Amiga Format CD 46

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

/ Amiga Format CD 46 / Amiga Format CD46 (1999-10-20)(Future Publishing)(GB)[!][issue 1999-12].iso / -serious- / misc / gettype / gt.doc < prev next >

Wrap

Text File | 1999-09-06 | 25KB | 604 lines

+--------------------------------------------------------------------------+ | Documentation for GetType 1.14b | | | | Program & related files are copyright © 1997-1999 Andrew Bell. | | | | Homepage: http://www.homeusers.prestel.co.uk/ab2000 | | | | email: andrew.ab2000@bigfoot.com | | | +--------------------------------------------------------------------------+ ---------------------------------------------------------------------------- DISCLAIMER ---------------------------------------------------------------------------- BY USING THIS PROGRAM THE USER ACCEPTS FULL LIABILITY FOR ANY DAMAGE THAT COULD POSSIBLY BE CAUSED BY ITS USAGE OR MISUSE. THE AUTHOR OF THIS SOFTWARE CANNOT BE HELD RESPONSIBLE FOR ANY DAMAGE. THIS PROGRAM IS PROVIDED AS-IS. USE THIS PROGRAM AT YOUR OWN RISK. (After all, you did get it for free!) ---------------------------------------------------------------------------- DISTRIBUTION ---------------------------------------------------------------------------- IT IS PROHIBITED TO DISTRIBUTE GETTYPE WITHOUT THE FOLLOWING FILES: GetType/GT GetType/GT.doc GetType/GT.readme GetType/GT_Types.txt GetType/Libs/Idm.library GetType/Libs/Idm.library.readme GetType/Libs/FileID.library GETTYPE IS FREE, AND NO PROFIT SHOULD BE MADE FROM DISTRIBUTING IT. ---------------------------------------------------------------------------- OVERVIEW ---------------------------------------------------------------------------- GetType is a CLI/Shell program that enables the user to identify files via a Shell/DOS prompt. It requires an 020 processor or better, and at least version 39 (aka OS3.x) of the Amiga operation system. It also needs Idm.library version 1 or later. The actual identification work is done by Idm.library (and FileID.library / xfdmaster.library), currently this library recognizes over 1100 filetypes. Using the UFID and UXFD, you check now check files with a database of over 2000 filetypes. I recommend using an 030 CPU or better. ---------------------------------------------------------------------------- INSTALLATION ---------------------------------------------------------------------------- Simply copy the executable (GT) to the C: drawer, and Idm.library to the LIBS: drawer. Copy FileID.library to LIBS: also, but check to see if you already have a higher version already installed, if so you don't need to install it. If you have xfdmaster.library or xvs.library installed then GetType can take advantage of these systems. See USAGE section for more details. ---------------------------------------------------------------------------- PROGRAM USAGE ---------------------------------------------------------------------------- Once installation has been completed, you can now start the program. Simply type GT at the Shell/DOS prompt to start it. The command line is as follows: ---> DIR/M <--- Use this parameter to provide the locations of the files and dirs to scan. ---> ALL/S <--- Enter into sub-directories. When using this option the output can be a little confusing on big HDs with deep directory trees. If you don't like the output, consider using the LFMT parameter. The ALL option is not really designed to be used with the default output. ---> HL_U/S <--- Highlight unknown files. ---> HL_K/S <--- Highlight known files. ---> HELP/S <--- A simple help page giving details on the parameters. ---> INFO/S <--- More information on the currently installed version of Idm.library. ---> TYPES/S <--- Display a (very long) list of all the filetypes known to the currently installed version of Idm.library. If this list is too long, then press CONTROL+C to break. ---> IDPAT/K <--- An AmigaDOS pattern/wildcard string, used for finding certain filetypes. For example, if you wanted to view only the filetypes recognized by FileID.library then type something similar to this: GT <PutPathHere> <PutOptionsHere> UFID IDPAT=FID:#? Note: Uppercase and lowercase are both treated as the same. ---> DGEN/S <--- Display general filetypes (Executable/Data) instead of a question mark '?'. ---> SC=SETCOMMENT/S <--- Set file's comment with identification string. Don't use this feature if you place great value on your file comments. ---> SI=SHOWICONS/S <--- Show #?.info files. By default GetType does not show icon files. ---> IOLEN/N <--- Specify a new read length. This value sets the amount of bytes to be loaded off the start of each file, while scanning. Idm.library requires GT to load about 2Kb from the start of each file in order to identify it. By using this option you can force GT to load a smaller amount of bytes from each file, doing this will result in less files being identified but a faster performance. Any number below 4 will become 4. NOTE: If you use the UFID option the IOLEN parameter will be forced to at least 1200 bytes. This is required because FileID.library needs a minimum of 1200 bytes to be loaded off the start of each file, to ensure a proper identification check. This is also the same for xfdmaster.library, about 5KB is required for proper identification when using the UXFD=USEXFD/S parameter. For the best result, don't use either the UFID or UXFD parameters with the IOLEN parameter. ---> UFID=USEFILEID/S <--- Tells GT to use FileID.library for any unknown filetype encountered. v8+ of this library is required. Any IDs made by this library will have a 'FID:' prefix for clarity (except for LFMT's %2 code). ---> LFMT/K <--- Perhaps one of GTs more powerful parameters, using this parameter allows the user to create simple script/batch files or configure the output. At the moment the default output is very ugly, so you might want to consider using this option :) especially if you are using DOpus, DirWork, etc. This parameter is similar to the LIST command's LFORMAT parameter. The user will pass a formatted string and GT will parse it with the desired information. So for example if we wanted to delete all StoneCracker compressed files in the C: drawer, we could type the following at a Shell prompt: GT c: all lfmt="delete %p%n FORCE" idpat=#?stc#? >ram:tempscriptfile execute ram:tempscriptfile delete ram:tempscriptfile In order to build the output string via the LFMT parameter we need to know what format components codes are available: %% = insert a literal % character %P = insert full path (same as %p%n, eg: dh0:mydrawer/myfile) %p = insert path string (eg: dh0:mydrawer/) %n = insert filename string (eg: myfile) %l = insert entry type string (eg: F, LF, PF, etc.) %s = insert size in bytes string (eg: 153432) %S = insert size string (rounds off lengths to K, MB, GB, etc.) %i = insert primary identifcation string. %1 = insert identification result of Idm.lib %2 = insert identification result of FileID.lib (without FID: prefix) %3 = insert identification result of GT name matching. %4 = insert identification result of xfdmaster.lib (without XFD: prefix) %5 = insert identification result of xvs.lib (without XVS: prefix) %c = insert file comment. %b = insert the amount of blocks the file takes up. If you insert a '-' and THREE decimal digits immediately after the '%' you can specify the width in characters you wish the string to fill. The remainder of the string will be padded with spaces. This will let you create more readable strings, but is not much use when creating script files (as shown above). Remember when using the '-' symbol that you MUST put three decimal digits after it. So NEVER use %-8n, use %-008n instead. If you want a full path use %P and not %p%n, remember these format codes are case sensitive. I hear you ask: What is the point of having %P when %p%n does the same job? Well the answer is: %P works better with the '-' symbol. Simple. If the string is bigger than the number you pass with the '-' symbol, then the bounds will be broken. If you want to limit the amount of characters inserted, then you can append a '.' (period) followed by a THREE digit number that will specify the maximum allowed characters. It you need spaces in your LFMT string remember to enclose it in "double quotes" (eg: LFMT="File %n is %i"). If all this seems complicated, thats because it is :) This parameter is for the advanced Amiga user only. Believe it or not, but you can create a more readable GT output by using: gt dh0: all lfmt="%-040.040P %-028.028i %s" dgen ---> FIDTYPES/S <--- Display a (very long) list of all the filetypes known to the currently installed version of FileID.library. If this list is too long, then press CONTROL+C to break. ---> NNM=NONAMEMATCH/S <--- Since version 1.12, GT now attempts to identify files via their names. Passing this option will force GT not to do any name matching. For example, GT is able to recognize a standard Amiga run-time libary by checking to see if the name matches *.library. The NONAMEMATCH option will disable this method of identification. ---> NAMETYPES/S <--- Display all known name prefixes/extensions (aka name types) types known internally to GetType. Since v1.12 GT has its own internal database of common filename prefixes/extensions, using this, GT can dertermine basic filetypes. Passing this option GT will enable you to see all known prefixes/extensions known to GT. ---> REQ/S <--- This will enable you to force the output to a requester instead of the Shell window. To use this option you must specify the LFMT parameter. The LFMT parameter is used to construct the body text of the requester. If you want a newline inside the requester use a *n or a \n inside the LFMT string. For example: gt c: LFMT="%P\n\nIdentification is:\n\n%i" REQ DGEN Notes: 1. Currently the requester is opened on the default public screen. 2. Each line is centred, there is currently *NO* way around this! But using the '-' after an LFMT code might solve this problem. Some people may prefer this option, especially if DOpus or DirWork is being used. ---> UXFD=USEXFD/S <--- If you supply this argument, GT will use the powerful xfdmaster.library to scan for compressed files. You *MUST* have the XFD system setup correctly on your system to take advantage of this parameter. The XFD system is currently maintained by Dirk Stöcker and the lastest version can be found on the Aminet at: util/pack/xfdmaster.lha. ---> XE=XFDEXTERN/S <--- Tell xfdmaster.library to use the external slaves. For more information on the external slave system used by XFD, please refer to the XFD documentation. There is no point in using this parameter if you are not using the UXFD=USEXFD/S parameter. ---> XFDTYPES/S <--- Display a (very long) list of all the cruncher types known to the currently installed version of xfdmaster.library. If this list is too long then press CTRL+C to break. ---> CFV=CHECKFORVIRUSES/S <--- Since GT 1.14, if you supply this argument, GT will use the powerful xvs.library to scan for viruses. You *MUST* have the XVS system setup correctly on your system to take advantage of this parameter. Any viruses encountered will be prefixed with XVS: (except when using the LFMT paramter). GetType only identifies viruses and does NOT kill them, nor does it attempt to fix your files. This is up to you. If GetType finds a virus on your system then I recommend that you check the files in question with an up to date virus killer. Note: Using this feature will greatly slow down GT, because the whole file is loaded into memory. The XVS system is currently maintained by Georg Hörmann and the lastest version can be found on the Aminet. ---> XVSTYPES/S <--- Display a (very long) list of viruses known to the current version of xvs.library. Bootblocks viruses are not shown. If this list is too long then press CONTROL+C to break. ---------------------------------------------------------------------------- INFORMATION & EXTRA NOTES ---------------------------------------------------------------------------- GT (by default) does not display icon files, use the SI=SHOWICONS/S option to override this. If you notice some short pauses while running GT, this is normally the icon files (#?.info) being skipped. For each release of Idm.library the IOLEN can change. For Idm.library v1.85 it stands at 1470 bytes. This might seem like a lot but since harddrives are getting faster these days, who cares? The output of GT will be faster if a console patch like KingCON is not used. (KingCON tends to slow down while updating its scroller gadget and history text buffer) If scanning big harddrives I recommend you redirect the output of GT, for example: GT dh0: all >ram:GT_output.txt, doing this will make life faster :) It will also give you a big output file! So make sure you have a lot of spare storage space in the destination drawer. ---------------------------------------------------------------------------- THE OUTPUT ---------------------------------------------------------------------------- Here is an example of GetTypes output in Shell window: Store:Visual/ Spin.IFF - IFF ILBM picture - F 5420 test1.anim8L - IFF Animation file - F 13080 test1.anim5 - IFF Animation file - F 9258 XenonII_sprites.ilbm - IFF ILBM picture - F 9110 Monsters of Terror - IFF ILBM picture - F 33068 walkman - IFF ILBM picture - F 18554 SampleFont_EMCVol8.ilbm - IFF ILBM picture - F 11570 samplegfx01.ilbm - IFF ILBM picture - F 32220 (END OF DIR) The output is divided into four sections these are: 1. File name - Name of the file processed. 2. Identification - Identification of the file. 3. File entry type - This will be one of the following: RT = ST_ROOT UD = ST_USERDIR SL = ST_SOFTLINK LD = ST_LINKDIR F = ST_FILE LF = ST_LINKFILE PF = ST_PIPEFILE ?? = Unknown type RT, UD, SL, LD will not appear in this version of GT. 4. File length - Length of the file in bytes ---------------------------------------------------------------------------- HOW GETTYPE WORKS ---------------------------------------------------------------------------- GetType uses a standard Amiga run-time library known as Idm.library, (short for IDentification Master). This library holds the actual information required by GT to determine a file's type. When you run GT, GT loads Idm.library into memory. GT then scans each file in the specified directory and loads a certain amount of bytes off every one it encounters. It then passes this data to Idm.library which then attempts to identify the file via a process called DNA (for lack of a better word) matching. Most data files produced by applications normally contain some sort of an ID string, for example an OctaMED module would contain the one of the following DNA strings at offset 0 into the file: MMD0, MMD1, MMD2 or MMD3. Idm.library contains many such DNA strings, each DNA string has a description associated with it, this is what you see if Idm.library is able to match a DNA string. But a major problem with this method is, there's a very high rate of incorrect matches! Say for example you create a text file and at the VERY beginning (that means no spaces or newlines) of this file you typed 'AMOS Basic V1.00' (must be exactly as you see it here without the quotes!) Idm.library would pick this up as an AMOS source code file when you scan the directory that this text file resides in with GT. But we know that this is just a simple text file. So the user (thats you!) must use his or her's common sense to determine what GT matches are correct and what matches are not correct. ---------------------------------------------------------------------------- COPYRIGHT ---------------------------------------------------------------------------- GetType is copyright © 1999 Andrew Bell. This software is FREEWARE. Commercial use of this program (and related files) is prohibited. FileID.library is copyright © 1998 Oliver Lange (BLOODROCK). Full package can be found on Aminet: utils/libs dir. ---------------------------------------------------------------------------- PROGRAM HISTORY ---------------------------------------------------------------------------- History for GT versions 1.0-1.9 not documented, early test versions. 1.10 (31.Jan.1999) - Aminet release. o First public release via Aminet. 1.11 (07.Feb.1999) - Aminet release. o Idm.library 1.19 had a bug in it that stop it from working with OS40 (WB3.1) or greater. This was a very stupid bug. Delete this version if you encounter it. o Some minor changes made to the main program. o Cleaned up the document, looks nicer :-) o GT now shows the amount of files hidden from view. o Removed tabs from the help page. o Added LFMT parameter. o Now includes version 8 of FileID.library. Thanks Darek! o Add FIDTYPES options. o The help page stated in 1.10 that GT required Idm.library version 0, this is wrong it should have been version 1. 1.12 (18.Feb.1999) - Aminet release. o GT is now able to determine a file's type via its filename. o Added NNM=NONAMEMATCH/S parameter. o Added NAMETYPES/S parameter. o Extended LFMT parameter, added %1, %2 and %3 format codes. o For completeness I added support for C style newlines \n in the LFMT parameter. o The output of GT can now be forced to a WB requester by using the REQ/S parameter. o More minor internal changes. o GT has more tolerance now when it comes to errors. It won't quit on the first error it encounters, instead it will attempt to continue with the scan. o You can now pass more then one file/dir name to GT in a single command line. This makes GT more powerful when used with DOpus or DirWork. o Extended the LFMT parameter by adding the %c format code. This will enable you to insert file comments into the output. o Also added the %b code to the LFMT parameter, this enables the user to insert the amount of blocks a file takes up. o For convenience, I added GT_Types.txt to the archive. 1.13 (26.Apr.1999) - Aminet release. o GT did not check the OS version before starting - Whoops! o You can now use the powerful xfdmaster.library by Georg Hörmann & Dirk Stöcker to check for compressed files that Idm.library does not know. The following parameters have been added to enable the user to bring in xfdmaster.library: UXFD=USEXFD/S, XE=XFDEXTERN/S and XFDTYPES/S. GT now has a large database of filetypes! o LFMT has been extended, it now supports XFD. Use LFMT code %4 to obtain the result of xfdmaster.library. o The INFO parameter has been extended. It now displays more information. o GT now breaks first time (CTRL+C) when the user specifies many #?TYPES parameters (eg: GT TYPES FIDTYPES NAMETYPES XFDTYPES). o GT now has EVEN MORE tolerance when it comes to errors. o Fixed some spelling mistakes in the document. I type too fast :) o Made some minor internal changes and optimizations. 1.14 (13.Aug.1999) - Limited beta release. o Reworked GT's exit and cleanup routines. o Source code ported to the Hisoft/Maxon C++ compiler. o I made a lot of internal changes. o Some extra nametypes added. o Due to requests, directories are now displayed in the output. o Virus checking added, via xvs.library. This feature can be activated with the CFV=CHECKFORVIRUSES/S parameter. o Added the XVSTYPES/S parameter for listing all virus types known to xvs.library. o Extended INFO/S option to include information about xvs.library. o Extended the LFMT parameter to support xvs, use LFMT code %5 to obtain the result of xvs.library. o The ALL/S now tells the user what directory it's entering, leaving or continuing with. o Removed some potential enforcer hits, when an empty file was encountered and the LFMT parameter was being used. o Fix a small bug concerning the USEXFD/S parameter. o Extended the HELP/S parameter list to include new xvs options. o If GT can't lock a dir, it now displays the name the dir with and error instead of just saying "object not found" or similar. o Totally reworked the printing routines inside GT. As a result output speed has increased by about 5-10% - /me thinks :). o Extended the internal nametypes routines. GetType now attempts to identify a file via it's prefix, not just the suffix/extension. o Added lots of filetype prefixes to the database. Mostly music module formats, for now. o Updated NAMETYPES/S parameter to show prefix database. o All the listing options (TYPES, NAMETYPE, XFDTYPES, ...) now count from 1. This makes life less confusing :) o Some extra bits added to the doc. o GetType now includes it's compile time, right down to the second - thanks to the Maxon compiler. This can be seen in the help page (use the HELP/S parameter). Hmm, I wonder if that can be considered a feature :) 1.14b (15.Aug.1999) - Aminet release. o Fixed a small bug. GetType was printing an incorrect figure at the end of the scan. The amount of entries processed figure, did not include directories. ---------------------------------------------------------------------------- ADDITION INFORMATION ---------------------------------------------------------------------------- GetType is not pure nor re-entrant. So don't even think about making it resident. Maybe future versions will support this, but this one does not, because generating such code makes the executable bigger (Well, using my methods anyway!). But what's the point in making programs resident these days when HDs are so fast? ---------------------------------------------------------------------------- TODO ---------------------------------------------------------------------------- AmigaGuide / HTML documentation! Maybe an installer script, but the package is still small. Config file, so the user does not have to type the same parameters at the Shell prompt every time when starting GT. Add a format code to the LFMT parameter that will enable the user to insert the file's date into the output string. ---------------------------------------------------------------------------- THANKS ---------------------------------------------------------------------------- To Oliver Lange (aka BLOODROCK) for creating FileID.library. To Dariusz Gac for helping me solve the problems with v1.10 and for giving me the idea for the LFMT parameter. Also, sorry for spelling your name wrong in the 1.11 docs :) To Rick Hodger for helping me test GT and his bug reports. To Dirk Stöcker for pointing out various problems Idm.library was having identifying some files. Thanks also for maintaining the XFD system and sharing your knowledge on various compressed (+ archive) file formats. To Georg Hörmann for creating the XFD system (And making it one of the easiest systems to program - wish I could say the same about Idm.library :) To StarBlazer for his suggestions on Idm.library. Maybe they will be implemented in v2 of Idm.library (Yes, I'm still working on it!). To James L Boyd for testing Idm.library with Blitz Basic 2. And sending loads of developer stuff for Blitz. To Richard Lane for allowing me to use filetypes from his excellent Magnum Opus package. <--- your name could be here :) ---> Thanks to all of those people who bothered to send bug reports and offered suggestions. Sorry if I forgot about you. ---------------------------------------------------------------------------- THE AUTHOR ---------------------------------------------------------------------------- send info, filetypes, bug reports, ideas, suggestions, anything, etc. too: Andrew Bell : andrew.ab2000@bigfoot.com (remember to put 'GetType' in the subject) Visit: http://www.homeusers.prestel.co.uk/ab2000 to get the latest release of this program and to see all of my other work.