2 * Copyright (c) Ian F. Darwin 1986-1995.
3 * Software written by Ian F. Darwin and others;
4 * maintained 1995-present by Christos Zoulas and others.
6 * Redistribution and use in source and binary forms, with or without
7 * modification, are permitted provided that the following conditions
9 * 1. Redistributions of source code must retain the above copyright
10 * notice immediately at the beginning of the file, without modification,
11 * 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.
16 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
17 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
18 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
19 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE FOR
20 * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
21 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
22 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
23 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
24 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
25 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
29 * ASCII magic -- try to detect text encoding.
31 * Extensively modified by Eric Fischer <enf@pobox.com> in July, 2000,
32 * to handle character codes other than ASCII on a unified basis.
38 FILE_RCSID("@(#)$File: ascmagic.c,v 1.81 2011/03/15 22:16:29 christos Exp $")
51 #define MAXLINELEN 300 /* longest sane line length */
52 #define ISSPC(x) ((x) == ' ' || (x) == '\t' || (x) == '\r' || (x) == '\n' \
53 || (x) == 0x85 || (x) == '\f')
55 private int ascmatch(const unsigned char *, const unichar *, size_t);
56 private unsigned char *encode_utf8(unsigned char *, size_t, unichar *, size_t);
57 private size_t trim_nuls(const unsigned char *, size_t);
60 * Undo the NUL-termination kindly provided by process()
61 * but leave at least one byte to look at
64 trim_nuls(const unsigned char *buf, size_t nbytes)
66 while (nbytes > 1 && buf[nbytes - 1] == '\0')
73 file_ascmagic(struct magic_set *ms, const unsigned char *buf, size_t nbytes)
79 const char *code = NULL;
80 const char *code_mime = NULL;
81 const char *type = NULL;
83 if (ms->flags & MAGIC_APPLE)
86 nbytes = trim_nuls(buf, nbytes);
88 /* If file doesn't look like any sort of text, give up. */
89 if (file_encoding(ms, buf, nbytes, &ubuf, &ulen, &code, &code_mime,
95 rv = file_ascmagic_with_encoding(ms, buf, nbytes, ubuf, ulen, code,
106 file_ascmagic_with_encoding(struct magic_set *ms, const unsigned char *buf,
107 size_t nbytes, unichar *ubuf, size_t ulen, const char *code,
110 unsigned char *utf8_buf = NULL, *utf8_end;
112 const struct names *p;
114 int mime = ms->flags & MAGIC_MIME;
116 const char *subtype = NULL;
117 const char *subtype_mime = NULL;
120 int has_backspace = 0;
127 int score, curtype, executable = 0;
129 size_t last_line_end = (size_t)-1;
130 int has_long_lines = 0;
132 if (ms->flags & MAGIC_APPLE)
135 nbytes = trim_nuls(buf, nbytes);
137 /* If we have fewer than 2 bytes, give up. */
143 if ((ms->flags & MAGIC_NO_CHECK_SOFT) == 0) {
144 /* Convert ubuf to UTF-8 and try text soft magic */
145 /* malloc size is a conservative overestimate; could be
146 improved, or at least realloced after conversion. */
148 if ((utf8_buf = CAST(unsigned char *, malloc(mlen))) == NULL) {
149 file_oomem(ms, mlen);
152 if ((utf8_end = encode_utf8(utf8_buf, mlen, ubuf, ulen))
155 if ((rv = file_softmagic(ms, utf8_buf,
156 (size_t)(utf8_end - utf8_buf), TEXTTEST)) != 0)
157 goto subtype_identified;
162 /* look for tokens from names.h - this is expensive! */
163 if ((ms->flags & MAGIC_NO_CHECK_TOKENS) != 0)
164 goto subtype_identified;
172 /* skip past any leading space */
173 while (i < ulen && ISSPC(ubuf[i]))
178 /* find the next whitespace */
179 for (end = i + 1; end < nbytes; end++)
180 if (ISSPC(ubuf[end]))
183 /* compare the word thus isolated against the token list */
184 for (p = names; p < names + NNAMES; p++) {
185 if (ascmatch((const unsigned char *)p->name, ubuf + i,
189 else if (curtype != p->type) {
195 subtype = types[p->type].human;
196 subtype_mime = types[p->type].mime;
197 goto subtype_identified;
207 /* Now try to discover other details about the file. */
208 for (i = 0; i < ulen; i++) {
209 if (ubuf[i] == '\n') {
218 seen_cr = (ubuf[i] == '\r');
222 if (ubuf[i] == 0x85) { /* X3.64/ECMA-43 "next line" character */
227 /* If this line is _longer_ than MAXLINELEN, remember it. */
228 if (i > last_line_end + MAXLINELEN)
231 if (ubuf[i] == '\033')
237 /* Beware, if the data has been truncated, the final CR could have
238 been followed by a LF. If we have HOWMANY bytes, it indicates
239 that the data might have been truncated, probably even before
240 this function was called. */
241 if (seen_cr && nbytes < HOWMANY)
244 if (strcmp(type, "binary") == 0) {
249 if (!file_printedlen(ms) && (mime & MAGIC_MIME_TYPE) != 0) {
251 if (file_printf(ms, "%s", subtype_mime) == -1)
254 if (file_printf(ms, "text/plain") == -1)
259 if (file_printedlen(ms)) {
260 switch (file_replace(ms, " text$", ", ")) {
262 switch (file_replace(ms, " text executable$",
265 if (file_printf(ms, ", ") == -1)
281 if (file_printf(ms, "%s", code) == -1)
285 if (file_printf(ms, " %s", subtype) == -1)
289 if (file_printf(ms, " %s", type) == -1)
293 if (file_printf(ms, " executable") == -1)
297 if (file_printf(ms, ", with very long lines") == -1)
301 * Only report line terminators if we find one other than LF,
302 * or if we find none at all.
304 if ((n_crlf == 0 && n_cr == 0 && n_nel == 0 && n_lf == 0) ||
305 (n_crlf != 0 || n_cr != 0 || n_nel != 0)) {
306 if (file_printf(ms, ", with") == -1)
309 if (n_crlf == 0 && n_cr == 0 && n_nel == 0 && n_lf == 0) {
310 if (file_printf(ms, " no") == -1)
314 if (file_printf(ms, " CRLF") == -1)
316 if (n_cr || n_lf || n_nel)
317 if (file_printf(ms, ",") == -1)
321 if (file_printf(ms, " CR") == -1)
324 if (file_printf(ms, ",") == -1)
328 if (file_printf(ms, " LF") == -1)
331 if (file_printf(ms, ",") == -1)
335 if (file_printf(ms, " NEL") == -1)
339 if (file_printf(ms, " line terminators") == -1)
344 if (file_printf(ms, ", with escape sequences") == -1)
347 if (file_printf(ms, ", with overstriking") == -1)
359 ascmatch(const unsigned char *s, const unichar *us, size_t ulen)
363 for (i = 0; i < ulen; i++) {
375 * Encode Unicode string as UTF-8, returning pointer to character
376 * after end of string, or NULL if an invalid character is found.
378 private unsigned char *
379 encode_utf8(unsigned char *buf, size_t len, unichar *ubuf, size_t ulen)
382 unsigned char *end = buf + len;
384 for (i = 0; i < ulen; i++) {
385 if (ubuf[i] <= 0x7f) {
388 *buf++ = (unsigned char)ubuf[i];
389 } else if (ubuf[i] <= 0x7ff) {
392 *buf++ = (unsigned char)((ubuf[i] >> 6) + 0xc0);
393 *buf++ = (unsigned char)((ubuf[i] & 0x3f) + 0x80);
394 } else if (ubuf[i] <= 0xffff) {
397 *buf++ = (unsigned char)((ubuf[i] >> 12) + 0xe0);
398 *buf++ = (unsigned char)(((ubuf[i] >> 6) & 0x3f) + 0x80);
399 *buf++ = (unsigned char)((ubuf[i] & 0x3f) + 0x80);
400 } else if (ubuf[i] <= 0x1fffff) {
403 *buf++ = (unsigned char)((ubuf[i] >> 18) + 0xf0);
404 *buf++ = (unsigned char)(((ubuf[i] >> 12) & 0x3f) + 0x80);
405 *buf++ = (unsigned char)(((ubuf[i] >> 6) & 0x3f) + 0x80);
406 *buf++ = (unsigned char)((ubuf[i] & 0x3f) + 0x80);
407 } else if (ubuf[i] <= 0x3ffffff) {
410 *buf++ = (unsigned char)((ubuf[i] >> 24) + 0xf8);
411 *buf++ = (unsigned char)(((ubuf[i] >> 18) & 0x3f) + 0x80);
412 *buf++ = (unsigned char)(((ubuf[i] >> 12) & 0x3f) + 0x80);
413 *buf++ = (unsigned char)(((ubuf[i] >> 6) & 0x3f) + 0x80);
414 *buf++ = (unsigned char)((ubuf[i] & 0x3f) + 0x80);
415 } else if (ubuf[i] <= 0x7fffffff) {
418 *buf++ = (unsigned char)((ubuf[i] >> 30) + 0xfc);
419 *buf++ = (unsigned char)(((ubuf[i] >> 24) & 0x3f) + 0x80);
420 *buf++ = (unsigned char)(((ubuf[i] >> 18) & 0x3f) + 0x80);
421 *buf++ = (unsigned char)(((ubuf[i] >> 12) & 0x3f) + 0x80);
422 *buf++ = (unsigned char)(((ubuf[i] >> 6) & 0x3f) + 0x80);
423 *buf++ = (unsigned char)((ubuf[i] & 0x3f) + 0x80);
424 } else /* Invalid character */