Date: Thu, 9 Apr 1998 15:00:56 +0200
Message-Id: <199804091300.PAA14541@ppp37.a2points.com>
From: Yann Dirson <ydirson@a2points.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
To: Debian bug-system submission <submit@bugs.debian.org>
Subject: file, magicfilter: duplicate information - magic numbers

Package: file, magicfilter
Version: 3.24-4, 1.2-20
Severity: wishlist

Both file and magicfilter use a magic number DB to identify file
formats.  This leads to duplicate information - the DB's can't be
expected to be in sync all of the time.

Further more, both text DB formats are limitted to each one's field of
application.

It would be great to have a general-purpose magic-number-handling
library, with a generic DB that would associate magics to cannonical
ID's, and maybe other recognizable infos, like file can identify.
"file" would be able to output its info in whatever language the user
wishes, and "magicfilter" would have the same knowledge base as "file"
does.

Maybe the work I did for the console tools (recently uploaded into
slink as package utils/console-tools) could be used as a base for
this, although it is not generic enough as such: magic definitions are
hardcoded in a struct - this can be easily fixed by using an external
DB; there are only basic informations that allow to recognize a file
with a magic+mask starting at offset 0 - the data structure have to be
extended / made more generic).
If your're interested in this idea, you will find the relevant code in
console-tools' source package, in files lib/findfile*.c.

-- 
Yann Dirson  <ydirson@a2points.com>      | Stop making M$-Bill richer & richer,
alt-email:     <dirson@univ-mlv.fr>      |     support Debian GNU/Linux:
debian-email:   <dirson@debian.org>      |         more powerful, more stable !
http://www.a2points.com/homepage/3475232 | Check <http://www.debian.org/>