file update to 4.24

The '-i' switch is now enabled so file(1) can output mime type strings.

ok ian@
builk ports build test on amd64 by jasper@

ok ray@ gilles@ on a almost identical diff
builk ports build test on sparc64 on this almost identical diff by ajacoutot@
also tested by landry@

106 files changed, 3586 insertions, 1735 deletions

diff --git a/usr.bin/file/MAINT b/usr.bin/file/MAINT
index 5749949e828..a7079c4b41f 100644
--- a/usr.bin/file/MAINT
+++ b/usr.bin/file/MAINT
@@ -1,4 +1,4 @@
-$OpenBSD: MAINT,v 1.4 2008/05/08 01:40:56 chl Exp $
+$OpenBSD: MAINT,v 1.5 2009/04/24 18:54:34 chl Exp $
 
 Maintenance notes:
 
@@ -41,5 +41,4 @@ ascmagic.c			file_ascmagic()
 readelf.c		file_tryelf()
 				"unknown"
 
-Christos Zoulas
-christos@astron.com
+Christos Zoulas (see README for email address)
diff --git a/usr.bin/file/README b/usr.bin/file/README
index e26ac69101f..0fada4f9f4f 100644
--- a/usr.bin/file/README
+++ b/usr.bin/file/README
@@ -1,5 +1,10 @@
-@(#) $OpenBSD: README,v 1.4 2008/05/08 01:40:56 chl Exp $
 ** README for file(1) Command **
+@(#) $OpenBSD: README,v 1.5 2009/04/24 18:54:34 chl Exp $
+
+E-mail: christos@astron.com
+Mailing List: file@mx.gw.com
+
+Phone: Do not even think of telephoning me about this program. Send cash first!
 
 This is Release 4.x of Ian Darwin's (copyright but distributable)
 file(1) command. This version is the standard "file" command for Linux,
@@ -71,6 +76,10 @@ readelf.[ch] - Stand-alone elf parsing code.
 compress.c - on-the-fly decompression.
 print.c - print results, errors, warnings.
 
+You can download the latest version of file from:
+
+	ftp://ftp.astron.com/pub/file/
+
 If your gzip sometimes fails to decompress things complaining about a short
 file, apply this patch [which is going to be in the next version of gzip]:
 *** -   Tue Oct 29 02:06:35 1996
@@ -85,10 +94,6 @@ file, apply this patch [which is going to be in the next version of gzip]:
       }
       bytes_in += (ulg)insize;
 
-E-mail: christos@astron.com
-
-Phone: Do not even think of telephoning me about this program. Send cash first!
-
 Parts of this software were developed at SoftQuad Inc., developers
 of SGML/HTML/XML publishing software, in Toronto, Canada.
 SoftQuad was swallowed up by Corel in 2002
diff --git a/usr.bin/file/apprentice.c b/usr.bin/file/apprentice.c
index fb5ff000544..46cb5f81ec4 100644
--- a/usr.bin/file/apprentice.c
+++ b/usr.bin/file/apprentice.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: apprentice.c,v 1.25 2008/05/08 01:40:56 chl Exp $ */
+/*	$OpenBSD: apprentice.c,v 1.26 2009/04/24 18:54:34 chl Exp $ */
 /*
  * Copyright (c) Ian F. Darwin 1986-1995.
  * Software written by Ian F. Darwin and others;
@@ -32,6 +32,7 @@
 
 #include "file.h"
 #include "magic.h"
+#include "patchlevel.h"
 #include <stdlib.h>
 #ifdef HAVE_UNISTD_H
 #include <unistd.h>
@@ -45,9 +46,11 @@
 #ifdef QUICK
 #include <sys/mman.h>
 #endif
+#include <sys/types.h>
+#include <dirent.h>
 
 #ifndef	lint
-FILE_RCSID("@(#)$Id: apprentice.c,v 1.25 2008/05/08 01:40:56 chl Exp $")
+FILE_RCSID("@(#)$Id: apprentice.c,v 1.26 2009/04/24 18:54:34 chl Exp $")
 #endif	/* lint */
 
 #define	EATAB {while (isascii((unsigned char) *l) && \
@@ -93,28 +96,34 @@ private const char *getstr(struct magic_set *, const char *, char *, int,
     int *, int);
 private int parse(struct magic_set *, struct magic_entry **, uint32_t *,
     const char *, size_t, int);
+private int parse_mime(struct magic_set *, struct magic_entry **, uint32_t *,
+    const char *);
 private void eatsize(const char **);
 private int apprentice_1(struct magic_set *, const char *, int, struct mlist *);
 private size_t apprentice_magic_strength(const struct magic *);
 private int apprentice_sort(const void *, const void *);
-private int apprentice_file(struct magic_set *, struct magic **, uint32_t *,
+private int apprentice_load(struct magic_set *, struct magic **, uint32_t *,
     const char *, int);
 private void byteswap(struct magic *, uint32_t);
 private void bs1(struct magic *);
 private uint16_t swap2(uint16_t);
 private uint32_t swap4(uint32_t);
 private uint64_t swap8(uint64_t);
-private char *mkdbname(const char *, char *, size_t, int);
+private void mkdbname(const char *, char **, int);
 private int apprentice_map(struct magic_set *, struct magic **, uint32_t *,
     const char *);
 private int apprentice_compile(struct magic_set *, struct magic **, uint32_t *,
     const char *);
 private int check_format_type(const char *, int);
 private int check_format(struct magic_set *, struct magic *);
+private int get_op(char);
 
 private size_t maxmagic = 0;
 private size_t magicsize = sizeof(struct magic);
 
+private const char usg_hdr[] = "cont\toffset\ttype\topcode\tmask\tvalue\tdesc";
+private const char mime_marker[] = "!:mime";
+private const size_t mime_marker_len = sizeof(mime_marker) - 1;
 
 #ifdef COMPILE_ONLY
 
@@ -150,13 +159,13 @@ main(int argc, char *argv[])
 #endif /* COMPILE_ONLY */
 
 static const struct type_tbl_s {
-	const char *name;
+	const char name[16];
 	const size_t len;
 	const int type;
 	const int format;
 } type_tbl[] = {
 # define XX(s)		s, (sizeof(s) - 1)
-# define XX_NULL	NULL, 0
+# define XX_NULL	"", 0
 	{ XX("byte"),		FILE_BYTE,		FILE_FMT_NUM },
 	{ XX("short"),		FILE_SHORT,		FILE_FMT_NUM },
 	{ XX("default"),	FILE_DEFAULT,		FILE_FMT_STR },
@@ -189,6 +198,12 @@ static const struct type_tbl_s {
 	{ XX("qldate"),		FILE_QLDATE,		FILE_FMT_STR },
 	{ XX("leqldate"),	FILE_LEQLDATE,		FILE_FMT_STR },
 	{ XX("beqldate"),	FILE_BEQLDATE,		FILE_FMT_STR },
+	{ XX("float"),		FILE_FLOAT,		FILE_FMT_FLOAT },
+	{ XX("befloat"),	FILE_BEFLOAT,		FILE_FMT_FLOAT },
+	{ XX("lefloat"),	FILE_LEFLOAT,		FILE_FMT_FLOAT },
+	{ XX("double"),		FILE_DOUBLE,		FILE_FMT_DOUBLE },
+	{ XX("bedouble"),	FILE_BEDOUBLE,		FILE_FMT_DOUBLE },
+	{ XX("ledouble"),	FILE_LEDOUBLE,		FILE_FMT_DOUBLE },
 	{ XX_NULL,		FILE_INVALID,		FILE_FMT_NONE },
 # undef XX
 # undef XX_NULL
@@ -199,7 +214,7 @@ get_type(const char *l, const char **t)
 {
 	const struct type_tbl_s *p;
 
-	for (p = type_tbl; p->name; p++) {
+	for (p = type_tbl; p->len; p++) {
 		if (strncmp(l, p->name, p->len) == 0) {
 			if (t)
 				*t = l + p->len;
@@ -219,7 +234,7 @@ init_file_tables(void)
 		return;
 	done++;
 
-	for (p = type_tbl; p->name; p++) {
+	for (p = type_tbl; p->len; p++) {
 		assert(p->type < FILE_NAMES_SIZE);
 		file_names[p->type] = p->name;
 		file_formats[p->type] = p->format;
@@ -227,7 +242,7 @@ init_file_tables(void)
 }
 
 /*
- * Handle one file.
+ * Handle one file or directory.
  */
 private int
 apprentice_1(struct magic_set *ms, const char *fn, int action,
@@ -247,7 +262,7 @@ apprentice_1(struct magic_set *ms, const char *fn, int action,
 	}
 
 	if (action == FILE_COMPILE) {
-		rv = apprentice_file(ms, &magic, &nmagic, fn, action);
+		rv = apprentice_load(ms, &magic, &nmagic, fn, action);
 		if (rv != 0)
 			return -1;
 		rv = apprentice_compile(ms, &magic, &nmagic, fn);
@@ -259,14 +274,14 @@ apprentice_1(struct magic_set *ms, const char *fn, int action,
 	if ((rv = apprentice_map(ms, &magic, &nmagic, fn)) == -1) {
 		if (ms->flags & MAGIC_CHECK)
 			file_magwarn(ms, "using regular magic file `%s'", fn);
-		rv = apprentice_file(ms, &magic, &nmagic, fn, action);
+		rv = apprentice_load(ms, &magic, &nmagic, fn, action);
 		if (rv != 0)
 			return -1;
 	}
 
 	mapped = rv;
 	     
-	if (magic == NULL || nmagic == 0) {
+	if (magic == NULL) {
 		file_delmagic(magic, mapped, nmagic);
 		return -1;
 	}
@@ -313,14 +328,13 @@ file_delmagic(struct magic *p, int type, size_t entries)
 	}
 }
 
-/* const char *fn: list of magic files */
+/* const char *fn: list of magic files and directories */
 protected struct mlist *
 file_apprentice(struct magic_set *ms, const char *fn, int action)
 {
-	char *p, *mfn, *afn = NULL;
+	char *p, *mfn;
 	int file_err, errs = -1;
 	struct mlist *mlist;
-	static const char mime[] = ".mime";
 
 	init_file_tables();
 
@@ -348,25 +362,8 @@ file_apprentice(struct magic_set *ms, const char *fn, int action)
 			*p++ = '\0';
 		if (*fn == '\0')
 			break;
-		if (ms->flags & MAGIC_MIME) {
-			size_t len = strlen(fn) + sizeof(mime);
-			if ((afn = malloc(len)) == NULL) {
-				free(mfn);
-				free(mlist);
-				file_oomem(ms, len);
-				return NULL;
-			}
-			(void)strlcpy(afn, fn, len);
-			(void)strlcat(afn, ".mime", len);
-			fn = afn;
-		}
 		file_err = apprentice_1(ms, fn, action, mlist);
-		if (file_err > errs)
-			errs = file_err;
-		if (afn) {
-			free(afn);
-			afn = NULL;
-		}
+		errs = MAX(errs, file_err);
 		fn = p;
 	}
 	if (errs == -1) {
@@ -422,7 +419,7 @@ apprentice_magic_strength(const struct magic *m)
 
 	case FILE_SEARCH:
 	case FILE_REGEX:
-		val += m->vallen;
+		val += m->vallen * MAX(MULT / m->vallen, 1);
 		break;
 
 	case FILE_DATE:
@@ -433,6 +430,9 @@ apprentice_magic_strength(const struct magic *m)
 	case FILE_LELDATE:
 	case FILE_BELDATE:
 	case FILE_MELDATE:
+	case FILE_FLOAT:
+	case FILE_BEFLOAT:
+	case FILE_LEFLOAT:
 		val += 4 * MULT;
 		break;
 
@@ -445,6 +445,9 @@ apprentice_magic_strength(const struct magic *m)
 	case FILE_QLDATE:
 	case FILE_LEQLDATE:
 	case FILE_BEQLDATE:
+	case FILE_DOUBLE:
+	case FILE_BEDOUBLE:
+	case FILE_LEDOUBLE:
 		val += 8 * MULT;
 		break;
 
@@ -456,10 +459,10 @@ apprentice_magic_strength(const struct magic *m)
 
 	switch (m->reln) {
 	case 'x':	/* matches anything penalize */
+	case '!':       /* matches almost anything penalize */
 		val = 0;
 		break;
 
-	case '!':
 	case '=':	/* Exact match, prefer */
 		val += MULT;
 		break;
@@ -503,36 +506,131 @@ apprentice_sort(const void *a, const void *b)
 		return 1;
 }
 
+private void
+set_test_type(struct magic *mstart, struct magic *m)
+{
+	switch (m->type) {
+	case FILE_BYTE:
+	case FILE_SHORT:
+	case FILE_LONG:
+	case FILE_DATE:
+	case FILE_BESHORT:
+	case FILE_BELONG:
+	case FILE_BEDATE:
+	case FILE_LESHORT:
+	case FILE_LELONG:
+	case FILE_LEDATE:
+	case FILE_LDATE:
+	case FILE_BELDATE:
+	case FILE_LELDATE:
+	case FILE_MEDATE:
+	case FILE_MELDATE:
+	case FILE_MELONG:
+	case FILE_QUAD:
+	case FILE_LEQUAD:
+	case FILE_BEQUAD:
+	case FILE_QDATE:
+	case FILE_LEQDATE:
+	case FILE_BEQDATE:
+	case FILE_QLDATE:
+	case FILE_LEQLDATE:
+	case FILE_BEQLDATE:
+	case FILE_FLOAT:
+	case FILE_BEFLOAT:
+	case FILE_LEFLOAT:
+	case FILE_DOUBLE:
+	case FILE_BEDOUBLE:
+	case FILE_LEDOUBLE:
+	case FILE_STRING:
+	case FILE_PSTRING:
+	case FILE_BESTRING16:
+	case FILE_LESTRING16:
+		/* binary test, set flag */
+		mstart->flag |= BINTEST;
+		break;
+	case FILE_REGEX:
+	case FILE_SEARCH:
+		/* binary test if pattern is not text */
+		if (file_looks_utf8(m->value.s, m->vallen, NULL, NULL) == 0)
+			mstart->flag |= BINTEST;
+		break;
+	case FILE_DEFAULT:
+		/* can't deduce anything; we shouldn't see this at the
+		   top level anyway */
+		break;
+	case FILE_INVALID:
+	default:
+		/* invalid search type, but no need to complain here */
+		break;
+	}
+}
+
 /*
- * parse from a file
- * const char *fn: name of magic file
+ * Load and parse one file.
  */
-private int
-apprentice_file(struct magic_set *ms, struct magic **magicp, uint32_t *nmagicp,
-    const char *fn, int action)
+private void
+load_1(struct magic_set *ms, int action, const char *fn, int *errs,
+   struct magic_entry **marray, uint32_t *marraycount)
 {
-	private const char hdr[] =
-		"cont\toffset\ttype\topcode\tmask\tvalue\tdesc";
-	FILE *f;
 	char line[BUFSIZ];
-	int errs = 0;
-	struct magic_entry *marray;
-	uint32_t marraycount, i, mentrycount = 0;
 	size_t lineno = 0;
-
-	ms->flags |= MAGIC_CHECK;	/* Enable checks for parsed files */
-
-	f = fopen(ms->file = fn, "r");
+	FILE *f = fopen(ms->file = fn, "r");
 	if (f == NULL) {
 		if (errno != ENOENT)
 			file_error(ms, errno, "cannot read magic file `%s'",
-			    fn);
-		return -1;
+				   fn);
+		(*errs)++;
+	} else {
+		/* read and parse this file */
+		for (ms->line = 1; fgets(line, sizeof(line), f) != NULL; ms->line++) {
+			size_t len;
+			len = strlen(line);
+			if (len == 0) /* null line, garbage, etc */
+				continue;
+			if (line[len - 1] == '\n') {
+				lineno++;
+				line[len - 1] = '\0'; /* delete newline */
+			}
+			if (line[0] == '\0')	/* empty, do not parse */
+				continue;
+			if (line[0] == '#')	/* comment, do not parse */
+				continue;
+			if (len > mime_marker_len &&
+			    memcmp(line, mime_marker, mime_marker_len) == 0) {
+				/* MIME type */
+				if (parse_mime(ms, marray, marraycount,
+					       line + mime_marker_len) != 0)
+					(*errs)++;
+				continue;
+			}
+			if (parse(ms, marray, marraycount, line, lineno, action) != 0)
+				(*errs)++;
+		}
+
+		(void)fclose(f);
 	}
+}
+
+/*
+ * parse a file or directory of files
+ * const char *fn: name of magic file or directory
+ */
+private int
+apprentice_load(struct magic_set *ms, struct magic **magicp, uint32_t *nmagicp,
+    const char *fn, int action)
+{
+	int errs = 0;
+	struct magic_entry *marray;
+	uint32_t marraycount, i, mentrycount = 0, starttest;
+	char subfn[MAXPATHLEN];
+	struct stat st;
+	DIR *dir;
+	struct dirent *d;
+
+	ms->flags |= MAGIC_CHECK;	/* Enable checks for parsed files */
 
         maxmagic = MAXMAGIS;
 	if ((marray = calloc(maxmagic, sizeof(*marray))) == NULL) {
-		(void)fclose(f);
 		file_oomem(ms, maxmagic * sizeof(*marray));
 		return -1;
 	}
@@ -540,32 +638,63 @@ apprentice_file(struct magic_set *ms, struct magic **magicp, uint32_t *nmagicp,
 
 	/* print silly verbose header for USG compat. */
 	if (action == FILE_CHECK)
-		(void)fprintf(stderr, "%s\n", hdr);
+		(void)fprintf(stderr, "%s\n", usg_hdr);
+
+	/* load directory or file */
+	if (stat(fn, &st) == 0 && S_ISDIR(st.st_mode)) {
+		dir = opendir(fn);
+		if (dir) {
+			while (d = readdir(dir)) {
+				snprintf(subfn, sizeof(subfn), "%s/%s",
+				    fn, d->d_name);
+				if (stat(subfn, &st) == 0 && S_ISREG(st.st_mode)) {
+					load_1(ms, action, subfn, &errs,
+					    &marray, &marraycount);
+				}
+			}
+			closedir(dir);
+		} else
+			errs++;
+	} else
+		load_1(ms, action, fn, &errs, &marray, &marraycount);
+	if (errs)
+		goto out;
 
-	/* read and parse this file */
-	for (ms->line = 1; fgets(line, sizeof(line), f) != NULL; ms->line++) {
-		size_t len;
-		len = strlen(line);
-		if (len == 0) /* null line, garbage, etc */
+	/* Set types of tests */
+	for (i = 0; i < marraycount; ) {
+		if (marray[i].mp->cont_level != 0) {
+			i++;
 			continue;
-		if (line[len - 1] == '\n') {
-			lineno++;
-			line[len - 1] = '\0'; /* delete newline */
 		}
-		if (line[0] == '\0')	/* empty, do not parse */
-			continue;
-		if (line[0] == '#')	/* comment, do not parse */
-			continue;
-		if (parse(ms, &marray, &marraycount, line, lineno, action) != 0)
-			errs++;
-	}
 
-	(void)fclose(f);
-	if (errs)
-		goto out;
+		starttest = i;
+		do {
+			set_test_type(marray[starttest].mp, marray[i].mp);
+			if (ms->flags & MAGIC_DEBUG) {
+				(void)fprintf(stderr, "%s%s%s: %s\n",
+					marray[i].mp->mimetype,
+					marray[i].mp->mimetype[0] == '\0' ? "" : "; ",
+					marray[i].mp->desc[0] ? marray[i].mp->desc : "(no description)",
+					marray[i].mp->flag & BINTEST ? "binary" : "text");
+				if (marray[i].mp->flag & BINTEST) {
+#define SYMBOL "text"
+#define SYMLEN sizeof(SYMBOL)
+					char *p = strstr(marray[i].mp->desc, "text");
+					if (p && (p == marray[i].mp->desc || isspace(p[-1])) &&
+					    (p + SYMLEN - marray[i].mp->desc == MAXstring ||
+					     (p[SYMLEN] == '\0' || isspace(p[SYMLEN])))) {
+						(void)fprintf(stderr,
+							      "*** Possible binary test for text type\n");
+					}
+#undef SYMBOL
+#undef SYMLEN
+				}
+			}
+		} while (++i < marraycount && marray[i].mp->cont_level != 0);
+	}
 
-#ifndef NOORDER
 	qsort(marray, marraycount, sizeof(*marray), apprentice_sort);
+
 	/*
 	 * Make sure that any level 0 "default" line is last (if one exists).
 	 */
@@ -583,7 +712,6 @@ apprentice_file(struct magic_set *ms, struct magic **magicp, uint32_t *nmagicp,
 			break;					    
 		}
 	}
-#endif
 
 	for (i = 0; i < marraycount; i++)
 		mentrycount += marray[i].cont_count;
@@ -648,6 +776,9 @@ file_signextend(struct magic_set *ms, struct magic *m, uint64_t v)
 		case FILE_BELONG:
 		case FILE_LELONG:
 		case FILE_MELONG:
+		case FILE_FLOAT:
+		case FILE_BEFLOAT:
+		case FILE_LEFLOAT:
 			v = (int32_t) v;
 			break;
 		case FILE_QUAD:
@@ -659,6 +790,9 @@ file_signextend(struct magic_set *ms, struct magic *m, uint64_t v)
 		case FILE_BEQLDATE:
 		case FILE_LEQDATE:
 		case FILE_LEQLDATE:
+		case FILE_DOUBLE:
+		case FILE_BEDOUBLE:
+		case FILE_LEDOUBLE:
 			v = (int64_t) v;
 			break;
 		case FILE_STRING:
@@ -680,7 +814,7 @@ file_signextend(struct magic_set *ms, struct magic *m, uint64_t v)
 }
 
 private int
-string_modifier_check(struct magic_set *ms, struct magic const *m)
+string_modifier_check(struct magic_set *ms, struct magic *m)
 {
 	if ((ms->flags & MAGIC_CHECK) == 0)
 		return 0;
@@ -689,19 +823,28 @@ string_modifier_check(struct magic_set *ms, struct magic const *m)
 	case FILE_BESTRING16:
 	case FILE_LESTRING16:
 		if (m->str_flags != 0) {
-			file_magwarn(ms, "no modifiers allowed for 16-bit strings\n");
+			file_magwarn(ms,
+			    "no modifiers allowed for 16-bit strings\n");
 			return -1;
 		}
 		break;
 	case FILE_STRING:
 	case FILE_PSTRING:
 		if ((m->str_flags & REGEX_OFFSET_START) != 0) {
-			file_magwarn(ms, "'/%c' only allowed on regex and search\n",
+			file_magwarn(ms,
+			    "'/%c' only allowed on regex and search\n",
 			    CHAR_REGEX_OFFSET_START);
 			return -1;
 		}
 		break;
 	case FILE_SEARCH:
+		if (m->str_range == 0) {
+			file_magwarn(ms,
+			    "missing range; defaulting to %d\n",
+                            STRING_DEFAULT_RANGE);
+			m->str_range = STRING_DEFAULT_RANGE;
+			return -1;
+		}
 		break;
 	case FILE_REGEX:
 		if ((m->str_flags & STRING_COMPACT_BLANK) != 0) {
@@ -752,19 +895,19 @@ get_op(char c)
 private int
 get_cond(const char *l, const char **t)
 {
-	static struct cond_tbl_s {
-		const char *name;
-		const size_t len;
-		const int cond;
+	static const struct cond_tbl_s {
+		char name[8];
+		size_t len;
+		int cond;
 	} cond_tbl[] = {
 		{ "if",		2,	COND_IF },
 		{ "elif",	4,	COND_ELIF },
 		{ "else",	4,	COND_ELSE },
-		{ NULL, 	0,	COND_NONE },
+		{ "",		0,	COND_NONE },
 	};
-	struct cond_tbl_s *p;
+	const struct cond_tbl_s *p;
 
-	for (p = cond_tbl; p->name; p++) {
+	for (p = cond_tbl; p->len; p++) {
 		if (strncmp(l, p->name, p->len) == 0 &&
 		    isspace((unsigned char)l[p->len])) {
 			if (t)
@@ -961,6 +1104,16 @@ parse(struct magic_set *ms, struct magic_entry **mentryp, uint32_t *nmentryp,
 			case 'B':
 				m->in_type = FILE_BYTE;
 				break;
+			case 'e':
+			case 'f':
+			case 'g':
+				m->in_type = FILE_LEDOUBLE;
+				break;
+			case 'E':
+			case 'F':
+			case 'G':
+				m->in_type = FILE_BEDOUBLE;
+				break;
 			default:
 				if (ms->flags & MAGIC_CHECK)
 					file_magwarn(ms,
@@ -1031,7 +1184,7 @@ parse(struct magic_set *ms, struct magic_entry **mentryp, uint32_t *nmentryp,
 			file_magwarn(ms, "'~' invalid for string types");
 		++l;
 	}
-	m->str_count = 0;
+	m->str_range = 0;
 	m->str_flags = 0;
 	m->num_mask = 0;
 	if ((op = get_op(*l)) != -1) {
@@ -1045,22 +1198,24 @@ parse(struct magic_set *ms, struct magic_entry **mentryp, uint32_t *nmentryp,
 			eatsize(&l);
 		}
 		else if (op == FILE_OPDIVIDE) {
-			int have_count = 0;
+			int have_range = 0;
 			while (!isspace((unsigned char)*++l)) {
 				switch (*l) {
-				/* for portability avoid "case '0' ... '9':" */
 				case '0':  case '1':  case '2':
 				case '3':  case '4':  case '5':
 				case '6':  case '7':  case '8':
-				case '9': {
-					if (have_count && ms->flags & MAGIC_CHECK)
+				case '9':
+					if (have_range &&
+					    (ms->flags & MAGIC_CHECK))
 						file_magwarn(ms,
-						    "multiple counts");
-					have_count = 1;
-					m->str_count = strtoul(l, &t, 0);
+						    "multiple ranges");
+					have_range = 1;
+					m->str_range = strtoul(l, &t, 0);
+					if (m->str_range == 0)
+						file_magwarn(ms,
+						    "zero range");
 					l = t - 1;
 					break;
-				}
 				case CHAR_COMPACT_BLANK:
 					m->str_flags |= STRING_COMPACT_BLANK;
 					break;
@@ -1085,7 +1240,8 @@ parse(struct magic_set *ms, struct magic_entry **mentryp, uint32_t *nmentryp,
 					return -1;
 				}
 				/* allow multiple '/' for readability */
-				if (l[1] == '/' && !isspace((unsigned char)l[2]))
+				if (l[1] == '/' &&
+				    !isspace((unsigned char)l[2]))
 					l++;
 			}
 			if (string_modifier_check(ms, m) == -1)
@@ -1148,13 +1304,12 @@ parse(struct magic_set *ms, struct magic_entry **mentryp, uint32_t *nmentryp,
 	EATAB;
 	if (l[0] == '\b') {
 		++l;
-		m->nospflag = 1;
+		m->flag |= NOSPACE;
 	} else if ((l[0] == '\\') && (l[1] == 'b')) {
 		++l;
 		++l;
-		m->nospflag = 1;
-	} else
-		m->nospflag = 0;
+		m->flag |= NOSPACE;
+	}
 	for (i = 0; (m->desc[i++] = *l++) != '\0' && i < sizeof(m->desc); )
 		continue;
 	if (i == sizeof(m->desc)) {
@@ -1176,11 +1331,59 @@ parse(struct magic_set *ms, struct magic_entry **mentryp, uint32_t *nmentryp,
 		file_mdump(m);
 	}
 #endif
+	m->mimetype[0] = '\0';		/* initialise MIME type to none */
 	if (m->cont_level == 0)
 		++(*nmentryp);		/* make room for next */
 	return 0;
 }
 
+/*
+ * parse a MIME annotation line from magic file, put into magic[index - 1]
+ * if valid
+ */
+private int
+parse_mime(struct magic_set *ms, struct magic_entry **mentryp,
+    uint32_t *nmentryp, const char *line)
+{
+	size_t i;
+	const char *l = line;
+	struct magic *m;
+	struct magic_entry *me;
+
+	if (*nmentryp == 0) {
+		file_error(ms, 0, "No current entry for MIME type");
+		return -1;
+	}
+
+	me = &(*mentryp)[*nmentryp - 1];
+	m = &me->mp[me->cont_count == 0 ? 0 : me->cont_count - 1];
+
+	if (m->mimetype[0] != '\0') {
+		file_error(ms, 0, "Current entry already has a MIME type: %s\n"
+		    "Description: %s\nNew type: %s", m->mimetype, m->desc, l);
+		return -1;
+	}	
+
+	EATAB;
+	for (i = 0;
+	     *l && ((isascii((unsigned char)*l) && isalnum((unsigned char)*l))
+	     || strchr("-+/.", *l)) && i < sizeof(m->mimetype);
+	     m->mimetype[i++] = *l++)
+		continue;
+	if (i == sizeof(m->mimetype)) {
+		m->desc[sizeof(m->mimetype) - 1] = '\0';
+		if (ms->flags & MAGIC_CHECK)
+			file_magwarn(ms, "MIME type `%s' truncated %zu",
+			    m->mimetype, i);
+	} else
+		m->mimetype[i] = '\0';
+
+	if (i > 0)
+		return 0;
+	else
+		return -1;
+}
+
 private int
 check_format_type(const char *ptr, int type)
 {
@@ -1254,6 +1457,31 @@ check_format_type(const char *ptr, int type)
 			return -1;
 		}
 		
+	case FILE_FMT_FLOAT:
+	case FILE_FMT_DOUBLE:
+		if (*ptr == '-')
+			ptr++;
+		if (*ptr == '.')
+			ptr++;
+		while (isdigit((unsigned char)*ptr)) ptr++;
+		if (*ptr == '.')
+			ptr++;
+		while (isdigit((unsigned char)*ptr)) ptr++;
+	
+		switch (*ptr++) {
+		case 'e':
+		case 'E':
+		case 'f':
+		case 'F':
+		case 'g':
+		case 'G':
+			return 0;
+			
+		default:
+			return -1;
+		}
+		
+
 	case FILE_FMT_STR:
 		if (*ptr == '-')
 			ptr++;
@@ -1300,12 +1528,12 @@ check_format(struct magic_set *ms, struct magic *m)
 	assert(file_nformats == file_nnames);
 
 	if (m->type >= file_nformats) {
-		file_error(ms, 0, "Internal error inconsistency between "
+		file_magwarn(ms, "Internal error inconsistency between "
 		    "m->type and format strings");		
 		return -1;
 	}
 	if (file_formats[m->type] == FILE_FMT_NONE) {
-		file_error(ms, 0, "No format string for `%s' with description "
+		file_magwarn(ms, "No format string for `%s' with description "
 		    "`%s'", m->desc, file_names[m->type]);
 		return -1;
 	}
@@ -1316,8 +1544,8 @@ check_format(struct magic_set *ms, struct magic *m)
 		 * TODO: this error message is unhelpful if the format
 		 * string is not one character long
 		 */
-		file_error(ms, 0, "Printf format `%c' is not valid for type "
-		    " `%s' in description `%s'", 
+		file_magwarn(ms, "Printf format `%c' is not valid for type "
+		    "`%s' in description `%s'",
 		    ptr && *ptr ? *ptr : '?',
 		    file_names[m->type], m->desc);
 		return -1;
@@ -1325,7 +1553,7 @@ check_format(struct magic_set *ms, struct magic *m)
 	
 	for (; *ptr; ptr++) {
 		if (*ptr == '%') {
-			file_error(ms, 0,
+			file_magwarn(ms,
 			    "Too many format strings (should have at most one) "
 			    "for `%s' with description `%s'",
 			    file_names[m->type], m->desc);
@@ -1360,6 +1588,30 @@ getvalue(struct magic_set *ms, struct magic *m, const char **p, int action)
 			return -1;
 		}
 		m->vallen = slen;
+		if (m->type == FILE_PSTRING)
+			m->vallen++;
+		return 0;
+	case FILE_FLOAT:
+	case FILE_BEFLOAT:
+	case FILE_LEFLOAT:
+		if (m->reln != 'x') {
+			char *ep;
+#ifdef HAVE_STRTOF
+			m->value.f = strtof(*p, &ep);
+#else
+			m->value.f = (float)strtod(*p, &ep);
+#endif
+			*p = ep;
+		}
+		return 0;
+	case FILE_DOUBLE:
+	case FILE_BEDOUBLE:
+	case FILE_LEDOUBLE:
+		if (m->reln != 'x') {
+			char *ep;
+			m->value.d = strtod(*p, &ep);
+			*p = ep;
+		}
 		return 0;
 	default:
 		if (m->reln != 'x') {
@@ -1634,40 +1886,40 @@ apprentice_map(struct magic_set *ms, struct magic **magicp, uint32_t *nmagicp,
 	uint32_t *ptr;
 	uint32_t version;
 	int needsbyteswap;
-	char buf[MAXPATHLEN];
-	char *dbname = mkdbname(fn, buf, sizeof(buf), 0);
+	char *dbname = NULL;
 	void *mm = NULL;
 
+	mkdbname(fn, &dbname, 0);
 	if (dbname == NULL)
-		return -1;
+		goto error2;
 
 	if ((fd = open(dbname, O_RDONLY|O_BINARY)) == -1)
-		return -1;
+		goto error2;
 
 	if (fstat(fd, &st) == -1) {
 		file_error(ms, errno, "cannot stat `%s'", dbname);
-		goto error;
+		goto error1;
 	}
-	if (st.st_size < 16) {
+	if (st.st_size < 8) {
 		file_error(ms, 0, "file `%s' is too small", dbname);
-		goto error;
+		goto error1;
 	}
 
 #ifdef QUICK
 	if ((mm = mmap(0, (size_t)st.st_size, PROT_READ|PROT_WRITE,
 	    MAP_PRIVATE|MAP_FILE, fd, (off_t)0)) == MAP_FAILED) {
 		file_error(ms, errno, "cannot map `%s'", dbname);
-		goto error;
+		goto error1;
 	}
 #define RET	2
 #else
 	if ((mm = malloc((size_t)st.st_size)) == NULL) {
 		file_oomem(ms, (size_t)st.st_size);
-		goto error;
+		goto error1;
 	}
 	if (read(fd, mm, (size_t)st.st_size) != (size_t)st.st_size) {
 		file_badread(ms);
-		goto error;
+		goto error1;
 	}
 #define RET	1
 #endif
@@ -1678,7 +1930,7 @@ apprentice_map(struct magic_set *ms, struct magic **magicp, uint32_t *nmagicp,
 	if (*ptr != MAGICNO) {
 		if (swap4(*ptr) != MAGICNO) {
 			file_error(ms, 0, "bad magic in `%s'");
-			goto error;
+			goto error1;
 		}
 		needsbyteswap = 1;
 	} else
@@ -1688,17 +1940,21 @@ apprentice_map(struct magic_set *ms, struct magic **magicp, uint32_t *nmagicp,
 	else
 		version = ptr[1];
 	if (version != VERSIONNO) {
-		file_error(ms, 0, "version mismatch (%d != %d) in `%s'",
-		    version, VERSIONNO, dbname);
-		goto error;
-	}
-	*nmagicp = (uint32_t)(st.st_size / sizeof(struct magic)) - 1;
+		file_error(ms, 0, "File %d.%d supports only %d version magic "
+		    "files. `%s' is version %d", FILE_VERSION_MAJOR, patchlevel,
+		    VERSIONNO, dbname, version);
+		goto error1;
+	}
+	*nmagicp = (uint32_t)(st.st_size / sizeof(struct magic));
+	if (*nmagicp > 0)
+		(*nmagicp)--;
 	(*magicp)++;
 	if (needsbyteswap)
 		byteswap(*magicp, *nmagicp);
+	free(dbname);
 	return RET;
 
-error:
+error1:
 	if (fd != -1)
 		(void)close(fd);
 	if (mm) {
@@ -1711,6 +1967,8 @@ error:
 		*magicp = NULL;
 		*nmagicp = 0;
 	}
+error2:
+	free(dbname);
 	return -1;
 }
 
@@ -1725,44 +1983,49 @@ apprentice_compile(struct magic_set *ms, struct magic **magicp,
     uint32_t *nmagicp, const char *fn)
 {
 	int fd;
-	char buf[MAXPATHLEN];
-	char *dbname = mkdbname(fn, buf, sizeof(buf), 1);
+	char *dbname;
+	int rv = -1;
+
+	mkdbname(fn, &dbname, 1);
 
 	if (dbname == NULL) 
-		return -1;
+		goto out;
 
 	if ((fd = open(dbname, O_WRONLY|O_CREAT|O_TRUNC|O_BINARY, 0644)) == -1) {
 		file_error(ms, errno, "cannot open `%s'", dbname);
-		return -1;
+		goto out;
 	}
 
 	if (write(fd, ar, sizeof(ar)) != (ssize_t)sizeof(ar)) {
 		file_error(ms, errno, "error writing `%s'", dbname);
-		return -1;
+		goto out;
 	}
 
 	if (lseek(fd, (off_t)sizeof(struct magic), SEEK_SET)
 	    != sizeof(struct magic)) {
 		file_error(ms, errno, "error seeking `%s'", dbname);
-		return -1;
+		goto out;
 	}
 
 	if (write(fd, *magicp, (sizeof(struct magic) * *nmagicp)) 
 	    != (ssize_t)(sizeof(struct magic) * *nmagicp)) {
 		file_error(ms, errno, "error writing `%s'", dbname);
-		return -1;
+		goto out;
 	}
 
 	(void)close(fd);
-	return 0;
+	rv = 0;
+out:
+	free(dbname);
+	return rv;
 }
 
 private const char ext[] = ".mgc";
 /*
  * make a dbname
  */
-private char *
-mkdbname(const char *fn, char *buf, size_t bufsiz, int strip)
+private void
+mkdbname(const char *fn, char **buf, int strip)
 {
 	if (strip) {
 		const char *p;
@@ -1770,8 +2033,11 @@ mkdbname(const char *fn, char *buf, size_t bufsiz, int strip)
 			fn = ++p;
 	}
 
-	(void)snprintf(buf, bufsiz, "%s%s", fn, ext);
-	return buf;
+	(void)asprintf(buf, "%s%s", fn, ext);
+	if (*buf && strlen(*buf) > MAXPATHLEN) {
+		free(*buf);
+		*buf = NULL;
+	}
 }
 
 /*
@@ -1824,6 +2090,7 @@ swap8(uint64_t sv)
 	uint32_t rv;
 	uint8_t *s = (uint8_t *)(void *)&sv; 
 	uint8_t *d = (uint8_t *)(void *)&rv; 
+#if 0
 	d[0] = s[3];
 	d[1] = s[2];
 	d[2] = s[1];
@@ -1832,6 +2099,16 @@ swap8(uint64_t sv)
 	d[5] = s[6];
 	d[6] = s[5];
 	d[7] = s[4];
+#else
+	d[0] = s[7];
+	d[1] = s[6];
+	d[2] = s[5];
+	d[3] = s[4];
+	d[4] = s[3];
+	d[5] = s[2];
+	d[6] = s[1];
+	d[7] = s[0];
+#endif
 	return rv;
 }
 
@@ -1846,7 +2123,7 @@ bs1(struct magic *m)
 	m->in_offset = swap4((uint32_t)m->in_offset);
 	m->lineno = swap4((uint32_t)m->lineno);
 	if (IS_STRING(m->type)) {
-		m->str_count = swap4(m->str_count);
+		m->str_range = swap4(m->str_range);
 		m->str_flags = swap4(m->str_flags);
 	}
 	else {
diff --git a/usr.bin/file/ascmagic.c b/usr.bin/file/ascmagic.c
index b5cc8b991d6..7dcf0d3d99a 100644
--- a/usr.bin/file/ascmagic.c
+++ b/usr.bin/file/ascmagic.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: ascmagic.c,v 1.9 2008/05/08 01:40:56 chl Exp $ */
+/*	$OpenBSD: ascmagic.c,v 1.10 2009/04/24 18:54:34 chl Exp $ */
 /*
  * Copyright (c) Ian F. Darwin 1986-1995.
  * Software written by Ian F. Darwin and others;
@@ -50,33 +50,35 @@
 #include "names.h"
 
 #ifndef	lint
-FILE_RCSID("@(#)$Id: ascmagic.c,v 1.9 2008/05/08 01:40:56 chl Exp $")
+FILE_RCSID("@(#)$Id: ascmagic.c,v 1.10 2009/04/24 18:54:34 chl 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_utf8_with_BOM(const unsigned char *, size_t, unichar *,
+    size_t *);
+protected int file_looks_utf8(const unsigned char *, size_t, unichar *, size_t *);
+private int looks_ucs16(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);
+private unsigned char *encode_utf8(unsigned char *, size_t, 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;
+	unsigned char *nbuf = NULL, *utf8_buf = NULL, *utf8_end;
 	unichar *ubuf = NULL;	
-	size_t ulen;
-	struct names *p;
+	size_t ulen, mlen;
+	const struct names *p;
 	int rv = -1;
+	int mime = ms->flags & MAGIC_MIME;
 
 	const char *code = NULL;
 	const char *code_mime = NULL;
@@ -118,11 +120,15 @@ file_ascmagic(struct magic_set *ms, const unsigned char *buf, size_t nbytes)
 		code = "ASCII";
 		code_mime = "us-ascii";
 		type = "text";
-	} else if (looks_utf8(buf, nbytes, ubuf, &ulen)) {
+	} else if (looks_utf8_with_BOM(buf, nbytes, ubuf, &ulen) > 0) {
+		code = "UTF-8 Unicode (with BOM)";
+		code_mime = "utf-8";
+		type = "text";
+	} else if (file_looks_utf8(buf, nbytes, ubuf, &ulen) > 1) {
 		code = "UTF-8 Unicode";
 		code_mime = "utf-8";
 		type = "text";
-	} else if ((i = looks_unicode(buf, nbytes, ubuf, &ulen)) != 0) {
+	} else if ((i = looks_ucs16(buf, nbytes, ubuf, &ulen)) != 0) {
 		if (i == 1)
 			code = "Little-endian UTF-16 Unicode";
 		else
@@ -160,40 +166,25 @@ file_ascmagic(struct magic_set *ms, const unsigned char *buf, size_t nbytes)
 		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;
-		}
+	/* Convert ubuf to UTF-8 and try text soft magic */
+	/* If original was ASCII or UTF-8, could use nbuf instead of
+	   re-converting. */
+	/* malloc size is a conservative overestimate; could be
+	   re-converting improved, or at least realloced after
+	   re-converting conversion. */
+	mlen = ulen * 6;
+	if ((utf8_buf = malloc(mlen)) == NULL) {
+		file_oomem(ms, mlen);
+		goto done;
 	}
-
-	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;
+	if ((utf8_end = encode_utf8(utf8_buf, mlen, ubuf, ulen)) == NULL)
+		goto done;
+	if (file_softmagic(ms, utf8_buf, utf8_end - utf8_buf, TEXTTEST) != 0) {
+		rv = 1;
+		goto done;
 	}
 
 	/* look for tokens from names.h - this is expensive! */
-
 	if ((ms->flags & MAGIC_NO_CHECK_TOKENS) != 0)
 		goto subtype_identified;
 
@@ -201,24 +192,18 @@ file_ascmagic(struct magic_set *ms, const unsigned char *buf, size_t nbytes)
 	while (i < ulen) {
 		size_t end;
 
-		/*
-		 * skip past any leading space
-		 */
+		/* skip past any leading space */
 		while (i < ulen && ISSPC(ubuf[i]))
 			i++;
 		if (i >= ulen)
 			break;
 
-		/*
-		 * find the next whitespace
-		 */
+		/* 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
-		 */
+		/* 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)) {
@@ -233,9 +218,7 @@ file_ascmagic(struct magic_set *ms, const unsigned char *buf, size_t nbytes)
 
 subtype_identified:
 
-	/*
-	 * Now try to discover other details about the file.
-	 */
+	/* Now try to discover other details about the file. */
 	for (i = 0; i < ulen; i++) {
 		if (ubuf[i] == '\n') {
 			if (seen_cr)
@@ -272,21 +255,27 @@ subtype_identified:
 	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 (mime) {
+		if (mime & MAGIC_MIME_TYPE) {
+			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)
+		if ((mime == 0 || mime == MAGIC_MIME) && code_mime) {
+			if ((mime & MAGIC_MIME_TYPE) &&
+			    file_printf(ms, " charset=") == -1)
 				goto done;
 			if (file_printf(ms, code_mime) == -1)
 				goto done;
 		}
+
+		if (mime == MAGIC_MIME_ENCODING)
+			file_printf(ms, "binary");
 	} else {
 		if (file_printf(ms, code) == -1)
 			goto done;
@@ -363,6 +352,8 @@ done:
 		free(nbuf);
 	if (ubuf)
 		free(ubuf);
+	if (utf8_buf)
+		free(utf8_buf);
 
 	return rv;
 }
@@ -521,15 +512,84 @@ looks_extended(const unsigned char *buf, size_t nbytes, unichar *ubuf,
 	return 1;
 }
 
-private int
-looks_utf8(const unsigned char *buf, size_t nbytes, unichar *ubuf, size_t *ulen)
+/*
+ * Encode Unicode string as UTF-8, returning pointer to character
+ * after end of string, or NULL if an invalid character is found.
+ */
+private unsigned char *
+encode_utf8(unsigned char *buf, size_t len, unichar *ubuf, size_t ulen)
+{
+	size_t i;
+	unsigned char *end = buf + len;
+
+	for (i = 0; i < ulen; i++) {
+		if (ubuf[i] <= 0x7f) {
+			if (end - buf < 1)
+				return NULL;
+			*buf++ = (unsigned char)ubuf[i];
+		} else if (ubuf[i] <= 0x7ff) {
+			if (end - buf < 2)
+				return NULL;
+			*buf++ = (unsigned char)((ubuf[i] >> 6) + 0xc0);
+			*buf++ = (unsigned char)((ubuf[i] & 0x3f) + 0x80);
+		} else if (ubuf[i] <= 0xffff) {
+			if (end - buf < 3)
+				return NULL;
+			*buf++ = (unsigned char)((ubuf[i] >> 12) + 0xe0);
+			*buf++ = (unsigned char)(((ubuf[i] >> 6) & 0x3f) + 0x80);
+			*buf++ = (unsigned char)((ubuf[i] & 0x3f) + 0x80);
+		} else if (ubuf[i] <= 0x1fffff) {
+			if (end - buf < 4)
+				return NULL;
+			*buf++ = (unsigned char)((ubuf[i] >> 18) + 0xf0);
+			*buf++ = (unsigned char)(((ubuf[i] >> 12) & 0x3f) + 0x80);
+			*buf++ = (unsigned char)(((ubuf[i] >>  6) & 0x3f) + 0x80);
+			*buf++ = (unsigned char)((ubuf[i] & 0x3f) + 0x80);
+		} else if (ubuf[i] <= 0x3ffffff) {
+			if (end - buf < 5)
+				return NULL;
+			*buf++ = (unsigned char)((ubuf[i] >> 24) + 0xf8);
+			*buf++ = (unsigned char)(((ubuf[i] >> 18) & 0x3f) + 0x80);
+			*buf++ = (unsigned char)(((ubuf[i] >> 12) & 0x3f) + 0x80);
+			*buf++ = (unsigned char)(((ubuf[i] >>  6) & 0x3f) + 0x80);
+			*buf++ = (unsigned char)((ubuf[i] & 0x3f) + 0x80);
+		} else if (ubuf[i] <= 0x7fffffff) {
+			if (end - buf < 6)
+				return NULL;
+			*buf++ = (unsigned char)((ubuf[i] >> 30) + 0xfc);
+			*buf++ = (unsigned char)(((ubuf[i] >> 24) & 0x3f) + 0x80);
+			*buf++ = (unsigned char)(((ubuf[i] >> 18) & 0x3f) + 0x80);
+			*buf++ = (unsigned char)(((ubuf[i] >> 12) & 0x3f) + 0x80);
+			*buf++ = (unsigned char)(((ubuf[i] >>  6) & 0x3f) + 0x80);
+			*buf++ = (unsigned char)((ubuf[i] & 0x3f) + 0x80);
+		} else /* Invalid character */
+			return NULL;
+	}
+
+	return buf;
+}
+
+/*
+ * Decide whether some text looks like UTF-8. Returns:
+ *
+ *     -1: invalid UTF-8
+ *      0: uses odd control characters, so doesn't look like text
+ *      1: 7-bit text
+ *      2: definitely UTF-8 text (valid high-bit set bytes)
+ *
+ * If ubuf is non-NULL on entry, text is decoded into ubuf, *ulen;
+ * ubuf must be big enough!
+ */
+protected int
+file_looks_utf8(const unsigned char *buf, size_t nbytes, unichar *ubuf, size_t *ulen)
 {
 	size_t i;
 	int n;
 	unichar c;
-	int gotone = 0;
+	int gotone = 0, ctrl = 0;
 
-	*ulen = 0;
+	if (ubuf)
+		*ulen = 0;
 
 	for (i = 0; i < nbytes; i++) {
 		if ((buf[i] & 0x80) == 0) {	   /* 0xxxxxxx is plain ASCII */
@@ -539,11 +599,12 @@ looks_utf8(const unsigned char *buf, size_t nbytes, unichar *ubuf, size_t *ulen)
 			 */
 
 			if (text_chars[buf[i]] != T)
-				return 0;
+				ctrl = 1;
 
-			ubuf[(*ulen)++] = buf[i];
+			if (ubuf)
+				ubuf[(*ulen)++] = buf[i];
 		} else if ((buf[i] & 0x40) == 0) { /* 10xxxxxx never 1st byte */
-			return 0;
+			return -1;
 		} else {			   /* 11xxxxxx begins UTF-8 */
 			int following;
 
@@ -563,7 +624,7 @@ looks_utf8(const unsigned char *buf, size_t nbytes, unichar *ubuf, size_t *ulen)
 				c = buf[i] & 0x01;
 				following = 5;
 			} else
-				return 0;
+				return -1;
 
 			for (n = 0; n < following; n++) {
 				i++;
@@ -571,21 +632,37 @@ looks_utf8(const unsigned char *buf, size_t nbytes, unichar *ubuf, size_t *ulen)
 					goto done;
 
 				if ((buf[i] & 0x80) == 0 || (buf[i] & 0x40))
-					return 0;
+					return -1;
 
 				c = (c << 6) + (buf[i] & 0x3f);
 			}
 
-			ubuf[(*ulen)++] = c;
+			if (ubuf)
+				ubuf[(*ulen)++] = c;
 			gotone = 1;
 		}
 	}
 done:
-	return gotone;   /* don't claim it's UTF-8 if it's all 7-bit */
+	return ctrl ? 0 : (gotone ? 2 : 1);
+}
+
+/*
+ * Decide whether some text looks like UTF-8 with BOM. If there is no
+ * BOM, return -1; otherwise return the result of looks_utf8 on the
+ * rest of the text.
+ */
+private int
+looks_utf8_with_BOM(const unsigned char *buf, size_t nbytes, unichar *ubuf,
+    size_t *ulen)
+{
+	if (nbytes > 3 && buf[0] == 0xef && buf[1] == 0xbb && buf[2] == 0xbf)
+		return file_looks_utf8(buf + 3, nbytes - 3, ubuf, ulen);
+	else
+		return -1;
 }
 
 private int
-looks_unicode(const unsigned char *buf, size_t nbytes, unichar *ubuf,
+looks_ucs16(const unsigned char *buf, size_t nbytes, unichar *ubuf,
     size_t *ulen)
 {
 	int bigend;
diff --git a/usr.bin/file/compress.c b/usr.bin/file/compress.c
index a17a1d32861..6da92b83202 100644
--- a/usr.bin/file/compress.c
+++ b/usr.bin/file/compress.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: compress.c,v 1.12 2008/05/08 01:40:56 chl Exp $ */
+/*	$OpenBSD: compress.c,v 1.13 2009/04/24 18:54:34 chl Exp $ */
 /*
  * Copyright (c) Ian F. Darwin 1986-1995.
  * Software written by Ian F. Darwin and others;
@@ -50,19 +50,20 @@
 #if defined(HAVE_SYS_TIME_H)
 #include <sys/time.h>
 #endif
-#ifdef HAVE_LIBZ
+#if defined(HAVE_ZLIB_H) && defined(HAVE_LIBZ)
+#define BUILTIN_DECOMPRESS
 #include <zlib.h>
 #endif
 
 
 #ifndef lint
-FILE_RCSID("@(#)$Id: compress.c,v 1.12 2008/05/08 01:40:56 chl Exp $")
+FILE_RCSID("@(#)$Id: compress.c,v 1.13 2009/04/24 18:54:34 chl Exp $")
 #endif
 
-private struct {
-	const char *magic;
+private const struct {
+	const char magic[8];
 	size_t maglen;
-	const char *const argv[3];
+	const char *argv[3];
 	int silent;
 } compr[] = {
 	{ "\037\235", 2, { "gzip", "-cdq", NULL }, 1 },		/* compressed */
@@ -87,7 +88,7 @@ private size_t ncompr = sizeof(compr) / sizeof(compr[0]);
 private ssize_t swrite(int, const void *, size_t);
 private size_t uncompressbuf(struct magic_set *, int, size_t,
     const unsigned char *, unsigned char **, size_t);
-#ifdef HAVE_LIBZ
+#ifdef BUILTIN_DECOMPRESS
 private size_t uncompressgzipped(struct magic_set *, const unsigned char *,
     unsigned char **, size_t);
 #endif
@@ -99,6 +100,7 @@ file_zmagic(struct magic_set *ms, int fd, const char *name,
 	unsigned char *newbuf = NULL;
 	size_t i, nsz;
 	int rv = 0;
+	int mime = ms->flags & MAGIC_MIME;
 
 	if ((ms->flags & MAGIC_COMPRESS) == 0)
 		return 0;
@@ -113,11 +115,18 @@ file_zmagic(struct magic_set *ms, int fd, const char *name,
 			rv = -1;
 			if (file_buffer(ms, -1, name, newbuf, nsz) == -1)
 				goto error;
-			if (file_printf(ms, " (") == -1)
-				goto error;
-			if (file_buffer(ms, -1, NULL, buf, nbytes) == -1)
+
+			if (mime == MAGIC_MIME || mime == 0) {
+				if (file_printf(ms, mime ?
+				    " compressed-encoding=" : " (") == -1)
+					goto error;
+			}
+
+			if ((mime == 0 || mime & MAGIC_MIME_ENCODING) &&
+			    file_buffer(ms, -1, NULL, buf, nbytes) == -1)
 				goto error;
-			if (file_printf(ms, ")") == -1)
+
+			if (!mime && file_printf(ms, ")") == -1)
 				goto error;
 			rv = 1;
 			break;
@@ -286,7 +295,7 @@ file_pipe2file(struct magic_set *ms, int fd, const void *startbuf,
 	return fd;
 }
 
-#ifdef HAVE_LIBZ
+#ifdef BUILTIN_DECOMPRESS
 
 #define FHCRC		(1 << 1)
 #define FEXTRA		(1 << 2)
@@ -365,7 +374,8 @@ uncompressbuf(struct magic_set *ms, int fd, size_t method,
 	int fdin[2], fdout[2];
 	int r;
 
-#ifdef HAVE_LIBZ
+#ifdef BUILTIN_DECOMPRESS
+        /* FIXME: This doesn't cope with bzip2 */
 	if (method == 2)
 		return uncompressgzipped(ms, old, newch, n);
 #endif
diff --git a/usr.bin/file/config.h b/usr.bin/file/config.h
index 242650c2d31..68a861e5c08 100644
--- a/usr.bin/file/config.h
+++ b/usr.bin/file/config.h
@@ -4,7 +4,7 @@
  * but can still carefully import stuff from Christos' version.
  *
  * This file is in the public domain. Original Author Ian F. Darwin.
- * $OpenBSD: config.h,v 1.3 2008/05/08 01:40:56 chl Exp $
+ * $OpenBSD: config.h,v 1.4 2009/04/24 18:54:34 chl Exp $
  */
 
 /* header file issues. */
@@ -16,7 +16,8 @@
 #define	HAVE_INTTYPES_H 1
 #define HAVE_GETOPT_H 1
 #define HAVE_LIMITS_H 1
-/* #define	HAVE_LIBZ 1  DO NOT ENABLE YET -- ian */
+/* #define	HAVE_ZLIB_H	1	DO NOT ENABLE YET -- chl */
+/* #define	HAVE_LIBZ	1	DO NOT ENABLE YET -- ian */
 
 #define HAVE_STRTOUL
 #define HAVE_STRERROR
diff --git a/usr.bin/file/elfclass.h b/usr.bin/file/elfclass.h
new file mode 100644
index 00000000000..7f6d8adce6d
--- /dev/null
+++ b/usr.bin/file/elfclass.h
@@ -0,0 +1,67 @@
+/*
+ * Copyright (c) Christos Zoulas 2008.
+ * All Rights Reserved.
+ * 
+ * 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.
+ */
+	if (nbytes <= sizeof(elfhdr))
+		return 0;
+
+	u.l = 1;
+	(void)memcpy(&elfhdr, buf, sizeof elfhdr);
+	swap = (u.c[sizeof(int32_t) - 1] + 1) != elfhdr.e_ident[EI_DATA];
+
+	type = elf_getu16(swap, elfhdr.e_type);
+	switch (type) {
+#ifdef ELFCORE
+	case ET_CORE:
+		if (dophn_core(ms, class, swap, fd,
+		    (off_t)elf_getu(swap, elfhdr.e_phoff),
+		    elf_getu16(swap, elfhdr.e_phnum), 
+		    (size_t)elf_getu16(swap, elfhdr.e_phentsize),
+		    fsize, &flags) == -1)
+			return -1;
+		break;
+#endif
+	case ET_EXEC:
+	case ET_DYN:
+		if (dophn_exec(ms, class, swap, fd,
+		    (off_t)elf_getu(swap, elfhdr.e_phoff),
+		    elf_getu16(swap, elfhdr.e_phnum), 
+		    (size_t)elf_getu16(swap, elfhdr.e_phentsize),
+		    fsize, &flags) == -1)
+			return -1;
+		/*FALLTHROUGH*/
+	case ET_REL:
+		if (doshn(ms, class, swap, fd,
+		    (off_t)elf_getu(swap, elfhdr.e_shoff),
+		    elf_getu16(swap, elfhdr.e_shnum),
+		    (size_t)elf_getu16(swap, elfhdr.e_shentsize),
+		    &flags) == -1)
+			return -1;
+		break;
+
+	default:
+		break;
+	}
+	return 1;
diff --git a/usr.bin/file/file.c b/usr.bin/file/file.c
index 8c06ffa732f..b4d05763cbf 100644
--- a/usr.bin/file/file.c
+++ b/usr.bin/file/file.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: file.c,v 1.18 2008/05/08 01:40:56 chl Exp $ */
+/*	$OpenBSD: file.c,v 1.19 2009/04/24 18:54:34 chl Exp $ */
 /*
  * Copyright (c) Ian F. Darwin 1986-1995.
  * Software written by Ian F. Darwin and others;
@@ -61,10 +61,9 @@
 #include <wchar.h>
 #endif
 
-#ifdef HAVE_GETOPT_H
-#include <getopt.h>	/* for long options (is this portable?)*/
-#else
-#undef HAVE_GETOPT_LONG
+#include <getopt.h>
+#ifndef HAVE_GETOPT_LONG
+int getopt_long(int argc, char * const *argv, const char *optstring, const struct option *longopts, int *longindex);
 #endif
 
 #include <netinet/in.h>		/* for byte swapping */
@@ -72,7 +71,7 @@
 #include "patchlevel.h"
 
 #ifndef	lint
-FILE_RCSID("@(#)$Id: file.c,v 1.18 2008/05/08 01:40:56 chl Exp $")
+FILE_RCSID("@(#)$Id: file.c,v 1.19 2009/04/24 18:54:34 chl Exp $")
 #endif	/* lint */
 
 
@@ -82,7 +81,7 @@ FILE_RCSID("@(#)$Id: file.c,v 1.18 2008/05/08 01:40:56 chl Exp $")
 #define SYMLINKFLAG ""
 #endif
 
-# define USAGE  "Usage: %s [-bck" SYMLINKFLAG "nNrsvz0] [-e test] [-f namefile] [-F separator] [-m magicfiles] file...\n" \
+# define USAGE  "Usage: %s [-bcik" SYMLINKFLAG "nNrsvz0] [-e test] [-f namefile] [-F separator] [-m magicfiles] file...\n" \
 		" 	%s [-m magicfiles] -C\n"
 
 #ifndef MAXPATHLEN
@@ -99,18 +98,13 @@ private const char *magicfile = 0;	/* where the magic is	*/
 private const char *default_magicfile = MAGIC;
 private const char *separator = ":";	/* Default field separator	*/
 
+extern char *__progname;		/* used throughout 		*/
+
 private struct magic_set *magic;
-extern char *__progname;
 
 private void unwrap(char *);
 private void usage(void);
-#ifdef HAVE_GETOPT_LONG
 private void help(void);
-#endif
-#if 0
-private int byteconv4(int, int, int);
-private short byteconv2(int, int, int);
-#endif
 
 int main(int, char *[]);
 private void process(const char *, int);
@@ -123,44 +117,26 @@ private void load(const char *, int);
 int
 main(int argc, char *argv[])
 {
-	int c, i;
+	int c;
+	size_t i;
 	int action = 0, didsomefiles = 0, errflg = 0;
 	int flags = 0;
 	char *home, *usermagic;
 	struct stat sb;
 	static const char hmagic[] = "/.magic";
-#define OPTSTRING	"bcCde:f:F:hkLm:nNprsvz0"
-#ifdef HAVE_GETOPT_LONG
+#define OPTSTRING	"bcCde:f:F:hikLm:nNprsvz0"
 	int longindex;
 	static const struct option long_options[] =
 	{
-		{"version", 0, 0, 'v'},
-		{"help", 0, 0, 0},
-		{"brief", 0, 0, 'b'},
-		{"checking-printout", 0, 0, 'c'},
-		{"debug", 0, 0, 'd'},
-		{"exclude", 1, 0, 'e' },
-		{"files-from", 1, 0, 'f'},
-		{"separator", 1, 0, 'F'},
-		{"keep-going", 0, 0, 'k'},
-#ifdef S_IFLNK
-		{"dereference", 0, 0, 'L'},
-		{"no-dereference", 0, 0, 'h'},
-#endif
-		{"magic-file", 1, 0, 'm'},
-#if defined(HAVE_UTIME) || defined(HAVE_UTIMES)
-		{"preserve-date", 0, 0, 'p'},
-#endif
-		{"uncompress", 0, 0, 'z'},
-		{"raw", 0, 0, 'r'},
-		{"no-buffer", 0, 0, 'n'},
-		{"no-pad", 0, 0, 'N'},
-		{"special-files", 0, 0, 's'},
-		{"compile", 0, 0, 'C'},
-		{"print0", 0, 0, '0'},
-		{0, 0, 0, 0},
-	};
-#endif
+#define OPT(shortname, longname, opt, doc)      \
+    {longname, opt, NULL, shortname},
+#define OPT_LONGONLY(longname, opt, doc)        \
+    {longname, opt, NULL, 0},
+#include "file_opts.h"
+#undef OPT
+#undef OPT_LONGONLY
+    {0, 0, NULL, 0}
+};
 
 	static const struct {
 		const char *name;
@@ -170,17 +146,13 @@ main(int argc, char *argv[])
 		{ "ascii",	MAGIC_NO_CHECK_ASCII },
 		{ "compress",	MAGIC_NO_CHECK_COMPRESS },
 		{ "elf",	MAGIC_NO_CHECK_ELF },
-		{ "fortran",	MAGIC_NO_CHECK_FORTRAN },
 		{ "soft",	MAGIC_NO_CHECK_SOFT },
 		{ "tar",	MAGIC_NO_CHECK_TAR },
 		{ "tokens",	MAGIC_NO_CHECK_TOKENS },
-		{ "troff",	MAGIC_NO_CHECK_TROFF },
 	};
 
-#ifdef LC_CTYPE
 	/* makes islower etc work for other langs */
 	(void)setlocale(LC_CTYPE, "");
-#endif
 
 #ifdef __EMX__
 	/* sh-like wildcard expansion! Shouldn't hurt at least ... */
@@ -206,24 +178,27 @@ main(int argc, char *argv[])
 #ifdef S_IFLNK
 	flags |= getenv("POSIXLY_CORRECT") ? MAGIC_SYMLINK : 0;
 #endif
-#ifndef HAVE_GETOPT_LONG
-	while ((c = getopt(argc, argv, OPTSTRING)) != -1)
-#else
 	while ((c = getopt_long(argc, argv, OPTSTRING, long_options,
 	    &longindex)) != -1)
-#endif
 		switch (c) {
-#ifdef HAVE_GETOPT_LONG
 		case 0 :
-			if (longindex == 1)
+			switch (longindex) {
+			case 0:
 				help();
+				break;
+			case 10:
+				flags |= MAGIC_MIME_TYPE;
+				break;
+			case 11:
+				flags |= MAGIC_MIME_ENCODING;
+				break;
+			}
 			break;
-#endif
 		case '0':
 			nulsep = 1;
 			break;
 		case 'b':
-			++bflag;
+			bflag++;
 			break;
 		case 'c':
 			action = FILE_CHECK;
@@ -255,6 +230,9 @@ main(int argc, char *argv[])
 		case 'F':
 			separator = optarg;
 			break;
+		case 'i':
+			flags |= MAGIC_MIME;
+			break;
 		case 'k':
 			flags |= MAGIC_CONTINUE;
 			break;
@@ -279,9 +257,9 @@ main(int argc, char *argv[])
 			flags |= MAGIC_DEVICES;
 			break;
 		case 'v':
-			(void)fprintf(stdout, "%s-%d.%.2d\n", __progname,
+			(void)fprintf(stderr, "%s-%d.%.2d\n", __progname,
 				       FILE_VERSION_MAJOR, patchlevel);
-			(void)fprintf(stdout, "magic file from %s\n",
+			(void)fprintf(stderr, "magic file from %s\n",
 				       magicfile);
 			return 1;
 		case 'z':
@@ -333,18 +311,26 @@ main(int argc, char *argv[])
 		}
 	}
 	else {
-		int i, wid, nw;
-		for (wid = 0, i = optind; i < argc; i++) {
-			nw = file_mbswidth(argv[i]);
+		size_t j, wid, nw;
+		for (wid = 0, j = (size_t)optind; j < (size_t)argc; j++) {
+			nw = file_mbswidth(argv[j]);
 			if (nw > wid)
 				wid = nw;
 		}
+		/*
+		 * If bflag is only set twice, set it depending on
+		 * number of files [this is undocumented, and subject to change]
+		 */
+		if (bflag == 2) {
+			bflag = optind >= argc - 1;
+		}
 		for (; optind < argc; optind++)
 			process(argv[optind], wid);
 	}
 
+	c = magic->haderr ? 1 : 0;
 	magic_close(magic);
-	return 0;
+	return c;
 }
 
 
@@ -432,64 +418,6 @@ process(const char *inname, int wid)
 		(void)printf("%s\n", type);
 }
 
-
-#if 0
-/*
- * byteconv4
- * Input:
- *	from		4 byte quantity to convert
- *	same		whether to perform byte swapping
- *	big_endian	whether we are a big endian host
- */
-private int
-byteconv4(int from, int same, int big_endian)
-{
-	if (same)
-		return from;
-	else if (big_endian) {		/* lsb -> msb conversion on msb */
-		union {
-			int i;
-			char c[4];
-		} retval, tmpval;
-
-		tmpval.i = from;
-		retval.c[0] = tmpval.c[3];
-		retval.c[1] = tmpval.c[2];
-		retval.c[2] = tmpval.c[1];
-		retval.c[3] = tmpval.c[0];
-
-		return retval.i;
-	}
-	else
-		return ntohl(from);	/* msb -> lsb conversion on lsb */
-}
-
-/*
- * byteconv2
- * Same as byteconv4, but for shorts
- */
-private short
-byteconv2(int from, int same, int big_endian)
-{
-	if (same)
-		return from;
-	else if (big_endian) {		/* lsb -> msb conversion on msb */
-		union {
-			short s;
-			char c[2];
-		} retval, tmpval;
-
-		tmpval.s = (short) from;
-		retval.c[0] = tmpval.c[1];
-		retval.c[1] = tmpval.c[0];
-
-		return retval.s;
-	}
-	else
-		return ntohs(from);	/* msb -> lsb conversion on lsb */
-}
-#endif
-
 size_t
 file_mbswidth(const char *s)
 {
@@ -528,47 +456,23 @@ private void
 usage(void)
 {
 	(void)fprintf(stderr, USAGE, __progname, __progname);
-#ifdef HAVE_GETOPT_LONG
 	(void)fputs("Try `file --help' for more information.\n", stderr);
-#endif
 	exit(1);
 }
 
-#ifdef HAVE_GETOPT_LONG
 private void
 help(void)
 {
-	(void)puts(
-"Usage: file [OPTION]... [FILE]...\n"
-"Determine file type of FILEs.\n"
-"\n"
-"  -m, --magic-file LIST      use LIST as a colon-separated list of magic\n"
-"                               number files\n"
-"  -z, --uncompress           try to look inside compressed files\n"
-"  -b, --brief                do not prepend filenames to output lines\n"
-"  -c, --checking-printout    print the parsed form of the magic file, use in\n"
-"                               conjunction with -m to debug a new magic file\n"
-"                               before installing it\n"
-"  -e, --exclude              exclude test from the list of test to be\n"
-"                               performed for file. Valid tests are:\n"
-"                               ascii, apptype, elf, compress, soft, tar\n"
-"  -f, --files-from FILE      read the filenames to be examined from FILE\n"
-"  -F, --separator string     use string as separator instead of `:'\n"
-"  -k, --keep-going           don't stop at the first match\n"
-"  -L, --dereference          causes symlinks to be followed\n"
-"  -n, --no-buffer            do not buffer output\n"
-"  -N, --no-pad               do not pad output\n"
-"  -p, --preserve-date        preserve access times on files\n"
-"  -r, --raw                  don't translate unprintable chars to \\ooo\n"
-"  -s, --special-files        treat special (block/char devices) files as\n"
-"                             ordinary ones\n"
-"or\n"
-"      --help                 display this help and exit\n"
-"or\n"
-"      --version              output version information and exit\n"
-"or\n"
-"  -C, --compile              compile file specified by -m\n"
-);
+	(void)fputs(
+"Usage: file [OPTION...] [FILE...]\n"
+"Determine type of FILEs.\n"
+"\n", stderr);
+#define OPT(shortname, longname, opt, doc)      \
+        fprintf(stderr, "  -%c, --" longname doc, shortname);
+#define OPT_LONGONLY(longname, opt, doc)        \
+        fprintf(stderr, "      --" longname doc);
+#include "file_opts.h"
+#undef OPT
+#undef OPT_LONGONLY
 	exit(0);
 }
-#endif
diff --git a/usr.bin/file/file.h b/usr.bin/file/file.h
index d6022fc5887..13d7ece44dc 100644
--- a/usr.bin/file/file.h
+++ b/usr.bin/file/file.h
@@ -1,4 +1,4 @@
-/*	$OpenBSD: file.h,v 1.19 2008/05/08 01:40:56 chl Exp $ */
+/*	$OpenBSD: file.h,v 1.20 2009/04/24 18:54:34 chl Exp $ */
 /*
  * Copyright (c) Ian F. Darwin 1986-1995.
  * Software written by Ian F. Darwin and others;
@@ -28,7 +28,7 @@
  */
 /*
  * file.h - definitions for file(1) program
- * @(#)$Id: file.h,v 1.19 2008/05/08 01:40:56 chl Exp $
+ * @(#)$Id: file.h,v 1.20 2009/04/24 18:54:34 chl Exp $
  */
 
 #ifndef __file_h__
@@ -51,6 +51,7 @@
 #include <sys/types.h>
 /* Do this here and now, because struct stat gets re-defined on solaris */
 #include <sys/stat.h>
+#include <stdarg.h>
 
 #define ENABLE_CONDITIONALS
 
@@ -84,16 +85,21 @@
 #define	MIN(a,b)	(((a) < (b)) ? (a) : (b))
 #endif
 
+#ifndef MAX
+#define	MAX(a,b)	(((a) > (b)) ? (a) : (b))
+#endif
+
 #ifndef HOWMANY
 # define HOWMANY (256 * 1024)	/* how much of the file to look at */
 #endif
-#define MAXMAGIS 8192		/* max entries in /etc/magic */
-#define MAXDESC	64		/* max leng of text description */
+#define MAXMAGIS 8192		/* max entries in any one magic file
+				   or directory */
+#define MAXDESC	64		/* max leng of text description/MIME type */
 #define MAXstring 32		/* max leng of "string" types */
 
 #define MAGICNO		0xF11E041C
-#define VERSIONNO	4
-#define FILE_MAGICSIZE	(32 * 4)
+#define VERSIONNO	5
+#define FILE_MAGICSIZE	(32 * 6)
 
 #define	FILE_LOAD	0
 #define FILE_CHECK	1
@@ -102,18 +108,23 @@
 struct magic {
 	/* Word 1 */
 	uint16_t cont_level;	/* level of ">" */
-	uint8_t nospflag;	/* suppress space character */
 	uint8_t flag;
-#define INDIR		1	/* if '(...)' appears */
-#define OFFADD		2	/* if '>&' or '>...(&' appears */
-#define INDIROFFADD	4	/* if '>&(' appears */
-#define	UNSIGNED	8	/* comparison is unsigned */
+#define INDIR		0x01	/* if '(...)' appears */
+#define OFFADD		0x02	/* if '>&' or '>...(&' appears */
+#define INDIROFFADD	0x04	/* if '>&(' appears */
+#define UNSIGNED	0x08	/* comparison is unsigned */
+#define NOSPACE		0x10	/* suppress space character before output */
+#define BINTEST		0x20	/* test is for a binary type (set only
+                                   for top-level tests) */
+#define TEXTTEST	0	/* for passing to file_softmagic */
+
+	uint8_t dummy1;
 
 	/* Word 2 */
 	uint8_t reln;		/* relation (0=eq, '>'=gt, etc) */
 	uint8_t vallen;		/* length of string value, if any */
-	uint8_t type;		/* int, short, long or string. */
-	uint8_t in_type;	/* type of indirrection */
+	uint8_t type;		/* comparison type (FILE_*) */
+	uint8_t in_type;	/* type of indirection */
 #define 			FILE_INVALID	0
 #define 			FILE_BYTE	1
 #define				FILE_SHORT	2
@@ -147,7 +158,13 @@ struct magic {
 #define				FILE_QLDATE	30
 #define				FILE_LEQLDATE	31
 #define				FILE_BEQLDATE	32
-#define				FILE_NAMES_SIZE	33/* size of array to contain all names */
+#define				FILE_FLOAT	33
+#define				FILE_BEFLOAT	34
+#define				FILE_LEFLOAT	35
+#define				FILE_DOUBLE	36
+#define				FILE_BEDOUBLE	37
+#define				FILE_LEDOUBLE	38
+#define				FILE_NAMES_SIZE	39/* size of array to contain all names */
 
 #define IS_STRING(t) \
 	((t) == FILE_STRING || \
@@ -162,16 +179,18 @@ struct magic {
 #define FILE_FMT_NUM  1 /* "cduxXi" */
 #define FILE_FMT_STR  2 /* "s" */
 #define FILE_FMT_QUAD 3 /* "ll" */
+#define FILE_FMT_FLOAT 4 /* "eEfFgG" */
+#define FILE_FMT_DOUBLE 5 /* "eEfFgG" */
 
 	/* Word 3 */
 	uint8_t in_op;		/* operator for indirection */
 	uint8_t mask_op;	/* operator for mask */
 #ifdef ENABLE_CONDITIONALS
 	uint8_t cond;		/* conditional type */
-	uint8_t dummy1;	
+	uint8_t dummy2;	
 #else
-	uint8_t dummy1;	
 	uint8_t dummy2;	
+	uint8_t dummy3;	
 #endif
 
 #define				FILE_OPS	"&|^+-*/%"
@@ -212,7 +231,7 @@ struct magic {
 		} _s;		/* for use with string types */
 	} _u;
 #define num_mask _u._mask
-#define str_count _u._s._count
+#define str_range _u._s._count
 #define str_flags _u._s._flags
 
 	/* Words 9-16 */
@@ -225,9 +244,13 @@ struct magic {
 		uint8_t hl[4];	/* 4 bytes of a fixed-endian "long" */
 		uint8_t hq[8];	/* 8 bytes of a fixed-endian "quad" */
 		char s[MAXstring];	/* the search string or regex pattern */
+		float f;
+		double d;
 	} value;		/* either number or string */
 	/* Words 17..31 */
 	char desc[MAXDESC];	/* description */
+	/* Words 32..47 */
+	char mimetype[MAXDESC]; /* MIME type */
 };
 
 #define BIT(A)   (1 << (A))
@@ -242,6 +265,7 @@ struct magic {
 #define CHAR_IGNORE_UPPERCASE		'C'
 #define CHAR_REGEX_OFFSET_START		's'
 #define STRING_IGNORE_CASE		(STRING_IGNORE_LOWERCASE|STRING_IGNORE_UPPERCASE)
+#define STRING_DEFAULT_RANGE		100
 
 
 /* list of magic entries */
@@ -268,14 +292,8 @@ struct magic_set {
 		} *li;
 	} c;
 	struct out {
-		/* Accumulation buffer */
-		char *buf;
-		char *ptr;
-		size_t left;
-		size_t size;
-		/* Printable buffer */
-		char *pbuf;
-		size_t psize;
+		char *buf;		/* Accumulation buffer */
+		char *pbuf;		/* Printable buffer */
 	} o;
 	uint32_t offset;
 	int error;
@@ -292,9 +310,14 @@ struct magic_set {
 		size_t rm_len;		/* match length */
 	} search;
 
+	/* FIXME: Make the string dynamically allocated so that e.g.
+	   strings matched in files can be longer than MAXstring */
 	union VALUETYPE ms_value;	/* either number or string */
 };
 
+/* Type for Unicode characters */
+typedef unsigned long unichar;
+
 struct stat;
 protected const char *file_fmttime(uint32_t, int);
 protected int file_buffer(struct magic_set *, int, const char *, const void *,
@@ -309,7 +332,7 @@ protected int file_zmagic(struct magic_set *, int, const char *,
     const unsigned char *, size_t);
 protected int file_ascmagic(struct magic_set *, const unsigned char *, size_t);
 protected int file_is_tar(struct magic_set *, const unsigned char *, size_t);
-protected int file_softmagic(struct magic_set *, const unsigned char *, size_t);
+protected int file_softmagic(struct magic_set *, const unsigned char *, size_t, int);
 protected struct mlist *file_apprentice(struct magic_set *, const char *, int);
 protected uint64_t file_signextend(struct magic_set *, struct magic *,
     uint64_t);
@@ -326,6 +349,7 @@ protected size_t file_mbswidth(const char *);
 protected const char *file_getbuffer(struct magic_set *);
 protected ssize_t sread(int, void *, size_t, int);
 protected int file_check_mem(struct magic_set *, unsigned int);
+protected int file_looks_utf8(const unsigned char *, size_t, unichar *, size_t *);
 
 #ifndef COMPILE_ONLY
 extern const char *file_names[];
@@ -343,8 +367,11 @@ extern char *sys_errlist[];
 #define strtoul(a, b, c)	strtol(a, b, c)
 #endif
 
-#ifndef HAVE_SNPRINTF
-int snprintf(char *, size_t, const char *, ...);
+#ifndef HAVE_VASPRINTF
+int vasprintf(char **ptr, const char *format_string, va_list vargs);
+#endif
+#ifndef HAVE_ASPRINTF
+int asprintf(char **ptr, const char *format_string, ...);
 #endif
 
 #if defined(HAVE_MMAP) && defined(HAVE_SYS_MMAN_H) && !defined(QUICK)
@@ -355,6 +382,9 @@ int snprintf(char *, size_t, const char *, ...);
 #define O_BINARY	0
 #endif
 
+#ifdef __GNUC__
+static const char *rcsid(const char *) __attribute__((__used__));
+#endif
 #define FILE_RCSID(id) \
 static const char *rcsid(const char *p) { \
 	return rcsid(p = id); \
diff --git a/usr.bin/file/file_opts.h b/usr.bin/file/file_opts.h
new file mode 100644
index 00000000000..46bc08ae4b4
--- /dev/null
+++ b/usr.bin/file/file_opts.h
@@ -0,0 +1,48 @@
+/*
+ * Table of command-line options
+ *
+ * The first column specifies the short name, if any, or 0 if none.
+ * The second column specifies the long name.
+ * The third column specifies whether it takes a parameter.
+ * The fourth column is the documentation.
+ *
+ * N.B. The long options' order must correspond to the code in file.c,
+ * and OPTSTRING must be kept up-to-date with the short options.
+ * Pay particular attention to the numbers of long-only options in the
+ * switch statement!
+ */
+
+OPT_LONGONLY("help", 0, "                 display this help and exit\n")
+OPT('v', "version", 0, "              output version information and exit\n")
+OPT('m', "magic-file", 1, " LIST      use LIST as a colon-separated list of magic\n"
+    "                               number files\n")
+OPT('z', "uncompress", 0, "           try to look inside compressed files\n")
+OPT('b', "brief", 0, "                do not prepend filenames to output lines\n")
+OPT('c', "checking-printout", 0, "    print the parsed form of the magic file, use in\n"
+    "                               conjunction with -m to debug a new magic file\n"
+    "                               before installing it\n")
+OPT('e', "exclude", 1, " TEST         exclude TEST from the list of test to be\n"
+    "                               performed for file. Valid tests are:\n"
+    "                               ascii, apptype, compress, elf, soft, tar, tokens, troff\n")
+OPT('f', "files-from", 1, " FILE      read the filenames to be examined from FILE\n")
+OPT('F', "separator", 1, " STRING     use string as separator instead of `:'\n")
+OPT('i', "mime", 0, "                 output MIME type strings (--mime-type and\n"
+    "                               --mime-encoding)\n")
+OPT_LONGONLY("mime-type", 0, "            output the MIME type\n")
+OPT_LONGONLY("mime-encoding", 0, "        output the MIME encoding\n")
+OPT('k', "keep-going", 0, "           don't stop at the first match\n")
+#ifdef S_IFLNK
+OPT('L', "dereference", 0, "          follow symlinks (default)\n")
+OPT('h', "no-dereference", 0, "       don't follow symlinks\n")
+#endif
+OPT('n', "no-buffer", 0, "            do not buffer output\n")
+OPT('N', "no-pad", 0, "               do not pad output\n")
+OPT('0', "print0", 0, "               terminate filenames with ASCII NUL\n")
+#if defined(HAVE_UTIME) || defined(HAVE_UTIMES)
+OPT('p', "preserve-date", 0, "        preserve access times on files\n")
+#endif
+OPT('r', "raw", 0, "                  don't translate unprintable chars to \\ooo\n")
+OPT('s', "special-files", 0, "        treat special (block/char devices) files as\n"
+    "                             ordinary ones\n")
+OPT('C', "compile", 0, "              compile file specified by -m\n")
+OPT('d', "debug", 0, "                print debugging messages\n")
diff --git a/usr.bin/file/fsmagic.c b/usr.bin/file/fsmagic.c
index c62428b6b6a..74f1349d331 100644
--- a/usr.bin/file/fsmagic.c
+++ b/usr.bin/file/fsmagic.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: fsmagic.c,v 1.12 2008/05/08 01:40:56 chl Exp $ */
+/*	$OpenBSD: fsmagic.c,v 1.13 2009/04/24 18:54:34 chl Exp $ */
 /*
  * Copyright (c) Ian F. Darwin 1986-1995.
  * Software written by Ian F. Darwin and others;
@@ -58,13 +58,31 @@
 #undef HAVE_MAJOR
 
 #ifndef	lint
-FILE_RCSID("@(#)$Id: fsmagic.c,v 1.12 2008/05/08 01:40:56 chl Exp $")
+FILE_RCSID("@(#)$Id: fsmagic.c,v 1.13 2009/04/24 18:54:34 chl Exp $")
 #endif	/* lint */
 
+private int
+bad_link(struct magic_set *ms, int err, char *buf)
+{
+	char *errfmt;
+	if (err == ELOOP)
+		errfmt = "symbolic link in a loop";
+	else
+		errfmt = "broken symbolic link to `%s'";
+	if (ms->flags & MAGIC_ERROR) {
+		file_error(ms, err, errfmt, buf);
+		return -1;
+	} 
+	if (file_printf(ms, errfmt, buf) == -1)
+		return -1;
+	return 1;
+}
+
 protected int
 file_fsmagic(struct magic_set *ms, const char *fn, struct stat *sb)
 {
 	int ret = 0;
+	int mime = ms->flags & MAGIC_MIME;
 #ifdef	S_IFLNK
 	char buf[BUFSIZ+4];
 	int nch;
@@ -96,11 +114,12 @@ file_fsmagic(struct magic_set *ms, const char *fn, struct stat *sb)
 		return 1;
 	}
 
-	if ((ms->flags & MAGIC_MIME) != 0) {
+	if (mime) {
 		if ((sb->st_mode & S_IFMT) != S_IFREG) {
-			if (file_printf(ms, "application/x-not-regular-file")
+			if ((mime & MAGIC_MIME_TYPE) &&
+			    file_printf(ms, "application/x-not-regular-file")
 			    == -1)
-				return -1;
+				    return -1;
 			return 1;
 		}
 	}
@@ -136,7 +155,7 @@ file_fsmagic(struct magic_set *ms, const char *fn, struct stat *sb)
 		 */
 		if ((ms->flags & MAGIC_DEVICES) != 0)
 			break;
-#ifdef HAVE_ST_RDEV
+#ifdef HAVE_STAT_ST_RDEV
 # ifdef dv_unit
 		if (file_printf(ms, "character special (%d/%d/%d)",
 		    major(sb->st_rdev), dv_unit(sb->st_rdev),
@@ -162,7 +181,7 @@ file_fsmagic(struct magic_set *ms, const char *fn, struct stat *sb)
 		 */
 		if ((ms->flags & MAGIC_DEVICES) != 0)
 			break;
-#ifdef HAVE_ST_RDEV
+#ifdef HAVE_STAT_ST_RDEV
 # ifdef dv_unit
 		if (file_printf(ms, "block special (%d/%d/%d)",
 		    major(sb->st_rdev), dv_unit(sb->st_rdev),
@@ -208,23 +227,13 @@ file_fsmagic(struct magic_set *ms, const char *fn, struct stat *sb)
 				return -1;
 			return 1;
 		}
-		buf[nch] = '\0';	/* readlink(2) forgets this */
+		buf[nch] = '\0';	/* readlink(2) does not do this */
 
 		/* If broken symlink, say so and quit early. */
 		if (*buf == '/') {
-		    if (stat(buf, &tstatbuf) < 0) {
-			    if (ms->flags & MAGIC_ERROR) {
-				    file_error(ms, errno, 
-					"broken symbolic link to `%s'", buf);
-				    return -1;
-			    } 
-			    if (file_printf(ms, "broken symbolic link to `%s'",
-				buf) == -1)
-				    return -1;
-			    return 1;
-		    }
-		}
-		else {
+			if (stat(buf, &tstatbuf) < 0)
+				return bad_link(ms, errno, buf);
+		} else {
 			char *tmp;
 			char buf2[BUFSIZ+BUFSIZ+4];
 
@@ -247,18 +256,8 @@ file_fsmagic(struct magic_set *ms, const char *fn, struct stat *sb)
 				(void)strlcat(buf2, buf, sizeof buf2); /* plus (rel) link */
 				tmp = buf2;
 			}
-			if (stat(tmp, &tstatbuf) < 0) {
-				if (ms->flags & MAGIC_ERROR) {
-					file_error(ms, errno, 
-					    "broken symbolic link to `%s'",
-					    buf);
-					return -1;
-				}
-				if (file_printf(ms,
-				    "broken symbolic link to `%s'", buf) == -1)
-					return -1;
-				return 1;
-			}
+			if (stat(tmp, &tstatbuf) < 0)
+				return bad_link(ms, errno, buf);
 		}
 
 		/* Otherwise, handle it. */
@@ -304,8 +303,9 @@ file_fsmagic(struct magic_set *ms, const char *fn, struct stat *sb)
 	 * when we read the file.)
 	 */
 	if ((ms->flags & MAGIC_DEVICES) == 0 && sb->st_size == 0) {
-		if (file_printf(ms, (ms->flags & MAGIC_MIME) ?
-		    "application/x-empty" : "empty") == -1)
+		if ((!mime || (mime & MAGIC_MIME_TYPE)) &&
+		    file_printf(ms, mime ? "application/x-empty" :
+		    "empty") == -1)
 			return -1;
 		return 1;
 	}
diff --git a/usr.bin/file/funcs.c b/usr.bin/file/funcs.c
index 7b8fdb182fa..2b4f34be86c 100644
--- a/usr.bin/file/funcs.c
+++ b/usr.bin/file/funcs.c
@@ -1,4 +1,4 @@
-/* $OpenBSD: funcs.c,v 1.5 2008/05/08 01:40:56 chl Exp $ */
+/* $OpenBSD: funcs.c,v 1.6 2009/04/24 18:54:34 chl Exp $ */
 /*
  * Copyright (c) Christos Zoulas 2003.
  * All Rights Reserved.
@@ -37,67 +37,40 @@
 #if defined(HAVE_WCTYPE_H)
 #include <wctype.h>
 #endif
-#if defined(HAVE_LIMITS_H)
-#include <limits.h>
-#endif
-#ifndef SIZE_T_MAX
-#ifdef __LP64__
-#define SIZE_T_MAX (size_t)0xfffffffffffffffffU
-#else
-#define SIZE_T_MAX (size_t)0xffffffffU
-#endif
-#endif
 
 #ifndef	lint
-FILE_RCSID("@(#)$Id: funcs.c,v 1.5 2008/05/08 01:40:56 chl Exp $")
+FILE_RCSID("@(#)$Id: funcs.c,v 1.6 2009/04/24 18:54:34 chl Exp $")
 #endif	/* lint */
 
-#ifndef HAVE_VSNPRINTF
-int vsnprintf(char *, size_t, const char *, va_list);
-#endif
-
 /*
- * Like printf, only we print to a buffer and advance it.
+ * Like printf, only we append to a buffer.
  */
 protected int
 file_printf(struct magic_set *ms, const char *fmt, ...)
 {
 	va_list ap;
-	size_t len, size;
-	char *buf;
+	int len;
+	char *buf, *newstr;
 
 	va_start(ap, fmt);
+	len = vasprintf(&buf, fmt, ap);
+	if (len < 0)
+		goto out;
+	va_end(ap);
 
-	len = vsnprintf(ms->o.ptr, ms->o.left, fmt, ap);
-	if (len == -1) {
-		file_error(ms, errno, "vsnprintf failed");
-		return -1;
-	} else if (len >= ms->o.left) {
-		long diff;	/* XXX: really ptrdiff_t */
-
-		va_end(ap);
-		size = (ms->o.size - ms->o.left) + len + 1024;
-		if ((buf = realloc(ms->o.buf, size)) == NULL) {
-			file_oomem(ms, size);
-			return -1;
-		}
-		diff = ms->o.ptr - ms->o.buf;
-		ms->o.ptr = buf + diff;
-		ms->o.buf = buf;
-		ms->o.left = size - diff;
-		ms->o.size = size;
-
-		va_start(ap, fmt);
-		len = vsnprintf(ms->o.ptr, ms->o.left, fmt, ap);
-		if (len == -1) {
-			file_error(ms, errno, "vsnprintf failed");
-			return -1;
-		}
+	if (ms->o.buf != NULL) {
+		len = asprintf(&newstr, "%s%s", ms->o.buf, buf);
+		free(buf);
+		if (len < 0)
+			goto out;
+		free(ms->o.buf);
+		buf = newstr;
 	}
-	va_end(ap);
-	ms->o.ptr += len;
-	ms->o.left -= len;
+	ms->o.buf = buf;
 	return 0;
+out:
+	file_error(ms, errno, "vasprintf failed");
+	return -1;
 }
 
 /*
@@ -108,21 +81,17 @@ private void
 file_error_core(struct magic_set *ms, int error, const char *f, va_list va,
     uint32_t lineno)
 {
-	size_t len;
 	/* Only the first error is ok */
 	if (ms->haderr)
 		return;
-	len = 0;
 	if (lineno != 0) {
-		(void)snprintf(ms->o.buf, ms->o.size, "line %u: ", lineno);
-		len = strlen(ms->o.buf);
-	}
-	(void)vsnprintf(ms->o.buf + len, ms->o.size - len, f, va);
-	if (error > 0) {
-		len = strlen(ms->o.buf);
-		(void)snprintf(ms->o.buf + len, ms->o.size - len, " (%s)",
-		    strerror(error));
+		free(ms->o.buf);
+		ms->o.buf = NULL;
+		file_printf(ms, "line %u: ", lineno);
 	}
+        file_printf(ms, f, va);
+	if (error > 0)
+		file_printf(ms, " (%s)", strerror(error));
 	ms->haderr++;
 	ms->error = error;
 }
@@ -173,59 +142,73 @@ protected int
 file_buffer(struct magic_set *ms, int fd, const char *inname, const void *buf,
     size_t nb)
 {
-    int m;
+	int m;
+	int mime = ms->flags & MAGIC_MIME;
+
+	if (nb == 0) {
+		if ((!mime || (mime & MAGIC_MIME_TYPE)) &&
+		    file_printf(ms, mime ? "application/x-empty" :
+		    "empty") == -1)
+			return -1;
+		return 1;
+	} else if (nb == 1) {
+		if ((!mime || (mime & MAGIC_MIME_TYPE)) &&
+		    file_printf(ms, mime ? "application/octet-stream" :
+		    "very short file (no magic)") == -1)
+			return -1;
+		return 1;
+	}
 
 #ifdef __EMX__
-    if ((ms->flags & MAGIC_NO_CHECK_APPTYPE) == 0 && inname) {
-	switch (file_os2_apptype(ms, inname, buf, nb)) {
-	case -1:
-	    return -1;
-	case 0:
-	    break;
-	default:
-	    return 1;
+	if ((ms->flags & MAGIC_NO_CHECK_APPTYPE) == 0 && inname) {
+		switch (file_os2_apptype(ms, inname, buf, nb)) {
+		case -1:
+			return -1;
+		case 0:
+			break;
+		default:
+			return 1;
+		}
 	}
-    }
 #endif
 
-    /* try compression stuff */
-    if ((ms->flags & MAGIC_NO_CHECK_COMPRESS) != 0 ||
-        (m = file_zmagic(ms, fd, inname, buf, nb)) == 0) {
-	/* Check if we have a tar file */
-	if ((ms->flags & MAGIC_NO_CHECK_TAR) != 0 ||
-	    (m = file_is_tar(ms, buf, nb)) == 0) {
-	    /* try tests in /etc/magic (or surrogate magic file) */
-	    if ((ms->flags & MAGIC_NO_CHECK_SOFT) != 0 ||
-		(m = file_softmagic(ms, buf, nb)) == 0) {
-		/* try known keywords, check whether it is ASCII */
-		if ((ms->flags & MAGIC_NO_CHECK_ASCII) != 0 ||
-		    (m = file_ascmagic(ms, buf, nb)) == 0) {
-		    /* abandon hope, all ye who remain here */
-		    if (file_printf(ms, ms->flags & MAGIC_MIME ?
-			(nb ? "application/octet-stream" :
-			    "application/empty") :
-			(nb ? "data" :
-			    "empty")) == -1)
-			    return -1;
-		    m = 1;
+	/* try compression stuff */
+	if ((ms->flags & MAGIC_NO_CHECK_COMPRESS) != 0 ||
+	    (m = file_zmagic(ms, fd, inname, buf, nb)) == 0) {
+	    /* Check if we have a tar file */
+	    if ((ms->flags & MAGIC_NO_CHECK_TAR) != 0 ||
+		(m = file_is_tar(ms, buf, nb)) == 0) {
+		/* try tests in /etc/magic (or surrogate magic file) */
+		if ((ms->flags & MAGIC_NO_CHECK_SOFT) != 0 ||
+		    (m = file_softmagic(ms, buf, nb, BINTEST)) == 0) {
+		    /* try known keywords, check whether it is ASCII */
+		    if ((ms->flags & MAGIC_NO_CHECK_ASCII) != 0 ||
+			(m = file_ascmagic(ms, buf, nb)) == 0) {
+			/* abandon hope, all ye who remain here */
+			if ((!mime || (mime & MAGIC_MIME_TYPE)) &&
+			    file_printf(ms, mime ? "application/octet-stream" :
+				"data") == -1)
+				return -1;
+			m = 1;
+		    }
 		}
 	    }
 	}
-    }
 #ifdef BUILTIN_ELF
-    if ((ms->flags & MAGIC_NO_CHECK_ELF) == 0 && m == 1 && nb > 5 && fd != -1) {
-	/*
-	 * We matched something in the file, so this *might*
-	 * be an ELF file, and the file is at least 5 bytes
-	 * long, so if it's an ELF file it has at least one
-	 * byte past the ELF magic number - try extracting
-	 * information from the ELF headers that cannot easily
-	 * be extracted with rules in the magic file.
-	 */
-	(void)file_tryelf(ms, fd, buf, nb);
-    }
+	if ((ms->flags & MAGIC_NO_CHECK_ELF) == 0 && m == 1 &&
+	    nb > 5 && fd != -1) {
+		/*
+		 * We matched something in the file, so this *might*
+		 * be an ELF file, and the file is at least 5 bytes
+		 * long, so if it's an ELF file it has at least one
+		 * byte past the ELF magic number - try extracting
+		 * information from the ELF headers that cannot easily
+		 * be extracted with rules in the magic file.
+		 */
+		(void)file_tryelf(ms, fd, buf, nb);
+	}
 #endif
-    return m;
+	return m;
 }
 #endif
 
@@ -236,8 +219,7 @@ file_reset(struct magic_set *ms)
 		file_error(ms, 0, "no magic files loaded");
 		return -1;
 	}
-	ms->o.ptr = ms->o.buf;
-	ms->o.left = ms->o.size;
+	ms->o.buf = NULL;
 	ms->haderr = 0;
 	ms->error = -1;
 	return 0;
@@ -263,21 +245,18 @@ file_getbuffer(struct magic_set *ms)
 	if (ms->flags & MAGIC_RAW)
 		return ms->o.buf;
 
-	len = ms->o.size - ms->o.left;
 	/* * 4 is for octal representation, + 1 is for NUL */
-	if (len > (SIZE_T_MAX - 1) / 4) {
+	len = strlen(ms->o.buf);
+	if (len > (SIZE_MAX - 1) / 4) {
 		file_oomem(ms, len);
 		return NULL;
 	}
 	psize = len * 4 + 1;
-	if (ms->o.psize < psize) {
-		if ((pbuf = realloc(ms->o.pbuf, psize)) == NULL) {
-			file_oomem(ms, psize);
-			return NULL;
-		}
-		ms->o.psize = psize;
-		ms->o.pbuf = pbuf;
+	if ((pbuf = realloc(ms->o.pbuf, psize)) == NULL) {
+		file_oomem(ms, psize);
+		return NULL;
 	}
+	ms->o.pbuf = pbuf;
 
 #if defined(HAVE_WCHAR_H) && defined(HAVE_MBRTOWC) && defined(HAVE_WCWIDTH)
 	{
@@ -290,7 +269,7 @@ file_getbuffer(struct magic_set *ms)
 
 		np = ms->o.pbuf;
 		op = ms->o.buf;
-		eop = op + strlen(ms->o.buf);
+		eop = op + len;
 
 		while (op < eop) {
 			bytesconsumed = mbrtowc(&nextchar, op,
@@ -350,28 +329,3 @@ file_check_mem(struct magic_set *ms, unsigned int level)
 #endif /* ENABLE_CONDITIONALS */
 	return 0;
 }
-/*
- * Yes these wrappers suffer from buffer overflows, but if your OS does not
- * have the real functions, maybe you should consider replacing your OS?
- */
-#ifndef HAVE_VSNPRINTF
-int
-vsnprintf(char *buf, size_t len, const char *fmt, va_list ap)
-{
-	return vsprintf(buf, fmt, ap);
-}
-#endif
-
-#ifndef HAVE_SNPRINTF
-/*ARGSUSED*/
-int
-snprintf(char *buf, size_t len, const char *fmt, ...)
-{
-	int rv;
-	va_list ap;
-	va_start(ap, fmt);
-	rv = vsprintf(buf, fmt, ap);
-	va_end(ap);
-	return rv;
-}
-#endif
diff --git a/usr.bin/file/is_tar.c b/usr.bin/file/is_tar.c
index cff746c55a4..8e0dc6d37c9 100644
--- a/usr.bin/file/is_tar.c
+++ b/usr.bin/file/is_tar.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: is_tar.c,v 1.8 2008/05/08 01:40:56 chl Exp $ */
+/*	$OpenBSD: is_tar.c,v 1.9 2009/04/24 18:54:34 chl Exp $ */
 /*
  * Copyright (c) Ian F. Darwin 1986-1995.
  * Software written by Ian F. Darwin and others;
@@ -46,7 +46,7 @@
 #include "tar.h"
 
 #ifndef lint
-FILE_RCSID("@(#)$Id: is_tar.c,v 1.8 2008/05/08 01:40:56 chl Exp $")
+FILE_RCSID("@(#)$Id: is_tar.c,v 1.9 2009/04/24 18:54:34 chl Exp $")
 #endif
 
 #define	isodigit(c)	( ((c) >= '0') && ((c) <= '7') )
@@ -54,6 +54,12 @@ FILE_RCSID("@(#)$Id: is_tar.c,v 1.8 2008/05/08 01:40:56 chl Exp $")
 private int is_tar(const unsigned char *, size_t);
 private int from_oct(int, const char *);	/* Decode octal number */
 
+static const char tartype[][32] = {
+	"tar archive",
+	"POSIX tar archive",
+	"POSIX tar archive (GNU)",
+};
+
 protected int
 file_is_tar(struct magic_set *ms, const unsigned char *buf, size_t nbytes)
 {
@@ -61,33 +67,27 @@ file_is_tar(struct magic_set *ms, const unsigned char *buf, size_t nbytes)
 	 * Do the tar test first, because if the first file in the tar
 	 * archive starts with a dot, we can confuse it with an nroff file.
 	 */
-	switch (is_tar(buf, nbytes)) {
-	case 1:
-	        if (file_printf(ms, (ms->flags & MAGIC_MIME) ?
-		    "application/x-tar" : "tar archive") == -1)
-			return -1;
-		return 1;
-	case 2:
-		if (file_printf(ms, (ms->flags & MAGIC_MIME) ?
-		    "application/x-tar, POSIX" : "POSIX tar archive") == -1)
-			return -1;
-		return 1;
-	case 3:
-		if (file_printf(ms, (ms->flags & MAGIC_MIME) ?
-		    "application/x-tar, POSIX (GNU)" :
-		    "POSIX tar archive (GNU)") == -1)
-			return -1;
-		return 1;
-	default:
+	int tar = is_tar(buf, nbytes);
+	int mime = ms->flags & MAGIC_MIME;
+
+	if (tar < 1 || tar > 3)
 		return 0;
-	}
+
+	if (mime == MAGIC_MIME_ENCODING)
+		return 0;
+
+	if (file_printf(ms, mime ? "application/x-tar" :
+	    tartype[tar - 1]) == -1)
+		return -1;
+	return 1;
 }
 
 /*
  * Return 
  *	0 if the checksum is bad (i.e., probably not a tar archive), 
  *	1 for old UNIX tar file,
- *	2 for Unix Std (POSIX) tar file.
+ *	2 for Unix Std (POSIX) tar file,
+ *	3 for GNU tar file.
  */
 private int
 is_tar(const unsigned char *buf, size_t nbytes)
diff --git a/usr.bin/file/magdir/acorn b/usr.bin/file/magdir/acorn
index 8245a96b777..156b0235521 100644
--- a/usr.bin/file/magdir/acorn
+++ b/usr.bin/file/magdir/acorn
@@ -1,4 +1,4 @@
-#	$OpenBSD: acorn,v 1.1 2004/06/03 03:14:19 tedu Exp $
+#	$OpenBSD: acorn,v 1.2 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # acorn:  file(1) magic for files found on Acorn systems
@@ -32,3 +32,38 @@
 0	string		Maestro\r	RISC OS music file
 >8	byte		x		version %d
 
+>8	byte		x		type %d
+
+# Digital Symphony data files
+# From: Bernard Jungen (bern8817@euphonynet.be)
+0		string	\x02\x01\x13\x13\x13\x01\x0d\x10	Digital Symphony sound sample (RISC OS),
+>8		byte	x	version %d,
+>9		pstring	x	named "%s",
+>(9.b+19)	byte	=0	8-bit logarithmic
+>(9.b+19)	byte	=1	LZW-compressed linear
+>(9.b+19)	byte	=2	8-bit linear signed
+>(9.b+19)	byte	=3	16-bit linear signed
+>(9.b+19)	byte	=4	SigmaDelta-compressed linear
+>(9.b+19)	byte	=5	SigmaDelta-compressed logarithmic
+>(9.b+19)	byte	>5	unknown format
+
+0	string	\x02\x01\x13\x13\x14\x12\x01\x0b	Digital Symphony song (RISC OS),
+>8	byte	x	version %d,
+>9	byte	=1	1 voice,
+>9	byte	!1	%d voices,
+>10	leshort	=1	1 track,
+>10	leshort	!1	%d tracks,
+>12	leshort	=1	1 pattern
+>12	leshort	!1	%d patterns
+
+0	string	\x02\x01\x13\x13\x10\x14\x12\x0e
+>9	byte	=0	Digital Symphony sequence (RISC OS),
+>>8	byte	x	version %d,
+>>10	byte	=1	1 line,
+>>10	byte	!1	%d lines,
+>>11	leshort	=1	1 position
+>>11	leshort	!1	%d positions
+>9	byte	=1	Digital Symphony pattern data (RISC OS),
+>>8	byte	x	version %d,
+>>10	leshort	=1	1 pattern
+>>10	leshort	!1	%d patterns
diff --git a/usr.bin/file/magdir/adventure b/usr.bin/file/magdir/adventure
index 9d8c985df6e..04ff0ceb8ae 100644
--- a/usr.bin/file/magdir/adventure
+++ b/usr.bin/file/magdir/adventure
@@ -1,4 +1,4 @@
-#	$OpenBSD: adventure,v 1.1 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: adventure,v 1.2 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # adventure: file(1) magic for Adventure game files
@@ -74,3 +74,13 @@
 >10	belong	 0x0A0D1A00
 >>14	string	>\0		%s saved game data
 
+# Danny Milosavljevic <danny.milo@gmx.net>
+# this are adrift (adventure game standard) game files, extension .taf
+# depending on version magic continues with 0x93453E6139FA (V 4.0)
+# 0x9445376139FA (V 3.90)
+# 0x9445366139FA (V 3.80)
+# this is from source (http://www.adrift.org.uk/) and I have some taf
+# files, and checked them.
+#0	belong	0x3C423FC9
+#>4	belong	0x6A87C2CF	Adrift game file
+#!:mime	application/x-adrift
diff --git a/usr.bin/file/magdir/animation b/usr.bin/file/magdir/animation
index e525bd137c3..d498d37c106 100644
--- a/usr.bin/file/magdir/animation
+++ b/usr.bin/file/magdir/animation
@@ -1,4 +1,4 @@
-#	$OpenBSD: animation,v 1.4 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: animation,v 1.5 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # animation:  file(1) magic for animation/movie formats
@@ -9,37 +9,58 @@
 
 # SGI and Apple formats
 0	string		MOVI		Silicon Graphics movie file
+!:mime	video/x-sgi-movie
 4       string          moov            Apple QuickTime
+!:mime	video/quicktime
 >12     string          mvhd            \b movie (fast start)
 >12     string          mdra            \b URL
 >12     string          cmov            \b movie (fast start, compressed header)
 >12     string          rmra            \b multiple URLs
 4       string          mdat            Apple QuickTime movie (unoptimized)
-4       string          wide            Apple QuickTime movie (unoptimized)
-4       string          skip            Apple QuickTime movie (modified)
-4       string          free            Apple QuickTime movie (modified)
+!:mime	video/quicktime
+#4       string          wide            Apple QuickTime movie (unoptimized)
+#!:mime	video/quicktime
+#4       string          skip            Apple QuickTime movie (modified)
+#!:mime	video/quicktime
+#4       string          free            Apple QuickTime movie (modified)
+#!:mime	video/quicktime
 4       string          idsc            Apple QuickTime image (fast start)
-4       string          idat            Apple QuickTime image (unoptimized)
+!:mime	image/x-quicktime
+#4       string          idat            Apple QuickTime image (unoptimized)
+#!:mime	image/x-quicktime
 4       string          pckg            Apple QuickTime compressed archive
+!:mime	application/x-quicktime-player
 4	string/B	jP		JPEG 2000 image
+!:mime	image/jp2
 4	string		ftyp		ISO Media
 >8	string		isom		\b, MPEG v4 system, version 1
+!:mime	video/mp4
 >8	string		iso2		\b, MPEG v4 system, part 12 revision
 >8	string		mp41		\b, MPEG v4 system, version 1
+!:mime	video/mp4
 >8	string		mp42		\b, MPEG v4 system, version 2
+!:mime	video/mp4
 >8	string		mp7t		\b, MPEG v4 system, MPEG v7 XML
 >8	string		mp7b		\b, MPEG v4 system, MPEG v7 binary XML
 >8	string/B	jp2		\b, JPEG 2000
+!:mime	image/jp2
 >8	string		3gp		\b, MPEG v4 system, 3GPP
+!:mime	video/3gpp
 >>11	byte		4		\b v4 (H.263/AMR GSM 6.10)
 >>11	byte		5		\b v5 (H.263/AMR GSM 6.10)
 >>11	byte		6		\b v6 (ITU H.264/AMR GSM 6.10)
 >8	string		mmp4		\b, MPEG v4 system, 3GPP Mobile
+!:mime	video/mp4
 >8	string		avc1		\b, MPEG v4 system, 3GPP JVT AVC
+!:mime	video/3gpp
 >8	string/B	M4A		\b, MPEG v4 system, iTunes AAC-LC
+!:mime	audio/mp4
+>8	string/B	M4V		\b, MPEG v4 system, iTunes AVC-LC
+!:mime	video/mp4
 >8	string/B	M4P		\b, MPEG v4 system, iTunes AES encrypted
 >8	string/B	M4B		\b, MPEG v4 system, iTunes bookmarked
 >8	string/B	qt		\b, Apple QuickTime movie
+!:mime	video/quicktime
 
 # MPEG sequences
 # Scans for all common MPEG header start codes
@@ -49,7 +70,7 @@
 >>5      byte               77             \b, main
 >>5      byte               88             \b, extended
 >>7      byte               x              \b @ L %u
-0        belong&0xFFFFFF00  0x00000100     MPEG sequence
+0        string             \000\000\001   MPEG sequence
 >3       byte               0xBA
 >>4      byte               &0x40          \b, v2, program multiplex
 >>4      byte               ^0x40          \b, v1, system multiplex
@@ -253,22 +274,39 @@
 # modified to fully support MPEG ADTS
 
 # MP3, M1A
-0       beshort&0xFFFE  0xFFFA         MPEG ADTS, layer III, v1
+# modified by Joerg Jenderek
+# GRR the original test are too common for many DOS files
+# so don't accept as MP3 until we've tested the rate
+0       beshort&0xFFFE
 # rates
->2      byte&0xF0       0x10           \b,  32 kBits
->2      byte&0xF0       0x20           \b,  40 kBits
->2      byte&0xF0       0x30           \b,  48 kBits
->2      byte&0xF0       0x40           \b,  56 kBits
->2      byte&0xF0       0x50           \b,  64 kBits
->2      byte&0xF0       0x60           \b,  80 kBits
->2      byte&0xF0       0x70           \b,  96 kBits
->2      byte&0xF0       0x80           \b, 112 kBits
->2      byte&0xF0       0x90           \b, 128 kBits
->2      byte&0xF0       0xA0           \b, 160 kBits
->2      byte&0xF0       0xB0           \b, 192 kBits
->2      byte&0xF0       0xC0           \b, 224 kBits
->2      byte&0xF0       0xD0           \b, 256 kBits
->2      byte&0xF0       0xE0           \b, 320 kBits
+>2      byte&0xF0       0x10           MPEG ADTS, layer III, v1,  32 kbps
+!:mime	audio/mpeg
+>2      byte&0xF0       0x20           MPEG ADTS, layer III, v1,  40 kbps
+!:mime	audio/mpeg
+>2      byte&0xF0       0x30           MPEG ADTS, layer III, v1,  48 kbps
+!:mime	audio/mpeg
+>2      byte&0xF0       0x40           MPEG ADTS, layer III, v1,  56 kbps
+!:mime	audio/mpeg
+>2      byte&0xF0       0x50           MPEG ADTS, layer III, v1,  64 kbps
+!:mime	audio/mpeg
+>2      byte&0xF0       0x60           MPEG ADTS, layer III, v1,  80 kbps
+!:mime	audio/mpeg
+>2      byte&0xF0       0x70           MPEG ADTS, layer III, v1,  96 kbps
+!:mime	audio/mpeg
+>2      byte&0xF0       0x80           MPEG ADTS, layer III, v1, 112 kbps
+!:mime	audio/mpeg
+>2      byte&0xF0       0x90           MPEG ADTS, layer III, v1, 128 kbps
+!:mime	audio/mpeg
+>2      byte&0xF0       0xA0           MPEG ADTS, layer III, v1, 160 kbps
+!:mime	audio/mpeg
+>2      byte&0xF0       0xB0           MPEG ADTS, layer III, v1, 192 kbps
+!:mime	audio/mpeg
+>2      byte&0xF0       0xC0           MPEG ADTS, layer III, v1, 224 kbps
+!:mime	audio/mpeg
+>2      byte&0xF0       0xD0           MPEG ADTS, layer III, v1, 256 kbps
+!:mime	audio/mpeg
+>2      byte&0xF0       0xE0           MPEG ADTS, layer III, v1, 320 kbps
+!:mime	audio/mpeg
 # timing
 >2      byte&0x0C       0x00           \b, 44.1 kHz
 >2      byte&0x0C       0x04           \b, 48 kHz
@@ -289,20 +327,20 @@
 # MP2, M1A
 0       beshort&0xFFFE  0xFFFC         MPEG ADTS, layer II, v1
 # rates
->2      byte&0xF0       0x10           \b,  32 kBits
->2      byte&0xF0       0x20           \b,  48 kBits
->2      byte&0xF0       0x30           \b,  56 kBits
->2      byte&0xF0       0x40           \b,  64 kBits
->2      byte&0xF0       0x50           \b,  80 kBits
->2      byte&0xF0       0x60           \b,  96 kBits
->2      byte&0xF0       0x70           \b, 112 kBits
->2      byte&0xF0       0x80           \b, 128 kBits
->2      byte&0xF0       0x90           \b, 160 kBits
->2      byte&0xF0       0xA0           \b, 192 kBits
->2      byte&0xF0       0xB0           \b, 224 kBits
->2      byte&0xF0       0xC0           \b, 256 kBits
->2      byte&0xF0       0xD0           \b, 320 kBits
->2      byte&0xF0       0xE0           \b, 384 kBits
+>2      byte&0xF0       0x10           \b,  32 kbps
+>2      byte&0xF0       0x20           \b,  48 kbps
+>2      byte&0xF0       0x30           \b,  56 kbps
+>2      byte&0xF0       0x40           \b,  64 kbps
+>2      byte&0xF0       0x50           \b,  80 kbps
+>2      byte&0xF0       0x60           \b,  96 kbps
+>2      byte&0xF0       0x70           \b, 112 kbps
+>2      byte&0xF0       0x80           \b, 128 kbps
+>2      byte&0xF0       0x90           \b, 160 kbps
+>2      byte&0xF0       0xA0           \b, 192 kbps
+>2      byte&0xF0       0xB0           \b, 224 kbps
+>2      byte&0xF0       0xC0           \b, 256 kbps
+>2      byte&0xF0       0xD0           \b, 320 kbps
+>2      byte&0xF0       0xE0           \b, 384 kbps
 # timing
 >2      byte&0x0C       0x00           \b, 44.1 kHz
 >2      byte&0x0C       0x04           \b, 48 kHz
@@ -323,58 +361,59 @@
 # MPA, M1A
 # updated by Joerg Jenderek
 # GRR the original test are too common for many DOS files, so test 32 <= kbits <= 448
-0	beshort&0xFFFE		0xFFFE	
->2	ubyte&0xF0	>0x0F		
->>2	ubyte&0xF0	<0xE1		MPEG ADTS, layer I, v1
-# rate
->>>2      byte&0xF0       0x10           \b,  32 kBits
->>>2      byte&0xF0       0x20           \b,  64 kBits
->>>2      byte&0xF0       0x30           \b,  96 kBits
->>>2      byte&0xF0       0x40           \b, 128 kBits
->>>2      byte&0xF0       0x50           \b, 160 kBits
->>>2      byte&0xF0       0x60           \b, 192 kBits
->>>2      byte&0xF0       0x70           \b, 224 kBits
->>>2      byte&0xF0       0x80           \b, 256 kBits
->>>2      byte&0xF0       0x90           \b, 288 kBits
->>>2      byte&0xF0       0xA0           \b, 320 kBits
->>>2      byte&0xF0       0xB0           \b, 352 kBits
->>>2      byte&0xF0       0xC0           \b, 384 kBits
->>>2      byte&0xF0       0xD0           \b, 416 kBits
->>>2      byte&0xF0       0xE0           \b, 448 kBits
-# timing
->>>2      byte&0x0C       0x00           \b, 44.1 kHz
->>>2      byte&0x0C       0x04           \b, 48 kHz
->>>2      byte&0x0C       0x08           \b, 32 kHz
-# channels/options
->>>3      byte&0xC0       0x00           \b, Stereo
->>>3      byte&0xC0       0x40           \b, JntStereo
->>>3      byte&0xC0       0x80           \b, 2x Monaural
->>>3      byte&0xC0       0xC0           \b, Monaural
-#>1     byte            ^0x01          \b, Data Verify
-#>2     byte            &0x02          \b, Packet Pad
-#>2     byte            &0x01          \b, Custom Flag
-#>3     byte            &0x08          \b, Copyrighted
-#>3     byte            &0x04          \b, Original Source
-#>3     byte&0x03       1              \b, NR: 50/15 ms
-#>3     byte&0x03       3              \b, NR: CCIT J.17
+# GRR this test is still too general as it catches a BOM of UTF-16 files (0xFFFE)
+#0	beshort&0xFFFE		0xFFFE	
+#>2	ubyte&0xF0	>0x0F		
+#>>2	ubyte&0xF0	<0xE1		MPEG ADTS, layer I, v1
+## rate
+#>>>2      byte&0xF0       0x10           \b,  32 kbps
+#>>>2      byte&0xF0       0x20           \b,  64 kbps
+#>>>2      byte&0xF0       0x30           \b,  96 kbps
+#>>>2      byte&0xF0       0x40           \b, 128 kbps
+#>>>2      byte&0xF0       0x50           \b, 160 kbps
+#>>>2      byte&0xF0       0x60           \b, 192 kbps
+#>>>2      byte&0xF0       0x70           \b, 224 kbps
+#>>>2      byte&0xF0       0x80           \b, 256 kbps
+#>>>2      byte&0xF0       0x90           \b, 288 kbps
+#>>>2      byte&0xF0       0xA0           \b, 320 kbps
+#>>>2      byte&0xF0       0xB0           \b, 352 kbps
+#>>>2      byte&0xF0       0xC0           \b, 384 kbps
+#>>>2      byte&0xF0       0xD0           \b, 416 kbps
+#>>>2      byte&0xF0       0xE0           \b, 448 kbps
+## timing
+#>>>2      byte&0x0C       0x00           \b, 44.1 kHz
+#>>>2      byte&0x0C       0x04           \b, 48 kHz
+#>>>2      byte&0x0C       0x08           \b, 32 kHz
+## channels/options
+#>>>3      byte&0xC0       0x00           \b, Stereo
+#>>>3      byte&0xC0       0x40           \b, JntStereo
+#>>>3      byte&0xC0       0x80           \b, 2x Monaural
+#>>>3      byte&0xC0       0xC0           \b, Monaural
+##>1     byte            ^0x01          \b, Data Verify
+##>2     byte            &0x02          \b, Packet Pad
+##>2     byte            &0x01          \b, Custom Flag
+##>3     byte            &0x08          \b, Copyrighted
+##>3     byte            &0x04          \b, Original Source
+##>3     byte&0x03       1              \b, NR: 50/15 ms
+##>3     byte&0x03       3              \b, NR: CCIT J.17
 
 # MP3, M2A
 0       beshort&0xFFFE  0xFFF2         MPEG ADTS, layer III, v2
 # rate
->2      byte&0xF0       0x10           \b,   8 kBits
->2      byte&0xF0       0x20           \b,  16 kBits
->2      byte&0xF0       0x30           \b,  24 kBits
->2      byte&0xF0       0x40           \b,  32 kBits
->2      byte&0xF0       0x50           \b,  40 kBits
->2      byte&0xF0       0x60           \b,  48 kBits
->2      byte&0xF0       0x70           \b,  56 kBits
->2      byte&0xF0       0x80           \b,  64 kBits
->2      byte&0xF0       0x90           \b,  80 kBits
->2      byte&0xF0       0xA0           \b,  96 kBits
->2      byte&0xF0       0xB0           \b, 112 kBits
->2      byte&0xF0       0xC0           \b, 128 kBits
->2      byte&0xF0       0xD0           \b, 144 kBits
->2      byte&0xF0       0xE0           \b, 160 kBits
+>2      byte&0xF0       0x10           \b,   8 kbps
+>2      byte&0xF0       0x20           \b,  16 kbps
+>2      byte&0xF0       0x30           \b,  24 kbps
+>2      byte&0xF0       0x40           \b,  32 kbps
+>2      byte&0xF0       0x50           \b,  40 kbps
+>2      byte&0xF0       0x60           \b,  48 kbps
+>2      byte&0xF0       0x70           \b,  56 kbps
+>2      byte&0xF0       0x80           \b,  64 kbps
+>2      byte&0xF0       0x90           \b,  80 kbps
+>2      byte&0xF0       0xA0           \b,  96 kbps
+>2      byte&0xF0       0xB0           \b, 112 kbps
+>2      byte&0xF0       0xC0           \b, 128 kbps
+>2      byte&0xF0       0xD0           \b, 144 kbps
+>2      byte&0xF0       0xE0           \b, 160 kbps
 # timing
 >2      byte&0x0C       0x00           \b, 22.05 kHz
 >2      byte&0x0C       0x04           \b, 24 kHz
@@ -395,20 +434,20 @@
 # MP2, M2A
 0       beshort&0xFFFE  0xFFF4         MPEG ADTS, layer II, v2
 # rate 
->2      byte&0xF0       0x10           \b,   8 kBits
->2      byte&0xF0       0x20           \b,  16 kBits 
->2      byte&0xF0       0x30           \b,  24 kBits
->2      byte&0xF0       0x40           \b,  32 kBits
->2      byte&0xF0       0x50           \b,  40 kBits
->2      byte&0xF0       0x60           \b,  48 kBits
->2      byte&0xF0       0x70           \b,  56 kBits
->2      byte&0xF0       0x80           \b,  64 kBits
->2      byte&0xF0       0x90           \b,  80 kBits
->2      byte&0xF0       0xA0           \b,  96 kBits
->2      byte&0xF0       0xB0           \b, 112 kBits
->2      byte&0xF0       0xC0           \b, 128 kBits
->2      byte&0xF0       0xD0           \b, 144 kBits
->2      byte&0xF0       0xE0           \b, 160 kBits
+>2      byte&0xF0       0x10           \b,   8 kbps
+>2      byte&0xF0       0x20           \b,  16 kbps 
+>2      byte&0xF0       0x30           \b,  24 kbps
+>2      byte&0xF0       0x40           \b,  32 kbps
+>2      byte&0xF0       0x50           \b,  40 kbps
+>2      byte&0xF0       0x60           \b,  48 kbps
+>2      byte&0xF0       0x70           \b,  56 kbps
+>2      byte&0xF0       0x80           \b,  64 kbps
+>2      byte&0xF0       0x90           \b,  80 kbps
+>2      byte&0xF0       0xA0           \b,  96 kbps
+>2      byte&0xF0       0xB0           \b, 112 kbps
+>2      byte&0xF0       0xC0           \b, 128 kbps
+>2      byte&0xF0       0xD0           \b, 144 kbps
+>2      byte&0xF0       0xE0           \b, 160 kbps
 # timing
 >2      byte&0x0C       0x00           \b, 22.05 kHz
 >2      byte&0x0C       0x04           \b, 24 kHz
@@ -429,20 +468,20 @@
 # MPA, M2A
 0       beshort&0xFFFE  0xFFF6         MPEG ADTS, layer I, v2
 # rate
->2      byte&0xF0       0x10           \b,  32 kBits
->2      byte&0xF0       0x20           \b,  48 kBits
->2      byte&0xF0       0x30           \b,  56 kBits
->2      byte&0xF0       0x40           \b,  64 kBits
->2      byte&0xF0       0x50           \b,  80 kBits
->2      byte&0xF0       0x60           \b,  96 kBits
->2      byte&0xF0       0x70           \b, 112 kBits
->2      byte&0xF0       0x80           \b, 128 kBits
->2      byte&0xF0       0x90           \b, 144 kBits
->2      byte&0xF0       0xA0           \b, 160 kBits
->2      byte&0xF0       0xB0           \b, 176 kBits
->2      byte&0xF0       0xC0           \b, 192 kBits
->2      byte&0xF0       0xD0           \b, 224 kBits
->2      byte&0xF0       0xE0           \b, 256 kBits
+>2      byte&0xF0       0x10           \b,  32 kbps
+>2      byte&0xF0       0x20           \b,  48 kbps
+>2      byte&0xF0       0x30           \b,  56 kbps
+>2      byte&0xF0       0x40           \b,  64 kbps
+>2      byte&0xF0       0x50           \b,  80 kbps
+>2      byte&0xF0       0x60           \b,  96 kbps
+>2      byte&0xF0       0x70           \b, 112 kbps
+>2      byte&0xF0       0x80           \b, 128 kbps
+>2      byte&0xF0       0x90           \b, 144 kbps
+>2      byte&0xF0       0xA0           \b, 160 kbps
+>2      byte&0xF0       0xB0           \b, 176 kbps
+>2      byte&0xF0       0xC0           \b, 192 kbps
+>2      byte&0xF0       0xD0           \b, 224 kbps
+>2      byte&0xF0       0xE0           \b, 256 kbps
 # timing
 >2      byte&0x0C       0x00           \b, 22.05 kHz
 >2      byte&0x0C       0x04           \b, 24 kHz
@@ -463,20 +502,20 @@
 # MP3, M25A
 0       beshort&0xFFFE  0xFFE2         MPEG ADTS, layer III,  v2.5
 # rate  
->2      byte&0xF0       0x10           \b,   8 kBits
->2      byte&0xF0       0x20           \b,  16 kBits
->2      byte&0xF0       0x30           \b,  24 kBits
->2      byte&0xF0       0x40           \b,  32 kBits
->2      byte&0xF0       0x50           \b,  40 kBits
->2      byte&0xF0       0x60           \b,  48 kBits
->2      byte&0xF0       0x70           \b,  56 kBits
->2      byte&0xF0       0x80           \b,  64 kBits
->2      byte&0xF0       0x90           \b,  80 kBits
->2      byte&0xF0       0xA0           \b,  96 kBits
->2      byte&0xF0       0xB0           \b, 112 kBits
->2      byte&0xF0       0xC0           \b, 128 kBits
->2      byte&0xF0       0xD0           \b, 144 kBits
->2      byte&0xF0       0xE0           \b, 160 kBits
+>2      byte&0xF0       0x10           \b,   8 kbps
+>2      byte&0xF0       0x20           \b,  16 kbps
+>2      byte&0xF0       0x30           \b,  24 kbps
+>2      byte&0xF0       0x40           \b,  32 kbps
+>2      byte&0xF0       0x50           \b,  40 kbps
+>2      byte&0xF0       0x60           \b,  48 kbps
+>2      byte&0xF0       0x70           \b,  56 kbps
+>2      byte&0xF0       0x80           \b,  64 kbps
+>2      byte&0xF0       0x90           \b,  80 kbps
+>2      byte&0xF0       0xA0           \b,  96 kbps
+>2      byte&0xF0       0xB0           \b, 112 kbps
+>2      byte&0xF0       0xC0           \b, 128 kbps
+>2      byte&0xF0       0xD0           \b, 144 kbps
+>2      byte&0xF0       0xE0           \b, 160 kbps
 # timing
 >2      byte&0x0C       0x00           \b, 11.025 kHz
 >2      byte&0x0C       0x04           \b, 12 kHz
@@ -498,6 +537,7 @@
 
 # Stored AAC streams (instead of the MP4 format)
 0       string          ADIF           MPEG ADIF, AAC
+!:mime	audio/x-hx-aac-adif
 >4      byte            &0x80
 >>13    byte            &0x10          \b, VBR
 >>13    byte            ^0x10          \b, CBR
@@ -522,6 +562,7 @@
 
 # Live or stored single AAC stream (used with MPEG-2 systems)
 0       beshort&0xFFF6  0xFFF0         MPEG ADTS, AAC
+!:mime	audio/x-hx-aac-adts
 >1      byte            &0x08          \b, v2
 >1      byte            ^0x08          \b, v4
 # profile
@@ -558,6 +599,7 @@
 
 # Live MPEG-4 audio streams (instead of RTP FlexMux)
 0       beshort&0xFFE0  0x56E0         MPEG-4 LOAS
+!:mime	audio/x-mp4a-latm
 #>1     beshort&0x1FFF  x              \b, %u byte packet
 >3      byte&0xE0       0x40
 >>4     byte&0x3C       0x04           \b, single stream
@@ -571,10 +613,13 @@
 >>4     byte&0x78       0x18           \b, 3 streams
 >>4     byte            &0x20          \b, 4 or more streams
 >>4     byte            &0x40          \b, 8 or more streams
-0       beshort         0x4DE1         MPEG-4 LO-EP audio stream
+# This magic isn't strong enough (matches plausible ISO-8859-1 text)
+#0       beshort         0x4DE1         MPEG-4 LO-EP audio stream
+#!:mime	audio/x-mp4a-latm
 
 # FLI animation format
 4	leshort		0xAF11			FLI file
+!:mime	video/x-fli
 >6	leshort		x			- %d frames,
 >8	leshort		x			width=%d pixels,
 >10	leshort		x			height=%d pixels,
@@ -582,6 +627,7 @@
 >16	leshort		x			ticks/frame=%d
 # FLC animation format
 4	leshort		0xAF12			FLC file
+!:mime	video/x-flc
 >6	leshort		x			- %d frames
 >8	leshort		x			width=%d pixels,
 >10	leshort		x			height=%d pixels,
@@ -600,9 +646,11 @@
 # 255 (hex FF)!  The DL format is really bad.
 #
 #0	byte	1	DL version 1, medium format (160x100, 4 images/screen)
+#!:mime	video/x-unknown
 #>42	byte	x	- %d screens,
 #>43	byte	x	%d commands
 #0	byte	2	DL version 2
+#!:mime	video/x-unknown
 #>1	byte	1	- large format (320x200,1 image/screen),
 #>1	byte	2	- medium format (160x100,4 images/screen),
 #>1	byte	>2	- unknown format,
@@ -642,6 +690,7 @@
 
 # MNG Video Format, <URL:http://www.libpng.org/pub/mng/spec/>
 0	string			\x8aMNG		MNG video data,
+!:mime	video/x-mng
 >4	belong			!0x0d0a1a0a	CORRUPTED,
 >4	belong			0x0d0a1a0a
 >>16    belong	x				%ld x
@@ -649,6 +698,7 @@
 
 # JNG Video Format, <URL:http://www.libpng.org/pub/mng/spec/>
 0	string			\x8bJNG		JNG video data,
+!:mime	video/x-jng
 >4	belong			!0x0d0a1a0a	CORRUPTED,
 >4	belong			0x0d0a1a0a
 >>16    belong	x				%ld x
@@ -659,7 +709,9 @@
 
 # VRML (Virtual Reality Modelling Language)
 0       string/b        #VRML\ V1.0\ ascii	VRML 1 file
+!:mime	model/vrml
 0	string/b	#VRML\ V2.0\ utf8	ISO/IEC 14772 VRML 97 file
+!:mime	model/vrml
 
 #---------------------------------------------------------------------------
 # HVQM4: compressed movie format designed by Hudson for Nintendo GameCube
@@ -679,3 +731,47 @@
 >0x21	byte		x		v%x
 0	string		DVDVIDEO-VMG	Video manager,
 >0x21	byte		x		v%x
+
+# From: Behan Webster <behanw@websterwood.com>
+# NuppelVideo used by Mythtv (*.nuv)
+# Note: there are two identical stanzas here differing only in the
+# initial string matched. It used to be done with a regex, but we're
+# trying to get rid of those.
+0	string		NuppelVideo	MythTV NuppelVideo
+>12	string		x		v%s
+>20	lelong		x		(%d
+>24	lelong		x		\bx%d),
+>36	string		P		\bprogressive,
+>36	string		I		\binterlaced,
+>40	ledouble	x		\baspect:%.2f,
+>48	ledouble	x		\bfps:%.2f
+0	string		MythTV		MythTV NuppelVideo
+>12	string		x		v%s
+>20	lelong		x		(%d
+>24	lelong		x		\bx%d),
+>36	string		P		\bprogressive,
+>36	string		I		\binterlaced,
+>40	ledouble	x		\baspect:%.2f,
+>48	ledouble	x		\bfps:%.2f
+
+#						MPEG file
+# MPEG sequences
+# FIXME: This section is from the old magic.mime file and needs integrating with the rest
+0       belong             0x000001BA
+>4      byte               &0x40
+!:mime	video/mp2p
+>4      byte               ^0x40
+!:mime	video/mpeg
+0       belong             0x000001BB
+!:mime	video/mpeg
+0       belong             0x000001B0
+!:mime	video/mp4v-es
+0       belong             0x000001B5
+!:mime	video/mp4v-es
+0       belong             0x000001B3
+!:mime	video/mpv
+0       belong&0xFF5FFF1F  0x47400010
+!:mime	video/mp2t
+0       belong             0x00000001
+>4      byte&0x1F	   0x07
+!:mime	video/h264
diff --git a/usr.bin/file/magdir/apple b/usr.bin/file/magdir/apple
index 381a59657f8..b68eb49d481 100644
--- a/usr.bin/file/magdir/apple
+++ b/usr.bin/file/magdir/apple
@@ -1,9 +1,9 @@
-#	$OpenBSD: apple,v 1.5 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: apple,v 1.6 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # apple:  file(1) magic for Apple file formats
 #
-0	string		FiLeStArTfIlEsTaRt	binscii (apple ][) text
+0	search/1	FiLeStArTfIlEsTaRt	binscii (apple ][) text
 0	string		\x0aGL			Binary II (apple ][) data
 0	string		\x76\xff		Squeezed (apple ][) data
 0	string		NuFile			NuFile archive (apple ][) data
@@ -11,6 +11,23 @@
 0	belong		0x00051600		AppleSingle encoded Macintosh file
 0	belong		0x00051607		AppleDouble encoded Macintosh file
 
+# Type: Apple Emulator 2IMG format
+# From: Radek Vokal <rvokal@redhat.com>
+0	string		2IMG	Apple ][ 2IMG Disk Image
+>4	string		XGS!	\b, XGS
+>4	string		CTKG	\b, Catakig
+>4	string		ShIm	\b, Sheppy's ImageMaker
+>4	string		WOOF	\b, Sweet 16
+>4	string		B2TR	\b, Bernie ][ the Rescue
+>4	string		!nfc	\b, ASIMOV2
+>4	string		x	\b, Unknown Format
+>0xc	byte		00	\b, DOS 3.3 sector order
+>>0x10	byte		00	\b, Volume 254
+>>0x10	byte&0x7f	x	\b, Volume %u
+>0xc	byte		01	\b, ProDOS sector order
+>>0x14	short		x	\b, %u Blocks
+>0xc	byte		02	\b, NIB data
+
 # magic for Newton PDA package formats
 # from Ruda Moura <ruda@helllabs.org>
 0	string	package0	Newton package, NOS 1.x,
@@ -119,11 +136,12 @@
 # Nevertheless this will manage to catch a lot of images that happen
 # to have a solid-colored line at the bottom of the screen.
 
-8144	string	\x7F\x7F\x7F\x7F\x7F\x7F\x7F\x7F	Apple II image with white background
-8144	string	\x55\x2A\x55\x2A\x55\x2A\x55\x2A	Apple II image with purple background
-8144	string	\x2A\x55\x2A\x55\x2A\x55\x2A\x55	Apple II image with green background
-8144	string	\xD5\xAA\xD5\xAA\xD5\xAA\xD5\xAA	Apple II image with blue background
-8144	string	\xAA\xD5\xAA\xD5\xAA\xD5\xAA\xD5	Apple II image with orange background
+# GRR: Magic too weak
+#8144	string	\x7F\x7F\x7F\x7F\x7F\x7F\x7F\x7F	Apple II image with white background
+#8144	string	\x55\x2A\x55\x2A\x55\x2A\x55\x2A	Apple II image with purple background
+#8144	string	\x2A\x55\x2A\x55\x2A\x55\x2A\x55	Apple II image with green background
+#8144	string	\xD5\xAA\xD5\xAA\xD5\xAA\xD5\xAA	Apple II image with blue background
+#8144	string	\xAA\xD5\xAA\xD5\xAA\xD5\xAA\xD5	Apple II image with orange background
 
 # Beagle Bros. Apple Mechanic fonts
 
@@ -190,3 +208,44 @@
 >0		byte		<5		\b
 >>13	byte		0x81	\b
 >>>14	uleshort	x		\b, system %hd
+
+#------------------------------------------------------------------------------
+# CAF: Apple CoreAudio File Format
+#
+# Container format for high-end audio purposes.
+# From: David Remahl <dremahl@apple.com>
+#
+0	string		caff		CoreAudio Format audio file
+>4	beshort		<10		version %d
+>6	beshort		x
+
+
+#------------------------------------------------------------------------------
+# Keychain database files
+0	string		kych		Mac OS X Keychain File
+
+#------------------------------------------------------------------------------
+# Code Signing related file types
+0	belong		0xfade0c00	Mac OS X Code Requirement
+>8	belong		1			(opExpr)
+>4	belong		x			- %d bytes
+
+0	belong		0xfade0c01	Mac OS X Code Requirement Set
+>8	belong		>1			containing %d items
+>4	belong		x			- %d bytes
+
+0	belong		0xfade0c02	Mac OS X Code Directory
+>8	belong		x			version %x
+>12	belong		>0			flags 0x%x
+>4	belong		x			- %d bytes
+
+0	belong		0xfade0cc0	Mac OS X Detached Code Signature (non-executable)
+>4	belong		x			- %d bytes
+
+0	belong		0xfade0cc1	Mac OS X Detached Code Signature
+>8	belong		>1			(%d elements)
+>4	belong		x			- %d bytes
+
+# From: "Nelson A. de Oliveira" <naoliv@gmail.com>
+# .vdi
+4	string innotek\ VirtualBox\ Disk\ Image %s
diff --git a/usr.bin/file/magdir/archive b/usr.bin/file/magdir/archive
index 6c9351b516f..4fd090e4aac 100644
--- a/usr.bin/file/magdir/archive
+++ b/usr.bin/file/magdir/archive
@@ -1,4 +1,4 @@
-#	$OpenBSD: archive,v 1.5 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: archive,v 1.6 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # archive:  file(1) magic for archive formats (see also "msdos" for self-
@@ -9,7 +9,9 @@
 
 # POSIX tar archives
 257	string		ustar\0		POSIX tar archive
+!:mime	application/x-tar # encoding: posix
 257	string		ustar\040\040\0	GNU tar archive
+!:mime	application/x-tar # encoding: gnu
 
 # cpio archives
 #
@@ -23,7 +25,9 @@
 # are defined as "short"s; I think all the new formats are
 # character-header formats and thus are strings, not numbers.
 0	short		070707		cpio archive
+!:mime	application/x-cpio
 0	short		0143561		byte-swapped cpio archive
+!:mime	application/x-cpio # encoding: swapped
 0	string		070707		ASCII cpio archive (pre-SVR4 or odc)
 0	string		070701		ASCII cpio archive (SVR4 with no CRC)
 0	string		070702		ASCII cpio archive (SVR4 with CRC)
@@ -31,8 +35,10 @@
 # Debian package (needs to go before regular portable archives)
 #
 0	string		=!<arch>\ndebian
+!:mime	application/x-debian-package
 >8	string		debian-split	part of multipart Debian package
 >8	string		debian-binary	Debian binary package
+>8	string		!debian
 >68	string		>\0		(format %s)
 # These next two lines do not work, because a bzip2 Debian archive
 # still uses gzip for the control.tar (first in the archive).  Only
@@ -50,6 +56,7 @@
 0	short		0177545		old PDP-11 archive
 0	long		0100554		apl workspace
 0	string		=<ar>		archive
+!:mime	application/x-archive
 
 # MIPS archive (needs to go before regular portable archives)
 #
@@ -61,7 +68,7 @@
 >19	string	B			and an EB hash table
 >22	string	X			-- out of date
 
-0	string		-h-		Software Tools format archive text
+0	search/1	-h-		Software Tools format archive text
 
 #
 # XXX - why are there multiple <ar> thingies?  Note that 0x213c6172 is
@@ -80,6 +87,7 @@
 # "ar" archives?
 #
 0	string		=!<arch>		current ar archive
+!:mime	application/x-archive
 >8	string		__.SYMDEF	random library
 >0	belong		=65538		- pre SR9.5
 >0	belong		=65539		- post SR9.5
@@ -122,15 +130,24 @@
 # we only test some types on basis of frequency: 0x08 (83%), 0x09 (5%),
 # 0x02 (5%), 0x03 (3%), 0x04 (2%), 0x06 (2%).  0x01 collides with terminfo.
 0	lelong&0x8080ffff	0x0000081a	ARC archive data, dynamic LZW
+!:mime	application/x-arc
 0	lelong&0x8080ffff	0x0000091a	ARC archive data, squashed
+!:mime	application/x-arc
 0	lelong&0x8080ffff	0x0000021a	ARC archive data, uncompressed
+!:mime	application/x-arc
 0	lelong&0x8080ffff	0x0000031a	ARC archive data, packed
+!:mime	application/x-arc
 0	lelong&0x8080ffff	0x0000041a	ARC archive data, squeezed
+!:mime	application/x-arc
 0	lelong&0x8080ffff	0x0000061a	ARC archive data, crunched
+!:mime	application/x-arc
 # [JW] stuff taken from idarc, obviously ARC successors:
 0	lelong&0x8080ffff	0x00000a1a	PAK archive data
+!:mime	application/x-arc
 0	lelong&0x8080ffff	0x0000141a	ARC+ archive data
+!:mime	application/x-arc
 0	lelong&0x8080ffff	0x0000481a	HYP archive data
+!:mime	application/x-arc
 
 # Acorn archive formats (Disaster prone simpleton, m91dps@ecs.ox.ac.uk)
 # I can't create either SPARK or ArcFS archives so I have not tested this stuff
@@ -436,6 +453,7 @@
 
 # ARJ archiver (jason@jarthur.Claremont.EDU)
 0	leshort		0xea60		ARJ archive data
+!:mime	application/x-arj
 >5	byte		x		\b, v%d,
 >8	byte		&0x04		multi-volume,
 >8	byte		&0x10		slash-switched,
@@ -488,19 +506,32 @@
 
 # LHARC/LHA archiver (Greg Roelofs, newt@uchicago.edu)
 2	string		-lh0-		LHarc 1.x/ARX archive data [lh0]
+!:mime	application/x-lharc
 2	string		-lh1-		LHarc 1.x/ARX archive data [lh1]
+!:mime	application/x-lharc
 2	string		-lz4-		LHarc 1.x archive data [lz4]
+!:mime	application/x-lharc
 2	string		-lz5-		LHarc 1.x archive data [lz5]
+!:mime	application/x-lharc
 #	[never seen any but the last; -lh4- reported in comp.compression:]
 2	string		-lzs-		LHa/LZS archive data [lzs]
+!:mime	application/x-lha
 2	string		-lh\40-		LHa 2.x? archive data [lh ]
+!:mime	application/x-lha
 2	string		-lhd-		LHa 2.x? archive data [lhd]
+!:mime	application/x-lha
 2	string		-lh2-		LHa 2.x? archive data [lh2]
+!:mime	application/x-lha
 2	string		-lh3-		LHa 2.x? archive data [lh3]
+!:mime	application/x-lha
 2	string		-lh4-		LHa (2.x) archive data [lh4]
+!:mime	application/x-lha
 2	string		-lh5-		LHa (2.x) archive data [lh5]
+!:mime	application/x-lha
 2	string		-lh6-		LHa (2.x) archive data [lh6]
+!:mime	application/x-lha
 2	string		-lh7-		LHa (2.x)/LHark archive data [lh7]
+!:mime	application/x-lha
 >20	byte		x		- header level %d
 # taken from idarc [JW]
 2   string      -lZ         PUT archive data
@@ -509,6 +540,7 @@
 
 # RAR archiver (Greg Roelofs, newt@uchicago.edu)
 0	string		Rar!		RAR archive data,
+!:mime	application/x-rar
 >44	byte		x		v%0x,
 >10	byte		>0		flags:
 >>10	byte		&0x01		Archive volume,
@@ -532,13 +564,19 @@
 
 # ZIP archives (Greg Roelofs, c/o zip-bugs@wkuvx1.wku.edu)
 0	string		PK\003\004
+>4	byte		0x00		Zip archive data
+!:mime	application/zip
 >4	byte		0x09		Zip archive data, at least v0.9 to extract
+!:mime	application/zip
 >4	byte		0x0a		Zip archive data, at least v1.0 to extract
+!:mime	application/zip
 >4	byte		0x0b		Zip archive data, at least v1.1 to extract
+!:mime	application/zip
+>0x161	string		WINZIP          Zip archive data, WinZIP self-extracting
+!:mime	application/zip
 >4	byte		0x14
 >>30	ubelong		!0x6d696d65	Zip archive data, at least v2.0 to extract
->0x161	string		WINZIP          Zip archive data, WinZIP self-extracting
-
+!:mime	application/zip
 
 # OpenOffice.org / KOffice / StarOffice documents
 # From: Abel Cheung <abel@oaka.org>
@@ -578,6 +616,7 @@
 >>>50	string	vnd.oasis.opendocument.	OpenDocument
 >>>>73	string	text
 >>>>>77	byte	!0x2d			Text
+!:mime	application/vnd.oasis.opendocument.text
 >>>>>77	string	-template		Text Template
 >>>>>77	string	-web			HTML Document Template
 >>>>>77	string	-master			Master Document
@@ -596,6 +635,7 @@
 
 # Zoo archiver
 20	lelong		0xfdc4a7dc	Zoo archive data
+!:mime	application/x-zoo
 >4	byte		>48		\b, v%c.
 >>6	byte		>47		\b%c
 >>>7	byte		>47		\b%c
@@ -607,6 +647,7 @@
 
 # Shell archives
 10	string		#\ This\ is\ a\ shell\ archive	shell archive text
+!:mime	application/octet-stream
 
 #
 # LBR. NB: May conflict with the questionable 
@@ -652,6 +693,7 @@
 
 # Felix von Leitner <felix-file@fefe.de>
 0	string	d8:announce	BitTorrent file
+!:mime	application/x-bittorrent
 
 # Atari MSA archive - Teemu Hukkanen <tjhukkan@iki.fi>
 0       beshort 0x0e0f          Atari MSA archive data
@@ -709,6 +751,7 @@
 # EET archive
 # From: Tilman Sauerbeck <tilman@code-monkey.de>
 0	belong	0x1ee7ff00	EET archive
+!:mime	application/x-eet
 
 # rzip archives
 0	string	RZIP		rzip compressed data
@@ -729,6 +772,36 @@
 #  http://www.thouky.co.uk/software/psifs/sis.html
 #  http://developer.symbian.com/main/downloads/papers/SymbianOSv91/softwareinstallsis.pdf
 8	lelong	0x10000419	Symbian installation file
+!:mime	application/vnd.symbian.install
 >4	lelong	0x1000006D	(EPOC release 3/4/5)
 >4	lelong	0x10003A12	(EPOC release 6)
 0	lelong	0x10201A7A	Symbian installation file (Symbian OS 9.x)
+!:mime	x-epoc/x-sisx-app
+
+# Pack200 Java archives, http://jcp.org/en/jsr/detail?id=200
+0	belong	0xcafed00d	Pack200 Java archive
+
+# From "Nelson A. de Oliveira" <naoliv@gmail.com>
+0	string	MPQ\032		MoPaQ (MPQ) archive
+
+# From: Dirk Jagdmann <doj@cubic.org>
+# xar archive format: http://code.google.com/p/xar/
+0	string	xar!		xar archive
+>6	beshort	x		- version %ld
+
+# From: "Nelson A. de Oliveira" <naoliv@gmail.com>
+# .kgb
+0	string KGB_arch		KGB Archiver file
+>10	string x		with compression level %.1s
+
+# xar (eXtensible ARchiver) archive
+# From: "David Remahl" <dremahl@apple.com>
+0	string	xar!		xar archive
+#>4	beshort	x		header size %d
+>6	beshort	x		version %d,
+#>8	quad	x		compressed TOC: %d,
+#>16	quad	x		uncompressed TOC: %d,
+>24	belong	0		no checksum
+>24	belong	1		SHA-1 checksum
+>24	belong	2		MD5 checksum
+
diff --git a/usr.bin/file/magdir/audio b/usr.bin/file/magdir/audio
index d668d13329e..7b38422a476 100644
--- a/usr.bin/file/magdir/audio
+++ b/usr.bin/file/magdir/audio
@@ -1,4 +1,4 @@
-#	$OpenBSD: audio,v 1.10 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: audio,v 1.11 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # audio:  file(1) magic for sound formats (see also "iff")
@@ -10,12 +10,19 @@
 # Sun/NeXT audio data
 0	string		.snd		Sun/NeXT audio data:
 >12	belong		1		8-bit ISDN mu-law,
+!:mime	audio/basic
 >12	belong		2		8-bit linear PCM [REF-PCM],
+!:mime	audio/basic
 >12	belong		3		16-bit linear PCM,
+!:mime	audio/basic
 >12	belong		4		24-bit linear PCM,
+!:mime	audio/basic
 >12	belong		5		32-bit linear PCM,
+!:mime	audio/basic
 >12	belong		6		32-bit IEEE floating point,
+!:mime	audio/basic
 >12	belong		7		64-bit IEEE floating point,
+!:mime	audio/basic
 >12	belong		8		Fragmented sample data,
 >12	belong		10		DSP program,
 >12	belong		11		8-bit fixed point,
@@ -27,6 +34,7 @@
 >12	belong		20		16-bit linear with emphasis and compression,
 >12	belong		21		Music kit DSP commands,
 >12	belong		23		8-bit ISDN mu-law compressed (CCITT G.721 ADPCM voice enc.),
+!:mime  audio/x-adpcm
 >12	belong		24		compressed (8-bit CCITT G.722 ADPCM)
 >12	belong		25		compressed (3-bit CCITT G.723.3 ADPCM),
 >12	belong		26		compressed (5-bit CCITT G.723.5 ADPCM),
@@ -40,12 +48,19 @@
 # that uses little-endian encoding and has a different magic number
 0	lelong		0x0064732E	DEC audio data:
 >12	lelong		1		8-bit ISDN mu-law,
+!:mime	audio/x-dec-basic
 >12	lelong		2		8-bit linear PCM [REF-PCM],
+!:mime	audio/x-dec-basic
 >12	lelong		3		16-bit linear PCM,
+!:mime	audio/x-dec-basic
 >12	lelong		4		24-bit linear PCM,
+!:mime	audio/x-dec-basic
 >12	lelong		5		32-bit linear PCM,
+!:mime	audio/x-dec-basic
 >12	lelong		6		32-bit IEEE floating point,
+!:mime	audio/x-dec-basic
 >12	lelong		7		64-bit IEEE floating point,
+!:mime	audio/x-dec-basic
 >12	belong		8		Fragmented sample data,
 >12	belong		10		DSP program,
 >12	belong		11		8-bit fixed point,
@@ -57,6 +72,7 @@
 >12	belong		20		16-bit linear with emphasis and compression,
 >12	belong		21		Music kit DSP commands,
 >12	lelong		23		8-bit ISDN mu-law compressed (CCITT G.721 ADPCM voice enc.),
+!:mime	audio/x-dec-basic
 >12	belong		24		compressed (8-bit CCITT G.722 ADPCM)
 >12	belong		25		compressed (3-bit CCITT G.723.3 ADPCM),
 >12	belong		26		compressed (5-bit CCITT G.723.5 ADPCM),
@@ -68,6 +84,7 @@
 
 # Creative Labs AUDIO stuff
 0	string	MThd			Standard MIDI data
+!:mime	audio/midi
 >8 	beshort	x			(format %d)
 >10	beshort	x			using %d track
 >10	beshort		>1		\bs
@@ -75,8 +92,11 @@
 >12	beshort&0x8000	>0		SMPTE
 
 0	string	CTMF			Creative Music (CMF) data
+!:mime	audio/x-unknown
 0	string	SBI			SoundBlaster instrument data
+!:mime	audio/x-unknown
 0	string	Creative\ Voice\ File	Creative Labs voice data
+!:mime	audio/x-unknown
 # is this next line right?  it came this way...
 >19	byte	0x1A
 >23	byte	>0			- version %d
@@ -97,7 +117,13 @@
 
 # Real Audio (Magic .ra\0375)
 0	belong		0x2e7261fd	RealAudio sound file
+!:mime	audio/x-pn-realaudio
 0	string		.RMF		RealMedia file
+!:mime	application/vnd.rn-realmedia
+#video/x-pn-realvideo
+#video/vnd.rn-realvideo
+#application/vnd.rn-realmedia
+#	sigh, there are many mimes for that but the above are the most common.
 
 # MTM/669/FAR/S3M/ULT/XM format checking [Aaron Eppert, aeppert@dialin.ind.net]
 # Oct 31, 1995
@@ -120,44 +146,84 @@
 0	string		GF1PATCH110\0ID#000002\0	GUS patch
 0	string		GF1PATCH100\0ID#000002\0	Old GUS	patch
 
+# mime types according to http://www.geocities.com/nevilo/mod.htm:
+#	audio/it	.it
+#	audio/x-zipped-it	.itz
+#	audio/xm	fasttracker modules
+#	audio/x-s3m	screamtracker modules
+#	audio/s3m	screamtracker modules
+#	audio/x-zipped-mod	mdz
+#	audio/mod	mod
+#	audio/x-mod	All modules (mod, s3m, 669, mtm, med, xm, it, mdz, stm, itz, xmz, s3z)
+
 #
 # Taken from loader code from mikmod version 2.14
 # by Steve McIntyre (stevem@chiark.greenend.org.uk)
 # <doj@cubic.org> added title printing on 2003-06-24
 0	string	MAS_UTrack_V00
 >14	string	>/0		ultratracker V1.%.1s module sound data
+!:mime	audio/x-mod
+#audio/x-tracker-module
 
 0	string	UN05		MikMod UNI format module sound data
 
 0	string	Extended\ Module: Fasttracker II module sound data
+!:mime	audio/x-mod
+#audio/x-tracker-module
 >17	string	>\0		Title: "%s"
 
 21	string/c	=!SCREAM!	Screamtracker 2 module sound data
+!:mime	audio/x-mod
+#audio/x-screamtracker-module
 21	string	BMOD2STM	Screamtracker 2 module sound data
+!:mime	audio/x-mod
+#audio/x-screamtracker-module
 1080	string	M.K.		4-channel Protracker module sound data
+!:mime	audio/x-mod
+#audio/x-protracker-module
 >0	string	>\0		Title: "%s"
 1080	string	M!K!		4-channel Protracker module sound data
+!:mime	audio/x-mod
+#audio/x-protracker-module
 >0	string	>\0		Title: "%s"
 1080	string	FLT4		4-channel Startracker module sound data
+!:mime	audio/x-mod
+#audio/x-startracker-module
 >0	string	>\0		Title: "%s"
 1080	string	FLT8		8-channel Startracker module sound data
+!:mime	audio/x-mod
+#audio/x-startracker-module
 >0	string	>\0		Title: "%s"
 1080	string	4CHN		4-channel Fasttracker module sound data
+!:mime	audio/x-mod
+#audio/x-fasttracker-module
 >0	string	>\0		Title: "%s"
 1080	string	6CHN		6-channel Fasttracker module sound data
+!:mime	audio/x-mod
+#audio/x-fasttracker-module
 >0	string	>\0		Title: "%s"
 1080	string	8CHN		8-channel Fasttracker module sound data
+!:mime	audio/x-mod
+#audio/x-fasttracker-module
 >0	string	>\0		Title: "%s"
 1080	string	CD81		8-channel Octalyser module sound data
+!:mime	audio/x-mod
+#audio/x-octalysertracker-module
 >0	string	>\0		Title: "%s"
-1080	string	OKTA		8-channel Oktalyzer module sound data
+1080	string	OKTA		8-channel Octalyzer module sound data
+!:mime	audio/x-mod
+#audio/x-octalysertracker-module
 >0	string	>\0		Title: "%s"
 # Not good enough.
 #1082	string	CH
 #>1080	string	>/0		%.2s-channel Fasttracker "oktalyzer" module sound data
 1080	string	16CN		16-channel Taketracker module sound data
+!:mime	audio/x-mod
+#audio/x-taketracker-module
 >0	string	>\0		Title: "%s"
 1080	string	32CN		32-channel Taketracker module sound data
+!:mime	audio/x-mod
+#audio/x-taketracker-module
 >0	string	>\0		Title: "%s"
 
 # TOC sound files -Trevor Johnson <trevor@jpj.net>
@@ -222,9 +288,11 @@
 # SGI SoundTrack <mpruett@sgi.com>
 0	string		_SGI_SoundTrack		SGI SoundTrack project file
 # ID3 version 2 tags <waschk@informatik.uni-rostock.de>
-0	string		ID3	Audio file with ID3 version 2
->3	ubyte	<0xff	\b%d.
->4	ubyte	<0xff	\b%d tag
+0	string		ID3	Audio file with ID3 version 2.
+# ??? Normally such a file is an MP3 file, but this will give false positives
+!:mime	audio/mpeg
+>3	ubyte	<0xff	\b%d
+#>4	ubyte	<0xff	\b%d tag
 >2584	string	fLaC		\b, FLAC encoding
 >>2588	byte&0x7f		>0		\b, unknown version
 >>2588	byte&0x7f		0		\b
@@ -271,6 +339,7 @@
 
 # Impulse tracker module (audio/x-it)
 0	string		IMPM		Impulse Tracker module sound data -
+!:mime	audio/x-mod
 >4	string		>\0		"%s"
 >40	leshort		!0		compatible w/ITv%x
 >42	leshort		!0		created w/ITv%x
@@ -381,6 +450,7 @@
 # Free lossless audio codec <http://flac.sourceforge.net>
 # From: Przemyslaw Augustyniak <silvathraec@rpg.pl>
 0	string			fLaC		FLAC audio bitstream data
+!:mime	audio/x-flac
 >4	byte&0x7f		>0		\b, unknown version
 >4	byte&0x7f		0		\b
 # some common bits/sample values
@@ -503,9 +573,9 @@
 # .preset
 0	string	[Equalizer\ preset]	XMMS equalizer preset
 # .m3u
-0	string	#EXTM3U			M3U playlist
+0	search/1	#EXTM3U 	M3U playlist text
 # .pls
-0	string	[playlist]		PLS playlist
+0	search/1	[playlist]	PLS playlist text
 # licq.conf
 1	string	[licq]			LICQ configuration file
 
@@ -554,6 +624,14 @@
 # http://www.wx800.com/msg/download/irda/iMelody.pdf
 0	string	BEGIN:IMELODY	iMelody Ringtone Format
 
-# From: Matthew Flaschen <matthew.flaschen@gatech.edu>
-0	string	#EXTM3U 	M3U playlist text
+# From: "Mateus Caruccio" <mateus@caruccio.com>
+# guitar pro v3,4,5 from http://filext.com/file-extension/gp3
+0	string	\030FICHIER\ GUITAR\ PRO\ v3.	Guitar Pro Ver. 3 Tablature
+
+# From: "Leslie P. Polzer" <leslie.polzer@gmx.net>
+60	string	SONG		SoundFX Module sound file
 
+# Type: Adaptive Multi-Rate Codec
+# URL:  http://filext.com/detaillist.php?extdetail=AMR
+# From: Russell Coker <russell@coker.com.au>
+0	string	#!AMR		Adaptive Multi-Rate Codec (GSM telephony)
diff --git a/usr.bin/file/magdir/blender b/usr.bin/file/magdir/blender
index 5dc3a3da9be..a78af50bb3d 100644
--- a/usr.bin/file/magdir/blender
+++ b/usr.bin/file/magdir/blender
@@ -1,19 +1,39 @@
-#	$OpenBSD: blender,v 1.1 2004/06/03 03:14:19 tedu Exp $
+#	$OpenBSD: blender,v 1.2 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
-# blender: file(1) magic for Blender 3D data files
+# blender: file(1) magic for Blender 3D related files
 #
-# Coded by Guillermo S. Romero <gsromero@alumnos.euitt.upm.es> using the
-# data from Ton Roosendaal <ton@blender.nl>. Ton or his company do not
-# support the rule, so mail GSR if problems with it. Rule version: 1.1.
-# You can get latest version with comments and details about the format
-# at http://acd.asoc.euitt.upm.es/~gsromero/3d/blender/magic.blender
+# Native format rule v1.2. For questions use the developers list 
+# http://lists.blender.org/mailman/listinfo/bf-committers
+# GLOB chunk was moved near start and provides subversion info since 2.42 
 
-0	string	=BLENDER	Blender3D,
->7	string	=_		saved as 32-bits
->7      string	=-		saved as 64-bits
->8	string	=v		little endian
->8	string	=V		big endian
->9	byte	x		with version %c.
->10	byte	x		\b%c
->11	byte	x		\b%c
+0		string	=BLENDER	Blender3D,
+>7		string	=_		saved as 32-bits
+>>8		string	=v		little endian
+>>>9		byte	x		with version %c.
+>>>10		byte	x		\b%c
+>>>11		byte	x		\b%c
+>>>0x40		string	=GLOB		\b.
+>>>>0x58	leshort	x		\b%.4d
+>>8		string	=V		big endian
+>>>9		byte	x		with version %c.
+>>>10		byte	x		\b%c
+>>>11		byte	x		\b%c
+>>>0x40		string	=GLOB		\b.
+>>>>0x58	beshort	x		\b%.4d
+>7		string	=-		saved as 64-bits
+>>8		string	=v		little endian
+>>9		byte	x		with version %c.
+>>10		byte	x		\b%c
+>>11		byte	x		\b%c
+>>0x44		string	=GLOB		\b.
+>>>0x60		leshort	x		\b%.4d
+>>8		string	=V		big endian
+>>>9		byte	x		with version %c.
+>>>10		byte	x		\b%c
+>>>11		byte	x		\b%c
+>>>0x44		string	=GLOB		\b.
+>>>>0x60	beshort	x		\b%.4d
+
+# Scripts that run in the embeded Python interpreter
+0		string	#!BPY		Blender3D BPython script
diff --git a/usr.bin/file/magdir/c-lang b/usr.bin/file/magdir/c-lang
index c3c1a2188f1..7211af3e073 100644
--- a/usr.bin/file/magdir/c-lang
+++ b/usr.bin/file/magdir/c-lang
@@ -1,4 +1,4 @@
-#	$OpenBSD: c-lang,v 1.4 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: c-lang,v 1.5 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # c-lang:  file(1) magic for C programs (or REXX)
@@ -7,11 +7,15 @@
 # XPM icons (Greg Roelofs, newt@uchicago.edu)
 # if you uncomment "/*" for C/REXX below, also uncomment this entry
 #0	string		/*\ XPM\ */	X pixmap image data
+#!:mime	image/x-xpmi
+
+# 3DS (3d Studio files) Conflicts with diff output 0x3d '='
+#16	beshort		0x3d3d		image/x-3ds
 
 # this first will upset you if you're a PL/1 shop...
 # in which case rm it; ascmagic will catch real C programs
-#0	string		/*		C or REXX program text
-#0	string		//		C++ program text
+#0	search/1	/*		C or REXX program text
+#0	search/1	//		C++ program text
 
 # From: Mikhail Teterin <mi@aldan.algebra.com> 
 0	string		cscope		cscope reference data
@@ -21,5 +25,5 @@
 # The inverted index functionality was added some time betwen
 # versions 11 and 15, so look for -q if version is above 14:
 >7	string		>14
->>10	regex		.+\ -q\ 	with inverted index
->10	regex		.+\ -c\ 	text (non-compressed)
+>>10	search/100	\ -q\ 		with inverted index
+>10	search/100	\ -c\ 		text (non-compressed)
diff --git a/usr.bin/file/magdir/cafebabe b/usr.bin/file/magdir/cafebabe
index 8adeb304369..b48652316c0 100644
--- a/usr.bin/file/magdir/cafebabe
+++ b/usr.bin/file/magdir/cafebabe
@@ -1,4 +1,4 @@
-#	$OpenBSD: cafebabe,v 1.2 2008/05/14 20:45:35 chl Exp $
+#	$OpenBSD: cafebabe,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # Cafe Babes unite!
@@ -14,10 +14,14 @@
 # (and use as a hack). Let's not use 18, because the Mach-O people
 # might add another one or two as time goes by...
 #
-0	belong		0xcafebabe
+0	beshort		0xcafe
+>2	beshort         0xbabe
+!:mime	application/x-java-applet
 >4	belong		>30		compiled Java class data,
->>6     beshort		x	        version %d.
->>4     beshort		x       	\b%d
->4	belong		1		Mach-O fat file with 1 architecture
->4	belong		>1
->>4	belong		<20		Mach-O fat file with %ld architectures
+>>6	beshort		x	        version %d.
+>>4	beshort		x       	\b%d
+>>4	belong		1		Mach-O fat file with 1 architecture
+>>4	belong		>1
+>>>4	belong		<20		Mach-O fat file with %ld architectures
+>2	beshort		0xd00d		JAR compressed with pack200
+!:mime	application/x-java-pack200
diff --git a/usr.bin/file/magdir/cddb b/usr.bin/file/magdir/cddb
index e0f555f0cde..17084c7a861 100644
--- a/usr.bin/file/magdir/cddb
+++ b/usr.bin/file/magdir/cddb
@@ -1,4 +1,4 @@
-#	$OpenBSD: cddb,v 1.1 2004/06/03 03:14:19 tedu Exp $
+#	$OpenBSD: cddb,v 1.2 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # CDDB: file(1) magic for CDDB(tm) format CD text data files
@@ -9,4 +9,4 @@
 # CDDB-enabled CD player applications.
 #
 
-0       string/b        #\040xmcd               CDDB(tm) format CD text data
+0	search/1/b	#\040xmcd	CDDB(tm) format CD text data
diff --git a/usr.bin/file/magdir/clarion b/usr.bin/file/magdir/clarion
new file mode 100644
index 00000000000..5f5f6e75613
--- /dev/null
+++ b/usr.bin/file/magdir/clarion
@@ -0,0 +1,26 @@
+
+#------------------------------------------------------------------------------
+# clarion:  file(1) magic for # Clarion Personal/Professional Developer
+# (v2 and above)
+# From: Julien Blache <jb@jblache.org>
+
+# Database files
+# signature
+0	leshort	0x3343	Clarion Developer (v2 and above) data file
+# attributes
+>2	leshort	&0x0001	\b, locked
+>2	leshort	&0x0004	\b, encrypted
+>2	leshort	&0x0008	\b, memo file exists
+>2	leshort	&0x0010	\b, compressed
+>2	leshort	&0x0040	\b, read only
+# number of records
+>5	lelong	x	\b, %ld records
+
+# Memo files
+0	leshort	0x334d	Clarion Developer (v2 and above) memo data
+
+# Key/Index files
+# No magic? :(
+
+# Help files
+0	leshort	0x49e0	Clarion Developer (v2 and above) help data
diff --git a/usr.bin/file/magdir/commands b/usr.bin/file/magdir/commands
index e03324b7f00..7748ce30413 100644
--- a/usr.bin/file/magdir/commands
+++ b/usr.bin/file/magdir/commands
@@ -1,42 +1,64 @@
-#	$OpenBSD: commands,v 1.4 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: commands,v 1.5 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # commands:  file(1) magic for various shells and interpreters
 #
-0	string		:			shell archive or script for antique kernel text
+#0	string		:			shell archive or script for antique kernel text
 0	string/b	#!\ /bin/sh		Bourne shell script text executable
+!:mime	application/x-shellscript
 0	string/b	#!\ /bin/csh		C shell script text executable
+!:mime	application/x-shellscript
 # korn shell magic, sent by George Wu, gwu@clyde.att.com
 0	string/b	#!\ /bin/ksh		Korn shell script text executable
+!:mime	application/x-shellscript
 0	string/b 	#!\ /bin/tcsh		Tenex C shell script text executable
+!:mime	application/x-shellscript
 0	string/b 	#!\ /usr/local/tcsh	Tenex C shell script text executable
+!:mime	application/x-shellscript
 0	string/b	#!\ /usr/local/bin/tcsh	Tenex C shell script text executable
+!:mime	application/x-shellscript
 
 #
 # zsh/ash/ae/nawk/gawk magic from cameron@cs.unsw.oz.au (Cameron Simpson)
 0	string/b	#!\ /bin/zsh		Paul Falstad's zsh script text executable
+!:mime	application/x-shellscript
 0	string/b	#!\ /usr/bin/zsh	Paul Falstad's zsh script text executable
+!:mime	application/x-shellscript
 0	string/b	#!\ /usr/local/bin/zsh	Paul Falstad's zsh script text executable
+!:mime	application/x-shellscript
 0	string/b	#!\ /usr/local/bin/ash	Neil Brown's ash script text executable
+!:mime	application/x-shellscript
 0	string/b	#!\ /usr/local/bin/ae	Neil Brown's ae script text executable
+!:mime	application/x-shellscript
 0	string/b	#!\ /bin/nawk		new awk script text executable
+!:mime	application/x-nawk
 0	string/b	#!\ /usr/bin/nawk	new awk script text executable
+!:mime	application/x-nawk
 0	string/b	#!\ /usr/local/bin/nawk	new awk script text executable
+!:mime	application/x-nawk
 0	string/b	#!\ /bin/gawk		GNU awk script text executable
+!:mime	application/x-gawk
 0	string/b	#!\ /usr/bin/gawk	GNU awk script text executable
+!:mime	application/x-gawk
 0	string/b	#!\ /usr/local/bin/gawk	GNU awk script text executable
+!:mime	application/x-gawk
 #
 0	string/b	#!\ /bin/awk		awk script text executable
+!:mime	application/x-awk
 0	string/b	#!\ /usr/bin/awk	awk script text executable
+!:mime	application/x-awk
 # update to distinguish from *.vcf files
-0	regex		BEGIN[[:space:]]*[{]	awk script text
+# this is broken because postscript has /EBEGIN{ for example.
+#0	search/Bb	BEGIN {			awk script text
 
 # AT&T Bell Labs' Plan 9 shell
 0	string/b	#!\ /bin/rc	Plan 9 rc shell script text executable
 
 # bash shell magic, from Peter Tobias (tobias@server.et-inf.fho-emden.de)
 0	string/b	#!\ /bin/bash	Bourne-Again shell script text executable
+!:mime	application/x-shellscript
 0	string/b	#!\ /usr/local/bin/bash	Bourne-Again shell script text executable
+!:mime	application/x-shellscript
 
 # using env
 0	string		#!/usr/bin/env		a
@@ -46,11 +68,11 @@
 
 # PHP scripts
 # Ulf Harnhammar <ulfh@update.uu.se>
-0	string/c	=<?php			PHP script text
-0	string		=<?\n			PHP script text
-0	string		=<?\r			PHP script text
-0	string/b	#!\ /usr/local/bin/php	PHP script text executable
-0	string/b	#!\ /usr/bin/php	PHP script text executable
+0	search/1/c	=<?php			PHP script text
+0	search/1	=<?\n			PHP script text
+0	search/1	=<?\r			PHP script text
+0	search/1/b	#!\ /usr/local/bin/php	PHP script text executable
+0	search/1/b	#!\ /usr/bin/php	PHP script text executable
 
 0	string		Zend\x00		PHP script Zend Optimizer data
 
diff --git a/usr.bin/file/magdir/compress b/usr.bin/file/magdir/compress
index 2a6809409ec..01e70249487 100644
--- a/usr.bin/file/magdir/compress
+++ b/usr.bin/file/magdir/compress
@@ -1,4 +1,4 @@
-#	$OpenBSD: compress,v 1.5 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: compress,v 1.6 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # compress:  file(1) magic for pure-compression formats (no archives)
@@ -10,7 +10,8 @@
 # because it tries to uncompress it to figure out what's inside.
 
 # standard unix compress
-0	string		\037\235	compress'd data
+#0	string		\037\235	compress'd data
+#!:mime	application/x-compress
 >2	byte&0x80	>0		block compressed
 >2	byte&0x1f	x		%d bits
 
@@ -20,6 +21,7 @@
 #	* Produce shorter output - notably, only report compression methods
 #	  other than 8 ("deflate", the only method defined in RFC 1952).
 0       string          \037\213        gzip compressed data
+!:mime	application/x-gzip
 >2	byte		<8		\b, reserved method
 >2	byte		>8		\b, unknown method
 >3	byte		&0x01		\b, ASCII
@@ -50,23 +52,29 @@
 
 # packed data, Huffman (minimum redundancy) codes on a byte-by-byte basis
 0	string		\037\036	packed data
+!:mime	application/octet-stream
 >2	belong		>1		\b, %d characters originally
 >2	belong		=1		\b, %d character originally
 #
 # This magic number is byte-order-independent.
 0	short		0x1f1f		old packed data
+!:mime	application/octet-stream
 
 # XXX - why *two* entries for "compacted data", one of which is
 # byte-order independent, and one of which is byte-order dependent?
 #
 0	short		0x1fff		compacted data
+!:mime	application/octet-stream
 # This string is valid for SunOS (BE) and a matching "short" is listed
 # in the Ultrix (LE) magic file.
 0	string		\377\037	compacted data
+!:mime	application/octet-stream
 0	short		0145405		huf output
+!:mime	application/octet-stream
 
 # bzip2
 0	string		BZh		bzip2 compressed data
+!:mime	application/x-bzip2
 >3	byte		>47		\b, block size = %c00k
 
 # squeeze and crunch
@@ -103,17 +111,17 @@
 # bzip	a block-sorting file compressor
 #	by Julian Seward <sewardj@cs.man.ac.uk> and others
 #
-0	string		BZ		bzip compressed data
->2	byte		x		\b, version: %c
->3	string		=1		\b, compression block size 100k
->3	string		=2		\b, compression block size 200k
->3	string		=3		\b, compression block size 300k
->3	string		=4		\b, compression block size 400k
->3	string		=5		\b, compression block size 500k
->3	string		=6		\b, compression block size 600k
->3	string		=7		\b, compression block size 700k
->3	string		=8		\b, compression block size 800k
->3	string		=9		\b, compression block size 900k
+#0	string		BZ		bzip compressed data
+#>2	byte		x		\b, version: %c
+#>3	string		=1		\b, compression block size 100k
+#>3	string		=2		\b, compression block size 200k
+#>3	string		=3		\b, compression block size 300k
+#>3	string		=4		\b, compression block size 400k
+#>3	string		=5		\b, compression block size 500k
+#>3	string		=6		\b, compression block size 600k
+#>3	string		=7		\b, compression block size 700k
+#>3	string		=8		\b, compression block size 800k
+#>3	string		=9		\b, compression block size 900k
 
 # lzop from <markus.oberhumer@jk.uni-linz.ac.at>
 0	string		\x89\x4c\x5a\x4f\x00\x0d\x0a\x1a\x0a	lzop compressed data
@@ -175,6 +183,13 @@
 >6	byte		x			version %d
 >7	byte		x			\b.%d
 
+# Type: LZMA
+# URL:  http://www.7-zip.org/sdk.html
+# From: Robert Millan <rmh@aybabtu.com> and Reuben Thomas <rrt@sc3d.org>
+# Commented out because apparently not reliable (according to Debian
+# bug #364260)
+#0	string		]\000\000\200\000	LZMA compressed data
+
 # AFX compressed files (Wolfram Kleff)
 2	string		-afx-		AFX compressed file data
 
diff --git a/usr.bin/file/magdir/console b/usr.bin/file/magdir/console
index c413d64dcce..e61d1820ba5 100644
--- a/usr.bin/file/magdir/console
+++ b/usr.bin/file/magdir/console
@@ -1,4 +1,4 @@
-#	$OpenBSD: console,v 1.2 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: console,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # Console game magic
@@ -193,3 +193,64 @@
 0	string	PPF10			Playstation Patch File version 1.0
 >5	byte	0			\b, Simple Encoding
 >6	string	x			\b, description: %s
+
+# From: Daniel Dawson <ddawson@icehouse.net>
+# SNES9x .smv "movie" file format.
+0		string		SMV\x1A	SNES9x input recording
+>0x4		lelong		x	\b, version %d
+# version 4 is latest so far 
+>0x4		lelong		<5
+>>0x8		ledate		x	\b, recorded at %s
+>>0xc		lelong		>0	\b, rerecorded %d times
+>>0x10		lelong		x	\b, %d frames long
+>>0x14		byte		>0	\b, data for controller(s):
+>>>0x14		byte		&0x1	#1
+>>>0x14		byte		&0x2	#2
+>>>0x14		byte		&0x4	#3
+>>>0x14		byte		&0x8	#4
+>>>0x14		byte		&0x10	#5
+>>0x15		byte		^0x1	\b, begins from snapshot
+>>0x15		byte		&0x1	\b, begins from reset
+>>0x15		byte		^0x2	\b, NTSC standard
+>>0x15		byte		&0x2	\b, PAL standard
+>>0x17		byte		&0x1    \b, settings:
+# WIP1Timing not used as of version 4
+>>>0x4		lelong		<4
+>>>>0x17	byte		&0x2	WIP1Timing
+>>>0x17		byte		&0x4	Left+Right
+>>>0x17		byte		&0x8	VolumeEnvX
+>>>0x17		byte		&0x10	FakeMute
+>>>0x17		byte		&0x20	SyncSound
+# New flag as of version 4
+>>>0x4		lelong		>3
+>>>>0x17	byte		&0x80	NoCPUShutdown
+>>0x4		lelong		<4
+>>>0x18		lelong		>0x23
+>>>>0x20	leshort		!0
+>>>>>0x20	lestring16	x	\b, metadata: "%s"
+>>0x4		lelong		>3
+>>>0x24		byte		>0	\b, port 1:
+>>>>0x24	byte		1	joypad
+>>>>0x24	byte		2	mouse
+>>>>0x24	byte		3	SuperScope
+>>>>0x24	byte		4	Justifier
+>>>>0x24	byte		5	multitap
+>>>0x24		byte		>0	\b, port 2:
+>>>>0x25	byte		1	joypad
+>>>>0x25	byte		2	mouse
+>>>>0x25	byte		3	SuperScope
+>>>>0x25	byte		4	Justifier
+>>>>0x25	byte		5	multitap
+>>>0x18		lelong		>0x43
+>>>>0x40	leshort		!0
+>>>>>0x40	lestring16	x	\b, metadata: "%s"
+>>0x17		byte		&0x40   \b, ROM:
+>>>(0x18.l-26)	lelong		x	CRC32 0x%08x
+>>>(0x18.l-23)	string		x	"%s"
+
+# From: "Nelson A. de Oliveira" <naoliv@gmail.com>
+# .w3g
+0	string Warcraft\ III\ recorded\ game	%s
+# .w3m
+0	string HM3W				Warcraft III map file
+
diff --git a/usr.bin/file/magdir/cracklib b/usr.bin/file/magdir/cracklib
index 973dfc3e7cc..75e27491fd0 100644
--- a/usr.bin/file/magdir/cracklib
+++ b/usr.bin/file/magdir/cracklib
@@ -1,4 +1,4 @@
-#	$OpenBSD: cracklib,v 1.2 2008/05/14 20:45:35 chl Exp $
+#	$OpenBSD: cracklib,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # cracklib:  file (1) magic for cracklib v2.7
@@ -10,5 +10,5 @@
 0	belong	0x70775631	Cracklib password index, big endian
 >4	belong	>-1		(%i words)
 # really bellong 0x0000000070775631
-4	belong	0x70775631	Cracklib password index, big endian ("64-bit")
+0	search/1	\0\0\0\0pwV1	Cracklib password index, big endian ("64-bit")
 >12	belong	>0		(%i words)
diff --git a/usr.bin/file/magdir/ctags b/usr.bin/file/magdir/ctags
index 3b64f023ef3..0f777a70cd9 100644
--- a/usr.bin/file/magdir/ctags
+++ b/usr.bin/file/magdir/ctags
@@ -1,6 +1,6 @@
-#	$OpenBSD: ctags,v 1.2 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: ctags,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 # ----------------------------------------------------------------------------
 # ctags:  file (1) magic for Exuberant Ctags files
 # From: Alexander Mai <mai@migdal.ikp.physik.tu-darmstadt.de>
-0       string  =!_TAG   Exuberant Ctags tag file text
+0	search/1	=!_TAG	Exuberant Ctags tag file text
diff --git a/usr.bin/file/magdir/database b/usr.bin/file/magdir/database
index 2faf6badb33..ea2407a560b 100644
--- a/usr.bin/file/magdir/database
+++ b/usr.bin/file/magdir/database
@@ -1,4 +1,4 @@
-#	$OpenBSD: database,v 1.5 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: database,v 1.6 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # database:  file(1) magic for various databases
@@ -10,8 +10,11 @@
 #  Will be maintained as part of the GDBM distribution in the future.
 #  <downsj@teeny.org>
 0	belong	0x13579ace	GNU dbm 1.x or ndbm database, big endian
+!:mime	application/x-gdbm
 0	lelong	0x13579ace	GNU dbm 1.x or ndbm database, little endian
+!:mime	application/x-gdbm
 0	string	GDBM		GNU dbm 2.x database
+!:mime	application/x-gdbm
 #
 # Berkeley DB
 #
@@ -22,6 +25,7 @@
 # Hash and Btree 2.X and later databases store the metadata in host byte order.
 
 0	long	0x00061561	Berkeley DB
+!:mime	application/x-dbm
 >8	belong	4321
 >>4	belong	>2		1.86
 >>4	belong	<3		1.85
@@ -108,74 +112,87 @@
 #>>0x04	  byte		6	non-incrementing secondary index .XGn file 
 #>>0x04	  byte		7	secondary index .YGn file 
 #>>>0x04	  byte		8	incrementing secondary index .XGn file 
+
 ## XBase database files
 #0      byte       0x02	
 #>8     leshort	  >0
 #>>12   leshort    0	FoxBase 
+#!:mime	application/x-dbf
 #>>>0x04	lelong		0		(no records)
 #>>>0x04	lelong		>0		(%ld records)
 #
 #0      byte       0x03	
+#!:mime	application/x-dbf
 #>8     leshort	  >0
 #>>12   leshort    0	FoxBase+, FoxPro, dBaseIII+, dBaseIV, no memo 
 #>>>0x04	lelong		0		(no records)
 #>>>0x04	lelong		>0		(%ld records)
 #
 #0      byte       0x04	
+#!:mime	application/x-dbf
 #>8     leshort	  >0
 #>>12   leshort    0	dBASE IV no memo file 
 #>>>0x04	lelong		0		(no records)
 #>>>0x04	lelong		>0		(%ld records)
 #
 #0      byte       0x05	
+#!:mime	application/x-dbf
 #>8     leshort	  >0
 #>>12   leshort    0	dBASE V no memo file 
 #>>>0x04	lelong		0		(no records)
 #>>>0x04	lelong		>0		(%ld records)
 #
 #0      byte       0x30
+#!:mime	application/x-dbf
 #>8     leshort	  >0
 #>>12   leshort    0	Visual FoxPro 
 #>>>0x04	lelong		0		(no records)
 #>>>0x04	lelong		>0		(%ld records)
 #
 #0      byte       0x43
+#!:mime	application/x-dbf
 #>8     leshort	  >0
 #>>12   leshort    0	FlagShip with memo var size 
 #>>>0x04	lelong		0		(no records)
 #>>>0x04	lelong		>0		(%ld records)
 #
 #0      byte       0x7b
+#!:mime	application/x-dbf
 #>8     leshort	  >0
 #>>12   leshort    0	dBASEIV with memo 
 #>>>0x04	lelong		0		(no records)
 #>>>0x04	lelong		>0		(%ld records)
 #
 #0      byte       0x83	
+#!:mime	application/x-dbf
 #>8     leshort	  >0
 #>>12   leshort    0	FoxBase+, dBaseIII+ with memo 
 #>>>0x04	lelong		0		(no records)
 #>>>0x04	lelong		>0		(%ld records)
 #
 #0      byte       0x8b
+#!:mime	application/x-dbf
 #>8     leshort	  >0
 #>>12   leshort    0	dBaseIV with memo 
 #>>>0x04	lelong		0		(no records)
 #>>>0x04	lelong		>0		(%ld records)
 #
 #0      byte       0x8e	
+#!:mime	application/x-dbf
 #>8     leshort	  >0
 #>>12   leshort    0	dBaseIV with SQL Table 
 #>>>0x04	lelong		0		(no records)
 #>>>0x04	lelong		>0		(%ld records)
 #
 #0      byte       0xb3
+#!:mime	application/x-dbf
 #>8     leshort	  >0
 #>>12   leshort    0	FlagShip with .dbt memo 
 #>>>0x04	lelong		0		(no records)
 #>>>0x04	lelong		>0		(%ld records)
 #
 #0      byte       0xf5
+#!:mime	application/x-dbf
 #>8     leshort	  >0
 #>>12   leshort    0	FoxPro with memo 
 #>>>0x04	lelong		0		(no records)
@@ -184,7 +201,8 @@
 #0	leshort		0x0006		DBase 3 index file
 
 # MS Access database
-4	  string	Standard\ Jet\ DB	Microsoft Access Database
+4	string	Standard\ Jet\ DB	Microsoft Access Database
+!:mime	application/x-msaccess
 
 # TDB database from Samba et al - Martin Pool <mbp@samba.org>
 0	string	TDB\ file		TDB database
@@ -211,3 +229,17 @@
 16	string		MIT-MAGIC-COOKIE-1	X11 Xauthority data
 17	string		MIT-MAGIC-COOKIE-1	X11 Xauthority data
 18	string		MIT-MAGIC-COOKIE-1	X11 Xauthority data
+
+# From: Maxime Henrion <mux@FreeBSD.org>
+# PostgreSQL's custom dump format, Maxime Henrion <mux@FreeBSD.org>
+0	string		PGDMP		PostgreSQL custom database dump
+>5	byte		x		- v%d
+>6	byte		x		\b.%d
+>5	beshort		<=0x100		\b-0
+>5	beshort		>0x100
+>>7	byte		x		\b-%d
+
+# Type: Advanced Data Format (ADF) database
+# URL:  http://www.grc.nasa.gov/WWW/cgns/adf/
+# From: Nicolas Chauvat <nicolas.chauvat@logilab.fr>
+0	string	@(#)ADF\ Database	CGNS Advanced Data Format
diff --git a/usr.bin/file/magdir/diff b/usr.bin/file/magdir/diff
index 7e951a1b2a3..8d5f889bc89 100644
--- a/usr.bin/file/magdir/diff
+++ b/usr.bin/file/magdir/diff
@@ -1,11 +1,11 @@
-#	$OpenBSD: diff,v 1.3 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: diff,v 1.4 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # diff:  file(1) magic for diff(1) output
 #
-0	string		diff\ 	'diff' output text
-0	string		***\ 		'diff' output text
-0	string		Only\ in\ 	'diff' output text
-0	string		Common\ subdirectories:\ 	'diff' output text
+0	search/1	diff\ 		diff output text
+0	search/1	***\ 		diff output text
+0	search/1	Only\ in\ 	diff output text
+0	search/1	Common\ subdirectories:\ 	diff output text
 
-0	string		Index:		RCS/CVS diff output text
+0	search/1	Index:		RCS/CVS diff output text
diff --git a/usr.bin/file/magdir/dump b/usr.bin/file/magdir/dump
index 57af29eb764..ae5ed1673a3 100644
--- a/usr.bin/file/magdir/dump
+++ b/usr.bin/file/magdir/dump
@@ -1,4 +1,4 @@
-#	$OpenBSD: dump,v 1.4 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: dump,v 1.5 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # dump:  file(1) magic for dump file format--for new and old dump filesystems
@@ -92,3 +92,41 @@
 >0	leshort	5		end of volume.
 >0	leshort	6		map of inodes deleted.
 >0	leshort	7		end of medium (for floppy).
+
+24	belong	0x19540119	new-fs dump file (ufs2, big endian),
+>896	beqdate	x		Previous dump %s,
+>904	beqdate	x		This dump %s,
+>12	belong	>0		Volume %ld,
+>692	belong	0		Level zero, type:
+>692	belong	>0		Level %d, type:
+>0	belong	1		tape header,
+>0	belong	2		beginning of file record,
+>0	belong	3		map of inodes on tape,
+>0	belong	4		continuation of file record,
+>0	belong	5		end of volume,
+>0	belong	6		map of inodes deleted,
+>0	belong	7		end of medium (for floppy),
+>676	string	>\0		Label %s,
+>696	string	>\0		Filesystem %s,
+>760	string	>\0		Device %s,
+>824	string	>\0		Host %s,
+>888	belong	>0		Flags %x
+
+24	lelong	0x19540119	new-fs dump file (ufs2, little endian),
+>896	leqdate	x		This dump %s,
+>904	leqdate	x		Previous dump %s,
+>12	lelong	>0		Volume %ld,
+>692	lelong	0		Level zero, type:
+>692	lelong	>0		Level %d, type:
+>0	lelong	1		tape header,
+>0	lelong	2		beginning of file record,
+>0	lelong	3		map of inodes on tape,
+>0	lelong	4		continuation of file record,
+>0	lelong	5		end of volume,
+>0	lelong	6		map of inodes deleted,
+>0	lelong	7		end of medium (for floppy),
+>676	string	>\0		Label %s,
+>696	string	>\0		Filesystem %s,
+>760	string	>\0		Device %s,
+>824	string	>\0		Host %s,
+>888	lelong	>0		Flags %x
diff --git a/usr.bin/file/magdir/efi b/usr.bin/file/magdir/efi
new file mode 100644
index 00000000000..4eec2eb9377
--- /dev/null
+++ b/usr.bin/file/magdir/efi
@@ -0,0 +1,14 @@
+
+#------------------------------------------------------------------------------
+# efi:  file(1) magic for Universal EFI binaries
+
+0	lelong	0x0ef1fab9
+>4	lelong	1		Universal EFI binary with 1 architecture
+>>&0	lelong	7		\b, i386
+>>&0	lelong	0x01000007	\b, x86_64
+>4	lelong	2		Universal EFI binary with 2 architectures
+>>&0	lelong	7		\b, i386
+>>&0	lelong	0x01000007	\b, x86_64
+>>&20	lelong	7		\b, i386
+>>&20	lelong	0x01000007	\b, x86_64
+>4	lelong	>2		Universal EFI binary with %ld architectures
diff --git a/usr.bin/file/magdir/elf b/usr.bin/file/magdir/elf
index 304d27a85ac..1058ab4c225 100644
--- a/usr.bin/file/magdir/elf
+++ b/usr.bin/file/magdir/elf
@@ -1,4 +1,4 @@
-#	$OpenBSD: elf,v 1.16 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: elf,v 1.17 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # elf:  file(1) magic for ELF executables
@@ -16,12 +16,17 @@
 >5	byte		0		invalid byte order
 >5	byte		1		LSB
 >>16	leshort		0		no file type,
+!:mime	application/octet-stream
 >>16	leshort		1		relocatable,
+!:mime	application/x-object
 >>16	leshort		2		executable,
+!:mime	application/x-executable
 >>16	leshort		3		shared object,
+!:mime	application/x-sharedlib
 # Core handling from Peter Tobias <tobias@server.et-inf.fho-emden.de>
 # corrections by Christian 'Dr. Disk' Hechelmann <drdisk@ds9.au.s.shuttle.de>
 >>16	leshort		4		core file
+!:mime	application/x-coredump
 # Core file detection is not reliable.
 #>>>(0x38+0xcc) string	>\0		of '%s'
 #>>>(0x38+0x10) lelong	>0		(signal %d),
@@ -52,10 +57,10 @@
 >>>>36  lelong&0xf0000000	0x20000000	MIPS-III
 >>>>36  lelong&0xf0000000	0x30000000	MIPS-IV
 >>>>36  lelong&0xf0000000	0x40000000	MIPS-V
->>>>36  lelong&0xf0000000	0x60000000	MIPS32
->>>>36  lelong&0xf0000000	0x70000000	MIPS64
->>>>36  lelong&0xf0000000	0x80000000	MIPS32 rel2
->>>>36  lelong&0xf0000000	0x90000000	MIPS64 rel2
+>>>>36  lelong&0xf0000000	0x50000000	MIPS32
+>>>>36  lelong&0xf0000000	0x60000000	MIPS64
+>>>>36  lelong&0xf0000000	0x70000000	MIPS32 rel2
+>>>>36  lelong&0xf0000000	0x80000000	MIPS64 rel2
 # only for 64-bit
 >>>4	byte		2
 >>>>48  lelong&0xf0000000	0x00000000	MIPS-I
@@ -63,10 +68,10 @@
 >>>>48  lelong&0xf0000000	0x20000000	MIPS-III
 >>>>48  lelong&0xf0000000	0x30000000	MIPS-IV
 >>>>48  lelong&0xf0000000	0x40000000	MIPS-V
->>>>48  lelong&0xf0000000	0x60000000	MIPS32
->>>>48  lelong&0xf0000000	0x70000000	MIPS64 
->>>>48  lelong&0xf0000000	0x80000000	MIPS32 rel2
->>>>48  lelong&0xf0000000	0x90000000	MIPS64 rel2
+>>>>48  lelong&0xf0000000	0x50000000	MIPS32
+>>>>48  lelong&0xf0000000	0x60000000	MIPS64
+>>>>48  lelong&0xf0000000	0x70000000	MIPS32 rel2
+>>>>48  lelong&0xf0000000	0x80000000	MIPS64 rel2
 >>18	leshort		9		Amdahl - invalid byte order,
 >>18	leshort		10		MIPS (deprecated),
 >>18	leshort		11		RS6000 - invalid byte order,
@@ -97,9 +102,46 @@
 >>18	leshort		51		Stanford MIPS-X,
 >>18	leshort		52		Motorola Coldfire,
 >>18	leshort		53		Motorola M68HC12,
+>>18	leshort		54		Fujitsu MMA,
+>>18	leshort		55		Siemens PCP,
+>>18	leshort		56		Sony nCPU,
+>>18	leshort		57		Denso NDR1,
+>>18	leshort		58		Start*Core,
+>>18	leshort		59		Toyota ME16,
+>>18	leshort		60		ST100,
+>>18	leshort		61		Tinyj emb.,
 >>18	leshort		62		x86-64,
+>>18	leshort		63		Sony DSP,
+>>18	leshort		66		FX66,
+>>18	leshort		67		ST9+ 8/16 bit,
+>>18	leshort		68		ST7 8 bit,
+>>18	leshort		69		MC68HC16,
+>>18	leshort		70		MC68HC11,
+>>18	leshort		71		MC68HC08,
+>>18	leshort		72		MC68HC05,
+>>18	leshort		73		SGI SVx,
+>>18	leshort		74		ST19 8 bit,
 >>18	leshort		75		Digital VAX,
+>>18	leshort		76		Axis cris,
+>>18	leshort		77		Infineon 32-bit embedded,
+>>18	leshort		78		Element 14 64-bit DSP,
+>>18	leshort		79		LSI Logic 16-bit DSP,
+>>18	leshort		80		MMIX,
+>>18	leshort		81		Harvard machine-independent,
+>>18	leshort		82		SiTera Prism,
+>>18	leshort		83		Atmel AVR 8-bit,
+>>18	leshort		84		Fujitsu FR30,
+>>18	leshort		85		Mitsubishi D10V,
+>>18	leshort		86		Mitsubishi D30V,
+>>18	leshort		87		NEC v850,
 >>18	leshort		88		Renesas M32R,
+>>18	leshort		89		Matsushita MN10300,
+>>18	leshort		90		Matsushita MN10200,
+>>18	leshort		91		picoJava,
+>>18	leshort		92		OpenRISC,
+>>18	leshort		93		ARC Cores Tangent-A5,
+>>18	leshort		0x3426		OpenRISC (obsolete),
+>>18	leshort		0x8472		OpenRISC (obsolete),
 >>18	leshort		94		Tensilica Xtensa,
 >>18	leshort		97		NatSemi 32k,
 >>18	leshort		106		Analog Devices Blackfin,
@@ -109,10 +151,15 @@
 >>36	lelong		1		MathCoPro/FPU/MAU Required
 >5	byte		2		MSB
 >>16	beshort		0		no file type,
+!:mime	application/octet-stream
 >>16	beshort		1		relocatable,
+!:mime	application/x-object
 >>16	beshort		2		executable,
+!:mime	application/x-executable
 >>16	beshort		3		shared object,
+!:mime	application/x-sharedlib
 >>16	beshort		4		core file,
+!:mime	application/x-coredump
 #>>>(0x38+0xcc) string	>\0		of '%s'
 #>>>(0x38+0x10) belong	>0		(signal %d),
 >>16	beshort		&0xff00		processor-specific,
@@ -140,10 +187,10 @@
 >>>>36  belong&0xf0000000	0x20000000	MIPS-III
 >>>>36  belong&0xf0000000	0x30000000	MIPS-IV
 >>>>36  belong&0xf0000000	0x40000000	MIPS-V
->>>>36  belong&0xf0000000	0x60000000	MIPS32
->>>>36  belong&0xf0000000	0x70000000	MIPS64
->>>>36  belong&0xf0000000	0x80000000	MIPS32 rel2
->>>>36  belong&0xf0000000	0x90000000	MIPS64 rel2
+>>>>36  belong&0xf0000000	0x50000000	MIPS32
+>>>>36  belong&0xf0000000	0x60000000	MIPS64
+>>>>36  belong&0xf0000000	0x70000000	MIPS32 rel2
+>>>>36  belong&0xf0000000	0x80000000	MIPS64 rel2
 # only for 64-bit
 >>>4	byte		2
 >>>>48	belong&0xf0000000	0x00000000	MIPS-I
@@ -151,10 +198,10 @@
 >>>>48	belong&0xf0000000	0x20000000	MIPS-III
 >>>>48	belong&0xf0000000	0x30000000	MIPS-IV
 >>>>48	belong&0xf0000000	0x40000000	MIPS-V
->>>>48	belong&0xf0000000	0x60000000	MIPS32
->>>>48	belong&0xf0000000	0x70000000	MIPS64 
->>>>48	belong&0xf0000000	0x80000000	MIPS32 rel2
->>>>48	belong&0xf0000000	0x90000000	MIPS64 rel2
+>>>>48	belong&0xf0000000	0x50000000	MIPS32
+>>>>48	belong&0xf0000000	0x60000000	MIPS64
+>>>>48	belong&0xf0000000	0x70000000	MIPS32 rel2
+>>>>48	belong&0xf0000000	0x80000000	MIPS64 rel2
 >>18	beshort		9		Amdahl,
 >>18	beshort		10		MIPS (deprecated),
 >>18	beshort		11		RS6000,
@@ -195,8 +242,12 @@
 >>18	beshort		73		Cray NV1,
 >>18	beshort		75		Digital VAX,
 >>18	beshort		88		Renesas M32R,
+>>18	leshort		92		OpenRISC,
+>>18	leshort		0x3426		OpenRISC (obsolete),
+>>18	leshort		0x8472		OpenRISC (obsolete),
 >>18	beshort		94		Tensilica Xtensa,
 >>18	beshort		97		NatSemi 32k,
+>>18	beshort		0x18ad		AVR32 (unofficial),
 >>18	beshort		0x9026		Alpha (unofficial),
 >>18	beshort		0xa390		IBM S/390 (obsolete),
 >>20	belong		0		invalid version
diff --git a/usr.bin/file/magdir/erlang b/usr.bin/file/magdir/erlang
new file mode 100644
index 00000000000..59f55ecb31b
--- /dev/null
+++ b/usr.bin/file/magdir/erlang
@@ -0,0 +1,18 @@
+
+#------------------------------------------------------------------------------
+# erlang:  file(1) magic for Erlang JAM and BEAM files
+# URL:  http://www.erlang.org/faq/x779.html#AEN812
+
+# OTP R3-R4
+0	string	\0177BEAM!	Old Erlang BEAM file
+>6	short	>0		- version %d
+
+# OTP R5 and onwards
+0	string	FOR1
+>8	string	BEAM		Erlang BEAM file
+
+# 4.2 version may have a copyright notice!
+4	string	Tue Jan 22 14:32:44 MET 1991	Erlang JAM file - version 4.2
+79	string	Tue Jan 22 14:32:44 MET 1991	Erlang JAM file - version 4.2
+
+4	string	1.0 Fri Feb 3 09:55:56 MET 1995	Erlang JAM file - version 4.3
diff --git a/usr.bin/file/magdir/filesystems b/usr.bin/file/magdir/filesystems
index 59318f5b92e..e4858017cbf 100644
--- a/usr.bin/file/magdir/filesystems
+++ b/usr.bin/file/magdir/filesystems
@@ -1,4 +1,4 @@
-#	$OpenBSD: filesystems,v 1.3 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: filesystems,v 1.4 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # filesystems:  file(1) magic for different filesystems
@@ -53,7 +53,13 @@
 >>>>11	ulelong	>0		\b, %d sectors/track
 >>>>15	ulelong	>0		\b, %d cylinders
 
+# updated by Joerg Jenderek at Sep 2007
+# only for sector sizes with 512 or more Bytes
 0x1FE	leshort	0xAA55			x86 boot sector
+# to do also for sectors < than 512 Bytes and some other files, GRR
+#30	search/481	\x55\xAA	x86 boot sector
+# not for BeOS floppy 1440k, MBRs
+#(11.s-2) uleshort	0xAA55		x86 boot sector
 >2	string	OSBS			\b, OS/BS MBR
 # J\xf6rg Jenderek <joerg dot jenderek at web dot de>
 >0x8C	string	Invalid\ partition\ table	\b, MS-DOS MBR
@@ -174,56 +180,59 @@
 >>498	string	BCDL\ \ \ \ BIN			\b, Bootable CD Loader (1.50Z)
 # mbr partion table entries
 # OEM-ID not Microsoft,SYSLINUX,or MTOOLs
->3			string		!MS	
+>3			string		!MS
 >>3			string		!SYSLINUX
 >>>3			string		!MTOOL
 # not FAT (32 bit)
 >>>>82			string		!FAT32	
 #not IO.SYS
 >>>>>472		string		!IO\ \ \ \ \ \ SYS
+>>>>>>480		string		!IO\ \ \ \ \ \ SYS
 #not Linux kernel
->>>>>>514		string		!HdrS		
+>>>>>>>514		string		!HdrS
+#not BeOS
+>>>>>>>>422		string		!Be\ Boot\ Loader
 # active flag 0 or 0x80 and type > 0
->>>>>>>446		ubyte		<0x81	
->>>>>>>>446		ubyte&0x7F	0	
+>>>>>>>>>446		ubyte		<0x81	
+>>>>>>>>>>446		ubyte&0x7F	0	
 >>>>>>>>>>>450		ubyte		>0	\b; partition 1: ID=0x%x
->>>>>>>>>>446		ubyte		0x80	\b, active
->>>>>>>>>>447		ubyte		x	\b, starthead %u
-#>>>>>>>>>>448		ubyte		x	\b, start C_S: 0x%x
-#>>>>>>>>>>448		ubeshort&1023	x	\b, startcylinder? %d
->>>>>>>>>>454		ulelong		x	\b, startsector %u
->>>>>>>>>>458		ulelong		x	\b, %u sectors
+>>>>>>>>>>>>446		ubyte		0x80	\b, active
+>>>>>>>>>>>>447		ubyte		x	\b, starthead %u
+#>>>>>>>>>>>>448		ubyte		x	\b, start C_S: 0x%x
+#>>>>>>>>>>>>448		ubeshort&1023	x	\b, startcylinder? %d
+>>>>>>>>>>>>454		ulelong		x	\b, startsector %u
+>>>>>>>>>>>>458		ulelong		x	\b, %u sectors
 #
->>>>>>>462		ubyte		<0x81	
->>>>>>>>462		ubyte&0x7F	0		
->>>>>>>>>466		ubyte		>0	\b; partition 2: ID=0x%x
->>>>>>>>>>462		ubyte		0x80	\b, active
->>>>>>>>>>463		ubyte		x	\b, starthead %u
-#>>>>>>>>>>464		ubyte		x	\b, start C_S: 0x%x
-#>>>>>>>>>>464		ubeshort&1023	x	\b, startcylinder? %d
->>>>>>>>>>470		ulelong		x	\b, startsector %u
->>>>>>>>>>474		ulelong		x	\b, %u sectors
+>>>>>>>>>462		ubyte		<0x81	
+>>>>>>>>>>462		ubyte&0x7F	0		
+>>>>>>>>>>>466		ubyte		>0	\b; partition 2: ID=0x%x
+>>>>>>>>>>>>462		ubyte		0x80	\b, active
+>>>>>>>>>>>>463		ubyte		x	\b, starthead %u
+#>>>>>>>>>>>>464		ubyte		x	\b, start C_S: 0x%x
+#>>>>>>>>>>>>464		ubeshort&1023	x	\b, startcylinder? %d
+>>>>>>>>>>>>470		ulelong		x	\b, startsector %u
+>>>>>>>>>>>>474		ulelong		x	\b, %u sectors
 #
->>>>>>>478		ubyte		<0x81		
->>>>>>>>478		ubyte&0x7F	0		
->>>>>>>>>482		ubyte		>0	\b; partition 3: ID=0x%x
->>>>>>>>>>478		ubyte		0x80	\b, active
->>>>>>>>>>479		ubyte		x	\b, starthead %u
-#>>>>>>>>>>480		ubyte		x	\b, start C_S: 0x%x
-#>>>>>>>>>>481		ubyte		x	\b, start C2S: 0x%x
-#>>>>>>>>>>480		ubeshort&1023	x	\b, startcylinder? %d
->>>>>>>>>>486		ulelong		x	\b, startsector %u
->>>>>>>>>>490		ulelong		x	\b, %u sectors
+>>>>>>>>>478		ubyte		<0x81		
+>>>>>>>>>>478		ubyte&0x7F	0		
+>>>>>>>>>>>482		ubyte		>0	\b; partition 3: ID=0x%x
+>>>>>>>>>>>>478		ubyte		0x80	\b, active
+>>>>>>>>>>>>479		ubyte		x	\b, starthead %u
+#>>>>>>>>>>>>480		ubyte		x	\b, start C_S: 0x%x
+#>>>>>>>>>>>>481		ubyte		x	\b, start C2S: 0x%x
+#>>>>>>>>>>>>480		ubeshort&1023	x	\b, startcylinder? %d
+>>>>>>>>>>>>486		ulelong		x	\b, startsector %u
+>>>>>>>>>>>>490		ulelong		x	\b, %u sectors
 #
->>>>>>>494		ubyte		<0x81	
->>>>>>>>494		ubyte&0x7F	0		
->>>>>>>>>498		ubyte		>0	\b; partition 4: ID=0x%x
->>>>>>>>>>494		ubyte		0x80	\b, active
->>>>>>>>>>495		ubyte		x	\b, starthead %u
-#>>>>>>>>>>496		ubyte		x	\b, start C_S: 0x%x
-#>>>>>>>>>>496		ubeshort&1023	x	\b, startcylinder? %d
->>>>>>>>>>502		ulelong		x	\b, startsector %u
->>>>>>>>>>506		ulelong		x	\b, %u sectors
+>>>>>>>>>494		ubyte		<0x81	
+>>>>>>>>>>494		ubyte&0x7F	0		
+>>>>>>>>>>>498		ubyte		>0	\b; partition 4: ID=0x%x
+>>>>>>>>>>>>494		ubyte		0x80	\b, active
+>>>>>>>>>>>>495		ubyte		x	\b, starthead %u
+#>>>>>>>>>>>>496		ubyte		x	\b, start C_S: 0x%x
+#>>>>>>>>>>>>496		ubeshort&1023	x	\b, startcylinder? %d
+>>>>>>>>>>>>502		ulelong		x	\b, startsector %u
+>>>>>>>>>>>>506		ulelong		x	\b, %u sectors
 # mbr partion table entries end
 # http://www.acronis.de/
 #FAT label=ACRONIS\ SZ
@@ -508,34 +517,52 @@
 >>>>>>>498	ubyte&0xDF	>0			
 >>>>>>>>498	string		x 			\b.%-.3s
 #
->486	ubyte&0xDF	>0
->>416	string	Non-System\ disk\ or\ 			
->>>435	string	disk\ error\r				
->>>>447	string	Replace\ and\ press\ any\ key\ 		
->>>>>473 string	when\ ready\r				\b, Microsoft DOS Bootloader
->480	ubyte&0xDF	>0			
->>393	string	Non-System\ disk\ or\ 			
->>>412	string	disk\ error\r				
->>>>424	string	Replace\ and\ press\ any\ key\ 		
->>>>>450 string	when\ ready\r				\b, Microsoft DOS bootloader
-#IO.SYS
->>>>>480	string		x 			\b %-.2s
->>>>>>482	ubyte&0xDF	>0			
->>>>>>>48	string		x 			\b%-.6s
->>>>>488	ubyte&0xDF	>0			
->>>>>>488	string		x 			\b.%-.3s
-#MSDOS.SYS
->>>>>>491	ubyte&0xDF	>0			\b+
->>>>>>>491	string		x 			\b%-.5s
->>>>>>>>496	ubyte&0xDF	>0			
->>>>>>>>>496	string		x 			\b%-.3s
->>>>>>>499	ubyte&0xDF	>0			
->>>>>>>>499	string		x 			\b.%-.3s
+>376	search/41	Non-System\ disk\ or\ 		
+>>395	search/41	disk\ error\r			
+>>>407	search/41	Replace\ and\ 			
+>>>>419	search/41	press\ 				\b,
+>>>>419	search/41	strike\ 			\b, old
+>>>>426	search/41	any\ key\ when\ ready\r		MS or PC-DOS bootloader
+#449			Disk\ Boot\ failure\r		MS 3.21
+#466			Boot\ Failure\r			MS 3.30
+>>>>>468 search/18	\0				
+#IO.SYS,IBMBIO.COM
+>>>>>>&0	string		x 			\b %-.2s
+>>>>>>>&-20	ubyte&0xDF	>0			
+>>>>>>>>&-1	string		x 			\b%-.4s
+>>>>>>>>>&-16	ubyte&0xDF	>0			
+>>>>>>>>>>&-1	string		x 			\b%-.2s
+>>>>>>&8	ubyte&0xDF	>0			\b.
+>>>>>>>&-1	string		x 			\b%-.3s
+#MSDOS.SYS,IBMDOS.COM
+>>>>>>&11	ubyte&0xDF	>0			\b+
+>>>>>>>&-1	string		x 			\b%-.5s
+>>>>>>>>&-6	ubyte&0xDF	>0			
+>>>>>>>>>&-1	string		x 			\b%-.1s
+>>>>>>>>>>&-5	ubyte&0xDF	>0			
+>>>>>>>>>>>&-1	string		x 			\b%-.2s
+>>>>>>>&7	ubyte&0xDF	>0			\b.
+>>>>>>>>&-1	string		x 			\b%-.3s
+>441	string	Cannot\ load\ from\ harddisk.\n\r
+>>469	string	Insert\ Systemdisk\ 			
+>>>487	string	and\ press\ any\ key.\n\r		\b, MS (2.11) DOS bootloader
 #>43	string	\224R-LOADER\ \ SYS			=label					
 >54	string	SYS
 >>324	string	VASKK
 >>>495	string	NEWLDR\0				\b, DR-DOS Bootloader (LOADER.SYS)
 #
+>98	string	Press\ a\ key\ to\ retry\0\r		
+>>120	string	Cannot\ find\ file\ \0\r		
+>>>139	string	Disk\ read\ error\0\r			
+>>>>156	string	Loading\ ...\0				\b, DR-DOS (3.41) Bootloader
+#DRBIOS.SYS
+>>>>>44		ubyte&0xDF	>0			
+>>>>>>44	string		x			\b %-.6s
+>>>>>>>50	ubyte&0xDF	>0			
+>>>>>>>>50	string		x 			\b%-.2s
+>>>>>>52	ubyte&0xDF	>0			
+>>>>>>>52	string		x 			\b.%-.3s
+#
 >70	string	IBMBIO\ \ COM				
 >>472	string	Cannot\ load\ DOS!\ 			
 >>>489	string	Any\ key\ to\ retry			\b, DR-DOS Bootloader
@@ -680,26 +707,43 @@
 #it also hangs with another message ("NF").
 >>>>>492		string		RENF		\b, FAT (12 bit)
 >>>>>495		string		RENF		\b, FAT (16 bit)
+# added by Joerg Jenderek
+# http://syslinux.zytor.com/iso.php
+0	ulelong	0x7c40eafa		isolinux Loader
+# http://syslinux.zytor.com/pxe.php
+0	ulelong	0x007c05ea		pxelinux Loader
+0	ulelong	0x60669c66		pxelinux Loader
 # loader end
-# Joerg Jenderek
->446	ubyte	0			
->>450	ubyte	>0			
->>>482	ubyte	0			
->>>>498	ubyte	0			
->>>>466	ubyte	0x05			\b, extended partition table
->>>>466	ubyte	0x0F			\b, extended partition table (LBA)
->>>>466	ubyte	0x0			\b, extended partition table (last)	
+# updated by Joerg Jenderek at Sep 2007
+>3	ubyte	0			
+#no active flag
+>>446	ubyte	0			
+# partition 1 not empty
+>>>450	ubyte	>0			
+# partitions 3,4 empty
+>>>>482		ubyte	0			
+>>>>>498	ubyte	0			
+# partition 2 ID=0,5,15
+>>>>>>466	ubyte	<0x10			
+>>>>>>>466	ubyte	0x05			\b, extended partition table
+>>>>>>>466	ubyte	0x0F			\b, extended partition table (LBA)
+>>>>>>>466	ubyte	0x0			\b, extended partition table (last)	
 # JuMP short     bootcodeoffset NOP assembler instructions will usually be EB xx 90
-# older drives may use E9 xx xx
+# http://mirror.href.com/thestarman/asm/2bytejumps.htmm#FWD
+# older drives may use Near JuMP instruction E9 xx xx
 >0		lelong&0x009000EB	0x009000EB 
 >0		lelong&0x000000E9	0x000000E9 
->>1		ubyte			>37	\b, code offset 0x%x
+# maximal short forward jump is 07fx
+>1		ubyte			<0xff	\b, code offset 0x%x
 # mtools-3.9.8/msdos.h
 # usual values are marked with comments to get only informations of strange FAT systems
-# valid sectorsize are from 32 to 2048
->>>11		uleshort	<2049	
->>>>11		uleshort	>31	
+# valid sectorsize must be a power of 2 from 32 to 32768
+>>11		uleshort&0x000f	x		
+>>>11		uleshort	<32769		
+>>>>11		uleshort	>31		
 >>>>>3		string		>\0		\b, OEM-ID "%8.8s"
+#http://mirror.href.com/thestarman/asm/debug/debug2.htm#IHC
+>>>>>>8		string		IHC		\b cached by Windows 9M
 >>>>>11		uleshort	>512		\b, Bytes/sector %u
 #>>>>>11	uleshort	=512		\b, Bytes/sector %u=512 (usual)
 >>>>>11		uleshort	<512		\b, Bytes/sector %u
@@ -726,17 +770,18 @@
 >>>>>26		ubyte		>2		\b, heads %u
 #>>>>>26	ubyte		=2		\b, heads %u (usual floppy)
 >>>>>26		ubyte		=1		\b, heads %u
->>>>>28		ulelong		>0		\b, hidden sectors %u
-#>>>>>28	ulelong		=0		\b, hidden sectors %u (usual floppy)
->>>>>32		ulelong		>0		\b, sectors %u (volumes > 32 MB) 
-#>>>>>32	ulelong		=0		\b, sectors %u (volumes > 32 MB) 
+#skip for Digital Research DOS (version 3.41) 1440 kB Bootdisk
+>>>>>38		ubyte		!0x70		
+>>>>>>28	ulelong		>0		\b, hidden sectors %u
+#>>>>>>28	ulelong		=0		\b, hidden sectors %u (usual floppy)
+>>>>>>32	ulelong		>0		\b, sectors %u (volumes > 32 MB) 
+#>>>>>>32	ulelong		=0		\b, sectors %u (volumes > 32 MB)
 # FAT<32 specific 
-# NOT le FAT3=NOT 3TAF=0xCCABBEB9
->>>>>82		ulelong&0xCCABBEB9	>0
->>>>>>36	ubyte		>0x80		\b, physical drive 0x%x
-#>>>>>>36	ubyte		=0x80		\b, physical drive 0x%x=0x80 (usual harddisk)
->>>>>>36	ubyte&0x7F	>0		\b, physical drive 0x%x
-#>>>>>>36	ubyte		=0		\b, physical drive 0x%x=0 (usual floppy)
+>>>>>82		string		!FAT32
+#>>>>>>36	ubyte		0x80		\b, physical drive 0x%x=0x80 (usual harddisk)
+#>>>>>>36	ubyte		0		\b, physical drive 0x%x=0 (usual floppy)
+>>>>>>36	ubyte		!0x80		
+>>>>>>>36	ubyte		!0		\b, physical drive 0x%x
 >>>>>>37	ubyte		>0		\b, reserved 0x%x
 #>>>>>>37	ubyte		=0		\b, reserved 0x%x
 >>>>>>38	ubyte		>0x29		\b, dos < 4.0 BootSector (0x%x)
@@ -789,7 +834,7 @@
 
 # Minix filesystems - Juan Cespedes <cespedes@debian.org>
 0x410	leshort		0x137f		Minix filesystem
-0x410	beshort		0x137f		Minix filesystem (big endian),
+0x410	beshort		0x137f		Minix filesystem (big endian)
 >0x402	beshort		!0		\b, %d zones
 >0x1e	string		minix		\b, bootable
 0x410	leshort		0x138f		Minix filesystem, 30 char names
@@ -981,19 +1026,42 @@
 >&-1248	belong		0		TIME optimization
 >&-1248	belong		1		SPACE optimization
 
-# ext2/ext3 filesystems - Andreas Dilger <adilger@turbolabs.com>
-0x438	leshort		0xEF53		Linux
->0x44c	lelong		x		rev %d
->0x43e	leshort		x		\b.%d
->0x45c	lelong		^0x0000004	ext2 filesystem data
->>0x43a	leshort		^0x0000001	(mounted or unclean)
->0x45c	lelong		&0x0000004	ext3 filesystem data
->>0x460	lelong		&0x0000004	(needs journal recovery)
->0x43a	leshort		&0x0000002	(errors)
->0x460	lelong		&0x0000001	(compressed)
-#>0x460	lelong		&0x0000002	(filetype)
-#>0x464	lelong		&0x0000001	(sparse_super)
->0x464	lelong		&0x0000002	(large files)
+# ext2/ext3 filesystems - Andreas Dilger <adilger@dilger.ca>
+# ext4 filesystem - Eric Sandeen <sandeen@sandeen.net>
+0x438   leshort         0xEF53          Linux
+>0x44c  lelong          x               rev %d
+>0x43e  leshort         x               \b.%d
+# No journal?  ext2
+>0x45c  lelong          ^0x0000004      ext2 filesystem data
+>>0x43a leshort         ^0x0000001      (mounted or unclean)
+# Has a journal?  ext3 or ext4
+>0x45c  lelong          &0x0000004
+#  and small INCOMPAT?
+>>0x460 lelong          <0x0000040
+#   and small RO_COMPAT?
+>>>0x464 lelong         <0x0000008      ext3 filesystem data
+#   else large RO_COMPAT?
+>>>0x464 lelong         >0x0000007      ext4 filesystem data
+#  else large INCOMPAT?
+>>0x460 lelong          >0x000003f      ext4 filesystem data
+# General flags for any ext* fs
+>0x460  lelong          &0x0000004      (needs journal recovery)
+>0x43a  leshort         &0x0000002      (errors)
+# INCOMPAT flags
+>0x460  lelong          &0x0000001      (compressed)
+#>0x460 lelong          &0x0000002      (filetype)
+#>0x460 lelong          &0x0000010      (meta bg)
+>0x460  lelong          &0x0000040      (extents)
+>0x460  lelong          &0x0000080      (64bit)
+#>0x460 lelong          &0x0000100      (mmp)
+#>0x460 lelong          &0x0000200      (flex bg)
+# RO_INCOMPAT flags
+#>0x464 lelong          &0x0000001      (sparse super)
+>0x464  lelong          &0x0000002      (large files)
+>0x464  lelong          &0x0000008      (huge files)
+#>0x464 lelong          &0x0000010      (gdt checksum)
+#>0x464 lelong          &0x0000020      (many subdirs)
+#>0x463 lelong          &0x0000040      (extra isize)
 
 # SGI disk labels - Nathan Scott <nathans@debian.org>
 0	belong		0x0BE5A941	SGI disk label (volume header)
@@ -1071,41 +1139,42 @@
 
 # CDROM Filesystems
 # Modified for UDF by gerardo.cacciari@gmail.com
-32769    string    CD001
->38913   string   !NSR0      ISO 9660 CD-ROM filesystem data
->38913   string    NSR01     UDF filesystem data (version 1.0)
->38913   string    NSR02     UDF filesystem data (version 1.5)
->38913   string    NSR03     UDF filesystem data (version 2.0)
->38913   string    >NSR03    UDF filesystem data (unknown version,
->>38917  byte      x         id 'NSR0%c')
->38913   string    <NSR01    UDF filesystem data (unknown version,
->>38917  byte      x         id 'NSR0%c')
+32769	string    CD001
+!:mime	application/x-iso9660-image
+>38913	string   !NSR0      ISO 9660 CD-ROM filesystem data
+>38913	string    NSR0      UDF filesystem data
+>>38917	string    1         (version 1.0)
+>>38917	string    2         (version 1.5)
+>>38917	string    3         (version 2.0)
+>>38917	byte     >0x33      (unknown version, ID 0x%X)
+>>38917	byte     <0x31      (unknown version, ID 0x%X)
 # "application id" which appears to be used as a volume label
->32808	 string    >\0       '%s'
->34816	 string    \000CD001\001EL\ TORITO\ SPECIFICATION    (bootable)
-37633    string    CD001     ISO 9660 CD-ROM filesystem data (raw 2352 byte sectors)
-32776    string    CDROM     High Sierra CD-ROM filesystem data
+>32808	string    >\0       '%s'
+>34816	string    \000CD001\001EL\ TORITO\ SPECIFICATION    (bootable)
+37633	string    CD001     ISO 9660 CD-ROM filesystem data (raw 2352 byte sectors)
+!:mime	application/x-iso9660-image
+32776	string    CDROM     High Sierra CD-ROM filesystem data
 
 # cramfs filesystem - russell@coker.com.au
 0       lelong    0x28cd3d45      Linux Compressed ROM File System data, little endian
->4      lelong  x size %d
+>4      lelong  x size %lu
 >8      lelong  &1 version #2
 >8      lelong  &2 sorted_dirs
 >8      lelong  &4 hole_support
 >32     lelong  x CRC 0x%x,
->36     lelong  x edition %d,
->40     lelong  x %d blocks,
->44     lelong  x %d files
+>36     lelong  x edition %lu,
+>40     lelong  x %lu blocks,
+>44     lelong  x %lu files
 
 0       belong    0x28cd3d45      Linux Compressed ROM File System data, big endian
->4      belong  x size %d
+>4      belong  x size %lu
 >8      belong  &1 version #2
 >8      belong  &2 sorted_dirs
 >8      belong  &4 hole_support
 >32     belong  x CRC 0x%x,
->36     belong  x edition %d,
->40     belong  x %d blocks,
->44     belong  x %d files
+>36     belong  x edition %lu,
+>40     belong  x %lu blocks,
+>44     belong  x %lu files
 
 # reiserfs - russell@coker.com.au
 0x10034		string	ReIsErFs	ReiserFS V3.5
@@ -1211,3 +1280,13 @@
 >525    byte            x               Level %d
 >525    byte            x               (ODS-%d OpenVMS file system),
 >984    string          x               volume label is '%-12.12s'
+
+# From: Thomas Klausner <wiz@NetBSD.org>
+# http://filext.com/file-extension/DAA
+# describes the daa file format. The magic would be:
+0	string		DAA\x0\x0\x0\x0\x0	PowerISO Direct-Access-Archive
+
+# From Albert Cahalan <acahalan@gmail.com>
+# really le32 operation,destination,payloadsize (but quite predictable)
+# 01 00 00 00 00 00 00 c0 00 02 00 00
+0	string		\1\0\0\0\0\0\0\300\0\2\0\0	Marvell Libertas firmware
diff --git a/usr.bin/file/magdir/flash b/usr.bin/file/magdir/flash
index 7d3d86aa10b..ea3e6fb20e9 100644
--- a/usr.bin/file/magdir/flash
+++ b/usr.bin/file/magdir/flash
@@ -1,4 +1,4 @@
-#	$OpenBSD: flash,v 1.2 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: flash,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # flash:	file(1) magic for Macromedia Flash file format
@@ -9,10 +9,13 @@
 #
 0	string		FWS		Macromedia Flash data,
 >3	byte		x		version %d
+!:mime	application/x-shockwave-flash
 0	string		CWS		Macromedia Flash data (compressed),
+!:mime	application/x-shockwave-flash
 >3	byte		x		version %d
 # From: Cal Peake <cp@absolutedigital.net>
 0	string		FLV		Macromedia Flash Video
+!:mime	video/x-flv
 
 #
 # From Dave Wilson
diff --git a/usr.bin/file/magdir/fonts b/usr.bin/file/magdir/fonts
index c3d32ab56b6..c2cf2d90394 100644
--- a/usr.bin/file/magdir/fonts
+++ b/usr.bin/file/magdir/fonts
@@ -1,15 +1,15 @@
-#	$OpenBSD: fonts,v 1.4 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: fonts,v 1.5 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # fonts:  file(1) magic for font data
 #
-0	string		FONT		ASCII vfont text
+0	search/1	FONT		ASCII vfont text
 0	short		0436		Berkeley vfont data
 0	short		017001		byte-swapped Berkeley vfont data
 
 # PostScript fonts (must precede "printer" entries), quinlan@yggdrasil.com
-0	string		%!PS-AdobeFont-1.	PostScript Type 1 font text
->20	string		>\0			(%s)
+0	search/1	%!PS-AdobeFont-1.	PostScript Type 1 font text
+>20	search/1	>\0			(%s)
 6	string		%!PS-AdobeFont-1.	PostScript Type 1 font program data
 
 # X11 font files in SNF (Server Natural Format) format
@@ -17,7 +17,7 @@
 0	lelong		00000004		X11 SNF font data, LSB first
 
 # X11 Bitmap Distribution Format, from Daniel Quinlan (quinlan@yggdrasil.com)
-0	string		STARTFONT\040		X11 BDF font text
+0	search/1	STARTFONT\ 		X11 BDF font text
 
 # X11 fonts, from Daniel Quinlan (quinlan@yggdrasil.com)
 # PCF must come before SGI additions ("MIPSEL MIPS-II COFF" collides)
diff --git a/usr.bin/file/magdir/fortran b/usr.bin/file/magdir/fortran
new file mode 100644
index 00000000000..3e490333e18
--- /dev/null
+++ b/usr.bin/file/magdir/fortran
@@ -0,0 +1,3 @@
+# FORTRAN source
+0	string/c	c\ 		FORTRAN program
+!:mime	text/x-fortran
diff --git a/usr.bin/file/magdir/frame b/usr.bin/file/magdir/frame
index 4b3405a3657..08f87e7c4b5 100644
--- a/usr.bin/file/magdir/frame
+++ b/usr.bin/file/magdir/frame
@@ -1,4 +1,4 @@
-#	$OpenBSD: frame,v 1.3 2004/06/03 03:14:20 tedu Exp $
+#	$OpenBSD: frame,v 1.4 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # frame:  file(1) magic for FrameMaker files
@@ -6,7 +6,11 @@
 # This stuff came on a FrameMaker demo tape, most of which is
 # copyright, but this file is "published" as witness the following:
 #
+# Note that this is the Framemaker Maker Interchange Format, not the
+# Normal format which would be application/vnd.framemaker.
+#
 0	string		\<MakerFile	FrameMaker document
+!:mime	application/x-mif
 >11	string		5.5		 (5.5
 >11	string		5.0		 (5.0
 >11	string		4.0		 (4.0
@@ -15,25 +19,32 @@
 >11	string		1.0		 (1.0
 >14	byte		x		  %c)
 0	string		\<MIFFile	FrameMaker MIF (ASCII) file
+!:mime	application/x-mif
 >9	string		4.0		 (4.0)
 >9	string		3.0		 (3.0)
 >9	string		2.0		 (2.0)
 >9	string		1.0		 (1.x)
-0	string		\<MakerDictionary	FrameMaker Dictionary text
+0	search/1	\<MakerDictionary	FrameMaker Dictionary text
+!:mime	application/x-mif
 >17	string		3.0		 (3.0)
 >17	string		2.0		 (2.0)
 >17	string		1.0		 (1.x)
 0	string		\<MakerScreenFont	FrameMaker Font file
+!:mime	application/x-mif
 >17	string		1.01		 (%s)
 0	string		\<MML		FrameMaker MML file
+!:mime	application/x-mif
 0	string		\<BookFile	FrameMaker Book file
+!:mime	application/x-mif
 >10	string		3.0		 (3.0
 >10	string		2.0		 (2.0
 >10	string		1.0		 (1.0
 >13	byte		x		  %c)
 # XXX - this book entry should be verified, if you find one, uncomment this
 #0	string		\<Book\ 	FrameMaker Book (ASCII) file
+#!:mime	application/x-mif
 #>6	string		3.0		 (3.0)
 #>6	string		2.0		 (2.0)
 #>6	string		1.0		 (1.0)
 0	string		\<Maker	Intermediate Print File	FrameMaker IPL file
+!:mime	application/x-mif
diff --git a/usr.bin/file/magdir/fsav b/usr.bin/file/magdir/fsav
index 799e19872d7..18395502fe6 100644
--- a/usr.bin/file/magdir/fsav
+++ b/usr.bin/file/magdir/fsav
@@ -1,4 +1,4 @@
-#	$OpenBSD: fsav,v 1.2 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: fsav,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # fsav:  file(1) magic for datafellows fsav virus definition files
@@ -47,7 +47,7 @@
 0	string		ClamAV-VDB:	
 >11	string		>\0		Clam AntiVirus database %-.23s
 >>34	string		:		
->>>35	regex		[^:]+		\b, version 
+>>>35		string		!:	\b, version 
 >>>>35		string		x 	\b%-.1s
 >>>>>36		string 		!:	
 >>>>>>36	string		x 	\b%-.1s
@@ -56,6 +56,10 @@
 >>>>>>>>>38	string		!:	
 >>>>>>>>>>38	string		x 	\b%-.1s
 >>>>512	string		\037\213	\b, gzipped
->>>>769	string		ustar\0		\b, tared
+>>>>769	string		ustar\0		\b, tarred
 >512	string		\037\213	\b, gzipped
->769	string		ustar\0		\b, tared
+>769	string		ustar\0		\b, tarred
+
+# Type: Grisoft AVG AntiVirus
+# From: David Newgas <david@newgas.net>
+0	string	AVG7_ANTIVIRUS_VAULT_FILE	AVG 7 Antivirus vault file data
diff --git a/usr.bin/file/magdir/games b/usr.bin/file/magdir/games
index aff8cef4c9e..7f0a79a3b61 100644
--- a/usr.bin/file/magdir/games
+++ b/usr.bin/file/magdir/games
@@ -1,14 +1,8 @@
-#	$OpenBSD: games,v 1.1 2004/06/03 03:14:20 tedu Exp $
+#	$OpenBSD: games,v 1.2 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # games:  file(1) for games
 
-# Thomas M. Ott (ThMO)
-1	string	=WAD		DOOM data,
->0	string	=I		main wad
->0	string	=P		patch wad
->0	byte	x		unknown junk
-
 # Fabio Bonelli <fabiobonelli@libero.it>
 # Quake II - III data files
 0       string  IDP2        	Quake II 3D Model file,
@@ -31,11 +25,6 @@
 # Doom and Quake
 # submitted by Nicolas Patrois
 
-# DOOM
-
-0       string  IWAD    DOOM or DOOM ][ world
-0       string  PWAD    DOOM or DOOM ][ extension world
-
 0       string  \xcb\x1dBoom\xe6\xff\x03\x01    Boom or linuxdoom demo
 # some doom lmp files don't match, I've got one beginning with \x6d\x02\x01\x01
 
@@ -157,3 +146,10 @@
 
 0	string	MComprHD	MAME CHD compressed hard disk image,
 >12	belong	x		version %lu
+
+# doom - submitted by Jon Dowland
+
+0	string	=IWAD		doom main IWAD data
+>4	lelong	x		containing %d lumps
+0	string	=PWAD		doom patch PWAD data
+>4	lelong	x		containing %d lumps
diff --git a/usr.bin/file/magdir/gnome-keyring b/usr.bin/file/magdir/gnome-keyring
new file mode 100644
index 00000000000..80a4f882c16
--- /dev/null
+++ b/usr.bin/file/magdir/gnome-keyring
@@ -0,0 +1,23 @@
+# GNOME keyring
+# Contributed by Josh Triplett
+# FIXME: Could be simplified if pstring supported two-byte counts
+0         string   GnomeKeyring\n\r\0\n GNOME keyring
+>&0       ubyte    0                    \b, major version 0
+>>&0      ubyte    0                    \b, minor version 0
+>>>&0     ubyte    0                    \b, crypto type 0 (AEL)
+>>>&0     ubyte    >0                   \b, crypto type %hhu (unknown)
+>>>&1     ubyte    0                    \b, hash type 0 (MD5)
+>>>&1     ubyte    >0                   \b, hash type %hhu (unknown)
+>>>&2     ubelong  0xFFFFFFFF           \b, name NULL
+>>>&2     ubelong  !0xFFFFFFFF
+>>>>&-4   ubelong  >255                 \b, name too long for file's pstring type
+>>>>&-4   ubelong  <256
+>>>>>&-1  pstring  x                    \b, name "%s"
+>>>>>>&0  ubeqdate x                    \b, last modified %s
+>>>>>>&8  ubeqdate x                    \b, created %s
+>>>>>>&16 ubelong  &1
+>>>>>>>&0 ubelong  x                    \b, locked if idle for %u seconds
+>>>>>>&16 ubelong  ^1                   \b, not locked if idle
+>>>>>>&24 ubelong  x                    \b, hash iterations %u
+>>>>>>&28 ubequad  x                    \b, salt %llu
+>>>>>>&52 ubelong  x                    \b, %u item(s)
diff --git a/usr.bin/file/magdir/gnu b/usr.bin/file/magdir/gnu
index baa9f4c03ac..4a34282ca04 100644
--- a/usr.bin/file/magdir/gnu
+++ b/usr.bin/file/magdir/gnu
@@ -1,4 +1,4 @@
-#	$OpenBSD: gnu,v 1.2 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: gnu,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # gnu:  file(1) magic for various GNU tools
@@ -19,10 +19,14 @@
 # The format is very similar to pgp
 0	string          \001gpg                 GPG key trust database
 >4	byte            x                       version %d
+# Note: magic.mime had 0x8501 for the next line instead of 0x8502
 0	beshort		0x8502			GPG encrypted data
+!:mime	text/PGP # encoding: data
+
 # This magic is not particularly good, as the keyrings don't have true
 # magic. Nevertheless, it covers many keyrings.
 0       beshort         0x9901                  GPG key public ring
+!:mime	application/x-gnupg-keyring
 
 # Gnumeric spreadsheet
 # This entry is only semi-helpful, as Gnumeric compresses its files, so
diff --git a/usr.bin/file/magdir/gnumeric b/usr.bin/file/magdir/gnumeric
new file mode 100644
index 00000000000..76dfa90684e
--- /dev/null
+++ b/usr.bin/file/magdir/gnumeric
@@ -0,0 +1,7 @@
+
+#------------------------------------------------------------------------------
+# gnumeric:  file(1) magic for Gnumeric spreadsheet
+# This entry is only semi-helpful, as Gnumeric compresses its files, so
+# they will ordinarily reported as "compressed", but at least -z helps
+39	string	=<gmr:Workbook	Gnumeric spreadsheet
+!:mime	application/x-gnumeric
diff --git a/usr.bin/file/magdir/hp b/usr.bin/file/magdir/hp
index ed131ba3e3f..49c6434cbd2 100644
--- a/usr.bin/file/magdir/hp
+++ b/usr.bin/file/magdir/hp
@@ -1,4 +1,4 @@
-#	$OpenBSD: hp,v 1.6 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: hp,v 1.7 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # hp:  file(1) magic for Hewlett Packard machines (see also "printer")
@@ -204,33 +204,36 @@
 >8	long		>0		%d messages
 
 # addendum to /etc/magic with HP-48sx file-types by phk@data.fls.dk 1jan92
-0	string		HPHP48-		HP48 binary
->7	byte		>0		- Rev %c
->8	beshort		0x1129		(ADR)
->8	beshort		0x3329		(REAL)
->8	beshort		0x5529		(LREAL)
->8	beshort		0x7729		(COMPLX)
->8	beshort		0x9d29		(LCOMPLX)
->8	beshort		0xbf29		(CHAR)
->8	beshort		0xe829		(ARRAY)
->8	beshort		0x0a2a		(LNKARRAY)
->8	beshort		0x2c2a		(STRING)
->8	beshort		0x4e2a		(HXS)
->8	beshort		0x742a		(LIST)
->8	beshort		0x962a		(DIR)
->8	beshort		0xb82a		(ALG)
->8	beshort		0xda2a		(UNIT)
->8	beshort		0xfc2a		(TAGGED)
->8	beshort		0x1e2b		(GROB)
->8	beshort		0x402b		(LIB)
->8	beshort		0x622b		(BACKUP)
->8	beshort		0x882b		(LIBDATA)
->8	beshort		0x9d2d		(PROG)
->8	beshort		0xcc2d		(CODE)
->8	beshort		0x482e		(GNAME)
->8	beshort		0x6d2e		(LNAME)
->8	beshort		0x922e		(XLIB)
-0	string		%%HP:		HP48 text
+0	string		HPHP4		HP
+>5	string		8		48 binary
+>5	string		9		49 binary
+>7	byte		>64		- Rev %c
+>8	leshort		0x2911		(ADR)
+>8	leshort		0x2933		(REAL)
+>8	leshort		0x2955		(LREAL)
+>8	leshort		0x2977		(COMPLX)
+>8	leshort		0x299d		(LCOMPLX)
+>8	leshort		0x29bf		(CHAR)
+>8	leshort		0x29e8		(ARRAY)
+>8	leshort		0x2a0a		(LNKARRAY)
+>8	leshort		0x2a2c		(STRING)
+>8	leshort		0x2a4e		(HXS)
+>8	leshort		0x2a74		(LIST)
+>8	leshort		0x2a96		(DIR)
+>8	leshort		0x2ab8		(ALG)
+>8	leshort		0x2ada		(UNIT)
+>8	leshort		0x2afc		(TAGGED)
+>8	leshort		0x2b1e		(GROB)
+>8	leshort		0x2b40		(LIB)
+>8	leshort		0x2b62		(BACKUP)
+>8	leshort		0x2b88		(LIBDATA)
+>8	leshort		0x2d9d		(PROG)
+>8	leshort		0x2dcc		(CODE)
+>8	leshort		0x2e48		(GNAME)
+>8	leshort		0x2e6d		(LNAME)
+>8	leshort		0x2e92		(XLIB)
+
+0	search/1	%%HP:		HP text
 >6	string		T(0)		- T(0)
 >6	string		T(1)		- T(1)
 >6	string		T(2)		- T(2)
@@ -241,6 +244,23 @@
 >14	string		F(.)		F(.);
 >14	string		F(,)		F(,);
 
+0	string		HP3		HP
+>3	string		8		38
+>3	string		9		39
+>4	string		Bin		binary
+>4	string		Asc		ASCII
+>7	string		A		(Directory List)
+>7	string		B		(Zaplet)
+>7	string		C		(Note)
+>7	string		D		(Program)
+>7	string		E		(Variable)
+>7	string		F		(List)
+>7	string		G		(Matrix)
+>7	string		H		(Library)
+>7	string		I		(Target List)
+>7	string		J		(ASCII Vector specification)
+>7	string		K		(wildcard)
+
 # hpBSD magic numbers
 0	beshort		200		hp200 (68010) BSD
 >2	beshort		0407		impure binary
@@ -392,5 +412,5 @@
 >>>>>>>>>0xC4	belong	34		- received SIGXFSZ
 
 # From: AMAKAWA Shuhei <sa264@cam.ac.uk>
-0	string	HPHP49-			HP49 binary
+#0	string	HPHP49-			HP49 binary
 
diff --git a/usr.bin/file/magdir/iff b/usr.bin/file/magdir/iff
index ca0b18a55bf..d1388b509d1 100644
--- a/usr.bin/file/magdir/iff
+++ b/usr.bin/file/magdir/iff
@@ -1,4 +1,4 @@
-#	$OpenBSD: iff,v 1.3 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: iff,v 1.4 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # iff:	file(1) magic for Interchange File Format (see also "audio" & "images")
@@ -14,8 +14,11 @@
 #>4	belong		x		\b, FORM is %d bytes long
 # audio formats
 >8	string		AIFF		\b, AIFF audio
+!:mime	audio/x-aiff
 >8	string		AIFC		\b, AIFF-C compressed audio
+!:mime	audio/x-aiff
 >8	string		8SVX		\b, 8SVX 8-bit sampled sound voice
+!:mime	audio/x-aiff
 >8	string		16SV		\b, 16SV 16-bit sampled sound voice
 >8	string		SAMP		\b, SAMP sampled audio
 >8	string		MAUD		\b, MAUD MacroSystem audio
diff --git a/usr.bin/file/magdir/images b/usr.bin/file/magdir/images
index 343e1ff23d8..2e47a8eb15a 100644
--- a/usr.bin/file/magdir/images
+++ b/usr.bin/file/magdir/images
@@ -1,7 +1,8 @@
-#	$OpenBSD: images,v 1.9 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: images,v 1.10 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
-# images:  file(1) magic for image formats (see also "iff")
+# images:  file(1) magic for image formats (see also "iff", and "c-lang" for
+# XPM bitmaps)
 #
 # originally from jef@helios.ee.lbl.gov (Jef Poskanzer),
 # additions by janl@ifi.uio.no as well as others. Jan also suggested
@@ -31,13 +32,20 @@
 
 # PBMPLUS images
 # The next byte following the magic is always whitespace.
-0	string		P1		Netpbm PBM image text
-0	string		P2		Netpbm PGM image text
-0	string		P3		Netpbm PPM image text
+0	search/1	P1		Netpbm PBM image text
+!:mime	image/x-portable-bitmap
+0	search/1	P2		Netpbm PGM image text
+!:mime	image/x-portable-greymap
+0	search/1	P3		Netpbm PPM image text
+!:mime	image/x-portable-pixmap
 0	string		P4		Netpbm PBM "rawbits" image data
+!:mime	image/x-portable-bitmap
 0	string		P5		Netpbm PGM "rawbits" image data
+!:mime	image/x-portable-greymap
 0	string		P6		Netpbm PPM "rawbits" image data
+!:mime	image/x-portable-pixmap
 0	string		P7		Netpbm PAM image file
+!:mime	image/x-portable-pixmap
 
 # From: bryanh@giraffe-data.com (Bryan Henderson)
 0	string		\117\072	Solitaire Image Recorder format
@@ -51,13 +59,17 @@
 >21	byte		51		version 3
 
 # NIFF (Navy Interchange File Format, a modification of TIFF) images
+# [GRR:  this *must* go before TIFF]
 0	string		IIN1		NIFF image data
+!:mime	image/x-niff
 
 # Tag Image File Format, from Daniel Quinlan (quinlan@yggdrasil.com)
 # The second word of TIFF files is the TIFF version number, 42, which has
 # never changed.  The TIFF specification recommends testing for it.
 0	string		MM\x00\x2a	TIFF image data, big-endian
+!:mime	image/tiff
 0	string		II\x2a\x00	TIFF image data, little-endian
+!:mime	image/tiff
 
 # PNG [Portable Network Graphics, or "PNG's Not GIF"] images
 # (Greg Roelofs, newt@uchicago.edu)
@@ -66,6 +78,7 @@
 # 137 P N G \r \n ^Z \n [4-byte length] H E A D [HEAD data] [HEAD crc] ...
 #
 0	string		\x89PNG		PNG image data,
+!:mime	image/png
 >4	belong		!0x0d0a1a0a	CORRUPTED,
 >4	belong		0x0d0a1a0a
 >>16	belong		x		%ld x
@@ -81,8 +94,27 @@
 >>28	byte		1		interlaced
 1	string		PNG		PNG image data, CORRUPTED
 
+# possible GIF replacements; none yet released!
+# (Greg Roelofs, newt@uchicago.edu)
+#
+# GRR 950115:  this was mine ("Zip GIF"):
+0	string		GIF94z		ZIF image (GIF+deflate alpha)
+!:mime	image/x-unknown
+#
+# GRR 950115:  this is Jeremy Wohl's Free Graphics Format (better):
+#					
+0	string		FGF95a		FGF image (GIF+deflate beta)
+!:mime	image/x-unknown
+#
+# GRR 950115:  this is Thomas Boutell's Portable Bitmap Format proposal
+# (best; not yet implemented):
+#					
+0	string		PBF		PBF image (deflate compression)
+!:mime	image/x-unknown
+
 # GIF
 0	string		GIF8		GIF image data
+!:mime	image/gif
 >4	string		7a		\b, version 8%s,
 >4	string		9a		\b, version 8%s,
 >6	leshort		>0		%hd x
@@ -114,7 +146,7 @@
 >4	long		3		\b, rectangular 32-bit (24-bit with matte)
 
 # FIG (Facility for Interactive Generation of figures), an object-based format
-0	string		#FIG		FIG image text
+0	search/1	#FIG		FIG image text
 >5	string		x		\b, version %.3s
 
 # PHIGS
@@ -131,8 +163,8 @@
 # CGM image files
 0	string		BEGMF		clear text Computer Graphics Metafile
 # XXX - questionable magic
-0	beshort&0xffe0	0x0020		binary Computer Graphics Metafile
-0	beshort		0x3020		character Computer Graphics Metafile
+#0	beshort&0xffe0	0x0020		binary Computer Graphics Metafile
+#0	beshort		0x3020		character Computer Graphics Metafile
 
 # MGR bitmaps  (Michael Haardt, u31b3hs@pool.informatik.rwth-aachen.de)
 0	string	yz	MGR bitmap, modern format, 8-bit aligned
@@ -153,8 +185,9 @@
 0	string		Sfff		structured fax file
 
 
-# PC bitmaps (OS/2, Windoze BMP files)  (Greg Roelofs, newt@uchicago.edu)
+# PC bitmaps (OS/2, Windows BMP files)  (Greg Roelofs, newt@uchicago.edu)
 0	string		BM		PC bitmap data
+!:mime	image/x-ms-bmp
 >14	leshort		12		\b, OS/2 1.x format
 >>18	leshort		x		\b, %d x
 >>20	leshort		x		%d
@@ -175,7 +208,7 @@
 
 # XPM icons (Greg Roelofs, newt@uchicago.edu)
 # note possible collision with C/REXX entry in c-lang; currently commented out
-0	string		/*\ XPM\ */	X pixmap image text
+0	search/1	/*\ XPM\ */	X pixmap image text
 
 # Utah Raster Toolkit RLE images (janl@ifi.uio.no)
 0	leshort		0xcc52		RLE image data,
@@ -261,7 +294,7 @@
 
 # other images
 0	string	This\ is\ a\ BitMap\ file	Lisp Machine bit-array-file
-0	string		=!!		Bennet Yee's "face" format
+#0	string		=!!		Bennet Yee's "face" format
 
 # From SunOS 5.5.1 "/etc/magic" - appeared right before Sun raster image
 # stuff.
@@ -270,6 +303,7 @@
 
 # DICOM medical imaging data
 128	string	DICM			DICOM medical imaging data
+!:mime	application/dicom
 
 # XWD - X Window Dump file.
 #   As described in /usr/X11R6/include/X11/XWDFile.h
@@ -368,6 +402,7 @@
 
 # Adobe Photoshop
 0	string		8BPS Adobe Photoshop Image
+!:mime	image/vnd.adobe.photoshop
 
 # XV thumbnail indicator (ThMO)
 0	string		P7\ 332		XV thumbnail image data
@@ -457,9 +492,9 @@
 >12     long    x               \b, track size %d bytes
 >16     byte    x               \b, device type 33%2.2X
 
-# Squeak images and - etoffi@softhome.net
-0 string \146\031\0\0  Squeak image data
-0 string 'From\040Squeak  Squeak program text
+# Squeak images and programs - etoffi@softhome.net
+0	string		\146\031\0\0	Squeak image data
+0	search/1	'From\040Squeak	Squeak program text
 
 # partimage: file(1) magic for PartImage files (experimental, incomplete)
 # Author: Hans-Joachim Baader <hjb@pro-linux.de>
@@ -509,9 +544,7 @@
 # From: stephane.loeuillet@tiscali.f
 # http://www.djvuzone.org/
 0	string	AT&TFORM		DjVu Image file
-
-# From: Jason Bacon <bacon@smithers.neuro.mcw.edu>
-0	beshort	0x3020			character Computer Graphics Metafile
+!:mime	image/vnd.djvu
 
 # From Marc Espie
 0	lelong	20000630		OpenEXR image data
@@ -524,9 +557,36 @@
 # Hierarchical Data Format, used to facilitate scientific data exchange
 # specifications at http://hdf.ncsa.uiuc.edu/
 0	belong	0x0e031301	Hierarchical Data Format (version 4) data
+!:mime	application/x-hdf
 0	string	\211HDF\r\n\032	Hierarchical Data Format (version 5) data
+!:mime	application/x-hdf
 
 # From: Tobias Burnus <burnus@net-b.de>
 # Xara (for a while: Corel Xara) is a graphic package, see
-# http://www.xara.com/ for Windows and as GPL application for
+# http://www.xara.com/ for Windows and as GPL application for Linux
 0	string	XARA\243\243	Xara graphics file
+
+# http://www.cartesianinc.com/Tech/
+0	string	CPC\262		Cartesian Perceptual Compression image
+!:mime	image/x-cpi
+
+# From Albert Cahalan <acahalan@gmail.com>
+# puredigital used it for the CVS disposable camcorder
+#8       lelong  4       ZBM bitmap image data
+#>4      leshort x       %u x
+#>6      leshort x       %u
+
+# From Albert Cahalan <acahalan@gmail.com>
+# uncompressed 5:6:5 HighColor image for OLPC XO firmware icons
+0       string C565     OLPC firmware icon image data
+>4      leshort x       %u x
+>6      leshort x       %u
+
+# Applied Images - Image files from Cytovision
+# Gustavo Junior Alves <gjalves@gjalves.com.br>
+0	string	\xce\xda\xde\xfa	Cytovision Metaphases file
+0	string	\xed\xad\xef\xac	Cytovision Karyotype file
+0	string	\x0b\x00\x03\x00	Cytovision FISH Probe file
+0	string	\xed\xfe\xda\xbe	Cytovision FLEX file
+0	string	\xed\xab\xed\xfe	Cytovision FLEX file
+0	string	\xad\xfd\xea\xad	Cytovision RATS file
diff --git a/usr.bin/file/magdir/inform b/usr.bin/file/magdir/inform
new file mode 100644
index 00000000000..ba35d61e046
--- /dev/null
+++ b/usr.bin/file/magdir/inform
@@ -0,0 +1,8 @@
+
+#------------------------------------------------------------------------------
+# inform:  file(1) magic for Inform interactive fiction language
+
+# URL:  http://www.inform-fiction.org/
+# From: Reuben Thomas <rrt@sc3d.org>
+
+0	search/cB/100	constant\ story		Inform source text
diff --git a/usr.bin/file/magdir/java b/usr.bin/file/magdir/java
index cdde1df7b0c..4951a66533c 100644
--- a/usr.bin/file/magdir/java
+++ b/usr.bin/file/magdir/java
@@ -1,4 +1,4 @@
-#	$OpenBSD: java,v 1.4 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: java,v 1.5 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------
 # Java ByteCode and Mach-O binaries (e.g., Mac OS X) use the
@@ -9,3 +9,18 @@
 # From Martin Pool (m.pool@pharos.com.au)
 0	beshort		0xaced		Java serialization data
 >2	beshort		>0x0004		\b, version %d
+
+0	belong		0xfeedfeed	Java KeyStore
+!:mime	application/x-java-keystore
+0	belong		0xcececece	Java JCE KeyStore
+!:mime	application/x-java-jce-keystore
+
+# Dalvik .dex format. http://retrodev.com/android/dexformat.html
+# From <mkf@google.com> "Mike Fleming"
+0	string	dex\n
+>0	regex	dex\n[0-9][0-9][0-9]\0	Dalvik dex file
+>4	string	>000			version %s
+0	string	dey\n
+>0	regex	dey\n[0-9][0-9][0-9]\0	Dalvik dex file (optimized for host)
+>4	string	>000			version %s
+
diff --git a/usr.bin/file/magdir/jpeg b/usr.bin/file/magdir/jpeg
index 5f40f0c8140..246ae162e57 100644
--- a/usr.bin/file/magdir/jpeg
+++ b/usr.bin/file/magdir/jpeg
@@ -1,4 +1,4 @@
-#	$OpenBSD: jpeg,v 1.3 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: jpeg,v 1.4 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # JPEG images
@@ -10,6 +10,7 @@
 # both of which turn into "JPEG image data" here.
 #
 0	beshort		0xffd8		JPEG image data
+!:mime	image/jpeg
 >6	string		JFIF		\b, JFIF standard
 # The following added by Erik Rossen <rossen@freesurf.ch> 1999-09-06
 # in a vain attempt to add image size reporting for JFIF.  Note that these
@@ -154,3 +155,8 @@
 
 # From: David Santinoli <david@santinoli.com>
 0	string		\x00\x00\x00\x0C\x6A\x50\x20\x20\x0D\x0A\x87\x0A	JPEG 2000 image data
+
+# Type: JPEG 2000 codesream
+# From: Mathieu Malaterre <mathieu.malaterre@gmail.com>
+0	belong		0xff4fff51						JPEG 2000 codestream
+45	beshort		0xff52
diff --git a/usr.bin/file/magdir/kde b/usr.bin/file/magdir/kde
new file mode 100644
index 00000000000..d81ee693fd5
--- /dev/null
+++ b/usr.bin/file/magdir/kde
@@ -0,0 +1,10 @@
+
+#------------------------------------------------------------------------------
+# kde:  file(1) magic for KDE
+
+0		string	[KDE\ Desktop\ Entry]	KDE desktop entry
+!:mime	application/x-kdelnk
+0		string	#\ KDE\ Config\ File	KDE config file
+!:mime	application/x-kdelnk
+0		string	#\ xmcd	xmcd database file for kscd
+!:mime	text/x-xmcd
diff --git a/usr.bin/file/magdir/lex b/usr.bin/file/magdir/lex
index c881e1e2b71..9059bf05bd4 100644
--- a/usr.bin/file/magdir/lex
+++ b/usr.bin/file/magdir/lex
@@ -1,12 +1,12 @@
-#	$OpenBSD: lex,v 1.3 2004/06/03 03:14:20 tedu Exp $
+#	$OpenBSD: lex,v 1.4 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # lex:  file(1) magic for lex
 #
 #	derived empirically, your offsets may vary!
-53	string		yyprevious	C program text (from lex)
->3	string		>\0		 for %s
+0	search/100	yyprevious	C program text (from lex)
+>3	search/1	>\0		 for %s
 # C program text from GNU flex, from Daniel Quinlan <quinlan@yggdrasil.com>
-21	string		generated\ by\ flex	C program text (from flex)
+0	search/100	generated\ by\ flex	C program text (from flex)
 # lex description file, from Daniel Quinlan <quinlan@yggdrasil.com>
-0	string		%{		lex description text
+0	search/1	%{		lex description text
diff --git a/usr.bin/file/magdir/linux b/usr.bin/file/magdir/linux
index 6d78266de78..76e45a49efe 100644
--- a/usr.bin/file/magdir/linux
+++ b/usr.bin/file/magdir/linux
@@ -1,4 +1,4 @@
-#	$OpenBSD: linux,v 1.5 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: linux,v 1.6 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # linux:  file(1) magic for Linux files
@@ -104,16 +104,18 @@
 >0x1e9		string	Loading		from prehistoric times
 
 # System.map files - Nicolás Lichtmaier <nick@debian.org>
-8	string	\ A\ _text	Linux kernel symbol map text
+8	search/1	\ A\ _text	Linux kernel symbol map text
 
 # LSM entries - Nicolás Lichtmaier <nick@debian.org>
-0	string	Begin3	Linux Software Map entry text
-0	string	Begin4	Linux Software Map entry text (new format)
-
-# From Matt Zimmerman
-0       belong  0x4f4f4f4d      User-mode Linux COW file
->4      belong  x               \b, version %d
->8      string  >\0             \b, backing file %s
+0	search/1	Begin3	Linux Software Map entry text
+0	search/1	Begin4	Linux Software Map entry text (new format)
+
+# From Matt Zimmerman, enhanced for v3 by Matthew Palmer
+0	belong	0x4f4f4f4d	User-mode Linux COW file
+>4	belong	<3		\b, version %d
+>>8	string	>\0		\b, backing file %s
+>4	belong	>2		\b, version %d
+>>32	string	>\0		\b, backing file %s
 
 ############################################################################
 # Linux kernel versions
diff --git a/usr.bin/file/magdir/lisp b/usr.bin/file/magdir/lisp
index 10f127265f0..b5c3bfa09d2 100644
--- a/usr.bin/file/magdir/lisp
+++ b/usr.bin/file/magdir/lisp
@@ -1,4 +1,4 @@
-#	$OpenBSD: lisp,v 1.4 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: lisp,v 1.5 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # lisp:  file(1) magic for lisp programs
@@ -6,11 +6,12 @@
 # various lisp types, from Daniel Quinlan (quinlan@yggdrasil.com)
 
 # updated by Joerg Jenderek
-0	string	;;			
+# GRR: This lot is too weak
+#0	string	;;			
 # windows INF files often begin with semicolon and use CRLF as line end
 # lisp files are mainly created on unix system with LF as line end
->2	search/2048	!\r		Lisp/Scheme program text
->2	search/2048	\r		Windows INF file
+#>2	search/2048	!\r		Lisp/Scheme program text
+#>2	search/2048	\r		Windows INF file
 
 0	search/256	(if\ 			Lisp/Scheme program text
 0	search/256	(setq\ 			Lisp/Scheme program text
@@ -22,15 +23,17 @@
 
 # Emacs 18 - this is always correct, but not very magical.
 0	string	\012(			Emacs v18 byte-compiled Lisp data
+!:mime	application/x-elc
 # Emacs 19+ - ver. recognition added by Ian Springer
-# Also applies to XEmacs 19+ .elc files; could tell them apart if we had regexp
-# support or similar - Chris Chittleborough <cchittleborough@yahoo.com.au>
-0	string	;ELC			
->4	byte	>19			
+# Also applies to XEmacs 19+ .elc files; could tell them apart with regexs
+# - Chris Chittleborough <cchittleborough@yahoo.com.au>
+0	string	;ELC	
+>4	byte	>18			
 >4	byte    <32			Emacs/XEmacs v%d byte-compiled Lisp data
+!:mime	application/x-elc		
 
 # Files produced by CLISP Common Lisp From: Bruno Haible <haible@ilog.fr>
-0	string	(SYSTEM::VERSION\040'	CLISP byte-compiled Lisp program text
+0	string	(SYSTEM::VERSION\040'	CLISP byte-compiled Lisp program
 0	long	0x70768BD2		CLISP memory image data
 0	long	0xD28B7670		CLISP memory image data, other endian
 
@@ -42,4 +45,5 @@
 0	string	\372\372\372\372	MIT scheme (library?)
 
 # From: David Allouche <david@allouche.net>
-0	string	\<TeXmacs|	TeXmacs document text
+0	search/1	\<TeXmacs|	TeXmacs document text
+!:mime	text/texmacs
diff --git a/usr.bin/file/magdir/llvm b/usr.bin/file/magdir/llvm
new file mode 100644
index 00000000000..8c1610cd10e
--- /dev/null
+++ b/usr.bin/file/magdir/llvm
@@ -0,0 +1,10 @@
+
+#------------------------------------------------------------------------------
+# llvm:  file(1) magic for LLVM byte-codes
+# URL:  http://llvm.cs.uiuc.edu/docs/BytecodeFormat.html#signature
+# From: Al Stone <ahs3@fc.hp.com>
+
+0	string	llvm	LLVM byte-codes, uncompressed
+0	string	llvc0	LLVM byte-codes, null compression
+0	string	llvc1	LLVM byte-codes, gzip compression
+0	string	llvc2	LLVM byte-codes, bzip2 compression
diff --git a/usr.bin/file/magdir/lua b/usr.bin/file/magdir/lua
new file mode 100644
index 00000000000..303fd3798e1
--- /dev/null
+++ b/usr.bin/file/magdir/lua
@@ -0,0 +1,15 @@
+#------------------------------------------------------------------------------
+# lua:  file(1) magic for Lua scripting language
+# URL:  http://www.lua.org/
+# From: Reuben Thomas <rrt@sc3d.org>, Seo Sanghyeon <tinuviel@sparcs.kaist.ac.kr>
+
+# Lua scripts
+0	search/1/b	#!\ /usr/bin/lua	Lua script text executable
+0	search/1/b	#!\ /usr/local/bin/lua	Lua script text executable
+0	search/1	#!/usr/bin/env\ lua	Lua script text executable
+0	search/1	#!\ /usr/bin/env\ lua	Lua script text executable
+
+# Lua bytecode
+0	string		\033Lua			Lua bytecode,
+>4	byte		0x50			version 5.0
+>4	byte		0x51			version 5.1
diff --git a/usr.bin/file/magdir/luks b/usr.bin/file/magdir/luks
new file mode 100644
index 00000000000..2ab23935ef6
--- /dev/null
+++ b/usr.bin/file/magdir/luks
@@ -0,0 +1,12 @@
+
+#------------------------------------------------------------------------------
+# luks:  file(1) magic for Linux Unified Key Setup
+# URL:	http://luks.endorphin.org/spec
+# From:	Anthon van der Neut <anthon@mnt.org>
+
+0	string		LUKS\xba\xbe	LUKS encrypted file,
+>6	beshort		x		ver %d
+>8	string		x		[%s,
+>40	string		x		%s,
+>72	string		x		%s]
+>168	string		x		UUID: %s
diff --git a/usr.bin/file/magdir/macintosh b/usr.bin/file/magdir/macintosh
index c7dbea573a7..3ad6da5b61a 100644
--- a/usr.bin/file/magdir/macintosh
+++ b/usr.bin/file/magdir/macintosh
@@ -1,4 +1,4 @@
-#	$OpenBSD: macintosh,v 1.3 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: macintosh,v 1.4 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # macintosh description
@@ -6,6 +6,7 @@
 # BinHex is the Macintosh ASCII-encoded file format (see also "apple")
 # Daniel Quinlan, quinlan@yggdrasil.com
 11	string	must\ be\ converted\ with\ BinHex	BinHex binary text
+!:mime	application/mac-binhex40
 >41	string	x					\b, version %.3s
 
 # Stuffit archives are the de facto standard of compression for Macintosh
@@ -19,37 +20,41 @@
 
 # Newer StuffIt archives (grant@netbsd.org)
 0	string		StuffIt			StuffIt Archive
+!:mime	application/x-stuffit
 #>162	string		>0			: %s
 
 # Macintosh Applications and Installation binaries (franklsm@tuns.ca)
-0	string		APPL			Macintosh Application (data)
->2	string		x			\b: %s
+# GRR: Too weak
+#0	string		APPL			Macintosh Application (data)
+#>2	string		x			\b: %s
 
 # Macintosh System files (franklsm@tuns.ca)
-0	string		zsys			Macintosh System File (data)
-0	string		FNDR			Macintosh Finder (data)
-0	string		libr			Macintosh Library (data)
->2	string		x			: %s
-0	string		shlb			Macintosh Shared Library (data)
->2	string		x			: %s
-0	string		cdev			Macintosh Control Panel (data)
->2	string		x			: %s
-0	string		INIT			Macintosh Extension (data)
->2	string		x			: %s
-0	string		FFIL			Macintosh Truetype Font (data)
->2	string		x			: %s
-0	string		LWFN			Macintosh Postscript Font (data)
->2	string		x			: %s
+# GRR: Too weak
+#0	string		zsys			Macintosh System File (data)
+#0	string		FNDR			Macintosh Finder (data)
+#0	string		libr			Macintosh Library (data)
+#>2	string		x			: %s
+#0	string		shlb			Macintosh Shared Library (data)
+#>2	string		x			: %s
+#0	string		cdev			Macintosh Control Panel (data)
+#>2	string		x			: %s
+#0	string		INIT			Macintosh Extension (data)
+#>2	string		x			: %s
+#0	string		FFIL			Macintosh Truetype Font (data)
+#>2	string		x			: %s
+#0	string		LWFN			Macintosh Postscript Font (data)
+#>2	string		x			: %s
 
 # Additional Macintosh Files (franklsm@tuns.ca)
-0	string		PACT			Macintosh Compact Pro Archive (data)
->2	string		x			: %s
-0	string		ttro			Macintosh TeachText File (data)
->2	string		x			: %s
-0	string		TEXT			Macintosh TeachText File (data)
->2	string		x			: %s
-0	string		PDF			Macintosh PDF File (data)
->2	string		x			: %s
+# GRR: Too weak
+#0	string		PACT			Macintosh Compact Pro Archive (data)
+#>2	string		x			: %s
+#0	string		ttro			Macintosh TeachText File (data)
+#>2	string		x			: %s
+#0	string		TEXT			Macintosh TeachText File (data)
+#>2	string		x			: %s
+#0	string		PDF			Macintosh PDF File (data)
+#>2	string		x			: %s
 
 # MacBinary format (Eric Fischer, enf@pobox.com)
 #
@@ -321,41 +326,50 @@
 # shorter than 32 bytes must be terminated with NULL" so I'll treat it as a 
 # cstring. Of course, partitions can contain more than four entries, but 
 # what're you gonna do?
-0x200		beshort		0x504D		Apple Partition data
->0x2		beshort		x		block size: %d,
->0x230		string		x		first type: %s,
->0x210		string		x		name: %s,
->0x254		belong		x		number of blocks: %d,
->0x400		beshort		0x504D		
->>0x430		string		x		second type: %s,
->>0x410		string		x		name: %s,
->>0x454		belong		x		number of blocks: %d,
->>0x800		beshort		0x504D		
->>>0x830	string		x		third type: %s,
->>>0x810	string		x		name: %s,
->>>0x854	belong		x		number of blocks: %d,
->>>0xa00	beshort		0x504D		
->>>>0xa30	string		x		fourth type: %s,
->>>>0xa10	string		x		name: %s,
->>>>0xa54	belong		x		number of blocks: %d
-# AFAIK, only the signature is different
-0x200		beshort		0x5453		Apple Old Partition data
->0x2		beshort		x		block size: %d,
->0x230		string		x		first type: %s,
->0x210		string		x		name: %s,
->0x254		belong		x		number of blocks: %d,
->0x400		beshort		0x504D		
->>0x430		string		x		second type: %s,
->>0x410		string		x		name: %s,
->>0x454		belong		x		number of blocks: %d,
->>0x800		beshort		0x504D		
->>>0x830	string		x		third type: %s,
->>>0x810	string		x		name: %s,
->>>0x854	belong		x		number of blocks: %d,
->>>0xa00	beshort		0x504D		
->>>>0xa30	string		x		fourth type: %s,
->>>>0xa10	string		x		name: %s,
->>>>0xa54	belong		x		number of blocks: %d
+# GRR: This magic is too weak, it is just "PM"
+#0x200		beshort		0x504D		Apple Partition data
+#>0x2		beshort		x		(block size: %d):
+#>0x230		string		x		first type: %s,
+#>0x210		string		x		name: %s,
+#>0x254		belong		x		number of blocks: %d,
+#>0x400		beshort		0x504D		
+#>>0x430		string		x		second type: %s,
+#>>0x410		string		x		name: %s,
+#>>0x454		belong		x		number of blocks: %d,
+#>>0x600		beshort		0x504D
+#>>>0x630	string		x		third type: %s,
+#>>>0x610	string		x		name: %s,
+#>>>0x654	belong		x		number of blocks: %d,
+#>>0x800		beshort		0x504D		
+#>>>0x830	string		x		fourth type: %s,
+#>>>0x810	string		x		name: %s,
+#>>>0x854	belong		x		number of blocks: %d,
+#>>>0xa00	beshort		0x504D		
+#>>>>0xa30	string		x		fifth type: %s,
+#>>>>0xa10	string		x		name: %s,
+#>>>>0xa54	belong		x		number of blocks: %d
+#>>>0xc00	beshort		0x504D
+#>>>>0xc30	string		x		sixth type: %s,
+#>>>>0xc10	string		x		name: %s,
+#>>>>0xc54	belong		x		number of blocks: %d
+## AFAIK, only the signature is different
+#0x200		beshort		0x5453		Apple Old Partition data
+#>0x2		beshort		x		block size: %d,
+#>0x230		string		x		first type: %s,
+#>0x210		string		x		name: %s,
+#>0x254		belong		x		number of blocks: %d,
+#>0x400		beshort		0x504D		
+#>>0x430		string		x		second type: %s,
+#>>0x410		string		x		name: %s,
+#>>0x454		belong		x		number of blocks: %d,
+#>>0x800		beshort		0x504D		
+#>>>0x830	string		x		third type: %s,
+#>>>0x810	string		x		name: %s,
+#>>>0x854	belong		x		number of blocks: %d,
+#>>>0xa00	beshort		0x504D		
+#>>>>0xa30	string		x		fourth type: %s,
+#>>>>0xa10	string		x		name: %s,
+#>>>>0xa54	belong		x		number of blocks: %d
 
 # From: Remi Mommsen <mommsen@slac.stanford.edu>
-0		string		BOMStore	Mac OS X bill of materials (BOM) fil
+0		string		BOMStore	Mac OS X bill of materials (BOM) file
diff --git a/usr.bin/file/magdir/mail.news b/usr.bin/file/magdir/mail.news
index 66b35309085..d8902728c71 100644
--- a/usr.bin/file/magdir/mail.news
+++ b/usr.bin/file/magdir/mail.news
@@ -1,4 +1,4 @@
-#	$OpenBSD: mail.news,v 1.5 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: mail.news,v 1.6 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # mail.news:  file(1) magic for mail and news
@@ -7,17 +7,28 @@
 #0	string		From 		mail text
 # There are tests to ascmagic.c to cope with mail and news.
 0	string		Relay-Version: 	old news text
+!:mime	message/rfc822
 0	string		#!\ rnews	batched news text
+!:mime	message/rfc822
 0	string		N#!\ rnews	mailed, batched news text
+!:mime	message/rfc822
 0	string		Forward\ to 	mail forwarding text
+!:mime	message/rfc822
 0	string		Pipe\ to 	mail piping text
+!:mime	message/rfc822
 0	string		Return-Path:	smtp mail text
+!:mime	message/rfc822
 0	string		Path:		news text
+!:mime	message/news
 0	string		Xref:		news text
+!:mime	message/news
 0	string		From:		news or mail text
+!:mime	message/rfc822
 0	string		Article 	saved news text
+!:mime	message/news
 0	string		BABYL		Emacs RMAIL text
 0	string		Received:	RFC 822 mail text
+!:mime	message/rfc822
 0	string		MIME-Version:	MIME entity text
 #0	string		Content-	MIME entity text
 
@@ -40,3 +51,5 @@
 # XXX: Weak magic
 #256	leshort	0xAFAE4453		Squish message area data file
 #>4	leshort	>0			(%d messages)
+
+#0	string		\<!--\ MHonArc		text/html; x-type=mhonarc
diff --git a/usr.bin/file/magdir/mathcad b/usr.bin/file/magdir/mathcad
new file mode 100644
index 00000000000..41801968888
--- /dev/null
+++ b/usr.bin/file/magdir/mathcad
@@ -0,0 +1,7 @@
+
+#------------------------------------------------------------------------------
+# mathcad:  file(1) magic for Mathcad documents
+# URL:	http://www.mathsoft.com/
+# From:	Josh Triplett <josh@freedesktop.org>
+
+0	string	.MCAD\t		Mathcad document
diff --git a/usr.bin/file/magdir/mathematica b/usr.bin/file/magdir/mathematica
index 0738918d1e4..2884d4eb44e 100644
--- a/usr.bin/file/magdir/mathematica
+++ b/usr.bin/file/magdir/mathematica
@@ -1,4 +1,4 @@
-#	$OpenBSD: mathematica,v 1.2 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: mathematica,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # mathematica:  file(1) magic for mathematica files
@@ -57,7 +57,8 @@
 0	string	(***********************	Mathematica 3.0 notebook
 
 # other (* matches it is a comment start in these langs
-0	string	(*	Mathematica, or Pascal,  Modula-2 or 3 code text
+# GRR: Too weak; also matches other languages e.g. ML
+#0	string	(*	Mathematica, or Pascal, Modula-2 or 3 code text
 
 #########################
 # MatLab v5
diff --git a/usr.bin/file/magdir/mercurial b/usr.bin/file/magdir/mercurial
new file mode 100644
index 00000000000..3f1bdfc8f9c
--- /dev/null
+++ b/usr.bin/file/magdir/mercurial
@@ -0,0 +1,12 @@
+
+#------------------------------------------------------------------------------
+# mercurial:  file(1) magic for Mercurial changeset bundles
+# http://www.selenic.com/mercurial/wiki/
+#
+# Jesse Glick (jesse.glick@sun.com)
+#
+
+0	string		HG10		Mercurial changeset bundle
+>4	string		UN		(uncompressed)
+>4	string		GZ		(gzip compressed)
+>4	string		BZ		(bzip2 compressed)
diff --git a/usr.bin/file/magdir/misctools b/usr.bin/file/magdir/misctools
index 9715119e4cc..cfbe2753c46 100644
--- a/usr.bin/file/magdir/misctools
+++ b/usr.bin/file/magdir/misctools
@@ -1,11 +1,12 @@
-#	$OpenBSD: misctools,v 1.2 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: misctools,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #-----------------------------------------------------------------------------
-# misctools:  file(1) magic for miscelanous UNIX tools.
+# misctools:  file(1) magic for miscellaneous UNIX tools.
 #
-0	string	%%!!			X-Post-It-Note text
+0	search/1	%%!!			X-Post-It-Note text
 0	string/c	BEGIN:VCALENDAR		vCalendar calendar file
 0	string/c	BEGIN:VCARD		vCard visiting card
+!:mime	text/x-vcard
 
 # From: Alex Beregszaszi <alex@fsn.hu>
 4	string	gtktalog		GNOME Catalogue (gtktalog)
diff --git a/usr.bin/file/magdir/mozilla b/usr.bin/file/magdir/mozilla
new file mode 100644
index 00000000000..069886064fb
--- /dev/null
+++ b/usr.bin/file/magdir/mozilla
@@ -0,0 +1,8 @@
+
+#------------------------------------------------------------------------------
+# mozilla:  file(1) magic for Mozilla XUL fastload files 
+# (XUL.mfasl and XPC.mfasl)
+# URL:	http://www.mozilla.org/
+# From:	Josh Triplett <josh@freedesktop.org>
+
+0	string	XPCOM\nMozFASL\r\n\x1A		Mozilla XUL fastload data
diff --git a/usr.bin/file/magdir/msdos b/usr.bin/file/magdir/msdos
index fe537ac4b88..2a707af65c8 100644
--- a/usr.bin/file/magdir/msdos
+++ b/usr.bin/file/magdir/msdos
@@ -1,4 +1,4 @@
-#	$OpenBSD: msdos,v 1.4 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: msdos,v 1.5 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # msdos:  file(1) magic for MS-DOS files
@@ -18,10 +18,6 @@
 100	regex/c =^[\ \t]{0,10}call[\ \t]{1,10}rxfunc OS/2 REXX batch file text
 100	regex/c =^[\ \t]{0,10}say\ ['"]	     OS/2 REXX batch file text
 
-
-100 regex/c =^\\s*call\\s+rxfuncadd.*sysloadfu OS/2 REXX batch file text
-100 regex/c =^\\s*say\ ['"] OS/2 REXX batch file text
-
 0	leshort		0x14c	MS Windows COFF Intel 80386 object file
 #>4	ledate		x	stamp %s
 0	leshort		0x166	MS Windows COFF MIPS R4000 object file
@@ -51,6 +47,7 @@
 # many of the compressed formats were extraced from IDARC 1.23 source code
 #
 0	string	MZ		MS-DOS executable
+!:mime	application/x-dosexec
 >0 string MZ\0\0\0\0\0\0\0\0\0\0PE\0\0 \b, PE for MS Windows
 >>&18	leshort&0x2000	>0	(DLL)
 >>&88	leshort		0	(unknown subsystem)
@@ -69,6 +66,11 @@
 >>&18	leshort&0x1000	>0	system file
 >>&0xf4 search/0x140 \x0\x40\x1\x0
 >>>(&0.l+(4)) string MSCF \b, WinHKI CAB self-extracting archive
+>30		string	Copyright\ 1989-1990\ PKWARE\ Inc.	Self-extracting PKZIP archive
+!:mime	application/zip
+# Is next line correct? One might expect "Corp." not "Copr." If it is right, add a note to that effect.
+>30		string	PKLITE\ Copr.	Self-extracting PKZIP archive
+!:mime	application/zip
 
 >0x18  leshort >0x3f
 >>(0x3c.l) string PE\0\0 PE 
@@ -220,7 +222,9 @@
 >0x1e	string	PKLITE \b, %s compressed
 >0x64   string  W\ Collis\0\0 \b, Compack compressed
 >0x24	string	LHa's\ SFX \b, LHa self-extracting archive
+!:mime	application/x-lha
 >0x24	string	LHA's\ SFX \b, LHa self-extracting archive
+!:mime	application/x-lha
 >0x24   string  \ $ARX \b, ARX self-extracting archive
 >0x24   string  \ $LHarc \b, LHarc self-extracting archive
 >0x20   string  SFX\ by\ LARC \b, LARC self-extracting archive
@@ -267,8 +271,8 @@
 # Uncommenting only the first two lines will cover about 2/3 of COM files,
 # but it isn't feasible to match all COM files since there must be at least
 # two dozen different one-byte "magics".
-0	byte		0xe9		DOS executable (COM)
->0x1FE	leshort		0xAA55		\b, boot code
+#0	byte		0xe9		DOS executable (COM)
+#>0x1FE	leshort		0xAA55		\b, boot code
 >6	string		SFX\ of\ LHarc	(%s)
 0	belong	0xffffffff		DOS executable (device driver)
 #CMD640X2.SYS
@@ -291,15 +295,15 @@
 >>77	string	>\x40			
 >>>77	string	<\x5B			
 >>>>77	string	x			\b, name: %.8s
-0	byte		0x8c		DOS executable (COM)
+#0	byte		0x8c		DOS executable (COM)
 # 0xeb conflicts with "sequent" magic
-0	byte		0xeb		DOS executable (COM)
->0x1FE	leshort		0xAA55		\b, boot code
->85	string		UPX		\b, UPX compressed
->4	string		\ $ARX		\b, ARX self-extracting archive
->4	string		\ $LHarc	\b, LHarc self-extracting archive
->0x20e	string		SFX\ by\ LARC	\b, LARC self-extracting archive
-0	byte		0xb8		COM executable
+#0	byte		0xeb		DOS executable (COM)
+#>0x1FE	leshort		0xAA55		\b, boot code
+#>85	string		UPX		\b, UPX compressed
+#>4	string		\ $ARX		\b, ARX self-extracting archive
+#>4	string		\ $LHarc	\b, LHarc self-extracting archive
+#>0x20e	string		SFX\ by\ LARC	\b, LARC self-extracting archive
+#0	byte		0xb8		COM executable
 # modified by Joerg Jenderek
 >1	lelong          !0x21cd4cff	for DOS
 # http://syslinux.zytor.com/comboot.php
@@ -364,38 +368,58 @@
 >30	byte	12		(4kB sectors)
 
 # Popular applications
-# False positive with PPT
-#0       string \xD0\xCF\x11\xE0\xA1\xB1\x1A\xE1\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x3E\x00\x03\x00\xFE\xFF Microsoft Installer
 2080	string	Microsoft\ Word\ 6.0\ Document	%s
+!:mime	application/msword
 2080	string	Documento\ Microsoft\ Word\ 6 Spanish Microsoft Word 6 document data
+!:mime	application/msword
 # Pawel Wiecek <coven@i17linuxb.ists.pwr.wroc.pl> (for polish Word)
 2112	string	MSWordDoc			Microsoft Word document data
+!:mime	application/msword
 #
 0	belong	0x31be0000			Microsoft Word Document
+!:mime	application/msword
 #
 0       string  PO^Q`				Microsoft Word 6.0 Document
+!:mime	application/msword
 #
 0	string	\376\067\0\043			Microsoft Office Document
+!:mime	application/msword
 0	string	\333\245-\0\0\0			Microsoft Office Document
+!:mime	application/msword
+512	string		\354\245\301		Microsoft Word Document
+!:mime	application/msword
 #
 2080	string	Microsoft\ Excel\ 5.0\ Worksheet	%s
+!:mime	application/vnd.ms-excel
+
 2080	string	Foglio\ di\ lavoro\ Microsoft\ Exce	%s
+!:mime	application/vnd.ms-excel
 #
 # Pawel Wiecek <coven@i17linuxb.ists.pwr.wroc.pl> (for polish Excel)
 2114	string	Biff5		Microsoft Excel 5.0 Worksheet
+!:mime	application/vnd.ms-excel
 # Italian MS-Excel
 2121	string	Biff5		Microsoft Excel 5.0 Worksheet
+!:mime	application/vnd.ms-excel
 0	string	\x09\x04\x06\x00\x00\x00\x10\x00	Microsoft Excel Worksheet
+!:mime	application/vnd.ms-excel
 #
 0	belong	0x00001a00	Lotus 1-2-3
+!:mime	application/x-123
 >4	belong	0x00100400	wk3 document data
 >4	belong	0x02100400	wk4 document data
 >4	belong	0x07800100	fm3 or fmb document data
 >4	belong	0x07800000	fm3 or fmb document data
 #
 0	belong	0x00000200 	Lotus 1-2-3
+!:mime	application/x-123
 >4	belong	0x06040600	wk1 document data
 >4	belong	0x06800200	fmt document data
+0	string		WordPro\0	Lotus WordPro
+!:mime	application/vnd.lotus-wordpro
+0	string		WordPro\r\373	Lotus WordPro
+!:mime	application/vnd.lotus-wordpro
+
 
 # Help files
 0	string	?_\3\0		MS Windows Help Data
@@ -449,6 +473,7 @@
 
 # Windows icons (Ian Springer <ips@fpk.hp.com>)
 0	string	\000\000\001\000	MS Windows icon resource
+!:mime	image/x-ico
 >4	byte	1			- 1 icon
 >4	byte	>1			- %d icons
 >>6	byte	>0			\b, %dx
@@ -523,7 +548,9 @@
 >>>>24	long		>0		length %d
 
 # TNEF magic From "Joomy" <joomy@se-ed.net> 
+# Microsoft Outlook's Transport Neutral Encapsulation Format (TNEF)
 0	leshort		0x223e9f78	TNEF
+!:mime	application/vnd.ms-tnef
 
 # HtmlHelp files (.chm)
 0	string  ITSF\003\000\000\000\x60\000\000\000\001\000\000\000	MS Windows HtmlHelp Data
@@ -587,17 +614,17 @@
 # syntax available by some file commands you could write:
 # 0 lelong 1
 # &40 ulelong 0x464D4520 Windows Enhanced Metafile (EMF) image data
-40	ulelong 0x464D4520	Windows Enhanced Metafile (EMF) image data
->44	ulelong x		version 0x%x.
+#40	ulelong 0x464D4520	Windows Enhanced Metafile (EMF) image data
+#>44	ulelong x		version 0x%x.
 # If the description has a length greater than zero, it exists and is 
 # found at offset (*64).
->64	ulelong >0		Description available at offset 0x%x
->>60	ulelong	>0		(length 0x%x)
+#>64	ulelong >0		Description available at offset 0x%x
+#>>60	ulelong	>0		(length 0x%x)
 # Note it would be better to print out the description, which is found 
 # as below. Unfortunately the following only prints out the first couple
 # of characters instead of all the "description length"
 # number of characters -- indicated by the ulelong at offset 60.
->>(64.l)  lestring16 >0 Description: %15.15s
+#>>(64.l)  lestring16 >0 Description: %15.15s
 
 # From: Alex Beregszaszi <alex@fsn.hu>
 0	string	COWD		VMWare3
@@ -611,10 +638,44 @@
 0	string	VMDK		 VMware4 disk image
 0	string	KDMV		 VMware4 disk image
 
-0	belong	0x514649fb	QEMU Copy-On-Write disk image
->4	belong	x		version %d,
->24	belong	x		size %d +
->28	belong	x		%d
+#--------------------------------------------------------------------
+# Qemu Emulator Images
+# Lines written by Friedrich Schwittay (f.schwittay@yousable.de)
+# Made by reading sources and doing trial and error on existing
+# qcow files
+0       string  QFI     Qemu Image, Format: Qcow
+
+# Uncomment the following line to display Magic (only used for debugging
+# this magic number)
+#>0     string  x       , Magic: %s
+
+# There are currently 2 Versions: "1" and "2"
+# I do not use Version 2 and therefor branch here
+# but can assure: it works (tested on both versions)
+# Also my Qemu 0.9.0 which uses this Version 2 refuses
+# to start in its bios
+>0x04   belong  2       , Version: 2
+>0x04   belong  1       , Version: 1
+
+# Using the existence of the Backing File Offset to Branch or not
+# to read Backing File Information
+>>0xc    belong  >0      , Backing File( Offset: %lu
+>>>(0xc.L)       string >\0     , Path: %s
+
+# Didn't get the trick here how qemu stores the "Size" at this Position
+# There is actually something stored but nothing makes sense
+# The header in the sources talks about it
+#>>>16   lelong  x       , Size: %lu
+
+# Modification time of the Backing File
+# Really useful if you want to know if your backing
+# file is still usable together with this image
+>>>20    bedate x       , Mtime: %s )
+
+# Don't know how to calculate in Magicfiles
+# Also: this Information is not reliably
+#       stored in image-files
+>>24     lelong  x       , Disk Size could be: %d * 256 bytes
 
 0	string	QEVM		QEMU's suspend to disk image
 
@@ -625,5 +686,43 @@
 0	lelong	0x02468ace			Bochs Sparse disk image
 
 # from http://filext.com by Derek M Jones <derek@knosof.co.uk>
-0	string	\xD0\xCF\x11\xE0\xA1\xB1\x1A\xE1\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x3E\x00\x03\x00\xFE\xFF	Microsoft Installer
+# False positive with PPT (also currently this string is too long)
+#0	string	\xD0\xCF\x11\xE0\xA1\xB1\x1A\xE1\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x3E\x00\x03\x00\xFE\xFF\x09\x00\x06	Microsoft Installer
 0	string	\320\317\021\340\241\261\032\341	Microsoft Office Document
+#>48	byte	0x1B	                                Excel Document
+#!:mime	application/vnd.ms-excel
+>546	string	bjbj			Microsoft Word Document
+!:mime	application/msword
+>546	string	jbjb			Microsoft Word Document
+!:mime	application/msword
+
+0       string	\224\246\056		Microsoft Word Document
+!:mime	application/msword
+
+512	string	R\0o\0o\0t\0\ \0E\0n\0t\0r\0y	Microsoft Word Document
+!:mime	application/msword
+
+# From: "Nelson A. de Oliveira" <naoliv@gmail.com>
+# Magic type for Dell's BIOS .hdr files
+# Dell's .hdr
+0	string $RBU
+>23	string Dell			%s system BIOS
+>48	string x			version %.3s
+
+# Type: Microsoft DirectDraw Surface
+# URL:  http://msdn.microsoft.com/library/default.asp?url=/library/en-us/directx9_c/directx/graphics/reference/DDSFileReference/ddsfileformat.asp
+# From: Morten Hustveit <morten@debian.org>
+0	string	DDS\040\174\000\000\000	Microsoft DirectDraw Surface (DDS),
+>16	lelong	>0			%hd x
+>12	lelong	>0			%hd,
+>84	string	x			%.4s
+
+# Type: Microsoft Document Imaging Format (.mdi)
+# URL:  http://en.wikipedia.org/wiki/Microsoft_Document_Imaging_Format
+# From: Daniele Sempione <scrows@oziosi.org>
+0	short	0x5045			Microsoft Document Imaging Format
+
+# MS eBook format (.lit)
+0	string  ITOLITLS		Microsoft Reader eBook Data
+>8	lelong	x			\b, version %u
+!:mime					application/x-ms-reader
diff --git a/usr.bin/file/magdir/mup b/usr.bin/file/magdir/mup
index d78242017f9..c7a1f0e0dd1 100644
--- a/usr.bin/file/magdir/mup
+++ b/usr.bin/file/magdir/mup
@@ -1,4 +1,4 @@
-#	$OpenBSD: mup,v 1.2 2008/05/14 20:45:35 chl Exp $
+#	$OpenBSD: mup,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 # ------------------------------------------------------------------------
 # mup: file(1) magic for Mup (Music Publisher) input file.
@@ -10,7 +10,7 @@
 # compatibility. Noteedit also use mup format, but is not forcing
 # user to use any header as well.
 #
-0		string		//!Mup		Mup music publication program input text
+0		search/1	//!Mup		Mup music publication program input text
 >6		string		-Arkkra		(Arkkra)
 >>13		string		-		
 >>>16		string		.		
diff --git a/usr.bin/file/magdir/netware b/usr.bin/file/magdir/netware
new file mode 100644
index 00000000000..fbc16054a32
--- /dev/null
+++ b/usr.bin/file/magdir/netware
@@ -0,0 +1,6 @@
+
+#------------------------------------------------------------------------------
+# netware:  file(1) magic for NetWare Loadable Modules (NLMs)
+# From: Mads Martin Joergensen <mmj@suse.de>
+
+0	string	NetWare\ Loadable\ Module	NetWare Loadable Module
diff --git a/usr.bin/file/magdir/ole2compounddocs b/usr.bin/file/magdir/ole2compounddocs
new file mode 100644
index 00000000000..ee07514ec5b
--- /dev/null
+++ b/usr.bin/file/magdir/ole2compounddocs
@@ -0,0 +1,13 @@
+
+#------------------------------------------------------------------------------
+# Microsoft OLE 2 Compound Documents : file(1) magic for Microsoft Structured
+# storage (http://en.wikipedia.org/wiki/Structured_Storage)
+# Additional tests for OLE 2 Compound Documents should be under this recipe.
+
+0   string  \320\317\021\340\241\261\032\341      OLE 2 Compound Document
+# - Microstation V8 DGN files (www.bentley.com)
+#   Last update on 10/23/2006 by Lester Hightower
+> 0x480  string  D\000g\000n\000~\000H                : Microstation V8 DGN
+# - Visio documents
+#   Last update on 10/23/2006 by Lester Hightower
+> 0x480  string  V\000i\000s\000i\000o\000D\000o\000c : Visio Document
diff --git a/usr.bin/file/magdir/os2 b/usr.bin/file/magdir/os2
index 91506ec9394..6622bb79f9a 100644
--- a/usr.bin/file/magdir/os2
+++ b/usr.bin/file/magdir/os2
@@ -1,4 +1,4 @@
-#	$OpenBSD: os2,v 1.2 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: os2,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # os2:  file(1) magic for OS/2 files
@@ -6,8 +6,8 @@
 
 # Provided 1998/08/22 by
 # David Mediavilla <davidme.news@REMOVEIFNOTSPAMusa.net>
-1	string	InternetShortcut	MS Windows 95 Internet shortcut text
->24	string	>\ 			(URL=<%s>)
+1	search/1	InternetShortcut	MS Windows 95 Internet shortcut text
+>24	search/1	>\ 			(URL=<%s>)
 
 # OS/2 URL objects
 # Provided 1998/08/22 by
diff --git a/usr.bin/file/magdir/palm b/usr.bin/file/magdir/palm
index 1c7afb77cde..903649af96d 100644
--- a/usr.bin/file/magdir/palm
+++ b/usr.bin/file/magdir/palm
@@ -1,4 +1,4 @@
-#	$OpenBSD: palm,v 1.2 2008/05/14 20:45:35 chl Exp $
+#	$OpenBSD: palm,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # palm:  file(1) magic for PalmOS {.prc,.pdb}: applications, docfiles, and hacks
@@ -46,7 +46,7 @@
 >0	string                  >\0             "%s"
 60	string	                SM01SMem	SuperMemo PalmOS document
 >0	string                  >\0             "%s"
-60	string	                DataTlPt	TealDoc PalmOS document
+60	string	                TEXtTlDc	TealDoc PalmOS document
 >0	string                  >\0             "%s"
 60	string	                InfoTlIf	TealInfo PalmOS document
 >0	string                  >\0             "%s"
diff --git a/usr.bin/file/magdir/pdf b/usr.bin/file/magdir/pdf
index a044c14f809..38d46e9dfed 100644
--- a/usr.bin/file/magdir/pdf
+++ b/usr.bin/file/magdir/pdf
@@ -1,9 +1,16 @@
-#	$OpenBSD: pdf,v 1.3 2001/01/29 01:57:57 niklas Exp $
+#	$OpenBSD: pdf,v 1.4 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # pdf:  file(1) magic for Portable Document Format
 #
 
 0	string		%PDF-		PDF document
+!:mime	application/pdf
 >5	byte		x		\b, version %c
 >7	byte		x		\b.%c
+
+# From: Nick Schmalenberger <nick@schmalenberger.us>
+# Forms Data Format
+0       string          %FDF-           FDF document
+>5      byte            x               \b, version %c
+>7      byte            x               \b.%c
diff --git a/usr.bin/file/magdir/perl b/usr.bin/file/magdir/perl
index 1fe81b042a4..5a9b8526f0b 100644
--- a/usr.bin/file/magdir/perl
+++ b/usr.bin/file/magdir/perl
@@ -1,27 +1,34 @@
-#	$OpenBSD: perl,v 1.2 2008/05/14 20:45:35 chl Exp $
+#	$OpenBSD: perl,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # perl:  file(1) magic for Larry Wall's perl language.
 #
-# The ``eval'' line recognizes an outrageously clever hack for USG systems.
+# The `eval' lines recognizes an outrageously clever hack.
 # Keith Waclena <keith@cerberus.uchicago.edu>
 # Send additions to <perl5-porters@perl.org>
-0	string/b	#!\ /bin/perl			perl script text executable
-0	string		eval\ "exec\ /bin/perl		perl script text
-0	string/b	#!\ /usr/bin/perl		perl script text executable
-0	string		eval\ "exec\ /usr/bin/perl	perl script text
-0	string/b	#!\ /usr/local/bin/perl		perl script text
-0	string		eval\ "exec\ /usr/local/bin/perl	perl script text executable
-0	string		eval\ '(exit\ $?0)'\ &&\ eval\ 'exec	perl script text
+0	search/1/b	#!\ /bin/perl			Perl script text executable
+!:mime	application/x-perl
+0	search/1	eval\ "exec\ /bin/perl		Perl script text
+!:mime	application/x-perl
+0	search/1/b	#!\ /usr/bin/perl		Perl script text executable
+!:mime	application/x-perl
+0	search/1	eval\ "exec\ /usr/bin/perl	Perl script text
+!:mime	application/x-perl
+0	search/1/b	#!\ /usr/local/bin/perl		Perl script text executable
+!:mime	application/x-perl
+0	search/1	eval\ "exec\ /usr/local/bin/perl	Perl script text
+!:mime	application/x-perl
+0	search/1	eval\ '(exit\ $?0)'\ &&\ eval\ 'exec	Perl script text
+!:mime	application/x-perl
 
 
 # by Dmitry V. Levin and Alexey Tourbin
 # check the first line
-0	string		package
->1	regex		\^package[\ \t]+[A-Za-z_]
->>1	regex		\^package[\ \t]+[0-9A-Za-z_:]*\ *;	Perl5 module source text
+0	search/1	package
+0	regex		\^package[\ \t]+[A-Za-z_]
+>0	regex		\^package[\ \t]+[0-9A-Za-z_:]+\ *;	Perl5 module source text
 # not 'p', check other lines
-0	byte		!0x70
+0	search/1	!p
 >0	regex		\^package[\ \t]+[0-9A-Za-z_:]+\ *;
 >>0	regex		\^1\ *;|\^(use|sub|my)\ .*[(;{=]	Perl5 module source text
 
@@ -29,21 +36,21 @@
 
 # Perl POD documents
 # From: Tom Hukins <tom@eborcom.com>
-0	string/B	\=pod\n		Perl POD document
-0	string/B	\n\=pod\n	Perl POD document
-0	string/B	\=head1\ 	Perl POD document
-0	string/B	\n\=head1\ 	Perl POD document
-0	string/B	\=head2\ 	Perl POD document
-0	string/B	\n\=head2\ 	Perl POD document
+0	search/1/B	\=pod\n		Perl POD document text
+0	search/1/B	\n\=pod\n	Perl POD document text
+0	search/1/B	\=head1\ 	Perl POD document text
+0	search/1/B	\n\=head1\ 	Perl POD document text
+0	search/1/B	\=head2\ 	Perl POD document text
+0	search/1/B	\n\=head2\ 	Perl POD document text
 
 # Perl Storable data files.
-0	string	perl-store	perl Storable(v0.6) data
+0	string	perl-store	perl Storable (v0.6) data
 >4	byte	>0	(net-order %d)
 >>4	byte	&01	(network-ordered)
 >>4	byte	=3	(major 1)
 >>4	byte	=2	(major 1)
 
-0	string	pst0	perl Storable(v0.7) data
+0	string	pst0	perl Storable (v0.7) data
 >4	byte	>0
 >>4	byte	&01	(network-ordered)
 >>4	byte	=5	(major 2)
diff --git a/usr.bin/file/magdir/pgp b/usr.bin/file/magdir/pgp
index e07fcdb4ced..f79b7848989 100644
--- a/usr.bin/file/magdir/pgp
+++ b/usr.bin/file/magdir/pgp
@@ -1,14 +1,27 @@
-#	$OpenBSD: pgp,v 1.4 2004/06/03 03:14:20 tedu Exp $
+#	$OpenBSD: pgp,v 1.5 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # pgp:  file(1) magic for Pretty Good Privacy
+# see http://lists.gnupg.org/pipermail/gnupg-devel/1999-September/016052.html
 #
 0       beshort         0x9900                  PGP key public ring
+!:mime	application/x-pgp-keyring
 0       beshort         0x9501                  PGP key security ring
+!:mime	application/x-pgp-keyring
 0       beshort         0x9500                  PGP key security ring
+!:mime	application/x-pgp-keyring
 0	beshort		0xa600			PGP encrypted data
-0       string          -----BEGIN\040PGP       PGP armored data
->15     string          PUBLIC\040KEY\040BLOCK- public key block
->15     string          MESSAGE-                message
->15     string          SIGNED\040MESSAGE-      signed message
->15     string          PGP\040SIGNATURE-       signature
+#!:mime	application/pgp-encrypted
+#0	string		-----BEGIN\040PGP	text/PGP armored data
+!:mime	text/PGP # encoding: armored data
+#>15	string	PUBLIC\040KEY\040BLOCK-	public key block
+#>15	string	MESSAGE-		message
+#>15	string	SIGNED\040MESSAGE-	signed message
+#>15	string	PGP\040SIGNATURE-	signature
+
+2	string	---BEGIN\ PGP\ PUBLIC\ KEY\ BLOCK-	PGP public key block
+!:mime	application/pgp-keys
+0	string	-----BEGIN\040PGP\40MESSAGE-		PGP message
+!:mime	application/pgp
+0	string	-----BEGIN\040PGP\40SIGNATURE-		PGP signature
+!:mime	application/pgp-signature
diff --git a/usr.bin/file/magdir/pkgadd b/usr.bin/file/magdir/pkgadd
index cd029422b70..602c5ace9f6 100644
--- a/usr.bin/file/magdir/pkgadd
+++ b/usr.bin/file/magdir/pkgadd
@@ -1,6 +1,7 @@
-#	$OpenBSD: pkgadd,v 1.2 2004/06/03 03:14:20 tedu Exp $
+#	$OpenBSD: pkgadd,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # pkgadd:  file(1) magic for SysV R4 PKG Datastreams
 #
 0       string          #\ PaCkAgE\ DaTaStReAm  pkg Datastream (SVR4)
+!:mime	application/x-svr4-package
diff --git a/usr.bin/file/magdir/printer b/usr.bin/file/magdir/printer
index fb1fe62b890..6afbe4849cd 100644
--- a/usr.bin/file/magdir/printer
+++ b/usr.bin/file/magdir/printer
@@ -1,4 +1,4 @@
-#	$OpenBSD: printer,v 1.6 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: printer,v 1.7 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # printer:  file(1) magic for printer-formatted files
@@ -6,21 +6,26 @@
 
 # PostScript, updated by Daniel Quinlan (quinlan@yggdrasil.com)
 0	string		%!		PostScript document text
+!:mime	application/postscript
 >2	string		PS-Adobe-	conforming
->>11	string		>\0		at level %.3s
->>>15	string		EPS		- type %s
->>>15	string		Query		- type %s
->>>15	string		ExitServer	- type %s
+>>11	string		>\0		DSC level %.3s
+>>>15	string		EPS		\b, type %s
+>>>15	string		Query		\b, type %s
+>>>15	string		ExitServer	\b, type %s
+>>>15   search/1000		%%LanguageLevel:\ 
+>>>>&0	string		>\0		\b, Level %s
 # Some PCs have the annoying habit of adding a ^D as a document separator
 0	string		\004%!		PostScript document text
+!:mime	application/postscript
 >3	string		PS-Adobe-	conforming
->>12	string		>\0		at level %.3s
->>>16	string		EPS		- type %s
->>>16	string		Query		- type %s
->>>16	string		ExitServer	- type %s
+>>12	string		>\0		DSC level %.3s
+>>>16	string		EPS		\b, type %s
+>>>16	string		Query		\b, type %s
+>>>16	string		ExitServer	\b, type %s
+>>>16   search/1000		%%LanguageLevel:\ 
+>>>>&0	string		>\0		\b, Level %s
 0	string		\033%-12345X%!PS	PostScript document
 
-
 # DOS EPS Binary File Header
 # From: Ed Sznyter <ews@Black.Market.NET>
 0       belong          0xC5D0D3C6      DOS EPS Binary File
diff --git a/usr.bin/file/magdir/revision b/usr.bin/file/magdir/revision
index 9ce0073cb41..0134cfc3d8a 100644
--- a/usr.bin/file/magdir/revision
+++ b/usr.bin/file/magdir/revision
@@ -1,4 +1,4 @@
-#	$OpenBSD: revision,v 1.2 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: revision,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # file(1) magic for revision control files
@@ -8,3 +8,13 @@
 # Conary changesets
 # From: Jonathan Smith <smithj@rpath.com>
 0	belong	0xea3f81bb	Conary changeset data
+
+# Type: Git bundles (git-bundle)
+# From: Josh Triplett <josh@freedesktop.org>
+0	string	#\ v2\ git\ bundle\n	Git bundle
+
+# Type:	Mercurial bundles
+# From:	Seo Sanghyeon <tinuviel@sparcs.kaist.ac.kr>
+0	string	HG10		Mercurial bundle,
+>4	string	UN		uncompressed
+>4	string	BZ		bzip2 compressed
diff --git a/usr.bin/file/magdir/riff b/usr.bin/file/magdir/riff
index 8eca0e3a3e2..2aa2b3f07b7 100644
--- a/usr.bin/file/magdir/riff
+++ b/usr.bin/file/magdir/riff
@@ -1,4 +1,4 @@
-#	$OpenBSD: riff,v 1.2 2008/05/14 20:45:35 chl Exp $
+#	$OpenBSD: riff,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # riff:  file(1) magic for RIFF format
@@ -34,6 +34,7 @@
 >8	string		RMP3		\b, MPEG Layer 3 audio
 # Microsoft WAVE format (*.wav)
 >8	string		WAVE		\b, WAVE audio
+!:mime	audio/x-wav
 >>20	leshort		1		\b, Microsoft PCM
 >>>34	leshort		>0		\b, %d bit
 >>20	leshort		2		\b, Microsoft ADPCM
@@ -51,8 +52,10 @@
 >>24	lelong		>0		%d Hz
 # Corel Draw Picture
 >8	string		CDRA		\b, Corel Draw Picture
+!:mime	image/x-coreldraw
 # AVI == Audio Video Interleave
 >8	string		AVI\040		\b, AVI
+!:mime	video/x-msvideo
 >>12    string          LIST
 >>>20   string          hdrlavih
 >>>>&36 lelong          x               \b, %lu x
@@ -131,6 +134,7 @@
 >>>>>>>(104.l+132)      string/c        divx    DivX 4
 >>>>>>>(104.l+132)      string/c        dx50    DivX 5
 >>>>>>>(104.l+132)      string/c        xvid    XviD
+>>>>>>>(104.l+132)	string/c	h264	H.264
 >>>>>>>(104.l+132)      string/c        wmv3    Windows Media Video 9
 >>>>>>>(104.l+132)      string/c        h264    X.264
 >>>>>>>(104.l+132)      lelong  0
diff --git a/usr.bin/file/magdir/rpm b/usr.bin/file/magdir/rpm
index eb87b23ea48..e8583cf457b 100644
--- a/usr.bin/file/magdir/rpm
+++ b/usr.bin/file/magdir/rpm
@@ -1,4 +1,4 @@
-#	$OpenBSD: rpm,v 1.5 2004/06/03 03:14:20 tedu Exp $
+#	$OpenBSD: rpm,v 1.6 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 #
@@ -6,6 +6,7 @@
 #
 0	beshort		0xedab
 >2	beshort		0xeedb		RPM
+!:mime	application/x-rpm
 >>4	byte		x		v%d
 >>6	beshort		0		bin
 >>6	beshort		1		src
@@ -15,7 +16,7 @@
 >>8	beshort		4		MIPS
 >>8	beshort		5		PowerPC
 >>8	beshort		6		68000
->>8	beshort		7	SGI
+>>8	beshort		7		SGI
 >>8	beshort		8		RS6000
 >>8	beshort		9		IA64
 >>8	beshort		10		Sparc64
diff --git a/usr.bin/file/magdir/rtf b/usr.bin/file/magdir/rtf
index d4c4014cea3..2c9ed84502f 100644
--- a/usr.bin/file/magdir/rtf
+++ b/usr.bin/file/magdir/rtf
@@ -1,4 +1,4 @@
-#	$OpenBSD: rtf,v 1.3 2004/06/03 03:14:20 tedu Exp $
+#	$OpenBSD: rtf,v 1.4 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # rtf:	file(1) magic for Rich Text Format (RTF)
@@ -6,8 +6,11 @@
 # Duncan P. Simpson, D.P.Simpson@dcs.warwick.ac.uk
 #
 0	string		{\\rtf		Rich Text Format data,
->5	byte		x		version %c,
->6	string		\\ansi		ANSI
->6	string		\\mac		Apple Macintosh
->6	string		\\pc		IBM PC, code page 437
->6	string		\\pca		IBM PS/2, code page 850
+!:mime	text/rtf
+>5	string		1		version 1,
+>>6	string		\\ansi		ANSI
+>>6	string		\\mac		Apple Macintosh
+>>6	string		\\pc		IBM PC, code page 437
+>>6	string		\\pca		IBM PS/2, code page 850
+>>6	default		x		unknown character set
+>5	default		x		unknown version
diff --git a/usr.bin/file/magdir/sc b/usr.bin/file/magdir/sc
index 23b0f710a7c..25c2c44ae3a 100644
--- a/usr.bin/file/magdir/sc
+++ b/usr.bin/file/magdir/sc
@@ -1,6 +1,7 @@
-#	$OpenBSD: sc,v 1.2 2004/06/03 03:14:20 tedu Exp $
+#	$OpenBSD: sc,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # sc:  file(1) magic for "sc" spreadsheet
 #
 38	string		Spreadsheet	sc spreadsheet file
+!:mime	application/x-sc
diff --git a/usr.bin/file/magdir/scientific b/usr.bin/file/magdir/scientific
index fdf9f54799c..9571b47ebae 100644
--- a/usr.bin/file/magdir/scientific
+++ b/usr.bin/file/magdir/scientific
@@ -1,4 +1,4 @@
-#	$OpenBSD: scientific,v 1.2 2008/05/14 20:45:35 chl Exp $
+#	$OpenBSD: scientific,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # scientific:  file(1) magic for scientific formats 
@@ -59,3 +59,15 @@
 >1104	ulong	>0		%d x
 >1108	ulong	>0		%d,
 >1120	ulong	>0		%d bits/pixel
+
+# Type: GEDCOM genealogical (family history) data
+# From: Giuseppe Bilotta
+0       search/1/c	0\ HEAD         GEDCOM genealogy text
+>&0     search		1\ GEDC
+>>&0    search		2\ VERS         version
+>>>&1   search/1	>\0		%s
+# From: Phil Endecott <phil05@chezphil.org>
+0	string	\000\060\000\040\000\110\000\105\000\101\000\104		GEDCOM data
+0	string	\060\000\040\000\110\000\105\000\101\000\104\000		GEDCOM data
+0	string	\376\377\000\060\000\040\000\110\000\105\000\101\000\104	GEDCOM data
+0	string	\377\376\060\000\040\000\110\000\105\000\101\000\104\000	GEDCOM data
diff --git a/usr.bin/file/magdir/securitycerts b/usr.bin/file/magdir/securitycerts
new file mode 100644
index 00000000000..aa55d2188fc
--- /dev/null
+++ b/usr.bin/file/magdir/securitycerts
@@ -0,0 +1,3 @@
+0	search/1		-----BEGIN\ CERTIFICATE------	RFC1421 Security Certificate text
+0	search/1		-----BEGIN\ NEW\ CERTIFICATE	RFC1421 Security Certificate Signing Request text
+0	belong	0xedfeedfe	Sun 'jks' Java Keystore File data
diff --git a/usr.bin/file/magdir/sgi b/usr.bin/file/magdir/sgi
index dddb9f08fbd..fa6a1edd32c 100644
--- a/usr.bin/file/magdir/sgi
+++ b/usr.bin/file/magdir/sgi
@@ -1,4 +1,4 @@
-#	$OpenBSD: sgi,v 1.4 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: sgi,v 1.5 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # sgi:  file(1) magic for Silicon Graphics applications
@@ -9,26 +9,37 @@
 0	string	PmNs				PCP compiled namespace (V.0)
 0	string	PmN				PCP compiled namespace
 >3	string	>\0				(V.%1.1s)
-3	lelong	0x84500526			PCP archive
+#3	lelong	0x84500526			PCP archive
+3	belong	0x84500526			PCP archive
 >7	byte	x				(V.%d)
->20	lelong	-2				temporal index
->20	lelong	-1				metadata
->20	lelong	0				log volume #0
->20	lelong	>0				log volume #%ld
+#>20	lelong	-2				temporal index
+#>20	lelong	-1				metadata
+#>20	lelong	0				log volume #0
+#>20	lelong	>0				log volume #%ld
+>20	belong	-2				temporal index
+>20	belong	-1				metadata
+>20	belong	0				log volume #0
+>20	belong	>0				log volume #%ld
 >24	string	>\0				host: %s
-0	string	PCPFolio			PCP 
+0	string	PCPFolio			PCP
 >9	string	Version:			Archive Folio
 >18	string	>\0				(V.%s)
 0	string	#pmchart			PCP pmchart view
 >9	string	Version
 >17	string	>\0				(V%-3.3s)
+0	string	#kmchart			PCP kmchart view
+>9	string	Version
+>17	string	>\0				(V.%s)
 0	string	pmview				PCP pmview config
 >7	string	Version
 >15	string	>\0				(V%-3.3s)
 0	string	#pmlogger			PCP pmlogger config
 >10	string	Version
 >18	string	>\0				(V%1.1s)
-0	string	PcPh				PCP Help 
+0	string	#pmdahotproc			PCP pmdahotproc config
+>13	string	Version
+>21	string	>\0				(V%-3.3s)
+0	string	PcPh				PCP Help
 >4	string	1				Index
 >4	string	2				Text
 >5	string	>\0				(V.%1.1s)
diff --git a/usr.bin/file/magdir/sgml b/usr.bin/file/magdir/sgml
index 0a891744bfe..60e9870fd6c 100644
--- a/usr.bin/file/magdir/sgml
+++ b/usr.bin/file/magdir/sgml
@@ -1,36 +1,50 @@
-#	$OpenBSD: sgml,v 1.4 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: sgml,v 1.5 2009/04/24 18:54:34 chl Exp $
+
+#------------------------------------------------------------------------------
+# Type:	SVG Vectorial Graphics
+# From:	Noel Torres <tecnico@ejerciciosresueltos.com>
+0	string		\<?xml\ version="
+>15	string		>\0
+>>23	search/400	\<svg			SVG Scalable Vector Graphics image
+!:mime	image/svg+xml
+>>23	search/400	\<gnc-v2		GnuCash file
+!:mime	application/x-gnucash
 
 #------------------------------------------------------------------------------
 # sgml:  file(1) magic for Standard Generalized Markup Language
 # HyperText Markup Language (HTML) is an SGML document type,
 # from Daniel Quinlan (quinlan@yggdrasil.com)
 # adapted to string extenstions by Anthon van der Neut <anthon@mnt.org)
-0   string/cB	\<!DOCTYPE\ html	HTML document text
-0   string/cb	\<head			HTML document text
-0   string/cb	\<title			HTML document text
-0   string/cb	\<html			HTML document text
+0	search/1/cB	\<!DOCTYPE\ html	HTML document text
+!:mime	text/html
+0	search/1/cb	\<head			HTML document text
+!:mime	text/html
+0	search/1/cb	\<title			HTML document text
+!:mime	text/html
+0	search/1/cb	\<html			HTML document text
+!:mime	text/html
 
 # Extensible markup language (XML), a subset of SGML
 # from Marc Prud'hommeaux (marc@apocalypse.org)
-0	string/cb	\<?xml			XML document text
+0	search/1/cb	\<?xml			XML document text
 0	string		\<?xml\ version\ "	XML
 0	string		\<?xml\ version="	XML
 0	string		\<?xml\ version='	XML
->15	string		>\0			%.3s document text
->>23	string		\<xsl:stylesheet	(XSL stylesheet)
->>24	string		\<xsl:stylesheet	(XSL stylesheet)
-0	string/b	\<?xml			XML document text
-0	string/cb	\<?xml			broken XML document text
+>15	search/1	>\0			%.3s document text
+>>23	search/1	\<xsl:stylesheet	(XSL stylesheet)
+>>24	search/1	\<xsl:stylesheet	(XSL stylesheet)
+0	search/1/b	\<?xml			XML document text
+0	search/1/b	\<?XML			broken XML document text
 
 
 # SGML, mostly from rph@sq
-0   string/cb	\<!doctype		exported SGML document text
-0   string/cb	\<!subdoc		exported SGML subdocument text
-0   string/cb	\<!--			exported SGML document text
+0	search/1/cb	\<!doctype		exported SGML document text
+0	search/1/cb	\<!subdoc		exported SGML subdocument text
+0	search/1/cb	\<!--			exported SGML document text
 
 # Web browser cookie files
 # (Mozilla, Galeon, Netscape 4, Konqueror..)
 # Ulf Harnhammar <ulfh@update.uu.se>
-0	string	#\ HTTP\ Cookie\ File	Web browser cookie text
-0	string	#\ Netscape\ HTTP\ Cookie\ File	Netscape cookie text
-0	string	#\ KDE\ Cookie\ File	Konqueror cookie text
+0	search/1	#\ HTTP\ Cookie\ File	Web browser cookie text
+0	search/1	#\ Netscape\ HTTP\ Cookie\ File	Netscape cookie text
+0	search/1	#\ KDE\ Cookie\ File	Konqueror cookie text
diff --git a/usr.bin/file/magdir/sharc b/usr.bin/file/magdir/sharc
index 6ac4b61a9d6..efb736d0682 100644
--- a/usr.bin/file/magdir/sharc
+++ b/usr.bin/file/magdir/sharc
@@ -1,4 +1,4 @@
-#	$OpenBSD: sharc,v 1.2 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: sharc,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------
 # file(1) magic for sharc files
@@ -7,14 +7,14 @@
 # FutureGroove Music (dsp@futuregroove.de)
 
 #------------------------------------------------------------------------
-0	string			Draw		RiscOS Drawfile
-0	string			PACK		RiscOS PackdDir archive
+#0	string			Draw		RiscOS Drawfile
+#0	string			PACK		RiscOS PackdDir archive
 
 #------------------------------------------------------------------------
 # SHARC DSP stuff (based on the FGM SHARC DSP SDK)
 
-0	string			=!		Assembler source
-0	string			Analog		ADi asm listing file
+#0	string			=!		Assembler source
+#0	string			Analog		ADi asm listing file
 0	string			.SYSTEM		SHARC architecture file
 0	string			.system		SHARC architecture file
 
diff --git a/usr.bin/file/magdir/sketch b/usr.bin/file/magdir/sketch
index 6addb7fa046..5d08adab003 100644
--- a/usr.bin/file/magdir/sketch
+++ b/usr.bin/file/magdir/sketch
@@ -1,6 +1,6 @@
-#	$OpenBSD: sketch,v 1.1 2004/06/03 03:14:20 tedu Exp $
+#	$OpenBSD: sketch,v 1.2 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # Sketch Drawings: http://sketch.sourceforge.net/ 
 # From: Edwin Mons <e@ik.nu>
-0	string	##Sketch	Sketch document text
+0	search/1	##Sketch	Sketch document text
diff --git a/usr.bin/file/magdir/softquad b/usr.bin/file/magdir/softquad
index 76e0b31a1e5..75dd69b786b 100644
--- a/usr.bin/file/magdir/softquad
+++ b/usr.bin/file/magdir/softquad
@@ -1,4 +1,4 @@
-#	$OpenBSD: softquad,v 1.4 2004/06/03 03:14:20 tedu Exp $
+#	$OpenBSD: softquad,v 1.5 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # softquad:  file(1) magic for SoftQuad Publishing Software
@@ -20,7 +20,7 @@
 0	short		0125252		SoftQuad DESC or font file binary
 >2	short		>0		- version %d
 # Bitmaps...
-0	string		SQ\ BITMAP1	SoftQuad Raster Format text
+0	search/1	SQ\ BITMAP1	SoftQuad Raster Format text
 #0	string		SQ\ BITMAP2	SoftQuad Raster Format data
 # sqtroff intermediate language (replacement for ditroff int. lang.)
 0	string		X\ 		SoftQuad troff Context intermediate
@@ -28,3 +28,10 @@
 >2	string		hp		for Hewlett-Packard LaserJet
 >2	string		impr		for IMAGEN imPRESS
 >2	string		ps		for PostScript
+
+# From: Michael Piefel <piefel@debian.org>
+# sqtroff intermediate language (replacement for ditroff int. lang.)
+0	string		X\ 495		SoftQuad troff Context intermediate for AT&T 495 laser printer
+0	string		X\ hp		SoftQuad troff Context intermediate for HP LaserJet
+0	string		X\ impr		SoftQuad troff Context intermediate for IMAGEN imPRESS
+0	string		X\ ps		SoftQuad troff Context intermediate for PostScript
diff --git a/usr.bin/file/magdir/spectrum b/usr.bin/file/magdir/spectrum
index 1b27ca24c43..7fcf867ffc4 100644
--- a/usr.bin/file/magdir/spectrum
+++ b/usr.bin/file/magdir/spectrum
@@ -1,4 +1,4 @@
-#	$OpenBSD: spectrum,v 1.1 2004/06/03 03:14:20 tedu Exp $
+#	$OpenBSD: spectrum,v 1.2 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # spectrum:  file(1) magic for Spectrum emulator files.
@@ -32,14 +32,19 @@
 # TZX tape images
 0      string          ZXTape!\x1a     Spectrum .TZX data
 >8     byte            x               version %d
->9     byte            x               .%d
+>9     byte            x               \b.%d
 
 # RZX input recording files
 0      string          RZX!            Spectrum .RZX data
 >4     byte            x               version %d
->5     byte            x               .%d
+>5     byte            x               \b.%d
 
-# And three sorts of disk image
+# Floppy disk images
 0      string          MV\ -\ CPCEMU\ Disk-Fil Amstrad/Spectrum .DSK data
 0      string          MV\ -\ CPC\ format\ Dis Amstrad/Spectrum DU54 .DSK data
 0      string          EXTENDED\ CPC\ DSK\ Fil Amstrad/Spectrum Extended .DSK data
+0      string          SINCLAIR        Spectrum .SCL Betadisk image
+
+# Hard disk images
+0      string          RS-IDE\x1a      Spectrum .HDF hard disk image
+>7     byte            x               \b, version 0x%02x
diff --git a/usr.bin/file/magdir/tex b/usr.bin/file/magdir/tex
index 1a0633d48ba..c77deddc6f3 100644
--- a/usr.bin/file/magdir/tex
+++ b/usr.bin/file/magdir/tex
@@ -1,47 +1,63 @@
-#	$OpenBSD: tex,v 1.4 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: tex,v 1.5 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # tex:  file(1) magic for TeX files
 #
+# XXX - needs byte-endian stuff (big-endian and little-endian DVI?)
+#
 # From <conklin@talisman.kaleida.com>
 
 # Although we may know the offset of certain text fields in TeX DVI
 # and font files, we can't use them reliably because they are not
 # zero terminated. [but we do anyway, christos]
 0	string		\367\002	TeX DVI file
+!:mime	application/x-dvi
 >16	string		>\0		(%s)
 0	string		\367\203	TeX generic font data
 0	string		\367\131	TeX packed font data
 >3	string		>\0		(%s)
 0	string		\367\312	TeX virtual font data
-0	string		This\ is\ TeX,	TeX transcript text
-0	string		This\ is\ METAFONT,	METAFONT transcript text
+0	search/1	This\ is\ TeX,	TeX transcript text
+0	search/1	This\ is\ METAFONT,	METAFONT transcript text
 
 # There is no way to detect TeX Font Metric (*.tfm) files without
 # breaking them apart and reading the data.  The following patterns
 # match most *.tfm files generated by METAFONT or afm2tfm.
 2	string		\000\021	TeX font metric data
+!:mime	application/x-tex-tfm
 >33	string		>\0		(%s)
 2	string		\000\022	TeX font metric data
+!:mime	application/x-tex-tfm
 >33	string		>\0		(%s)
 
 # Texinfo and GNU Info, from Daniel Quinlan (quinlan@yggdrasil.com)
-0	string		\\input\ texinfo	Texinfo source text
-0	string		This\ is\ Info\ file	GNU Info text
+0	search/1	\\input\ texinfo	Texinfo source text
+!:mime	text/x-texinfo
+0	search/1	This\ is\ Info\ file	GNU Info text
+!:mime	text/x-info
 
 # TeX documents, from Daniel Quinlan (quinlan@yggdrasil.com)
 0	search/400	\\input		TeX document text
+!:mime	text/x-tex
 0	search/400	\\section	LaTeX document text
+!:mime	text/x-tex
 0	search/400	\\setlength	LaTeX document text
+!:mime	text/x-tex
 0	search/400	\\documentstyle	LaTeX document text
+!:mime	text/x-tex
 0	search/400	\\chapter	LaTeX document text
+!:mime	text/x-tex
 0	search/400	\\documentclass	LaTeX 2e document text
+!:mime	text/x-tex
 0	search/400	\\relax		LaTeX auxiliary file
+!:mime	text/x-tex
 0	search/400	\\contentsline	LaTeX table of contents
+!:mime	text/x-tex
 0	search/400	%\ -*-latex-*-	LaTeX document text
+!:mime	text/x-tex
 
 # Tex document, from Hendrik Scholz <hendrik@scholz.net>
-0   string      \\ifx       TeX document text
+0   	search/1	\\ifx		TeX document text
 
 # Index and glossary files
 0	search/400	\\indexentry	LaTeX raw index file
@@ -56,25 +72,24 @@
 # file(1) magic for BibTex text files
 # From Hendrik Scholz <hendrik@scholz.net>
 
-0   string/c @article{        BibTeX text file
-0   string/c @book{           BibTeX text file
-0   string/c @inbook{         BibTeX text file
-0   string/c @incollection{   BibTeX text file
-0   string/c @inproceedings{  BibTeX text file
-0   string/c @manual{         BibTeX text file
-0   string/c @misc{           BibTeX text file
-0   string/c @preamble{       BibTeX text file
-0   string/c @phdthesis{      BibTeX text file
-0   string/c @techreport{     BibTeX text file
-0   string/c @unpublished{    BibTeX text file
-
-73  string %%%\ \ BibTeX-file{ BibTex text file (with full header)
+0	search/1/c	@article{	BibTeX text file
+0	search/1/c	@book{		BibTeX text file
+0	search/1/c	@inbook{	BibTeX text file
+0	search/1/c	@incollection{	BibTeX text file
+0	search/1/c	@inproceedings{	BibTeX text file
+0	search/1/c	@manual{	BibTeX text file
+0	search/1/c	@misc{		BibTeX text file
+0	search/1/c	@preamble{	BibTeX text file
+0	search/1/c	@phdthesis{	BibTeX text file
+0	search/1/c	@techreport{	BibTeX text file
+0	search/1/c	@unpublished{	BibTeX text file
 
-73  string %%%\ \ @BibTeX-style-file{   BibTeX style text file (with full header)
+73	search/1	%%%\ \ 		BibTeX-file{ BibTex text file (with full header)
 
-0   string %\ BibTeX\ standard\ bibliography\      BibTeX standard bibliography style text file
+73	search/1	%%%\ \ @BibTeX-style-file{   BibTeX style text file (with full header)
 
-0   string %\ BibTeX\ `     BibTeX custom bibliography style text file
+0	search/1	%\ BibTeX\ standard\ bibliography\ 	BibTeX standard bibliography style text file
 
-0   string  @c\ @mapfile{   TeX font aliases text file
+0	search/1	%\ BibTeX\ `	BibTeX custom bibliography style text file
 
+0	search/1	@c\ @mapfile{	TeX font aliases text file
diff --git a/usr.bin/file/magdir/troff b/usr.bin/file/magdir/troff
index 9f6a87c874b..9fd3f5a423c 100644
--- a/usr.bin/file/magdir/troff
+++ b/usr.bin/file/magdir/troff
@@ -1,4 +1,4 @@
-#	$OpenBSD: troff,v 1.3 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: troff,v 1.4 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # troff:  file(1) magic for *roff
@@ -6,24 +6,29 @@
 # updated by Daniel Quinlan (quinlan@yggdrasil.com)
 
 # troff input
-0	string		.\\"		troff or preprocessor input text
-0	string		'\\"		troff or preprocessor input text
-0	string		'.\\"		troff or preprocessor input text
-0	string		\\"		troff or preprocessor input text
-0	string		'''		troff or preprocessor input text
+0	search/1	.\\"		troff or preprocessor input text
+!:mime	text/troff
+0	search/1	'\\"		troff or preprocessor input text
+!:mime	text/troff
+0	search/1	'.\\"		troff or preprocessor input text
+!:mime	text/troff
+0	search/1	\\"		troff or preprocessor input text
+!:mime	text/troff
+0	search/1	'''		troff or preprocessor input text
+!:mime	text/troff
 
 # ditroff intermediate output text
-0	string		x\ T		ditroff output text
->4	string		cat		for the C/A/T phototypesetter
->4	string		ps		for PostScript
->4	string		dvi		for DVI
->4	string		ascii		for ASCII
->4	string		lj4		for LaserJet 4
->4	string		latin1		for ISO 8859-1 (Latin 1)
->4	string		X75		for xditview at 75dpi
->>7	string		-12		(12pt)
->4	string		X100		for xditview at 100dpi
->>8	string		-12		(12pt)
+0	search/1	x\ T		ditroff output text
+>4	search/1	cat		for the C/A/T phototypesetter
+>4	search/1	ps		for PostScript
+>4	search/1	dvi		for DVI
+>4	search/1	ascii		for ASCII
+>4	search/1	lj4		for LaserJet 4
+>4	search/1	latin1		for ISO 8859-1 (Latin 1)
+>4	search/1	X75		for xditview at 75dpi
+>>7	search/1	-12		(12pt)
+>4	search/1	X100		for xditview at 100dpi
+>>8	search/1	-12		(12pt)
 
 # output data formats
 0	string		\100\357	very old (C/A/T) troff output data
diff --git a/usr.bin/file/magdir/unicode b/usr.bin/file/magdir/unicode
index 184fa7fe5a8..a00ebcdce08 100644
--- a/usr.bin/file/magdir/unicode
+++ b/usr.bin/file/magdir/unicode
@@ -1,16 +1,16 @@
-#	$OpenBSD: unicode,v 1.2 2008/05/14 20:45:35 chl Exp $
+#	$OpenBSD: unicode,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #---------------------------------------------------------------------------
 # Unicode:  BOM prefixed text files - Adrian Havill <havill@turbolinux.co.jp>
+# GRR: These types should be recognised in file_ascmagic so these
+# encodings can be treated by text patterns.
+# Missing types are already dealt with internally.
 #
 0	string	+/v8			Unicode text, UTF-7
 0	string	+/v9			Unicode text, UTF-7
 0	string	+/v+			Unicode text, UTF-7
 0	string	+/v/			Unicode text, UTF-7
-0	string	\357\273\277		Unicode text, UTF-8
 0	string	\335\163\146\163	Unicode text, UTF-8-EBCDIC
 0	string	\376\377\000\000	Unicode text, UTF-32, big-endian
 0	string	\377\376\000\000	Unicode text, UTF-32, little-endian
-0	string	\376\377		Unicode text, UTF-16, big-endian
-0	string	\377\376		Unicode text, UTF-16, little-endian
 0	string	\016\376\377		Unicode text, SCSU (Standard Compression Scheme for Unicode)
diff --git a/usr.bin/file/magdir/uuencode b/usr.bin/file/magdir/uuencode
index aab736b5ce7..a1aa0f4d5cf 100644
--- a/usr.bin/file/magdir/uuencode
+++ b/usr.bin/file/magdir/uuencode
@@ -1,4 +1,4 @@
-#	$OpenBSD: uuencode,v 1.3 2004/06/03 03:14:20 tedu Exp $
+#	$OpenBSD: uuencode,v 1.4 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # uuencode:  file(1) magic for ASCII-encoded files
@@ -10,22 +10,22 @@
 # punctuation and survives BITNET gateways better.)  If regular expressions
 # were supported, this entry could possibly be split into two with
 # "begin\040\.\*\012M" or "begin\040\.\*\012h" (where \. and \* are REs).
-0	string		begin\040	uuencoded or xxencoded text
+0	search/1	begin\ 		uuencoded or xxencoded text
 
 # btoa(1) is an alternative to uuencode that requires less space.
-0	string		xbtoa\ Begin	btoa'd text
+0	search/1	xbtoa\ Begin	btoa'd text
 
 # ship(1) is another, much cooler alternative to uuencode.
 # Greg Roelofs, newt@uchicago.edu
-0	string		$\012ship	ship'd binary text
+0	search/1	$\012ship	ship'd binary text
 
 # bencode(8) is used to encode compressed news batches (Bnews/Cnews only?)
 # Greg Roelofs, newt@uchicago.edu
-0	string	Decode\ the\ following\ with\ bdeco	bencoded News text
+0	search/1	Decode\ the\ following\ with\ bdeco	bencoded News text
 
 # BinHex is the Macintosh ASCII-encoded file format (see also "apple")
 # Daniel Quinlan, quinlan@yggdrasil.com
-11	string	must\ be\ converted\ with\ BinHex	BinHex binary text
->41	string	x					\b, version %.3s
+11	search/1	must\ be\ converted\ with\ BinHex	BinHex binary text
+>41	search/1	x					\b, version %.3s
 
-# GRR:  is MIME BASE64 encoding handled somewhere?
+# GRR: handle BASE64
diff --git a/usr.bin/file/magdir/varied.script b/usr.bin/file/magdir/varied.script
index 3e2ada75e1c..c00f72b70e6 100644
--- a/usr.bin/file/magdir/varied.script
+++ b/usr.bin/file/magdir/varied.script
@@ -1,4 +1,4 @@
-#	$OpenBSD: varied.script,v 1.2 2008/05/14 20:45:35 chl Exp $
+#	$OpenBSD: varied.script,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # varied.script:  file(1) magic for various interpreter scripts
@@ -12,3 +12,9 @@
 0	string		#!\ 			script text executable
 >3	string		>\0			for %s
 
+# From: arno <arenevier@fdn.fr>
+# mozilla xpconnect typelib
+# see http://www.mozilla.org/scriptable/typelib_file.html
+0	string 		XPCOM\nTypeLib\r\n\032		XPConnect Typelib
+>0x10  byte        x       version %d
+>>0x11 byte        x      \b.%d
diff --git a/usr.bin/file/magdir/vorbis b/usr.bin/file/magdir/vorbis
index b7b6d47ceac..b2537643919 100644
--- a/usr.bin/file/magdir/vorbis
+++ b/usr.bin/file/magdir/vorbis
@@ -1,4 +1,4 @@
-#	$OpenBSD: vorbis,v 1.2 2008/05/08 01:40:57 chl Exp $
+#	$OpenBSD: vorbis,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # vorbis:  file(1) magic for Ogg/Vorbis files
@@ -12,16 +12,38 @@
 # probable things advanced users would want to uncomment are probably
 # the number of comments and the encoder version.
 #
+# FIXME: The first match has been made a search, so that it can skip
+# over prepended ID3 tags. This will work for MIME type detection, but
+# won't work for detecting other properties of the file (they all need
+# to be made relative to the search). In any case, if the file has ID3
+# tags, the ID3 information will be printed, not the Ogg information,
+# so until that's fixed, this doesn't matter.
+#
 # --- Ogg Framing ---
-0		string		OggS		Ogg data
+0		search/1000	OggS		Ogg data
+!:mime		application/ogg
 >4		byte		!0		UNKNOWN REVISION %u
 ##>4		byte		0		revision 0
 >4		byte		0
 ##>>14		lelong		x		(Serial %lX)
 # non-Vorbis content: FLAC (Free Lossless Audio Codec, http://flac.sourceforge.net)
->>28		string		fLaC		\b, FLAC audio
+>>28		string		\x7fFLAC	\b, FLAC audio
 # non-Vorbis content: Theora
 >>28		string		\x80theora	\b, Theora video
+# non-Vorbis content: Kate
+>>28		string		\x80kate\0\0\0\0	\b, Kate
+>>>37		ubyte		x		v%u
+>>>38		ubyte		x		\b.%u,
+>>>40		byte		0		utf8 encoding,
+>>>40		byte		!0		unknown character encoding,
+>>>60		string		>\0		language %s,
+>>>60		string		\0		no language set,
+>>>76		string		>\0		category %s
+>>>76		string		\0		no category set
+# non-Vorbis content: Skeleton
+>>28		string		fishead\0	\b, Skeleton
+>>>36		short		x		v%u
+>>>40		short		x		\b.%u
 # non-Vorbis content: Speex
 >>28		string		Speex\ \ \ 	\b, Speex audio
 # non-Vorbis content: OGM
diff --git a/usr.bin/file/magdir/wordprocessors b/usr.bin/file/magdir/wordprocessors
index e25af223a5c..f0afe538da7 100644
--- a/usr.bin/file/magdir/wordprocessors
+++ b/usr.bin/file/magdir/wordprocessors
@@ -1,4 +1,4 @@
-#	$OpenBSD: wordprocessors,v 1.2 2008/05/14 20:45:35 chl Exp $
+#	$OpenBSD: wordprocessors,v 1.3 2009/04/24 18:54:34 chl Exp $
 
 #------------------------------------------------------------------------------
 # wordprocessors:  file(1) magic fo word processors.
@@ -13,7 +13,7 @@
 >26	byte	0x46	\b, A4
 
 #WordPerfect type files Version 1.6 - PLEASE DO NOT REMOVE THIS LINE
-0	string	\377WPC\020\000\000\000\022\012\001\001\000\000\000\000	(WP) loadable text
+0	string	\377WPC\020\000\000\000\022\012\001\001\000\000\000\000	(WP) loadable file
 >15	byte	0	Optimized for Intel
 >15	byte	1	Optimized for Non-Intel
 1	string	WPC	(Corel/WP)
@@ -108,6 +108,7 @@
 0	string	HWP\ Document\ File	Hangul (Korean) Word Processor File 3.0
 # From: Won-Kyu Park <wkpark@kldp.org>
 512	string		R\0o\0o\0t\0	Hangul (Korean) Word Processor File 2000
+!:mime	application/x-hwp
 
 # CosmicBook, from Benoît Rouits
 0       string  CSBK    Ted Neslson's CosmicBook hypertext file
@@ -123,15 +124,29 @@
 2	string	IIXPR3			Intel Quark Express Document (English)
 2	string	IIXPRa			Intel Quark Express Document (Korean)
 2	string	MMXPR3			Motorola Quark Express Document (English)
+!:mime	application/x-quark-xpress-3
 2	string	MMXPRa			Motorola Quark Express Document (Korean)
 
 # adobe indesign (document, whatever...) from querkan
 0	belong	0x0606edf5		Adobe InDesign
 >16	string	DOCUMENT		Document
 
-# From: Michael Piefel <piefel@debian.org>
-# sqtroff intermediate language (replacement for ditroff int. lang.)
-0	string		X\ 495		SoftQuad troff Context intermediate for AT&T 495 laser printer
-0	string		X\ hp		SoftQuad troff Context intermediate for HP LaserJet
-0	string		X\ impr		SoftQuad troff Context intermediate for IMAGEN imPRESS
-0	string		X\ ps		SoftQuad troff Context intermediate for PostScript
+#------------------------------------------------------------------------------
+# ichitaro456: file(1) magic for Just System Word Processor Ichitaro
+#
+# Contributor kenzo-:
+# Reversed-engineered JS Ichitaro magic numbers
+#
+
+0	string		DOC
+>43	byte		0x14	Just System Word Processor Ichitaro v4
+!:mime	application/x-ichitaro4
+>144	string	JDASH		application/x-ichitaro4
+
+0	string		DOC
+>43	byte		0x15	Just System Word Processor Ichitaro v5
+!:mime	application/x-ichitaro5
+
+0	string		DOC
+>43	byte		0x16	Just System Word Processor Ichitaro v6
+!:mime	application/x-ichitaro6
diff --git a/usr.bin/file/magdir/xilinx b/usr.bin/file/magdir/xilinx
new file mode 100644
index 00000000000..8f411ce21dd
--- /dev/null
+++ b/usr.bin/file/magdir/xilinx
@@ -0,0 +1,34 @@
+
+#------------------------------------------------------------------------------
+# This is Aaron's attempt at a MAGIC file for Xilinx .bit files.
+# Xilinx-Magic@RevRagnarok.com
+# Got the info from FPGA-FAQ 0026
+#
+# First there is the sync header and its length
+0	beshort 0x0009
+>2 	belong	=0x0ff00ff0
+>>&0	belong  =0x0ff00ff0
+>>>&0	beshort =0x0000	
+>>>>&0	pstring	a	Xilinx BIT data
+# Next is a Pascal-style string with the NCD name. We want to capture that.
+>>>>0x0F	pstring	x	- from %s
+# It is followed by a NUL
+>>>>>&1	byte	0x00
+# And then 'b'
+>>>>>&2	string b
+# With the part number:
+#>>>>>&5 string	4v	(Virtex4)
+#>>>>>&5 string  2v	(Virtex II
+#>>>>>>&0	string	!p	\b)
+#>>>>>>&0	string	p	Pro)
+>>>>>&4	pstring x	- for %s
+# And then NUL / 'c' / Build Data / NUL / 'd' / Date / NUL / 'e' / Data Length
+>>>>>>&1	byte	0x00
+>>>>>>&2	string	c
+>>>>>>&4	pstring	x	- built %s
+>>>>>>>&1	byte	0x00
+>>>>>>>&2	string	d
+>>>>>>>&4	pstring	x	\b(%s)
+>>>>>>>>&1	byte	0x00
+>>>>>>>>&2	string	e
+>>>>>>>>&4	belong	x	- data length 0x%lx
diff --git a/usr.bin/file/magic.c b/usr.bin/file/magic.c
index a09b987a182..e28186a9639 100644
--- a/usr.bin/file/magic.c
+++ b/usr.bin/file/magic.c
@@ -1,4 +1,4 @@
-/* $OpenBSD: magic.c,v 1.5 2008/05/08 01:40:56 chl Exp $ */
+/* $OpenBSD: magic.c,v 1.6 2009/04/24 18:54:34 chl Exp $ */
 /*
  * Copyright (c) Christos Zoulas 2003.
  * All Rights Reserved.
@@ -64,9 +64,18 @@
 #include "patchlevel.h"
 
 #ifndef	lint
-FILE_RCSID("@(#)$Id: magic.c,v 1.5 2008/05/08 01:40:56 chl Exp $")
+FILE_RCSID("@(#)$Id: magic.c,v 1.6 2009/04/24 18:54:34 chl Exp $")
 #endif	/* lint */
 
+#ifndef PIPE_BUF 
+/* Get the PIPE_BUF from pathconf */
+#ifdef _PC_PIPE_BUF
+#define PIPE_BUF pathconf(".", _PC_PIPE_BUF)
+#else
+#define PIPE_BUF 512
+#endif
+#endif
+
 #ifdef __EMX__
 private char *apptypeName = NULL;
 protected int file_os2_apptype(struct magic_set *ms, const char *fn,
@@ -77,6 +86,9 @@ private void free_mlist(struct mlist *);
 private void close_and_restore(const struct magic_set *, const char *, int,
     const struct stat *);
 private int info_from_stat(struct magic_set *, mode_t);
+#ifndef COMPILE_ONLY
+private const char *file_or_fd(struct magic_set *, const char *, int);
+#endif
 
 #ifndef	STDIN_FILENO
 #define	STDIN_FILENO	0
@@ -92,20 +104,14 @@ magic_open(int flags)
 
 	if (magic_setflags(ms, flags) == -1) {
 		errno = EINVAL;
-		goto free1;
+		goto free;
 	}
 
-	ms->o.ptr = ms->o.buf = malloc(ms->o.left = ms->o.size = 1024);
-	if (ms->o.buf == NULL)
-		goto free1;
-
-	ms->o.pbuf = malloc(ms->o.psize = 1024);
-	if (ms->o.pbuf == NULL)
-		goto free2;
+	ms->o.buf = ms->o.pbuf = NULL;
 
 	ms->c.li = malloc((ms->c.len = 10) * sizeof(*ms->c.li));
 	if (ms->c.li == NULL)
-		goto free3;
+		goto free;
 	
 	ms->haderr = 0;
 	ms->error = -1;
@@ -113,11 +119,7 @@ magic_open(int flags)
 	ms->file = "unknown";
 	ms->line = 0;
 	return ms;
-free3:
-	free(ms->o.pbuf);
-free2:
-	free(ms->o.buf);
-free1:
+free:
 	free(ms);
 	return NULL;
 }
@@ -216,6 +218,7 @@ close_and_restore(const struct magic_set *ms, const char *name, int fd,
 		 */
 #ifdef HAVE_UTIMES
 		struct timeval  utsbuf[2];
+		(void)memset(utsbuf, 0, sizeof(utsbuf));
 		utsbuf[0].tv_sec = sb->st_atime;
 		utsbuf[1].tv_sec = sb->st_mtime;
 
@@ -223,6 +226,7 @@ close_and_restore(const struct magic_set *ms, const char *name, int fd,
 #elif defined(HAVE_UTIME_H) || defined(HAVE_SYS_UTIME_H)
 		struct utimbuf  utbuf;
 
+		(void)memset(utbuf, 0, sizeof(utbuf));
 		utbuf.actime = sb->st_atime;
 		utbuf.modtime = sb->st_mtime;
 		(void) utime(name, &utbuf); /* don't care if loses */
@@ -231,13 +235,28 @@ close_and_restore(const struct magic_set *ms, const char *name, int fd,
 }
 
 #ifndef COMPILE_ONLY
+
+/*
+ * find type of descriptor
+ */
+public const char *
+magic_descriptor(struct magic_set *ms, int fd)
+{
+	return file_or_fd(ms, NULL, fd);
+}
+
 /*
  * find type of named file
  */
 public const char *
 magic_file(struct magic_set *ms, const char *inname)
 {
-	int	fd = 0;
+	return file_or_fd(ms, inname, STDIN_FILENO);
+}
+
+private const char *
+file_or_fd(struct magic_set *ms, const char *inname, int fd)
+{
 	int	rv = -1;
 	unsigned char *buf;
 	struct stat	sb;
@@ -266,7 +285,6 @@ magic_file(struct magic_set *ms, const char *inname)
 	}
 
 	if (inname == NULL) {
-		fd = STDIN_FILENO;
 		if (fstat(fd, &sb) == 0 && S_ISFIFO(sb.st_mode))
 			ispipe = 1;
 	} else {
@@ -280,16 +298,18 @@ magic_file(struct magic_set *ms, const char *inname)
 		errno = 0;
 		if ((fd = open(inname, flags)) < 0) {
 #ifdef __CYGWIN__
-		    char *tmp = alloca(strlen(inname) + 5);
-		    (void)strcat(strcpy(tmp, inname), ".exe");
-		    if ((fd = open(tmp, flags)) < 0) {
+			/* FIXME: Do this with EXEEXT from autotools */
+			char *tmp = alloca(strlen(inname) + 5);
+			(void)strcat(strcpy(tmp, inname), ".exe");
+			if ((fd = open(tmp, flags)) < 0) {
 #endif
-			if (info_from_stat(ms, sb.st_mode) == -1)
-			    goto done;
-			rv = 0;
-			goto done;
+				fprintf(stderr, "couldn't open file\n");
+				if (info_from_stat(ms, sb.st_mode) == -1)
+					goto done;
+				rv = 0;
+				goto done;
 #ifdef __CYGWIN__
-		    }
+			}
 #endif
 		}
 #ifdef O_NONBLOCK
@@ -327,18 +347,9 @@ magic_file(struct magic_set *ms, const char *inname)
 		}
 	}
 
-	if (nbytes == 0) {
-		if (file_printf(ms, (ms->flags & MAGIC_MIME) ?
-		    "application/x-empty" : "empty") == -1)
-			goto done;
-	} else if (nbytes == 1) {
-		if (file_printf(ms, "very short file (no magic)") == -1)
-			goto done;
-	} else {
-		(void)memset(buf + nbytes, 0, SLOP); /* NUL terminate */
-		if (file_buffer(ms, fd, inname, buf, (size_t)nbytes) == -1)
-			goto done;
-	}
+	(void)memset(buf + nbytes, 0, SLOP); /* NUL terminate */
+	if (file_buffer(ms, fd, inname, buf, (size_t)nbytes) == -1)
+		goto done;
 	rv = 0;
 done:
 	free(buf);
diff --git a/usr.bin/file/magic.h b/usr.bin/file/magic.h
index 1731bffbcb6..03d0b1b0d6c 100644
--- a/usr.bin/file/magic.h
+++ b/usr.bin/file/magic.h
@@ -1,4 +1,4 @@
-/* $OpenBSD: magic.h,v 1.3 2008/05/08 01:40:56 chl Exp $ */
+/* $OpenBSD: magic.h,v 1.4 2009/04/24 18:54:34 chl Exp $ */
 /*
  * Copyright (c) Christos Zoulas 2003.
  * All Rights Reserved.
@@ -35,21 +35,26 @@
 #define	MAGIC_SYMLINK		0x000002 /* Follow symlinks */
 #define	MAGIC_COMPRESS		0x000004 /* Check inside compressed files */
 #define	MAGIC_DEVICES		0x000008 /* Look at the contents of devices */
-#define	MAGIC_MIME		0x000010 /* Return a mime string */
+#define	MAGIC_MIME_TYPE		0x000010 /* Return only the MIME type */
 #define	MAGIC_CONTINUE		0x000020 /* Return all matches */
 #define	MAGIC_CHECK		0x000040 /* Print warnings to stderr */
 #define	MAGIC_PRESERVE_ATIME	0x000080 /* Restore access time on exit */
-#define	MAGIC_RAW		0x000100 /* Don't translate unprintable chars */
+#define	MAGIC_RAW		0x000100 /* Don't translate unprint chars */
 #define	MAGIC_ERROR		0x000200 /* Handle ENOENT etc as real errors */
-#define MAGIC_NO_CHECK_COMPRESS	0x001000 /* Don't check for compressed files */
-#define MAGIC_NO_CHECK_TAR	0x002000 /* Don't check for tar files */
-#define MAGIC_NO_CHECK_SOFT	0x004000 /* Don't check magic entries */
-#define MAGIC_NO_CHECK_APPTYPE	0x008000 /* Don't check application type */
-#define MAGIC_NO_CHECK_ELF	0x010000 /* Don't check for elf details */
-#define MAGIC_NO_CHECK_ASCII	0x020000 /* Don't check for ascii files */
-#define MAGIC_NO_CHECK_TROFF	0x040000 /* Don't check ascii/troff */
-#define MAGIC_NO_CHECK_FORTRAN	0x080000 /* Don't check ascii/fortran */
-#define MAGIC_NO_CHECK_TOKENS	0x100000 /* Don't check ascii/tokens */
+#define	MAGIC_MIME_ENCODING	0x000400 /* Return only the MIME encoding */
+#define MAGIC_MIME		(MAGIC_MIME_TYPE|MAGIC_MIME_ENCODING)
+#define	MAGIC_NO_CHECK_COMPRESS	0x001000 /* Don't check for compressed files */
+#define	MAGIC_NO_CHECK_TAR	0x002000 /* Don't check for tar files */
+#define	MAGIC_NO_CHECK_SOFT	0x004000 /* Don't check magic entries */
+#define	MAGIC_NO_CHECK_APPTYPE	0x008000 /* Don't check application type */
+#define	MAGIC_NO_CHECK_ELF	0x010000 /* Don't check for elf details */
+#define	MAGIC_NO_CHECK_ASCII	0x020000 /* Don't check for ascii files */
+#define	MAGIC_NO_CHECK_TOKENS	0x100000 /* Don't check ascii/tokens */
+
+/* Defined for backwards compatibility; do nothing */
+#define	MAGIC_NO_CHECK_FORTRAN	0x000000 /* Don't check ascii/fortran */
+#define	MAGIC_NO_CHECK_TROFF	0x000000 /* Don't check ascii/troff */
+
 
 #ifdef __cplusplus
 extern "C" {
@@ -60,6 +65,7 @@ magic_t magic_open(int);
 void magic_close(magic_t);
 
 const char *magic_file(magic_t, const char *);
+const char *magic_descriptor(magic_t, int);
 const char *magic_buffer(magic_t, const void *, size_t);
 
 const char *magic_error(magic_t);
diff --git a/usr.bin/file/names.h b/usr.bin/file/names.h
index afc018d0c41..ef5fdf86d98 100644
--- a/usr.bin/file/names.h
+++ b/usr.bin/file/names.h
@@ -1,4 +1,4 @@
-/*	$OpenBSD: names.h,v 1.7 2008/05/08 01:40:56 chl Exp $ */
+/*	$OpenBSD: names.h,v 1.8 2009/04/24 18:54:34 chl Exp $ */
 /*
  * Copyright (c) Ian F. Darwin 1986-1995.
  * Software written by Ian F. Darwin and others;
@@ -33,7 +33,7 @@
  * appear at fixed offsets into the file. Don't make HOWMANY
  * too high unless you have a very fast CPU.
  *
- * $Id: names.h,v 1.7 2008/05/08 01:40:56 chl Exp $
+ * $Id: names.h,v 1.8 2009/04/24 18:54:34 chl Exp $
  */
 
 /*
@@ -44,27 +44,25 @@
 /* these types are used to index the table 'types': keep em in sync! */
 #define	L_C	0		/* first and foremost on UNIX */
 #define	L_CC	1		/* Bjarne's postincrement */
-#define	L_FORT	2		/* the oldest one */
-#define	L_MAKE	3		/* Makefiles */
-#define	L_PLI	4		/* PL/1 */
-#define	L_MACH	5		/* some kinda assembler */
-#define	L_ENG	6		/* English */
-#define	L_PAS	7		/* Pascal */
-#define	L_MAIL	8		/* Electronic mail */
-#define	L_NEWS	9		/* Usenet Netnews */
-#define	L_JAVA	10		/* Java code */
-#define	L_HTML	11		/* HTML */
-#define	L_BCPL	12		/* BCPL */
-#define	L_M4	13		/* M4 */
-#define	L_PO	14		/* PO */
+#define	L_MAKE	2		/* Makefiles */
+#define	L_PLI	3		/* PL/1 */
+#define	L_MACH	4		/* some kinda assembler */
+#define	L_ENG	5		/* English */
+#define	L_PAS	6		/* Pascal */
+#define	L_MAIL	7		/* Electronic mail */
+#define	L_NEWS	8		/* Usenet Netnews */
+#define	L_JAVA	9		/* Java code */
+#define	L_HTML	10		/* HTML */
+#define	L_BCPL	11		/* BCPL */
+#define	L_M4	12		/* M4 */
+#define	L_PO	13		/* PO */
 
 static const struct {
-	const char *human;
-	const char *mime;
+	char human[48];
+	char mime[16];
 } types[] = {
 	{ "C program",					"text/x-c", },
 	{ "C++ program",				"text/x-c++" },
-	{ "FORTRAN program",				"text/x-fortran" },
 	{ "make commands",				"text/x-makefile" },
 	{ "PL/1 program",				"text/x-pl1" },
 	{ "assembler program",				"text/x-asm" },
@@ -77,8 +75,7 @@ static const struct {
 	{ "BCPL program",				"text/x-bcpl" },
 	{ "M4 macro language pre-processor",		"text/x-m4" },
 	{ "PO (gettext message catalogue)",             "text/x-po" },
-	{ "cannot happen error on names.h/types",	"error/x-error" },
-	{ 0, 0}
+	{ "cannot happen error on names.h/types",	"error/x-error" }
 };
 
 /*
@@ -117,8 +114,8 @@ static const struct {
  * as Java, as it comes after "the" and "The".  Perhaps we need a fancier
  * heuristic to identify Java?
  */
-static struct names {
-	const char *name;
+static const struct names {
+	char name[14];
 	short type;
 } names[] = {
 	/* These must be sorted by eye for optimal hit rate */
@@ -148,18 +145,6 @@ static struct names {
 	{"LDFLAGS",	L_MAKE},
 	{"all:",	L_MAKE},
 	{".PRECIOUS",	L_MAKE},
-/* Too many files of text have these words in them.  Find another way
- * to recognize Fortrash.
- */
-#ifdef	NOTDEF
-	{"subroutine",	L_FORT},
-	{"function",	L_FORT},
-	{"block",	L_FORT},
-	{"common",	L_FORT},
-	{"dimension",	L_FORT},
-	{"integer",	L_FORT},
-	{"data",	L_FORT},
-#endif	/*NOTDEF*/
 	{".ascii",	L_MACH},
 	{".asciiz",	L_MACH},
 	{".byte",	L_MACH},
@@ -184,6 +169,6 @@ static struct names {
 	{"<BODY",	L_HTML},
 	{"<html",	L_HTML},
 	{"<HTML",	L_HTML},
-	{NULL,		0}
+	{"<!--",	L_HTML},
 };
-#define NNAMES ((sizeof(names)/sizeof(struct names)) - 1)
+#define NNAMES (sizeof(names)/sizeof(struct names))
diff --git a/usr.bin/file/patchlevel.h b/usr.bin/file/patchlevel.h
index 103cde05235..4e540831195 100644
--- a/usr.bin/file/patchlevel.h
+++ b/usr.bin/file/patchlevel.h
@@ -1,19 +1,33 @@
-/*	$OpenBSD: patchlevel.h,v 1.8 2008/05/08 01:40:56 chl Exp $	*/
+/*	$OpenBSD: patchlevel.h,v 1.9 2009/04/24 18:54:34 chl Exp $	*/
 
 #define	FILE_VERSION_MAJOR	4
-#define	patchlevel		21
+#define	patchlevel		24
 
 /*
  * Patchlevel file for Ian Darwin's MAGIC command.
- * $File: patchlevel.h,v 1.65 2007/05/24 17:22:27 christos Exp $
+ * $File: patchlevel.h,v 1.68 2008/03/22 21:39:43 christos Exp $
  *
  * $Log: patchlevel.h,v $
- * Revision 1.8  2008/05/08 01:40:56  chl
- * Update file to 4.21.
+ * Revision 1.9  2009/04/24 18:54:34  chl
+ * file update to 4.24
  *
- * tested by ian@ sthen@
+ * The '-i' switch is now enabled so file(1) can output mime type strings.
  *
- * OK ian@, "move ahead with it" deraadt@
+ * ok ian@
+ * builk ports build test on amd64 by jasper@
+ *
+ * ok ray@ gilles@ on a almost identical diff
+ * builk ports build test on sparc64 on this almost identical diff by ajacoutot@
+ * also tested by landry@
+ *
+ * Revision 1.68  2008/03/22 21:39:43  christos
+ * file 4.24
+ *
+ * Revision 1.67  2007/12/28 20:08:40  christos
+ * welcome to 4.23.
+ *
+ * Revision 1.66  2007/12/27 16:38:24  christos
+ * welcome to 4.22
  *
  * Revision 1.65  2007/05/24 17:22:27  christos
  * Welcome to 4.21
diff --git a/usr.bin/file/print.c b/usr.bin/file/print.c
index 438aaa00712..32dee304645 100644
--- a/usr.bin/file/print.c
+++ b/usr.bin/file/print.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: print.c,v 1.14 2008/05/08 01:40:56 chl Exp $ */
+/*	$OpenBSD: print.c,v 1.15 2009/04/24 18:54:34 chl Exp $ */
 /*
  * Copyright (c) Ian F. Darwin 1986-1995.
  * Software written by Ian F. Darwin and others;
@@ -42,7 +42,7 @@
 #include <time.h>
 
 #ifndef lint
-FILE_RCSID("@(#)$Id: print.c,v 1.14 2008/05/08 01:40:56 chl Exp $")
+FILE_RCSID("@(#)$Id: print.c,v 1.15 2009/04/24 18:54:34 chl Exp $")
 #endif  /* lint */
 
 #define SZOF(a)	(sizeof(a) / sizeof(a[0]))
@@ -90,8 +90,8 @@ file_mdump(struct magic *m)
 			if (m->str_flags & REGEX_OFFSET_START) 
 				(void) fputc(CHAR_REGEX_OFFSET_START, stderr);
 		}
-		if (m->str_count)
-			(void) fprintf(stderr, "/%u", m->str_count);
+		if (m->str_range)
+			(void) fprintf(stderr, "/%u", m->str_range);
 	}
 	else {
 		if ((m->mask_op & FILE_OPS_MASK) < SZOF(optyp))
@@ -158,6 +158,16 @@ file_mdump(struct magic *m)
 			(void)fprintf(stderr, "%s,",
 			    file_fmttime((uint32_t)m->value.q, 0));
 			break;
+		case FILE_FLOAT:
+		case FILE_BEFLOAT:
+		case FILE_LEFLOAT:
+			(void) fprintf(stderr, "%G", m->value.f);
+			break;
+		case FILE_DOUBLE:
+		case FILE_BEDOUBLE:
+		case FILE_LEDOUBLE:
+			(void) fprintf(stderr, "%G", m->value.d);
+			break;
 		case FILE_DEFAULT:
 			/* XXX - do anything here? */
 			break;
@@ -175,13 +185,15 @@ protected void
 file_magwarn(struct magic_set *ms, const char *f, ...)
 {
 	va_list va;
-	va_start(va, f);
 
 	/* cuz we use stdout for most, stderr here */
 	(void) fflush(stdout); 
 
-	(void) fprintf(stderr, "%s, %lu: Warning ", ms->file,
-	    (unsigned long)ms->line);
+	if (ms->file)
+		(void) fprintf(stderr, "%s, %lu: ", ms->file,
+		    (unsigned long)ms->line);
+	(void) fprintf(stderr, "Warning: ");
+	va_start(va, f);
 	(void) vfprintf(stderr, f, va);
 	va_end(va);
 	(void) fputc('\n', stderr);
diff --git a/usr.bin/file/readelf.c b/usr.bin/file/readelf.c
index da2d53ad4b5..e41f217ec51 100644
--- a/usr.bin/file/readelf.c
+++ b/usr.bin/file/readelf.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: readelf.c,v 1.9 2008/05/08 01:40:56 chl Exp $ */
+/*	$OpenBSD: readelf.c,v 1.10 2009/04/24 18:54:34 chl Exp $ */
 /*
  * Copyright (c) Christos Zoulas 2003.
  * All Rights Reserved.
@@ -36,9 +36,10 @@
 #endif
 
 #include "readelf.h"
+#include "magic.h"
 
 #ifndef lint
-FILE_RCSID("@(#)$Id: readelf.c,v 1.9 2008/05/08 01:40:56 chl Exp $")
+FILE_RCSID("@(#)$Id: readelf.c,v 1.10 2009/04/24 18:54:34 chl Exp $")
 #endif
 
 #ifdef	ELFCORE
@@ -124,84 +125,106 @@ getu64(int swap, uint64_t value)
 		return value;
 }
 
+#define elf_getu16(swap, value) getu16(swap, value)
+#define elf_getu32(swap, value) getu32(swap, value)
 #ifdef USE_ARRAY_FOR_64BIT_TYPES
 # define elf_getu64(swap, array) \
-	((swap ? ((uint64_t)getu32(swap, array[0])) << 32 : getu32(swap, array[0])) + \
-	 (swap ? getu32(swap, array[1]) : ((uint64_t)getu32(swap, array[1]) << 32)))
+	((swap ? ((uint64_t)elf_getu32(swap, array[0])) << 32 : elf_getu32(swap, array[0])) + \
+	 (swap ? elf_getu32(swap, array[1]) : ((uint64_t)elf_getu32(swap, array[1]) << 32)))
 #else
 # define elf_getu64(swap, value) getu64(swap, value)
 #endif
 
-#define xsh_addr	(class == ELFCLASS32		\
-			 ? (void *) &sh32		\
+#define xsh_addr	(class == ELFCLASS32			\
+			 ? (void *) &sh32			\
 			 : (void *) &sh64)
-#define xsh_sizeof	(class == ELFCLASS32		\
-			 ? sizeof sh32			\
+#define xsh_sizeof	(class == ELFCLASS32			\
+			 ? sizeof sh32				\
 			 : sizeof sh64)
-#define xsh_size	(class == ELFCLASS32		\
-			 ? getu32(swap, sh32.sh_size)	\
-			 : getu64(swap, sh64.sh_size))
-#define xsh_offset	(class == ELFCLASS32		\
-			 ? getu32(swap, sh32.sh_offset)	\
-			 : getu64(swap, sh64.sh_offset))
-#define xsh_type	(class == ELFCLASS32		\
-			 ? getu32(swap, sh32.sh_type)	\
-			 : getu32(swap, sh64.sh_type))
-#define xph_addr	(class == ELFCLASS32		\
-			 ? (void *) &ph32		\
+#define xsh_size	(class == ELFCLASS32			\
+			 ? elf_getu32(swap, sh32.sh_size)	\
+			 : elf_getu64(swap, sh64.sh_size))
+#define xsh_offset	(class == ELFCLASS32			\
+			 ? elf_getu32(swap, sh32.sh_offset)	\
+			 : elf_getu64(swap, sh64.sh_offset))
+#define xsh_type	(class == ELFCLASS32			\
+			 ? elf_getu32(swap, sh32.sh_type)	\
+			 : elf_getu32(swap, sh64.sh_type))
+#define xph_addr	(class == ELFCLASS32			\
+			 ? (void *) &ph32			\
 			 : (void *) &ph64)
-#define xph_sizeof	(class == ELFCLASS32		\
-			 ? sizeof ph32			\
+#define xph_sizeof	(class == ELFCLASS32			\
+			 ? sizeof ph32				\
 			 : sizeof ph64)
-#define xph_type	(class == ELFCLASS32		\
-			 ? getu32(swap, ph32.p_type)	\
-			 : getu32(swap, ph64.p_type))
-#define xph_offset	(off_t)(class == ELFCLASS32	\
-			 ? getu32(swap, ph32.p_offset)	\
-			 : getu64(swap, ph64.p_offset))
-#define xph_align	(size_t)((class == ELFCLASS32	\
-			 ? (off_t) (ph32.p_align ? 	\
-			    getu32(swap, ph32.p_align) : 4) \
-			 : (off_t) (ph64.p_align ?	\
-			    getu64(swap, ph64.p_align) : 4)))
-#define xph_filesz	(size_t)((class == ELFCLASS32	\
-			 ? getu32(swap, ph32.p_filesz)	\
-			 : getu64(swap, ph64.p_filesz)))
-#define xnh_addr	(class == ELFCLASS32		\
-			 ? (void *) &nh32		\
+#define xph_type	(class == ELFCLASS32			\
+			 ? elf_getu32(swap, ph32.p_type)	\
+			 : elf_getu32(swap, ph64.p_type))
+#define xph_offset	(off_t)(class == ELFCLASS32		\
+			 ? elf_getu32(swap, ph32.p_offset)	\
+			 : elf_getu64(swap, ph64.p_offset))
+#define xph_align	(size_t)((class == ELFCLASS32		\
+			 ? (off_t) (ph32.p_align ? 		\
+			    elf_getu32(swap, ph32.p_align) : 4) \
+			 : (off_t) (ph64.p_align ?		\
+			    elf_getu64(swap, ph64.p_align) : 4)))
+#define xph_filesz	(size_t)((class == ELFCLASS32		\
+			 ? elf_getu32(swap, ph32.p_filesz)	\
+			 : elf_getu64(swap, ph64.p_filesz)))
+#define xnh_addr	(class == ELFCLASS32			\
+			 ? (void *) &nh32			\
 			 : (void *) &nh64)
-#define xph_memsz	(size_t)((class == ELFCLASS32	\
-			 ? getu32(swap, ph32.p_memsz)	\
-			 : getu64(swap, ph64.p_memsz)))
-#define xnh_sizeof	(class == ELFCLASS32		\
-			 ? sizeof nh32			\
+#define xph_memsz	(size_t)((class == ELFCLASS32		\
+			 ? elf_getu32(swap, ph32.p_memsz)	\
+			 : elf_getu64(swap, ph64.p_memsz)))
+#define xnh_sizeof	(class == ELFCLASS32			\
+			 ? sizeof nh32				\
 			 : sizeof nh64)
-#define xnh_type	(class == ELFCLASS32		\
-			 ? getu32(swap, nh32.n_type)	\
-			 : getu32(swap, nh64.n_type))
-#define xnh_namesz	(class == ELFCLASS32		\
-			 ? getu32(swap, nh32.n_namesz)	\
-			 : getu32(swap, nh64.n_namesz))
-#define xnh_descsz	(class == ELFCLASS32		\
-			 ? getu32(swap, nh32.n_descsz)	\
-			 : getu32(swap, nh64.n_descsz))
-#define prpsoffsets(i)	(class == ELFCLASS32		\
-			 ? prpsoffsets32[i]		\
+#define xnh_type	(class == ELFCLASS32			\
+			 ? elf_getu32(swap, nh32.n_type)	\
+			 : elf_getu32(swap, nh64.n_type))
+#define xnh_namesz	(class == ELFCLASS32			\
+			 ? elf_getu32(swap, nh32.n_namesz)	\
+			 : elf_getu32(swap, nh64.n_namesz))
+#define xnh_descsz	(class == ELFCLASS32			\
+			 ? elf_getu32(swap, nh32.n_descsz)	\
+			 : elf_getu32(swap, nh64.n_descsz))
+#define prpsoffsets(i)	(class == ELFCLASS32			\
+			 ? prpsoffsets32[i]			\
 			 : prpsoffsets64[i])
 
 #ifdef ELFCORE
-size_t	prpsoffsets32[] = {
-	8,		/* FreeBSD */
+/*
+ * Try larger offsets first to avoid false matches
+ * from earlier data that happen to look like strings.
+ */
+static const size_t	prpsoffsets32[] = {
+#ifdef USE_NT_PSINFO
+	104,		/* SunOS 5.x (command line) */
+	88,		/* SunOS 5.x (short name) */
+#endif /* USE_NT_PSINFO */
+
+	100,		/* SunOS 5.x (command line) */
+	84,		/* SunOS 5.x (short name) */
+
+	44,		/* Linux (command line) */
 	28,		/* Linux 2.0.36 (short name) */
-	44,		/* Linux (path name) */
-	84,		/* SunOS 5.x */
+
+	8,		/* FreeBSD */
 };
 
-size_t	prpsoffsets64[] = {
-	16,		/* FreeBSD, 64-bit */
+static const size_t	prpsoffsets64[] = {
+#ifdef USE_NT_PSINFO
+	152,		/* SunOS 5.x (command line) */
+	136,		/* SunOS 5.x (short name) */
+#endif /* USE_NT_PSINFO */
+
+	136,		/* SunOS 5.x, 64-bit (command line) */
+	120,		/* SunOS 5.x, 64-bit (short name) */
+
+	56,		/* Linux (command line) */
 	40,             /* Linux (tested on core from 2.4.x, short name) */
-	56,		/* Linux (path name) */
-	120,		/* SunOS 5.x, 64-bit */
+
+	16,		/* FreeBSD, 64-bit */
 };
 
 #define	NOFFSETS32	(sizeof prpsoffsets32 / sizeof prpsoffsets32[0])
@@ -221,6 +244,14 @@ size_t	prpsoffsets64[] = {
  * SVR4-flavored systems, and Linux) containing the start of the
  * command line for that program.
  *
+ * SunOS 5.x core files contain two PT_NOTE sections, with the types
+ * NT_PRPSINFO (old) and NT_PSINFO (new).  These structs contain the
+ * same info about the command name and command line, so it probably
+ * isn't worthwhile to look for NT_PSINFO, but the offsets are provided
+ * above (see USE_NT_PSINFO), in case we ever decide to do so.  The
+ * NT_PRPSINFO and NT_PSINFO sections are always in order and adjacent;
+ * the SunOS 5.x file command relies on this (and prefers the latter).
+ *
  * The signal number probably appears in a section of type NT_PRSTATUS,
  * but that's also rather OS-dependent, in ways that are harder to
  * dissect with heuristics, so I'm not bothering with the signal number.
@@ -234,7 +265,7 @@ size_t	prpsoffsets64[] = {
 #define	OS_STYLE_FREEBSD	1
 #define	OS_STYLE_NETBSD		2
 
-private const char *os_style_names[] = {
+private const char os_style_names[][8] = {
 	"SVR4",
 	"FreeBSD",
 	"NetBSD",
@@ -242,6 +273,7 @@ private const char *os_style_names[] = {
 
 #define FLAGS_DID_CORE		1
 #define FLAGS_DID_NOTE		2
+#define FLAGS_DID_CORE_STYLE	4
 
 private int
 dophn_core(struct magic_set *ms, int class, int swap, int fd, off_t off,
@@ -384,7 +416,7 @@ donote(struct magic_set *ms, unsigned char *nbuf, size_t offset, size_t size,
 
 		if (file_printf(ms, ", for GNU/") == -1)
 			return size;
-		switch (getu32(swap, desc[0])) {
+		switch (elf_getu32(swap, desc[0])) {
 		case GNU_OS_LINUX:
 			if (file_printf(ms, "Linux") == -1)
 				return size;
@@ -397,12 +429,20 @@ donote(struct magic_set *ms, unsigned char *nbuf, size_t offset, size_t size,
 			if (file_printf(ms, "Solaris") == -1)
 				return size;
 			break;
+		case GNU_OS_KFREEBSD:
+			if (file_printf(ms, "kFreeBSD") == -1)
+				return size;
+			break;
+		case GNU_OS_KNETBSD:
+			if (file_printf(ms, "kNetBSD") == -1)
+				return size;
+			break;
 		default:
 			if (file_printf(ms, "<unknown>") == -1)
 				return size; 
 		}
-		if (file_printf(ms, " %d.%d.%d", getu32(swap, desc[1]),
-		    getu32(swap, desc[2]), getu32(swap, desc[3])) == -1)
+		if (file_printf(ms, " %d.%d.%d", elf_getu32(swap, desc[1]),
+		    elf_getu32(swap, desc[2]), elf_getu32(swap, desc[3])) == -1)
 			return size;
 		*flags |= FLAGS_DID_NOTE;
 		return size;
@@ -412,7 +452,7 @@ donote(struct magic_set *ms, unsigned char *nbuf, size_t offset, size_t size,
 	    xnh_type == NT_NETBSD_VERSION && descsz == 4) {
 		uint32_t desc;
 		(void)memcpy(&desc, &nbuf[doff], sizeof(desc));
-		desc = getu32(swap, desc);
+		desc = elf_getu32(swap, desc);
 
 		if (file_printf(ms, ", for NetBSD") == -1)
 			return size;
@@ -458,7 +498,7 @@ donote(struct magic_set *ms, unsigned char *nbuf, size_t offset, size_t size,
 	    xnh_type == NT_FREEBSD_VERSION && descsz == 4) {
 		uint32_t desc;
 		(void)memcpy(&desc, &nbuf[doff], sizeof(desc));
-		desc = getu32(swap, desc);
+		desc = elf_getu32(swap, desc);
 		if (file_printf(ms, ", for FreeBSD") == -1)
 			return size;
 
@@ -547,7 +587,7 @@ donote(struct magic_set *ms, unsigned char *nbuf, size_t offset, size_t size,
 		if (file_printf(ms, ", for DragonFly") == -1)
 			return size;
 		(void)memcpy(&desc, &nbuf[doff], sizeof(desc));
-		desc = getu32(swap, desc);
+		desc = elf_getu32(swap, desc);
 		if (file_printf(ms, " %d.%d.%d", desc / 100000,
 		    desc / 10000 % 10, desc % 10000) == -1)
 			return size;
@@ -588,10 +628,11 @@ core:
 	if ((*flags & FLAGS_DID_CORE) != 0)
 		return size;
 
-	if (os_style != -1) {
+	if (os_style != -1 && (*flags & FLAGS_DID_CORE_STYLE) == 0) {
 		if (file_printf(ms, ", %s-style", os_style_names[os_style])
 		    == -1)
 			return size;
+		*flags |= FLAGS_DID_CORE_STYLE;
 	}
 
 	switch (os_style) {
@@ -614,8 +655,9 @@ core:
 			(void)memcpy(&signo, &nbuf[doff + 0x08],
 			    sizeof(signo));
 			if (file_printf(ms, " (signal %u)",
-			    getu32(swap, signo)) == -1)
+			    elf_getu32(swap, signo)) == -1)
 				return size;
+			*flags |= FLAGS_DID_CORE;
 			return size;
 		}
 		break;
@@ -635,6 +677,7 @@ core:
 			 * reject it.
 			 */
 			for (i = 0; i < NOFFSETS; i++) {
+				unsigned char *cname, *cp;
 				size_t reloffset = prpsoffsets(i);
 				size_t noffset = doff + reloffset;
 				for (j = 0; j < 16; j++, noffset++,
@@ -682,9 +725,20 @@ core:
 				/*
 				 * Well, that worked.
 				 */
-				if (file_printf(ms, ", from '%.16s'",
-				    &nbuf[doff + prpsoffsets(i)]) == -1)
+				cname = (unsigned char *)
+				    &nbuf[doff + prpsoffsets(i)];
+				for (cp = cname; *cp && isprint(*cp); cp++)
+					continue;
+				/*
+				 * Linux apparently appends a space at the end
+				 * of the command line: remove it.
+				 */
+				while (cp > cname && isspace(cp[-1]))
+					cp--;
+				if (file_printf(ms, ", from '%.*s'",
+				    (int)(cp - cname), cname) == -1)
 					return size;
+				*flags |= FLAGS_DID_CORE;
 				return size;
 
 			tryanother:
@@ -694,7 +748,6 @@ core:
 		break;
 	}
 #endif
-	*flags |= FLAGS_DID_CORE;
 	return offset;
 }
 
@@ -805,7 +858,7 @@ dophn_exec(struct magic_set *ms, int class, int swap, int fd, off_t off,
 	
 	if (size != xph_sizeof) {
 		if (file_printf(ms, ", corrupted program header size") == -1)
-		    return -1;
+			return -1;
 		return 0;
 	}
 
@@ -885,6 +938,8 @@ dophn_exec(struct magic_set *ms, int class, int swap, int fd, off_t off,
 				return -1;
 			}
 			break;
+		default:
+			break;
 		}
 	}
 	if (file_printf(ms, ", %s linked%s", linking_style, shared_libraries)
@@ -907,112 +962,55 @@ file_tryelf(struct magic_set *ms, int fd, const unsigned char *buf,
 	struct stat st;
 	off_t fsize;
 	int flags = 0;
+	Elf32_Ehdr elf32hdr;
+	Elf64_Ehdr elf64hdr;
+	uint16_t type;
 
-	/*
-	 * If we cannot seek, it must be a pipe, socket or fifo.
-	 */
-	if((lseek(fd, (off_t)0, SEEK_SET) == (off_t)-1) && (errno == ESPIPE))
-		fd = file_pipe2file(ms, fd, buf, nbytes);
-
-	if (fstat(fd, &st) == -1) {
-  		file_badread(ms);
-		return -1;
-	}
-	fsize = st.st_size;
-
+	if (ms->flags & MAGIC_MIME)
+		return 0;
 	/*
 	 * ELF executables have multiple section headers in arbitrary
 	 * file locations and thus file(1) cannot determine it from easily.
 	 * Instead we traverse thru all section headers until a symbol table
 	 * one is found or else the binary is stripped.
+	 * Return immediately if it's not ELF (so we avoid pipe2file unless needed).
 	 */
 	if (buf[EI_MAG0] != ELFMAG0
 	    || (buf[EI_MAG1] != ELFMAG1 && buf[EI_MAG1] != OLFMAG1)
 	    || buf[EI_MAG2] != ELFMAG2 || buf[EI_MAG3] != ELFMAG3)
-	    return 0;
-
-
-	class = buf[EI_CLASS];
-
-	if (class == ELFCLASS32) {
-		Elf32_Ehdr elfhdr;
-		if (nbytes <= sizeof (Elf32_Ehdr))
-			return 0;
-
+		return 0;
 
-		u.l = 1;
-		(void) memcpy(&elfhdr, buf, sizeof elfhdr);
-		swap = (u.c[sizeof(int32_t) - 1] + 1) != elfhdr.e_ident[EI_DATA];
+	/*
+	 * If we cannot seek, it must be a pipe, socket or fifo.
+	 */
+	if((lseek(fd, (off_t)0, SEEK_SET) == (off_t)-1) && (errno == ESPIPE))
+		fd = file_pipe2file(ms, fd, buf, nbytes);
 
-		if (getu16(swap, elfhdr.e_type) == ET_CORE) {
-#ifdef ELFCORE
-			if (dophn_core(ms, class, swap, fd,
-			    (off_t)getu32(swap, elfhdr.e_phoff),
-			    getu16(swap, elfhdr.e_phnum), 
-			    (size_t)getu16(swap, elfhdr.e_phentsize),
-			    fsize, &flags) == -1)
-				return -1;
-#else
-			;
-#endif
-		} else {
-			if (getu16(swap, elfhdr.e_type) == ET_EXEC) {
-				if (dophn_exec(ms, class, swap,
-				    fd, (off_t)getu32(swap, elfhdr.e_phoff),
-				    getu16(swap, elfhdr.e_phnum), 
-				    (size_t)getu16(swap, elfhdr.e_phentsize),
-				    fsize, &flags)
-				    == -1)
-					return -1;
-			}
-			if (doshn(ms, class, swap, fd,
-			    (off_t)getu32(swap, elfhdr.e_shoff),
-			    getu16(swap, elfhdr.e_shnum),
-			    (size_t)getu16(swap, elfhdr.e_shentsize),
-			    &flags) == -1)
-				return -1;
-		}
-		return 1;
+	if (fstat(fd, &st) == -1) {
+  		file_badread(ms);
+		return -1;
 	}
+	fsize = st.st_size;
 
-        if (class == ELFCLASS64) {
-		Elf64_Ehdr elfhdr;
-		if (nbytes <= sizeof (Elf64_Ehdr))
-			return 0;
-
-
-		u.l = 1;
-		(void) memcpy(&elfhdr, buf, sizeof elfhdr);
-		swap = (u.c[sizeof(int32_t) - 1] + 1) != elfhdr.e_ident[EI_DATA];
+	class = buf[EI_CLASS];
 
-		if (getu16(swap, elfhdr.e_type) == ET_CORE) {
-#ifdef ELFCORE
-			if (dophn_core(ms, class, swap, fd,
-			    (off_t)elf_getu64(swap, elfhdr.e_phoff),
-			    getu16(swap, elfhdr.e_phnum), 
-			    (size_t)getu16(swap, elfhdr.e_phentsize),
-			    fsize, &flags) == -1)
-				return -1;
-#else
-			;
-#endif
-		} else {
-			if (getu16(swap, elfhdr.e_type) == ET_EXEC) {
-				if (dophn_exec(ms, class, swap, fd,
-				    (off_t)elf_getu64(swap, elfhdr.e_phoff),
-				    getu16(swap, elfhdr.e_phnum), 
-				    (size_t)getu16(swap, elfhdr.e_phentsize),
-				    fsize, &flags) == -1)
-					return -1;
-			}
-			if (doshn(ms, class, swap, fd,
-			    (off_t)elf_getu64(swap, elfhdr.e_shoff),
-			    getu16(swap, elfhdr.e_shnum),
-			    (size_t)getu16(swap, elfhdr.e_shentsize), &flags)
-			    == -1)
-				return -1;
-		}
-		return 1;
+	switch (class) {
+	case ELFCLASS32:
+#undef elf_getu
+#define elf_getu(a, b)	elf_getu32(a, b)
+#undef elfhdr
+#define elfhdr elf32hdr
+#include "elfclass.h"
+	case ELFCLASS64:
+#undef elf_getu
+#define elf_getu(a, b)	elf_getu64(a, b)
+#undef elfhdr
+#define elfhdr elf64hdr
+#include "elfclass.h"
+	default:
+	    if (file_printf(ms, ", unknown class %d", class) == -1)
+		    return -1;
+	    break;
 	}
 	return 0;
 }
diff --git a/usr.bin/file/readelf.h b/usr.bin/file/readelf.h
index 1ffab5a97a7..6d5bc30158f 100644
--- a/usr.bin/file/readelf.h
+++ b/usr.bin/file/readelf.h
@@ -1,4 +1,4 @@
-/*	$OpenBSD: readelf.h,v 1.6 2008/05/08 01:40:56 chl Exp $ */
+/*	$OpenBSD: readelf.h,v 1.7 2009/04/24 18:54:34 chl Exp $ */
 /*
  * Copyright (c) Christos Zoulas 2003.
  * All Rights Reserved.
@@ -45,7 +45,7 @@ typedef uint16_t	Elf32_Half;
 typedef uint32_t	Elf32_Word;
 typedef uint8_t		Elf32_Char;
 
-#if SIZEOF_UINT64_T != 8
+#if SIZEOF_LONG_LONG != 8
 #define USE_ARRAY_FOR_64BIT_TYPES
 typedef	uint32_t 	Elf64_Addr[2];
 typedef	uint32_t 	Elf64_Off[2];
@@ -97,7 +97,9 @@ typedef struct {
 } Elf64_Ehdr;
 
 /* e_type */
+#define ET_REL		1
 #define ET_EXEC		2
+#define ET_DYN		3
 #define ET_CORE		4
 
 /* sh_type */
@@ -191,12 +193,6 @@ typedef struct {
     Elf64_Off	sh_entsize;
 } Elf64_Shdr;
 
-/* Notes used in ET_CORE */
-#define NT_PRSTATUS	1
-#define NT_PRFPREG	2
-#define NT_PRPSINFO	3
-#define NT_TASKSTRUCT	4
-
 #define	NT_NETBSD_CORE_PROCINFO		1
 
 /* Note header in a PT_NOTE section */
@@ -212,10 +208,12 @@ typedef struct {
     Elf64_Word	n_type;
 } Elf64_Nhdr;
 
+/* Notes used in ET_CORE */
 #define	NT_PRSTATUS	1
 #define	NT_PRFPREG	2
 #define	NT_PRPSINFO	3
 #define	NT_PRXREG	4
+#define NT_TASKSTRUCT	4
 #define	NT_PLATFORM	5
 #define	NT_AUXV		6
 
@@ -233,5 +231,7 @@ typedef struct {
 #define GNU_OS_LINUX	0
 #define GNU_OS_HURD	1
 #define GNU_OS_SOLARIS	2
+#define GNU_OS_KFREEBSD	3
+#define GNU_OS_KNETBSD	4
 
 #endif
diff --git a/usr.bin/file/softmagic.c b/usr.bin/file/softmagic.c
index 701ebe41646..5290b613334 100644
--- a/usr.bin/file/softmagic.c
+++ b/usr.bin/file/softmagic.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: softmagic.c,v 1.13 2008/05/08 01:40:56 chl Exp $ */
+/*	$OpenBSD: softmagic.c,v 1.14 2009/04/24 18:54:34 chl Exp $ */
 /*
  * Copyright (c) Ian F. Darwin 1986-1995.
  * Software written by Ian F. Darwin and others;
@@ -39,11 +39,11 @@
 
 
 #ifndef	lint
-FILE_RCSID("@(#)$Id: softmagic.c,v 1.13 2008/05/08 01:40:56 chl Exp $")
+FILE_RCSID("@(#)$Id: softmagic.c,v 1.14 2009/04/24 18:54:34 chl Exp $")
 #endif	/* lint */
 
 private int match(struct magic_set *, struct magic *, uint32_t,
-    const unsigned char *, size_t);
+    const unsigned char *, size_t, int);
 private int mget(struct magic_set *, const unsigned char *,
     struct magic *, size_t, unsigned int);
 private int magiccheck(struct magic_set *, struct magic *);
@@ -59,17 +59,23 @@ private void cvt_32(union VALUETYPE *, const struct magic *);
 private void cvt_64(union VALUETYPE *, const struct magic *);
 
 /*
+ * Macro to give description string according to whether we want plain
+ * text or MIME type
+ */
+#define MAGIC_DESC ((ms->flags & MAGIC_MIME) ? m->mimetype : m->desc)
+
+/*
  * softmagic - lookup one file in parsed, in-memory copy of database
  * Passed the name and FILE * of one file to be typed.
  */
 /*ARGSUSED1*/		/* nbytes passed for regularity, maybe need later */
 protected int
-file_softmagic(struct magic_set *ms, const unsigned char *buf, size_t nbytes)
+file_softmagic(struct magic_set *ms, const unsigned char *buf, size_t nbytes, int mode)
 {
 	struct mlist *ml;
 	int rv;
 	for (ml = ms->mlist->next; ml != ms->mlist; ml = ml->next)
-		if ((rv = match(ms, ml->magic, ml->nmagic, buf, nbytes)) != 0)
+		if ((rv = match(ms, ml->magic, ml->nmagic, buf, nbytes, mode)) != 0)
 			return rv;
 
 	return 0;
@@ -104,7 +110,7 @@ file_softmagic(struct magic_set *ms, const unsigned char *buf, size_t nbytes)
  */
 private int
 match(struct magic_set *ms, struct magic *magic, uint32_t nmagic,
-    const unsigned char *s, size_t nbytes)
+    const unsigned char *s, size_t nbytes, int mode)
 {
 	uint32_t magindex = 0;
 	unsigned int cont_level = 0;
@@ -118,17 +124,26 @@ match(struct magic_set *ms, struct magic *magic, uint32_t nmagic,
 
 	for (magindex = 0; magindex < nmagic; magindex++) {
 		int flush;
+		struct magic *m = &magic[magindex];
+
+		if ((m->flag & BINTEST) != mode) {
+			/* Skip sub-tests */
+			while (magic[magindex + 1].cont_level != 0 &&
+			       ++magindex < nmagic)
+				continue;
+			continue; /* Skip to next top-level test*/
+		}
 
-		ms->offset = magic[magindex].offset;
-		ms->line = magic[magindex].lineno;
+		ms->offset = m->offset;
+		ms->line = m->lineno;
 
 		/* if main entry matches, print it... */
-		flush = !mget(ms, s, &magic[magindex], nbytes, cont_level);
+		flush = !mget(ms, s, m, nbytes, cont_level);
 		if (flush) {
-			if (magic[magindex].reln == '!')
+			if (m->reln == '!')
 				flush = 0;
 		} else {	
-			switch (magiccheck(ms, &magic[magindex])) {
+			switch (magiccheck(ms, m)) {
 			case -1:
 				return -1;
 			case 0:
@@ -153,15 +168,14 @@ match(struct magic_set *ms, struct magic *magic, uint32_t nmagic,
 		 * If we are going to print something, we'll need to print
 		 * a blank before we print something else.
 		 */
-		if (magic[magindex].desc[0]) {
+		if (*MAGIC_DESC) {
 			need_separator = 1;
 			printed_something = 1;
 			if (print_sep(ms, firstline) == -1)
 				return -1;
 		}
 
-		if ((ms->c.li[cont_level].off = mprint(ms, &magic[magindex]))
-		    == -1)
+		if ((ms->c.li[cont_level].off = mprint(ms, m)) == -1)
 			return -1;
 
 		/* and any continuations that match */
@@ -170,36 +184,36 @@ match(struct magic_set *ms, struct magic *magic, uint32_t nmagic,
 
 		while (magic[magindex+1].cont_level != 0 &&
 		    ++magindex < nmagic) {
-			ms->line = magic[magindex].lineno; /* for messages */
+			m = &magic[magindex];
+			ms->line = m->lineno; /* for messages */
 
-			if (cont_level < magic[magindex].cont_level)
+			if (cont_level < m->cont_level)
 				continue;
-			if (cont_level > magic[magindex].cont_level) {
+			if (cont_level > m->cont_level) {
 				/*
 				 * We're at the end of the level
 				 * "cont_level" continuations.
 				 */
-				cont_level = magic[magindex].cont_level;
+				cont_level = m->cont_level;
 			}
-			ms->offset = magic[magindex].offset;
-			if (magic[magindex].flag & OFFADD) {
+			ms->offset = m->offset;
+			if (m->flag & OFFADD) {
 				ms->offset +=
 				    ms->c.li[cont_level - 1].off;
 			}
 
 #ifdef ENABLE_CONDITIONALS
-			if (magic[magindex].cond == COND_ELSE ||
-			    magic[magindex].cond == COND_ELIF) {
+			if (m->cond == COND_ELSE ||
+			    m->cond == COND_ELIF) {
 				if (ms->c.li[cont_level].last_match == 1)
 					continue;
 			}
 #endif
-			flush = !mget(ms, s, &magic[magindex], nbytes,
-			    cont_level);
-			if (flush && magic[magindex].reln != '!')
+			flush = !mget(ms, s, m, nbytes, cont_level);
+			if (flush && m->reln != '!')
 				continue;
 				
-			switch (flush ? 1 : magiccheck(ms, &magic[magindex])) {
+			switch (flush ? 1 : magiccheck(ms, m)) {
 			case -1:
 				return -1;
 			case 0:
@@ -211,7 +225,7 @@ match(struct magic_set *ms, struct magic *magic, uint32_t nmagic,
 #ifdef ENABLE_CONDITIONALS
 				ms->c.li[cont_level].last_match = 1;
 #endif
-				if (magic[magindex].type != FILE_DEFAULT)
+				if (m->type != FILE_DEFAULT)
 					ms->c.li[cont_level].got_match = 1;
 				else if (ms->c.li[cont_level].got_match) {
 					ms->c.li[cont_level].got_match = 0;
@@ -221,7 +235,7 @@ match(struct magic_set *ms, struct magic *magic, uint32_t nmagic,
 				 * If we are going to print something,
 				 * make sure that we have a separator first.
 				 */
-				if (magic[magindex].desc[0]) {
+				if (*MAGIC_DESC) {
 					printed_something = 1;
 					if (print_sep(ms, firstline) == -1)
 						return -1;
@@ -234,15 +248,15 @@ match(struct magic_set *ms, struct magic *magic, uint32_t nmagic,
 				 */
 				/* space if previous printed */
 				if (need_separator
-				    && (magic[magindex].nospflag == 0)
-				    && (magic[magindex].desc[0] != '\0')) {
+				    && ((m->flag & NOSPACE) == 0)
+				    && *MAGIC_DESC) {
 					if (file_printf(ms, " ") == -1)
 						return -1;
 					need_separator = 0;
 				}
-				if ((ms->c.li[cont_level].off = mprint(ms, &magic[magindex])) == -1)
+				if ((ms->c.li[cont_level].off = mprint(ms, m)) == -1)
 					return -1;
-				if (magic[magindex].desc[0])
+				if (*MAGIC_DESC)
 					need_separator = 1;
 
 				/*
@@ -255,9 +269,10 @@ match(struct magic_set *ms, struct magic *magic, uint32_t nmagic,
 				break;
 			}
 		}
-		firstline = 0;
-		if (printed_something)
+		if (printed_something) {
+			firstline = 0;
 			returnval = 1;
+		}
 		if ((ms->flags & MAGIC_CONTINUE) == 0 && printed_something) {
 			return 1; /* don't keep searching */
 		}			
@@ -271,7 +286,7 @@ check_fmt(struct magic_set *ms, struct magic *m)
 	regex_t rx;
 	int rc;
 
-	if (strchr(m->desc, '%') == NULL)
+	if (strchr(MAGIC_DESC, '%') == NULL)
 		return 0;
 
 	rc = regcomp(&rx, "%[-0-9\\.]*s", REG_EXTENDED|REG_NOSUB);
@@ -281,7 +296,7 @@ check_fmt(struct magic_set *ms, struct magic *m)
 		file_magerror(ms, "regex error %d, (%s)", rc, errmsg);
 		return -1;
 	} else {
-		rc = regexec(&rx, m->desc, 0, 0, 0);
+		rc = regexec(&rx, MAGIC_DESC, 0, 0, 0);
 		regfree(&rx);
 		return !rc;
 	}
@@ -311,8 +326,10 @@ private int32_t
 mprint(struct magic_set *ms, struct magic *m)
 {
 	uint64_t v;
+	float vf;
+	double vd;
 	int64_t t = 0;
- 	char buf[512];
+ 	char *buf;
 	union VALUETYPE *p = &ms->ms_value;
 
   	switch (m->type) {
@@ -322,14 +339,13 @@ mprint(struct magic_set *ms, struct magic *m)
 		case -1:
 			return -1;
 		case 1:
-			if (snprintf(buf, sizeof(buf), "%c",
-			    (unsigned char)v) < 0)
+			if (asprintf(&buf, "%c", (unsigned char)v) < 0)
 				return -1;
-			if (file_printf(ms, m->desc, buf) == -1)
+			if (file_printf(ms, MAGIC_DESC, buf) == -1)
 				return -1;
 			break;
 		default:
-			if (file_printf(ms, m->desc, (unsigned char) v) == -1)
+			if (file_printf(ms, MAGIC_DESC, (unsigned char) v) == -1)
 				return -1;
 			break;
 		}
@@ -344,14 +360,13 @@ mprint(struct magic_set *ms, struct magic *m)
 		case -1:
 			return -1;
 		case 1:
-			if (snprintf(buf, sizeof(buf), "%hu",
-			    (unsigned short)v) < 0)
+			if (asprintf(&buf, "%hu", (unsigned short)v) < 0)
 				return -1;
-			if (file_printf(ms, m->desc, buf) == -1)
+			if (file_printf(ms, MAGIC_DESC, buf) == -1)
 				return -1;
 			break;
 		default:
-			if (file_printf(ms, m->desc, (unsigned short) v) == -1)
+			if (file_printf(ms, MAGIC_DESC, (unsigned short) v) == -1)
 				return -1;
 			break;
 		}
@@ -367,13 +382,13 @@ mprint(struct magic_set *ms, struct magic *m)
 		case -1:
 			return -1;
 		case 1:
-			if (snprintf(buf, sizeof(buf), "%u", (uint32_t)v) < 0)
+			if (asprintf(&buf, "%u", (uint32_t)v) < 0)
 				return -1;
-			if (file_printf(ms, m->desc, buf) == -1)
+			if (file_printf(ms, MAGIC_DESC, buf) == -1)
 				return -1;
 			break;
 		default:
-			if (file_printf(ms, m->desc, (uint32_t) v) == -1)
+			if (file_printf(ms, MAGIC_DESC, (uint32_t) v) == -1)
 				return -1;
 			break;
 		}
@@ -384,7 +399,7 @@ mprint(struct magic_set *ms, struct magic *m)
   	case FILE_BEQUAD:
   	case FILE_LEQUAD:
 		v = file_signextend(ms, m, p->q);
-		if (file_printf(ms, m->desc, (uint64_t) v) == -1)
+		if (file_printf(ms, MAGIC_DESC, (uint64_t) v) == -1)
 			return -1;
 		t = ms->offset + sizeof(int64_t);
   		break;
@@ -394,16 +409,18 @@ mprint(struct magic_set *ms, struct magic *m)
   	case FILE_BESTRING16:
   	case FILE_LESTRING16:
 		if (m->reln == '=' || m->reln == '!') {
-			if (file_printf(ms, m->desc, m->value.s) == -1)
+			if (file_printf(ms, MAGIC_DESC, m->value.s) == -1)
 				return -1;
 			t = ms->offset + m->vallen;
 		}
 		else {
 			if (*m->value.s == '\0')
 				p->s[strcspn(p->s, "\n")] = '\0';
-			if (file_printf(ms, m->desc, p->s) == -1)
+			if (file_printf(ms, MAGIC_DESC, p->s) == -1)
 				return -1;
 			t = ms->offset + strlen(p->s);
+			if (m->type == FILE_PSTRING)
+				t++;
 		}
 		break;
 
@@ -411,7 +428,7 @@ mprint(struct magic_set *ms, struct magic *m)
 	case FILE_BEDATE:
 	case FILE_LEDATE:
 	case FILE_MEDATE:
-		if (file_printf(ms, m->desc, file_fmttime(p->l, 1)) == -1)
+		if (file_printf(ms, MAGIC_DESC, file_fmttime(p->l, 1)) == -1)
 			return -1;
 		t = ms->offset + sizeof(time_t);
 		break;
@@ -420,7 +437,7 @@ mprint(struct magic_set *ms, struct magic *m)
 	case FILE_BELDATE:
 	case FILE_LELDATE:
 	case FILE_MELDATE:
-		if (file_printf(ms, m->desc, file_fmttime(p->l, 0)) == -1)
+		if (file_printf(ms, MAGIC_DESC, file_fmttime(p->l, 0)) == -1)
 			return -1;
 		t = ms->offset + sizeof(time_t);
 		break;
@@ -428,7 +445,7 @@ mprint(struct magic_set *ms, struct magic *m)
 	case FILE_QDATE:
 	case FILE_BEQDATE:
 	case FILE_LEQDATE:
-		if (file_printf(ms, m->desc, file_fmttime((uint32_t)p->q, 1))
+		if (file_printf(ms, MAGIC_DESC, file_fmttime((uint32_t)p->q, 1))
 		    == -1)
 			return -1;
 		t = ms->offset + sizeof(uint64_t);
@@ -437,12 +454,54 @@ mprint(struct magic_set *ms, struct magic *m)
 	case FILE_QLDATE:
 	case FILE_BEQLDATE:
 	case FILE_LEQLDATE:
-		if (file_printf(ms, m->desc, file_fmttime((uint32_t)p->q, 0))
+		if (file_printf(ms, MAGIC_DESC, file_fmttime((uint32_t)p->q, 0))
 		    == -1)
 			return -1;
 		t = ms->offset + sizeof(uint64_t);
 		break;
 
+  	case FILE_FLOAT:
+  	case FILE_BEFLOAT:
+  	case FILE_LEFLOAT:
+		vf = p->f;
+		switch (check_fmt(ms, m)) {
+		case -1:
+			return -1;
+		case 1:
+			if (asprintf(&buf, "%g", vf) < 0)
+				return -1;
+			if (file_printf(ms, MAGIC_DESC, buf) == -1)
+				return -1;
+			break;
+		default:
+			if (file_printf(ms, MAGIC_DESC, vf) == -1)
+				return -1;
+			break;
+		}
+		t = ms->offset + sizeof(float);
+  		break;
+
+  	case FILE_DOUBLE:
+  	case FILE_BEDOUBLE:
+  	case FILE_LEDOUBLE:
+		vd = p->d;
+		switch (check_fmt(ms, m)) {
+		case -1:
+			return -1;
+		case 1:
+			if (asprintf(&buf, "%g", vd) < 0)
+				return -1;
+			if (file_printf(ms, MAGIC_DESC, buf) == -1)
+				return -1;
+			break;
+		default:
+			if (file_printf(ms, MAGIC_DESC, vd) == -1)
+				return -1;
+			break;
+		}
+		t = ms->offset + sizeof(double);
+  		break;
+
 	case FILE_REGEX: {
 		char *cp;
 		int rval;
@@ -452,7 +511,7 @@ mprint(struct magic_set *ms, struct magic *m)
 			file_oomem(ms, ms->search.rm_len);
 			return -1;
 		}
-		rval = file_printf(ms, m->desc, cp);
+		rval = file_printf(ms, MAGIC_DESC, cp);
 		free(cp);
 
 		if (rval == -1)
@@ -466,7 +525,7 @@ mprint(struct magic_set *ms, struct magic *m)
 	}
 
 	case FILE_SEARCH:
-	  	if (file_printf(ms, m->desc, m->value.s) == -1)
+	  	if (file_printf(ms, MAGIC_DESC, m->value.s) == -1)
 			return -1;
 		if ((m->str_flags & REGEX_OFFSET_START))
 			t = ms->search.offset;
@@ -475,7 +534,7 @@ mprint(struct magic_set *ms, struct magic *m)
 		break;
 
 	case FILE_DEFAULT:
-	  	if (file_printf(ms, m->desc, m->value.s) == -1)
+	  	if (file_printf(ms, MAGIC_DESC, m->value.s) == -1)
 			return -1;
 		t = ms->offset;
 		break;
@@ -543,6 +602,35 @@ cvt_64(union VALUETYPE *p, const struct magic *m)
 	DO_CVT(q, (uint64_t));
 }
 
+#define DO_CVT2(fld, cast) \
+	if (m->num_mask) \
+		switch (m->mask_op & FILE_OPS_MASK) { \
+		case FILE_OPADD: \
+			p->fld += cast m->num_mask; \
+			break; \
+		case FILE_OPMINUS: \
+			p->fld -= cast m->num_mask; \
+			break; \
+		case FILE_OPMULTIPLY: \
+			p->fld *= cast m->num_mask; \
+			break; \
+		case FILE_OPDIVIDE: \
+			p->fld /= cast m->num_mask; \
+			break; \
+		} \
+
+private void
+cvt_float(union VALUETYPE *p, const struct magic *m)
+{
+	DO_CVT2(f, (float));
+}
+
+private void
+cvt_double(union VALUETYPE *p, const struct magic *m)
+{
+	DO_CVT2(d, (double));
+}
+
 /*
  * Convert the byte order of the data we are looking at
  * While we're here, let's apply the mask operation
@@ -609,10 +697,11 @@ mconvert(struct magic_set *ms, struct magic *m)
 	case FILE_BEQUAD:
 	case FILE_BEQDATE:
 	case FILE_BEQLDATE:
-		p->q = (int64_t)
-		    (((int64_t)p->hq[0]<<56)|((int64_t)p->hq[1]<<48)|
-		     ((int64_t)p->hq[2]<<40)|((int64_t)p->hq[3]<<32)|
-		     (p->hq[4]<<24)|(p->hq[5]<<16)|(p->hq[6]<<8)|(p->hq[7]));
+		p->q = (uint64_t)
+		    (((uint64_t)p->hq[0]<<56)|((uint64_t)p->hq[1]<<48)|
+		     ((uint64_t)p->hq[2]<<40)|((uint64_t)p->hq[3]<<32)|
+		     ((uint64_t)p->hq[4]<<24)|((uint64_t)p->hq[5]<<16)|
+		     ((uint64_t)p->hq[6]<<8)|((uint64_t)p->hq[7]));
 		cvt_64(p, m);
 		return 1;
 	case FILE_LESHORT:
@@ -629,10 +718,11 @@ mconvert(struct magic_set *ms, struct magic *m)
 	case FILE_LEQUAD:
 	case FILE_LEQDATE:
 	case FILE_LEQLDATE:
-		p->q = (int64_t)
-		    (((int64_t)p->hq[7]<<56)|((int64_t)p->hq[6]<<48)|
-		     ((int64_t)p->hq[5]<<40)|((int64_t)p->hq[4]<<32)|
-		     (p->hq[3]<<24)|(p->hq[2]<<16)|(p->hq[1]<<8)|(p->hq[0]));
+		p->q = (uint64_t)
+		    (((uint64_t)p->hq[7]<<56)|((uint64_t)p->hq[6]<<48)|
+		     ((uint64_t)p->hq[5]<<40)|((uint64_t)p->hq[4]<<32)|
+		     ((uint64_t)p->hq[3]<<24)|((uint64_t)p->hq[2]<<16)|
+		     ((uint64_t)p->hq[1]<<8)|((uint64_t)p->hq[0]));
 		cvt_64(p, m);
 		return 1;
 	case FILE_MELONG:
@@ -642,6 +732,36 @@ mconvert(struct magic_set *ms, struct magic *m)
 		    ((p->hl[1]<<24)|(p->hl[0]<<16)|(p->hl[3]<<8)|(p->hl[2]));
 		cvt_32(p, m);
 		return 1;
+	case FILE_FLOAT:
+		cvt_float(p, m);
+		return 1;
+	case FILE_BEFLOAT:
+		p->l =  ((uint32_t)p->hl[0]<<24)|((uint32_t)p->hl[1]<<16)|
+			((uint32_t)p->hl[2]<<8) |((uint32_t)p->hl[3]);
+		cvt_float(p, m);
+		return 1;
+	case FILE_LEFLOAT:
+		p->l =  ((uint32_t)p->hl[3]<<24)|((uint32_t)p->hl[2]<<16)|
+			((uint32_t)p->hl[1]<<8) |((uint32_t)p->hl[0]);
+		cvt_float(p, m);
+		return 1;
+	case FILE_DOUBLE:
+		cvt_double(p, m);
+		return 1;
+	case FILE_BEDOUBLE:
+		p->q =  ((uint64_t)p->hq[0]<<56)|((uint64_t)p->hq[1]<<48)|
+			((uint64_t)p->hq[2]<<40)|((uint64_t)p->hq[3]<<32)|
+			((uint64_t)p->hq[4]<<24)|((uint64_t)p->hq[5]<<16)|
+			((uint64_t)p->hq[6]<<8) |((uint64_t)p->hq[7]);
+		cvt_double(p, m);
+		return 1;
+	case FILE_LEDOUBLE:
+		p->q =  ((uint64_t)p->hq[7]<<56)|((uint64_t)p->hq[6]<<48)|
+			((uint64_t)p->hq[5]<<40)|((uint64_t)p->hq[4]<<32)|
+			((uint64_t)p->hq[3]<<24)|((uint64_t)p->hq[2]<<16)|
+			((uint64_t)p->hq[1]<<8) |((uint64_t)p->hq[0]);
+		cvt_double(p, m);
+		return 1;
 	case FILE_REGEX:
 	case FILE_SEARCH:
 	case FILE_DEFAULT:
@@ -675,13 +795,10 @@ mcopy(struct magic_set *ms, union VALUETYPE *p, int type, int indir,
 		case FILE_SEARCH:
 			ms->search.s = (const char *)s + offset;
 			ms->search.s_len = nbytes - offset;
+			ms->search.offset = offset;
 			return 0;
 
 		case FILE_REGEX: {
-			/*
-			 * offset is interpreted as last line to search,
-			 * (starting at 1), not as bytes-from start-of-file
-			 */
 			const char *b;
 			const char *c;
 			const char *last;	/* end of search region */
@@ -728,13 +845,17 @@ mcopy(struct magic_set *ms, union VALUETYPE *p, int type, int indir,
 				    offset);
 				return -1;
 			}
-			for (/*EMPTY*/; src < esrc; src++, dst++) {
+			for (/*EMPTY*/; src < esrc; src += 2, dst++) {
 				if (dst < edst)
-					*dst = *src++;
+					*dst = *src;
 				else
 					break;
-				if (*dst == '\0')
-					*dst = ' ';
+				if (*dst == '\0') {
+					if (type == FILE_BESTRING16 ?
+					    *(src - 1) != '\0' :
+					    *(src + 1) != '\0')
+						*dst = ' ';
+				}
 			}
 			*edst = '\0';
 			return 0;
@@ -772,7 +893,7 @@ mget(struct magic_set *ms, const unsigned char *s,
     struct magic *m, size_t nbytes, unsigned int cont_level)
 {
 	uint32_t offset = ms->offset;
-	uint32_t count = m->str_count;
+	uint32_t count = m->str_range;
 	union VALUETYPE *p = &ms->ms_value;
 
 	if (mcopy(ms, p, m->type, m->flag & INDIR, s, offset, nbytes, count) == -1)
@@ -1233,13 +1354,6 @@ mget(struct magic_set *ms, const unsigned char *s,
 				case FILE_OPMODULO:
 					offset = p->l % off;
 					break;
-			/*	case TOOMANYSWITCHBLOCKS:
-			 *		ugh = p->eye % m->strain;
-			 *		rub;
-			 *	case BEER:
-			 *		off = p->tab & m->in_gest;
-			 *		sleep;
-			 */
 				}
 			} else
 				offset = p->l;
@@ -1287,10 +1401,20 @@ mget(struct magic_set *ms, const unsigned char *s,
 	case FILE_BELDATE:
 	case FILE_LELDATE:
 	case FILE_MELDATE:
+	case FILE_FLOAT:
+	case FILE_BEFLOAT:
+	case FILE_LEFLOAT:
 		if (nbytes < (offset + 4))
 			return 0;
 		break;
 		
+	case FILE_DOUBLE:
+	case FILE_BEDOUBLE:
+	case FILE_LEDOUBLE:
+		if (nbytes < (offset + 8))
+			return 0;
+		break;
+
 	case FILE_STRING:
 	case FILE_PSTRING:
 	case FILE_SEARCH:
@@ -1326,10 +1450,8 @@ file_strncmp(const char *s1, const char *s2, size_t len, uint32_t flags)
 	uint64_t v;
 
 	/*
-	 * What we want here is:
-	 * v = strncmp(m->value.s, p->s, m->vallen);
-	 * but ignoring any nulls.  bcmp doesn't give -/+/0
-	 * and isn't universally available anyway.
+	 * What we want here is v = strncmp(s1, s2, len),
+	 * but ignoring any nulls.
 	 */
 	v = 0;
 	if (0L == flags) { /* normal string: do it fast */
@@ -1393,6 +1515,8 @@ magiccheck(struct magic_set *ms, struct magic *m)
 {
 	uint64_t l = m->value.q;
 	uint64_t v;
+	float fl, fv;
+	double dl, dv;
 	int matched;
 	union VALUETYPE *p = &ms->ms_value;
 
@@ -1434,6 +1558,72 @@ magiccheck(struct magic_set *ms, struct magic *m)
 		v = p->q;
 		break;
 
+	case FILE_FLOAT:
+	case FILE_BEFLOAT:
+	case FILE_LEFLOAT:
+		fl = m->value.f;
+		fv = p->f;
+		switch (m->reln) {
+		case 'x':
+			matched = 1;
+			break;
+	
+		case '!':
+			matched = fv != fl;
+			break;
+	
+		case '=':
+			matched = fv == fl;
+			break;
+	
+		case '>':
+			matched = fv > fl;
+			break;
+	
+		case '<':
+			matched = fv < fl;
+			break;
+	
+		default:
+			matched = 0;
+			file_magerror(ms, "cannot happen with float: invalid relation `%c'", m->reln);
+			return -1;
+		}
+		return matched;
+
+	case FILE_DOUBLE:
+	case FILE_BEDOUBLE:
+	case FILE_LEDOUBLE:
+		dl = m->value.d;
+		dv = p->d;
+		switch (m->reln) {
+		case 'x':
+			matched = 1;
+			break;
+	
+		case '!':
+			matched = dv != dl;
+			break;
+	
+		case '=':
+			matched = dv == dl;
+			break;
+	
+		case '>':
+			matched = dv > dl;
+			break;
+	
+		case '<':
+			matched = dv < dl;
+			break;
+	
+		default:
+			matched = 0;
+			file_magerror(ms, "cannot happen with double: invalid relation `%c'", m->reln);
+			return -1;
+		}
+		return matched;
+
 	case FILE_DEFAULT:
 		l = 0;
 		v = 0;
@@ -1461,15 +1651,14 @@ magiccheck(struct magic_set *ms, struct magic *m)
 		slen = MIN(m->vallen, sizeof(m->value.s));
 		l = 0;
 		v = 0;
-		ms->search.offset = m->offset;
 
-		for (idx = 0; m->str_count == 0 || idx < m->str_count; idx++) {
+		for (idx = 0; m->str_range == 0 || idx < m->str_range; idx++) {
 			if (slen + idx > ms->search.s_len)
 				break;
 
 			v = file_strncmp(m->value.s, ms->search.s + idx, slen, m->str_flags);
 			if (v == 0) {	/* found match */
-				ms->search.offset = m->offset + idx;
+				ms->search.offset += idx;
 				break;
 			}
 		}
diff --git a/usr.bin/file/tar.h b/usr.bin/file/tar.h
index 0a10238695b..9e6f3a84f3b 100644
--- a/usr.bin/file/tar.h
+++ b/usr.bin/file/tar.h
@@ -1,4 +1,4 @@
-/*	$OpenBSD: tar.h,v 1.6 2008/05/08 01:40:56 chl Exp $ */
+/*	$OpenBSD: tar.h,v 1.7 2009/04/24 18:54:34 chl Exp $ */
 /*
  * Copyright (c) Ian F. Darwin 1986-1995.
  * Software written by Ian F. Darwin and others;
@@ -33,20 +33,10 @@
  *
  * Created 25 August 1985 by John Gilmore, ihnp4!hoptoad!gnu.
  *
- * $Id: tar.h,v 1.6 2008/05/08 01:40:56 chl Exp $ # checkin only
+ * $Id: tar.h,v 1.7 2009/04/24 18:54:34 chl Exp $ # checkin only
  */
 
 /*
- * Kludge for handling systems that cannot cope with multiple
- * external definitions of a variable.  In ONE routine (tar.c),
- * we #define TAR_EXTERN to null; here, we set it to "extern" if
- * it is not already set.
- */
-#ifndef TAR_EXTERN
-#define TAR_EXTERN extern
-#endif
-
-/*
  * Header block on tape.
  *
  * I'm going to use traditional DP naming conventions here.
@@ -79,58 +69,6 @@ union record {
 	} header;
 };
 
-/* The checksum field is filled with this while the checksum is computed. */
-#define	CHKBLANKS	"        "	/* 8 blanks, no null */
-
 /* The magic field is filled with this if uname and gname are valid. */
 #define	TMAGIC		"ustar"		/* 5 chars and a null */
 #define	GNUTMAGIC	"ustar  "	/* 7 chars and a null */
-
-/* The linkflag defines the type of file */
-#define	LF_OLDNORMAL	'\0'		/* Normal disk file, Unix compat */
-#define	LF_NORMAL	'0'		/* Normal disk file */
-#define	LF_LINK		'1'		/* Link to previously dumped file */
-#define	LF_SYMLINK	'2'		/* Symbolic link */
-#define	LF_CHR		'3'		/* Character special file */
-#define	LF_BLK		'4'		/* Block special file */
-#define	LF_DIR		'5'		/* Directory */
-#define	LF_FIFO		'6'		/* FIFO special file */
-#define	LF_CONTIG	'7'		/* Contiguous file */
-/* Further link types may be defined later. */
-
-/*
- * Exit codes from the "tar" program
- */
-#define	EX_SUCCESS	0		/* success! */
-#define	EX_ARGSBAD	1		/* invalid args */
-#define	EX_BADFILE	2		/* invalid filename */
-#define	EX_BADARCH	3		/* bad archive */
-#define	EX_SYSTEM	4		/* system gave unexpected error */
-
-/*
- * Structure for keeping track of filenames and lists thereof.
- */
-struct name {
-	struct name	*next;
-	short		length;
-	char		found;
-	char		name[NAMSIZ+1];
-};
-
-/*
- *
- * Due to the next struct declaration, each routine that includes
- * "tar.h" must also include <sys/types.h>.  I tried to make it automatic,
- * but System V has no defines in <sys/types.h>, so there is no way of
- * knowing when it has been included.  In addition, it cannot be included
- * twice, but must be included exactly once.  Argghh!
- *
- * Thanks, typedef.  Thanks, USG.
- */
-struct link {
-	struct link	*next;
-	dev_t		dev;
-	ino_t		ino;
-	short		linkcount;
-	char		name[NAMSIZ+1];
-};