From: dg AT dcs DOT st-and DOT ac DOT uk
Message-Id: <1445.9702022229@dufftown.dcs.st-andrews.ac.uk>
To: opendos AT mail DOT tacoma DOT net
Subject: Re: [opendos] OpenDOS + Win95 w/FAT32? 
In-Reply-To: jamesl@mail.albany.net's message of Sun, 02 Feb 97 15:08:57 +0000.
	     <199702022015 DOT PAA28583 AT keeper DOT albany DOT net> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Sun, 02 Feb 97 22:29:06 +0000
Sender: owner-opendos AT mail DOT tacoma DOT net
Precedence: bulk

[...]
>A TSR wouldn't address the weakness of the current FS, and could 
>cause problems. For example, lookup tables result in a multitude of 
>files that have totally meaningless names, and the loss or corruption 
>of that tsr would make the disk a maze to the average user.  It would 
>be best to use an FS that naturally supports LFN and perform only 
>translation (ie function call <-> native function call and space to 
>dot if necessary, etc...).
[...]

A TSR would allow long-filename-aware programs to run under standard FAT, 
which would let us see if the whole thing was going to work. It's not entirely 
necessary to have totally meaningless names. We could, for example, keep the 
first eight letters of the filename and use the extension for the index into 
the lookup file. Of course, to allow long filenames in system files, it would 
have to be built into the kernel.

*Are* there any standard DOS long-filename-aware programs? Under Win95, what 
happens if you type, say:

	EDIT "This is a long filename"

...?

According to Ralf Brown's interrupt list, there are three interrupts. The 
multifunction one I mentioned earlier, a long-filename FindFile, and a 
long-filename FindNext. It would be relatively easy to modify the standard C 
libraries to use these if possible and if not fall back on the old-style 8.3 
ones. Then a simple recompile would convert a program. The difficulty is in 
fixing those programs that assume that a filename is a maximum of 11 bytes, 
and that would have to be done one at a time.

The next stage would be to build it into the kernel so all programs could use 
the long-filename interrupts. Then, changing to a different file system would 
be relatively trivial. Instead of the long-filename interrupts being 
implemented on top of the 8.3-filename interrupts, it would be the other way 
round.

-- 
------------------- http://www-hons-cs.cs.st-and.ac.uk/~dg --------------------
   If you're up against someone more intelligent than you are, do something
    totally insane and let him think himself to death.  --- Pyanfar Chanur
---------------- Sun-Earther David Daton Given of Lochcarron ------------------