2 * Copyright (c) Christos Zoulas 2003.
5 * Redistribution and use in source and binary forms, with or without
6 * modification, are permitted provided that the following conditions
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.
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
30 FILE_RCSID("@(#)$File: funcs.c,v 1.72 2014/05/14 23:15:42 christos Exp $")
39 #if defined(HAVE_WCHAR_H)
42 #if defined(HAVE_WCTYPE_H)
45 #if defined(HAVE_LIMITS_H)
48 #if defined(HAVE_LOCALE_H)
53 #define SIZE_MAX ((size_t)~0)
57 * Like printf, only we append to a buffer.
60 file_vprintf(struct magic_set *ms, const char *fmt, va_list ap)
65 if (ms->event_flags & EVENT_HAD_ERR)
67 len = vasprintf(&buf, fmt, ap);
71 if (ms->o.buf != NULL) {
72 len = asprintf(&newstr, "%s%s", ms->o.buf, buf);
82 file_error(ms, errno, "vasprintf failed");
87 file_printf(struct magic_set *ms, const char *fmt, ...)
93 rv = file_vprintf(ms, fmt, ap);
99 * error - print best error message possible
102 __attribute__((__format__(__printf__, 3, 0)))
104 file_error_core(struct magic_set *ms, int error, const char *f, va_list va,
107 /* Only the first error is ok */
108 if (ms->event_flags & EVENT_HAD_ERR)
113 file_printf(ms, "line %" SIZE_T_FORMAT "u: ", lineno);
115 file_vprintf(ms, f, va);
117 file_printf(ms, " (%s)", strerror(error));
118 ms->event_flags |= EVENT_HAD_ERR;
124 file_error(struct magic_set *ms, int error, const char *f, ...)
128 file_error_core(ms, error, f, va, 0);
133 * Print an error with magic line number.
137 file_magerror(struct magic_set *ms, const char *f, ...)
141 file_error_core(ms, 0, f, va, ms->line);
146 file_oomem(struct magic_set *ms, size_t len)
148 file_error(ms, errno, "cannot allocate %" SIZE_T_FORMAT "u bytes",
153 file_badseek(struct magic_set *ms)
155 file_error(ms, errno, "error seeking");
159 file_badread(struct magic_set *ms)
161 file_error(ms, errno, "error reading");
166 file_buffer(struct magic_set *ms, int fd, const char *inname __attribute__ ((unused)),
167 const void *buf, size_t nb)
169 int m = 0, rv = 0, looks_text = 0;
170 int mime = ms->flags & MAGIC_MIME;
171 const unsigned char *ubuf = CAST(const unsigned char *, buf);
172 unichar *u8buf = NULL;
174 const char *code = NULL;
175 const char *code_mime = "binary";
176 const char *type = "application/octet-stream";
177 const char *def = "data";
178 const char *ftype = NULL;
182 type = "application/x-empty";
184 } else if (nb == 1) {
185 def = "very short file (no magic)";
189 if ((ms->flags & MAGIC_NO_CHECK_ENCODING) == 0) {
190 looks_text = file_encoding(ms, ubuf, nb, &u8buf, &ulen,
191 &code, &code_mime, &ftype);
195 if ((ms->flags & MAGIC_NO_CHECK_APPTYPE) == 0 && inname) {
196 switch (file_os2_apptype(ms, inname, buf, nb)) {
207 /* try compression stuff */
208 if ((ms->flags & MAGIC_NO_CHECK_COMPRESS) == 0)
209 if ((m = file_zmagic(ms, fd, inname, ubuf, nb)) != 0) {
210 if ((ms->flags & MAGIC_DEBUG) != 0)
211 (void)fprintf(stderr, "zmagic %d\n", m);
215 /* Check if we have a tar file */
216 if ((ms->flags & MAGIC_NO_CHECK_TAR) == 0)
217 if ((m = file_is_tar(ms, ubuf, nb)) != 0) {
218 if ((ms->flags & MAGIC_DEBUG) != 0)
219 (void)fprintf(stderr, "tar %d\n", m);
223 /* Check if we have a CDF file */
224 if ((ms->flags & MAGIC_NO_CHECK_CDF) == 0)
225 if ((m = file_trycdf(ms, fd, ubuf, nb)) != 0) {
226 if ((ms->flags & MAGIC_DEBUG) != 0)
227 (void)fprintf(stderr, "cdf %d\n", m);
231 /* try soft magic tests */
232 if ((ms->flags & MAGIC_NO_CHECK_SOFT) == 0)
233 if ((m = file_softmagic(ms, ubuf, nb, 0, BINTEST,
235 if ((ms->flags & MAGIC_DEBUG) != 0)
236 (void)fprintf(stderr, "softmagic %d\n", m);
238 if ((ms->flags & MAGIC_NO_CHECK_ELF) == 0 && m == 1 &&
239 nb > 5 && fd != -1) {
241 * We matched something in the file, so this
242 * *might* be an ELF file, and the file is at
243 * least 5 bytes long, so if it's an ELF file
244 * it has at least one byte past the ELF magic
245 * number - try extracting information from the
246 * ELF headers that cannot easily * be
247 * extracted with rules in the magic file.
249 if ((m = file_tryelf(ms, fd, ubuf, nb)) != 0)
250 if ((ms->flags & MAGIC_DEBUG) != 0)
251 (void)fprintf(stderr,
258 /* try text properties */
259 if ((ms->flags & MAGIC_NO_CHECK_TEXT) == 0) {
261 if ((m = file_ascmagic(ms, ubuf, nb, looks_text)) != 0) {
262 if ((ms->flags & MAGIC_DEBUG) != 0)
263 (void)fprintf(stderr, "ascmagic %d\n", m);
271 if ((!mime || (mime & MAGIC_MIME_TYPE)) &&
272 file_printf(ms, "%s", mime ? type : def) == -1) {
276 if ((ms->flags & MAGIC_MIME_ENCODING) != 0) {
277 if (ms->flags & MAGIC_MIME_TYPE)
278 if (file_printf(ms, "; charset=") == -1)
280 if (file_printf(ms, "%s", code_mime) == -1)
295 file_reset(struct magic_set *ms)
297 if (ms->mlist[0] == NULL) {
298 file_error(ms, 0, "no magic files loaded");
309 ms->event_flags &= ~EVENT_HAD_ERR;
314 #define OCTALIFY(n, o) \
316 (void)(*(n)++ = '\\', \
317 *(n)++ = (((uint32_t)*(o) >> 6) & 3) + '0', \
318 *(n)++ = (((uint32_t)*(o) >> 3) & 7) + '0', \
319 *(n)++ = (((uint32_t)*(o) >> 0) & 7) + '0', \
322 protected const char *
323 file_getbuffer(struct magic_set *ms)
325 char *pbuf, *op, *np;
328 if (ms->event_flags & EVENT_HAD_ERR)
331 if (ms->flags & MAGIC_RAW)
334 if (ms->o.buf == NULL)
337 /* * 4 is for octal representation, + 1 is for NUL */
338 len = strlen(ms->o.buf);
339 if (len > (SIZE_MAX - 1) / 4) {
344 if ((pbuf = CAST(char *, realloc(ms->o.pbuf, psize))) == NULL) {
345 file_oomem(ms, psize);
350 #if defined(HAVE_WCHAR_H) && defined(HAVE_MBRTOWC) && defined(HAVE_WCWIDTH)
355 size_t bytesconsumed;
357 (void)memset(&state, 0, sizeof(mbstate_t));
364 bytesconsumed = mbrtowc(&nextchar, op,
365 (size_t)(eop - op), &state);
366 if (bytesconsumed == (size_t)(-1) ||
367 bytesconsumed == (size_t)(-2)) {
372 if (iswprint(nextchar)) {
373 (void)memcpy(np, op, bytesconsumed);
377 while (bytesconsumed-- > 0)
383 /* Parsing succeeded as a multi-byte sequence */
389 for (np = ms->o.pbuf, op = ms->o.buf; *op;) {
390 if (isprint((unsigned char)*op)) {
401 file_check_mem(struct magic_set *ms, unsigned int level)
405 if (level >= ms->c.len) {
406 len = (ms->c.len += 20) * sizeof(*ms->c.li);
407 ms->c.li = CAST(struct level_info *, (ms->c.li == NULL) ?
409 realloc(ms->c.li, len));
410 if (ms->c.li == NULL) {
415 ms->c.li[level].got_match = 0;
416 #ifdef ENABLE_CONDITIONALS
417 ms->c.li[level].last_match = 0;
418 ms->c.li[level].last_cond = COND_NONE;
419 #endif /* ENABLE_CONDITIONALS */
424 file_printedlen(const struct magic_set *ms)
426 return ms->o.buf == NULL ? 0 : strlen(ms->o.buf);
430 file_replace(struct magic_set *ms, const char *pat, const char *rep)
435 rc = file_regcomp(&rx, pat, REG_EXTENDED);
437 file_regerror(&rx, rc, ms);
441 while (file_regexec(&rx, ms->o.buf, 1, &rm, 0) == 0) {
442 ms->o.buf[rm.rm_so] = '\0';
443 if (file_printf(ms, "%s%s", rep,
444 rm.rm_eo != 0 ? ms->o.buf + rm.rm_eo : "") == -1)
456 file_regcomp(file_regex_t *rx, const char *pat, int flags)
458 rx->old_lc_ctype = setlocale(LC_CTYPE, NULL);
459 assert(rx->old_lc_ctype != NULL);
460 rx->old_lc_ctype = strdup(rx->old_lc_ctype);
461 assert(rx->old_lc_ctype != NULL);
464 (void)setlocale(LC_CTYPE, "C");
465 return rx->rc = regcomp(&rx->rx, pat, flags);
469 file_regexec(file_regex_t *rx, const char *str, size_t nmatch,
470 regmatch_t* pmatch, int eflags)
473 return regexec(&rx->rx, str, nmatch, pmatch, eflags);
477 file_regfree(file_regex_t *rx)
481 (void)setlocale(LC_CTYPE, rx->old_lc_ctype);
482 free(rx->old_lc_ctype);
486 file_regerror(file_regex_t *rx, int rc, struct magic_set *ms)
490 (void)regerror(rc, &rx->rx, errmsg, sizeof(errmsg));
491 file_magerror(ms, "regex error %d for `%s', (%s)", rc, rx->pat,