2 #------------------------------------------------------------------------------
3 # $File: msdos,v 1.124 2018/07/10 04:05:50 christos Exp $
4 # msdos: file(1) magic for MS-DOS files
7 # .BAT files (Daniel Quinlan, quinlan@yggdrasil.com)
8 # updated by Joerg Jenderek at Oct 2008,Apr 2011
10 >1 string/cW \ echo\ off DOS batch file text
11 !:mime text/x-msdos-batch
12 >1 string/cW echo\ off DOS batch file text
13 !:mime text/x-msdos-batch
14 >1 string/cW rem DOS batch file text
15 !:mime text/x-msdos-batch
16 >1 string/cW set\ DOS batch file text
17 !:mime text/x-msdos-batch
20 # OS/2 batch files are REXX. the second regex is a bit generic, oh well
21 # the matched commands seem to be common in REXX and uncommon elsewhere
22 100 search/0xffff rxfuncadd
23 >100 regex/c =^[\ \t]{0,10}call[\ \t]{1,10}rxfunc OS/2 REXX batch file text
25 >100 regex/c =^[\ \t]{0,10}say\ ['"] OS/2 REXX batch file text
27 # updated by Joerg Jenderek at Oct 2015
28 # https://de.wikipedia.org/wiki/Common_Object_File_Format
29 # http://www.delorie.com/djgpp/doc/coff/filhdr.html
30 # ./intel already labeled COFF type 0x14c=0514 as "80386 COFF executable"
31 #0 leshort 0x14c MS Windows COFF Intel 80386 object file
33 0 leshort 0x166 MS Windows COFF MIPS R4000 object file
35 0 leshort 0x184 MS Windows COFF Alpha object file
37 0 leshort 0x268 MS Windows COFF Motorola 68000 object file
39 0 leshort 0x1f0 MS Windows COFF PowerPC object file
41 0 leshort 0x290 MS Windows COFF PA-RISC object file
44 # Tests for various EXE types.
46 # Many of the compressed formats were extraced from IDARC 1.23 source code.
49 # All non-DOS EXE extensions have the relocation table more than 0x40 bytes into the file.
50 >0x18 leshort <0x40 MS-DOS executable
51 !:mime application/x-dosexec
52 # These traditional tests usually work but not always. When test quality support is
53 # implemented these can be turned on.
54 #>>0x18 leshort 0x1c (Borland compiler)
55 #>>0x18 leshort 0x1e (MS compiler)
57 # If the relocation table is 0x40 or more bytes into the file, it's definitely
62 >>(0x3c.l) string PE\0\0 PE
63 !:mime application/x-dosexec
64 >>>(0x3c.l+24) leshort 0x010b \b32 executable
65 >>>(0x3c.l+24) leshort 0x020b \b32+ executable
66 >>>(0x3c.l+24) leshort 0x0107 ROM image
67 >>>(0x3c.l+24) default x Unknown PE signature
69 >>>(0x3c.l+22) leshort&0x2000 >0 (DLL)
70 >>>(0x3c.l+92) leshort 1 (native)
71 >>>(0x3c.l+92) leshort 2 (GUI)
72 >>>(0x3c.l+92) leshort 3 (console)
73 >>>(0x3c.l+92) leshort 7 (POSIX)
74 >>>(0x3c.l+92) leshort 9 (Windows CE)
75 >>>(0x3c.l+92) leshort 10 (EFI application)
76 >>>(0x3c.l+92) leshort 11 (EFI boot service driver)
77 >>>(0x3c.l+92) leshort 12 (EFI runtime driver)
78 >>>(0x3c.l+92) leshort 13 (EFI ROM)
79 >>>(0x3c.l+92) leshort 14 (XBOX)
80 >>>(0x3c.l+92) leshort 15 (Windows boot application)
81 >>>(0x3c.l+92) default x (Unknown subsystem
82 >>>>&0 leshort x 0x%x)
83 >>>(0x3c.l+4) leshort 0x14c Intel 80386
84 >>>(0x3c.l+4) leshort 0x166 MIPS R4000
85 >>>(0x3c.l+4) leshort 0x168 MIPS R10000
86 >>>(0x3c.l+4) leshort 0x184 Alpha
87 >>>(0x3c.l+4) leshort 0x1a2 Hitachi SH3
88 >>>(0x3c.l+4) leshort 0x1a6 Hitachi SH4
89 >>>(0x3c.l+4) leshort 0x1c0 ARM
90 >>>(0x3c.l+4) leshort 0x1c2 ARM Thumb
91 >>>(0x3c.l+4) leshort 0x1c4 ARMv7 Thumb
92 >>>(0x3c.l+4) leshort 0x1f0 PowerPC
93 >>>(0x3c.l+4) leshort 0x200 Intel Itanium
94 >>>(0x3c.l+4) leshort 0x266 MIPS16
95 >>>(0x3c.l+4) leshort 0x268 Motorola 68000
96 >>>(0x3c.l+4) leshort 0x290 PA-RISC
97 >>>(0x3c.l+4) leshort 0x366 MIPSIV
98 >>>(0x3c.l+4) leshort 0x466 MIPS16 with FPU
99 >>>(0x3c.l+4) leshort 0xebc EFI byte code
100 >>>(0x3c.l+4) leshort 0x8664 x86-64
101 >>>(0x3c.l+4) leshort 0xc0ee MSIL
102 >>>(0x3c.l+4) default x Unknown processor type
103 >>>>&0 leshort x 0x%x
104 >>>(0x3c.l+22) leshort&0x0200 >0 (stripped to external PDB)
105 >>>(0x3c.l+22) leshort&0x1000 >0 system file
106 >>>(0x3c.l+24) leshort 0x010b
107 >>>>(0x3c.l+232) lelong >0 Mono/.Net assembly
108 >>>(0x3c.l+24) leshort 0x020b
109 >>>>(0x3c.l+248) lelong >0 Mono/.Net assembly
111 # hooray, there's a DOS extender using the PE format, with a valid PE
112 # executable inside (which just prints a message and exits if run in win)
113 >>>(8.s*16) string 32STUB \b, 32rtm DOS extender
114 >>>(8.s*16) string !32STUB \b, for MS Windows
115 >>>(0x3c.l+0xf8) string UPX0 \b, UPX compressed
116 >>>(0x3c.l+0xf8) search/0x140 PEC2 \b, PECompact2 compressed
117 >>>(0x3c.l+0xf8) search/0x140 UPX2
118 >>>>(&0x10.l+(-4)) string PK\3\4 \b, ZIP self-extracting archive (Info-Zip)
119 >>>(0x3c.l+0xf8) search/0x140 .idata
120 >>>>(&0xe.l+(-4)) string PK\3\4 \b, ZIP self-extracting archive (Info-Zip)
121 >>>>(&0xe.l+(-4)) string ZZ0 \b, ZZip self-extracting archive
122 >>>>(&0xe.l+(-4)) string ZZ1 \b, ZZip self-extracting archive
123 >>>(0x3c.l+0xf8) search/0x140 .rsrc
124 >>>>(&0x0f.l+(-4)) string a\\\4\5 \b, WinHKI self-extracting archive
125 >>>>(&0x0f.l+(-4)) string Rar! \b, RAR self-extracting archive
126 >>>>(&0x0f.l+(-4)) search/0x3000 MSCF \b, InstallShield self-extracting archive
127 >>>>(&0x0f.l+(-4)) search/32 Nullsoft \b, Nullsoft Installer self-extracting archive
128 >>>(0x3c.l+0xf8) search/0x140 .data
129 >>>>(&0x0f.l) string WEXTRACT \b, MS CAB-Installer self-extracting archive
130 >>>(0x3c.l+0xf8) search/0x140 .petite\0 \b, Petite compressed
131 >>>>(0x3c.l+0xf7) byte x
132 >>>>>(&0x104.l+(-4)) string =!sfx! \b, ACE self-extracting archive
133 >>>(0x3c.l+0xf8) search/0x140 .WISE \b, WISE installer self-extracting archive
134 >>>(0x3c.l+0xf8) search/0x140 .dz\0\0\0 \b, Dzip self-extracting archive
135 >>>&(0x3c.l+0xf8) search/0x100 _winzip_ \b, ZIP self-extracting archive (WinZip)
136 >>>&(0x3c.l+0xf8) search/0x100 SharedD \b, Microsoft Installer self-extracting archive
137 >>>0x30 string Inno \b, InnoSetup self-extracting archive
139 # Hmm, not a PE but the relocation table is too high for a traditional DOS exe,
140 # must be one of the unusual subformats.
141 >>(0x3c.l) string !PE\0\0 MS-DOS executable
142 !:mime application/x-dosexec
144 >>(0x3c.l) string NE \b, NE
145 !:mime application/x-dosexec
146 >>>(0x3c.l+0x36) byte 1 for OS/2 1.x
147 >>>(0x3c.l+0x36) byte 2 for MS Windows 3.x
148 >>>(0x3c.l+0x36) byte 3 for MS-DOS
149 >>>(0x3c.l+0x36) byte 4 for Windows 386
150 >>>(0x3c.l+0x36) byte 5 for Borland Operating System Services
151 >>>(0x3c.l+0x36) default x
152 >>>>(0x3c.l+0x36) byte x (unknown OS %x)
153 >>>(0x3c.l+0x36) byte 0x81 for MS-DOS, Phar Lap DOS extender
154 >>>(0x3c.l+0x0c) leshort&0x8003 0x8002 (DLL)
155 >>>(0x3c.l+0x0c) leshort&0x8003 0x8001 (driver)
156 >>>&(&0x24.s-1) string ARJSFX \b, ARJ self-extracting archive
157 >>>(0x3c.l+0x70) search/0x80 WinZip(R)\ Self-Extractor \b, ZIP self-extracting archive (WinZip)
159 >>(0x3c.l) string LX\0\0 \b, LX
160 !:mime application/x-dosexec
161 >>>(0x3c.l+0x0a) leshort <1 (unknown OS)
162 >>>(0x3c.l+0x0a) leshort 1 for OS/2
163 >>>(0x3c.l+0x0a) leshort 2 for MS Windows
164 >>>(0x3c.l+0x0a) leshort 3 for DOS
165 >>>(0x3c.l+0x0a) leshort >3 (unknown OS)
166 >>>(0x3c.l+0x10) lelong&0x28000 =0x8000 (DLL)
167 >>>(0x3c.l+0x10) lelong&0x20000 >0 (device driver)
168 >>>(0x3c.l+0x10) lelong&0x300 0x300 (GUI)
169 >>>(0x3c.l+0x10) lelong&0x28300 <0x300 (console)
170 >>>(0x3c.l+0x08) leshort 1 i80286
171 >>>(0x3c.l+0x08) leshort 2 i80386
172 >>>(0x3c.l+0x08) leshort 3 i80486
173 >>>(8.s*16) string emx \b, emx
175 >>>&(&0x54.l-3) string arjsfx \b, ARJ self-extracting archive
177 # MS Windows system file, supposedly a collection of LE executables
178 >>(0x3c.l) string W3 \b, W3 for MS Windows
179 !:mime application/x-dosexec
181 >>(0x3c.l) string LE\0\0 \b, LE executable
182 !:mime application/x-dosexec
183 >>>(0x3c.l+0x0a) leshort 1
184 # some DOS extenders use LE files with OS/2 header
185 >>>>0x240 search/0x100 DOS/4G for MS-DOS, DOS4GW DOS extender
186 >>>>0x240 search/0x200 WATCOM\ C/C++ for MS-DOS, DOS4GW DOS extender
187 >>>>0x440 search/0x100 CauseWay\ DOS\ Extender for MS-DOS, CauseWay DOS extender
188 >>>>0x40 search/0x40 PMODE/W for MS-DOS, PMODE/W DOS extender
189 >>>>0x40 search/0x40 STUB/32A for MS-DOS, DOS/32A DOS extender (stub)
190 >>>>0x40 search/0x80 STUB/32C for MS-DOS, DOS/32A DOS extender (configurable stub)
191 >>>>0x40 search/0x80 DOS/32A for MS-DOS, DOS/32A DOS extender (embedded)
192 # this is a wild guess; hopefully it is a specific signature
193 >>>>&0x24 lelong <0x50
194 >>>>>(&0x4c.l) string \xfc\xb8WATCOM
195 >>>>>>&0 search/8 3\xdbf\xb9 \b, 32Lite compressed
196 # another wild guess: if real OS/2 LE executables exist, they probably have higher start EIP
197 #>>>>(0x3c.l+0x1c) lelong >0x10000 for OS/2
198 # fails with DOS-Extenders.
199 >>>(0x3c.l+0x0a) leshort 2 for MS Windows
200 >>>(0x3c.l+0x0a) leshort 3 for DOS
201 >>>(0x3c.l+0x0a) leshort 4 for MS Windows (VxD)
202 >>>(&0x7c.l+0x26) string UPX \b, UPX compressed
203 >>>&(&0x54.l-3) string UNACE \b, ACE self-extracting archive
205 # looks like ASCII, probably some embedded copyright message.
206 # and definitely not NE/LE/LX/PE
207 >>0x3c lelong >0x20000000
208 >>>(4.s*512) leshort !0x014c \b, MZ for MS-DOS
209 !:mime application/x-dosexec
210 # header data too small for extended executable
213 >>>(4.s*512) leshort !0x014c
215 >>>>&(2.s-514) string !LE
216 >>>>>&-2 string !BW \b, MZ for MS-DOS
217 !:mime application/x-dosexec
218 >>>>&(2.s-514) string LE \b, LE
219 >>>>>0x240 search/0x100 DOS/4G for MS-DOS, DOS4GW DOS extender
220 # educated guess since indirection is still not capable enough for complex offset
221 # calculations (next embedded executable would be at &(&2*512+&0-2)
222 # I suspect there are only LE executables in these multi-exe files
223 >>>>&(2.s-514) string BW
224 >>>>>0x240 search/0x100 DOS/4G \b, LE for MS-DOS, DOS4GW DOS extender (embedded)
225 >>>>>0x240 search/0x100 !DOS/4G \b, BW collection for MS-DOS
227 # This sequence skips to the first COFF segment, usually .text
228 >(4.s*512) leshort 0x014c \b, COFF
229 !:mime application/x-dosexec
230 >>(8.s*16) string go32stub for MS-DOS, DJGPP go32 DOS extender
231 >>(8.s*16) string emx
232 >>>&1 string x for DOS, Win or OS/2, emx %s
233 >>&(&0x42.l-3) byte x
234 >>>&0x26 string UPX \b, UPX compressed
235 # and yet another guess: small .text, and after large .data is unusal, could be 32lite
236 >>&0x2c search/0xa0 .text
237 >>>&0x0b lelong <0x2000
238 >>>>&0 lelong >0x6000 \b, 32lite compressed
240 >(8.s*16) string $WdX \b, WDos/X DOS extender
242 # By now an executable type should have been printed out. The executable
243 # may be a self-uncompressing archive, so look for evidence of that and
246 # Some signatures below from Greg Roelofs, newt@uchicago.edu.
248 >0x35 string \x8e\xc0\xb9\x08\x00\xf3\xa5\x4a\x75\xeb\x8e\xc3\x8e\xd8\x33\xff\xbe\x30\x00\x05 \b, aPack compressed
249 >0xe7 string LH/2\ Self-Extract \b, %s
250 >0x1c string UC2X \b, UCEXE compressed
251 >0x1c string WWP\ \b, WWPACK compressed
252 >0x1c string RJSX \b, ARJ self-extracting archive
253 >0x1c string diet \b, diet compressed
254 >0x1c string LZ09 \b, LZEXE v0.90 compressed
255 >0x1c string LZ91 \b, LZEXE v0.91 compressed
256 >0x1c string tz \b, TinyProg compressed
257 >0x1e string Copyright\ 1989-1990\ PKWARE\ Inc. Self-extracting PKZIP archive
258 !:mime application/zip
259 # Yes, this really is "Copr", not "Corp."
260 >0x1e string PKLITE\ Copr. Self-extracting PKZIP archive
261 !:mime application/zip
262 # winarj stores a message in the stub instead of the sig in the MZ header
263 >0x20 search/0xe0 aRJsfX \b, ARJ self-extracting archive
265 >>0x23 string 2 \b, AIN 2.x compressed
266 >>0x23 string <2 \b, AIN 1.x compressed
267 >>0x23 string >2 \b, AIN 1.x compressed
268 >0x24 string LHa's\ SFX \b, LHa self-extracting archive
269 !:mime application/x-lha
270 >0x24 string LHA's\ SFX \b, LHa self-extracting archive
271 !:mime application/x-lha
272 >0x24 string \ $ARX \b, ARX self-extracting archive
273 >0x24 string \ $LHarc \b, LHarc self-extracting archive
274 >0x20 string SFX\ by\ LARC \b, LARC self-extracting archive
275 >0x40 string aPKG \b, aPackage self-extracting archive
276 >0x64 string W\ Collis\0\0 \b, Compack compressed
277 >0x7a string Windows\ self-extracting\ ZIP \b, ZIP self-extracting archive
278 >>&0xf4 search/0x140 \x0\x40\x1\x0
279 >>>(&0.l+(4)) string MSCF \b, WinHKI CAB self-extracting archive
280 >1638 string -lh5- \b, LHa self-extracting archive v2.13S
281 >0x17888 string Rar! \b, RAR self-extracting archive
283 # Skip to the end of the EXE. This will usually work fine in the PE case
284 # because the MZ image is hardcoded into the toolchain and almost certainly
285 # won't match any of these signatures.
288 >>>&0 string PK\3\4 \b, ZIP self-extracting archive
289 >>>&0 string Rar! \b, RAR self-extracting archive
290 >>>&0 string =!\x11 \b, AIN 2.x self-extracting archive
291 >>>&0 string =!\x12 \b, AIN 2.x self-extracting archive
292 >>>&0 string =!\x17 \b, AIN 1.x self-extracting archive
293 >>>&0 string =!\x18 \b, AIN 1.x self-extracting archive
294 >>>&7 search/400 **ACE** \b, ACE self-extracting archive
295 >>>&0 search/0x480 UC2SFX\ Header \b, UC2 self-extracting archive
297 # a few unknown ZIP sfxes, no idea if they are needed or if they are
298 # already captured by the generic patterns above
299 >(8.s*16) search/0x20 PKSFX \b, ZIP self-extracting archive (PKZIP)
300 # TODO: how to add this? >FileSize-34 string Windows\ Self-Installing\ Executable \b, ZIP self-extracting archive
303 # TELVOX Teleinformatica CODEC self-extractor for OS/2:
304 >49801 string \x79\xff\x80\xff\x76\xff \b, CODEC archive v3.21
305 >>49824 leshort =1 \b, 1 file
306 >>49824 leshort >1 \b, %u files
308 # added by Joerg Jenderek of http://www.freedos.org/software/?prog=kc
309 # and http://www.freedos.org/software/?prog=kpdos
310 # for FreeDOS files like KEYBOARD.SYS, KEYBRD2.SYS, KEYBRD3.SYS, *.KBD
311 0 string/b KCF FreeDOS KEYBoard Layout collection
312 # only version=0x100 found
313 >3 uleshort x \b, version 0x%x
314 # length of string containing author,info and special characters
316 #>>6 pstring x \b, name=%s
317 >>7 string >\0 \b, author=%-.14s
318 >>7 search/254 \xff \b, info=
319 #>>>&0 string x \b%-s
320 >>>&0 string x \b%-.15s
321 # for FreeDOS *.KL files
322 0 string/b KLF FreeDOS KEYBoard Layout file
323 # only version=0x100 or 0x101 found
324 >3 uleshort x \b, version 0x%x
327 >>8 string x \b, name=%-.2s
328 0 string \xffKEYB\ \ \ \0\0\0\0
329 >12 string \0\0\0\0`\004\360 MS-DOS KEYBoard Layout file
331 # DOS device driver updated by Joerg Jenderek at May 2011,Mar 2017
332 # https://amaus.net/static/S100/IBM/software/DOS/DOS%20techref/CHAPTER.009
333 0 ulequad&0x07a0ffffffff 0xffffffff
335 0 name msdos-driver DOS executable (
336 #!:mime application/octet-stream
337 !:mime application/x-dosdriver
338 # also found FreeDOS print driver SPOOL.DEV and disc compression driver STACLOAD.BIN
340 >40 search/7 UPX! \bUPX compressed
341 # DOS device driver attributes
342 >4 uleshort&0x8000 0x0000 \bblock device driver
344 >4 uleshort&0x8000 0x8000 \b
345 >>4 uleshort&0x0008 0x0008 \bclock
346 # fast video output by int 29h
347 >>4 uleshort&0x0010 0x0010 \bfast
348 # standard input/output device
349 >>4 uleshort&0x0003 >0 \bstandard
350 >>>4 uleshort&0x0001 0x0001 \binput
351 >>>4 uleshort&0x0003 0x0003 \b/
352 >>>4 uleshort&0x0002 0x0002 \boutput
353 >>4 uleshort&0x8000 0x8000 \bcharacter device driver
355 # upx compressed device driver has garbage instead of real in name field of header
358 # leading/trailing nulls, zeros or non ASCII characters in 8-byte name field at offset 10 are skipped
362 >>>>>>10 ubyte !0x2A \b%c
364 >>>>>11 ubyte !0x2E \b%c
367 >>>>>>12 ubyte !0x2E \b%c
369 >>>>13 ubyte !0x2E \b%c
371 >>>>>14 ubyte !0x2E \b%c
373 >>>>>15 ubyte !0x2E \b%c
376 >>>>>>16 ubyte <0xCB \b%c
379 >>>>>>17 ubyte <0x90 \b%c
380 # some character device drivers like ASPICD.SYS, btcdrom.sys and Cr_atapi.sys contain only spaces or points in name field
382 # they have their real name at offset 22
383 # also block device drivers like DUMBDRV.SYS
384 >>>>22 string >\056 %-.6s
385 >4 uleshort&0x8000 0x0000
386 # 32 bit sector addressing ( > 32 MB) for block devices
387 >>4 uleshort&0x0002 0x0002 \b,32-bit sector-
388 # support by driver functions 13h, 17h, 18h
389 >4 uleshort&0x0040 0x0040 \b,IOCTL-
390 # open, close, removable media support by driver functions 0Dh, 0Eh, 0Fh
391 >4 uleshort&0x0800 0x0800 \b,close media-
392 # output until busy support by int 10h for character device driver
393 >4 uleshort&0x8000 0x8000
394 >>4 uleshort&0x2000 0x2000 \b,until busy-
395 # direct read/write support by driver functions 03h,0Ch
396 >4 uleshort&0x4000 0x4000 \b,control strings-
397 >4 uleshort&0x8000 0x8000
398 >>4 uleshort&0x6840 >0 \bsupport
399 >4 uleshort&0x8000 0x0000
400 >>4 uleshort&0x4842 >0 \bsupport
402 # DOS driver cmd640x.sys has 0x12 instead of 0xffffffff for pointer field to next device header
403 0 ulequad 0x0513c00000000012
405 # DOS drivers DC2975.SYS, DUMBDRV.SYS, ECHO.SYS has also none 0xffffffff for pointer field
406 0 ulequad 0x32f28000ffff0016
408 0 ulequad 0x007f00000000ffff
410 0 ulequad 0x001600000000ffff
412 # DOS drivers LS120.SYS, MKELS120.SYS use reserved bits of attribute field
413 0 ulequad 0x0bf708c2ffffffff
415 0 ulequad 0x07bd08c2ffffffff
418 # updated by Joerg Jenderek
419 # GRR: line below too general as it catches also
420 # rt.lib DYADISKS.PIC and many more
421 # start with assembler instruction MOV
423 # skip "AppleWorks word processor data" like ARTICLE.1 ./apple
425 # skip some unknown basic binaries like RocketRnger.SHR
427 # skip "GPG symmetrically encrypted data" ./gnu
428 # skip "PGP symmetric key encrypted data" ./pgp
429 # openpgpdefs.h: fourth byte < 14 indicate cipher algorithm type
430 >>>4 ubyte >13 DOS executable (COM, 0x8C-variant)
431 # the remaining files should be DOS *.COM executables
432 # dosshell.COM 8cc0 2ea35f07 e85211 e88a11 b80058 cd
433 # hmload.COM 8cc8 8ec0 bbc02b 89dc 83c30f c1eb04 b4
434 # UNDELETE.COM 8cca 2e8916 6503 b430 cd21 8b 2e0200 8b
435 # BOOTFIX.COM 8cca 2e8916 9603 b430 cd21 8b 2e0200 8b
436 # RAWRITE3.COM 8cca 2e8916 d602 b430 cd21 8b 2e0200 8b
437 # SHARE.COM 8cca 2e8916 d602 b430 cd21 8b 2e0200 8b
438 # validchr.COM 8cca 2e8916 9603 b430 cd21 8b 2e028b1e
439 # devload.COM 8cca 8916ad01 b430 cd21 8b2e0200 892e
440 !:mime application/x-dosexec
443 # updated by Joerg Jenderek at Oct 2008
444 0 ulelong 0xffff10eb DR-DOS executable (COM)
445 # byte 0xeb conflicts with "sequent" magic leshort 0xn2eb
446 0 ubeshort&0xeb8d >0xeb00
447 # DR-DOS STACKER.COM SCREATE.SYS missed
450 >0 byte x DOS executable (COM)
451 >6 string SFX\ of\ LHarc \b, %s
452 >0x1FE leshort 0xAA55 \b, boot code
453 >85 string UPX \b, UPX compressed
454 >4 string \ $ARX \b, ARX self-extracting archive
455 >4 string \ $LHarc \b, LHarc self-extracting archive
456 >0x20e string SFX\ by\ LARC \b, LARC self-extracting archive
460 # allow forward jumps only
462 # that offset must be accessible
470 # that offset must be accessible
473 # negative offset, must not lead into PSP
475 # that offset must be accessible
479 # updated by Joerg Jenderek at Oct 2008,2015
480 # following line is too general
482 # skip 2 linux kernels like memtest.bin with "\xb8\xc0\x07\x8e" in ./linux
483 >0 string !\xb8\xc0\x07\x8e
484 # modified by Joerg Jenderek
485 # syslinux COM32 or COM32R executable
486 >>1 lelong&0xFFFFFFFe 0x21CD4CFe COM executable (32-bit COMBOOT
487 # http://www.syslinux.org/wiki/index.php/Comboot_API
488 # Since version 5.00 c32 modules switched from the COM32 object format to ELF
489 !:mime application/x-c32-comboot-syslinux-exec
491 # http://syslinux.zytor.com/comboot.php
492 # older syslinux version ( <4 )
493 # (32-bit COMBOOT) programs *.C32 contain 32-bit code and run in flat-memory 32-bit protected mode
494 # start with assembler instructions mov eax,21cd4cffh
495 >>>1 lelong 0x21CD4CFf \b)
496 # syslinux:doc/comboot.txt
497 # A COM32R program must start with the byte sequence B8 FE 4C CD 21 (mov
498 # eax,21cd4cfeh) as a magic number.
499 # syslinux version (4.x)
500 # "COM executable (COM32R)" or "Syslinux COM32 module" by TrID
501 >>>1 lelong 0x21CD4CFe \b, relocatable)
502 # remaining are DOS COM executables starting with assembler instruction MOV
503 # like FreeDOS BANNER*.COM FINDDISK.COM GIF2RAW.COM WINCHK.COM
504 # MS-DOS SYS.COM RESTART.COM
505 # SYSLINUX.COM (version 1.40 - 2.13)
506 # GFXBOOT.COM (version 3.75)
507 # COPYBS.COM POWEROFF.COM INT18.COM
508 >>1 default x COM executable for DOS
509 !:mime application/x-dosexec
510 #!:mime application/x-ms-dos-executable
511 #!:mime application/x-msdos-program
515 >4 string \x77\x02\xcd\x20\xb9
516 >>36 string UPX! FREE-DOS executable (COM), UPX compressed
517 252 string Must\ have\ DOS\ version DR-DOS executable (COM)
518 # added by Joerg Jenderek at Oct 2008
519 # GRR search is not working
520 #34 search/2 UPX! FREE-DOS executable (COM), UPX compressed
521 34 string UPX! FREE-DOS executable (COM), UPX compressed
522 35 string UPX! FREE-DOS executable (COM), UPX compressed
523 # GRR search is not working
524 #2 search/28 \xcd\x21 COM executable for MS-DOS
526 2 string \xcd\x21 COM executable for DOS
527 #DELTREE.cOM DELTREE2.cOM
528 4 string \xcd\x21 COM executable for DOS
529 #IFMEMDSK.cOM ASSIGN.cOM COMP.cOM
530 5 string \xcd\x21 COM executable for DOS
531 #DELTMP.COm HASFAT32.cOM
533 >0 byte !0xb8 COM executable for DOS
536 >5 string !\xcd\x21 COM executable for DOS
538 13 string \xcd\x21 COM executable for DOS
540 18 string \xcd\x21 COM executable for MS-DOS
542 23 string \xcd\x21 COM executable for MS-DOS
543 #LOADFIX.cOm LOADFIX.cOm
544 30 string \xcd\x21 COM executable for MS-DOS
546 70 string \xcd\x21 COM executable for DOS
547 # many compressed/converted COMs start with a copy loop instead of a jump
548 0x6 search/0xa \xfc\x57\xf3\xa5\xc3 COM executable for MS-DOS
549 0x6 search/0xa \xfc\x57\xf3\xa4\xc3 COM executable for DOS
550 >0x18 search/0x10 \x50\xa4\xff\xd5\x73 \b, aPack compressed
551 0x3c string W\ Collis\0\0 COM executable for MS-DOS, Compack compressed
552 # FIXME: missing diet .com compression
554 # miscellaneous formats
555 0 string/b LZ MS-DOS executable (built-in)
556 #0 byte 0xf0 MS-DOS program library data
560 # <stuartc@rd.bbc.co.uk> Stuart Cunningham
561 0 string/b \320\317\021\340\241\261\032\341AAFB\015\000OM\006\016\053\064\001\001\001\377 AAF legacy file using MS Structured Storage
562 >30 byte 9 (512B sectors)
563 >30 byte 12 (4kB sectors)
564 0 string/b \320\317\021\340\241\261\032\341\001\002\001\015\000\002\000\000\006\016\053\064\003\002\001\001 AAF file using MS Structured Storage
565 >30 byte 9 (512B sectors)
566 >30 byte 12 (4kB sectors)
568 # Popular applications
569 2080 string Microsoft\ Word\ 6.0\ Document %s
570 !:mime application/msword
571 2080 string Documento\ Microsoft\ Word\ 6 Spanish Microsoft Word 6 document data
572 !:mime application/msword
573 # Pawel Wiecek <coven@i17linuxb.ists.pwr.wroc.pl> (for polish Word)
574 2112 string MSWordDoc Microsoft Word document data
575 !:mime application/msword
577 0 belong 0x31be0000 Microsoft Word Document
578 !:mime application/msword
580 0 string/b PO^Q` Microsoft Word 6.0 Document
581 !:mime application/msword
584 >0 belong 0xfe320000 Microsoft Word for Macintosh 1.0
585 !:mime application/msword
587 >0 belong 0xfe340000 Microsoft Word for Macintosh 3.0
588 !:mime application/msword
590 >0 belong 0xfe37001c Microsoft Word for Macintosh 4.0
591 !:mime application/msword
593 >0 belong 0xfe370023 Microsoft Word for Macintosh 5.0
594 !:mime application/msword
597 0 string/b \333\245-\0\0\0 Microsoft Word 2.0 Document
598 !:mime application/msword
600 # Note: seems already recognized as "OLE 2 Compound Document" in ./ole2compounddocs
601 #512 string/b \354\245\301 Microsoft Word Document
602 #!:mime application/msword
605 0 string/b \xDB\xA5\x2D\x00 Microsoft WinWord 2.0 Document
606 !:mime application/msword
608 2080 string Microsoft\ Excel\ 5.0\ Worksheet %s
609 !:mime application/vnd.ms-excel
611 0 string/b \xDB\xA5\x2D\x00 Microsoft WinWord 2.0 Document
612 !:mime application/msword
614 2080 string Foglio\ di\ lavoro\ Microsoft\ Exce %s
615 !:mime application/vnd.ms-excel
617 # Pawel Wiecek <coven@i17linuxb.ists.pwr.wroc.pl> (for polish Excel)
618 2114 string Biff5 Microsoft Excel 5.0 Worksheet
619 !:mime application/vnd.ms-excel
621 2121 string Biff5 Microsoft Excel 5.0 Worksheet
622 !:mime application/vnd.ms-excel
623 0 string/b \x09\x04\x06\x00\x00\x00\x10\x00 Microsoft Excel Worksheet
624 !:mime application/vnd.ms-excel
626 # Update: Joerg Jenderek
627 # URL: https://en.wikipedia.org/wiki/Lotus_1-2-3
628 # Reference: http://www.aboutvb.de/bas/formate/pdf/wk3.pdf
629 # Note: newer Lotus versions >2 use longer BOF record
630 # record type (BeginningOfFile=0000h) + length (001Ah)
632 # reserved should be 0h but 8c0dh for TUTMAC.WK3, 5h for SAMPADNS.WK3, 1h for a_readme.wk3, 1eh for K&G86.WK3
633 #>18 uleshort&0x73E0 0
634 # Lotus Multi Byte Character Set (LMBCS=1-31)
636 >>20 ubyte <32 Lotus 1-2-3
637 #!:mime application/x-123
638 !:mime application/vnd.lotus-1-2-3
640 # (version 5.26) labeled the entry as "Lotus 1-2-3 wk3 document data"
641 >>>4 uleshort 0x1000 WorKsheet, version 3
643 # (version 5.26) labeled the entry as "Lotus 1-2-3 wk4 document data"
644 >>>4 uleshort 0x1002 WorKsheet, version 4
645 # also worksheet template 4 (.wt4)
647 # no example or documentation for wk5
648 #>>4 uleshort 0x???? WorKsheet, version 4
650 # only MacrotoScript.123 example
651 >>>4 uleshort 0x1003 WorKsheet, version 97
652 # also worksheet template Smartmaster (.12M)?
654 # only Set_Y2K.123 example
655 >>>4 uleshort 0x1005 WorKsheet, version 9.8 Millennium
657 # no example for this version
658 >>>4 uleshort 0x8001 FoRMatting data
660 # (version 5.26) labeled the entry as "Lotus 1-2-3 fm3 or fmb document data"
661 # TrID labeles the entry as "Formatting Data for Lotus 1-2-3 worksheet"
662 >>>4 uleshort 0x8007 ForMatting data, version 3
664 >>>4 default x unknown
665 # file revision sub code 0004h for worksheets
666 >>>>6 uleshort =0x0004 worksheet
668 >>>>6 uleshort !0x0004 formatting data
670 # main revision number
671 >>>>4 uleshort x \b, revision 0x%x
672 >>>6 uleshort =0x0004 \b, cell range
673 # active cellcoord range (start row, page,column ; end row, page, column)
674 # start values normally 0~1st sheet A1
676 >>>>>10 ubyte >0 \b%d*
677 >>>>>8 uleshort x \b%d,
678 >>>>>11 ubyte x \b%d-
680 >>>>14 ubyte >0 \b%d*
681 # end raw, column normally not 0
682 >>>>12 uleshort x \b%d,
684 # Lotus Multi Byte Character Set (1~cp850,2~cp851,...,16~japan,...,31~??)
685 >>>>20 ubyte >1 \b, character set 0x%x
687 >>>>21 ubyte x \b, flags 0x%x
688 >>>6 uleshort !0x0004
689 # record type (FONTNAME=00AEh)
690 >>>>30 search/29 \0\xAE
691 # variable length m (2) + entries (1) + ?? (1) + LCMBS string (n)
692 >>>>>&4 string >\0 \b, 1st font "%s"
694 # Update: Joerg Jenderek
695 # URL: http://fileformats.archiveteam.org/wiki/Lotus_1-2-3
696 # Reference: http://www.schnarff.com/file-formats/lotus-1-2-3/WSFF2.TXT
697 # Note: Used by both old Lotus 1-2-3 and Lotus Symphony (DOS) til version 2.x
698 # record type (BeginningOfFile=0000h) + length (0002h)
700 # GRR: line above is too general as it catches also MS Windows CURsor
701 # to display MS Windows cursor (strength=70) before Lotus 1-2-3 (strength=70-1)
703 # skip Windows cursors with image height <256 and keep Lotus with low opcode 0001-0083h
705 # skip Windows cursors with image width 256 and keep Lotus with positiv opcode
707 # !:mime application/x-123
708 !:mime application/vnd.lotus-1-2-3
710 # revision number (0404h = 123 1A, 0405h = Lotus Symphony , 0406h = 123 2.x wk1 , 8006h = fmt , ...)
711 # undocumented; (version 5.26) labeled the configurations as "Lotus 1-2-3"
712 >>>4 uleshort 0x0007 1-2-3 CoNFiguration, version 2.x (PGRAPH.CNF)
714 >>>4 uleshort 0x0C05 1-2-3 CoNFiguration, version 2.4J
716 >>>4 uleshort 0x0801 1-2-3 CoNFiguration, version 1-2.1
718 >>>4 uleshort 0x0802 Symphony CoNFiguration
720 >>>4 uleshort 0x0804 1-2-3 CoNFiguration, version 2.2
722 >>>4 uleshort 0x080A 1-2-3 CoNFiguration, version 2.3-2.4
724 >>>4 uleshort 0x1402 1-2-3 CoNFiguration, version 3.x
726 >>>4 uleshort 0x1450 1-2-3 CoNFiguration, version 4.x
728 # (version 5.26) labeled the entry as "Lotus 123"
729 # TrID labeles the entry as "Lotus 123 Worksheet (generic)"
730 >>>4 uleshort 0x0404 1-2-3 WorKSheet, version 1
731 # extension "wks" also for Microsoft Works document
733 # (version 5.26) labeled the entry as "Lotus 123"
734 # TrID labeles the entry as "Lotus 123 Worksheet (generic)"
735 >>>4 uleshort 0x0405 Symphony WoRksheet, version 1.0
737 # (version 5.26) labeled the entry as "Lotus 1-2-3 wk1 document data"
738 # TrID labeles the entry as "Lotus 123 Worksheet (V2)"
739 >>>4 uleshort 0x0406 1-2-3/Symphony worksheet, version 2
742 # no example for this japan version
743 >>>4 uleshort 0x0600 1-2-3 WorKsheet, version 1.xJ
745 # no example or documentation for wk2
746 #>>>4 uleshort 0x???? 1-2-3 WorKsheet, version 2
748 # undocumented japan version
749 >>>4 uleshort 0x0602 1-2-3 worksheet, version 2.4J
751 # (version 5.26) labeled the entry as "Lotus 1-2-3 fmt document data"
752 >>>4 uleshort 0x8006 1-2-3 ForMaTting data, version 2.x
753 # japan version 2.4J (fj3)
755 # no example for this version
756 >>>4 uleshort 0x8007 1-2-3 FoRMatting data, version 2.0
758 # (version 5.26) labeled the entry as "Lotus 1-2-3"
759 >>>4 default x unknown worksheet or configuration
761 >>>>4 uleshort x \b, revision 0x%x
762 # 2nd record for most worksheets describes cells range
764 # 3nd record for most japan worksheets describes cells range
765 >>>(8.s+10) use lotus-cells
766 # check and then display Lotus worksheet cells range
768 # look for type (RANGE=0006h) + length (0008h) at record begin
769 >0 ubelong 0x06000800 \b, cell range
770 # cell range (start column, row, end column, row) start values normally 0,0~A1 cell
772 >>>4 uleshort x \b%d,
773 >>>6 uleshort x \b%d-
778 0 string/b WordPro\0 Lotus WordPro
779 !:mime application/vnd.lotus-wordpro
780 0 string/b WordPro\r\373 Lotus WordPro
781 !:mime application/vnd.lotus-wordpro
784 # Summary: Script used by InstallScield to uninstall applications
786 # Submitted by: unknown
787 # Modified by (1): Abel Cheung <abelcheung@gmail.com> (replace useless entry)
788 0 string \x71\xa8\x00\x00\x01\x02
789 >12 string Stirling\ Technologies, InstallShield Uninstall Script
792 #0 string Nullsoft\ AVS\ Preset\ \060\056\061\032 A plug in for Winamp ms-windows Freeware media player
793 0 string/b Nullsoft\ AVS\ Preset\ Winamp plug in
795 # Windows Metafile .WMF
796 0 string/b \327\315\306\232 Windows metafile
799 0 string/b \002\000\011\000 Windows metafile
802 0 string/b \001\000\011\000 Windows metafile
806 #tz3 files whatever that is (MS Works files)
807 0 string/b \003\001\001\004\070\001\000\000 tz3 ms-works file
808 0 string/b \003\002\001\004\070\001\000\000 tz3 ms-works file
809 0 string/b \003\003\001\004\070\001\000\000 tz3 ms-works file
812 #0 string \211\000\077\003\005\000\063\237\127 065 to \027\266\151\064\005\045\101\233\021\002 PGP sig
813 0 string \211\000\077\003\005\000\063\237\127\065\027\266\151\064\005\045\101\233\021\002 PGP sig
814 0 string \211\000\077\003\005\000\063\237\127\066\027\266\151\064\005\045\101\233\021\002 PGP sig
815 0 string \211\000\077\003\005\000\063\237\127\067\027\266\151\064\005\045\101\233\021\002 PGP sig
816 0 string \211\000\077\003\005\000\063\237\127\070\027\266\151\064\005\045\101\233\021\002 PGP sig
817 0 string \211\000\077\003\005\000\063\237\127\071\027\266\151\064\005\045\101\233\021\002 PGP sig
818 0 string \211\000\225\003\005\000\062\122\207\304\100\345\042 PGP sig
820 # windows zips files .dmf
821 0 string/b MDIF\032\000\010\000\000\000\372\046\100\175\001\000\001\036\001\000 MS Windows special zipped file
825 0 string/b \102\101\050\000\000\000\056\000\000\000\000\000\000\000 Icon for MS Windows
828 # Update: Joerg Jenderek
829 # URL: https://en.wikipedia.org/wiki/CUR_(file_format)
830 # Note: similar to Windows CURsor. container for BMP (only DIB part) or PNG
838 # displays number of icons and information for icon or cursor
840 # skip some Lotus 1-2-3 worksheets, CYCLE.PIC and keep Windows cursors with
841 # 1st data offset = dir header size + n * dir entry size = 6 + n * 10h = ?6h
842 >18 ulelong &0x00000006
843 # skip remaining worksheets, because valid only for DIB image (40) or PNG image (\x89PNG)
844 >>(18.l) ulelong x MS Windows
845 >>>0 ubelong 0x00000100 icon resource
846 #!:mime image/vnd.microsoft.icon
849 >>>>4 uleshort x - %d icon
851 >>>>4 uleshort >1 \bs
853 >>>>0x06 use ico-entry
856 >>>>>0x16 use ico-entry
857 >>>0 ubelong 0x00000200 cursor resource
859 !:mime image/x-win-bitmap
861 >>>>4 uleshort x - %d icon
862 >>>>4 uleshort >1 \bs
864 >>>>0x06 use cur-entry
865 #>>>>0x16 use cur-entry
866 # display information of one cursor entry
869 >4 uleshort x \b, hotspot @%dx
871 # display information of one icon entry
874 # normally 0 1 but also found 14
875 >4 uleshort >1 \b, %d planes
876 # normally 0 1 but also found some 3, 4, some 6, 8, 24, many 32, two 256
877 >6 uleshort >1 \b, %d bits/pixel
878 # display shared information of cursor or icon entry
884 # number of colors in palette
885 >2 ubyte !0 \b, %d colors
887 #>3 ubyte x \b, reserved %x
888 #>8 ulelong x \b, image size %d
889 # offset of PNG or DIB image
890 #>12 ulelong x \b, offset 0x%x
891 # PNG header (\x89PNG)
892 >(12.l) ubelong =0x89504e47
893 >>&-4 indirect x \b with
895 >(12.l) ubelong !0x89504e47
898 # Windows non-animated cursors
899 # Update: Joerg Jenderek
900 # URL: https://en.wikipedia.org/wiki/CUR_(file_format)
901 # Note: similar to Windows ICOn. container for BMP ( only DIB part)
902 # GRR: line below is too general as it catches also Lotus 1-2-3 files
910 0 string/b PK\010\010BGI Borland font
912 # then there is a copyright notice
916 0 string/b pk\010\010BGI Borland device
918 # then there is a copyright notice
921 # Windows Recycle Bin record file (named INFO2)
922 # By Abel Cheung (abelcheung AT gmail dot com)
923 # Version 4 always has 280 bytes (0x118) per record, version 5 has 800 bytes
924 # Since Vista uses another structure, INFO2 structure probably won't change
925 # anymore. Detailed analysis in:
926 # http://www.cybersecurityinstitute.biz/downloads/INFO2.pdf
928 >12 lelong 0x00000118 Windows Recycle Bin INFO2 file (Win98 or below)
931 >12 lelong 0x00000320 Windows Recycle Bin INFO2 file (Win2k - WinXP)
933 # From Doug Lee via a FreeBSD pr
934 9 string GERBILDOC First Choice document
935 9 string GERBILDB First Choice database
936 9 string GERBILCLIP First Choice database
937 0 string GERBIL First Choice device file
938 9 string RABBITGRAPH RabbitGraph file
939 0 string DCU1 Borland Delphi .DCU file
940 0 string =!<spell> MKS Spell hash list (old format)
941 0 string =!<spell2> MKS Spell hash list
943 #0 string AH Halo(TM) bitmapped font file
944 0 lelong 0x08086b70 TurboC BGI file
945 0 lelong 0x08084b50 TurboC Font file
947 # Debian#712046: The magic below identifies "Delphi compiled form data".
948 # An additional source of information is available at:
949 # http://www.woodmann.com/fravia/dafix_t1.htm
951 >4 pstring >\0 Delphi compiled form '%s'
953 # tests for DBase files moved, updated and merged to database
955 0 string PMCC Windows 3.x .GRP file
956 1 string RDC-meg MegaDots
957 >8 byte >0x2F version %c
958 >9 byte >0x2F \b.%c file
960 >4 lelong 0x00021401 Windows shortcut file
962 # .PIF files added by Joerg Jenderek from http://smsoft.ru/en/pifdoc.htm
963 # only for windows versions equal or greater 3.0
964 0x171 string MICROSOFT\ PIFEX\0 Windows Program Information File
965 !:mime application/x-dosexec
966 #>2 string >\0 \b, Title:%.30s
967 >0x24 string >\0 \b for %.63s
968 >0x65 string >\0 \b, directory=%.64s
969 >0xA5 string >\0 \b, parameters=%.64s
970 #>0x181 leshort x \b, offset %x
971 #>0x183 leshort x \b, offsetdata %x
972 #>0x185 leshort x \b, section length %x
973 >0x187 search/0xB55 WINDOWS\ VMM\ 4.0\0
975 >>>&-1 string <PIFMGR.DLL \b, icon=%s
976 #>>>&-1 string PIFMGR.DLL \b, icon=%s
977 >>>&-1 string >PIFMGR.DLL \b, icon=%s
979 >>>&-1 string <Terminal \b, font=%.32s
980 #>>>&-1 string =Terminal \b, font=%.32s
981 >>>&-1 string >Terminal \b, font=%.32s
983 >>>&-1 string <Lucida\ Console \b, TrueTypeFont=%.32s
984 #>>>&-1 string =Lucida\ Console \b, TrueTypeFont=%.32s
985 >>>&-1 string >Lucida\ Console \b, TrueTypeFont=%.32s
986 #>0x187 search/0xB55 WINDOWS\ 286\ 3.0\0 \b, Windows 3.X standard mode-style
987 #>0x187 search/0xB55 WINDOWS\ 386\ 3.0\0 \b, Windows 3.X enhanced mode-style
988 >0x187 search/0xB55 WINDOWS\ NT\ \ 3.1\0 \b, Windows NT-style
989 #>0x187 search/0xB55 WINDOWS\ NT\ \ 4.0\0 \b, Windows NT-style
990 >0x187 search/0xB55 CONFIG\ \ SYS\ 4.0\0 \b +CONFIG.SYS
991 #>>&06 string x \b:%s
992 >0x187 search/0xB55 AUTOEXECBAT\ 4.0\0 \b +AUTOEXEC.BAT
993 #>>&06 string x \b:%s
995 # DOS EPS Binary File Header
996 # From: Ed Sznyter <ews@Black.Market.NET>
997 0 belong 0xC5D0D3C6 DOS EPS Binary File
999 >4 long >0 Postscript starts at byte %d
1000 >>8 long >0 length %d
1001 >>>12 long >0 Metafile starts at byte %d
1002 >>>>16 long >0 length %d
1003 >>>20 long >0 TIFF starts at byte %d
1004 >>>>24 long >0 length %d
1006 # TNEF magic From "Joomy" <joomy@se-ed.net>
1007 # Microsoft Outlook's Transport Neutral Encapsulation Format (TNEF)
1008 0 lelong 0x223e9f78 TNEF
1009 !:mime application/vnd.ms-tnef
1011 # Norton Guide (.NG , .HLP) files added by Joerg Jenderek from source NG2HTML.C
1012 # of http://www.davep.org/norton-guides/ng2h-105.tgz
1013 # http://en.wikipedia.org/wiki/Norton_Guides
1015 # only value 0x100 found at offset 2
1016 >2 ulelong 0x00000100 Norton Guide
1018 >>8 string >\0 "%-.40s"
1019 #>>6 uleshort x \b, MenuCount=%u
1021 >>48 string >\0 \b, %-.66s
1022 >>114 string >\0 %-.66s
1024 # 4DOS help (.HLP) files added by Joerg Jenderek from source TPHELP.PAS
1025 # of http://www.4dos.info/
1026 # pointer,HelpID[8]=4DHnnnmm
1027 0 ulelong 0x48443408 4DOS help file
1028 >4 string x \b, version %-4.4s
1030 # old binary Microsoft (.HLP) files added by Joerg Jenderek from http://file-extension.net/seeker/file_extension_hlp
1031 0 ulequad 0x3a000000024e4c MS Advisor help file
1033 # HtmlHelp files (.chm)
1034 0 string/b ITSF\003\000\000\000\x60\000\000\000 MS Windows HtmlHelp Data
1036 # GFA-BASIC (Wolfram Kleff)
1037 2 string/b GFA-BASIC3 GFA-BASIC 3 data
1039 #------------------------------------------------------------------------------
1040 # From Stuart Caie <kyzer@4u.net> (developer of cabextract)
1041 # Update: Joerg Jenderek
1042 # URL: https://en.wikipedia.org/wiki/Cabinet_(file_format)
1043 # Reference: https://msdn.microsoft.com/en-us/library/bb267310.aspx
1044 # Note: verified by `7z l *.cab`
1045 # Microsoft Cabinet files
1046 0 string/b MSCF\0\0\0\0 Microsoft Cabinet archive data
1048 # https://support.microsoft.com/en-us/help/973559/frequently-asked-questions-about-the-microsoft-support-diagnostic-tool
1049 # CAB with *.{diagcfg,diagpkg} is used by Microsoft Support Diagnostic Tool MSDT.EXE
1050 # because some archive does not have *.diag* as 1st or 2nd archive member like
1051 # O15CTRRemove.diagcab or AzureStorageAnalyticsLogs_global.DiagCab
1052 # brute looking after header for filenames with diagcfg or diagpkg extension in CFFILE section
1053 >0x2c search/980/c .diag \b, Diagnostic
1054 !:mime application/vnd.ms-cab-compressed
1056 # http://fileformats.archiveteam.org/wiki/PUZ
1057 # Microsoft Publisher version about 2003 has a "Pack and Go" feature that
1058 # bundles a Publisher document *PNG.pub with all links into a CAB
1059 >0x2c search/300/c png.pub\0 \b, Publisher Packed and Go
1060 !:mime application/vnd.ms-cab-compressed
1062 # ppz variant with Microsoft PowerPoint Viewer ppview32.exe to play PowerPoint presentation
1063 >0x2c search/17/c ppview32.exe\0 \b, PowerPoint Viewer Packed and Go
1064 !:mime application/vnd.ms-powerpoint
1065 #!:mime application/mspowerpoint
1067 # http://www.incredimail.com/
1068 # IncrediMail CAB contains an initialisation file "content.ini" like in im2.ims
1069 >0x2c search/3369/c content.ini\0 \b, IncrediMail
1070 !:mime application/x-incredimail
1071 # member Flavor.htm implies IncrediMail ecard like in tell_a_friend.imf
1072 >>0x2c search/83/c Flavor.htm\0 ecard
1074 # member Macromedia Flash data *.swf implies IncrediMail skin like in im2.ims
1075 >>0x2c search/211/c .swf\0 skin
1077 # member anim.im3 implies IncrediMail animation like in letter_fold.ima
1078 >>0x2c search/92/c anim.im3\0 animation
1080 # other IncrediMail cab archive
1082 >>>0x2c search/116/c thumb ecard, image, notifier or skin
1083 !:ext imf/imi/imn/ims
1084 # http://file-extension.net/seeker/file_extension_ime
1085 >>>0x2c default x emoticons or sound
1087 # no Diagnostic and IncrediMail
1089 # look for 1st member name
1091 # https://en.wikipedia.org/wiki/SNP_file_format
1092 >>>&-1 string/c _accrpt_.snp \b, Access report snapshot
1093 !:mime application/msaccess
1095 # https://www.cabextract.org.uk/wince_cab_format/
1096 # extension of DOS 8+3 name with ".000" of 1st archive member name implies Windows CE installer
1097 >>>&7 string =.000 \b, WinCE install
1098 !:mime application/vnd.ms-cab-compressed
1101 # http://support.microsoft.com/kb/934307/en-US
1102 # All inspected MSU contain a file with name WSUSSCAN.cab
1103 # that is called "Windows Update meta data" by Microsoft
1104 >>>&-1 string/c wsusscan.cab \b, Microsoft Standalone Update
1105 !:mime application/vnd.ms-cab-compressed
1108 # look at point charcter of 1st archive member name for file name extension
1109 >>>>&-1 search/255 .
1110 # http://www.pptfaq.com/FAQ00164_What_is_a_PPZ_file-.htm
1111 # PPZ were created using Pack & Go feature of PowerPoint versions 97 - 2002
1112 # packs optional files, a PowerPoint presentation *.ppt with optional PLAYLIST.LST to CAB
1113 >>>>>&0 string/c ppt\0 \b, PowerPoint Packed and Go
1114 !:mime application/vnd.ms-powerpoint
1115 #!:mime application/mspowerpoint
1117 # https://msdn.microsoft.com/en-us/library/windows/desktop/bb773190(v=vs.85).aspx
1118 # first member *.theme implies Windows 7 Theme Pack like in CommunityShowcaseAqua3.themepack
1119 # or Windows 8 Desktop Theme Pack like in PanoramicGlaciers.deskthemepack
1120 >>>>>&0 string/c theme \b, Windows
1121 !:mime application/x-windows-themepack
1122 # http://www.drewkeller.com/content/using-theme-both-windows-7-and-windows-8
1123 # 1st member Panoramic.theme or Panoramas.theme implies Windows 8-10 Theme Pack
1124 # with MTSM=RJSPBS in [MasterThemeSelector] inside *.theme
1125 >>>>>>(16.l+16) string =Panoram 8
1127 >>>>>>(16.l+16) string !Panoram 7 or 8
1128 !:ext themepack/deskthemepack
1129 >>>>>>(16.l+16) ubyte x Theme Pack
1131 # look for null terminator of 1st member name
1132 >>>>>>&0 search/255 \0
1133 # 2nd member name WSUSSCAN.cab like in Microsoft-Windows-MediaFeaturePack-OOB-Package.msu
1134 >>>>>>>&16 string/c wsusscan.cab \b, Microsoft Standalone Update
1135 !:mime application/vnd.ms-cab-compressed
1137 >>>>>>>&16 default x
1138 # archive with more then one file need some output in version 5.32 to avoid error message like
1139 # Magdir/msdos, 1138: Warning: Current entry does not yet have a description for adding a MIME type
1140 # Magdir/msdos, 1139: Warning: Current entry does not yet have a description for adding a EXTENSION type
1141 # file: could not find any valid magic files!
1142 >>>>>>>>28 uleshort >1 \b, many
1143 !:mime application/vnd.ms-cab-compressed
1145 # remaining archives with just one file
1146 >>>>>>>>28 uleshort =1
1147 # neither extra bytes nor cab chain implies Windows 2000,XP setup files in directory i386
1148 >>>>>>>>>30 uleshort =0x0000 \b, Windows 2000/XP setup
1149 # cut of last char of source extension and add underscore to generate extension
1150 # TERMCAP._ ... FXSCOUNT.H_ ... L3CODECA.AC_ ... NPDRMV2.ZI_
1151 !:mime application/vnd.ms-cab-compressed
1153 # archive need some output like "single" in version 5.32 to avoid error messages
1154 >>>>>>>>>30 uleshort !0x0000 \b, single
1155 !:mime application/vnd.ms-cab-compressed
1157 # TODO: additional extensions like
1158 # .xsn InfoPath Dynamic Form
1159 # .xtp InfoPath Template Part
1160 # .lvf Logitech Video Effects Face Accessory
1161 >8 ulelong x \b, %u bytes
1162 >28 uleshort 1 \b, 1 file
1163 >28 uleshort >1 \b, %u files
1164 # Reserved fields, set to zero
1165 #>4 belong !0 \b, reserved1 %x
1166 #>12 belong !0 \b, reserved2 %x
1167 # offset of the first CFFILE entry coffFiles: minimal 2Ch
1168 >16 ulelong x \b, at 0x%x
1169 >(16.l) use cab-file
1170 # at least also 2nd member
1174 # second member info
1176 #>20 belong !0 \b, reserved %x
1177 # Cabinet file format version. Currently, versionMajor = 1 and versionMinor = 3
1178 >24 ubeshort !0x0301 \b version 0x%x
1179 # number of CFFOLDER entries
1180 >26 uleshort >1 \b, %u cffolders
1181 # cabinet file option indicators 1~PREVIOUS, 2~NEXT, 4~reserved fields
1182 # only found for flags 0 1 2 3 4 not 7
1183 >30 uleshort >0 \b, flags 0x%x
1184 # Cabinet files have a 16-bit cabinet setID field that is designed for application use.
1185 # default is zero, however, the -i option of cabarc can be used to set this field
1186 >32 uleshort >0 \b, ID %u
1187 # iCabinet is number of this cabinet file in a set, where 0 for the first cabinet
1188 #>34 uleshort x \b, iCabinet %u
1189 # add one for display because humans start numbering by 1 and also fit to name of disk szDisk*
1190 >34 uleshort+1 x \b, number %u
1191 >30 uleshort &0x0004 \b, extra bytes
1192 # cbCFHeader optional size of per-cabinet reserved area 14h 1800h
1193 >>36 uleshort >0 %u in head
1194 # cbCFFolder is optional size of per-folder reserved area
1195 >>38 ubyte >0 %u in folder
1196 # cbCFData is optional size of per-datablock reserved area
1197 >>39 ubyte >0 %u in data block
1198 # optional per-cabinet reserved area abReserve[cbCFHeader]
1200 # 1st CFFOLDER after reserved area in header
1201 >>>(36.s+40) use cab-folder
1202 # no reserved area in header
1203 >30 uleshort ^0x0004
1204 # no previous and next cab archive
1205 >>30 uleshort =0x0000
1206 >>>36 use cab-folder
1207 # only previous cab archive
1208 >>30 uleshort =0x0001 \b, previous
1209 >>>36 use cab-anchor
1210 # only next cab archive
1211 >>30 uleshort =0x0002 \b, next
1212 >>>36 use cab-anchor
1213 # previous+next cab archive
1214 # can not use sub routine cab-anchor to display previous and next cabinet together
1215 #>>>36 use cab-anchor
1216 #>>>>&0 use cab-anchor
1217 >>30 uleshort =0x0003 \b, previous
1219 # optional name of previous disk szDisk*
1220 >>>>&1 string x disk %s
1221 >>>>>&1 string x \b, next %s
1222 # optional name of previous disk szDisk*
1223 >>>>>>&1 string x disk %s
1224 >>>>>>>&1 use cab-folder
1225 # display filename and disk name of previous or next cabinet
1227 # optional name of previous/next cabinet file szCabinet*[255]
1229 # optional name of previous/next disk szDisk*[255]
1230 >>&1 string x disk %s
1231 # display folder structure CFFOLDER information like compression of cabinet
1233 # offset of the CFDATA block in this folder
1234 #>0 ulelong x \b, coffCabStart 0x%x
1235 # number of CFDATA blocks in folder
1236 >4 uleshort x \b, %u datablock
1239 # compression typeCompress: 0~None 1~MSZIP 0x1503~LZX:21 0x1003~LZX:16 0x0f03~LZX:15
1240 >6 uleshort x \b, 0x%x compression
1241 # optional per-folder reserved area
1242 #>8 ubequad x \b, abReserve 0x%llx
1243 # display member structure CFFILE information like member name of cabinet
1245 # cbFile is uncompressed size of file in bytes
1246 #>0 ulelong x \b, cbFile %u
1247 # uoffFolderStart is uncompressed offset of file in folder
1248 #>4 ulelong >0 \b, uoffFolderStart 0x%x
1249 # iFolder is index into the CFFOLDER area. 0 indicates first folder in cabinet
1250 # define ifoldCONTINUED_FROM_PREV (0xFFFD)
1251 # define ifoldCONTINUED_TO_NEXT (0xFFFE)
1252 # define ifoldCONTINUED_PREV_AND_NEXT (0xFFFF)
1253 >8 uleshort >0 \b, iFolder 0x%x
1254 # date stamp for file
1255 #>10 uleshort x \b, date 0x%x
1256 # time stamp for file
1257 #>12 uleshort x \b, time 0x%x
1258 # attribs is attribute flags for file
1259 # define _A_RDONLY (0x01) file is read-only
1260 # define _A_HIDDEN (0x02) file is hidden
1261 # define _A_SYSTEM (0x04) file is a system file
1262 # define _A_ARCH (0x20) file modified since last backup
1263 # example http://sebastien.kirche.free.fr/pebuilder_plugins/depends.cab
1264 # define _A_EXEC (0x40) run after extraction
1265 # define _A_NAME_IS_UTF (0x80) szName[] contains UTF
1266 # define UNKNOWN (0x0100) undocumented or accident
1267 #>14 uleshort x \b, attribs 0x%x
1269 >>14 uleshort &0x0001 \bR
1270 >>14 uleshort &0x0002 \bH
1271 >>14 uleshort &0x0004 \bS
1272 >>14 uleshort &0x0020 \bA
1273 >>14 uleshort &0x0040 \bX
1274 >>14 uleshort &0x0080 \bUtf
1275 # unknown 0x0100 flag found on one XP_CD:\I386\DRIVER.CAB
1276 >>14 uleshort &0x0100 \b?
1277 # szName is name of archive member
1279 # next archive member name if more files
1280 #>>&17 string >\0 \b, NEXT NAME %-.50s
1282 # InstallShield Cabinet files
1283 0 string/b ISc( InstallShield Cabinet archive data
1284 >5 byte&0xf0 =0x60 version 6,
1285 >5 byte&0xf0 !0x60 version 4/5,
1286 >(12.l+40) lelong x %u files
1288 # Windows CE package files
1289 0 string/b MSCE\0\0\0\0 Microsoft WinCE install header
1290 >20 lelong 0 \b, architecture-independent
1291 >20 lelong 103 \b, Hitachi SH3
1292 >20 lelong 104 \b, Hitachi SH4
1293 >20 lelong 0xA11 \b, StrongARM
1294 >20 lelong 4000 \b, MIPS R4000
1295 >20 lelong 10003 \b, Hitachi SH3
1296 >20 lelong 10004 \b, Hitachi SH3E
1297 >20 lelong 10005 \b, Hitachi SH4
1298 >20 lelong 70001 \b, ARM 7TDMI
1299 >52 leshort 1 \b, 1 file
1300 >52 leshort >1 \b, %u files
1301 >56 leshort 1 \b, 1 registry entry
1302 >56 leshort >1 \b, %u registry entries
1305 # Windows Enhanced Metafile (EMF)
1306 # See msdn.microsoft.com/archive/en-us/dnargdi/html/msdn_enhmeta.asp
1307 # for further information.
1309 >40 string \ EMF Windows Enhanced Metafile (EMF) image data
1310 >>44 ulelong x version 0x%x
1313 0 string/b \224\246\056 Microsoft Word Document
1314 !:mime application/msword
1316 512 string R\0o\0o\0t\0\ \0E\0n\0t\0r\0y Microsoft Word Document
1317 !:mime application/msword
1319 # From: "Nelson A. de Oliveira" <naoliv@gmail.com>
1320 # Magic type for Dell's BIOS .hdr files
1323 >23 string Dell %s system BIOS
1325 >>48 byte x version %d.
1329 >>48 string x version %.3s
1331 # Type: Microsoft Document Imaging Format (.mdi)
1332 # URL: http://en.wikipedia.org/wiki/Microsoft_Document_Imaging_Format
1333 # From: Daniele Sempione <scrows@oziosi.org>
1335 #0 short 0x5045 Microsoft Document Imaging Format
1337 # MS eBook format (.lit)
1338 0 string/b ITOLITLS Microsoft Reader eBook Data
1339 >8 lelong x \b, version %u
1340 !:mime application/x-ms-reader
1342 # Windows CE Binary Image Data Format
1343 # From: Dr. Jesus <j@hug.gs>
1344 0 string/b B000FF\n Windows Embedded CE binary image
1346 # Windows Imaging (WIM) Image
1347 0 string/b MSWIM\000\000\000 Windows imaging (WIM) image
1348 0 string/b WLPWM\000\000\000 Windows imaging (WIM) image, wimlib pipable format
1350 # The second byte of these signatures is a file version; I don't know what,
1351 # if anything, produced files with version numbers 0-2.
1352 # From: John Elliott <johne@seasip.demon.co.uk>
1353 0 string \xfc\x03\x00 Mallard BASIC program data (v1.11)
1354 0 string \xfc\x04\x00 Mallard BASIC program data (v1.29+)
1355 0 string \xfc\x03\x01 Mallard BASIC protected program data (v1.11)
1356 0 string \xfc\x04\x01 Mallard BASIC protected program data (v1.29+)
1358 0 string MIOPEN Mallard BASIC Jetsam data
1359 0 string Jetsam0 Mallard BASIC Jetsam index data
1361 # DOS backup 2.0 to 3.2
1365 # plausibility check for date
1369 # actually 121 nul bytes
1370 >>>0x7 string \0\0\0\0\0\0\0\0
1371 >>>>0x1 ubyte x DOS 2.0 backup id file, sequence %d
1373 >>>>0x0 ubyte 0xff \b, last disk
1377 # skip some AppleWorks word like Tomahawk.Awp, WIN98SE-DE.vhd
1378 # by looking for trailing nul of maximal file name string
1380 # test for flag byte: FFh~complete file, 00h~split file
1381 # FFh -127 = -1 -127 = -128
1382 # 00h -127 = 0 -127 = -127
1384 # plausibility check for file name length
1386 # looking for terminating nul of file name string
1387 >>>(0x53.b+4) ubyte 0
1388 # looking if last char of string is valid DOS file name
1389 >>>>(0x53.b+3) ubyte >0x1F
1390 # actually 44 nul bytes
1391 # but sometimes garbage according to Ralf Quint. So can not be used as test
1392 #>0x54 string \0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0
1393 # first char of full file name is DOS (5Ch) or UNIX (2Fh) path separator
1394 # only DOS variant found. UNIX variant according to V32SLASH.TXT in archive PD0315.EXE
1395 >>>>>5 ubyte&0x8C 0x0C
1396 # ./msdos (version 5.30) labeled the entry as
1397 # "DOS 2.0 backed up file %s, split file, sequence %d" or
1398 # "DOS 2.0 backed up file %s, complete file"
1399 >>>>>>0 ubyte x DOS 2.0-3.2 backed up
1400 #>>>>>>0 ubyte 0xff complete
1402 >>>>>>>1 uleshort x sequence %d of
1403 # full file name with path but without drive letter and colon stored from 0x05 til 0x52
1404 >>>>>>0x5 string x file %s
1405 # backup name is original filename
1407 # magic/Magdir/msdos, 1169: Warning: EXTENSION type ` *' has bad char '*'
1408 # file: line 1169: Bad magic entry ' *'
1409 # after header original file content
1410 >>>>>>128 indirect x \b;
1413 # DOS backup 3.3 to 5.x
1416 0 string \x8bBACKUP\x20
1417 # actually 128 nul bytes
1418 >0xa string \0\0\0\0\0\0\0\0
1419 >>0x9 ubyte x DOS 3.3 backup control file, sequence %d
1420 >>0x8a ubyte 0xff \b, last disk
1422 # NB: The BACKUP.nnn files consist of the files backed up,