Eagle Eye CD-ROM BBS Network

home *** CD-ROM | disk | FTP | other *** search

/ Eagle Eye CD-ROM BBS Network / eagle-eye-cdrom-bbs-network.iso / recent / gus_170.arj / GUS.DOC < prev next >

Wrap

Text File | 1993-04-14 | 58KB | 1,184 lines

╔══════════════════════════════════════════╗ ║ █▀▀▀▀▀▀█ █ █ █▀▀▀▀▀▀█ ║ ║ █ █ █ █ ║ ║ █ ▀▀██ █ ██ ▀▀▀▀▀▀██ ║ ║ █ ██ █ ██ ▄ ██ ║ ║ ████████ ████████ ████████ ║ ║ ≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡ ║ ║ General Unpack Shell ║ ║ version 1.70 ║ ╚══╤═════════════════════════════════════╤═╝ │Copyright (C) 1993 by TRI-SYSTEMS co.│ │ - ALL RIGHTS RESERVED - │ │ Written by Johan Zwiekhorst │ └─────────────────────────────────────┘ T A B L E O F C O N T E N T S ********************************* 1. LEGAL STUFF The no-nonsense licence statement Warranty Contact Payment What you should have received 2. INTRODUCTION 3. USAGE 3.1. General Usage under DOS SYSTEM REQUIREMENTS COMMAND-LINE PARAMETERS NOTE EXAMPLES EXIT CODES 3.2. Unpacking Mailarchives 3.3. Identifying Archive Types 4. BUILT-IN DEFINITIONS 4.1. Built-in Unpacker Definitions 4.2. How To Define Other Unpackers 5. GUS & OTHER SHELLS 6. HOW GUS IDENTIFIES ARCHIVES 6.1. Recognition patterns as used by GUS 6.2. Record layout of ARC/PAK/ARC+ 6.3. How GUS identifies SFX (self-extracting) archives 6.4. Mandatory order of scanning the patterns 7. RUNTIME MESSAGES [A] General information messages [B] Warning messages [C] Fatal error messages 8. ACKNOWLEDGEMENTS 9. REVISION HISTORY ═══════════════════════════════════════════════════════════════════════════════ ╒════════════════╕ │ 1. LEGAL STUFF │ ╘════════════════╛ This software is copyrighted (C) 1989, 1990, 1991, 1992, 1993 by TRI-SYSTEMS co., hereafter called the Owner. All Rights Reserved. The software was written by Johan Zwiekhorst, hereafter called the Author. The No-Nonsense Licence Statement ================================= This software and everything enclosed with it are protected by both Belgian copyright law and international treaty provisions. It is called "freeware". FREEWARE software may be used, copied and distributed freely for NONCOMMERCIAL use only IF: ▀▀▀▀▀▀▀▀▀▀▀▀▀ NO FEE IS CHARGED FOR USE, COPYING OR DISTRIBUTION. IT IS NOT MODIFIED IN ANY WAY. It may be distributed ONLY in it's original, unmodified compressed package file. ~~~~~~~~~~ This means you may not add comments to the compressed package file (also known as an archive file or simply an archive), nor may you delete files from or add files to the archive file, UNLESS YOU HAVE A WRITTEN PERMISSION TO DO SO. Converting the archive file to another compression method or another archive file format is allowed, provided that the above conditions are met. The original package as released by me is in Robert Jung's ARJ archive format. (See below for what you should have received.) In order to extract the files from an ARJ archive, you will need to get the file ARJ***.EXE, where '***' stands for the version number of the program ARJ. At the time this is written, the latest version is 2.30, so look for ARJ230.EXE. Note, that recompressing the archive will nearly always result in a bigger archive. The use of FREEWARE software is prohibited in a governmental or commercial situation. In these cases, this software must be purchased and a Commercial Licence Statement will then be provided for. You may write to TRI-SYSTEMS at the address below for more information. Warranty ======== This software is provided AS IS without any warranty, expressed or implied, including but not limited to fitness for a particular purpose. IN NO EVENT SHALL THE AUTHOR/OWNER OF THIS PRODUCT BE LIABLE FOR ANY DIRECT OR CONSEQUENTIAL LOSS OR DAMAGES WHICH MAY HAVE ARISEN FROM THE USE OF THIS PRODUCT. If your local law does not permit any of the statements made above, or if you do not agree with any of them yourself, THEN YOU ARE NOT LICENCED TO USE THIS PROGRAM! Contact ======= The Author can be reached via electronic mail at phone number +32-89-762626. This is the Tripod BBS. Network addresses: Internet jz@f118.n292.z2.fidonet.org Compuserve >INTERNET:jz@f118.n292.z2.fidonet.org FIDOnet 2:292/118 SIGnet 27:332/0 The Owner can be contacted at the following address: TRI-SYSTEMS co. Langstraat 89 3630 MAASMECHELEN (Belgium) Phone +32-89-764905 during office hours, Central European Time. Payment ======= If you would like to use this product in a commercial or governmental situation, please contact TRI-SYSTEMS at the address above. You will then learn the price of the product and a Commercial Licence Statement will be made available to you. For all others, this product is free, as mentioned before. But if you would like to support the author and encourage him to write more useful software, you're welcome to pay some money. You may pay whatever you feel the product is worth to you. Note that this kind of freeware products is developed entirely in the author's leisure time and he receives absolutely no compensation for it, apart from what you as a user would pay him. If you pay at least U.S. $15 (BEF 500, NLG 30, DEM 25), you will receive, when available, a 5.25" or 3.5" floppy diskette with the next version. Please specify which. Immediately after receiving your payment, I will send you an acknowledgement and a list of the latest versions of all freeware I wrote. Payments to the author can be sent in cash to the address mentioned above or transferred to one of the following bank accounts: Bank Brussel Lambert (Belgium) - account number 335-0076382-89 Rabobank (Holland) - account number 1059.19.519 ***NOTE THAT THIS IS FOR NON-COMMERCIAL SITUATIONS ONLY! For all payments made: please specify NAME and VERSION NUMBER of the product! What you should have received: ============================== You should have received the file GUS_170.ARJ - (32432 bytes) with the following contents: ┌──────────┐ ┌─────┐ ┌─────────────────────────┐ │ filename │ │bytes│ │ description │ └══════════┘ └═════┘ └═════════════════════════┘ GUS .DOC 58872 This documentation. GUS .EXE 11873 The program file. CRC/32 = dec3b859 GUS_WCFG.PAS 6185 TP source for a program CRC/32 = 66948e00 that writes a new configur- ation into GUS.EXE. You may also use the program VALIDATE from McAfee Associates for the purpose of checking the authenticity of the program file(s). It should produce the following: File Name: gus.exe gus_wcfg.pas Size: 11,873 6,185 Date: 4-14-1993 4-14-1993 File Authentication: Check Method 1 - 713A 5944 Check Method 2 - 1FA2 184D ╒═════════════════╕ │ 2. INTRODUCTION │ ╘═════════════════╛ The General Unpack Shell, or GUS, identifies compressed file types and calls the correct unpacker in order to extract the files from them. Its main purpose is, of course, to take work out of YOUR hands. You can use GUS with its straight-forward and easy to remember commands instead of having to learn a new set of commands each time a new archiver sees the daylight. GUS will also work nicely in automated tasks, where any type of archive should be uncompressed, or where a certain file has to be added to any given archive. GUS was made to be command-line compatible with the ARCE.COM unpacker program by Vernon Buerg. This makes it possible for you to rename GUS.EXE to ARCE.COM and have it invoked by any program that expects both ARCE and SEA's ARC-type compressed files, so that such a program will in fact work with any archive format YOU choose. Ben Baker's MAKENL is but one example of such a program. GUS does not require you to fiddle with cumbersome and difficult configuration files: it's just a single EXE file. You copy it into your favourite utility directory and you can immediately start using it, no hassles at all. ╒══════════╕ │ 3. USAGE │ ╘══════════╛ GUS assumes you have located all archiver programs it has to invoke somewhere in your system PATH. GUS is small and it will only occupy about 21K while shelling out to other programs, which should leave more than enough memory for those archiver programs. 3.1. General Usage under DOS ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ SYSTEM REQUIREMENTS: -------------------- GUS will run on any IBM PC compatible computer running DOS 3.0 or greater, provided that at least 25K plus the memory the largest archiver program to invoke would need is available. COMMAND-LINE PARAMETERS: ------------------------ As said before, GUS is command-line compatible with ARCE. Hence, the general syntax is: GUS «compressed_filespec» [filespec(s)] [target_path] [switch(es)] (Entries enclosed within [] are optional, those within « » are mandatory. The [] and « » signs serve to indicate this only and should never be typed!) «compressed_filespec» ::= this specifies where to find the compressed file. If an extension is not given, GUS will assume '.*'. Currently, the following archive types are supported: ARC and ARC+, ARJ, DWC, HA, HAP, HPK, HYP, LZH (both LHarc and LHA), PAK, SQZ, ZIP and ZOO. [filespec(s)] ::= specifies which files should be unpacked. You may give more than one file specification, all of which may contain wild cards. [target_path] ::= specifies where to locate the unpacked files. In order to allow GUS to be as flexible as possible, the ordering of the file specifications and the target path is not important. You may define the target path first and then the files to be extracted. It is even allowed to put the target path in the middle of a number of specifications of files to extract! If you give more than one directory, GUS will ignore all but the last. [switch(es)] ::= specifies one or more of the following switches... /D : Delete archive after successful unpacking /I : Identify only, don't shell out (see 3.3 below) /M : unpack Mailarchives only (see 3.2 below) /N : do Not use embedded path while extracting (for the sake of compatibility with ARCE, /5 may also be used) /P : Print file(s) on standard output device /Q : Quiet mode, suppresses shell output /R : Replace existing files /T : Test archive integrity /Gpswd : supply password 'pswd' for Garbled archive All parameters have to separated by at least one blank. Switches may be joined together without spaces, but the '/' character must be present for each switch. GUS does not support the dash '-' instead of the slash '/'. Options may be given in no matter what case. Only the Delete option _has_ to be in uppercase for safety reasons. NOTE: ----- Consistent with ARCE's behaviour, GUS will create any directories contained within an archive if they do not exist. Both ARCE and GUS have a commandline switch with which you can prevent this and have them extract to the current directory. If you are using the /M (mail archive) switch however, use of the /N (No embedded paths) switch is automatically assumed and unpacking will always be done in the current directory. With /M, the option /D (Delete archive after successful unpacking) is also automatically selected. The /N switch is equally automatically invoked with /P (print) and /T (test). EXAMPLES: --------- 1) Extract all files from an archive CFILE.ANY: > GUS CFILE.ANY 2) Extract all *.COM and *.EXE files from an archive UTILS.ANY into a target directory D:\Utils and replace all existing ones: > GUS UTILS.ANY *.COM *.EXE D:\UTILS /R/N 3) A batchfile LA.BAT which will list any text file inside any archive: LA.BAT: @echo off GUS %1 %2 %3 %4 %5 %6 %7 %8 %9 /P|LIST/S Then you can use this like: LA KBUI_200 *.DOC 4) Test the integrity of all archives in the directory F:\Arcs > GUS F:\ARCS\* /T 5) Unpack all ZIPfiles PW*.ZIP protected with password JiMmY into the directory D:\PWS, not replacing any existing files and using any embedded paths: > GUS PW*.ZIP D:\PWS /GJiMmY EXIT CODES: ----------- GUS will yield an errorlevel of 0 if all operations succeeded. If something's wrong, it will pass on the errorlevel returned by the child program invoked. If the child program could not be started, GUS will return errorlevel 202 = no such program file found in PATH 203 = non-existing directory 204 = too many open files (increase FILES=... parameter in your CONFIG.SYS file) 205 = access denied 206 = invalid handle 208 = not enough memory to start the child program 210 = invalid environment 211 = invalid format 218 = no more files Note, that GUS will not be able to handle a PATH of more than 255 characters properly. 3.2. Unpacking Mailarchives ~~~~~~~~~~~~~~~~~~~~~~~~~~~ If you are a BBS SysOp or a Point connected to a Fidonet Technology compatible electronic mail network, you receive mail packets compressed and bundled within mailarchives. You obviously need to unpack those mailarchives. You can do that in two ways: a) Have your mailprocessor call GUS to unpack the mailarchives. + If your mailprocessor allows you to specify which unpacker it should invoke to decompress mailarchives, have it call GUS. Example for ConfMail: > CM IMPORT AREAS.BBS -K -N -F Conf.Imp -A GUS This approach has the disadvantage that GUS will create any directories embedded in the archive, should they not exist. Another disadvantage is, that the mailprocessor is still in memory. This occupies a lot more memory than is necessary. + If your mailprocessor does not allow you to specify which unpacker to use, it will most likely expect ARCE or PKXARC/ PKUNPAK. Rename GUS.EXE to ARCE.COM or PKXARC.EXE or PKUNPAK.EXE, whatever your mailprocessor wants. If your mailprocessor wants to use an unpacker that needs a specific decompress command (like PAK E Archive), you cannot have it call GUS, since GUS would interpret its first command-line argument as the archive name. Use the method described in item b) to have GUS unpack your mailarchives BEFORE your mailprocessor is started. Since your mailprocessor will only see mail packets then and no mailarchives, there will be no problem. b) Have GUS unpack all mailarchives BEFORE you invoke your mailprocessor. This is the preferred method. You this by starting GUS with the following command-line: GUS «Inbound_directory» /M Instead of the archive name, you specify the path to the directory where your inbound mail is located. Suppose your inbound directory is D:\Opus\NetFiles. GUS will unpack all mailarchives in that directory with the following command: > GUS D:\OPUS\NETFILES /M After a mailarchive has been unpacked succesfully, GUS will delete it automatically. If a mailarchive cannot be unpacked succesfully, then GUS will create a subdirectory BADARC.GUS in your inbound directory and move that mailarchive to it. This allows you to inspect the problematic mailarchives later on while retaining their original name. Other unpack shells always rename a faulty archive to BADARC.001, which makes it very difficult if you would like to restore the archive to its original name after you have inspected and maybe repaired it. I didn't like that procedure, so I decided to let GUS move problematic archives to a special subdirectory instead. Let me know how you feel. 3.3. Identifying Archive Types ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ With the /I switch, you can have GUS simply report which archive type is at hand and do nothing else. This switch can be used just to get a list of archive types, like this: > GUS J:\Outbound\*.MO? /I Type Archive Filename ---- ---------------- LZH J:\OUTBOUND\0009FE64.MO1 ??? J:\OUTBOUND\8FDCB1E2.MO1 ZIP J:\OUTBOUND\8FDCB243.MO0 ARJ J:\OUTBOUND\FF24FFE9.MO0 LZH J:\OUTBOUND\8FDCB241.MO0 GUS yields something like the above list. The '???' means that GUS was unable to determine the archive type, possibly because it isn't an archive at all. In this case, it was a zero-length file. A much more useful way of using this switch is to determine the type of just one archive and act upon that. For instance, to add a file to any archive that comes along. If used with /I, GUS will return an errorlevel from 0 to 13, indicating the archive type. Archive Type: Unknown ARC ARC+ ARJ DWC HA HAP HPK HYP LZH PAK SQZ ZIP ZOO Errorlevel : 0 1 2 3 4 5 6 7 8 9 10 11 12 13 Example: suppose you want to add a header text LOGO.TXT to ZIP and ARJ files, and leave other archive types alone. These batchfile instructions will take care of that: GUS %1 /I if not errorlevel 13 if errorlevel 12 goto IsZIP if not errorlevel 4 if errorlevel 3 goto IsARJ goto Finish :IsZIP PKZIP -z %1 <LOGO.TXT goto Finish :IsARJ ARJ c -zLOGO.TXT %1 :Finish Please be advised that the errorlevel codes assigned to each archive type may change! If a new archive type is added to GUS, its extension will be merged into the list shown above so that the order of the extensions is alphabetical. If you don't like this and would rather see new types simply be added to the end of the list, let me know. I will comply with your wishes if enough people feel like that. At this time, I like to have an alphabetical list and so do most of the people I asked about this. ╒═════════════════════════╕ │ 4. BUILT-IN DEFINITIONS │ ╘═════════════════════════╛ 4.1. Built-In Unpacker Definitions ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ The following unpackers are defined by default within GUS.EXE: ╔════╤════════════╤═══════╤═══════╤═══════╤═════╤═════╤════════╗ ║Type│Program │Extract│Replace│Display│Test │Path │Password║ ╟────┼────────────┼───────┼───────┼───────┼─────┼─────┼────────╢ ║ARC │PKUNPAK .EXE│-n │-r │-c │-t │*****│g ║ ║A7+ │XARC .EXE│ │/o │***** │*****│*****│/g ║ ║ARJ │ARJ .EXE│e -uy │e -y │p │t │<x │─g ║ ║DWC │DWC .EXE│xow │xw │p │t │r │g ║ ║HA │HA .EXE│et │ety │***** │t │<x │***** ║ ║HAP │PAH .EXE│e │e │***** │**** │*****│***** ║ ║HPK │HPACK .EXE│x -on-r│x -oa-r│p -r │t -r │-p │-c ║ ║HYP │HYPER .EXE│-x │-xo │***** │*****│p │***** ║ ║LZH │LHA .EXE│e /m+ │e /m+c+│p /m+ │t /m+│x+ │***** ║ ║PAK │PAK .EXE│e/WO │e/WA │p │t │/PATH│/g= ║ ║SQZ │SQZ .EXE│e /o0 │e /o1 │p │t │<x │***** ║ ║ZIP │PKUNZIP .EXE│-n │-o │-c │-t │-d │-s ║ ║ZOO │ZOO .EXE│e:O │e:OS │e:p │e:N │// │***** ║ ╟────┼────────────┼───────┼───────┼───────┼─────┼─────┼────────╢ ║ 3 │ 12 │ 10 │ 10 │ 10 │ 10 │ 5 │ 5 ║ ╚════╧════════════╧═══════╧═══════╧═══════╧═════╧═════╧════════╝ The numbers above indicate the number of characters provided for each string. Note that Extract, Replace, Display and Test are COMMANDS, while Path and Password are OPTIONS. The difference is, that an OPTION is always combined with a COMMAND and cannot be used alone. If the first character of the Path option is a `<', however, it means that the second character should be used to replace the first command string character. This is currently only the case with ARJ, HA and SQZ, who all need the `e' command to be replaced with `x' in order to extract files while using embedded path information. All spaces means that that particular program does not need any parameters for that particular command. All stars means that that particular program does not support that command or that option. ATTENTION! When encountering an ARC sfx made by SEA's MKSARC program, GUS will identify it correctly, but PKUNPAK can't handle this sfx and will therefore exit with an error. If you expect to handle a lot of MKSARC sfx files, then you'd better replace PKUNPAK by ARC 6.02 in GUS' configuration segment. You may want to redefine some of those built-in definitions for various reasons. To use another unpacker program, for instance. Or to change some parameters. 4.2. How To Define Other Unpackers ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ There are two ways to modify the built-in definitions. You can grab a hex editor and change GUS.EXE directly. If you do, please take note of the lengths of the strings as listed above in section 4.1. All strings occupy the specified number of characters. To achieve that, they were padded with spaces where necessary. If you edit them, make sure you retain the number of characters and do NOT change the funny looking character at the start of each string (that's in fact the string length indicator byte). If a particular parameter is not supported by a certain unpacker program, then you should edit the appropriate field to contain all stars. If you own a Turbo Pascal compiler v5.0 or later from Borland International, you can edit the file GUS_WCG.PAS and re-compile it. This small program will update the configuration information at the end of GUS.EXE for you. Do *NOT* modify the record layout of that information, since GUS will not recognize it anymore if you do. Future versions of GUS may come with a complete setup program to edit and save a new configuration. ╒═══════════════════════╕ │ 5. GUS & OTHER SHELLS │ ╘═══════════════════════╛ Since GUS was first created, other authors have joined the club and released their own versions of a utility that identifies archive types and shells out to the appropriate unpacker programs. Some of those other shell programs come with source, others don't. Some have configuration files, others don't. Some are large, others small. Only one is GUS and all the others "ain't"! :-) Why should you use GUS? 1. GUS is small and fast. Other shell programs typically use a lot more memory than GUS does. 2. GUS provides you with all possible commands to allow not only automatic use, but also easy DOS command-line usage. 3. While scanning a file to determine the archive type, the identification bytes have to be investigated in a well-defined order. Only then, the program will not be fooled by things like archives within archives. GUS is the only program that does this flawlessly: it will never be fooled. 4. GUS has built-in code especially designed for archives that have their identification code at the end of the file. If such an archive has been transmitted by means of a protocol like X-modem, some junk may have been appended to the file to make it grow to the next 128 byte or even 1 K-byte boundary! GUS is the *only* program that will recognize an archive with appended junk because it can skip that while scanning. (At this time, this is only needed for DWC archives, but you never know...) 5. The HAP&PAH, HPack and DWC archivers require their compressed files to have the respective extension '.HAP', 'HPK' and '.DWC' or else they won't work with them. GUS knows this and will rename any such archive that does not have the proper extension before calling the dearchiver program and rename it back to the original name afterwards. 6. Because *I* wrote it! <grin> ╒════════════════════════════════╕ │ 6. HOW GUS IDENTIFIES ARCHIVES │ ╘════════════════════════════════╛ GUS recognizes archives by searching for well-defined patterns in the archive file. Such a pattern can be from 1 to 5 bytes in length and it is extremely important that they be checked in the PROPER ORDER! That is what distinguishes GUS from all it's competitors: most programs do search for the right patterns (with the exception of the pattern for ZOO, which is almost always wrong), but don't do this in the proper order. That can result in faulty identifications, specifically when encoutering archives within archives. 6.1. Recognition patterns as used by GUS ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ArcType Offset Pattern Comment ------- ------ ------------ ------------------------------- ARC 0 0x1A ARC+ 0 0x1A Method byte (offset 1) of all PAK entries needs to be scanned: if HYP >= 0x0A then PAK; >= 0x48 then HYP; == 0x14 then ARC+ Note: PAK can also be recognized by locating the byte 0xFE at offset EOF-2, but GUS doesn't use that because it is less accurate than scanning the method bytes, which has to be done anyway for identi- fying ARC+ and HYP. For completeness, the record layout of an ARC archive will be given in item 2. ARJ 0 0x60 0xEA HA 0 'HA' Offset 4 binary ANDed with 0xFC should yield 0x20. This is an additional check. DWC -3 'DWC' Offset -3 means the third LAST byte of the archive file. It is possible that some junk is present at the end of an archive, because of Xmodem transmissions for example. In order to avoid GUS not recognizing the archive because of this, the last 1028 bytes (or 343 triplets) are read into a buffer and if that contains the string 'DWC', then we have a DWC archive. An additional check will be done, however. The `DWC' string will have to be the last item in a 27 byte structure of which the first two items are ArcStrucSize=27 (2 bytes) and DirStrucSize=34 (1 byte) before GUS will accept the file to be a DWC archive. LZH 2 '-l??-' The '?' specifies a wildcard character. HAP 0 0x91 '3HF' HPK 0 'HPAK' ZIP 0 'PK' 0x03 0x04 ZOO 20 0xDC 0xA7 0xC4 0xFD Most other programs search for the string 'ZOO' at the front of the archive, but that is wrong! Only the ZOO archives made using MS-DOS would be recognized this way. ZOO archives made by an Amiga or a computer running Unix would not be necessarily be recognized this way. SQZ 0 'HLSQZ' 6.2. Record layout of ARC/ARC+/PAK ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ var ArcHeader : record Marker: Byte; Method: Byte; Name : array [1..13] of char; Size : DWord; Stamp : DWord; CRC : Word; Length: DWord; end; Procedure to scan all archive entries: |begin | seek(F,0); | notOK:=false; | YieldARC:=ARC; | repeat | {$I-} | blockread(F,ArcHeader,sizeof(ArcHeader)); | {$I+} | if IOresult=0 | then begin | if ArcHeader.Method>=PAKid | then begin | notOK:=true; | YieldARC:=PAK; | if ArcHeader.Method>=HYPid | then YieldARC:=HYP | else if ArcHeader.Method=ARPid | then YieldARC:=ARP | end | else MoveFilePtr(F,ArcHeader.Size); | end | else notOK:=true | until notOK |end; This is of course all in Turbo Pascal, the language in which GUS was written. The above are in fact literal excerpts from GUS's source code. 6.3. How GUS identifies SFX (self-extracting) archives ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ The basic principle is simple. A self-extracting archive consists of an extraction program in EXE form followed by the archive itself as appended data. The header of an EXE file contains information to determine the size of the EXE portion of the file and hence the offset where the appended data starts. This proved to be true for all archive types, except for SFXs made by MKSARC, the ZIP/sfx as used in PKLTE115.EXE and the ZIP/sfx for OS/2. GUS has those offset values hardcoded. 6.4. Mandatory order for scanning the patterns ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1 - SQZ 2 - ZIP 3 - HPK 4 - HAP 5 - ZOO 6 - LZH 7 - HA 8 - ARJ 9 - DWC 10 - ARC/PAK/ARC+/HYP This order is mandatory because it guarantees the greatest chance for a correct recognition. Every other order would enlarge the chance for a faulty result. This is also the reason why the archive specifications are still built into GUS and not given in a seperate configuration file (like PolyXarc, for example): I still haven't found a good method to have GUS determine automatically in which order the patterns have to be scanned, if a possibility exists that new patterns would be added to the list. I can't expect the users to include new patterns in the proper order themselves, can I? Therefore, I don't think providing GUS with a CFG file is very important at this time. I see no problem for providing a new GUS when a new and exciting archiver is released. That's it folks! If you're curious: the TP source for GUS is about 870 lines in length. Those lines are `filled' in the same way as those of the procedure quoted above. NOTE: you may use the scanning and identification method as used by GUS and as described above in your own programs, but please be so kind and don't forget the reference indicating where you got the information! ╒═════════════════════╕ │ 7. RUNTIME MESSAGES │ ╘═════════════════════╛ GUS may produce a number of messages while it's working. I will list all messages below, with an explanation what's wrong. [A] General information messages ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ MESSAGE: Child program returns exit code # -> this message is given when the invoked unpacker program returns control to GUS. It shows the errorlevel returned by the unpacker program. The "#" will be replaced by the actual exit code. [CHILD]: «unpacker_commandline» -> this is shown in Quiet mode (/Q switch) instead of the archiver's screen output. The «unpacker_commandline» will be replaced by just that. Extract: ALL files (XARC cannot extract specific files) -> this message is shown when the first four characters of the unpacker program definition string are "XARC". Any files to extract specified on the command-line will be ignored. [B] Warning messages ~~~~~~~~~~~~~~~~~~~~ WARNING: multiple target directories defined -- will use the LAST one! -> you have defined more than one target path on the command-line. GUS warns you that it will ignore all paths but the last. WARNING: you need to use /D (capital!) to have the unpacked archive deleted. -> you have specified "/d" in lower case. For security reasons, GUS will only accept a capital D for the Delete switch. WARNING: unknown switch /X ignored -> you have specified a switch that GUS doesn't know. In the above warning message, the "X" will be replaced by the actual character (converted to upper case) that you used. GUS will continue but ignores this unknown switch. WARNING: a slash by itself is not a valid option -- ignored -> you have typed a slash "/" followed by a space or an end-of-line. GUS will continue and ignore this. WARNING: unknown parameter #9 «X» ignored -> you have typed something on the command-line that GUS can't decipher. The actual word you typed will be inserted instead of "X" in the above message, and the number of that parameter on the commandline will be shown instead of the "9" above. WARNING: unable to determine archive type due to error while opening file -> for some reason, GUS can't open the file it has to investigate. Bummer! GUS will simply skip it and continue with the next one, if any. WARNING: can't open NUL device -- Quiet command ignored -> this occurs most often if the /Q (Quiet) command is used and when one or more TSRs were loaded with their output redirected to NUL. This yields a DOS sharing violation error, hence this message from GUS. WARNING: moving bad archive HHHHHHHH.XX9 to X:\Inbound\BADARC.GUS\ -> in mail unpack mode (/M switch), GUS was unable to unpack an archive and warns you that it will be moved to the BADARC.GUS subdirectory that GUS creates in the mail inbound directory. The actual name of the bad archive will be inserted in the message instead of "HHHHHHHH.XX9" and "X:\Inbound" will be replaced by the path to your mail inbound directory. WARNING: unsupported command -- will do normal extract instead -> you tried to perform an action not supported by the particular archiver defined within GUS (i.e., issue a /T [test] command with the HYPER archiver program). GUS warns you it will ignore that command and do a normal extract instead. WARNING: XXX type cannot be unpacked into embedded directories! -> the unpacker program has no option to enable using embedded directories or creating them, so all unpacking will be done into the current directory, since that is the only way. WARNING: XXX type cannot be garbled - ignoring password... -> you supplied an extraction password for an archive whose unpacker program does not support password-protection. GUS will continue the command while ignoring the /G switch. The "XXX" will be replaced by the actual archive type detected. [C] Fatal error messages ~~~~~~~~~~~~~~~~~~~~~~~~ >ERROR<: cannot read configuration information! MESSAGE: aborting with exit code 255... -> GUS complains it can't find the configuration information at the back of it's EXE file. This means something is terribly wrong with that EXE file. You better delete it and get the original release archive unpacked again! (You *did* save that one, didn't you?) >ERROR<: cannot create new filename for rename or move! -> while trying to rename (like when an archiver need a fixed extension) or move (like when a mailarchive couldn't be unpacked) a file, GUS increments the filename (in case of a fixed extension) or the extension (in case of a bad mailarchive) when it at first doesn't succeed with the rename or move. If the filename or extension cannot be incremented anymore and no other options are left, GUS issues the above error message. >ERROR<: cannot erase unpacked archive! -> in mail unpack mode (/M switch) or if the Delete option (/D switch) was used, GUS is unable to delete the archive after it has been unpacked successfully. This more than likely means that the archive was marked Read/Only. You will have to unlock and delete it manually. As mailarchives are created fresh upon receipt, it is very unlikely that they would be marked R/O. >ERROR<: can't rename fixed extension back to original, leaving as is... -> GUS had to rename an archive to a name with the fixed extension required by archivers like DWC, HAP or HPACK and now it can't rename the file back to the original name. This can normally only happen in a multitasking environment, for instance when the archive file was renamed, moved or deleted before GUS could rename it back. GUS issues this error message and leaves the file as it was. >ERROR<: DOS couldn't execute «XXX» due to: YYY -> GUS was unable to load and execute the unpacker program. The path and name of that unpacker will be inserted in the error message instead of "XXX" and the reason will be shown instead of "YYY". That reason will be one of 9 possible problems described in section 3.1. General Usage under DOS, EXIT CODES. If the error code returned by DOS should be unknown to GUS, it will display "DOS ERROR" followed by the error number instead. >ERROR<: error locating directory XXX will unpack in current directory. -> You specified a target directory GUS was unable to find. The target path specification will be ignored and the unpacking will be done in the current directory. >ERROR<: «XXX» is no archive file or a type unkown to GUS! -> GUS encountered a file that is not one of the known archive types. GUS will continue with the next file, if there is one. "XXX" will be replaced by the actual archive name. >ERROR<: no such file(s)! -> GUS was started with an archive filename specification, but no such file could be found. GUS will abort with errorlevel 1. >ERROR<: no such mail directory! -> you specified a mail inbound directory (/M switch) that GUS was unable to locate. GUS will end with errorlevel 0. ╒═════════════════════╕ │ 8. ACKNOWLEDGEMENTS │ ╘═════════════════════╛ + PKUNPAK FAST! Archive Extract Utility Version 3.61 08-02-88 Copyright (c) 1986-1988 PKWARE Inc. All Rights Reserved. + PKUNZIP (R) FAST! Extract Utility Version 2.04g 02-01-93 Copr. 1989-1993 PKWARE Inc. All Rights Reserved. Shareware Version PKUNZIP Reg. U.S. Pat. and Tm. Off. + XARC - to decompress a standard ARC Format Archive, Ver 7.1, October, 1990 Copyright 1990 by System Enhancement Associates, Inc.; ALL RIGHTS RESERVED + ARJ 2.39d PRE-RELEASE Copyright (c) 1990-93 Robert K Jung. Mar 06 1993 All Rights Reserved. U.S. Patent No. 5,140,321 and patent pending. + DWC - Archive utility, Release 5.10, Created 3/07/90 (C) Copyright 1986-90 by Dean W. Cooper; All rights reserved. + HA 0.98 Copyright (c) 1993 Harri Hirvola + Hamarsoft (R) Hap&Pah TM 3.00 Copyright (C) 1992 By Harald Feldmann. Publicly Distributed evaluation copy. + HPACK - The multi-system archiver Version 0.78a0 (shareware version) For Amiga, Archimedes, Macintosh, MSDOS, OS/2, and UNIX Copyright (c) Peter Gutmann 1989 - 1992. Release date: 1 Sept 1992 + Hyper - Pack Utility 2.5 Copyright (c) 1989,1990 P. Sawatzki and K.P. Nischke + LHA version 2.55b Copyright (c) Haruyasu Yoshizaki, 1988-92 + Pak 2.51 Copyright 1988-90 NoGate Consulting + SQZ -- Squeeze It(1.08.3), Jan 24 1993, Copyright J I Hammarberg + Zoo archiver, Version 2.10 (1991/07/09 02:10:34) (C) Copyright 1991 Rahul Dhesi -- Noncommercial use permitted + ARCE Copyright (c) 1986-91 Vernon D. Buerg. Extract ARC files, Version 4.0g, 4/12/91. All rights reserved. + Conference Mail - Revision: 4.07 by Bob Hartman, FidoNet Node 132/101 (C) Copyright 1986, 1987 by Spark Software Inc. All rights reserved. ╒═════════════════════╕ │ 9. REVISION HISTORY │ ╘═════════════════════╛ Ver. Comment ~~~~ ~~~~~~~ 1.70 - Added support for the Dutch HAP archive format, which compresses at least 10% better than PKZIP v2.04. - Added support for the Finnish HA archiver, which beats PKZIP v2.04 by more than 20%! - Added the "/D" or Delete option. - Added detection of SFX (self-extracting) archives. GUS will recognize all EXE variants of the known archive types. Note: GUS can't handle COM self-extractors made by LHarc 1.xx, only the EXE sfx version. - If the drive is full (0 bytes free), GUS is unable to move a bad mailarchive away. It would take quite some time before GUS would give up trying, though. Fixed: GUS will check the drive space and abort the move if less than 32 bytes are free. (32 bytes is what's need to create a new directory entry.) - If a file could not be opened for the purpose of determining the archive type, a runtime error 103 resulted. Fixed. Added a message: «WARNING: unable to determine archive type due to error while opening file». - Some people experienced strange `runtime 162' errors. Here's what happened... GUS tried to open the NUL device and create it. Some environments may have objected to that. I changed the code so that the NUL device is opened for writing only (no creation) and only if the /Q option is used, not always like before. This still caused some systems to yield that runtime error, so I added an error check and disabled the Quiet mode in case the NUL device cannot be opened. That should solve the `runtime 162' problem once and for all... Added a message: «WARNING: can't open NUL device -- Quiet command ignored». (This problem appeared mostly with system that loaded TSRs and redirected their output to the NUL device. That causes DOS to open but never close the NUL device, so when GUS tries to open it, a sharing violation (runtime 162) occurs.) 1.61 - Fixed a bug which caused GUS to use the wrong archive name when operating with an archiver that requires a fixed archive extension (DWC and HPK at this time). 1.60 - Added support for the HPACK archiver from Peter Gutmann, which makes the absolutely smallest archives at this time. - Added support for the Swedish SQZ archiver, which compresses better than ARJ 2.30 or the new PKZIP 1.93a! - Previously, the first three archive types were: ARC (#1), ARJ (#2) and ARC+ (#3). From now on, ARC+ will be #2 and ARJ #3. This is more logical. (The reason for the previous order was that GUS uses the abbreviation ARp internally for ARC+, and ARp comes after ARJ alphabetically.) - Due to a string length mismatch, GUS couldn't tell whether UsePath or UsePassword options were supported or not. The UsePath problem was reported by Wim Van Sebroeck (2:292/862) and by checking that, I discovered that the same was true for the UsePassword option. Corrected. - Added code to allow the Path option to replace a command instead of being added to it. (Indicated by `<' as the first character in the Path option in GUS_WCFG.PAS) Only needed for ARJ and SQZ at this time. (The previously used `e -jf' for ARJ doesn't seem to work equal to `x', so that was changed.) - GUS `forgot' about the specification indicating which files had to be extracted, once an ARC+ type archive had been worked on. Reported by Wim Van Sebroeck (2:292/862) and fixed. - If the specified target directory ended with a backslash (\), GUS wouldn't recognize it as the target directory. Reported by Wim Van Sebroeck (2:292/862) and fixed. - Extract command for ARJ changed from `e -n' to `e -uy' and modified some other archiver parameters in order to make the behaviour of various archiver programs more homogenious. GUS without /R option: (1) XARC and HYPER will ask you if they should overwrite older files. Unfortunately, these programs have no command options to work in batchmode and avoid this. (2) HPACK will never overwrite older files because it doesn't have an option for this. (3) all other archivers will overwrite older files and skip the rest. GUS with /R will cause existing files always to be overwritten. - If someone would combine several real commands on the GUS commandline, the effects might not be what one desires. Example: GUS * /T /R would start unpacking when one would expect it to ignore the replace command and do a test only. So I have changed the behaviour. The /T command now has priority over all other commands. This means that if you specify multiple commands, the /T (test) command will be executed and all others ignored. If you specify /P and /R, the /P has a higher priority and will be executed. The /R will be ignored. The priority of a command increases if it can do less damage. So the priority order for the GUS commands is in descending order: /I /T /P /R (remember: the other switches are options, not commands!). This problem was discovered after a tip by Alex Cleynhens (2:292/500). - Additional check for DWC archives. Apart from the string `DWC' in the last 1K of the file, it is now required that this string is the last item in a 27 byte structure for the file to be identified as a DWC archive. - GUS reported a runtime error when trying to rename an archive to a fixed extension and if the new name existed already. Reported by Wim Van Sebroeck (2:292/862). Fixed: the name is now incremented. Modified the routine which moves bad mailarchives to the BADARC.GUS subdirectory as well: if a file with the same name already exists in the BADARC.GUS subdirectory, the extension of the file to be moved is incremented. - GUS had a problem if the target directory was a root directory. In that case, it specified only the drive instead of the root directory. Reported by Wim Van Sebroeck (2:292/862). Fixed. - HPACK seems to require that its archives have a fixed extension of .HPK (like DWC does). GUS directory /M /T doesn't move bad archives into the BAD_ARC.GUS directory. Reported by Peter Smink (2:285/1). Fixed. - Various changes to the documentation. 1.50 - Added the /N (/5) switch to prevent GUS using or creating paths embedded within archives. - Made this switch automatic while working with mail archives (/M). Thanks to John Lots (2:512/36.3@fidonet.org) and Eef Hartman (2:281/603.5) for suggesting this and detecting the problem with this in GUS 1.40. /N is now also automatically invoked with /P and /T. - Changed the way the configuration information is stored a bit, since there was a useless amount of space being reserved for the "Unknown" type, which of course shouldn't have been saved. - Fixed a minor problem which caused a runtime error when GUS.EXE was given a read/only attribute. Thanks to Rob Essers (2:283/406.2) for reporting this. - Cleaned up the batchfile listing in section 3.3 a bit. Thanks to Roelof Heuvel (27:3331/5000@signet.ftn) for the suggestion. - Fixed a minor problem which caused GUS to not append '.*' to a filename given without an extension when the pathame would contain a dot somewhere. Thanks to Hans Siemons (2:285/214@fidonet.org) for reporting it. - Because of a space inserted between the appropriate switch to supply a password to an unarchiver and the actual password it- self, encrypted archives could never be unpacked. This is now fixed. - Made minor modifications for PKUNZIP 2.00, due to some changes in the way that one handles its command options. - The `use path' option was always supplied with the ZOO unpack commands. Corrected. 1.40 - Removed MDCD archive support again, since nobody was likely to use it -- unless you're looking for the worst performing archiver ever, of course. - Added support for the new LHA version 2 archiver from Yoshi, which succeeds LHARC. The previous version of GUS could already handle the new compression, but I didn't expect the name of the program to be changed. - Added support for the ARJ archiver program from Robert K. Jung, which yields nearly always the best compression and has a lot of features. - Added support for the HYPER archiver program from Germany, which seems to outperform every other archiver on 600..800K logfiles ONLY. Weird. - Added support for ARCfiles made by the new ARC version 7 compressor from SEA. At this time, only one public domain extractor is available, which unfortunately lacks almost every feature GUS has to offer. My thanks to Donn Bly (1:236/7@fidonet.org) and Jeffrey Nonken (1:273/715@fidonet.org) for providing me with all the information on the ARC7+ archive format and the XARC program. - GUS is now fully commandline compatible with Vernon Buerg's ARCE program. All of ARCE's switches are supported - except for /5, which prevents ARCE from creating subdirectories contained within ARCfile entries. - GUS provides two extra options: /I will identify an archive type by means of the exitcode (errorlevel) and /M will unpack and delete mailarchives in Fidonet Technology Networks. - BUGFIXES: * cleaned up handling zero-length and read/only files. From now on, GUS won't abort with a runtime error on those. * you could only specify one single file to extract on GUS's commandline, although the help screen and manual suggested you could give more than one filespec. That's also corrected now. - OTHER IMPROVEMENTS: * the code which detects the archive type has been completely re-written and now is a *lot* faster than before! 1.31 - This version was never released, but mentioned in the documentation of the ARCA*Simulator v2.31 (ASIM_231.LZH). 1.30 - This version was never released, but mentioned in the "Latest Software Versions" column of the FidoNews magazine. 1.20M Added MDCD archives, corrected an error which made GUS not recognize uncompressed file entries in an LZH archive, made sure the new compression method of PAK is supported, added features to allow selection of files and target directory for unpacking. This is a maintenance release, hence the 'M' behind the version number. 1.10 - Added LZH archives, and changed the way DWC archives are identified in order to identify them even if up to 1K of rubbish is appended to the end of a DWC archive. This is useful for DWC archives which have been transferred by means of an Xmodem protocol. 1.00 - Base version. (Turbo Pascal 5.0) _______________________________________________________________________(eof)__