contrib/file-4/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 #include "magic.h"
  30
  31 #include <stdio.h>
  32 #include <stdlib.h>
  33 #include <unistd.h>
  34 #include <string.h>
  35 #include <sys/types.h>
  36 #include <sys/param.h>  /* for MAXPATHLEN */
  37 #include <sys/stat.h>
  38 #include <fcntl.h>      /* for open() */
  39 #ifdef QUICK
  40 #include <sys/mman.h>
  41 #endif
  42
  43 #if defined(HAVE_UTIMES)
  44 # include <sys/time.h>
  45 #elif defined(HAVE_UTIME)
  46 # if defined(HAVE_SYS_UTIME_H)
  47 #  include <sys/utime.h>
  48 # elif defined(HAVE_UTIME_H)
  49 #  include <utime.h>
  50 # endif
  51 #endif
  52
  53 #ifdef HAVE_UNISTD_H
  54 #include <unistd.h>     /* for read() */
  55 #endif
  56
  57 #ifdef HAVE_LOCALE_H
  58 #include <locale.h>
  59 #endif
  60
  61 #include <netinet/in.h>         /* for byte swapping */
  62
  63 #include "patchlevel.h"
  64
  65 #ifndef lint
  66 FILE_RCSID("@(#)$Id: magic.c,v 1.25 2005/01/07 19:17:27 christos Exp $")
  67 #endif  /* lint */
  68
  69 #ifdef __EMX__
  70 private char *apptypeName = NULL;
  71 protected int file_os2_apptype(struct magic_set *ms, const char *fn,
  72     const void *buf, size_t nb);
  73 #endif /* __EMX__ */
  74
  75 private void free_mlist(struct mlist *);
  76 private void close_and_restore(const struct magic_set *, const char *, int,
  77     const struct stat *);
  78
  79 public struct magic_set *
  80 magic_open(int flags)
  81 {
  82         struct magic_set *ms;
  83
  84         if ((ms = malloc(sizeof(struct magic_set))) == NULL)
  85                 return NULL;
  86
  87         if (magic_setflags(ms, flags) == -1) {
  88                 errno = EINVAL;
  89                 goto free1;
  90         }
  91
  92         ms->o.ptr = ms->o.buf = malloc(ms->o.size = 1024);
  93         if (ms->o.buf == NULL)
  94                 goto free1;
  95
  96         ms->o.pbuf = malloc(ms->o.psize = 1024);
  97         if (ms->o.pbuf == NULL)
  98                 goto free2;
  99
 100         ms->c.off = malloc((ms->c.len = 10) * sizeof(*ms->c.off));
 101         if (ms->c.off == NULL)
 102                 goto free3;
 103
 104         ms->o.len = 0;
 105         ms->haderr = 0;
 106         ms->error = -1;
 107         ms->mlist = NULL;
 108         return ms;
 109 free3:
 110         free(ms->o.pbuf);
 111 free2:
 112         free(ms->o.buf);
 113 free1:
 114         free(ms);
 115         return NULL;
 116 }
 117
 118 private void
 119 free_mlist(struct mlist *mlist)
 120 {
 121         struct mlist *ml;
 122
 123         if (mlist == NULL)
 124                 return;
 125
 126         for (ml = mlist->next; ml != mlist;) {
 127                 struct mlist *next = ml->next;
 128                 struct magic *mg = ml->magic;
 129                 file_delmagic(mg, ml->mapped, ml->nmagic);
 130                 free(ml);
 131                 ml = next;
 132         }
 133         free(ml);
 134 }
 135
 136 public void
 137 magic_close(ms)
 138     struct magic_set *ms;
 139 {
 140         free_mlist(ms->mlist);
 141         free(ms->o.pbuf);
 142         free(ms->o.buf);
 143         free(ms->c.off);
 144         free(ms);
 145 }
 146
 147 /*
 148  * load a magic file
 149  */
 150 public int
 151 magic_load(struct magic_set *ms, const char *magicfile)
 152 {
 153         struct mlist *ml = file_apprentice(ms, magicfile, FILE_LOAD);
 154         if (ml) {
 155                 free_mlist(ms->mlist);
 156                 ms->mlist = ml;
 157                 return 0;
 158         }
 159         return -1;
 160 }
 161
 162 public int
 163 magic_compile(struct magic_set *ms, const char *magicfile)
 164 {
 165         struct mlist *ml = file_apprentice(ms, magicfile, FILE_COMPILE);
 166         free_mlist(ml);
 167         return ml ? 0 : -1;
 168 }
 169
 170 public int
 171 magic_check(struct magic_set *ms, const char *magicfile)
 172 {
 173         struct mlist *ml = file_apprentice(ms, magicfile, FILE_CHECK);
 174         free_mlist(ml);
 175         return ml ? 0 : -1;
 176 }
 177
 178 private void
 179 close_and_restore(const struct magic_set *ms, const char *name, int fd,
 180     const struct stat *sb)
 181 {
 182         (void) close(fd);
 183         if (fd != STDIN_FILENO && (ms->flags & MAGIC_PRESERVE_ATIME) != 0) {
 184                 /*
 185                  * Try to restore access, modification times if read it.
 186                  * This is really *bad* because it will modify the status
 187                  * time of the file... And of course this will affect
 188                  * backup programs
 189                  */
 190 #ifdef HAVE_UTIMES
 191                 struct timeval  utsbuf[2];
 192                 utsbuf[0].tv_sec = sb->st_atime;
 193                 utsbuf[1].tv_sec = sb->st_mtime;
 194
 195                 (void) utimes(name, utsbuf); /* don't care if loses */
 196 #elif defined(HAVE_UTIME_H) || defined(HAVE_SYS_UTIME_H)
 197                 struct utimbuf  utbuf;
 198
 199                 utbuf.actime = sb->st_atime;
 200                 utbuf.modtime = sb->st_mtime;
 201                 (void) utime(name, &utbuf); /* don't care if loses */
 202 #endif
 203         }
 204 }
 205
 206 #ifndef COMPILE_ONLY
 207 /*
 208  * find type of named file
 209  */
 210 public const char *
 211 magic_file(struct magic_set *ms, const char *inname)
 212 {
 213         int     fd = 0;
 214         unsigned char buf[HOWMANY+1];   /* one extra for terminating '\0' */
 215         struct stat     sb;
 216         ssize_t nbytes = 0;     /* number of bytes read from a datafile */
 217
 218         if (file_reset(ms) == -1)
 219                 return NULL;
 220
 221         switch (file_fsmagic(ms, inname, &sb)) {
 222         case -1:
 223                 return NULL;
 224         case 0:
 225                 break;
 226         default:
 227                 return file_getbuffer(ms);
 228         }
 229
 230 #ifndef STDIN_FILENO
 231 #define STDIN_FILENO    0
 232 #endif
 233         if (inname == NULL)
 234                 fd = STDIN_FILENO;
 235         else if ((fd = open(inname, O_RDONLY)) < 0) {
 236                 /* We cannot open it, but we were able to stat it. */
 237                 if (sb.st_mode & 0222)
 238                         if (file_printf(ms, "writable, ") == -1)
 239                                 return NULL;
 240                 if (sb.st_mode & 0111)
 241                         if (file_printf(ms, "executable, ") == -1)
 242                                 return NULL;
 243                 if (S_ISREG(sb.st_mode))
 244                         if (file_printf(ms, "regular file, ") == -1)
 245                                 return NULL;
 246                 if (file_printf(ms, "no read permission") == -1)
 247                         return NULL;
 248                 return file_getbuffer(ms);
 249         }
 250
 251         /*
 252          * try looking at the first HOWMANY bytes
 253          */
 254         if ((nbytes = read(fd, (char *)buf, HOWMANY)) == -1) {
 255                 file_error(ms, errno, "cannot read `%s'", inname);
 256                 goto done;
 257         }
 258
 259         if (nbytes == 0) {
 260                 if (file_printf(ms, (ms->flags & MAGIC_MIME) ?
 261                     "application/x-empty" : "empty") == -1)
 262                         goto done;
 263                 goto gotit;
 264         } else if (nbytes == 1) {
 265                 if (file_printf(ms, "very short file (no magic)") == -1)
 266                         goto done;
 267                 goto gotit;
 268         } else {
 269                 buf[nbytes] = '\0';     /* null-terminate it */
 270 #ifdef __EMX__
 271                 switch (file_os2_apptype(ms, inname, buf, nbytes)) {
 272                 case -1:
 273                         goto done;
 274                 case 0:
 275                         break;
 276                 default:
 277                         goto gotit;
 278                 }
 279 #endif
 280                 if (file_buffer(ms, fd, buf, (size_t)nbytes) == -1)
 281                         goto done;
 282 #ifdef BUILTIN_ELF
 283                 if (nbytes > 5) {
 284                         /*
 285                          * We matched something in the file, so this *might*
 286                          * be an ELF file, and the file is at least 5 bytes
 287                          * long, so if it's an ELF file it has at least one
 288                          * byte past the ELF magic number - try extracting
 289                          * information from the ELF headers that cannot easily
 290                          * be extracted with rules in the magic file.
 291                          */
 292                         file_tryelf(ms, fd, buf, (size_t)nbytes);
 293                 }
 294 #endif
 295         }
 296 gotit:
 297         close_and_restore(ms, inname, fd, &sb);
 298         return file_getbuffer(ms);
 299 done:
 300         close_and_restore(ms, inname, fd, &sb);
 301         return NULL;
 302 }
 303
 304
 305 public const char *
 306 magic_buffer(struct magic_set *ms, const void *buf, size_t nb)
 307 {
 308         if (file_reset(ms) == -1)
 309                 return NULL;
 310         /*
 311          * The main work is done here!
 312          * We have the file name and/or the data buffer to be identified.
 313          */
 314         if (file_buffer(ms, -1, buf, nb) == -1) {
 315                 return NULL;
 316         }
 317         return file_getbuffer(ms);
 318 }
 319 #endif
 320
 321 public const char *
 322 magic_error(struct magic_set *ms)
 323 {
 324         return ms->haderr ? ms->o.buf : NULL;
 325 }
 326
 327 public int
 328 magic_errno(struct magic_set *ms)
 329 {
 330         return ms->haderr ? ms->error : 0;
 331 }
 332
 333 public int
 334 magic_setflags(struct magic_set *ms, int flags)
 335 {
 336 #if !defined(HAVE_UTIME) && !defined(HAVE_UTIMES)
 337         if (flags & MAGIC_PRESERVE_ATIME)
 338                 return -1;
 339 #endif
 340         ms->flags = flags;
 341         return 0;
 342 }