X-Authentication-Warning: delorie.com: mailnull set sender to djgpp-bounces using -f
X-Authentication-Warning: kendall.sfbr.org: jeffw set sender to jeffw AT darwin DOT sfbr DOT org using -f
Date: Wed, 1 May 2002 15:23:54 -0500
From: JT Williams <jeffw AT darwin DOT sfbr DOT org>
To: djgpp AT delorie DOT com
Subject: Re: LFN on pure DOS
Message-ID: <20020501202354.GB3136@kendall.sfbr.org>
Mail-Followup-To: djgpp AT delorie DOT com
References: <c21a43ff DOT 0204241352 DOT 74858db3 AT posting DOT google DOT com> <20020425010620 DOT GA17090 AT taniwha DOT org> <1019895346 DOT 635577 AT queeg DOT ludd DOT luth DOT se> <20020430132704 DOT GA1671 AT kendall DOT sfbr DOT org> <aam6v9$60m$1 AT antares DOT lu DOT erisoft DOT se>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <aam6v9$60m$1@antares.lu.erisoft.se>
User-Agent: Mutt/1.3.27i
Reply-To: djgpp AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

-: -: http://taniwha.org/lfn.zip
-: 
-: -: If there is interest, I can upload my WIP version of Bill's lfn.zip to
-: -: DJ's site (it's very small).  I still tinker with it from time to time.

(With Bill's permission, I should add.)


-: What changes have you made?

I split the source code into more manageable units, reformatted it,
added comments, etc.

I ripped out support for DOS != 5.0 and FAT != 16-bit.

I removed everything related to the LFN cache (cleverly tucked away in
the PSP).

I replaced the CRC algorithm for LFN->SFN mapping with simple LFN
truncation to 8.3 (just for testing; CRC or any other mapping can be
introduced later).

I threw out support for Unicode.

Rather than devour unused directory entries as M$ does, I write the LFN
information to a file in the current directory.

This was ages ago; I haven't worked with the driver for some time.
The idea was to burn away everything but the bare essentials and get the
basic LFN TSR engine stable.  Anything I chopped can be restored later.

There is a weird interaction with FNCASE, but I can't fully describe
the problem yet.