out = dest->ptr; 959 slash = dest->ptr; 960 961 962 #if defined(__WIN32) || defined(__CYGWIN__) 963 /* cygwin is treating \ and / the same, so we have to that too 964 */ 965 966 for (walk = src->ptr; *walk; walk++) { 967 if (*walk == '\\') *walk = '/'; 968 } 969 walk = src->ptr; 970 #endif 971 972 while (*walk == ' ') { 973 walk++; 974 } 975 976 pre1 = *(walk++); 977 c = *(walk++); 978 pre = pre1; 979 if (pre1 != '/') { 980 pre = ('/' << 8) | pre1; 981 *(out++) = '/'; 982 } 983 *(out++) = pre1; 984 985 if (pre1 == '\0') { 986 dest->used = (out - start) + 1; 987 return 0; 988 } 989 990 while (1) { 991 if (c == '/' || c == '\0') { 992 toklen = out - slash; 993 if (toklen == 3 && pre == (('.' << 8) | '.')) { 994 out = slash; 995 if (out > start) { 996 out--; 997 while (out > start && *out != '/') { 998 out--; 999 } 1000 } 1001 1002 if (c == '\0') 1003 out++; 1004 } else if (toklen == 1 || pre == (('/' << 8) | '.')) { 1005 out = slash; 1006 if (c == '\0') 1007 out++; 1008 } 1009 1010 slash = out; 1011 } 1012 1013 if (c == '\0') 1014 break; 1015 1016 pre1 = c; 1017 pre = (pre << 8) | pre1; 1018 c = *walk; 1019 *out = pre1; 1020 1021 out++; 1022 walk++; 1023 } 1024 1025 *out = '\0'; 1026 dest->used = (out - start) + 1; 1027 1028 return 0; 1029 } 1030 1031 int light_isdigit(int c) { 1032 return (c >= '0' && c <= '9'); 1033 } 1034 1035 int light_isxdigit(int c) { 1036 if (light_isdigit(c)) return 1; 1037 1038 c |= 32; 1039 return (c >= 'a' && c <= 'f'); 1040 } 1041 1042 int light_isalpha(int c) { 1043 c |= 32; 1044 return (c >= 'a' && c <= 'z'); 1045 } 1046 1047 int light_isalnum(int c) { 1048 return light_isdigit(c) || light_isalpha(c); 1049 } 1050 1051 int buffer_to_lower(buffer *b) { 1052 char *c; 1053 1054 if (b->used == 0) return 0; 1055 1056 for (c = b->ptr; *c; c++) { 1057 if (*c >= 'A' && *c <= 'Z') { 1058 *c |= 32; 1059 } 1060 } 1061 1062 return 0; 1063 } 1064 1065 1066 int buffer_to_upper(buffer *b) { 1067 char *c; 1068 1069 if (b->used == 0) return 0; 1070 1071 for (c = b->ptr; *c; c++) { 1072 if (*c >= 'a' && *c <= 'z') { 1073 *c &= ~32; 1074 } 1075 } 1076 1077 return 0; 1078 } 1079