contrib/file/src/magic.c

   1 /*
   2  * Copyright (c) Christos Zoulas 2003.
   3  * All Rights Reserved.
   4  *
   5  * Redistribution and use in source and binary forms, with or without
   6  * modification, are permitted provided that the following conditions
   7  * are met:
   8  * 1. Redistributions of source code must retain the above copyright
   9  *    notice immediately at the beginning of the file, without modification,
  10  *    this list of conditions, and the following disclaimer.
  11  * 2. Redistributions in binary form must reproduce the above copyright
  12  *    notice, this list of conditions and the following disclaimer in the
  13  *    documentation and/or other materials provided with the distribution.
  14  *
  15  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
  16  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  17  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  18  * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE FOR
  19  * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  20  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  21  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  22  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  23  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  24  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  25  * SUCH DAMAGE.
  26  */
  27
  28 #include "file.h"
  29
  30 #ifndef lint
  31 FILE_RCSID("@(#)$File: magic.c,v 1.65 2009/09/14 17:50:38 christos Exp $")
  32 #endif  /* lint */
  33
  34 #include "magic.h"
  35
  36 #include <stdlib.h>
  37 #include <unistd.h>
  38 #include <string.h>
  39 #ifdef QUICK
  40 #include <sys/mman.h>
  41 #endif
  42 #ifdef HAVE_LIMITS_H
  43 #include <limits.h>     /* for PIPE_BUF */
  44 #endif
  45
  46 #if defined(HAVE_UTIMES)
  47 # include <sys/time.h>
  48 #elif defined(HAVE_UTIME)
  49 # if defined(HAVE_SYS_UTIME_H)
  50 #  include <sys/utime.h>
  51 # elif defined(HAVE_UTIME_H)
  52 #  include <utime.h>
  53 # endif
  54 #endif
  55
  56 #ifdef HAVE_UNISTD_H
  57 #include <unistd.h>     /* for read() */
  58 #endif
  59
  60 #include "patchlevel.h"
  61
  62 #ifndef PIPE_BUF
  63 /* Get the PIPE_BUF from pathconf */
  64 #ifdef _PC_PIPE_BUF
  65 #define PIPE_BUF pathconf(".", _PC_PIPE_BUF)
  66 #else
  67 #define PIPE_BUF 512
  68 #endif
  69 #endif
  70
  71 private void free_mlist(struct mlist *);
  72 private void close_and_restore(const struct magic_set *, const char *, int,
  73     const struct stat *);
  74 private int unreadable_info(struct magic_set *, mode_t, const char *);
  75 private const char* get_default_magic(void);
  76 #ifndef COMPILE_ONLY
  77 private const char *file_or_fd(struct magic_set *, const char *, int);
  78 #endif
  79
  80 #ifndef STDIN_FILENO
  81 #define STDIN_FILENO    0
  82 #endif
  83
  84 private const char *
  85 get_default_magic(void)
  86 {
  87         static const char hmagic[] = "/.magic";
  88         static char default_magic[2 * MAXPATHLEN + 2];
  89         char *home;
  90         char hmagicpath[MAXPATHLEN + 1];
  91
  92         if ((home = getenv("HOME")) == NULL)
  93                 return MAGIC;
  94
  95         (void)snprintf(hmagicpath, sizeof(hmagicpath), "%s%s", home, hmagic);
  96
  97         if (access(hmagicpath, R_OK) == -1)
  98                 return MAGIC;
  99
 100         (void)snprintf(default_magic, sizeof(default_magic), "%s:%s",
 101             hmagicpath, MAGIC);
 102
 103         return default_magic;
 104 }
 105
 106 public const char *
 107 magic_getpath(const char *magicfile, int action)
 108 {
 109         if (magicfile != NULL)
 110                 return magicfile;
 111
 112         magicfile = getenv("MAGIC");
 113         if (magicfile != NULL)
 114                 return magicfile;
 115
 116         return action == FILE_LOAD ? get_default_magic() : MAGIC;
 117 }
 118
 119 public struct magic_set *
 120 magic_open(int flags)
 121 {
 122         struct magic_set *ms;
 123         size_t len;
 124
 125         if ((ms = CAST(struct magic_set *, calloc((size_t)1,
 126             sizeof(struct magic_set)))) == NULL)
 127                 return NULL;
 128
 129         if (magic_setflags(ms, flags) == -1) {
 130                 errno = EINVAL;
 131                 goto free;
 132         }
 133
 134         ms->o.buf = ms->o.pbuf = NULL;
 135         len = (ms->c.len = 10) * sizeof(*ms->c.li);
 136
 137         if ((ms->c.li = CAST(struct level_info *, malloc(len))) == NULL)
 138                 goto free;
 139
 140         ms->event_flags = 0;
 141         ms->error = -1;
 142         ms->mlist = NULL;
 143         ms->file = "unknown";
 144         ms->line = 0;
 145         return ms;
 146 free:
 147         free(ms);
 148         return NULL;
 149 }
 150
 151 private void
 152 free_mlist(struct mlist *mlist)
 153 {
 154         struct mlist *ml;
 155
 156         if (mlist == NULL)
 157                 return;
 158
 159         for (ml = mlist->next; ml != mlist;) {
 160                 struct mlist *next = ml->next;
 161                 struct magic *mg = ml->magic;
 162                 file_delmagic(mg, ml->mapped, ml->nmagic);
 163                 free(ml);
 164                 ml = next;
 165         }
 166         free(ml);
 167 }
 168
 169 private int
 170 unreadable_info(struct magic_set *ms, mode_t md, const char *file)
 171 {
 172         /* We cannot open it, but we were able to stat it. */
 173         if (access(file, W_OK) == 0)
 174                 if (file_printf(ms, "writable, ") == -1)
 175                         return -1;
 176         if (access(file, X_OK) == 0)
 177                 if (file_printf(ms, "executable, ") == -1)
 178                         return -1;
 179         if (S_ISREG(md))
 180                 if (file_printf(ms, "regular file, ") == -1)
 181                         return -1;
 182         if (file_printf(ms, "no read permission") == -1)
 183                 return -1;
 184         return 0;
 185 }
 186
 187 public void
 188 magic_close(struct magic_set *ms)
 189 {
 190         free_mlist(ms->mlist);
 191         free(ms->o.pbuf);
 192         free(ms->o.buf);
 193         free(ms->c.li);
 194         free(ms);
 195 }
 196
 197 /*
 198  * load a magic file
 199  */
 200 public int
 201 magic_load(struct magic_set *ms, const char *magicfile)
 202 {
 203         struct mlist *ml = file_apprentice(ms, magicfile, FILE_LOAD);
 204         if (ml) {
 205                 free_mlist(ms->mlist);
 206                 ms->mlist = ml;
 207                 return 0;
 208         }
 209         return -1;
 210 }
 211
 212 public int
 213 magic_compile(struct magic_set *ms, const char *magicfile)
 214 {
 215         struct mlist *ml = file_apprentice(ms, magicfile, FILE_COMPILE);
 216         free_mlist(ml);
 217         return ml ? 0 : -1;
 218 }
 219
 220 public int
 221 magic_check(struct magic_set *ms, const char *magicfile)
 222 {
 223         struct mlist *ml = file_apprentice(ms, magicfile, FILE_CHECK);
 224         free_mlist(ml);
 225         return ml ? 0 : -1;
 226 }
 227
 228 private void
 229 close_and_restore(const struct magic_set *ms, const char *name, int fd,
 230     const struct stat *sb)
 231 {
 232         if (fd == STDIN_FILENO)
 233                 return;
 234         (void) close(fd);
 235
 236         if ((ms->flags & MAGIC_PRESERVE_ATIME) != 0) {
 237                 /*
 238                  * Try to restore access, modification times if read it.
 239                  * This is really *bad* because it will modify the status
 240                  * time of the file... And of course this will affect
 241                  * backup programs
 242                  */
 243 #ifdef HAVE_UTIMES
 244                 struct timeval  utsbuf[2];
 245                 (void)memset(utsbuf, 0, sizeof(utsbuf));
 246                 utsbuf[0].tv_sec = sb->st_atime;
 247                 utsbuf[1].tv_sec = sb->st_mtime;
 248
 249                 (void) utimes(name, utsbuf); /* don't care if loses */
 250 #elif defined(HAVE_UTIME_H) || defined(HAVE_SYS_UTIME_H)
 251                 struct utimbuf  utbuf;
 252
 253                 (void)memset(&utbuf, 0, sizeof(utbuf));
 254                 utbuf.actime = sb->st_atime;
 255                 utbuf.modtime = sb->st_mtime;
 256                 (void) utime(name, &utbuf); /* don't care if loses */
 257 #endif
 258         }
 259 }
 260
 261 #ifndef COMPILE_ONLY
 262
 263 /*
 264  * find type of descriptor
 265  */
 266 public const char *
 267 magic_descriptor(struct magic_set *ms, int fd)
 268 {
 269         return file_or_fd(ms, NULL, fd);
 270 }
 271
 272 /*
 273  * find type of named file
 274  */
 275 public const char *
 276 magic_file(struct magic_set *ms, const char *inname)
 277 {
 278         return file_or_fd(ms, inname, STDIN_FILENO);
 279 }
 280
 281 private const char *
 282 file_or_fd(struct magic_set *ms, const char *inname, int fd)
 283 {
 284         int     rv = -1;
 285         unsigned char *buf;
 286         struct stat     sb;
 287         ssize_t nbytes = 0;     /* number of bytes read from a datafile */
 288         int     ispipe = 0;
 289
 290         /*
 291          * one extra for terminating '\0', and
 292          * some overlapping space for matches near EOF
 293          */
 294 #define SLOP (1 + sizeof(union VALUETYPE))
 295         if ((buf = CAST(unsigned char *, malloc(HOWMANY + SLOP))) == NULL)
 296                 return NULL;
 297
 298         if (file_reset(ms) == -1)
 299                 goto done;
 300
 301         switch (file_fsmagic(ms, inname, &sb)) {
 302         case -1:                /* error */
 303                 goto done;
 304         case 0:                 /* nothing found */
 305                 break;
 306         default:                /* matched it and printed type */
 307                 rv = 0;
 308                 goto done;
 309         }
 310
 311         if (inname == NULL) {
 312                 if (fstat(fd, &sb) == 0 && S_ISFIFO(sb.st_mode))
 313                         ispipe = 1;
 314         } else {
 315                 int flags = O_RDONLY|O_BINARY;
 316
 317                 if (stat(inname, &sb) == 0 && S_ISFIFO(sb.st_mode)) {
 318                         flags |= O_NONBLOCK;
 319                         ispipe = 1;
 320                 }
 321
 322                 errno = 0;
 323                 if ((fd = open(inname, flags)) < 0) {
 324                         if (unreadable_info(ms, sb.st_mode, inname) == -1)
 325                                 goto done;
 326                         rv = 0;
 327                         goto done;
 328                 }
 329 #ifdef O_NONBLOCK
 330                 if ((flags = fcntl(fd, F_GETFL)) != -1) {
 331                         flags &= ~O_NONBLOCK;
 332                         (void)fcntl(fd, F_SETFL, flags);
 333                 }
 334 #endif
 335         }
 336
 337         /*
 338          * try looking at the first HOWMANY bytes
 339          */
 340         if (ispipe) {
 341                 ssize_t r = 0;
 342
 343                 while ((r = sread(fd, (void *)&buf[nbytes],
 344                     (size_t)(HOWMANY - nbytes), 1)) > 0) {
 345                         nbytes += r;
 346                         if (r < PIPE_BUF) break;
 347                 }
 348
 349                 if (nbytes == 0) {
 350                         /* We can not read it, but we were able to stat it. */
 351                         if (unreadable_info(ms, sb.st_mode, inname) == -1)
 352                                 goto done;
 353                         rv = 0;
 354                         goto done;
 355                 }
 356
 357         } else {
 358                 if ((nbytes = read(fd, (char *)buf, HOWMANY)) == -1) {
 359                         file_error(ms, errno, "cannot read `%s'", inname);
 360                         goto done;
 361                 }
 362         }
 363
 364         (void)memset(buf + nbytes, 0, SLOP); /* NUL terminate */
 365         if (file_buffer(ms, fd, inname, buf, (size_t)nbytes) == -1)
 366                 goto done;
 367         rv = 0;
 368 done:
 369         free(buf);
 370         close_and_restore(ms, inname, fd, &sb);
 371         return rv == 0 ? file_getbuffer(ms) : NULL;
 372 }
 373
 374
 375 public const char *
 376 magic_buffer(struct magic_set *ms, const void *buf, size_t nb)
 377 {
 378         if (file_reset(ms) == -1)
 379                 return NULL;
 380         /*
 381          * The main work is done here!
 382          * We have the file name and/or the data buffer to be identified.
 383          */
 384         if (file_buffer(ms, -1, NULL, buf, nb) == -1) {
 385                 return NULL;
 386         }
 387         return file_getbuffer(ms);
 388 }
 389 #endif
 390
 391 public const char *
 392 magic_error(struct magic_set *ms)
 393 {
 394         return (ms->event_flags & EVENT_HAD_ERR) ? ms->o.buf : NULL;
 395 }
 396
 397 public int
 398 magic_errno(struct magic_set *ms)
 399 {
 400         return (ms->event_flags & EVENT_HAD_ERR) ? ms->error : 0;
 401 }
 402
 403 public int
 404 magic_setflags(struct magic_set *ms, int flags)
 405 {
 406 #if !defined(HAVE_UTIME) && !defined(HAVE_UTIMES)
 407         if (flags & MAGIC_PRESERVE_ATIME)
 408                 return -1;
 409 #endif
 410         ms->flags = flags;
 411         return 0;
 412 }