summaryrefslogtreecommitdiff
path: root/libbb/unarchive.c
blob: e4150bda23f835fbb737acba7431ddc3b319fff1 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
/*
 *  Copyright (C) 2000 by Glenn McGrath
 *  Copyright (C) 2001 by Laurence Anderson
 *	
 *  Based on previous work by busybox developers and others.
 *
 *  This program is free software; you can redistribute it and/or modify
 *  it under the terms of the GNU General Public License as published by
 *  the Free Software Foundation; either version 2 of the License, or
 *  (at your option) any later version.
 *
 *  This program is distributed in the hope that it will be useful,
 *  but WITHOUT ANY WARRANTY; without even the implied warranty of
 *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 *  GNU Library General Public License for more details.
 *
 *  You should have received a copy of the GNU General Public License
 *  along with this program; if not, write to the Free Software
 *  Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
 */

#include <stdio.h>
#include <errno.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#include <utime.h>
#include "libbb.h"

typedef struct file_headers_s {
	char *name;
	char *link_name;
	off_t size;
	uid_t uid;
	gid_t gid;
	mode_t mode;
	time_t mtime;
	dev_t device;
} file_header_t;

off_t archive_offset;

void seek_sub_file(FILE *src_stream, const int count)
{
	int i;
	/* Try to fseek as faster */
	archive_offset += count;
	if (fseek(src_stream, count, SEEK_CUR) != 0 && errno == ESPIPE) {
	for (i = 0; i < count; i++) {
		fgetc(src_stream);
		}
	}
	return;
}


/* Extract the data postioned at src_stream to either filesystem, stdout or 
 * buffer depending on the value of 'function' which is defined in libbb.h 
 *
 * prefix doesnt have to be just a directory, it may prefix the filename as well.
 *
 * e.g. '/var/lib/dpkg/info/dpkg.' will extract all files to the base bath 
 * '/var/lib/dpkg/info/' and all files/dirs created in that dir will have 
 * 'dpkg.' as their prefix
 *
 * For this reason if prefix does point to a dir then it must end with a
 * trailing '/' or else the last dir will be assumed to be the file prefix 
 */
char *extract_archive(FILE *src_stream, FILE *out_stream, const file_header_t *file_entry,
 const int function, const char *prefix)
{
	FILE *dst_stream = NULL;
	char *full_name = NULL;
	char *buffer = NULL;
	struct utimbuf t;

	/* prefix doesnt have to be a proper path it may prepend 
	 * the filename as well */
	if (prefix != NULL) {
		/* strip leading '/' in filename to extract as prefix may not be dir */
		/* Cant use concat_path_file here as prefix might not be a directory */
		char *path = file_entry->name;
		if (*path == '/') {
			path++;
		}
		full_name = xmalloc(strlen(prefix) + strlen(path) + 1);
		strcpy(full_name, prefix);
		strcat(full_name, path);
	} else {
		full_name = file_entry->name;
	}

	if (function & extract_to_stdout) {
		if (S_ISREG(file_entry->mode)) {
			copy_file_chunk(src_stream, out_stream, file_entry->size);			
			archive_offset += file_entry->size;
		}
	}
	else if (function & extract_one_to_buffer) { 
		if (S_ISREG(file_entry->mode)) {
			buffer = (char *) xmalloc(file_entry->size + 1);
			fread(buffer, 1, file_entry->size, src_stream);
			archive_offset += file_entry->size;
			return(buffer);
		}
	}
	else if (function & extract_all_to_fs) {
#if 0
		struct stat oldfile;
		if ( (S_ISLNK(file_entry->mode) ? lstat (full_name, &oldfile) : stat (full_name, &oldfile)) == 0) { /* The file already exists */
			if (function & extract_unconditional || oldfile.st_mtime < file_entry->mtime) {
				if (!S_ISDIR(oldfile.st_mode)) {
					unlink(full_name); /* Directories might not be empty etc */
				}
			} else {
				error_msg("%s not created: newer or same age file exists", file_entry->name);
				if (S_ISREG(file_entry->mode)) {
					seek_sub_file(src_stream, file_entry->size);
				}
				return (NULL);
			}
		}
#endif
		switch(file_entry->mode & S_IFMT) {
			case S_IFREG:
				if (file_entry->link_name) { /* Found a cpio hard link */
					if (link(file_entry->link_name, full_name) != 0) {
						perror_msg("Cannot link from %s to '%s'",
							file_entry->name, file_entry->link_name);
					}
				} else {
					if ((dst_stream = wfopen(full_name, "w")) == NULL) {
						seek_sub_file(src_stream, file_entry->size);
						return NULL;
					}
					archive_offset += file_entry->size;
					copy_file_chunk(src_stream, dst_stream, file_entry->size);			
					fclose(dst_stream);
				}
				break;
			case S_IFDIR:
				/* Use make_directory instead of mkdir in case prefix path hasn't been created */
				if (function & extract_create_dirs) {
					if (make_directory(full_name, file_entry->mode, FILEUTILS_RECUR) < 0) {
						return NULL;
					}
				}
				break;
			case S_IFLNK:
				if (symlink(file_entry->link_name, full_name) < 0) {
					perror_msg("Cannot create symlink from %s to '%s'", file_entry->name, file_entry->link_name); 
					return NULL;
				}
				break;
			case S_IFSOCK:
			case S_IFBLK:
			case S_IFCHR:
			case S_IFIFO:
				if (mknod(full_name, file_entry->mode, file_entry->device) == -1) {
					perror_msg("Cannot create node %s", file_entry->name);
					return NULL;
				}
				break;
		}
		if (function & extract_preserve_date) {
			t.actime = file_entry->mtime;
			t.modtime = file_entry->mtime;
			utime(full_name, &t);
		}
		chmod(full_name, file_entry->mode);
		lchown(full_name, file_entry->uid, file_entry->gid);
	} else {
		/* If we arent extracting data we have to skip it, 
		 * if data size is 0 then then just do it anyway
		 * (saves testing for it) */
		seek_sub_file(src_stream, file_entry->size);
	}

	/* extract_list and extract_verbose_list can be used in conjunction
	 * with one of the above four extraction functions, so do this seperately */
	if (function & extract_verbose_list) {
		fprintf(out_stream, "%s %d/%d %8d %s ", mode_string(file_entry->mode), 
			file_entry->uid, file_entry->gid,
			(int) file_entry->size, time_string(file_entry->mtime));
	}
	if ((function & extract_list) || (function & extract_verbose_list)){
		/* fputs doesnt add a trailing \n, so use fprintf */
		fprintf(out_stream, "%s\n", file_entry->name);
	}

	free(full_name);

	return(NULL); /* Maybe we should say if failed */
}

char *unarchive(FILE *src_stream, void *(*get_headers)(FILE *),
	const int extract_function, const char *prefix, char **extract_names)
{
	file_header_t *file_entry;
	int found;
	int i;
	char *buffer = NULL;

	archive_offset = 0;
	while ((file_entry = (file_header_t *) get_headers(src_stream)) != NULL) {
		found = FALSE;
		if (extract_names[0] != NULL) {
			for(i = 0; extract_names[i] != 0; i++) {
				if (strcmp(extract_names[i], file_entry->name) == 0) {
					found = TRUE;
				}
			}
			if (!found) {
				/* seek past the data entry */
				if (!S_ISLNK(file_entry->mode) && file_entry->link_name && file_entry->size == 0) {
					error_msg("You should extract %s as other files are hardlinked to it", file_entry->name);
				}
				seek_sub_file(src_stream, file_entry->size);
				continue;
			}
		}
		buffer = extract_archive(src_stream, stdout, file_entry, extract_function, prefix);
	}
	return(buffer);
}

#if defined BB_AR || defined BB_DPKG_DEB || defined BB_DPKG
void *get_header_ar(FILE *src_stream)
{
	file_header_t *typed;
	union {
		char raw[60];
	 	struct {
 			char name[16];
 			char date[12];
 			char uid[6];
 			char gid[6];
 			char mode[8];
 			char size[10];
 			char magic[2];
 		} formated;
	} ar;
	static char *ar_long_names;

	if (fread(ar.raw, 1, 60, src_stream) != 60) {
		return(NULL);
	}
	archive_offset += 60;
	/* align the headers based on the header magic */
	if ((ar.formated.magic[0] != '`') || (ar.formated.magic[1] != '\n')) {
		/* some version of ar, have an extra '\n' after each data entry,
		 * this puts the next header out by 1 */
		if (ar.formated.magic[1] != '`') {
			error_msg("Invalid magic");
			return(NULL);
		}
		/* read the next char out of what would be the data section,
		 * if its a '\n' then it is a valid header offset by 1*/
		archive_offset++;
		if (fgetc(src_stream) != '\n') {
			error_msg("Invalid magic");
			return(NULL);
		}
		/* fix up the header, we started reading 1 byte too early */
		/* raw_header[60] wont be '\n' as it should, but it doesnt matter */
		memmove(ar.raw, &ar.raw[1], 59);
	}
		
	typed = (file_header_t *) xcalloc(1, sizeof(file_header_t));

	typed->size = (size_t) atoi(ar.formated.size);
	/* long filenames have '/' as the first character */
	if (ar.formated.name[0] == '/') {
		if (ar.formated.name[1] == '/') {
			/* If the second char is a '/' then this entries data section
			 * stores long filename for multiple entries, they are stored
			 * in static variable long_names for use in future entries */
			ar_long_names = (char *) xrealloc(ar_long_names, typed->size);
			fread(ar_long_names, 1, typed->size, src_stream);
			archive_offset += typed->size;
			/* This ar entries data section only contained filenames for other records
			 * they are stored in the static ar_long_names for future reference */
			return(NULL);
		} else {
			/* The number after the '/' indicates the offset in the ar data section
			(saved in variable long_name) that conatains the real filename */
			if (!ar_long_names) {
				error_msg("Cannot resolve long file name");
				return (NULL);
			}
			typed->name = xstrdup(ar_long_names + atoi(&ar.formated.name[1]));
		}
	} else {
		/* short filenames */
		typed->name = xcalloc(1, 16);
		strncpy(typed->name, ar.formated.name, 16);
	}
	typed->name[strcspn(typed->name, " /")]='\0';

	/* convert the rest of the now valid char header to its typed struct */	
	parse_mode(ar.formated.mode, &typed->mode);
	typed->mtime = atoi(ar.formated.date);
	typed->uid = atoi(ar.formated.uid);
	typed->gid = atoi(ar.formated.gid);

	return(typed);
}
#endif

#if defined BB_CPIO
void *get_header_cpio(FILE *src_stream)
{
	file_header_t *cpio_entry = NULL;
	char cpio_header[110];
	char dummy[14];
	int namesize;
	int major, minor, nlink;

	/* There can be padding before archive header */
	seek_sub_file(src_stream, (4 - (archive_offset % 4)) % 4);
	if (fread(cpio_header, 1, 110, src_stream) == 110) {
		archive_offset += 110;
		if (strncmp(cpio_header, "07070", 5) != 0) {
			error_msg("Unsupported format or invalid magic");
			return(NULL);
		}
		switch (cpio_header[5]) {
			case '2': /* "crc" header format */
				/* Doesnt do the crc check yet */
			case '1': /* "newc" header format */
				cpio_entry = (file_header_t *) xcalloc(1, sizeof(file_header_t));
				sscanf(cpio_header, "%14c%8x%8x%8x%8x%8lx%8lx%16c%8x%8x%8x%8c",
					dummy, &cpio_entry->mode, &cpio_entry->uid, &cpio_entry->gid,
					&nlink, &cpio_entry->mtime, &cpio_entry->size,
					dummy, &major, &minor, &namesize, dummy);

				cpio_entry->name = (char *) xcalloc(1, namesize);
				fread(cpio_entry->name, 1, namesize, src_stream); /* Read in filename */
				archive_offset += namesize;
				/* Skip padding before file contents */
				seek_sub_file(src_stream, (4 - (archive_offset % 4)) % 4);
				if (strcmp(cpio_entry->name, "TRAILER!!!") == 0) {
					printf("%d blocks\n", (int) (archive_offset % 512 ? (archive_offset / 512) + 1 : archive_offset / 512)); /* Always round up */
					return(NULL);
				}

				if (S_ISLNK(cpio_entry->mode)) {
					cpio_entry->link_name = (char *) xcalloc(1, cpio_entry->size + 1);
					fread(cpio_entry->link_name, 1, cpio_entry->size, src_stream);
					archive_offset += cpio_entry->size;
				}
				if (nlink > 1 && !S_ISDIR(cpio_entry->mode) && cpio_entry->size == 0) {
					error_msg("%s not extracted: Cannot handle hard links yet", cpio_entry->name);
					return(get_header_cpio(src_stream)); /* Recurse to next file */
				}
				cpio_entry->device = (major << 8) | minor;
				break;
			default:
				error_msg("Unsupported format");
				return(NULL);
		}
		if (ferror(src_stream) || feof(src_stream)) {
			perror_msg("Stream error");
			return(NULL);
		}
	}
	return(cpio_entry);
}
#endif

#if defined BB_UNTAR || defined BB_DPKG_DEB || defined BB_DPKG
void *get_header_tar(FILE *tar_stream)
{
	union {
		unsigned char raw[512];
		struct {
			char name[100];		/*   0-99 */
			char mode[8];		/* 100-107 */
			char uid[8];		/* 108-115 */
			char gid[8];		/* 116-123 */
			char size[12];		/* 124-135 */
			char mtime[12];		/* 136-147 */
			char chksum[8];		/* 148-155 */
			char typeflag;		/* 156-156 */
			char linkname[100];	/* 157-256 */
			char magic[6];		/* 257-262 */
			char version[2];	/* 263-264 */
			char uname[32];		/* 265-296 */
			char gname[32];		/* 297-328 */
			char devmajor[8];	/* 329-336 */
			char devminor[8];	/* 337-344 */
			char prefix[155];	/* 345-499 */
			char padding[12];	/* 500-512 */
		} formated;
	} tar;
	file_header_t *tar_entry = NULL;
	long i;
	long sum = 0;

	if (archive_offset % 512 != 0) {
		seek_sub_file(tar_stream, 512 - (archive_offset % 512));
	}

	if (fread(tar.raw, 1, 512, tar_stream) != 512) {
		error_msg("Couldnt read header");
		return(NULL);
	}
	archive_offset += 512;

	/* Check header has valid magic, unfortunately some tar files
	 * have empty (0'ed) tar entries at the end, which will
	 * cause this to fail, so fail silently for now
	 */
	if (strncmp(tar.formated.magic, "ustar", 5) != 0) {
		return(NULL);
	}

	/* Do checksum on headers */
	for (i =  0; i < 148 ; i++) {
		sum += tar.raw[i];
	}
	sum += ' ' * 8;
	for (i =  156; i < 512 ; i++) {
		sum += tar.raw[i];
	}
	if (sum != strtol(tar.formated.chksum, NULL, 8)) {
		error_msg("Invalid tar header checksum");
		return(NULL);
	}

	/* convert to type'ed variables */
	tar_entry = xcalloc(1, sizeof(file_header_t));
	tar_entry->name = xstrdup(tar.formated.name);

	parse_mode(tar.formated.mode, &tar_entry->mode);
	tar_entry->uid   = strtol(tar.formated.uid, NULL, 8);
	tar_entry->gid   = strtol(tar.formated.gid, NULL, 8);
	tar_entry->size  = strtol(tar.formated.size, NULL, 8);
	tar_entry->mtime = strtol(tar.formated.mtime, NULL, 8);
	tar_entry->link_name  = strlen(tar.formated.linkname) ? xstrdup(tar.formated.linkname) : NULL;
	tar_entry->device = (strtol(tar.formated.devmajor, NULL, 8) << 8) +
		strtol(tar.formated.devminor, NULL, 8);

	return(tar_entry);
}
#endif

#if defined BB_DPKG || defined BB_DPKG_DEB
char *deb_extract(const char *package_filename, FILE *out_stream, const int extract_function,
	const char *prefix, const char *filename)
{
	FILE *deb_stream;
	FILE *uncompressed_stream = NULL;
	file_header_t *ar_header = NULL;
	char *output_buffer = NULL;
	char *ared_file = NULL;
	char ar_magic[8];
	char **file_list;
	int gunzip_pid;

	file_list = malloc(sizeof(char *));
	file_list[0] = xstrdup(filename);
	file_list[1] = NULL;

	if (extract_function & extract_control_tar_gz) {
		ared_file = xstrdup("control.tar.gz");
	}
	else if (extract_function & extract_data_tar_gz) {		
		ared_file = xstrdup("data.tar.gz");
	}

	/* open the debian package to be worked on */
	deb_stream = wfopen(package_filename, "r");

	/* check ar magic */
	fread(ar_magic, 1, 8, deb_stream);
	if (strncmp(ar_magic,"!<arch>",7) != 0) {
		error_msg_and_die("invalid magic");
	}
	archive_offset = 8;

	while ((ar_header = get_header_ar(deb_stream)) != NULL) {
		if (strcmp(ared_file, ar_header->name) == 0) {
			/* open a stream of decompressed data */
			uncompressed_stream = gz_open(deb_stream, &gunzip_pid);
			archive_offset = 0;
			output_buffer = unarchive(uncompressed_stream, get_header_tar, extract_function, prefix, file_list);
		}
		seek_sub_file(deb_stream, ar_header->size);
	}
	gz_close(gunzip_pid);
	fclose(deb_stream);
	fclose(uncompressed_stream);
	free(ared_file);
	return(output_buffer);
}
#endif