archive_read_support_format_mtree.c revision 302295
1/*- 2 * Copyright (c) 2003-2007 Tim Kientzle 3 * Copyright (c) 2008 Joerg Sonnenberger 4 * Copyright (c) 2011-2012 Michihiro NAKAJIMA 5 * All rights reserved. 6 * 7 * Redistribution and use in source and binary forms, with or without 8 * modification, are permitted provided that the following conditions 9 * are met: 10 * 1. Redistributions of source code must retain the above copyright 11 * notice, this list of conditions and the following disclaimer. 12 * 2. Redistributions in binary form must reproduce the above copyright 13 * notice, this list of conditions and the following disclaimer in the 14 * documentation and/or other materials provided with the distribution. 15 * 16 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR(S) ``AS IS'' AND ANY EXPRESS OR 17 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES 18 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. 19 * IN NO EVENT SHALL THE AUTHOR(S) BE LIABLE FOR ANY DIRECT, INDIRECT, 20 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT 21 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, 22 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY 23 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT 24 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF 25 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 26 */ 27 28#include "archive_platform.h" 29__FBSDID("$FreeBSD: stable/10/contrib/libarchive/libarchive/archive_read_support_format_mtree.c 302295 2016-06-30 12:44:15Z mm $"); 30 31#ifdef HAVE_SYS_STAT_H 32#include <sys/stat.h> 33#endif 34#ifdef HAVE_ERRNO_H 35#include <errno.h> 36#endif 37#ifdef HAVE_FCNTL_H 38#include <fcntl.h> 39#endif 40#include <stddef.h> 41/* #include <stdint.h> */ /* See archive_platform.h */ 42#ifdef HAVE_STDLIB_H 43#include <stdlib.h> 44#endif 45#ifdef HAVE_STRING_H 46#include <string.h> 47#endif 48 49#include "archive.h" 50#include "archive_entry.h" 51#include "archive_private.h" 52#include "archive_read_private.h" 53#include "archive_string.h" 54#include "archive_pack_dev.h" 55 56#ifndef O_BINARY 57#define O_BINARY 0 58#endif 59#ifndef O_CLOEXEC 60#define O_CLOEXEC 0 61#endif 62 63#define MTREE_HAS_DEVICE 0x0001 64#define MTREE_HAS_FFLAGS 0x0002 65#define MTREE_HAS_GID 0x0004 66#define MTREE_HAS_GNAME 0x0008 67#define MTREE_HAS_MTIME 0x0010 68#define MTREE_HAS_NLINK 0x0020 69#define MTREE_HAS_PERM 0x0040 70#define MTREE_HAS_SIZE 0x0080 71#define MTREE_HAS_TYPE 0x0100 72#define MTREE_HAS_UID 0x0200 73#define MTREE_HAS_UNAME 0x0400 74 75#define MTREE_HAS_OPTIONAL 0x0800 76#define MTREE_HAS_NOCHANGE 0x1000 /* FreeBSD specific */ 77 78struct mtree_option { 79 struct mtree_option *next; 80 char *value; 81}; 82 83struct mtree_entry { 84 struct mtree_entry *next; 85 struct mtree_option *options; 86 char *name; 87 char full; 88 char used; 89}; 90 91struct mtree { 92 struct archive_string line; 93 size_t buffsize; 94 char *buff; 95 int64_t offset; 96 int fd; 97 int archive_format; 98 const char *archive_format_name; 99 struct mtree_entry *entries; 100 struct mtree_entry *this_entry; 101 struct archive_string current_dir; 102 struct archive_string contents_name; 103 104 struct archive_entry_linkresolver *resolver; 105 106 int64_t cur_size; 107 char checkfs; 108}; 109 110static int bid_keycmp(const char *, const char *, ssize_t); 111static int cleanup(struct archive_read *); 112static int detect_form(struct archive_read *, int *); 113static int mtree_bid(struct archive_read *, int); 114static int parse_file(struct archive_read *, struct archive_entry *, 115 struct mtree *, struct mtree_entry *, int *); 116static void parse_escapes(char *, struct mtree_entry *); 117static int parse_line(struct archive_read *, struct archive_entry *, 118 struct mtree *, struct mtree_entry *, int *); 119static int parse_keyword(struct archive_read *, struct mtree *, 120 struct archive_entry *, struct mtree_option *, int *); 121static int read_data(struct archive_read *a, 122 const void **buff, size_t *size, int64_t *offset); 123static ssize_t readline(struct archive_read *, struct mtree *, char **, ssize_t); 124static int skip(struct archive_read *a); 125static int read_header(struct archive_read *, 126 struct archive_entry *); 127static int64_t mtree_atol10(char **); 128static int64_t mtree_atol8(char **); 129static int64_t mtree_atol(char **); 130 131/* 132 * There's no standard for TIME_T_MAX/TIME_T_MIN. So we compute them 133 * here. TODO: Move this to configure time, but be careful 134 * about cross-compile environments. 135 */ 136static int64_t 137get_time_t_max(void) 138{ 139#if defined(TIME_T_MAX) 140 return TIME_T_MAX; 141#else 142 /* ISO C allows time_t to be a floating-point type, 143 but POSIX requires an integer type. The following 144 should work on any system that follows the POSIX 145 conventions. */ 146 if (((time_t)0) < ((time_t)-1)) { 147 /* Time_t is unsigned */ 148 return (~(time_t)0); 149 } else { 150 /* Time_t is signed. */ 151 /* Assume it's the same as int64_t or int32_t */ 152 if (sizeof(time_t) == sizeof(int64_t)) { 153 return (time_t)INT64_MAX; 154 } else { 155 return (time_t)INT32_MAX; 156 } 157 } 158#endif 159} 160 161static int64_t 162get_time_t_min(void) 163{ 164#if defined(TIME_T_MIN) 165 return TIME_T_MIN; 166#else 167 if (((time_t)0) < ((time_t)-1)) { 168 /* Time_t is unsigned */ 169 return (time_t)0; 170 } else { 171 /* Time_t is signed. */ 172 if (sizeof(time_t) == sizeof(int64_t)) { 173 return (time_t)INT64_MIN; 174 } else { 175 return (time_t)INT32_MIN; 176 } 177 } 178#endif 179} 180 181static int 182archive_read_format_mtree_options(struct archive_read *a, 183 const char *key, const char *val) 184{ 185 struct mtree *mtree; 186 187 mtree = (struct mtree *)(a->format->data); 188 if (strcmp(key, "checkfs") == 0) { 189 /* Allows to read information missing from the mtree from the file system */ 190 if (val == NULL || val[0] == 0) { 191 mtree->checkfs = 0; 192 } else { 193 mtree->checkfs = 1; 194 } 195 return (ARCHIVE_OK); 196 } 197 198 /* Note: The "warn" return is just to inform the options 199 * supervisor that we didn't handle it. It will generate 200 * a suitable error if no one used this option. */ 201 return (ARCHIVE_WARN); 202} 203 204static void 205free_options(struct mtree_option *head) 206{ 207 struct mtree_option *next; 208 209 for (; head != NULL; head = next) { 210 next = head->next; 211 free(head->value); 212 free(head); 213 } 214} 215 216int 217archive_read_support_format_mtree(struct archive *_a) 218{ 219 struct archive_read *a = (struct archive_read *)_a; 220 struct mtree *mtree; 221 int r; 222 223 archive_check_magic(_a, ARCHIVE_READ_MAGIC, 224 ARCHIVE_STATE_NEW, "archive_read_support_format_mtree"); 225 226 mtree = (struct mtree *)malloc(sizeof(*mtree)); 227 if (mtree == NULL) { 228 archive_set_error(&a->archive, ENOMEM, 229 "Can't allocate mtree data"); 230 return (ARCHIVE_FATAL); 231 } 232 memset(mtree, 0, sizeof(*mtree)); 233 mtree->fd = -1; 234 235 r = __archive_read_register_format(a, mtree, "mtree", 236 mtree_bid, archive_read_format_mtree_options, read_header, read_data, skip, NULL, cleanup, NULL, NULL); 237 238 if (r != ARCHIVE_OK) 239 free(mtree); 240 return (ARCHIVE_OK); 241} 242 243static int 244cleanup(struct archive_read *a) 245{ 246 struct mtree *mtree; 247 struct mtree_entry *p, *q; 248 249 mtree = (struct mtree *)(a->format->data); 250 251 p = mtree->entries; 252 while (p != NULL) { 253 q = p->next; 254 free(p->name); 255 free_options(p->options); 256 free(p); 257 p = q; 258 } 259 archive_string_free(&mtree->line); 260 archive_string_free(&mtree->current_dir); 261 archive_string_free(&mtree->contents_name); 262 archive_entry_linkresolver_free(mtree->resolver); 263 264 free(mtree->buff); 265 free(mtree); 266 (a->format->data) = NULL; 267 return (ARCHIVE_OK); 268} 269 270static ssize_t 271get_line_size(const char *b, ssize_t avail, ssize_t *nlsize) 272{ 273 ssize_t len; 274 275 len = 0; 276 while (len < avail) { 277 switch (*b) { 278 case '\0':/* Non-ascii character or control character. */ 279 if (nlsize != NULL) 280 *nlsize = 0; 281 return (-1); 282 case '\r': 283 if (avail-len > 1 && b[1] == '\n') { 284 if (nlsize != NULL) 285 *nlsize = 2; 286 return (len+2); 287 } 288 /* FALL THROUGH */ 289 case '\n': 290 if (nlsize != NULL) 291 *nlsize = 1; 292 return (len+1); 293 default: 294 b++; 295 len++; 296 break; 297 } 298 } 299 if (nlsize != NULL) 300 *nlsize = 0; 301 return (avail); 302} 303 304static ssize_t 305next_line(struct archive_read *a, 306 const char **b, ssize_t *avail, ssize_t *ravail, ssize_t *nl) 307{ 308 ssize_t len; 309 int quit; 310 311 quit = 0; 312 if (*avail == 0) { 313 *nl = 0; 314 len = 0; 315 } else 316 len = get_line_size(*b, *avail, nl); 317 /* 318 * Read bytes more while it does not reach the end of line. 319 */ 320 while (*nl == 0 && len == *avail && !quit) { 321 ssize_t diff = *ravail - *avail; 322 size_t nbytes_req = (*ravail+1023) & ~1023U; 323 ssize_t tested; 324 325 /* Increase reading bytes if it is not enough to at least 326 * new two lines. */ 327 if (nbytes_req < (size_t)*ravail + 160) 328 nbytes_req <<= 1; 329 330 *b = __archive_read_ahead(a, nbytes_req, avail); 331 if (*b == NULL) { 332 if (*ravail >= *avail) 333 return (0); 334 /* Reading bytes reaches the end of file. */ 335 *b = __archive_read_ahead(a, *avail, avail); 336 quit = 1; 337 } 338 *ravail = *avail; 339 *b += diff; 340 *avail -= diff; 341 tested = len;/* Skip some bytes we already determinated. */ 342 len = get_line_size(*b, *avail, nl); 343 if (len >= 0) 344 len += tested; 345 } 346 return (len); 347} 348 349/* 350 * Compare characters with a mtree keyword. 351 * Returns the length of a mtree keyword if matched. 352 * Returns 0 if not matched. 353 */ 354static int 355bid_keycmp(const char *p, const char *key, ssize_t len) 356{ 357 int match_len = 0; 358 359 while (len > 0 && *p && *key) { 360 if (*p == *key) { 361 --len; 362 ++p; 363 ++key; 364 ++match_len; 365 continue; 366 } 367 return (0);/* Not match */ 368 } 369 if (*key != '\0') 370 return (0);/* Not match */ 371 372 /* A following character should be specified characters */ 373 if (p[0] == '=' || p[0] == ' ' || p[0] == '\t' || 374 p[0] == '\n' || p[0] == '\r' || 375 (p[0] == '\\' && (p[1] == '\n' || p[1] == '\r'))) 376 return (match_len); 377 return (0);/* Not match */ 378} 379 380/* 381 * Test whether the characters 'p' has is mtree keyword. 382 * Returns the length of a detected keyword. 383 * Returns 0 if any keywords were not found. 384 */ 385static int 386bid_keyword(const char *p, ssize_t len) 387{ 388 static const char *keys_c[] = { 389 "content", "contents", "cksum", NULL 390 }; 391 static const char *keys_df[] = { 392 "device", "flags", NULL 393 }; 394 static const char *keys_g[] = { 395 "gid", "gname", NULL 396 }; 397 static const char *keys_il[] = { 398 "ignore", "inode", "link", NULL 399 }; 400 static const char *keys_m[] = { 401 "md5", "md5digest", "mode", NULL 402 }; 403 static const char *keys_no[] = { 404 "nlink", "nochange", "optional", NULL 405 }; 406 static const char *keys_r[] = { 407 "resdevice", "rmd160", "rmd160digest", NULL 408 }; 409 static const char *keys_s[] = { 410 "sha1", "sha1digest", 411 "sha256", "sha256digest", 412 "sha384", "sha384digest", 413 "sha512", "sha512digest", 414 "size", NULL 415 }; 416 static const char *keys_t[] = { 417 "tags", "time", "type", NULL 418 }; 419 static const char *keys_u[] = { 420 "uid", "uname", NULL 421 }; 422 const char **keys; 423 int i; 424 425 switch (*p) { 426 case 'c': keys = keys_c; break; 427 case 'd': case 'f': keys = keys_df; break; 428 case 'g': keys = keys_g; break; 429 case 'i': case 'l': keys = keys_il; break; 430 case 'm': keys = keys_m; break; 431 case 'n': case 'o': keys = keys_no; break; 432 case 'r': keys = keys_r; break; 433 case 's': keys = keys_s; break; 434 case 't': keys = keys_t; break; 435 case 'u': keys = keys_u; break; 436 default: return (0);/* Unknown key */ 437 } 438 439 for (i = 0; keys[i] != NULL; i++) { 440 int l = bid_keycmp(p, keys[i], len); 441 if (l > 0) 442 return (l); 443 } 444 return (0);/* Unknown key */ 445} 446 447/* 448 * Test whether there is a set of mtree keywords. 449 * Returns the number of keyword. 450 * Returns -1 if we got incorrect sequence. 451 * This function expects a set of "<space characters>keyword=value". 452 * When "unset" is specified, expects a set of "<space characters>keyword". 453 */ 454static int 455bid_keyword_list(const char *p, ssize_t len, int unset, int last_is_path) 456{ 457 int l; 458 int keycnt = 0; 459 460 while (len > 0 && *p) { 461 int blank = 0; 462 463 /* Test whether there are blank characters in the line. */ 464 while (len >0 && (*p == ' ' || *p == '\t')) { 465 ++p; 466 --len; 467 blank = 1; 468 } 469 if (*p == '\n' || *p == '\r') 470 break; 471 if (p[0] == '\\' && (p[1] == '\n' || p[1] == '\r')) 472 break; 473 if (!blank && !last_is_path) /* No blank character. */ 474 return (-1); 475 if (last_is_path && len == 0) 476 return (keycnt); 477 478 if (unset) { 479 l = bid_keycmp(p, "all", len); 480 if (l > 0) 481 return (1); 482 } 483 /* Test whether there is a correct key in the line. */ 484 l = bid_keyword(p, len); 485 if (l == 0) 486 return (-1);/* Unknown keyword was found. */ 487 p += l; 488 len -= l; 489 keycnt++; 490 491 /* Skip value */ 492 if (*p == '=') { 493 int value = 0; 494 ++p; 495 --len; 496 while (len > 0 && *p != ' ' && *p != '\t') { 497 ++p; 498 --len; 499 value = 1; 500 } 501 /* A keyword should have a its value unless 502 * "/unset" operation. */ 503 if (!unset && value == 0) 504 return (-1); 505 } 506 } 507 return (keycnt); 508} 509 510static int 511bid_entry(const char *p, ssize_t len, ssize_t nl, int *last_is_path) 512{ 513 int f = 0; 514 static const unsigned char safe_char[256] = { 515 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 00 - 0F */ 516 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 10 - 1F */ 517 /* !"$%&'()*+,-./ EXCLUSION:( )(#) */ 518 0, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 20 - 2F */ 519 /* 0123456789:;<>? EXCLUSION:(=) */ 520 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, /* 30 - 3F */ 521 /* @ABCDEFGHIJKLMNO */ 522 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 40 - 4F */ 523 /* PQRSTUVWXYZ[\]^_ */ 524 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 50 - 5F */ 525 /* `abcdefghijklmno */ 526 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 60 - 6F */ 527 /* pqrstuvwxyz{|}~ */ 528 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, /* 70 - 7F */ 529 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 80 - 8F */ 530 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 90 - 9F */ 531 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* A0 - AF */ 532 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* B0 - BF */ 533 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* C0 - CF */ 534 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* D0 - DF */ 535 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* E0 - EF */ 536 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* F0 - FF */ 537 }; 538 ssize_t ll; 539 const char *pp = p; 540 const char * const pp_end = pp + len; 541 542 *last_is_path = 0; 543 /* 544 * Skip the path-name which is quoted. 545 */ 546 for (;pp < pp_end; ++pp) { 547 if (!safe_char[*(const unsigned char *)pp]) { 548 if (*pp != ' ' && *pp != '\t' && *pp != '\r' 549 && *pp != '\n') 550 f = 0; 551 break; 552 } 553 f = 1; 554 } 555 ll = pp_end - pp; 556 557 /* If a path-name was not found at the first, try to check 558 * a mtree format(a.k.a form D) ``NetBSD's mtree -D'' creates, 559 * which places the path-name at the last. */ 560 if (f == 0) { 561 const char *pb = p + len - nl; 562 int name_len = 0; 563 int slash; 564 565 /* The form D accepts only a single line for an entry. */ 566 if (pb-2 >= p && 567 pb[-1] == '\\' && (pb[-2] == ' ' || pb[-2] == '\t')) 568 return (-1); 569 if (pb-1 >= p && pb[-1] == '\\') 570 return (-1); 571 572 slash = 0; 573 while (p <= --pb && *pb != ' ' && *pb != '\t') { 574 if (!safe_char[*(const unsigned char *)pb]) 575 return (-1); 576 name_len++; 577 /* The pathname should have a slash in this 578 * format. */ 579 if (*pb == '/') 580 slash = 1; 581 } 582 if (name_len == 0 || slash == 0) 583 return (-1); 584 /* If '/' is placed at the first in this field, this is not 585 * a valid filename. */ 586 if (pb[1] == '/') 587 return (-1); 588 ll = len - nl - name_len; 589 pp = p; 590 *last_is_path = 1; 591 } 592 593 return (bid_keyword_list(pp, ll, 0, *last_is_path)); 594} 595 596#define MAX_BID_ENTRY 3 597 598static int 599mtree_bid(struct archive_read *a, int best_bid) 600{ 601 const char *signature = "#mtree"; 602 const char *p; 603 604 (void)best_bid; /* UNUSED */ 605 606 /* Now let's look at the actual header and see if it matches. */ 607 p = __archive_read_ahead(a, strlen(signature), NULL); 608 if (p == NULL) 609 return (-1); 610 611 if (memcmp(p, signature, strlen(signature)) == 0) 612 return (8 * (int)strlen(signature)); 613 614 /* 615 * There is not a mtree signature. Let's try to detect mtree format. 616 */ 617 return (detect_form(a, NULL)); 618} 619 620static int 621detect_form(struct archive_read *a, int *is_form_d) 622{ 623 const char *p; 624 ssize_t avail, ravail; 625 ssize_t detected_bytes = 0, len, nl; 626 int entry_cnt = 0, multiline = 0; 627 int form_D = 0;/* The archive is generated by `NetBSD mtree -D' 628 * (In this source we call it `form D') . */ 629 630 if (is_form_d != NULL) 631 *is_form_d = 0; 632 p = __archive_read_ahead(a, 1, &avail); 633 if (p == NULL) 634 return (-1); 635 ravail = avail; 636 for (;;) { 637 len = next_line(a, &p, &avail, &ravail, &nl); 638 /* The terminal character of the line should be 639 * a new line character, '\r\n' or '\n'. */ 640 if (len <= 0 || nl == 0) 641 break; 642 if (!multiline) { 643 /* Leading whitespace is never significant, 644 * ignore it. */ 645 while (len > 0 && (*p == ' ' || *p == '\t')) { 646 ++p; 647 --avail; 648 --len; 649 } 650 /* Skip comment or empty line. */ 651 if (p[0] == '#' || p[0] == '\n' || p[0] == '\r') { 652 p += len; 653 avail -= len; 654 continue; 655 } 656 } else { 657 /* A continuance line; the terminal 658 * character of previous line was '\' character. */ 659 if (bid_keyword_list(p, len, 0, 0) <= 0) 660 break; 661 if (multiline == 1) 662 detected_bytes += len; 663 if (p[len-nl-1] != '\\') { 664 if (multiline == 1 && 665 ++entry_cnt >= MAX_BID_ENTRY) 666 break; 667 multiline = 0; 668 } 669 p += len; 670 avail -= len; 671 continue; 672 } 673 if (p[0] != '/') { 674 int last_is_path, keywords; 675 676 keywords = bid_entry(p, len, nl, &last_is_path); 677 if (keywords >= 0) { 678 detected_bytes += len; 679 if (form_D == 0) { 680 if (last_is_path) 681 form_D = 1; 682 else if (keywords > 0) 683 /* This line is not `form D'. */ 684 form_D = -1; 685 } else if (form_D == 1) { 686 if (!last_is_path && keywords > 0) 687 /* This this is not `form D' 688 * and We cannot accept mixed 689 * format. */ 690 break; 691 } 692 if (!last_is_path && p[len-nl-1] == '\\') 693 /* This line continues. */ 694 multiline = 1; 695 else { 696 /* We've got plenty of correct lines 697 * to assume that this file is a mtree 698 * format. */ 699 if (++entry_cnt >= MAX_BID_ENTRY) 700 break; 701 } 702 } else 703 break; 704 } else if (strncmp(p, "/set", 4) == 0) { 705 if (bid_keyword_list(p+4, len-4, 0, 0) <= 0) 706 break; 707 /* This line continues. */ 708 if (p[len-nl-1] == '\\') 709 multiline = 2; 710 } else if (strncmp(p, "/unset", 6) == 0) { 711 if (bid_keyword_list(p+6, len-6, 1, 0) <= 0) 712 break; 713 /* This line continues. */ 714 if (p[len-nl-1] == '\\') 715 multiline = 2; 716 } else 717 break; 718 719 /* Test next line. */ 720 p += len; 721 avail -= len; 722 } 723 if (entry_cnt >= MAX_BID_ENTRY || (entry_cnt > 0 && len == 0)) { 724 if (is_form_d != NULL) { 725 if (form_D == 1) 726 *is_form_d = 1; 727 } 728 return (32); 729 } 730 731 return (0); 732} 733 734/* 735 * The extended mtree format permits multiple lines specifying 736 * attributes for each file. For those entries, only the last line 737 * is actually used. Practically speaking, that means we have 738 * to read the entire mtree file into memory up front. 739 * 740 * The parsing is done in two steps. First, it is decided if a line 741 * changes the global defaults and if it is, processed accordingly. 742 * Otherwise, the options of the line are merged with the current 743 * global options. 744 */ 745static int 746add_option(struct archive_read *a, struct mtree_option **global, 747 const char *value, size_t len) 748{ 749 struct mtree_option *opt; 750 751 if ((opt = malloc(sizeof(*opt))) == NULL) { 752 archive_set_error(&a->archive, errno, "Can't allocate memory"); 753 return (ARCHIVE_FATAL); 754 } 755 if ((opt->value = malloc(len + 1)) == NULL) { 756 free(opt); 757 archive_set_error(&a->archive, errno, "Can't allocate memory"); 758 return (ARCHIVE_FATAL); 759 } 760 memcpy(opt->value, value, len); 761 opt->value[len] = '\0'; 762 opt->next = *global; 763 *global = opt; 764 return (ARCHIVE_OK); 765} 766 767static void 768remove_option(struct mtree_option **global, const char *value, size_t len) 769{ 770 struct mtree_option *iter, *last; 771 772 last = NULL; 773 for (iter = *global; iter != NULL; last = iter, iter = iter->next) { 774 if (strncmp(iter->value, value, len) == 0 && 775 (iter->value[len] == '\0' || 776 iter->value[len] == '=')) 777 break; 778 } 779 if (iter == NULL) 780 return; 781 if (last == NULL) 782 *global = iter->next; 783 else 784 last->next = iter->next; 785 786 free(iter->value); 787 free(iter); 788} 789 790static int 791process_global_set(struct archive_read *a, 792 struct mtree_option **global, const char *line) 793{ 794 const char *next, *eq; 795 size_t len; 796 int r; 797 798 line += 4; 799 for (;;) { 800 next = line + strspn(line, " \t\r\n"); 801 if (*next == '\0') 802 return (ARCHIVE_OK); 803 line = next; 804 next = line + strcspn(line, " \t\r\n"); 805 eq = strchr(line, '='); 806 if (eq > next) 807 len = next - line; 808 else 809 len = eq - line; 810 811 remove_option(global, line, len); 812 r = add_option(a, global, line, next - line); 813 if (r != ARCHIVE_OK) 814 return (r); 815 line = next; 816 } 817} 818 819static int 820process_global_unset(struct archive_read *a, 821 struct mtree_option **global, const char *line) 822{ 823 const char *next; 824 size_t len; 825 826 line += 6; 827 if (strchr(line, '=') != NULL) { 828 archive_set_error(&a->archive, ARCHIVE_ERRNO_MISC, 829 "/unset shall not contain `='"); 830 return ARCHIVE_FATAL; 831 } 832 833 for (;;) { 834 next = line + strspn(line, " \t\r\n"); 835 if (*next == '\0') 836 return (ARCHIVE_OK); 837 line = next; 838 len = strcspn(line, " \t\r\n"); 839 840 if (len == 3 && strncmp(line, "all", 3) == 0) { 841 free_options(*global); 842 *global = NULL; 843 } else { 844 remove_option(global, line, len); 845 } 846 847 line += len; 848 } 849} 850 851static int 852process_add_entry(struct archive_read *a, struct mtree *mtree, 853 struct mtree_option **global, const char *line, ssize_t line_len, 854 struct mtree_entry **last_entry, int is_form_d) 855{ 856 struct mtree_entry *entry; 857 struct mtree_option *iter; 858 const char *next, *eq, *name, *end; 859 size_t name_len, len; 860 int r, i; 861 862 if ((entry = malloc(sizeof(*entry))) == NULL) { 863 archive_set_error(&a->archive, errno, "Can't allocate memory"); 864 return (ARCHIVE_FATAL); 865 } 866 entry->next = NULL; 867 entry->options = NULL; 868 entry->name = NULL; 869 entry->used = 0; 870 entry->full = 0; 871 872 /* Add this entry to list. */ 873 if (*last_entry == NULL) 874 mtree->entries = entry; 875 else 876 (*last_entry)->next = entry; 877 *last_entry = entry; 878 879 if (is_form_d) { 880 /* Filename is last item on line. */ 881 /* Adjust line_len to trim trailing whitespace */ 882 while (line_len > 0) { 883 char last_character = line[line_len - 1]; 884 if (last_character == '\r' 885 || last_character == '\n' 886 || last_character == '\t' 887 || last_character == ' ') { 888 line_len--; 889 } else { 890 break; 891 } 892 } 893 /* Name starts after the last whitespace separator */ 894 name = line; 895 for (i = 0; i < line_len; i++) { 896 if (line[i] == '\r' 897 || line[i] == '\n' 898 || line[i] == '\t' 899 || line[i] == ' ') { 900 name = line + i + 1; 901 } 902 } 903 name_len = line + line_len - name; 904 end = name; 905 } else { 906 /* Filename is first item on line */ 907 name_len = strcspn(line, " \t\r\n"); 908 name = line; 909 line += name_len; 910 end = line + line_len; 911 } 912 /* name/name_len is the name within the line. */ 913 /* line..end brackets the entire line except the name */ 914 915 if ((entry->name = malloc(name_len + 1)) == NULL) { 916 archive_set_error(&a->archive, errno, "Can't allocate memory"); 917 return (ARCHIVE_FATAL); 918 } 919 920 memcpy(entry->name, name, name_len); 921 entry->name[name_len] = '\0'; 922 parse_escapes(entry->name, entry); 923 924 for (iter = *global; iter != NULL; iter = iter->next) { 925 r = add_option(a, &entry->options, iter->value, 926 strlen(iter->value)); 927 if (r != ARCHIVE_OK) 928 return (r); 929 } 930 931 for (;;) { 932 next = line + strspn(line, " \t\r\n"); 933 if (*next == '\0') 934 return (ARCHIVE_OK); 935 if (next >= end) 936 return (ARCHIVE_OK); 937 line = next; 938 next = line + strcspn(line, " \t\r\n"); 939 eq = strchr(line, '='); 940 if (eq == NULL || eq > next) 941 len = next - line; 942 else 943 len = eq - line; 944 945 remove_option(&entry->options, line, len); 946 r = add_option(a, &entry->options, line, next - line); 947 if (r != ARCHIVE_OK) 948 return (r); 949 line = next; 950 } 951} 952 953static int 954read_mtree(struct archive_read *a, struct mtree *mtree) 955{ 956 ssize_t len; 957 uintmax_t counter; 958 char *p; 959 struct mtree_option *global; 960 struct mtree_entry *last_entry; 961 int r, is_form_d; 962 963 mtree->archive_format = ARCHIVE_FORMAT_MTREE; 964 mtree->archive_format_name = "mtree"; 965 966 global = NULL; 967 last_entry = NULL; 968 969 (void)detect_form(a, &is_form_d); 970 971 for (counter = 1; ; ++counter) { 972 len = readline(a, mtree, &p, 65536); 973 if (len == 0) { 974 mtree->this_entry = mtree->entries; 975 free_options(global); 976 return (ARCHIVE_OK); 977 } 978 if (len < 0) { 979 free_options(global); 980 return ((int)len); 981 } 982 /* Leading whitespace is never significant, ignore it. */ 983 while (*p == ' ' || *p == '\t') { 984 ++p; 985 --len; 986 } 987 /* Skip content lines and blank lines. */ 988 if (*p == '#') 989 continue; 990 if (*p == '\r' || *p == '\n' || *p == '\0') 991 continue; 992 if (*p != '/') { 993 r = process_add_entry(a, mtree, &global, p, len, 994 &last_entry, is_form_d); 995 } else if (strncmp(p, "/set", 4) == 0) { 996 if (p[4] != ' ' && p[4] != '\t') 997 break; 998 r = process_global_set(a, &global, p); 999 } else if (strncmp(p, "/unset", 6) == 0) { 1000 if (p[6] != ' ' && p[6] != '\t') 1001 break; 1002 r = process_global_unset(a, &global, p); 1003 } else 1004 break; 1005 1006 if (r != ARCHIVE_OK) { 1007 free_options(global); 1008 return r; 1009 } 1010 } 1011 1012 archive_set_error(&a->archive, ARCHIVE_ERRNO_FILE_FORMAT, 1013 "Can't parse line %ju", counter); 1014 free_options(global); 1015 return (ARCHIVE_FATAL); 1016} 1017 1018/* 1019 * Read in the entire mtree file into memory on the first request. 1020 * Then use the next unused file to satisfy each header request. 1021 */ 1022static int 1023read_header(struct archive_read *a, struct archive_entry *entry) 1024{ 1025 struct mtree *mtree; 1026 char *p; 1027 int r, use_next; 1028 1029 mtree = (struct mtree *)(a->format->data); 1030 1031 if (mtree->fd >= 0) { 1032 close(mtree->fd); 1033 mtree->fd = -1; 1034 } 1035 1036 if (mtree->entries == NULL) { 1037 mtree->resolver = archive_entry_linkresolver_new(); 1038 if (mtree->resolver == NULL) 1039 return ARCHIVE_FATAL; 1040 archive_entry_linkresolver_set_strategy(mtree->resolver, 1041 ARCHIVE_FORMAT_MTREE); 1042 r = read_mtree(a, mtree); 1043 if (r != ARCHIVE_OK) 1044 return (r); 1045 } 1046 1047 a->archive.archive_format = mtree->archive_format; 1048 a->archive.archive_format_name = mtree->archive_format_name; 1049 1050 for (;;) { 1051 if (mtree->this_entry == NULL) 1052 return (ARCHIVE_EOF); 1053 if (strcmp(mtree->this_entry->name, "..") == 0) { 1054 mtree->this_entry->used = 1; 1055 if (archive_strlen(&mtree->current_dir) > 0) { 1056 /* Roll back current path. */ 1057 p = mtree->current_dir.s 1058 + mtree->current_dir.length - 1; 1059 while (p >= mtree->current_dir.s && *p != '/') 1060 --p; 1061 if (p >= mtree->current_dir.s) 1062 --p; 1063 mtree->current_dir.length 1064 = p - mtree->current_dir.s + 1; 1065 } 1066 } 1067 if (!mtree->this_entry->used) { 1068 use_next = 0; 1069 r = parse_file(a, entry, mtree, mtree->this_entry, 1070 &use_next); 1071 if (use_next == 0) 1072 return (r); 1073 } 1074 mtree->this_entry = mtree->this_entry->next; 1075 } 1076} 1077 1078/* 1079 * A single file can have multiple lines contribute specifications. 1080 * Parse as many lines as necessary, then pull additional information 1081 * from a backing file on disk as necessary. 1082 */ 1083static int 1084parse_file(struct archive_read *a, struct archive_entry *entry, 1085 struct mtree *mtree, struct mtree_entry *mentry, int *use_next) 1086{ 1087 const char *path; 1088 struct stat st_storage, *st; 1089 struct mtree_entry *mp; 1090 struct archive_entry *sparse_entry; 1091 int r = ARCHIVE_OK, r1, parsed_kws; 1092 1093 mentry->used = 1; 1094 1095 /* Initialize reasonable defaults. */ 1096 archive_entry_set_filetype(entry, AE_IFREG); 1097 archive_entry_set_size(entry, 0); 1098 archive_string_empty(&mtree->contents_name); 1099 1100 /* Parse options from this line. */ 1101 parsed_kws = 0; 1102 r = parse_line(a, entry, mtree, mentry, &parsed_kws); 1103 1104 if (mentry->full) { 1105 archive_entry_copy_pathname(entry, mentry->name); 1106 /* 1107 * "Full" entries are allowed to have multiple lines 1108 * and those lines aren't required to be adjacent. We 1109 * don't support multiple lines for "relative" entries 1110 * nor do we make any attempt to merge data from 1111 * separate "relative" and "full" entries. (Merging 1112 * "relative" and "full" entries would require dealing 1113 * with pathname canonicalization, which is a very 1114 * tricky subject.) 1115 */ 1116 for (mp = mentry->next; mp != NULL; mp = mp->next) { 1117 if (mp->full && !mp->used 1118 && strcmp(mentry->name, mp->name) == 0) { 1119 /* Later lines override earlier ones. */ 1120 mp->used = 1; 1121 r1 = parse_line(a, entry, mtree, mp, 1122 &parsed_kws); 1123 if (r1 < r) 1124 r = r1; 1125 } 1126 } 1127 } else { 1128 /* 1129 * Relative entries require us to construct 1130 * the full path and possibly update the 1131 * current directory. 1132 */ 1133 size_t n = archive_strlen(&mtree->current_dir); 1134 if (n > 0) 1135 archive_strcat(&mtree->current_dir, "/"); 1136 archive_strcat(&mtree->current_dir, mentry->name); 1137 archive_entry_copy_pathname(entry, mtree->current_dir.s); 1138 if (archive_entry_filetype(entry) != AE_IFDIR) 1139 mtree->current_dir.length = n; 1140 } 1141 1142 if (mtree->checkfs) { 1143 /* 1144 * Try to open and stat the file to get the real size 1145 * and other file info. It would be nice to avoid 1146 * this here so that getting a listing of an mtree 1147 * wouldn't require opening every referenced contents 1148 * file. But then we wouldn't know the actual 1149 * contents size, so I don't see a really viable way 1150 * around this. (Also, we may want to someday pull 1151 * other unspecified info from the contents file on 1152 * disk.) 1153 */ 1154 mtree->fd = -1; 1155 if (archive_strlen(&mtree->contents_name) > 0) 1156 path = mtree->contents_name.s; 1157 else 1158 path = archive_entry_pathname(entry); 1159 1160 if (archive_entry_filetype(entry) == AE_IFREG || 1161 archive_entry_filetype(entry) == AE_IFDIR) { 1162 mtree->fd = open(path, O_RDONLY | O_BINARY | O_CLOEXEC); 1163 __archive_ensure_cloexec_flag(mtree->fd); 1164 if (mtree->fd == -1 && 1165 (errno != ENOENT || 1166 archive_strlen(&mtree->contents_name) > 0)) { 1167 archive_set_error(&a->archive, errno, 1168 "Can't open %s", path); 1169 r = ARCHIVE_WARN; 1170 } 1171 } 1172 1173 st = &st_storage; 1174 if (mtree->fd >= 0) { 1175 if (fstat(mtree->fd, st) == -1) { 1176 archive_set_error(&a->archive, errno, 1177 "Could not fstat %s", path); 1178 r = ARCHIVE_WARN; 1179 /* If we can't stat it, don't keep it open. */ 1180 close(mtree->fd); 1181 mtree->fd = -1; 1182 st = NULL; 1183 } 1184 } else if (lstat(path, st) == -1) { 1185 st = NULL; 1186 } 1187 1188 /* 1189 * Check for a mismatch between the type in the specification 1190 * and the type of the contents object on disk. 1191 */ 1192 if (st != NULL) { 1193 if (((st->st_mode & S_IFMT) == S_IFREG && 1194 archive_entry_filetype(entry) == AE_IFREG) 1195#ifdef S_IFLNK 1196 ||((st->st_mode & S_IFMT) == S_IFLNK && 1197 archive_entry_filetype(entry) == AE_IFLNK) 1198#endif 1199#ifdef S_IFSOCK 1200 ||((st->st_mode & S_IFSOCK) == S_IFSOCK && 1201 archive_entry_filetype(entry) == AE_IFSOCK) 1202#endif 1203#ifdef S_IFCHR 1204 ||((st->st_mode & S_IFMT) == S_IFCHR && 1205 archive_entry_filetype(entry) == AE_IFCHR) 1206#endif 1207#ifdef S_IFBLK 1208 ||((st->st_mode & S_IFMT) == S_IFBLK && 1209 archive_entry_filetype(entry) == AE_IFBLK) 1210#endif 1211 ||((st->st_mode & S_IFMT) == S_IFDIR && 1212 archive_entry_filetype(entry) == AE_IFDIR) 1213#ifdef S_IFIFO 1214 ||((st->st_mode & S_IFMT) == S_IFIFO && 1215 archive_entry_filetype(entry) == AE_IFIFO) 1216#endif 1217 ) { 1218 /* Types match. */ 1219 } else { 1220 /* Types don't match; bail out gracefully. */ 1221 if (mtree->fd >= 0) 1222 close(mtree->fd); 1223 mtree->fd = -1; 1224 if (parsed_kws & MTREE_HAS_OPTIONAL) { 1225 /* It's not an error for an optional 1226 * entry to not match disk. */ 1227 *use_next = 1; 1228 } else if (r == ARCHIVE_OK) { 1229 archive_set_error(&a->archive, 1230 ARCHIVE_ERRNO_MISC, 1231 "mtree specification has different" 1232 " type for %s", 1233 archive_entry_pathname(entry)); 1234 r = ARCHIVE_WARN; 1235 } 1236 return (r); 1237 } 1238 } 1239 1240 /* 1241 * If there is a contents file on disk, pick some of the 1242 * metadata from that file. For most of these, we only 1243 * set it from the contents if it wasn't already parsed 1244 * from the specification. 1245 */ 1246 if (st != NULL) { 1247 if (((parsed_kws & MTREE_HAS_DEVICE) == 0 || 1248 (parsed_kws & MTREE_HAS_NOCHANGE) != 0) && 1249 (archive_entry_filetype(entry) == AE_IFCHR || 1250 archive_entry_filetype(entry) == AE_IFBLK)) 1251 archive_entry_set_rdev(entry, st->st_rdev); 1252 if ((parsed_kws & (MTREE_HAS_GID | MTREE_HAS_GNAME)) 1253 == 0 || 1254 (parsed_kws & MTREE_HAS_NOCHANGE) != 0) 1255 archive_entry_set_gid(entry, st->st_gid); 1256 if ((parsed_kws & (MTREE_HAS_UID | MTREE_HAS_UNAME)) 1257 == 0 || 1258 (parsed_kws & MTREE_HAS_NOCHANGE) != 0) 1259 archive_entry_set_uid(entry, st->st_uid); 1260 if ((parsed_kws & MTREE_HAS_MTIME) == 0 || 1261 (parsed_kws & MTREE_HAS_NOCHANGE) != 0) { 1262#if HAVE_STRUCT_STAT_ST_MTIMESPEC_TV_NSEC 1263 archive_entry_set_mtime(entry, st->st_mtime, 1264 st->st_mtimespec.tv_nsec); 1265#elif HAVE_STRUCT_STAT_ST_MTIM_TV_NSEC 1266 archive_entry_set_mtime(entry, st->st_mtime, 1267 st->st_mtim.tv_nsec); 1268#elif HAVE_STRUCT_STAT_ST_MTIME_N 1269 archive_entry_set_mtime(entry, st->st_mtime, 1270 st->st_mtime_n); 1271#elif HAVE_STRUCT_STAT_ST_UMTIME 1272 archive_entry_set_mtime(entry, st->st_mtime, 1273 st->st_umtime*1000); 1274#elif HAVE_STRUCT_STAT_ST_MTIME_USEC 1275 archive_entry_set_mtime(entry, st->st_mtime, 1276 st->st_mtime_usec*1000); 1277#else 1278 archive_entry_set_mtime(entry, st->st_mtime, 0); 1279#endif 1280 } 1281 if ((parsed_kws & MTREE_HAS_NLINK) == 0 || 1282 (parsed_kws & MTREE_HAS_NOCHANGE) != 0) 1283 archive_entry_set_nlink(entry, st->st_nlink); 1284 if ((parsed_kws & MTREE_HAS_PERM) == 0 || 1285 (parsed_kws & MTREE_HAS_NOCHANGE) != 0) 1286 archive_entry_set_perm(entry, st->st_mode); 1287 if ((parsed_kws & MTREE_HAS_SIZE) == 0 || 1288 (parsed_kws & MTREE_HAS_NOCHANGE) != 0) 1289 archive_entry_set_size(entry, st->st_size); 1290 archive_entry_set_ino(entry, st->st_ino); 1291 archive_entry_set_dev(entry, st->st_dev); 1292 1293 archive_entry_linkify(mtree->resolver, &entry, 1294 &sparse_entry); 1295 } else if (parsed_kws & MTREE_HAS_OPTIONAL) { 1296 /* 1297 * Couldn't open the entry, stat it or the on-disk type 1298 * didn't match. If this entry is optional, just 1299 * ignore it and read the next header entry. 1300 */ 1301 *use_next = 1; 1302 return ARCHIVE_OK; 1303 } 1304 } 1305 1306 mtree->cur_size = archive_entry_size(entry); 1307 mtree->offset = 0; 1308 1309 return r; 1310} 1311 1312/* 1313 * Each line contains a sequence of keywords. 1314 */ 1315static int 1316parse_line(struct archive_read *a, struct archive_entry *entry, 1317 struct mtree *mtree, struct mtree_entry *mp, int *parsed_kws) 1318{ 1319 struct mtree_option *iter; 1320 int r = ARCHIVE_OK, r1; 1321 1322 for (iter = mp->options; iter != NULL; iter = iter->next) { 1323 r1 = parse_keyword(a, mtree, entry, iter, parsed_kws); 1324 if (r1 < r) 1325 r = r1; 1326 } 1327 if (r == ARCHIVE_OK && (*parsed_kws & MTREE_HAS_TYPE) == 0) { 1328 archive_set_error(&a->archive, ARCHIVE_ERRNO_FILE_FORMAT, 1329 "Missing type keyword in mtree specification"); 1330 return (ARCHIVE_WARN); 1331 } 1332 return (r); 1333} 1334 1335/* 1336 * Device entries have one of the following forms: 1337 * - raw dev_t 1338 * - format,major,minor[,subdevice] 1339 * When parsing succeeded, `pdev' will contain the appropriate dev_t value. 1340 */ 1341 1342/* strsep() is not in C90, but strcspn() is. */ 1343/* Taken from http://unixpapa.com/incnote/string.html */ 1344static char * 1345la_strsep(char **sp, const char *sep) 1346{ 1347 char *p, *s; 1348 if (sp == NULL || *sp == NULL || **sp == '\0') 1349 return(NULL); 1350 s = *sp; 1351 p = s + strcspn(s, sep); 1352 if (*p != '\0') 1353 *p++ = '\0'; 1354 *sp = p; 1355 return(s); 1356} 1357 1358static int 1359parse_device(dev_t *pdev, struct archive *a, char *val) 1360{ 1361#define MAX_PACK_ARGS 3 1362 unsigned long numbers[MAX_PACK_ARGS]; 1363 char *p, *dev; 1364 int argc; 1365 pack_t *pack; 1366 dev_t result; 1367 const char *error = NULL; 1368 1369 memset(pdev, 0, sizeof(*pdev)); 1370 if ((dev = strchr(val, ',')) != NULL) { 1371 /* 1372 * Device's major/minor are given in a specified format. 1373 * Decode and pack it accordingly. 1374 */ 1375 *dev++ = '\0'; 1376 if ((pack = pack_find(val)) == NULL) { 1377 archive_set_error(a, ARCHIVE_ERRNO_FILE_FORMAT, 1378 "Unknown format `%s'", val); 1379 return ARCHIVE_WARN; 1380 } 1381 argc = 0; 1382 while ((p = la_strsep(&dev, ",")) != NULL) { 1383 if (*p == '\0') { 1384 archive_set_error(a, ARCHIVE_ERRNO_FILE_FORMAT, 1385 "Missing number"); 1386 return ARCHIVE_WARN; 1387 } 1388 if (argc >= MAX_PACK_ARGS) { 1389 archive_set_error(a, ARCHIVE_ERRNO_FILE_FORMAT, 1390 "Too many arguments"); 1391 return ARCHIVE_WARN; 1392 } 1393 numbers[argc++] = (unsigned long)mtree_atol(&p); 1394 } 1395 if (argc < 2) { 1396 archive_set_error(a, ARCHIVE_ERRNO_FILE_FORMAT, 1397 "Not enough arguments"); 1398 return ARCHIVE_WARN; 1399 } 1400 result = (*pack)(argc, numbers, &error); 1401 if (error != NULL) { 1402 archive_set_error(a, ARCHIVE_ERRNO_FILE_FORMAT, 1403 "%s", error); 1404 return ARCHIVE_WARN; 1405 } 1406 } else { 1407 /* file system raw value. */ 1408 result = (dev_t)mtree_atol(&val); 1409 } 1410 *pdev = result; 1411 return ARCHIVE_OK; 1412#undef MAX_PACK_ARGS 1413} 1414 1415/* 1416 * Parse a single keyword and its value. 1417 */ 1418static int 1419parse_keyword(struct archive_read *a, struct mtree *mtree, 1420 struct archive_entry *entry, struct mtree_option *opt, int *parsed_kws) 1421{ 1422 char *val, *key; 1423 1424 key = opt->value; 1425 1426 if (*key == '\0') 1427 return (ARCHIVE_OK); 1428 1429 if (strcmp(key, "nochange") == 0) { 1430 *parsed_kws |= MTREE_HAS_NOCHANGE; 1431 return (ARCHIVE_OK); 1432 } 1433 if (strcmp(key, "optional") == 0) { 1434 *parsed_kws |= MTREE_HAS_OPTIONAL; 1435 return (ARCHIVE_OK); 1436 } 1437 if (strcmp(key, "ignore") == 0) { 1438 /* 1439 * The mtree processing is not recursive, so 1440 * recursion will only happen for explicitly listed 1441 * entries. 1442 */ 1443 return (ARCHIVE_OK); 1444 } 1445 1446 val = strchr(key, '='); 1447 if (val == NULL) { 1448 archive_set_error(&a->archive, ARCHIVE_ERRNO_FILE_FORMAT, 1449 "Malformed attribute \"%s\" (%d)", key, key[0]); 1450 return (ARCHIVE_WARN); 1451 } 1452 1453 *val = '\0'; 1454 ++val; 1455 1456 switch (key[0]) { 1457 case 'c': 1458 if (strcmp(key, "content") == 0 1459 || strcmp(key, "contents") == 0) { 1460 parse_escapes(val, NULL); 1461 archive_strcpy(&mtree->contents_name, val); 1462 break; 1463 } 1464 if (strcmp(key, "cksum") == 0) 1465 break; 1466 case 'd': 1467 if (strcmp(key, "device") == 0) { 1468 /* stat(2) st_rdev field, e.g. the major/minor IDs 1469 * of a char/block special file */ 1470 int r; 1471 dev_t dev; 1472 1473 *parsed_kws |= MTREE_HAS_DEVICE; 1474 r = parse_device(&dev, &a->archive, val); 1475 if (r == ARCHIVE_OK) 1476 archive_entry_set_rdev(entry, dev); 1477 return r; 1478 } 1479 case 'f': 1480 if (strcmp(key, "flags") == 0) { 1481 *parsed_kws |= MTREE_HAS_FFLAGS; 1482 archive_entry_copy_fflags_text(entry, val); 1483 break; 1484 } 1485 case 'g': 1486 if (strcmp(key, "gid") == 0) { 1487 *parsed_kws |= MTREE_HAS_GID; 1488 archive_entry_set_gid(entry, mtree_atol10(&val)); 1489 break; 1490 } 1491 if (strcmp(key, "gname") == 0) { 1492 *parsed_kws |= MTREE_HAS_GNAME; 1493 archive_entry_copy_gname(entry, val); 1494 break; 1495 } 1496 case 'i': 1497 if (strcmp(key, "inode") == 0) { 1498 archive_entry_set_ino(entry, mtree_atol10(&val)); 1499 break; 1500 } 1501 case 'l': 1502 if (strcmp(key, "link") == 0) { 1503 archive_entry_copy_symlink(entry, val); 1504 break; 1505 } 1506 case 'm': 1507 if (strcmp(key, "md5") == 0 || strcmp(key, "md5digest") == 0) 1508 break; 1509 if (strcmp(key, "mode") == 0) { 1510 if (val[0] >= '0' && val[0] <= '9') { 1511 *parsed_kws |= MTREE_HAS_PERM; 1512 archive_entry_set_perm(entry, 1513 (mode_t)mtree_atol8(&val)); 1514 } else { 1515 archive_set_error(&a->archive, 1516 ARCHIVE_ERRNO_FILE_FORMAT, 1517 "Symbolic mode \"%s\" unsupported", val); 1518 return ARCHIVE_WARN; 1519 } 1520 break; 1521 } 1522 case 'n': 1523 if (strcmp(key, "nlink") == 0) { 1524 *parsed_kws |= MTREE_HAS_NLINK; 1525 archive_entry_set_nlink(entry, 1526 (unsigned int)mtree_atol10(&val)); 1527 break; 1528 } 1529 case 'r': 1530 if (strcmp(key, "resdevice") == 0) { 1531 /* stat(2) st_dev field, e.g. the device ID where the 1532 * inode resides */ 1533 int r; 1534 dev_t dev; 1535 1536 r = parse_device(&dev, &a->archive, val); 1537 if (r == ARCHIVE_OK) 1538 archive_entry_set_dev(entry, dev); 1539 return r; 1540 } 1541 if (strcmp(key, "rmd160") == 0 || 1542 strcmp(key, "rmd160digest") == 0) 1543 break; 1544 case 's': 1545 if (strcmp(key, "sha1") == 0 || strcmp(key, "sha1digest") == 0) 1546 break; 1547 if (strcmp(key, "sha256") == 0 || 1548 strcmp(key, "sha256digest") == 0) 1549 break; 1550 if (strcmp(key, "sha384") == 0 || 1551 strcmp(key, "sha384digest") == 0) 1552 break; 1553 if (strcmp(key, "sha512") == 0 || 1554 strcmp(key, "sha512digest") == 0) 1555 break; 1556 if (strcmp(key, "size") == 0) { 1557 archive_entry_set_size(entry, mtree_atol10(&val)); 1558 break; 1559 } 1560 case 't': 1561 if (strcmp(key, "tags") == 0) { 1562 /* 1563 * Comma delimited list of tags. 1564 * Ignore the tags for now, but the interface 1565 * should be extended to allow inclusion/exclusion. 1566 */ 1567 break; 1568 } 1569 if (strcmp(key, "time") == 0) { 1570 int64_t m; 1571 int64_t my_time_t_max = get_time_t_max(); 1572 int64_t my_time_t_min = get_time_t_min(); 1573 long ns = 0; 1574 1575 *parsed_kws |= MTREE_HAS_MTIME; 1576 m = mtree_atol10(&val); 1577 /* Replicate an old mtree bug: 1578 * 123456789.1 represents 123456789 1579 * seconds and 1 nanosecond. */ 1580 if (*val == '.') { 1581 ++val; 1582 ns = (long)mtree_atol10(&val); 1583 } else 1584 ns = 0; 1585 if (m > my_time_t_max) 1586 m = my_time_t_max; 1587 else if (m < my_time_t_min) 1588 m = my_time_t_min; 1589 archive_entry_set_mtime(entry, (time_t)m, ns); 1590 break; 1591 } 1592 if (strcmp(key, "type") == 0) { 1593 switch (val[0]) { 1594 case 'b': 1595 if (strcmp(val, "block") == 0) { 1596 archive_entry_set_filetype(entry, AE_IFBLK); 1597 break; 1598 } 1599 case 'c': 1600 if (strcmp(val, "char") == 0) { 1601 archive_entry_set_filetype(entry, 1602 AE_IFCHR); 1603 break; 1604 } 1605 case 'd': 1606 if (strcmp(val, "dir") == 0) { 1607 archive_entry_set_filetype(entry, 1608 AE_IFDIR); 1609 break; 1610 } 1611 case 'f': 1612 if (strcmp(val, "fifo") == 0) { 1613 archive_entry_set_filetype(entry, 1614 AE_IFIFO); 1615 break; 1616 } 1617 if (strcmp(val, "file") == 0) { 1618 archive_entry_set_filetype(entry, 1619 AE_IFREG); 1620 break; 1621 } 1622 case 'l': 1623 if (strcmp(val, "link") == 0) { 1624 archive_entry_set_filetype(entry, 1625 AE_IFLNK); 1626 break; 1627 } 1628 default: 1629 archive_set_error(&a->archive, 1630 ARCHIVE_ERRNO_FILE_FORMAT, 1631 "Unrecognized file type \"%s\"; " 1632 "assuming \"file\"", val); 1633 archive_entry_set_filetype(entry, AE_IFREG); 1634 return (ARCHIVE_WARN); 1635 } 1636 *parsed_kws |= MTREE_HAS_TYPE; 1637 break; 1638 } 1639 case 'u': 1640 if (strcmp(key, "uid") == 0) { 1641 *parsed_kws |= MTREE_HAS_UID; 1642 archive_entry_set_uid(entry, mtree_atol10(&val)); 1643 break; 1644 } 1645 if (strcmp(key, "uname") == 0) { 1646 *parsed_kws |= MTREE_HAS_UNAME; 1647 archive_entry_copy_uname(entry, val); 1648 break; 1649 } 1650 default: 1651 archive_set_error(&a->archive, ARCHIVE_ERRNO_FILE_FORMAT, 1652 "Unrecognized key %s=%s", key, val); 1653 return (ARCHIVE_WARN); 1654 } 1655 return (ARCHIVE_OK); 1656} 1657 1658static int 1659read_data(struct archive_read *a, const void **buff, size_t *size, 1660 int64_t *offset) 1661{ 1662 size_t bytes_to_read; 1663 ssize_t bytes_read; 1664 struct mtree *mtree; 1665 1666 mtree = (struct mtree *)(a->format->data); 1667 if (mtree->fd < 0) { 1668 *buff = NULL; 1669 *offset = 0; 1670 *size = 0; 1671 return (ARCHIVE_EOF); 1672 } 1673 if (mtree->buff == NULL) { 1674 mtree->buffsize = 64 * 1024; 1675 mtree->buff = malloc(mtree->buffsize); 1676 if (mtree->buff == NULL) { 1677 archive_set_error(&a->archive, ENOMEM, 1678 "Can't allocate memory"); 1679 return (ARCHIVE_FATAL); 1680 } 1681 } 1682 1683 *buff = mtree->buff; 1684 *offset = mtree->offset; 1685 if ((int64_t)mtree->buffsize > mtree->cur_size - mtree->offset) 1686 bytes_to_read = (size_t)(mtree->cur_size - mtree->offset); 1687 else 1688 bytes_to_read = mtree->buffsize; 1689 bytes_read = read(mtree->fd, mtree->buff, bytes_to_read); 1690 if (bytes_read < 0) { 1691 archive_set_error(&a->archive, errno, "Can't read"); 1692 return (ARCHIVE_WARN); 1693 } 1694 if (bytes_read == 0) { 1695 *size = 0; 1696 return (ARCHIVE_EOF); 1697 } 1698 mtree->offset += bytes_read; 1699 *size = bytes_read; 1700 return (ARCHIVE_OK); 1701} 1702 1703/* Skip does nothing except possibly close the contents file. */ 1704static int 1705skip(struct archive_read *a) 1706{ 1707 struct mtree *mtree; 1708 1709 mtree = (struct mtree *)(a->format->data); 1710 if (mtree->fd >= 0) { 1711 close(mtree->fd); 1712 mtree->fd = -1; 1713 } 1714 return (ARCHIVE_OK); 1715} 1716 1717/* 1718 * Since parsing backslash sequences always makes strings shorter, 1719 * we can always do this conversion in-place. 1720 */ 1721static void 1722parse_escapes(char *src, struct mtree_entry *mentry) 1723{ 1724 char *dest = src; 1725 char c; 1726 1727 if (mentry != NULL && strcmp(src, ".") == 0) 1728 mentry->full = 1; 1729 1730 while (*src != '\0') { 1731 c = *src++; 1732 if (c == '/' && mentry != NULL) 1733 mentry->full = 1; 1734 if (c == '\\') { 1735 switch (src[0]) { 1736 case '0': 1737 if (src[1] < '0' || src[1] > '7') { 1738 c = 0; 1739 ++src; 1740 break; 1741 } 1742 /* FALLTHROUGH */ 1743 case '1': 1744 case '2': 1745 case '3': 1746 if (src[1] >= '0' && src[1] <= '7' && 1747 src[2] >= '0' && src[2] <= '7') { 1748 c = (src[0] - '0') << 6; 1749 c |= (src[1] - '0') << 3; 1750 c |= (src[2] - '0'); 1751 src += 3; 1752 } 1753 break; 1754 case 'a': 1755 c = '\a'; 1756 ++src; 1757 break; 1758 case 'b': 1759 c = '\b'; 1760 ++src; 1761 break; 1762 case 'f': 1763 c = '\f'; 1764 ++src; 1765 break; 1766 case 'n': 1767 c = '\n'; 1768 ++src; 1769 break; 1770 case 'r': 1771 c = '\r'; 1772 ++src; 1773 break; 1774 case 's': 1775 c = ' '; 1776 ++src; 1777 break; 1778 case 't': 1779 c = '\t'; 1780 ++src; 1781 break; 1782 case 'v': 1783 c = '\v'; 1784 ++src; 1785 break; 1786 case '\\': 1787 c = '\\'; 1788 ++src; 1789 break; 1790 } 1791 } 1792 *dest++ = c; 1793 } 1794 *dest = '\0'; 1795} 1796 1797/* 1798 * Note that this implementation does not (and should not!) obey 1799 * locale settings; you cannot simply substitute strtol here, since 1800 * it does obey locale. 1801 */ 1802static int64_t 1803mtree_atol8(char **p) 1804{ 1805 int64_t l, limit, last_digit_limit; 1806 int digit, base; 1807 1808 base = 8; 1809 limit = INT64_MAX / base; 1810 last_digit_limit = INT64_MAX % base; 1811 1812 l = 0; 1813 digit = **p - '0'; 1814 while (digit >= 0 && digit < base) { 1815 if (l>limit || (l == limit && digit > last_digit_limit)) { 1816 l = INT64_MAX; /* Truncate on overflow. */ 1817 break; 1818 } 1819 l = (l * base) + digit; 1820 digit = *++(*p) - '0'; 1821 } 1822 return (l); 1823} 1824 1825/* 1826 * Note that this implementation does not (and should not!) obey 1827 * locale settings; you cannot simply substitute strtol here, since 1828 * it does obey locale. 1829 */ 1830static int64_t 1831mtree_atol10(char **p) 1832{ 1833 int64_t l, limit, last_digit_limit; 1834 int base, digit, sign; 1835 1836 base = 10; 1837 1838 if (**p == '-') { 1839 sign = -1; 1840 limit = ((uint64_t)(INT64_MAX) + 1) / base; 1841 last_digit_limit = ((uint64_t)(INT64_MAX) + 1) % base; 1842 ++(*p); 1843 } else { 1844 sign = 1; 1845 limit = INT64_MAX / base; 1846 last_digit_limit = INT64_MAX % base; 1847 } 1848 1849 l = 0; 1850 digit = **p - '0'; 1851 while (digit >= 0 && digit < base) { 1852 if (l > limit || (l == limit && digit > last_digit_limit)) 1853 return (sign < 0) ? INT64_MIN : INT64_MAX; 1854 l = (l * base) + digit; 1855 digit = *++(*p) - '0'; 1856 } 1857 return (sign < 0) ? -l : l; 1858} 1859 1860/* Parse a hex digit. */ 1861static int 1862parsehex(char c) 1863{ 1864 if (c >= '0' && c <= '9') 1865 return c - '0'; 1866 else if (c >= 'a' && c <= 'f') 1867 return c - 'a'; 1868 else if (c >= 'A' && c <= 'F') 1869 return c - 'A'; 1870 else 1871 return -1; 1872} 1873 1874/* 1875 * Note that this implementation does not (and should not!) obey 1876 * locale settings; you cannot simply substitute strtol here, since 1877 * it does obey locale. 1878 */ 1879static int64_t 1880mtree_atol16(char **p) 1881{ 1882 int64_t l, limit, last_digit_limit; 1883 int base, digit, sign; 1884 1885 base = 16; 1886 1887 if (**p == '-') { 1888 sign = -1; 1889 limit = ((uint64_t)(INT64_MAX) + 1) / base; 1890 last_digit_limit = ((uint64_t)(INT64_MAX) + 1) % base; 1891 ++(*p); 1892 } else { 1893 sign = 1; 1894 limit = INT64_MAX / base; 1895 last_digit_limit = INT64_MAX % base; 1896 } 1897 1898 l = 0; 1899 digit = parsehex(**p); 1900 while (digit >= 0 && digit < base) { 1901 if (l > limit || (l == limit && digit > last_digit_limit)) 1902 return (sign < 0) ? INT64_MIN : INT64_MAX; 1903 l = (l * base) + digit; 1904 digit = parsehex(*++(*p)); 1905 } 1906 return (sign < 0) ? -l : l; 1907} 1908 1909static int64_t 1910mtree_atol(char **p) 1911{ 1912 if (**p != '0') 1913 return mtree_atol10(p); 1914 if ((*p)[1] == 'x' || (*p)[1] == 'X') { 1915 *p += 2; 1916 return mtree_atol16(p); 1917 } 1918 return mtree_atol8(p); 1919} 1920 1921/* 1922 * Returns length of line (including trailing newline) 1923 * or negative on error. 'start' argument is updated to 1924 * point to first character of line. 1925 */ 1926static ssize_t 1927readline(struct archive_read *a, struct mtree *mtree, char **start, 1928 ssize_t limit) 1929{ 1930 ssize_t bytes_read; 1931 ssize_t total_size = 0; 1932 ssize_t find_off = 0; 1933 const void *t; 1934 void *nl; 1935 char *u; 1936 1937 /* Accumulate line in a line buffer. */ 1938 for (;;) { 1939 /* Read some more. */ 1940 t = __archive_read_ahead(a, 1, &bytes_read); 1941 if (t == NULL) 1942 return (0); 1943 if (bytes_read < 0) 1944 return (ARCHIVE_FATAL); 1945 nl = memchr(t, '\n', bytes_read); 1946 /* If we found '\n', trim the read to end exactly there. */ 1947 if (nl != NULL) { 1948 bytes_read = ((const char *)nl) - ((const char *)t) + 1; 1949 } 1950 if (total_size + bytes_read + 1 > limit) { 1951 archive_set_error(&a->archive, 1952 ARCHIVE_ERRNO_FILE_FORMAT, 1953 "Line too long"); 1954 return (ARCHIVE_FATAL); 1955 } 1956 if (archive_string_ensure(&mtree->line, 1957 total_size + bytes_read + 1) == NULL) { 1958 archive_set_error(&a->archive, ENOMEM, 1959 "Can't allocate working buffer"); 1960 return (ARCHIVE_FATAL); 1961 } 1962 /* Append new bytes to string. */ 1963 memcpy(mtree->line.s + total_size, t, bytes_read); 1964 __archive_read_consume(a, bytes_read); 1965 total_size += bytes_read; 1966 mtree->line.s[total_size] = '\0'; 1967 1968 for (u = mtree->line.s + find_off; *u; ++u) { 1969 if (u[0] == '\n') { 1970 /* Ends with unescaped newline. */ 1971 *start = mtree->line.s; 1972 return total_size; 1973 } else if (u[0] == '#') { 1974 /* Ends with comment sequence #...\n */ 1975 if (nl == NULL) { 1976 /* But we've not found the \n yet */ 1977 break; 1978 } 1979 } else if (u[0] == '\\') { 1980 if (u[1] == '\n') { 1981 /* Trim escaped newline. */ 1982 total_size -= 2; 1983 mtree->line.s[total_size] = '\0'; 1984 break; 1985 } else if (u[1] != '\0') { 1986 /* Skip the two-char escape sequence */ 1987 ++u; 1988 } 1989 } 1990 } 1991 find_off = u - mtree->line.s; 1992 } 1993} 1994