ascmagic.c

sleuthit-2.09 一个磁盘的工具集

/*
 * Copyright (c) Ian F. Darwin 1986-1995.
 * Software written by Ian F. Darwin and others;
 * maintained 1995-present by Christos Zoulas and others.
 * 
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice immediately at the beginning of the file, without modification,
 *    this list of conditions, and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 *  
 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE FOR
 * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 */
/*
 * ASCII magic -- file types that we know based on keywords
 * that can appear anywhere in the file.
 *
 * Extensively modified by Eric Fischer <[e-mail]> in July, 2000,
 * to handle character codes other than ASCII on a unified basis.
 *
 * Joerg Wunsch <[e-mail]> wrote the original support for 8-bit
 * international characters, now subsumed into this file.
 */

#include "file.h"
#include "magic.h"
#include <stdio.h>
#include <string.h>
#include <memory.h>
#include <ctype.h>
#include <stdlib.h>
#ifdef HAVE_UNISTD_H
#include <unistd.h>
#endif
#include "names.h"

#ifndef	lint
FILE_RCSID("@(#)$File: ascmagic.c,v 1.50 2007/03/15 14:51:00 christos Exp $")
#endif	/* lint */

typedef unsigned long unichar;

#define MAXLINELEN 300	/* longest sane line length */
#define ISSPC(x) ((x) == ' ' || (x) == '\t' || (x) == '\r' || (x) == '\n' \
		  || (x) == 0x85 || (x) == '\f')

private int looks_ascii(const unsigned char *, size_t, unichar *, size_t *);
private int looks_utf8(const unsigned char *, size_t, unichar *, size_t *);
private int looks_unicode(const unsigned char *, size_t, unichar *, size_t *);
private int looks_latin1(const unsigned char *, size_t, unichar *, size_t *);
private int looks_extended(const unsigned char *, size_t, unichar *, size_t *);
private void from_ebcdic(const unsigned char *, size_t, unsigned char *);
private int ascmatch(const unsigned char *, const unichar *, size_t);


protected int
file_ascmagic(struct magic_set *ms, const unsigned char *buf, size_t nbytes)
{
	size_t i;
	unsigned char *nbuf = NULL;
	unichar *ubuf = NULL;	
	size_t ulen;
	struct names *p;
	int rv = -1;

	const char *code = NULL;
	const char *code_mime = NULL;
	const char *type = NULL;
	const char *subtype = NULL;
	const char *subtype_mime = NULL;

	int has_escapes = 0;
	int has_backspace = 0;
	int seen_cr = 0;

	int n_crlf = 0;
	int n_lf = 0;
	int n_cr = 0;
	int n_nel = 0;

	size_t last_line_end = (size_t)-1;
	int has_long_lines = 0;

	/*
	 * Undo the NUL-termination kindly provided by process()
	 * but leave at least one byte to look at
	 */
	while (nbytes > 1 && buf[nbytes - 1] == '\0')
		nbytes--;

	if ((nbuf = calloc(1, (nbytes + 1) * sizeof(nbuf[0]))) == NULL)
		goto done;
	if ((ubuf = calloc(1, (nbytes + 1) * sizeof(ubuf[0]))) == NULL)
		goto done;

	/*
	 * Then try to determine whether it's any character code we can
	 * identify.  Each of these tests, if it succeeds, will leave
	 * the text converted into one-unichar-per-character Unicode in
	 * ubuf, and the number of characters converted in ulen.
	 */
	if (looks_ascii(buf, nbytes, ubuf, &ulen)) {
		code = "ASCII";
		code_mime = "us-ascii";
		type = "text";
	} else if (looks_utf8(buf, nbytes, ubuf, &ulen)) {
		code = "UTF-8 Unicode";
		code_mime = "utf-8";
		type = "text";
	} else if ((i = looks_unicode(buf, nbytes, ubuf, &ulen)) != 0) {
		if (i == 1)
			code = "Little-endian UTF-16 Unicode";
		else
			code = "Big-endian UTF-16 Unicode";

		type = "character data";
		code_mime = "utf-16";    /* is this defined? */
	} else if (looks_latin1(buf, nbytes, ubuf, &ulen)) {
		code = "ISO-8859";
		type = "text";
		code_mime = "iso-8859-1"; 
	} else if (looks_extended(buf, nbytes, ubuf, &ulen)) {
		code = "Non-ISO extended-ASCII";
		type = "text";
		code_mime = "unknown";
	} else {
		from_ebcdic(buf, nbytes, nbuf);

		if (looks_ascii(nbuf, nbytes, ubuf, &ulen)) {
			code = "EBCDIC";
			type = "character data";
			code_mime = "ebcdic";
		} else if (looks_latin1(nbuf, nbytes, ubuf, &ulen)) {
			code = "International EBCDIC";
			type = "character data";
			code_mime = "ebcdic";
		} else {
			rv = 0;
			goto done;  /* doesn't look like text at all */
		}
	}

	if (nbytes <= 1) {
		rv = 0;
		goto done;
	}

	/*
	 * for troff, look for . + letter + letter or .\";
	 * this must be done to disambiguate tar archives' ./file
	 * and other trash from real troff input.
	 *
	 * I believe Plan 9 troff allows non-ASCII characters in the names
	 * of macros, so this test might possibly fail on such a file.
	 */
	if ((ms->flags & MAGIC_NO_CHECK_TROFF) == 0 && *ubuf == '.') {
		unichar *tp = ubuf + 1;

		while (ISSPC(*tp))
			++tp;	/* skip leading whitespace */
		if ((tp[0] == '\\' && tp[1] == '\"') ||
		    (isascii((unsigned char)tp[0]) &&
		     isalnum((unsigned char)tp[0]) &&
		     isascii((unsigned char)tp[1]) &&
		     isalnum((unsigned char)tp[1]) &&
		     ISSPC(tp[2]))) {
			subtype_mime = "text/troff";
			subtype = "troff or preprocessor input";
			goto subtype_identified;
		}
	}

	if ((ms->flags & MAGIC_NO_CHECK_FORTRAN) == 0 &&
	    (*buf == 'c' || *buf == 'C') && ISSPC(buf[1])) {
		subtype_mime = "text/fortran";
		subtype = "fortran program";
		goto subtype_identified;
	}

	/* look for tokens from names.h - this is expensive! */

	if ((ms->flags & MAGIC_NO_CHECK_TOKENS) != 0)
		goto subtype_identified;

	i = 0;
	while (i < ulen) {
		size_t end;

		/*
		 * skip past any leading space
		 */
		while (i < ulen && ISSPC(ubuf[i]))
			i++;
		if (i >= ulen)
			break;

		/*
		 * find the next whitespace
		 */
		for (end = i + 1; end < nbytes; end++)
			if (ISSPC(ubuf[end]))
				break;

		/*
		 * compare the word thus isolated against the token list
		 */
		for (p = names; p < names + NNAMES; p++) {
			if (ascmatch((const unsigned char *)p->name, ubuf + i,
			    end - i)) {
				subtype = types[p->type].human;
				subtype_mime = types[p->type].mime;
				goto subtype_identified;
			}
		}

		i = end;
	}

subtype_identified:

	/*
	 * Now try to discover other details about the file.
	 */
	for (i = 0; i < ulen; i++) {
		if (ubuf[i] == '\n') {
			if (seen_cr)
				n_crlf++;
			else
				n_lf++;
			last_line_end = i;
		} else if (seen_cr)
			n_cr++;

		seen_cr = (ubuf[i] == '\r');
		if (seen_cr)
			last_line_end = i;

		if (ubuf[i] == 0x85) { /* X3.64/ECMA-43 "next line" character */
			n_nel++;
			last_line_end = i;
		}

		/* If this line is _longer_ than MAXLINELEN, remember it. */
		if (i > last_line_end + MAXLINELEN)
			has_long_lines = 1;

		if (ubuf[i] == '\033')
			has_escapes = 1;
		if (ubuf[i] == '\b')
			has_backspace = 1;
	}

	/* Beware, if the data has been truncated, the final CR could have
	   been followed by a LF.  If we have HOWMANY bytes, it indicates
	   that the data might have been truncated, probably even before
	   this function was called. */
	if (seen_cr && nbytes < HOWMANY)
		n_cr++;

	if ((ms->flags & MAGIC_MIME)) {
		if (subtype_mime) {
			if (file_printf(ms, subtype_mime) == -1)
				goto done;
		} else {
			if (file_printf(ms, "text/plain") == -1)
				goto done;
		}

		if (code_mime) {
			if (file_printf(ms, "; charset=") == -1)
				goto done;
			if (file_printf(ms, code_mime) == -1)
				goto done;
		}
	} else {
		if (file_printf(ms, code) == -1)
			goto done;

		if (subtype) {
			if (file_printf(ms, " ") == -1)
				goto done;
			if (file_printf(ms, subtype) == -1)
				goto done;
		}

		if (file_printf(ms, " ") == -1)
			goto done;
		if (file_printf(ms, type) == -1)
			goto done;

		if (has_long_lines)
			if (file_printf(ms, ", with very long lines") == -1)
				goto done;

		/*
		 * Only report line terminators if we find one other than LF,
		 * or if we find none at all.
		 */
		if ((n_crlf == 0 && n_cr == 0 && n_nel == 0 && n_lf == 0) ||
		    (n_crlf != 0 || n_cr != 0 || n_nel != 0)) {
			if (file_printf(ms, ", with") == -1)
				goto done;

			if (n_crlf == 0 && n_cr == 0 && n_nel == 0 && n_lf == 0)			{
				if (file_printf(ms, " no") == -1)
					goto done;
			} else {
				if (n_crlf) {
					if (file_printf(ms, " CRLF") == -1)
						goto done;
					if (n_cr || n_lf || n_nel)
						if (file_printf(ms, ",") == -1)
							goto done;
				}
				if (n_cr) {
					if (file_printf(ms, " CR") == -1)
						goto done;
					if (n_lf || n_nel)
						if (file_printf(ms, ",") == -1)
							goto done;
				}
				if (n_lf) {
					if (file_printf(ms, " LF") == -1)
						goto done;
					if (n_nel)
						if (file_printf(ms, ",") == -1)
							goto done;
				}
				if (n_nel)
					if (file_printf(ms, " NEL") == -1)
						goto done;
			}

			if (file_printf(ms, " line terminators") == -1)
				goto done;
		}

		if (has_escapes)
			if (file_printf(ms, ", with escape sequences") == -1)
				goto done;
		if (has_backspace)
			if (file_printf(ms, ", with overstriking") == -1)
				goto done;
	}