What MS did with SFN's and LFN's is, IMO, a sobering
example of a bad design, in that having two FN's/file, with
both being valid ways to refer to the file, has a lot of ramifications.
Someone mentioned earlier the problem of a program creating a
temp file, deleting the original, and renaming the temp file to the
original file's name--a very common sequence in many applications.
How does the pair of names travel from the old instance of the file
to the new one? There's no API in Windows to set just one of the
names, after all. Believe it or not, when you delete a file, Windows
remembers the names for some number of seconds (I forget the
setting), and if you then create or rename a file using one of the
names from the deleted file, the other one is also used.
This is how MS solved this problem back in Win95, and I haven't
checked lately to see if it's still the same in Win98 and Win2K. I
don't know which would be worse--if it's the same, or if they came
up with a more "creative" solution...
Lou
-----Original Message-----
From: owner-linux-kernel-digest@vger.rutgers.edu
[SMTP:owner-linux-kernel-digest@vger.rutgers.edu]
Sent: Tuesday, August 03, 1999 1:03 AM
To: linux-kernel-digest@vger.rutgers.edu
Subject: linux-kernel-digest V1 #4240
From: Steve Dodd <dirk@loth.demon.co.uk>
Date: Mon, 2 Aug 1999 20:52:19 +0100
Subject: Re: VFAT naming stuff
On Mon, Aug 02, 1999 at 11:34:22AM +0100, degs wrote:
> No long-filename aware windows program should access any file by its short
> name, except backup software and file managers.
But I think that's the problem -- they do (apparently).
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/