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 1.107 2020/06/08 19:58:36 christos Exp $")
49 #define MAXLINELEN 300 /* longest sane line length */
50 #define ISSPC(x) ((x) == ' ' || (x) == '\t' || (x) == '\r' || (x) == '\n' \
51 || (x) == 0x85 || (x) == '\f')
53 private unsigned char *encode_utf8(unsigned char *, size_t, unichar *, size_t);
54 private size_t trim_nuls(const unsigned char *, size_t);
57 * Undo the NUL-termination kindly provided by process()
58 * but leave at least one byte to look at
61 trim_nuls(const unsigned char *buf, size_t nbytes)
63 while (nbytes > 1 && buf[nbytes - 1] == '\0')
70 file_ascmagic(struct magic_set *ms, const struct buffer *b, int text)
77 const char *code = NULL;
78 const char *code_mime = NULL;
79 const char *type = NULL;
82 bb.flen = trim_nuls(CAST(const unsigned char *, b->fbuf), b->flen);
84 * Avoid trimming at an odd byte if the original buffer was evenly
85 * sized; this avoids losing the last character on UTF-16 LE text
87 if ((bb.flen & 1) && !(b->flen & 1))
90 /* If file doesn't look like any sort of text, give up. */
91 if (file_encoding(ms, &bb, &ubuf, &ulen, &code, &code_mime,
95 rv = file_ascmagic_with_encoding(ms, &bb,
96 ubuf, ulen, code, type, text);
104 file_ascmagic_with_encoding(struct magic_set *ms,
105 const struct buffer *b, unichar *ubuf, size_t ulen, const char *code,
106 const char *type, int text)
109 const unsigned char *buf = CAST(const unsigned char *, b->fbuf);
110 size_t nbytes = b->flen;
111 unsigned char *utf8_buf = NULL, *utf8_end;
114 int mime = ms->flags & MAGIC_MIME;
115 int need_separator = 0;
117 const char *subtype = NULL;
120 int has_backspace = 0;
129 size_t last_line_end = CAST(size_t, -1);
130 int has_long_lines = 0;
132 nbytes = trim_nuls(buf, nbytes);
134 /* If we have fewer than 2 bytes, give up. */
140 if (ulen > 0 && (ms->flags & MAGIC_NO_CHECK_SOFT) == 0) {
141 /* Convert ubuf to UTF-8 and try text soft magic */
142 /* malloc size is a conservative overestimate; could be
143 improved, or at least realloced after conversion. */
145 if ((utf8_buf = CAST(unsigned char *, malloc(mlen))) == NULL) {
146 file_oomem(ms, mlen);
149 if ((utf8_end = encode_utf8(utf8_buf, mlen, ubuf, ulen))
152 buffer_init(&bb, b->fd, &b->st, utf8_buf,
153 CAST(size_t, utf8_end - utf8_buf));
155 if ((rv = file_softmagic(ms, &bb, NULL, NULL,
156 TEXTTEST, text)) == 0)
161 if ((ms->flags & (MAGIC_APPLE|MAGIC_EXTENSION))) {
162 rv = rv == -1 ? 0 : 1;
167 if ((ms->flags & (MAGIC_APPLE|MAGIC_EXTENSION))) {
172 /* Now try to discover other details about the file. */
173 for (i = 0; i < ulen; i++) {
174 if (ubuf[i] == '\n') {
183 seen_cr = (ubuf[i] == '\r');
187 if (ubuf[i] == 0x85) { /* X3.64/ECMA-43 "next line" character */
192 /* If this line is _longer_ than MAXLINELEN, remember it. */
193 if (i > last_line_end + MAXLINELEN)
196 if (ubuf[i] == '\033')
202 /* Beware, if the data has been truncated, the final CR could have
203 been followed by a LF. If we have ms->bytes_max bytes, it indicates
204 that the data might have been truncated, probably even before
205 this function was called. */
206 if (seen_cr && nbytes < ms->bytes_max)
209 if (strcmp(type, "binary") == 0) {
213 len = file_printedlen(ms);
215 if ((mime & MAGIC_MIME_TYPE) != 0) {
218 * Softmagic printed something, we
219 * are either done, or we need a separator
221 if ((ms->flags & MAGIC_CONTINUE) == 0) {
225 if (need_separator && file_separator(ms) == -1)
228 if (file_printf(ms, "text/plain") == -1)
234 switch (file_replace(ms, " text$", ", ")) {
236 switch (file_replace(ms, " text executable$",
239 if (file_printf(ms, ", ") == -1)
256 if (file_printf(ms, "%s", code) == -1)
260 if (file_printf(ms, " %s", subtype) == -1)
264 if (file_printf(ms, " %s", type) == -1)
268 if (file_printf(ms, " executable") == -1)
272 if (file_printf(ms, ", with very long lines") == -1)
276 * Only report line terminators if we find one other than LF,
277 * or if we find none at all.
279 if ((n_crlf == 0 && n_cr == 0 && n_nel == 0 && n_lf == 0) ||
280 (n_crlf != 0 || n_cr != 0 || n_nel != 0)) {
281 if (file_printf(ms, ", with") == -1)
284 if (n_crlf == 0 && n_cr == 0 && n_nel == 0 && n_lf == 0) {
285 if (file_printf(ms, " no") == -1)
289 if (file_printf(ms, " CRLF") == -1)
291 if (n_cr || n_lf || n_nel)
292 if (file_printf(ms, ",") == -1)
296 if (file_printf(ms, " CR") == -1)
299 if (file_printf(ms, ",") == -1)
303 if (file_printf(ms, " LF") == -1)
306 if (file_printf(ms, ",") == -1)
310 if (file_printf(ms, " NEL") == -1)
314 if (file_printf(ms, " line terminators") == -1)
319 if (file_printf(ms, ", with escape sequences") == -1)
322 if (file_printf(ms, ", with overstriking") == -1)
333 * Encode Unicode string as UTF-8, returning pointer to character
334 * after end of string, or NULL if an invalid character is found.
336 private unsigned char *
337 encode_utf8(unsigned char *buf, size_t len, unichar *ubuf, size_t ulen)
340 unsigned char *end = buf + len;
342 for (i = 0; i < ulen; i++) {
343 if (ubuf[i] <= 0x7f) {
346 *buf++ = CAST(unsigned char, ubuf[i]);
347 } else if (ubuf[i] <= 0x7ff) {
350 *buf++ = CAST(unsigned char, (ubuf[i] >> 6) + 0xc0);
351 *buf++ = CAST(unsigned char, (ubuf[i] & 0x3f) + 0x80);
352 } else if (ubuf[i] <= 0xffff) {
355 *buf++ = CAST(unsigned char, (ubuf[i] >> 12) + 0xe0);
356 *buf++ = CAST(unsigned char, ((ubuf[i] >> 6) & 0x3f) + 0x80);
357 *buf++ = CAST(unsigned char, (ubuf[i] & 0x3f) + 0x80);
358 } else if (ubuf[i] <= 0x1fffff) {
361 *buf++ = CAST(unsigned char, (ubuf[i] >> 18) + 0xf0);
362 *buf++ = CAST(unsigned char, ((ubuf[i] >> 12) & 0x3f) + 0x80);
363 *buf++ = CAST(unsigned char, ((ubuf[i] >> 6) & 0x3f) + 0x80);
364 *buf++ = CAST(unsigned char, (ubuf[i] & 0x3f) + 0x80);
365 } else if (ubuf[i] <= 0x3ffffff) {
368 *buf++ = CAST(unsigned char, (ubuf[i] >> 24) + 0xf8);
369 *buf++ = CAST(unsigned char, ((ubuf[i] >> 18) & 0x3f) + 0x80);
370 *buf++ = CAST(unsigned char, ((ubuf[i] >> 12) & 0x3f) + 0x80);
371 *buf++ = CAST(unsigned char, ((ubuf[i] >> 6) & 0x3f) + 0x80);
372 *buf++ = CAST(unsigned char, (ubuf[i] & 0x3f) + 0x80);
373 } else if (ubuf[i] <= 0x7fffffff) {
376 *buf++ = CAST(unsigned char, (ubuf[i] >> 30) + 0xfc);
377 *buf++ = CAST(unsigned char, ((ubuf[i] >> 24) & 0x3f) + 0x80);
378 *buf++ = CAST(unsigned char, ((ubuf[i] >> 18) & 0x3f) + 0x80);
379 *buf++ = CAST(unsigned char, ((ubuf[i] >> 12) & 0x3f) + 0x80);
380 *buf++ = CAST(unsigned char, ((ubuf[i] >> 6) & 0x3f) + 0x80);
381 *buf++ = CAST(unsigned char, (ubuf[i] & 0x3f) + 0x80);
382 } else /* Invalid character */